The Law Commission

APPENDIX A REPORT ON PUBLIC SURVEY OF MURDER AND MANDATORY SENTENCING IN CRIMINAL HOMICIDES^*

Barry Mitchell

Professor of Criminal Law and Criminal Justice

Coventry University

INTRODUCTION – OUTLINE AIMS AND METHODOLOGY

A.1 This is a report of a short public survey carried out between 1 and 28 September 2005 on mandatory sentencing in criminal homicides. The central aim of the survey was to determine whether there appears to be any evidence of the likely support amongst members of the public for mandatory sentencing of persons convicted of unlawful homicide, (probably murder). Further, the survey sought to identify (1) the kinds of offence characteristics in which mandatory sentencing is favoured, and (2) the reasons which underpin participants' views (for or against mandatory sentencing).

A.2 Fuller details of the methodology and the participants in the survey are set out at the end of this report. In essence, the survey was conducted by convening five groups of people who were drawn from different parts of England, with one each from the north, south, east, west and midlands. Each group met on two occasions, with a week between meetings. In total, 56 participants attended both meetings.^[1] The first meetings concentrated on discussing different kinds of homicides so as to encourage and enable participants to think about the different kinds of circumstances in which homicides might occur and then to identify variations in the seriousness of homicides – how individual factors (offence characteristics) might affect their perceptions of the seriousness of the killing. During the week between meetings participants were free to talk about the issues with family, friends, colleagues and neighbours etc., if they so chose but were not asked to do so. Discussions in the second meetings focussed on the sentencing of convicted killers – were there any kinds of homicides in which participants favoured some form of mandatory sentencing. Whatever the responses, participants were invited to articulate the reasons for their views.

TWO "CAVEATS"

A.3 Whilst the general public are aware that criminal homicides are committed and offenders prosecuted and sentenced in the courts,^[2] it is extremely unlikely that any of them will have examined the issues in any depth: this survey was almost certainly the first time that they will have been encouraged to do so. Moreover, one of the main reasons for holding group discussions (as opposed to individual interviews) was to provide an opportunity for participants to listen to the views of others and to consider different and sometimes opposing arguments. Thus, it is possible that some participants might change their mind whilst considering and reflecting on the issues. Similarly, they might contradict themselves in the course of the discussion. In addition, not every participant responded in detail on every issue, but tended to indicate a general agreement or disagreement with what had been said by someone else. Indeed, as criminal lawyers are only too well aware, some of the issues here are extremely complex, calling on moral, ethical and legal analyses, and so it was not surprising that occasionally participants felt unable to express a view on a point. In consequence it is not possible to produce a definitively quantitative set of results.

OPINIONS

Bad homicides and mandatory sentencing

A.4 Support for mandatory sentencing seemed to vary quite considerably between groups. In group 1, at least six and up to eight of the eleven participants spoke in support of mandatory sentencing – two appeared to vary in their views at different times in the discussion. It is worth noting that in the latter stages of their discussion several members of this group, including some who had earlier indicated support for mandatory sentencing, appeared to favour a different approach, namely the creation of a set of guidelines for different kinds of circumstances (such as battered spouse homicides, killing in self-defence, killing during pub or club fights, gangland killings etc.) within which judges should sentence. The principal reasons were that cases inevitably varied from one another and there would usually be some, albeit minimal, element of mitigation which might affect the sentence. Even relatively small adjustments in sentencing were thought to be desirable. Similarly, in group 2, up to eight of the twelve participants thought there were some instances where the trial judge should have no choice as to sentence, but one (of these eight) was less than sure about this. In groups 3 and 4, the corresponding figures were seven and six out of twelve respectively, but in the final group only one person advocated mandatory sentencing. Thus, taking account of the slight uncertainty, between 48.2 and 53.6% of the participants felt that in some kinds of homicides the sentence should be fixed by law.

A.5 Predictably, each group felt that homicides would vary in their seriousness because there would almost inevitably be a mixture of aggravating and mitigating characteristics. The obviously crucial issue is the extent to which this recognition of a cocktail of aggravation and mitigation might lead participants away from mandatory sentencing. Indeed, the kinds of homicides which were identified as deserving a mandatory sentence is (unsurprisingly) of potential significance here. In the first group three participants identified premeditated killings, three more referred to serial killings, whilst other individuals mentioned murders in the course of rape and gangland killings. Interestingly, in contrast to the other groups, no-one in this group referred at this stage to the general category of killing children as especially serious, although there was widespread agreement that certain notorious offenders warranted the toughest penalties because they had killed children. In the first meeting of this group, the majority of participants expressed the view that killing multiple victims through a single act (such as detonating a bomb or setting fire to a building) was particularly bad, as was a killing where the offender inflicted torture or made the victim endure additional suffering as well as death. This latter point substantiates the opinion expressed during the second meeting of the group that murder in the course of rape is very serious; in each instance there is an additional element of (serious) harm and accompanying fault. At the first meeting, participants had suggested that killing not only a child but also the elderly or the handicapped was very serious. On the other hand, there was some disagreement in the group about killing a police officer, or anyone else in a "vulnerable occupation".^[3]

A.6 In the second meetings of group 2, at least two participants identified homicides in the course of acts of terrorism as warranting a mandatory sentence, four mentioned hired or professional killers, two spoke of premeditated killing, and four referred to different kinds of child killing (adults killing children, children killing children, adults committing sexual killings of children, and parents killing their own children), as calling for a mandatory sentence. As with all five groups, there was unanimous agreement during the first meeting that killing a child or an elderly person or a handicapped person was likely to be regarded as one of the more serious homicides, although there was no apparent support for treating the killing of those in vulnerable occupations as unusually bad.

A.7 In all the first meetings and in some of the second meetings of all groups there was apparent agreement that premeditation, in the form of nothing less than an intent to kill,^[4] is likely to indicate a particularly serious homicide, subject to there being no "good motive" – the most obvious example of the latter being a mercy killing. Further, the means by which death is caused was recognised as possible circumstantial evidence of the killer's intent, and there was widespread opinion in the groups that the use of a gun (and perhaps a knife) indicates an intent to kill – though it depends on which part of the victim's body the weapon is aimed. On the other hand, hitting someone on the arm with a wooden club with the intent to cause serious harm but not to inflict any life-threatening injury was seen as less serious. If the victim in these latter cases subsequently died, perhaps in the course of treatment for a broken arm, participants frequently suggested that this should not be viewed as murder because death was "accidental", since the offender's act carried no apparent risk of death.

A.8 Similar unanimity was expressed regarding homicides where the victim was tortured in some way before being killed,^[5] and in cases where the offender demonstrated a "total disregard for human life". An example of the latter is the Russian-roulette player, in which the killer fires a gun containing one bullet rather than a full complement of six. The killer acts in order to expose his victim to the risk of death.^[6] As indicated earlier, the first group associated the killing of more than one individual – either through serial killings or by a single act -with the worst homicides. Three of the other groups took the same view, but the fourth group seemed to regard them as not especially serious.

A.9 The reasons given for either favouring or rejecting mandatory sentencing were foreseeable. Those who supported it pointed to the sheer gravity of the wrongdoing, and they were keen to ensure that judges should not be allowed to pass an unduly lenient sentence. Indeed, in all groups there was clear evidence of concern that judges can "have an off day", or that some simply are "too soft" or "out of touch" with the views and standards of ordinary people. Several participants talked generally about inconsistencies in sentencing practice, and suggested that this might be alleviated by mandatory sentencing. Requiring the judge to impose the death penalty or "natural life imprisonment" would provide better protection for the public and effect greater deterrence. Conversely, those against mandatory sentencing felt that the circumstances of a homicide would, as in other forms of criminality, vary from case to case and thus felt the judge should be able to reflect these variations when determining the appropriate punishment. They did not feel that judges are so out of touch with contemporary society; rather judges are aware of the relevant principles of sentencing law and policy, and of the facts of the case and are therefore in a good position to decide the most appropriate sentence. Any possible errors – especially unduly light sentences – can and should be corrected by appeal.

A.10 Finally on this point, in all group discussions, regardless of their personal preference for or against mandatory sentencing, participants came to appreciate the importance of the definition of offences, of ensuring that offenders are convicted of the "right offences" – not so much for reasons of fair labelling, but more to ensure that they received the appropriate sentence. In this respect, they appeared to be echoing the comments of senior judges that the current definitions – for example, of murder – are not as accurate as they should be.^[7] Cases resulting in murder convictions vary so considerably in moral culpability that they should not all be classified in the same way.

Specific sentences

A.11 Of the 56 participants, 25 (44.64%) expressed their clear support for the availability of the death penalty in certain kinds of homicides, and a further three were undecided on what should be the maximum or mandatory sentence. (Not all of these 25 favoured mandatory sentencing.) Those who were against capital punishment felt that offenders convicted of the worst homicides should be sentenced to life imprisonment, by which they meant that they should serve the rest of their natural lives in prison. Indeed, it is important to record here the very strong criticism by participants of the current system whereby life sentence prisoners are likely to be released into the community on licence after serving the first part of the sentence in custody. Participants thought that this did not merit the description "life imprisonment".

Other ways of controlling judicial sentencing decisions

A.12 Regardless of the specific opinions on mandatory sentencing, there was a good deal of support for controlling the extent to which judges should be able to determine the punishment through the use of upper and lower limits. Although they were not encouraged to be clear about how many separate offences they would like the law to recognise and how they should be defined, several participants thought that there should be a range of sentences available for the different categories of homicide, so that there was a much greater likelihood that the "punishment would fit the crime." What was being advocated here seemed to be broadly analogous to the guidelines set down in Schedule 21 of the Criminal Justice Act 2003 which indicate the length of time to be spent in prison before first release on licence according to specific offence characteristics. Allied to this, there was considerable support for the idea of having the sentence determined by a panel of judges, so that if one judge had an "off day" or wanted to be unduly lenient, the other judges could redress this and thereby ensure a more justifiable penalty.

The perennial problem of mercy killing

A.13 In both meetings the groups considered the familiar question of how the criminal justice system should regard and respond to mercy killing. It was invariably accepted that provided there is clear evidence of the victim's desire to die, such cases are amongst the least serious of homicides. Where there is no such evidence, opinions were less clear, and that meant that where the victim is unable to indicate a desire to die participants found it more difficult to express a view on the gravity of the killing, even assuming the killer was motivated solely by compassion. In general, they thought that the homicide would be more serious, though not necessarily amongst the most serious. Participants said it would obviously be vital to know whether the case was a "genuine mercy killing" – had the victim truly and freely wanted to die, and was the killer's motive a "good" one? It was this that concentrated participants' minds most of all. Virtually all suggested that there ought to be some form of official enquiry into what had happened, and that a formal prosecution or police investigation might serve this purpose. Where the case was one of genuine mercy killing, the most punitive suggestion was for a short period of imprisonment, and many participants felt that a community-based disposal, with the emphasis on counselling for the killer, would be appropriate.

MISCELLANEOUS FURTHER ISSUES

A.14 In addition to the above issues which were addressed by all groups, there were various other points which naturally came out of the discussions by some participants in one or two groups but which were not raised in others.

(1) Motorists who know they have been drinking and ought not to drive but nonetheless do so and kill another road-user – for example, by driving too fast or failing to negotiate a bend safely – were condemned by several participants, (literally) one or two of whom advocated mandatory sentencing by, at least as part of the punishment, automatic disqualification from driving indefinitely.

(2) In an effort to ensure that "the punishment fits the crime", some participants suggested that the jury should explain the factual basis of their verdict to the judge so that there could be no doubt in his mind as to the nature of the homicide of which the defendant had been convicted.

(3) Some participants thought it would be helpful to have a lawyer retire with the jury so that (s)he could (a) advise or guide them on any legal issues which they did not understand and (b) might (again) then make the judge aware of the basis of the jury's verdict.

(4) The first group meetings were held at about the time when it was announced through the media that consideration might be given to the deceased's next of kin – the secondary victims – having the opportunity to be formally involved in the legal process. Several participants commented on this and it was recognised that such involvement might, theoretically at least, take two forms. Secondary victims might be allowed to suggest an appropriate sentence, but this possibility received a mixed and usually critical response with many participants fearing that the victims would be unable to be sufficiently objective about something so serious and comparatively recent. Alternatively, secondary victims could talk about the impact of the homicide on them, about their pain and suffering, and thereby provide the court with more (precise) evidence of the harm caused by the offence. This possibility was treated less critically by participants but it is unclear just how far they would positively support its introduction.

(5) Whilst many participants, especially in the latter stages of the discussions, spoke favourably about minimum sentencing and sentencing ranges for different categories of homicides, there was also a recognition of the potential injustice that these could bring about because individual cases might fit a general description but contain an additional factor which required them to be treated differently when sentence was passed.

(6) Participants' primary concerns when passing sentence on convicted killers were to give the defendant his just deserts and to protect the public. In relation to those who commit homicides which are not amongst the most serious but which merit a fixed term of imprisonment, some desire was expressed for the time in custody to be used constructively, to

rehabilitate the offender and help him re-establish himself in the community.

(7) At the same time, a number of those who rejected capital punishment felt that execution would be "too easy" for the killer who should, instead, be made to suffer through spending the rest of his life in prison.

IMPLICATIONS

A.15 Obviously, this survey is based on a small number of participants, but in view of the methodology employed the views expressed can be regarded as providing further valuable evidence of how the public would like the criminal justice system to respond to criminal homicides.

A.16 The principal implications of these discussions might be presented thus:

(1) The survey appears to confirm the findings of previous studies that members of the public share the law's view that there are important variations in the seriousness of homicides – that any unlawful homicide is a serious matter, but some are worse than others – and that these variations should be reflected in the law.^[8]

(2) This research supports the findings of another recent public survey that there does not appear to be a clear swell of opinion in favour of mandatory sentencing for what are regarded as the most serious homicides.^[9]

(3) There are variations in the kinds of homicides in which mandatory sentences are favoured.

(4) There seems to be clearer support for adopting alternative means of controlling the use of judicial discretion in sentencing convicted killers, namely by substituting judicial panels for single judges, and/or by formally recognising sentencing ranges for different categories of homicides – though there would have to be a degree of flexibility to avoid injustice.

METHODOLOGY

A.17 The survey is based on the views expressed during meetings of five groups of participants in different parts of England. In an attempt to provide a broad range of backgrounds, men and women who had reached adulthood with a variety of educational experiences and religious affiliations were recruited by PH Research. The participants in the five groups were drawn from the north, south, east, west and centre of England. Each group met – in rooms hired in local hotels^[10] - on two occasions with exactly seven days in between.^[11] The meetings took place between 7.00 and 9.00 pm^[12] and the discussions were tape-recorded. Participants were recompensed for their time and travel expenses.

A.18 The personal characteristics of the participants were as follows:

(1) There were 26 men and 30 women.

(2) Ages ranged from 21 to 68.

(3) Nearly two-thirds (36 out of 56) regarded themselves as Christians; there were six Muslims; three Hindus; one Jew; one agnostic; eight atheists, and one who had "no religion".

(4) There were 16 (28.6%) graduates; 14 (25%) who had no formal education qualifications, and the remainder had GCSEs, A levels, HNCs etc.

A.19 In addition, a short screening questionnaire was used to gather further background data on participants and this revealed that:

(1) Two-thirds (37 of the 56) lived in a town; four lived in a city, and 15 in a village.

(2) Over 80% (45 of the 56) lived with their family; three lived with friends, and the other eight lived alone.

(3) Just over two-thirds (38) had children.

(4) Four participants had themselves been a victim of a crime of violence; 10 said that other members of their family had been victims of violent crime; a relative of one participant had been unlawfully killed; two participants had had property offences committed against them, and two more were related to victims of property crimes.

(5) Seven had served on a jury in a criminal trial.

A.20 The first meetings were devoted to encouraging participants to think about the different kinds of circumstances in which homicides take place and to consider how and why they regarded the seriousness of the different cases. It was stressed that they should put aside whatever they knew or thought about the current law, and concentrate instead on their own personal opinions with a view to formulating ideas about how they would like the criminal justice system to respond to unlawful homicides. One of the devices used in this process was to present participants with three brief scenarios and invite them to (1) consider how serious they regarded them, and why; and (2) to change individual factors in the scenarios in order to see how, if at all, that affected their view of their seriousness.

A.21 The scenarios are as follows:

(1) John found out that David, aged 11, had been bullying his daughter Julie (also 11). John kidnapped David and locked him in a remote disused shed. After torturing him there for three days, John shot David dead.

(2) Two men, Mike and Peter, were members of a local amateur football team. Their team was doing very poorly and they were arguing about which of them had been more responsible for the team's poor performances. As the argument became more heated, Mike punched Peter in the face, causing him to lose his balance and fall over. As he did so, Peter hit his head against a brick wall, fracturing his skull, and he later died from his injuries.

(3) Sue was suffering from a terminal illness and was constantly in considerable pain. The doctors were unanimous that there was no hope of recovery, and they were able to give her very little pain relief. For several months she had regularly begged her husband Graham to "put her out of her misery". He had always refused to do so and had tried to comfort her. Eventually though, he gave in and gave her a fatal overdose of tablets.

A.22 In view of the nature and purpose of the survey the discussions were loosely structured, to permit time and opportunity for the discussions to follow individual responses and consider the various arguments and counter-arguments. At the same time the discussions were conducted so that each group addressed the same broad set of issues. Each group was invited to think in particular about what they regarded as the most serious homicides and how the presence or absence of individual factors might affect their views. In the light of previous surveys on the subject, regardless of their immediate responses, participants were invited to comment on the impact of the following potentially aggravating features:-the significance of premeditation (in contrast to spur of the moment killings); killing multiple victims (either by a series of attacks or by a single act); killing vulnerable victims such as children, the elderly or the handicapped, or alternatively those in vulnerable occupations such as police officers); killing combined with torture of victims; and finally cases where the killer shows a contemptible disregard for life, such as the Russian-roulette killer. The potentially mitigating features addressed were: a less morally culpable state of mind on the part of the killer, ranging from no more than an awareness of the risk of doing minor harm to an intent to seriously (but not fatally) injure; and the presence of what might be regarded as a "good motive" such as mercy or self-defence.

A.23 At the end of the first meeting participants were told that they were free to discuss the issues between then and convening for the second meeting, if they chose to do so.

A.24 The second meeting was devoted to discussing how participants thought that convicted killers ought to be sentenced, and especially whether there were any kinds of homicides in which they felt that the judge should have no choice but be required by law to impose a specific penalty. Whichever view they adopted, they were also encouraged to indicate the reasons for their opinions. The loose structure of the discussions enabled the participants to examine other issues apart from the stark choice between mandatory versus discretionary sentencing.

APPENDIX B SUMMARY OF FINDINGS OF SURVEY OF CROWN PROSECUTORS

B.1 To assist in its review of the law of murder, the Law Commission sought the views of prosecutors on potential reforms to the law of murder in relation to the definition of murder, diminished responsibility and charging decisions. To this end, the Law Commission sent a questionnaire to each branch of the Crown Prosecution Service (CPS). This appendix provides a quantitative and qualitative analysis of the CPS responses.

METHODOLOGY

B.2 The Commission sent the Chief Crown Prosecutor of each branch of the CPS (42 in all) a questionnaire titled "Murder Project: Prosecution Decisions in Homicide Cases". The questionnaire consisted of three sections, which covered the definition of murder, diminished responsibility and charging decisions. The Commission received 27 responses. Of those responses, some Chief Crown Prosecutors answered the questionnaire alone, whereas other CPS branches appeared to have considered and answered the questionnaire as a group.

B.3 Each section of the questionnaire provided background information on each issue and reasons for reform. Each section then asked specific questions, seeking a mixture of yes and no answers, rankings and general comments.

B.4 The following sections of this appendix set out a quantitative and qualitative analysis of the responses. The quantitative analysis was complicated by the following factors:

(1) The responses did not always provide answers to all the questions asked.

(2) The questionnaire did not always ask for a yes or no response to particular questions. Consequently, determining whether a response was for or against a proposal sometimes required an assessment of the general comments provided.

(3) If the comments did not clearly favour or reject a proposal, the response was generally classed as 'mixed' in the quantitative data.

(4) Many responses qualified their approval or disapproval of a particular proposal. Therefore, although a response may have been classed as for or against a proposal for the purposes of quantitative analysis, it is necessary to read the quantitative results in the context of the qualitative analysis.

B.5 Given these complications and qualifications, the quantitative data should be treated as indicative only, rather than as a precise assessment of the prosecutors' views.

B.6 A copy of the questionnaire sent to prosecutors is included at the end of this Appendix.

QUANTITATIVE ANALYSIS

B.7 The three tables below provide a quantitative analysis of CPS responses to the Commission's questionnaire. The tables set out each question asked and the corresponding percentages of those responses which answered yes or no, for or against etc.

B.8 In each table, 'n' refers to the number of responses to a particular question. In most cases 'n' includes only those responses which provided an answer to that particular question (as noted above, not all prosecutors answered all questions). Any variation to or qualification of this approach is noted in the table.

B.9 When assessing the quantitative data, it is important to read it in the context of the information included in the questionnaire and the qualitative analysis. For example, when considering alternative definitions to murder, prosecutors were not informed that on option the Commission was considering was the creation of an intermediate grade of homicide ("second degree murder") which would include cases of unlawful killing where the defendant had intended to effect grievous (serious) bodily harm (the 'GBH rule'). Consequently, prosecutors may have assumed that the GBH rule might be abolished or included in the manslaughter category. The possibility of including the GBH rule in "second degree murder" was only raised in the final question on the questionnaire. Therefore, any assessment of the responses needs to take this factor into account.

B.10 Finally, to reiterate, the percentages should be treated as indicative only, not as absolute.

The Definition of Murder	The Definition of Murder	The Definition of Murder	The Definition of Murder	The Definition of Murder
Alternatives: Comments on each definition of murder.	Alternatives: Comments on each definition of murder.	Alternatives: Comments on each definition of murder.	Alternatives: Comments on each definition of murder.	Alternatives: Comments on each definition of murder.
	For	Against	Mixed	Mixed
1. D is guilty of murder if he intended to kill, or if he intended to cause what he realised at the time might be a life-threatening injury. (n=25)	4%	80%	16%	16%
2. D is guilty of murder if he intended to kill, or if he intended to cause harm the jury regards as inherently life-threatening at the time it was done. (n=25)	12%	64%	24%	24%
3. D is guilty of murder if he intends to kill, or is recklessly indifferent at the time to the causing of death. (n=25)	16%	60%	24%	24%
Scenario: Based on the scenario outlined, which of the above definitions is likely to focus on what are the significant issues?	Scenario: Based on the scenario outlined, which of the above definitions is likely to focus on what are the significant issues?	Scenario: Based on the scenario outlined, which of the above definitions is likely to focus on what are the significant issues?	Scenario: Based on the scenario outlined, which of the above definitions is likely to focus on what are the significant issues?	Scenario: Based on the scenario outlined, which of the above definitions is likely to focus on what are the significant issues?
	Most likely	Less likely	Least likely	Equally likely
Current law (n=21)	52%	19%	5%	24%
Alternative 1 (n=20)	15%	30%	45%	10%
Alternative 2 (n=20)	25%	40%	10%	25%
Alternative 3 (n=23)	30%	26%	30%	13%

Diminished Responsibility
Proposals/Questions	Yes	No	Mixed
1. Retain diminished responsibility (DR) as a partial defence to murder. A successful plea of DR or provocation would reduce the crime to "second degree murder" not to voluntary manslaughter. 'Manslaughter' would be reserved for 'involuntary' manslaughter. (n=23)	83%	13%	4%
2. The Commission is concerned that the ability of D to run provocation and DR defences together may lead to unwarranted acquittals because as a result of the decision in Smith [2001] 1 AC 146, separation of the issues relevant to each defence at trial is difficult. The PDM Report recommended reforming provocation to make separation easier. An alternative would be to create a separate offence of 'DR killing', leaving provocation alone as a partial defence to murder. (a) Can you see the advantages of this approach? (n=22) (b) Can you see the disadvantages of this approach? (n=22)of this approach? (n=22)
	50%	50%
	50%	50%
3. The Commission is concerned about the adequacy of the procedure for assessing whether D's responsibility is diminished. In the French system, the judge can commission an examination of D by court-appointed medical experts whose written reports are disclosed to both sides well before trial. (a) Are the procedures for assessing D's mental state prior to and/or at the trial adequate? (n=23) (b) In what way, as through adoption/variation of the French system, could the procedures be improved? (n=15)* * The percentages given are of those responses that indicated approval, disapproval or a mixed response regarding the French system or a system like it. It does not include responses which did not directly comment on the French system or a system like it.
	30%	57%	13%
13%	73%	7%	20%

Charging Decisions
Questions	Comments
1. In what circumstances is it right to accept a plea of guilty to manslaughter?** (n=27) ** The percentages given are not mutually exclusive.	Out of 27 responses, the following number thought that it may be appropriate to accept a plea of manslaughter: • 15 (56%) in DR cases. • 11 (41%) in provocation cases. • 4 (15%) involuntary manslaughter (no intent) cases. • 2 (7%) excessive self defence cases. Almost all responses emphasised that accepting such a plea only occurred rarely and in the clearest cases.
2. How would it affect charging decisions if there was an intermediate grade of homicide – say "second degree murder" – between murder and manslaughter, with a discretionary life sentence, that encompassed killings where the D's intent was to do only serious harm?*** (n=18) *** The percentages given are of those responses that gave a better/worse/no change response. It does not include responses which did not comment directly on whether such a change would be an improvement, be worse or have little impact.	• 3 (17%) thought this would be better at the charge stage. • 5 (28%) thought that this would be worse at the charge stage. • 10 (56%) thought that this would have little impact at the charge stage. Many responses were mixed. Many thought that such a change would lead to many pleas of "second degree murder". Views were split as to whether this was a good thing.

QUALITATIVE ANALYSIS

B.11 The following sections analyse the CPS responses to each question, drawing out themes, qualifications and suggestions. The qualitative analysis aims to assist the interpretation of the quantitative data and to provide a broader understanding of the responses.

The Definition of Murder

B.12 The first section of the questionnaire focused on the definition of murder. The preface to the section set out the current definition of murder.^[13] It then explained that the Law Commission sought prosecutors' views on three alternatives to the current law (in particular, modification of the GBH rule) which were designed to reach a fairer outcome, without making the process of prosecution unduly complex.

B.13 In assessing these responses, it should be noted that the questionnaire participants were not told at this point that the Commission was considering whether to provisionally propose that the GBH rule become "second degree murder". This may have influenced the prosecutors' responses.

First alternative

B.14 The first alternative definition of murder provides that:

D is guilty of murder if he intended to kill, or if he intended to cause what he realised at the time might be a life-threatening injury.

B.15 A large majority of responses (80%) were against this alternative. Only one prosecutor was in favour of it. The remainder (16%) were mixed.

B.16 The reasons given for objecting to this alternative are as follows:

(1) Many prosecutors thought that the subjective requirement that the defendant realised at the time that the injury he intended to cause might be life-threatening was problematic on two grounds:

(a) First, whether a defendant realised that the harm he intended to cause might be life-threatening would depend on the defendant's knowledge of medical science, and so would vary depending on factors such as age and low intelligence. It may also be affected by drugs or alcohol.

(b) Secondly, proving that the defendant realised that the harm he intended to cause might be life-threatening would be very difficult. Whether something is life-threatening is more specific than proving a realisation that it may cause an unspecified serious injury. Therefore, it would be much harder to infer this realisation in the face of a denial by the defendant.

As a consequence of 1(a) and 1(b), many prosecutors were concerned that this alternative would lead to too many acquittals because the threshold for intent was too high. A high number of acquittals would undermine public confidence in the judicial system and be unjust to victims. A number of prosecutors thought this alternative moved too far in favour of the defendant at the expense of victims and their families.

(2) Many prosecutors questioned whether a jury is qualified to determine whether an injury was life-threatening. One asked whether it would be enough proof that the victim died; whether other considerations like haemophilia would be accounted for; and whether expert evidence would be necessary on this issue. Such questions were considered to be likely to lead to significant legal argument, especially if expert evidence was required. One response pointed out that the jury would not only have to determine the degree of injury (as they already do in relation to serious injury), but also the likely effect of such of injury (ie whether it was life-threatening). (Arguably, this objection is stronger in relation to the second alternative in which the jury is asked to decide whether the injury was in fact life-threatening, rather than whether the defendant realised it might be life-threatening.)

(3) One prosecutor thought this alternative would overcomplicate the role of juries by requiring them to decide whether the defendant intended to kill, to cause life-threatening injury, to cause serious bodily harm or something else.

(4) One prosecutor argued that it was difficult to establish the meaning of 'might'. For example, did the likelihood of death have to be 'very likely' or 'pretty likely'?

B.17 Overall, a typical negative response to this alternative was:

Our view is that this would narrow the scope of murder and rather than simplifying it would present further difficulty in that argument will ensue over the meaning of "life-threatening". Also one would have to consider the question of the level of understanding that a D had regarding the consequence of injury. For these reasons we think that this is not a viable option in the sense of leading to a fairer outcome.

B.18 A small number of responses were mixed (16%). These mixed responses made their acceptance of this definition conditional upon modification to deal with some of the problems noted above.

B.19 Only one prosecutor (4%) thought that this alternative was a good idea without modification. This prosecutor argued that such a serious offence carrying a mandatory life sentence should focus on intent. The prosecutor noted that:

Some people might be concerned that the subjectivity that's built into this alternative might result in unjust acquittals. I would not agree; the trial process is well equipped to deal with spurious "I didn't think that my stabbing (or stamping/shooting/kicking/burning) might have put X's life in danger" defences!

B.20 In summary, this was the least popular alternative by a considerable margin.

Second Alternative

B.21 The second alternative definition of murder provides that:

D is guilty of murder if he intended to kill, or if he intended to cause harm the jury regards as inherently life-threatening at the time it was done.

B.22 A large majority of responses (64%) were against this alternative. However, there were more in favour of it (albeit most with some caution) (12%) than were in favour of the first alternative. There was also a greater proportion of mixed responses (24%).

B.23 Prosecutors raised the following objections and concerns regarding the second alternative:

(1) The most recurrent concern was that a jury is not qualified to make an assessment of whether an injury might be 'life-threatening'. This would necessitate the use of expert evidence, which would lead to added complexity and significant legal argument. As one response put it: "This is better [than the first alternative] but open to expert evidence, statistics on survival rates, quality of care, etc. Do not like it."

(2) Many responses criticised the use of the word 'inherently' as meaningless. One suggested that the fact that the victim had died would surely lead the jury to infer that the injury was life-threatening. This would be potentially unfair in cases of minor injuries that result in death, which are currently covered by manslaughter. In contrast, others thought that the narrowness of the test would lead to unjust acquittals and thus a lack of justice for the victim.

(3) One response considered that the mixture of the defendant's intention and the jury's objective view would lead to unnecessary complexity similar to that which arises in provocation cases. This would lead to injustice.

(4) Other responses criticised the test as too subjective in terms of varying views among jurors as to what injuries are inherently life-threatening. One response thought this might lead to more "hung" juries. Another queried why what the jury regards as life-threatening was relevant rather than the view of the 'reasonable person'.

(5) Finally, the response that favoured the first alternative was uneasy about the objective element in this alternative on the basis that such a serious offence with a mandatory life sentence should focus on intent. This response considered that the second alternative might lead to injustice in cases in which the defendant lacks mental capacity but falls short of diminished responsibility. Another response also criticised the use of an objective test, stating:

The difficulty here is that this approach imparts a wholly objective assessment (akin to the 'reasonable man') upon the severity of the harm, regardless of whether the defendant foresaw the likely consequences of his actions. The Courts have shrunk from this construct since DPP v Smith (1961).

This response further noted that the subjective approach was also supported in relation to recklessness in R v G&R (2003) HL.

B.24 In the mixed responses, a number of prosecutors approved of the use of an objective standard in this alternative, rather than the subjective approach in the first alternative. However, many still thought that the use of 'life-threatening' would lead to extra argument.

B.25 Three responses (12%) had a generally favourable view of this option. One thought it would "probably be easier to prove and arguably fairer" (though it was unclear whether this was in comparison to the current law or to the first alternative). Another thought that juries would find this approach easier to apply and would lead to more appropriate verdicts.

B.26 In summary, the second alternative was more popular than the first alternative. However, the majority of responses were against this as an alternative to the current law. Many believed that the second alternative would lead to greater complexity (particularly in relation to the question of what was 'life-threatening') and was an inappropriate question for a jury to answer.

Third Alternative

B.27 The third alternative definition of murder provides that:

D is guilty of murder if he intends to kill, or is recklessly indifferent at the time to the causing of death.

B.28 This was the most popular of the three alternatives, though support was still low (16%). The majority of responses were against it (60%), and a significant proportion gave a mixed response (24%).

B.29 The most frequent criticism made by prosecutors of this alternative was that it lacks precision. A common sentiment was: "The concept of "recklessness" has bedevilled the criminal law. … We cannot seriously consider using this." A number of responses argued that it would encroach on the law of involuntary manslaughter and so would lead to confusion. One response queried what "at the time" meant. Many prosecutors thought that it would be difficult to direct the jury adequately on the precise terms of the offence. This difficulty would result in a range of outcomes. In contrast, another response stated: "this would be the easiest definition for a jury to understand and for a judge to direct upon." Another thought this alternative would assist the jury.

B.30 A number of responses considered that the third alternative broadened the current scope of murder. Some saw this as a positive, others as a negative. One prosecutor argued that the third alternative would be more likely to capture cases where a defendant "takes a highly culpable risk, but without foreseeing death or life-threatening injury as virtually certain." This prosecutor believed that the third alternative would capture arsonists who believe the premises to be unoccupied or the terrorist who issues a warning. Others criticised the broadening of murder on the basis that it would capture those who we would not necessarily wish to classify as murderers.

B.31 A number of responses thought that an offence of such seriousness should not be based on recklessness but rather on intent, especially given the Government's commitment to the mandatory life sentence. One response argued that although a conviction might be easier, it would be less fair. Further, the mandatory life sentence did not sufficiently differentiate between different offenders' culpability. One thought that if murder was extended to include recklessness, then the mandatory life sentence should be replaced by a maximum sentence.

B.32 In contrast, a number of responses believed that the third alternative narrowed the current test and was too restrictive. One response thought that it would lead to more unjust acquittals as the defendant could simply deny that he had turned his mind to the possibility of causing death. Similarly, another thought that the definition was too restrictive and that the GBH rule should be retained instead.

B.33 Of the mixed responses, most suggested some sort of modification to or clarification of this alternative. For example, one stated that "indifferent" should require that the defendant appreciated that more than merely trivial harm may result which he disregarded, rather than a total failure by the defendant to consider the possibility of any harm resulting. One response thought that this alternative would be closer to the public's perception of what ought to constitute murder.

B.34 A small number of responses thought without qualification that this was a good approach. One response thought that this alternative would bring the law of England and Wales into line with the European states. Another response stated:

This is the most favoured approach, as it provides "just deserts" for the D who acts totally unreasonably. Why should someone who could not care less about the consequences and does not bother to think about them escape prosecution for murder? This approach provides justice for V in these circumstances.

B.35 Overall, prosecutors were generally against this approach, though many preferred it to the other alternatives. The comments were mixed and some were directly contradictory. The main criticism focused on the imprecision of the notion of recklessness and the confusion that was likely to result.

Scenario

B.36 To help prosecutors assess the three alternative definitions of murder, the questionnaire provided a scenario involving a defendant who kills his wife after he loses his temper in an argument during which he believes his wife confessed to adultery. The wife died of complications after the defendant stabbed her, the knife penetrating 3 inches into her stomach. The defendant admits to the killing, but refuses to plead provocation. Instead he claims he lacked the mental element for murder.

B.37 The questionnaire asked prosecutors which legal definition of murder (including the current law) is most/more/less/least likely to focus on what are, in the respondent's view, the significant issues (answer: 'most likely/'less likely'/'least likely'/'equally likely').

B.38 In answering this question, a number of respondents only ranked some but not all of the alternatives. Some respondents created their own categories or placed two alternatives in the same category. Some respondents ignored this question entirely, or argued that it was the wrong question to ask. Another respondent changed the scenario and gave an alternative ranking based on the new scenario. Not all respondents who ranked the alternatives gave any further comments as to their reasons. A number of respondents reiterated the reasons they gave for and against the alternatives noted above. These reasons have not been repeated below. The figures noted below (and in the table) should be taken as indicative only, given the variety of responses outside the set categories.

B.39 The most popular definition of murder was the current law (ranked 'most likely': 55%). The least popular definition was the first alternative (ranked 'least likely': 45%, and 'less likely': 30%. A minority ranked it 'most likely': 15%). The second and third alternatives fell in the middle in terms of popularity. One prosecutor felt so strongly against all three alternatives, the response stated: "I would not wish to validate these options by ranking them."

B.40 Of the majority who favoured the current law, the general feeling was that the current definition worked well and that the GBH rule should be retained. (As noted above, the questionnaire did not at this point raise the possibility of a revised GBH rule being moved to "second degree murder").

B.41 A number of responses noted that while it would be possible to prove intent to cause serious harm in this scenario, it would be more difficult to prove intent to cause life-threatening injury in the first and second alternatives. It would also be difficult to direct the jury on life-threatening harm. One response, which favoured the second alternative, suggested that an extra category of "second degree murder" should be created if the defendant only intended serious harm.

B.42 A number of responses thought that the defendant could be convicted under the third alternative, though some questioned whether murder should extend to recklessness at all. Some argued that the third alternative would be complex for the jury. One stated: "3^rd alternative will tie the jury in knots."

B.43 The third alternative attracted the most extreme responses, either for or against it. Prosecutors' reluctance to include recklessness in the definition of murder was reflected in the high number who ranked this alternative in the 'least likely' category (30%). However, there was an equally high number (30%) who thought that recklessness was a good option, ranking it as 'most likely'.

B.44 The second alternative attracted less extreme responses, with 40% categorising it as 'less likely'. The number of responses who ranked it as 'least likely' (10%) was considerably lower than in respect of the third alternative. However, the number who ranked it as 'most likely' (25%) was also lower than in respect of the third alternative.

B.45 A number of responses took issue with the question itself. One response argued that each of the alternatives would be equally likely to focus on the significant issues. What would differ is the likelihood of conviction. Another argued that 'the issue' would differ depending on which definition was applied.

B.46 Many responses thought that the likelihood of acquittal (or conviction for manslaughter) in the scenario was higher if any of the three alternatives were applied compared to the current law. Most disapproved of this outcome. One thought this would send the wrong message to violent offenders to tailor their defence to secure acquittals.

B.47 Overall, the current definition of murder was preferred by the majority on the basis that it was easier to apply and was likely to result in just convictions. The response to the three alternatives was mixed. However, the first alternative was the least favoured.

Diminished Responsibility

B.48 The second part of the questionnaire explained that the Law Commission would be considering the partial defence of diminished responsibility in some detail. Accordingly, the questionnaire contained three questions on diminished responsibility.

Question 1

B.49 The questionnaire noted that the Commission was considering the retention of the partial defence of diminished responsibility in its current form (as a partial defence to murder). However, rather than diminished responsibility and provocation reducing the offence to voluntary manslaughter, it was likely that the Commission would suggest that these partial defences reduce murder to "second degree murder". The respondents were asked if they supported this proposal. The questionnaire provided for a yes or no response, without space for comments. 83% of responses supported the proposal.

B.50 Although the questionnaire did not ask for comments, a number of responses included comments. One response did not support the introduction of "second degree murder", but instead suggested that involuntary manslaughter become culpable homicide. Another rejected the Law Commission's proposal on the basis that it would be a change in terminology only, unless the penalty was different to manslaughter. In contrast, a number of responses supported a change in labelling on the basis that it would be more acceptable to victims' families. One response thought the fact that a mandatory life sentence would not apply is less significant for victims' families. This response stated: "Grief will often be fuelled by headlines in the press stating that the offender "got away with murder"." Similarly, another response stated: "The family of the victim always finds it difficult to understand how a 'diminished responsibility killing' becomes manslaughter."

Question 2

B.51 Question 2 was prefaced by the following:

The Commission is concerned that the ability of persons accused of murder to run provocation and diminished responsibility together may lead to unwarranted acquittals, because it is no longer possible adequately to separate the issues relevant to each defence at trial (R v Smith [2001] 1 AC 146). The Commission has already recommended that provocation be reformed to make such separation easier (Law Com 290, Partial Defences to Murder). Another approach would be to create a separate offence of 'diminished responsibility killing', leaving provocation alone as a partial defence to murder.

B.52 The questionnaire participants were asked whether they could see the advantages of this approach (yes or no) and the disadvantages of this approach (yes or no). They were asked for further comments.

B.53 It should be noted that the questionnaire was distributed prior to the Privy Council decision in A-G for Jersey v Holley.^[14] A number of responses argued that Holley sufficiently differentiated provocation from diminished responsibility, so the problem identified by the Law Commission was no longer an issue. Those responses that noted the impact of Holley took this into account when answering yes or no to the questions noted in paragraph B.52. In contrast, many other responses did not refer to Holley, nor did they appear to take it into account when answering these questions. Consequently, it is difficult to draw any significant conclusions from the answers. This i9s complicated by the fact that a number of responses made no further comments so did not indicate whether they had taken account of Holley when answering yes or no. Other responses did not answer the questions at all.

B.54 Of those who did respond:

(1) 50% could see the advantages of introducing a separate offence of diminished responsibility killing.

(2) 50 % could not see the advantages of introducing a separate offence of diminished responsibility killing.

(3) 50% could see the disadvantages of introducing a separate offence of diminished responsibility killing.

(4) 50% could not see the disadvantages of introducing a separate offence of diminished responsibility killing.

B.55 Clearly, the prosecutors were evenly split over the question of whether a separate offence of diminished responsibility killing should be introduced. However, it is difficult to draw any strong conclusions from these figures, particularly in light of the impact of Holley on the responses and on the law itself.

B.56 The responses that provided further comments identified the following advantages:

(1) Charging defendants with two separate offences may help the jury to separate and understand the issues better, and so lead to more appropriate verdicts.

(2) The likelihood of an acquittal may be reduced if the defendant is charged with two separate offences. It may also eliminate acquittals arising from confusion on the part of the jury.

(3) A clear verdict would result because the jury would identify whether it was diminished responsibility killing, murder or provocation. <

(4) A separate offence would establish "as early as possible if "mental health" is a live issue."

(5) The separate offence would help focus the court on the relevant issues and make judicial directions more straightforward.

B.57 The responses that provided further comments identified the following disadvantages:

(1) The new offence would not remove the problem of juries that cannot decide between provocation and diminished responsibility. This will lead to acquittals as it does now.

(2) If provocation and diminished responsibility are retained, there will still be a problem of separating them. The issues raised by provocation and diminished responsibility "are often inextricably linked and sometimes opposites sides of the same coin."

(3) It was likely that both issues would still be heard together (this would depend on the judge's direction).

(4) Provocation and diminished responsibility are often pleaded in tandem on the basis of the same psychiatric evidence. This gives the jury the opportunity for a compassionate verdict.

(5) The new offence would only complicate matters further and confuse juries. A clearer definition of each defence is preferable to address the overlap.

(6) The new offence would have little impact. Most defendants would be charged with murder as well, unless they offered an acceptable plea.

(7) It is likely that the new offence would be perceived as downgrading a deliberate killing (albeit by a person not in full command of their faculties at the time).

(8) Given that the new offence may be perceived as a lesser offence, it would fall into disrepute if the CPS decision making process about what to charge were not regarded as robust. Therefore the decision should be restricted to a limited number of authorised prosecutors.

(9) The old law of joint defences did not result in unwarranted acquittals. Changing the name will not address any perceived problem.

(10) Provocation should not be an offence to murder.

B.58 The advantages and disadvantages identified by prosecutors arguably do not point in a particular direction. However, as noted above, this issue is less pressing since the decision in Holley.

Question 3

B.59 Question 3 was prefaced by the following statement:

The Commission has concerns about the adequacy of the procedure for assessing whether D's responsibility is diminished. Under French criminal procedure, if there are questions about D's mental state at the usual preliminary examination of the issues by the judge, the latter can commission an examination of D by court appointed experts whose written reports are disclosed to both sides well before the trial. Both sides are free to ask for a further report on D's mental state, and all the experts give evidence at trial.

B.60 The questionnaire participants were asked two questions. The first question asked:

Are the procedures for assessing D's mental state prior to and/or at trial adequate?

B.61 The majority of responses thought that the current procedures were not adequate (57%). However, a large minority thought the current procedures were adequate (30%). A minority gave a mixed response (13%).

B.62 Most responses to the first question only gave a yes or no answer without further comment. However, of those who gave no further comment on the first question, many gave further comments on the second question which are also applicable to the first question. Consequently, the comments on the first question are considered in our analysis of the second question.

B.63 The second question asked:

In what way, through the adoption/variation of the French system, could the procedures be improved?

B.64 The questionnaire participants were not asked to say yes or no to the French system. However, many responses indicated their approval or disapproval of the French system or of a system like the French system. Of those who directly commented on the French system or on a system like it, a large majority (73%) indicated approval for the French system or a system like it. A small minority (7%) indicated disapproval. The remainder (20%) gave a mixed response.

B.65 A key concern of those who did not think the current procedures were adequate was that conflicting expert reports on complex issues placed the jury in a difficult position. Further, it was thought that the adversarial process of using separate defence and prosecution expert witnesses was more likely to lead to conflict. Withdrawing the issue from the prosecution and defence was thought by those in favour of change to be more likely to result in consensus.

B.66 Another key concern among those who thought that current procedures were not adequate was the fact that the defence was entitled to pick and choose which reports to disclose, thereby avoiding disclosure of reports that were unfavourable to the defence.

B.67 A number of responses noted that problems arise when a mental health issue is only drawn to the prosecution's attention "at some remove from the arrest and occasionally just before trial." This puts the prosecution at a disadvantage and can result in delays. It was thought that if a court requested a report at an early stage of a homicide case, "it is possible that any mental health issues would be crystallised and even resolved at an earlier stage in the proceedings." Therefore any process which ensured that reports were obtained and disclosed well before the trial would be welcomed.

B.68 Many responses suggested possible solutions to the problems associated with the current procedure, including the following:

(1) The judge could make a preliminary ruling on the defendant's mental state, thus withdrawing the issue from the jury.

(2) The defence and prosecution's expert witnesses could meet before trial to discuss their findings.

(3) Defence and prosecution experts should be compelled to disclose their reports.

(4) Experts should be given more time to adequately assess defendants than they are given under the current system. ("A system could perhaps be adopted whereby the court-appointed expert has far more time to continually observe and assess the defendant, better still in a hospital environment rather than prison.")

(5) Existing provisions concerning disclosure should be brought into force, specifically sections 6C (notice of intention to call witnesses) and 6D (notification of names of experts instructed by the accused) of the Criminal Procedure and Investigation Act 1996.

A number of responses thought that a more transparent system, such as reports requested the court, would inspire greater public confidence in the system and would appear fairer to victims' families.

B.69 A number of responses explicitly supported the French approach. One response stated:

I would suggest adapting something approaching the French system – which is essentially an "objective" enquiry to establish the "truth" of the situation – this would avoid the "hired gun" approach which is an inevitable consequence of the "adversarial" approach – court approved expert with full medical disclosure may also reduce costs!

Another stated:

The French idea appears excellent. Too many trials currently revolve around trying to find an expert who will support one side or the other, rather than reaching a just conclusion as to D's mental state.

One response suggested that the use of expert reports commissioned by the court could extend beyond reports on mental health, to other areas where expert reports are used.

B.70 In contrast, a small minority thought the current system was adequate. One response stated:

Whether medical experts are instructed by the Crown, the defence or the court matters not. They all appear as expert witnesses independent of the party that pays their fees, to give impartial evidence to assist the court.

This directly contradicts the criticisms noted above about perceived bias and nondisclosure of reports in practice.

B.71 A number of responses were mixed. One noted that the current system works well in cases where a mental health issue is identified early on. At this point reports can be commissioned by the prosecution and defence to clarify the issues and assist in making an informed case disposal. However, the system fell down when the mental health issue emerged at a late stage in proceedings.

B.72 Finally, one response noted that in the English system, courts used to request a report on the defendant. However, "[t]he problem with this approach arose when defendants were advised not to see the psychiatrist." Further, according to a number of responses, reports are sought in any event when the defendant is in custody. Therefore the main problems arise when the defendant is on bail.

B.73 Overall, a large majority of prosecutors did not think that the current procedures for assessing a defendant's mental state were adequate. The key problems arose in relation to conflicting reports, perceived bias, non-disclosure and late notice by the defence raising a mental health issue. The majority favoured a system similar to the French system.

Charging Decisions

B.74 The final part of the questionnaire focused on what impact proposed changes may have on charging decisions. The section was prefaced with the following statement:

Since the mid-1960s, whilst the rate of recorded homicides has roughly doubled (c.400-800), receptions into prison for murder have increased four-fold (58-208). Receptions doubled between 1965-1975 (58-103), and doubled again between 1975 and 1991 (103-208). Juries have clearly been convicting of murder in an ever larger proportion of homicide cases. This may suggest that pleas of provocation or diminished responsibility are increasingly less successful before the jury. It may suggest pleas of guilty to manslaughter are accepted less readily in serious cases by the prosecution.

B.75 The questionnaire asked two questions on charging decisions. The questions were expressed in an open-ended way, rather than as a yes or no answer. Consequently, the percentages referred to in the quantitative analysis section are the percentage of respondents who raised a particular matter in response to the question.

Question 1

B.76 The first question asked:

In what circumstances is it right to accept a plea of guilty to manslaughter?

B.77 A small majority of respondents (56%) said that it may be appropriate to accept a plea of guilty to manslaughter in cases of diminished responsibility. However, almost all stated that they would only accept this plea in limited circumstances. For example, this plea would only be accepted in the "clearest cases", when there was "unanimous agreement" between the experts, "where there was no realistic prospect of conviction" or when there was a psychiatric report supported by independent evidence indicating diminished responsibility.

B.78 A large minority of respondents (41%) said that it may be appropriate to accept a plea of guilty to manslaughter in cases of provocation. Again, this was only in limited cases. Respondents prefaced their comments with "rarely", "hardly ever", "where there was no realistic prospect of conviction", "where there was insufficient evidence" to prosecute for murder or in "clear cases" that cannot be rebutted.

B.79 A minority of respondents (15%) said that it may be appropriate to accept a plea of guilty to manslaughter in cases of involuntary manslaughter, but only "occasionally", "where there was no intent" or where there was "no intent on clear evidence".

B.80 Finally, a small minority (7%) said that it may be appropriate to accept a plea of guilty to manslaughter in cases of excessive self-defence. Once again, this acceptance was prefaced with the words "rarely" or "where there is no realistic prospect" of conviction for murder.

B.81 Regarding prosecution practice, one prosecutor stated:

Most pleas to manslaughter (in my experience) [are] accepted on "bird in the hand" basis when it was known that witnesses [were] unreliable and offer and acceptance of plea was made on basis of risk assessment by both sides – whether this is proper or not is perhaps immaterial – it is realistic.

B.82 There was a general reluctance among prosecutors to drop murder to manslaughter on a plea. One response suggested the introduction of an intermediate grade of homicide for diminished responsibility and provocation.

B.83 A number of responses commented on the preface to the question regarding the increase in the rate of homicide and receptions into prison. One response thought the question was ill-founded, stating:

This question ignores the issue that violent crime has changed in the course of the last 40 years and it is naïve to suggest that the changes are because fewer pleas to manslaughter are accepted. It is equally likely that in a more violent society the proportion only of genuine diminished cases has fallen not the actual numbers.

Another response suggested other possible reasons for the increases, stating:

It may be right that the prosecution is accepting pleas to manslaughter less readily in the light of past judicial criticism of the Crown for being too ready to do so. There may well be other reasons – better detection, better forensic evidence, more murders being committed.

B.84 One response noted that there are guidelines in the CPS Manual regarding when to accept a plea of manslaughter.

B.85 Another response noted that since the introduction of sentencing guidelines for the murder tariff, there had been a significant increase in the number of guilty pleas to murder. The respondent couldn't remember any guilty pleas to murder before the change.

B.86 Finally, one prosecutor noted that difficulties could arise in explaining the grounds for a manslaughter plea to a victim's family. This factor may deter prosecutors from accepting a plea of manslaughter. The prosecutor stated:

It has always been my view that diminished responsibility and provocation are jury issues but it is difficult for juries to understand the issues around mental state, and diminished [victims'?] families also find it difficult to understand how their murdered relative was the victim of manslaughter, but I have accepted it when 3 consultants all agreed diminished, when the judge said if it was accepted, life would be the sentence and would mean life and when [name deleted] has spent 2 hours explaining to the victims family the whys and wherefores – however two weeks later the family were on a television programme saying that they didn't understand why he wasn't convicted of murder and why they felt let down by the system. I have never since had such a clear cut case and have not been in the position again – but the victim's families put me in such a position that I would have difficulty accepting such a plea again. … But there must be a clear indicator for change.

B.87 Overall, prosecutors were willing to accept pleas to manslaughter, but only in very limited cases where the prospect of conviction for murder was low. Even in such cases, a number of prosecutors believed that it may still be appropriate to charge murder in order to test the evidence at trial. There seemed to be a greater willingness among prosecutors to accept pleas of diminished responsibility compared to provocation. However, prosecutors' strong preference was to charge defendants with murder.

Question 2

B.88 The second question asked:

How would it affect charging decisions if there was an intermediate grade of homicide – say, 2^nd degree murder – between murder and manslaughter, with a discretionary life maximum sentence, that encompassed killings where the defendant's intent was to do only serious harm?

B.89 Of the questionnaire participants that directly commented on the effect such a change would have on charging and plea decisions, a minority (17%) thought that the proposal would be an improvement at the charge stage. One respondent stated: "It would be a major improvement and far fairer." 28% thought that it would make the situation worse at the charge stage. 56% thought that this proposal would have little impact at the charge stage. The remainder of respondents did not comment directly on whether they thought the proposal would be an improvement at the point of charge or not.

B.90 Generally, the responses indicated that the introduction of an intermediate grade of homicide would not have a great impact at the point of charge. Charging occurs very early on in a matter, before the prosecution has necessarily accumulated sufficient evidence to determine whether the defendant intended to kill or cause serious harm. As a result, defendants are generally charged with the higher level offence on the basis that it can be downgraded later (either prior to trial or on the basis of a plea). Consequently introducing a lesser offence would make little difference at the point of charging a defendant as it is likely that the prosecution would continue to charge with murder regardless.

B.91 Of those who did consider the potential change at the charge stage, one indicated that the decision-making process needed to be robust, and so the decision to charge with a lesser offence should be limited to a number of authorised prosecutors. This was likely to cause delays in gathering sufficient evidence and it would be unlikely that it would be available at the point the charging decision needed to be made. Another prosecutor thought that the new offence would make the charging process more difficult by creating a new "borderline" between "first degree murder" and "second degree murder" which would be difficult to negotiate. This would be complicated by the wishes of the victim's family. Similarly, another prosecutor stated that the proposal would be unlikely to have an impact on charging, "[g]iven the expectation of relatives of victims for a trial, Prosecutors and police offices would be under pressure to charge murder."

B.92 A number of prosecutors noted that the introduction of an intermediate grade of murder may make a difference to pleas. If the defendant had the option of pleading guilty to "second degree murder", this might encourage pleas given that the mandatory life sentence would not apply. Further, the prosecution might be more willing to accept such a plea if the alternative was having to prove intent to kill. One response stated: "This would result in nearly all cases being resolved as 2^nd degree."

B.93 Some prosecutors approved of such a change to the operation of pleas. However, others disapproved of the prospect of more defendants being found guilty of a lesser offence carrying a lesser sentence than they would under the current system. Some thought there was potential that such a change would bring the system into disrepute by introducing a 'soft option'. A number of prosecutors were concerned that victims' families would perceive it in this light. Further, another respondent was concerned that defence lawyers would tell clients to say they only intended to cause serious harm, not to kill.

B.94 Other prosecutors raised concerns over labelling. One thought that "second degree murder" sounded too American, but conceded that this may make it more easily understandable for the public who are already familiar with the term from television. Another prosecutor thought that if intent to cause serious harm was withdrawn from murder, it was really downgrading it to manslaughter, "calling it second degree murder is a fig leaf to prevent the outcry when what the public would see as a murder is charged as manslaughter." Another asked: "Are we not going through hoops to deal with the problem of mandatory life sentence for murder?"

B.95 Similarly, another prosecutor noted that this was a back door way around the mandatory life sentence. However, this prosecutor thought that the proposal did not address cases where the defendant intended to kill out of compassion, not malice.

B.96 One response was suspicious of the motives behind this proposal. The prosecutor stated:

This proposal is presumably to allow judges greater flexibility in sentencing. It cannot be to increase the conviction rate if the statistics in this section are correct. Why do we not trust the judgment of juries which convict of murder? If flexibility in sentencing is the motive behind the proposed change how is that consistent with the increased use of indeterminate sentences introduced by the CJA in 2003?

B.97 Finally, other responses questioned whether it was fair to reduce intent to cause serious harm to a lesser offence. One argued that introducing a lesser offence was not appropriate.

B.98 Overall, most prosecutors did not think the introduction of "second degree murder" for cases of intent to cause serious harm would have a great impact upon the charging process and decisions. However, some suggested that it may affect the offer and acceptance of pleas to the lesser offence. Some prosecutors thought this was a good thing. However, others were critical of this outcome as defendants would arguably get away with a lesser offence and sentence than they deserved.

QUESTIONNAIRE SENT TO PROSECUTORS MURDER PROJECT: PROSECUTION DECISIONS IN HOMICIDE CASES

The Law Commission is currently embarking on a review of the law of murder. We would very much like to have your observations on some possibilities for change.

(A) THE DEFINITION OF MURDER

CURRENT LAW: Murder may be committed when a killing stemmed from either an intention to kill or from an intention to inflict grievous harm. The latter can be called the 'GBH rule'.

The Law Commission would find it invaluable to have the views of prosecutors on whether a fairer outcome, but one that does not make the process of prosecution unduly complex, would be more likely if there were a change to the GBH rule along one or other of the lines suggested below.

The first alternative below is meant to be much more restrictive than the GBH rule.

The second alternative is meant to be slightly more restrictive than the GBH rule.

The third alternative reflects the law in many other common law jurisdictions.

The first and third alternatives are wholly subjective in nature, whereas the second alternative has an element of objective judgement to be passed by the jury on whether the harm was life-threatening.

Unlike the first and second alternatives, the third alternative does not employ the term 'intention'.

FIRST ALTERNATIVE: D is guilty of murder if he intended to kill, or if he

intended to cause what he realised at the time might be a life-threatening injury.

Comments:

SECOND ALTERNATIVE: D is guilty of murder if he intended to kill, or if he

intended to cause a harm the jury regards as inherently life-threatening at the

time it was done.

Comments:

THIRD ALTERNATIVE: D is guilty of murder if he intends to kill, or is recklessly

indifferent at the time to the causing of death.

Comments:

It may help to focus on a scenario:

A 25-year-old man (D) of average build, strength and intelligence admits killing the victim (V) whilst sober, but denies that he had the mental element required for murder. D will not plead provocation.

D had an argument with his wife V who he thought he heard confessing to adultery. He lost his temper, picked up a small kitchen knife lying nearby on the table, and stabbed V once in the stomach, the blade penetrating about three inches. Regaining his temper, D immediately called an ambulance.

Through no one's fault, complications arose with regard to V's treatment and she died. D's claim is that, at the time, he did not think about the possible consequences of stabbing V.

Which of the legal definitions above is most/more/less/least likely to focus on what are, in your view, the significant issues (answer: 'most likely'/'less likely'/'least likely'/'equally likely'):

Current law:

First Alternative:

Second Alternative:

Third Alternative:

(B) DIMINISHED RESPONSIBILITY

1. The Commission will be considering this partial defence in some detail. Suppose that the defence is preserved in its current form (as a partial defence that can be raised when murder is charged). It is likely that the Commission will be recommending that a successful plea of diminished responsibility or provocation reduces the crime to "second-degree murder", and not to so-called voluntary manslaughter, as now. The term 'manslaughter' would be reserved for 'involuntary' manslaughter.

Do you support this proposal, if no other changes are made?:

2 The Commission is concerned that the ability of persons accused of murder to run provocation and diminished responsibility together may lead to unwarranted acquittals, because it is no longer possible adequately to separate the issues relevant to each defence at trial (R v Smith [2001] 1 AC 146). The Commission has already recommended that provocation be reformed to make such separation easier (Law Com 290, Partial Defences to Murder). Another approach would be to create a separate offence of 'diminished responsibility killing', leaving provocation alone as a partial defence to murder.

Can you see advantages to this approach?:

Can you see disadvantages to this approach?:

Comments: 3. The Commission has concerns about the adequacy of the procedure for assessing whether D's responsibility is diminished. Under French criminal procedure, if there are questions about D's mental state at the usual preliminary examination of the issues by a judge, the latter can commission an examination of D by court-appointed medical experts whose written reports are disclosed to both sides well before trial. Both sides are free to ask for a further report on D's mental state, and all the experts give evidence at trial.

Are the procedures for assessing D's mental state prior to and/or at the trial adequate?:

In what way, as through adoption/variation of the French system, could the procedures be improved?:

(C) CHARGING DECISIONS

Since the mid-1960s, whilst the rate of recorded homicides has roughly doubled (c.400-800), receptions into prison for murder have increased four-fold (58-208). Receptions doubled between 1965 -1975 (58-103), and doubled again between 1975 and 1991 (103-208). Juries have clearly been convicting of murder in an ever larger proportion of homicide cases. This may suggest that pleas of provocation or diminished responsibility are increasingly less successful before the jury. It may suggest pleas of guilty to manslaughter are accepted less readily in serious cases by the prosecution.

1.In what circumstances is it right to accept a plea of guilty to manslaughter?

2. How would it affect charging decisions if there was an intermediate grade of homicide – say, '2^nd degree murder' – between murder and manslaughter, with a discretionary life maximum sentence, that encompassed killings where the defendant's intent was to do only serious harm?

FEEL FREE TO ADD FURTHER COMMENTS ON ANY OF THE ABOVE QUESTIONS ON THE BACK OF THIS SHEET

APPENDIX C SUBMISSIONS FROM JUDGES ON THE FAULT ELEMENT IN MURDER

C.1 The Law Commission sent a question paper to a random sample of judges who have experience of murder trials to ask their opinions on whether reform of the fault element of murder was desirable and, if desirable, what changes needed to be made. We asked the judges to express their views on the following options for reform of the fault element of murder:

(1) D can be guilty of murder only where D had an intention to kill or cause serious harm (the current law);

(2) D can be guilty of murder only where D had an intention to kill and nothing else;

(3) D can be guilty of murder only where D had an intention to kill or cause risk to life;

(4) D can be guilty of murder only where D had an intention to kill or cause harm that the jury regards as being inherently life-threatening; or

(5) D can be guilty of murder only where D had an intention to kill or he indifferently ran a foreseen risk of causing death.

We also provided a space for any other comments that the judges may have had.

C.2 The question paper did not expressly offer the option of having degrees of murder. In other words it did not make clear whether "murder" meant the single highest level homicide offence, as it currently does, or whether it might include several offences, for example both "first degree murder" and "second degree murder". This means that some judges may have proceeded on the basis of the first interpretation of "murder" and some the latter.

C.3 Thirteen judges responded to our question paper. The majority, nine, believed the current law was inadequate and favoured change. Only four judges believed that the current law was superior to all the offered alternatives.

C.4 Of the nine judges favouring reform six, or two-thirds, favoured limiting murder solely to those killings that were committed with an intention to kill. Two judges would include intention to cause risk to life. Finally, one judge favoured a definition of murder that consisted of an intention to kill or cause harm that the jury regarded as inherently life-threatening.

C.5 The additional comments made by the judges favouring reform were also interesting. Three of the nine, or a third, suggested that an intention to kill ought to lead to a conviction for "first degree murder" and an intention to cause serious harm a conviction for "second degree murder". Another judge noted that it was inappropriate to convict D of murder where he never intended harm that is life-threatening.

C.6 Those judges who preferred limiting murder to intentional killings tended to believe that the other options, such as "intending to cause a risk to life" or "intending to cause harm that the jury regard as inherently life-threatening", would be too complicated for juries to apply easily in practice.

C.7 Interestingly, half of the four judges who preferred the current definition of murder over any of our proposed reforms expressed some preference for changes to the law in their additional comments. One said that he preferred the current definition because it worked in practice but said he would like to see the mandatory life sentence abolished. Another said that a further jury decision on "degrees of murder" may have merits.

C.8 The judges' answers are presented in tabular form below:

Option for the fault element of murder to include	Judges in support
1) Intent to kill or do serious harm (current law)	4
2) Intent to kill	6
3) Intent to kill or to cause risk to life	2
4) Intent to kill or to cause harm the jury regard as inherently life threatening	1
5) Intent to kill or indifferently run a foreseen risk of death	0
TOTAL	13

APPENDIX D

THE LAW OF HOMICIDE IN OTHER

JURISDICTIONS

SCOTLAND

D.1 There are two offences of homicide in Scotland: murder and culpable homicide, which is equivalent to manslaughter. In appropriate cases a judge may withdraw culpable homicide from the jury and direct that they can either convict of murder or acquit.^[15] Culpable homicide may be charged when death results from an unlawful act, from recklessness, or found following the successful raising of either provocation or diminished responsibility. Provocation and diminished responsibility are no longer partial defences but denials of mens rea, which prevent the prosecution proving murder.^[16] There is no offence or defence of infanticide. Murder carries a mandatory life sentence whilst judges have complete discretion in sentencing following a conviction of culpable homicide.

D.2 The definition of murder can be stated as:

…any wilful act causing the destruction of life, whether [wickedly] intended to kill, or displaying such wicked recklessness as to imply a disposition depraved enough to be regardless of the consequences.^[17]

D.3 "Destruction of life" includes any shortening of life. As for the mens rea requirements, the courts are yet to fully develop the terms "wicked intention" and "displaying such wicked recklessness as to imply a disposition depraved enough to be regardless of the consequences".

D.4 "Intention" is particularly under developed, partly because of the availability of the alternative mens rea "wicked recklessness". Hence, it is unclear whether intention is limited to direct intention or includes oblique intention. Moreover, the redefining of murder's mens rea to require "wicked intention" means that those who were previously entitled to a defence now fall outside the scope of the offence instead. However, it is unclear whether this is all the redefinition does, or whether some defendants who previously lacked a defence may now also fall outside the definition of murder. Indeed, nothing suggests that the broader interpretation is not the correct one. This has implications for the partial defences (discussed below).

D.5 "Displaying such wicked recklessness as to imply a disposition depraved enough to be regardless of the consequences" also remains unclear. It is said that "wicked recklessness" is interpreted objectively.^[18] However, in this context the word "objective" does not bear its normal meaning, as an accidental stabbing certainly is not murder. Moreover, the acts from which "wicked recklessness" can be inferred must be done intentionally.^[19] Therefore, the term is not so much "objective" as, to some extent, "constructive". However, the extent to which this is the case is also unclear: previously judges have directed some juries that intention to seriously assault is a third type of mens rea for murder.^[20] Now, however, it appears that, if this suffices at all, it will only suffice in cases of armed robbery. The better view, however, is that intention to seriously assault is merely one means of establishing wicked recklessness,^[21] not an alternative form of mens rea.

D.6 In any event, the central issue is the accused's disposition. The prosecution need to prove "…wicked recklessness as to imply a disposition depraved enough to be regardless of the consequences". Therefore, it seems that intentional use of life-threatening violence alone is insufficient to fulfil the mens rea requirement of murder if the defendant had some regard for the consequences. Imagine that A and B carryout identical violent assaults which include kicking their victim whilst on the ground but do not include any blows to the head. A avoids blows to the head because he wants to reduce the chance of death. B gives the issue no thought and avoids blows to the head purely by chance. From B's attack the jury may be able to "imply a disposition depraved enough to be regardless of the consequences" and convict of murder. However, the same jury may not be able to imply the same from A's attack, and instead may only convict of culpable homicide. This is further proof, if needed, that wicked recklessness is not "objective" in the conventional sense of the word.^[22]

D.7 In Scottish law, like English law, murder can be committed in the second degree, that is by being the perpetrator's accomplice. Murder "art and part" is the equivalent to aiding and abetting in English law. Under this doctrine, the defendant will be guilty of murder if the perpetrator kills and the defendant actively associated himself with the perpetrator in a common criminal purpose which was, or included, the taking of human life or carried an obvious risk of this happening.^[23] Accordingly, the defendant can be guilty of murder art and part without having any foresight of injury, never mind death. Moreover, if the defendant associates himself with some lesser common criminal purpose and the perpetrator kills, the defendant will still be guilty of culpable homicide. However, this only applies to antecedent concert, not to spontaneous concert, where, for murder and culpable homicide, the defendant himself must be found to have the necessary mens rea for the offence.^[24] However, there appears to be no principled reason why antecedent and spontaneous concert should be treated differently.

D.8 Scottish law has two partial defences to murder, namely, provocation and diminished responsibility. Provocation may take the form of violence or the discovery of infidelity by one from whom the defendant could reasonably expect fidelity. In both cases the response must be immediate and, in the case of the former, proportionate too.^[25] Of course, under the broader interpretation of wicked intent, cases falling outside these restrictions may still succeed on a basis of denying the required mens rea.

D.9 Diminished Responsibility has recently been broadened.^[26] It now appears that any psychological or psychiatric disorder will suffice, and may even be triggered by sunstroke, alcoholism, thyroid problems, prescribed drugs or abuse and still be the basis of a diminished responsibility plea. However, voluntary intoxication and personality disorders cannot form the foundation of the plea.^[27] In order for the plea to succeed, the mental disorder must cause a substantial impairment of the defendant's ability to control his actions. Previously, judges have been able to withdraw the issue from the jury. However, this will only remain possible if the effect of the redefinition of the mens rea of murder to require wicked intent is limited only to excluding those who previously had a defence from the definition of the offence.

FRANCE

D.10 Despite its similar name, meurte is narrower than murder in English law: it requires an intentional killing. Where death is caused but the defendant only intended to inflict grievous bodily harm the death is regarded as an aggravating factor relevant only to sentencing following conviction for one of the lesser offences against the person.

D.11 Intention is not defined in the Code Penal or accompanying case law. However, French legal writing reveals that intention can be divided into direct intention, where the intended result must be desired, and indirect intention, where the defendant knows his act will have the "intended" result. Accordingly, intention in French law is similar to intention in English law.

D.12 Turning to sentencing, custodial sentences in France can be either réclusion criminelle (criminal detention) or "mere" emprisonnement. In practice there is little difference between the two and réclusion criminelle was almost abolished in 1994 when the new Code Penal was passed.

D.13 In French law there is no mandatory life sentence. The old Code Penal stipulated both a maximum and a minimum sentence for each offence. The new Code abandoned this practice and only stipulates maximum sentences. However, for the most serious offences, those potentially punishable by réclusion criminelle for life, the custodial term passed must be at least two years, and where the crime is potentially punishable by a determinate period of réclusion criminelle the custodial term passed must be at least one year.^[28] However, wherever the term passed is less than five years a court is free to suspend it.^[29] Finally, there is also the issue of the périod de sureté or safety period, during which D is ineligible for early release on parole. This has become the centrepiece on the crime debate in French politics, especially since the abolition of the death penalty in 1981.

D.14 Meurte carries a maximum sentence of 30 years réclusion criminelle without a safety period.^[30] However, there are also "aggravated murders", including murders committed in connection with another crime or where the victim falls within a range of stipulated categories, such as children under 15 years old, state officials, or minorities victimised by virtue of their membership of a minority group. For these murders the maximum sentence available is increased to life and a safety period of half the determinate term, or in the case of life sentences, 18 years, applies.^[31]

D.15 Where the killing is not only intended but also premeditated the crime ceases to be meurte and becomes assassination, which carries a maximum sentence of réclusion criminelle for life with a safety period.^[32]

D.16 So far as defences are concerned, provocation, self-defence, "mercy" killing and insanity raise issues worthy of brief mention.

D.17 The New Code penal, which was passed in 1994, abolished minimum penalties, and the partial defence of provocation was, therefore, deemed unnecessary and abolished.

D.18 The defence of self-defence will be available to the defendant where the action taken was an immediate response to an unjustified attack, including an attack of another, and the force used was not disproportionate.^[33] Moreover, the defendant will be judged according to the situation as it was reasonably believed to be. It will also be presumed that the defendant acted in self-defence where he or she acted: to repulse an entry to an inhabited place made by breaking in using violence or deception at night; or to prevent themselves becoming the victim of theft or pillage carried out by violence. However, this presumption has now ceased to be irrebuttable.^[34]

D.19 There is no defence of "mercy" killing in French law.

D.20 France has provisions that broadly equate to the English defences of insanity and diminished responsibility. Defendants will not be criminally liable where they offended whilst suffering from a psychological or neuropsychological disorder which destroyed their discernment or their ability to control their actions. Under this provision, unlike under English law, those suffering from irresistible impulses and those who are, at least involuntarily, intoxicated will not be criminally liable. Furthermore, while those suffering from a psychological or neuropsychological disorder that merely reduces their discernment or impedes their ability to control their actions, remain punishable, the sentencing judge is expressly told to consider the disorder when sentencing^[35].

GERMANY

D.21 The basic homicide offence in German law is Totschlag. Totschlag is defined as any killing of a human being carried out with an intention to kill that is not murder.^[36] It is broadly equivalent to voluntary manslaughter in English law. Because Totschlag requires an intention to kill, cases where death is caused with an intention only to do serious bodily harm will be charged as inflicting bodily injury resulting in death,^[37] an offence against the person not a homicide offence.

D.22 Meanwhile, murder is very narrowly defined in German law. Totschlag becomes murder when the killing is committed:

…out of a lust for killing, [or] in order to satisfy his sexual desires, [or] motivated by greed or other despicable reasons, [or] deviously or cruelly or with means capable of causing widespread mayhem, or in order to enable or to cover up the commission of another crime.^[38]

D.23 A killing is committed deviously where the defendant relies on the trusting nature and consequent helplessness of the victim to commit the crime. This range of situations is designed to catch those who commit murder in a way that evidences a higher anti-social personal disposition. Simply put, in these situations the defendant and his actions are more evil.

D.24 In German criminal trials nothing separates conviction and the passing of sentence. There is no plea in mitigation, instead the trial judge sentences the defendant on the basis of the information placed before the jury. Therefore, all defences and mitigating factors that the defendant wishes to rely upon must go before the jury. Accordingly, those mitigating factors that can only be placed before the court by the defendant himself will be left unconsidered if the defendant denies the offence completely.

D.25 Murder carries a mandatory life sentence.^[39] Unless the general provisions on diminished responsibility apply (see paragraphs D.29-D.31) the court has no discretion to pass a lesser sentence. This has caused problems where an upgrading element, which makes the offence murder coincides with mitigating elements that would downgrade the sentence if the crime had not become one to which a mandatory sentence applies. A paradigm example is where a battered woman suffering from provocation (but not diminished responsibility) "deviously" kills, that is murders, her abusive partner. The judicially engineered solution to this problem has been to hold that the courts are constitutionally authorised to disregard the mandatory life sentence in those cases where to impose it would be disproportionate.

D.26 Totschlag carries a sentence of not less than five years and, in the worse cases, life.^[40]

D.27 Meanwhile, where the crime displays the stipulated characteristics or circumstances to bring it within the provision governing less serious cases of manslaughter, the sentence will be between one and ten years. This provision primarily caters for killings that would come under the partial defence of provocation in English law. For the defendant to utilise this provision the victim must have directed physical or verbal abuse at the defendant or one of the defendant's close relatives and consequently the defendant must have suffered a loss of temper and thereby have been carried away to commit the offence. There must be no grounds for blaming the defendant for the loss of temper, for example the defendant must not have acted in a way to give the victim a reason for acting provocatively.^[41] Essentially, the provision requires an evaluation of the role of the defendant and the victim in the killing. Only where the defendant was the wronged party will this provision be available. However, this provision can also be relied on in "an otherwise less serious case". This need not be synonymous with provocation, the provision being a catch-all catering for cases where the defendant's ability to control his or her actions was reduced by factors like fear, duress, or compassion.

D.28 Killing on request is a special form of lesser manslaughter. Where the killing is performed at the victim's express and earnest request the defendant will receive a sentence of between six months and five years.^[42]

D.29 The Defendant's sentence will be mitigated where the offence is committed as a result of a:

…pathological emotional disorder,…severe mental disturbance,

severe mental retardation or other severe mental abnormality…^[43]

which substantially reduced D's ability to appreciate the wrongfulness of his or her conduct or his or her ability to act according to their appreciation of the conduct's wrongfulness.^[44] This covers the mental states covered in English law, as well as states not so covered, such as transitory states and self-induced states, even if culpably self-induced. Therefore, severe intoxication may lead to an abnormal mental state that can found a claim of diminished responsibility.

D.30 Where the diminished responsibility provision applies the sentence will be reduced, unless mitigation is refused because other aggravating factors outweigh the diminished responsibility based mitigation. Therefore, assuming that the mitigation is granted, a life sentence is reduced to a sentence not shorter than three years, whilst any determinate term will be reduced by one quarter.^[45]

D.31 Where double mitigation applies, for example where the defendant suffers from both provocation and diminished responsibility, the sentence will be between three months and seven and a half years. However, if the double mitigation arises from a killing on request committing under diminished responsibility the sentence will be at least one month and up to three years and nine months.

D.32 Turning to defences, German law provides two excusatory yet complete defences: excessive self-defence^[46] and exculpatory necessity.^[47] The former requires that the force used in response to an attack be excessive due to confusion, fear or panic. Meanwhile, the latter applies where the defendant commits an unlawful act in order to protect himself or herself, or a close relative from an imminent danger to life, limb or freedom that could not otherwise be averted.

D.33 The prosecution must disprove these defences, as no burden, evidential or legal, rests upon the defence.

CANADA

D.34 Sections 222-240 of the Canadian Criminal Code^[48] address homicide. These must, however, be read alongside the jurisprudence of the Canadian Supreme Court, as the Code has not been amended since numerous successful Charter^[49] challenges.

D.35 Culpable homicide is Canadian law's starting point. This is a genus of two species, manslaughter or murder, the two being separated by murder's additional mens rea requirement.

D.36 Culpable homicide is committed where death is caused by an unlawful act, criminal negligence, the victim's own actions (if triggered by violence or deception) or, in the case of a child or sick person, wilful frightening.^[50] However, the defendant will be found guilty of murder rather than manslaughter when they mean to cause death or bodily harm that is known to be likely to cause death (even if someone other than the intended victim dies), or they do anything that is known to be likely to cause death during the pursuit of an unlawful object, even if death is not desired.^[51] Importantly, subjective foresight of the risk of death is a constitutional pre-requisite for a murder conviction.^[52] These differing states of mind are briefly examined in turn below.

D.37 Motive is irrelevant when deciding whether the defendant meant to cause death: mercy killers are murderers.^[53] Oblique intent cases are invariably brought under the "unlawful object" limb but orthodox opinion suggests that the intention limb is also applicable. Therefore, this limb probably covers the oft-cited plane bomber seeking to claim under their cargo insurance.

D.38 The "likely" in "means to cause bodily harm that is known to be likely to cause death" is a simple word requiring no elaboration.^[54] Moreover, where a series of acts leads to the victim's death the requisite knowledge need only be present momentarily during the series of acts.

D.39 Finally, what might be called the "unlawful object" mental state covers cases where the defendant engages in a scheme that, if completed, would amount to a serious crime, and during this scheme does something which is known to be likely to cause death. This will catch the terrorist who plants a bomb giving a barely sufficient warning, who may not fall within the "means to kill" limb.

D.40 The sentence for murder depends on the type of murder: first degree or second degree. First degree murderers are ineligible for parole for 25 years. Second degree murderers have a minimum tariff set between 10 and 25 years by the trial judge, who may receive a recommendation from the jury.^[55]

D.41 All murders are second degree murders, except planned and deliberate murders (where a period of planning separates the intention's formation and the act's commission), contract killings, murders of stipulated officials (for example police officers) where the defendant knew the victim was a member of the said category, or murders committed during the commission of certain stipulated offences.^[56]

D.42 In Canadian law, as in English law, it is possible to commit murder as an accomplice. Sections 21-23 of the Code address aiding, abetting and counselling.

D.43 Aiding and abetting can take many forms. For example, aiding includes supplying an instrument or acting as look out, while abetting, which is separate to aiding but similar to counselling, consists of encouraging or supporting the principal. The secondary party must also share the principal's level of mens rea: they must intend to aid the killing of the victim or intend to abet the causing of bodily harm known to be likely to cause death. Vitally, the secondary party must act with the intention to assist the principal commit murder.

D.44 The secondary party can be convicted of manslaughter and the principal of murder, or vice versa. Moreover, if A assists B to murder C but B kills D by mistake, A is probably liable for the death via extended transferred malice. Finally, parties to a joint unlawful enterprise are liable for all crimes committed during that enterprise if foreseen as the enterprise's likely consequence.

D.45 Meanwhile, counselling is treated identically to aiding and abetting, except that for counselling the Code states that the counsellor remains liable where the principal adopts a different means of committing the offence. This would probably be applied by implication to aiders and abettors too; however, the point is academic, as counsellors are normally abettors too.

D.46 Under Canadian law, the defendant can rely on a range of defences, including provocation, mental disorder, automatism, duress, necessity, intoxication, self-defence, and what is called "the rolled up plea".

D.47 Provocation is the only partial defence to murder. The killing must occur during a sudden loss of self-control following a wrongful act or insult sufficient to make the ordinary person lose self-control.^[57] It is unnecessary to show that the ordinary person would have done as D did, only that he would have lost self-control. There is a liberal approach to "sudden" and the "ordinary person" will share the defendant's age, sex and any other factors that would give the provocation special significance.^[58]

D.48 If D is rendered incapable of appreciating the nature or quality of his or her act or that it was wrong by a mental disorder he will be found "not criminally responsible on account of mental disorder".^[59] This is broader than the M'Naughten rules: "mental disorder" includes any illness or abnormality that impairs the mind's functioning, except self-induced or transitory states.^[60] Meanwhile, "appreciate the nature and quality of the act" includes estimating the act's consequences.^[61] Whilst "wrong" means morally wrong according to society's standards.^[62] A finding of mental disorder entitles D to the least restrictive disposal possible: an absolute discharge if D is not a danger to the public.^[63]

D.49 To avail themselves of the defence of automatism the defendant must establish the act's involuntariness on the balance of probabilities.^[64] Then the judge will decide whether the plea is one of automatism or mental disorder. In deciding this, the judge will examine whether the act's cause was internal and assess any danger of repetition.

D.50 Two duress defences existed: one at common law, one in the Code.^[65] It now appears that the statutory provision is redundant, bar its list of excluded offences.^[66] The common law defence is available where the defendant faces threats of death or serious harm, has no safe avenue of escape, has not exposed himself to the threats, and (it seems) acts proportionately. A Charter challenge to the statutory list of excluded offences, which applies to the common-law defence, is awaited.^[67]

D.51 The defence of necessity permits prima facie unlawful action where a peril or danger is imminent and the defendant has no reasonable legal alternative course of action, so long as any harm caused is proportionate to that avoided.

D.52 Murder is a crime of specific intent so intoxication that prevents mens rea being formed reduces murder to manslaughter. The Legislature abolished a judicially created general defence of extreme intoxication.^[68] A Charter challenge to this enactment is awaited.

D.53 The Code's complicated and much maligned^[69] provisions do not contain a defence of excessive self-defence. However, the liberal approach to pre-emptive self-defence benefits battered women who kill.^[70]

D.54 Finally, whilst Supreme Court authorities conflict,^[71] it appears that the cumulative effect of failed defences can negate mens rea: the so called "rolled up plea".

UNITED STATES OF AMERICA

D.55 The states in America divide into two: those with murder statutes based on the English Common Law, of which California is typical; and those with murder statutes based on the American Law Institute's Model Penal Code, of which New York is typical.

D.56 The California Penal Code largely retains the old common law definition of murder:

The unlawful killing of a human being, or a foetus, with malice aforethought.^[72]

D.57 There is also a rebuttable presumption that the killing was lawful if death occurs more than three years and a day after the defendant's act.^[73]

D.58 In contrast, New York's crimes of homicide have a very different structure. Homicide is defined as:

Conduct which causes the death of a person …under circumstances constituting murder, manslaughter in the first degree, manslaughter in the second degree, [or] criminally negligent homicide….^[74]

D.59 Notably, "causing death" rather than "killing" is required under the New York Penal Code. This better covers cases of complicity. Subsequent provisions then add the mental elements that dictate which crime of homicide is committed.

D.60 Turning to issues of mens rea, under the California Penal Code malice can be express or implied. Express malice equates to an intention to kill.^[75] However, this excludes cases of knowledge rather than intent and cases where grievous bodily harm rather than death is intended. Despite the California Penal Code's narrow definition of implied malice, which states that implied malice exists where "no considerable provocation appears, or when the circumstances attending the killing show an abandoned and malignant heart",^[76] such cases do in fact fall within implied malice by virtue of the California Jury Instructions, which state that implied malice exists where:

(1) the killing resulted from an intentional act;

(2) the natural consequences of the act were dangerous to human life; and

(3) the act was deliberately performed with knowledge of the danger to, and with conscious disregard for, human life.^[77]

D.61 Accordingly, both the person who blows up a plane to claim under their cargo insurance policy and the person who intends to inflict grievous bodily harm but not death will have malice implied.

D.62 Implied malice is also the statutory source for the judicially developed doctrine of felony murder. This states that any killing in the course of an inherently dangerous felony will be murder, unless the felony is so close in nature to homicide as to merge with it, in which case it ceases to be able to be an independent basis for a finding of murder.

D.63 Additionally, second degree murder will be elevated to first degree murder where certain circumstances exist, including premeditation, the use of "weapons of mass destruction", or where the killing occurs during the commission of certain offences.^[78] The victim's status does not elevate the murder to first degree murder but does increase the judge's sentencing powers.^[79]

D.64 Turning to mens rea under New York's Penal Law, unlike in California, there is not a single mens rea for murder. Instead the Penal Law says homicide will be second degree murder where:

(1) with intent to cause the death of a person the defendant causes the death of that person or another person;

(2) under circumstances evincing a depraved indifference to human life, the defendant recklessly engages in conduct which creates a grave risk of death to another person, and thereby causes the death of another person;

(3) the defendant, during the commission of a stipulated crime, or flight from the scene of a stipulated crime, causes the death of another not involved in the crime; or

(4) under circumstances evincing a depraved indifference to human life, and being eighteen years old or more the defendant recklessly engages in conduct which creates a grave risk of serious physical injury or death to another person less than eleven years old and thereby causes the death of such person.^[80]

D.65 Furthermore a person acts:

(1) intentionally, where it is their conscious objective to cause a result or to engage in such conduct;

(2) knowingly, where he is aware that his or her conduct is of such a nature or a circumstance exists; and

(3) recklessly, where he is aware of, and consciously disregards, a substantial and unjustifiable risk that such a result will occur or such a circumstance exists.^[81]

D.66 The failure to mention "knowingly" in D.64(1) appears to be an oversight, especially since it is included in the Model Penal Code, but one of limited importance since such cases will normally fall within D.64(2). Unlike in California, felony murder has a statutory basis and the courts have held that the list of offences in the Penal Law are exhaustive, thus supposedly preventing the issue of merger arising.^[82] The felony murder provision also departs from the Model Penal Code, which creates a rebuttable presumption that felony murders are committed with the extreme indifference required by D.64(2), rather than establishing them as a distinct means of committing second degree murder as in D.64(3).

D.67 Meanwhile, second degree murder will be elevated to first degree murder where the killing displays certain characteristics, including the victim being of a certain status (for example a law enforcement officer or a witnesses), and the killing occurring during the commission of certain stipulated offences.^[83]

D.68 Finally, California and New York have different approaches to defences. Under the Californian Penal Code, once murder is prima facie proved by the prosecution, the burden to establish any defence moves to the defendant.^[84] The following defences are specific to homicide:

(1) Excusable Homicide,^[85] which requires the homicide to be committed: by accident during a lawful act performed without any unlawful intent; or by accident and misfortune, in the heat of passion,^[86] upon any sudden and sufficient provocation, or upon a sudden combat, when no undue advantage is taken, nor any dangerous weapon used, and when the killing is not done in a cruel or unusual manner.

(2) Justified Homicide,^[87] which includes killings in the course of the defence of oneself or another from a reasonably anticipated felony, or in defence of property or habitation. Therefore, the thief who intends to forcibly steal a wallet risks a lethal response in self-defence.

D.69 Meanwhile under New York Penal Law, first degree murder is reduced to second degree manslaughter by the defence of acting under extreme emotional disturbance. This appears to be a combination of provocation and diminished responsibility, which focuses on the defendant's state of mind. Meanwhile, second degree murder can also be reduced to second degree manslaughter if the killing occurs in the course of aiding a suicide.^[88]

AUSTRALIA

D.70 The law of homicide is a matter for State jurisdiction. The laws adopted by the States on murder do vary to some degree. However, in most significant respects there is a large degree of convergence. This is in part because the High Court of Australia is the final court of appeal for questions on murder regardless of which State's law the question arises under. Currently Victoria and South Australia retain the Australian common law, New South Wales and the Australian Capital Territory have statutory enactments, and Queensland, the Northern Territory, Western Australia, Tasmania, and the Commonwealth have a completely codified criminal law (although they still rely in part on the common law when interpreting their code).

D.71 Typically, murder is defined as a killing committed with an intention to kill or to inflict grievous bodily harm, or being "reckless" as to killing (but not as to inflicting grievous bodily harm). These mental elements shall be considered in turn.

D.72 Intention has repeatedly been held to be a simple word that forms part of everyday English. As such the appellate courts have frowned upon any attempt to explain or define the word, preferring juries to be left to apply it as their common sense dictates. The Australian courts have successfully avoided the difficulties experienced by the English courts when dealing with "intent" largely because there has been virtually no need to consider the concept of indirect intention in Australia.

D.73 A major reason for this has been that the Australian States have some concept of "reckless" murder. However, "reckless" in this concept does not have the same meaning as in English law or indeed the Australian law on non-fatal offences against the person. To avoid confusion the High Court of Australia did not use the word in its seminal statement of the mental element for murder and has made clear that the word "reckless" ought not to be used when directing the jury in murder cases.^[89] It seems that what Australian lawyers mean when they refer, for want of a better universally recognised phrase, as "recklessness" in the context of murder means something vaguely akin to indirect intention in English law. In Crabbe it was held:

If an accused knows when he does an act that death or grievous bodily harm is a probable consequence, he does the act expecting death or grievous bodily harm will be the likely result, for the word "probable" means likely to happen. That state of mind is comparable with an intention to kill or to do grievous bodily harm… a person who, without lawful justification or excuse, does an act knowing that it is probable that death or grievous bodily harm will result, is guilty of murder if death in fact results.

D.74 The Australian High Court has since confirmed that "probable" or "likely" does not mean more probable or likely than not: a substantial or real chance, even if less than 50%, suffices.^[90] This is, of course, a far lower threshold than English law's "foresight of a virtual certainty".

D.75 The statement of principle in Crabbe has been left largely unchallenged since. Indeed, the practices of the sentencing courts seem to affirm the idea that "reckless" killers are "comparable" to intentional killers. Whilst reckless killings are more likely to be seen as less bad, the courts have shown themselves willing to treat the worst cases of "reckless" killings as harshly as intentional killings. It appears that reckless killings are most likely to be viewed as being of the worst type where the recklessness displays indifference to life, that is, suggests that the defendant would have acted as he did even if he had known that death was a certainty.

D.76 Whilst the courts have differentiated between what may be called "types" or "degrees" of recklessness at the sentencing stage, attempts to place a gloss on the word reckless in the definition of the substantive offence of murder have failed. In the past those who have argued that the law should focus on the defendant's attitude to the risk, rather than his mere awareness of it, have seized upon the use of the word "expecting" in "expecting death or grievous bodily harm will be the likely result". They have argued that "recklessness" should require an "expectation" of death rather than a mere "appreciation". However, it has become clear that the use of the word "expecting" was of limited significance and the correct approach is to simply consider the defendant's awareness of the degree of risk being run.

D.77 Similarly, the repeated assertion that it is the degree of risk foreseen that is relevant, combined with the fact that the definition of the mental element deliberately declines to use the word "recklessness", has doomed any attempt to make the fault element of murder include the running of a risk which is unjustified. Questions of justification or excuse in cases of reckless killings, like intentional ones, arise solely at the defence stage.

D.78 However, whilst the above may be clearly established, the major difficulty with the definition of the mental element of murder is that fact that the line between murder and manslaughter is hazy. It is unclear precisely where "recklessness" sufficient for a murder conviction becomes simple everyday recklessness (that is, the unjustified running of a foreseen risk of any degree) rendering the killing only manslaughter. Meanwhile, manslaughter can be committed by gross negligence or by foreseeing a risk that is insufficient to come within "reckless" murder.

D.79 So far as sentencing is concerned in murder cases some but not all States retain the mandatory life sentence. All States empower their judges to set non-parole periods, although some have taken legislative action to curtail the judicial discretion when doing so. In manslaughter cases those States that do not have a discretionary life sentence have a maximum sentence of at least 20 years.

D.80 Turning to defences, self-defence is a complete defence to murder and the States each provide for a varying range of partial defences, such as provocation, duress and diminished responsibility. Interestingly, Crabbe suggested that the Australian common law will allow necessity as a defence to "reckless" murder and murder committed with an intention to inflict grievous bodily harm but not to murder committed with an intention to kill.^[91] However, case law since Crabbe has offered no further guidance on the issue.

D.81 Finally, a brief word about complicity. A defendant will be guilty for a murder committed by his accomplice wherever the offence came within the scope of their criminal plan, no matter how unlikely the commission of the crime seemed when the plan was formulated. Moreover, the defendant will be guilty of a murder committed by his accomplice outside the scope of their plan if the defendant foresaw the commission of the offence as possible. The harsh width of this doctrine has been justified by analogy with the doctrine of unlawful act manslaughter, found in one form or another in all the States.

APPENDIX E

ANALYSIS OF "LIFER" CASES

INTRODUCTION

E.1 Whilst all murders are by definition serious, some murders can be seen as being "more" serious. This is reflected in the tariff that is set for each murderer. The tariff is the period of the mandatory life sentence that the convicted murderer must serve before release. It is designed to reflect the period of time that must be spent in custody for the purposes of retribution and deterrence. Once this period has been served, the murderer may be released on licence. Such licences are by definition conditional and revocable. The murderer will only be released on licence if the parole board is satisfied that the murderer does not pose an unacceptable risk to the public. It is important to appreciate that the tariff's expiry is a necessary, but not sufficient, condition of a murderer's release on licence.

E.2 The Law Commission has conducted research into murderers whose tariff was, comparatively speaking, "short", with a view to investigating what common factors recurred in these cases. We decided to examine the cases of all murderers who were convicted between 1994 and 1996, whose tariff was set at 10 years or less, and who had been released by 1 August 2005. Having reviewed the files held by the National Offender Management Lifer Review and Recall Team^[92] 53 such cases came to our attention.

SUMMARY OF OUR FINDINGS

E.3 All the cases in the sample had some feature that reduced the defendant's culpability for the killing. This tended to be because the defendant was easy to sympathise with; or, alternatively, because the victim was more difficult to sympathise with, normally because they were, to some extent, the author of their own misfortune.

E.4 The most common mitigating factor was some form of provocation. On occasions this verged on being sufficient to provide a partial defence that would reduce the crime from murder to manslaughter. However, even where this was not the case, the provocation was invariably sufficient to constitute a genuine mitigating circumstance.

E.5 After provocation the most common mitigating factors were diminished responsibility and self-defence. Those cases involving a claim of diminished responsibility that appear to have verged on providing the defendant with a partial defence almost invariably involved a conflict between experts, one of whom testified that the defendant came within the partial defence and ought, therefore, be found guilty only of manslaughter. Those cases where the defendant's mental disorder was not sufficient to form a potentially viable partial defence almost invariably involved a low scale mental disorder, often some form of reactive depression.

E.6 Meanwhile, whilst excessive force used in self-defence may not have been the largest category it did contain some of the most striking cases. One particularly striking case illustrated that the gap between complete acquittal and a murder conviction can be as little as a second "unnecessary" stab at the aggressor.

E.7 There was also a noticeable subset of cases which, whilst involving no suggestion of a partial defence whatsoever, were brought within the sample by virtue of the fact that they were committed by young defendants who were frequently not just young but also immature for their age.

E.8 It appeared that the most common relationship between defendant and victim was that of man and wife, cohabiting couple, or parting couple. Meanwhile, a knife was the most common weapon used. Frequently this would be picked up by the defendant in the heat of the moment.

DETAILED BREAKDOWN OF OUR FINDINGS

E.9 So far as potential defences are concerned, of the 53 cases in the sample:

(1) 8 showed a degree of provocation that appeared to verge on providing a partial defence; and

(2) 12 showed a degree of provocation that whilst being insufficient to provide a partial defence offered some degree of mitigation.

(3) 5 showed a degree of diminished responsibility that appeared to verge on providing a partial defence; and

(4) 3 showed a degree of diminished responsibility that whilst being insufficient to provide a partial defence offered some degree of mitigation.

(5) 4 showed what appeared to be barely excessive degrees of force used in self-defence; and

(6) 9 showed the use of a degree of force that whilst being clearly excessive nonetheless would have provided some degree of mitigation.

(7) 10 involved young and immature defendants.

(8) 2 involved defendants who were extremely intoxicated.

E.10 So far as the relationship between victim and defendant is concerned:

(1) 11 cases involved couples, whether married or cohabiting;

(2) 4 involved ex-partners;

(3) 3 involved other family relationships such as parent-child;

(4) 7 involved friends;

(5) 17 involved acquaintances not falling into preceding categories, such as employer-employee or fellow employees; and

(6) 11 involved complete strangers.

E.11 Finally, so far as the weapon used is concerned:

(1) 31 cases involved the use of a knife;

(2) 7 involved the use of punching or kicking or both;

(3) 4 involved blunt, non-penetrating trauma;

(4) 7 involved strangulation; and

(5) 4 involved the use of a gun.

E.12 This information is summarised in tabular form below.

		Couple	Family	Friends	Acquaintance	Stranger
Provocation	Major	A1 B1 C2	A0 B0 C0	A1 B1 C0	A0 B0 C1	A0 B0 C0
		D0 E0	D0 E0	D0 E0	D0 E1	D0 E0
D0 E0	Minor	A2 B0 C0	A0 B0 C0	A1 B0 C0	A3 B0 C0	A1 B0 C0
		D4 E0	D0 E0	D0 E0	D0 E0	D0 E1
Diminished	Major	A0 B0 C1	A0 B0 C0	A1 B0 C0	A0 B0 C0	A0 B0 C0
		D3 E0	D0 E0	D0 E0	D0 E0	D0 E0
D0 E0	Minor	A1 B0 C0	A0 B0 C0	A2 B0 C0	A0 B0 C0	A0 B0 C0
		D0 E0	D0 E0	D0 E0	D0 E0	D0 E0
Self-Defence	Major	A0 B0 C0	A1 B0 C0	A0 B0 C0	A2 B0 C0	A1 B0 C0
		D0 E0	D0 E0	D0 E0	D0 E0	D0 E0
D0 E0	Minor	A0 B0 C0	A1 B0 C0	A0 B0 C0	A3(1) B1 C0	A2 B1 C0
		D0 E0	D0 E1	D0 E0	D0 E0	D0 E0
Youth	A0 B0 C0	A0 B0 C0	A0 B0 C0	A3(2) B2(1) C0	A3(1) B1 C0	A3(1) B1 C0
	D0 E0	D0 E0	D0 E0	D0 E0	D0 E1(1)	D0 E1(1)
Intoxication	A0 B0 C0	A0 B0 C0	A1 B0 C0	A1 B0 C0	A0 B0 C0	A0 B0 C0
	D0 E0	D0 E0	D0 E0	D0 E0	D0 E0	D0 E0

KEY

A: Knife

B: Fist/Boot

C: Blunt Object

D: Strangulation

E: Gun

(x): Of whom are secondary parties.

APPENDIX F

A CASE STUDY

INTRODUCTION

F.1 This section looks at the recent case of S.^[93] The facts do not conform to the archetypal killing by a woman who is the victim of domestic violence.^[94] However it was argued that the appellant in S was suffering from a personality disorder that had arisen in part from a history of physical and sexual abuse. The case was affected by the procedural and substantive difficulties common to cases in which such disorders are at issue.

THE FACTS OF THE CASE

Facts

F.2 D, in her thirties, killed V, a diabetic in his sixties, by administering several syringes of insulin. The two were described as having a complex relationship. V had been a lodger in D's house during her childhood. Since she was a child, D had stayed at V's flat from time to time when she was without accommodation. D was addicted to alcohol since her early teenage years, and later became addicted to other drugs. When D was desperate for money or drugs, V would pay D to perform sexual acts upon him. Their relationship was amicable at the relevant time. D assisted V with managing his diabetes. D was heavily intoxicated by alcohol and other drugs at the time of the offence.

At trial

F.3 D's son gave evidence that D prepared and administered the insulin to V saying that she was going to kill him. V was found dead the following morning.

F.4 D denied having administered the insulin at all. Shortly after V's death, while suffering the effects of withdrawing from drugs, D said that her son had administered the insulin. At trial, she said she had no recollection of attacking V, or of wanting to harm him. She may have injected V under the impression that it was medically necessary, but she was of the opinion that she would have remembered doing so. The second limb of her defence was that causation had not been made out – insulin was not in itself toxic and, owing to V's weak heart, death could have occurred at any time.

F.5 No diminished responsibility argument was advanced. This was, firstly, because the available medical assessments disclosed only minimal evidence of a relevant mental disorder. Secondly, diminished responsibility was not compatible with the other defences advanced: diminished responsibility is predicated on acceptance of responsibility for death coupled with the requisite intent,^[95] whereas D's other arguments variously sought to negate the actus reus, causation and intent.

On appeal

F.6 The Court examined two new psychiatric reports indicating that the appellant had a viable defence of diminished responsibility, as well as a report prepared at the request of the Crown which suggested the contrary.

F.7 The first new psychiatric report was prepared in 2004 by Dr B, who had previously prepared a report for the purposes of a bail application in 1996. Dr B found that the appellant was suffering from a long-standing personality disorder. The disorder caused her to maintain that she was not responsible for the death at all. Thus the very condition from which the appellant was suffering had effectively removed the possibility of putting the issue of diminished responsibility before the Court. Dr B had not addressed diminished responsibility in her 1996 report because she had only been instructed in respect of a bail application. In 1996 she had no access to the relevant medical history; it would therefore have been impossible to comment on diminished responsibility.

F.8 The second new psychiatric report was prepared in 2003 by Dr M. He was also of the opinion that the appellant was suffering from a personality disorder at the time of the offence. The disorder was substantially responsible for her lack of memory, albeit in conjunction with her intoxication. Her personality disorder had caused her to pursue an implausible defence.

F.9 The third report, obtained by the Crown, was prepared by Dr J in 2004. It was his opinion that, if the appellant's amnesia was genuine, it resulted from intoxication. She did not have a personality disorder. In any case, denial and memory loss were not particular features of the alleged disorder. Furthermore, a genuine disorder would have manifested itself in similar behaviour in prison. This had not happened. Her alcohol and drug use was solely responsible for her "chaotic lifestyle". There was simply no explanation for the killing.

F.10 The Court declined to admit the new psychiatric evidence, on the grounds that:

(1) the appellant's lack of recollection was not consistent;

(2) denial was not a documented feature of borderline personality disorder;

(3) there were more persuasive explanations for her denial/lack of recollection, namely that she was not telling the truth or that intoxication had caused her lack of recollection.

(4) critically, it had not been established on appeal that the reason the appellant had not run diminished responsibility at trial was integral to her abnormality of mind. There was no reasonable explanation for failing to provide the relevant evidence at trial.

The appeal was therefore dismissed.

COMMENTARY

Introduction

F.11 Personality disorders, post traumatic stress disorder (PTSD) and depression all commonly ground a plea or appeal by women who kill. In theory, such conditions may ground a successful partial defence to murder.^[96] However, there are particular limitations in the substantive law and in the criminal justice system where such conditions are concerned. They arise from a number of features peculiar to the way in which these conditions may manifest themselves. In practice it is particularly difficult to run successfully a partial defence based on such disorders. Some of these difficulties are highlighted by the trial and appeal in S.

At trial

F.12 On appeal, two experts gave a psychiatric explanation for S's claim at trial that she could not remember the killing and that she was not responsible for it.^[97] When put forward at trial, however, a jury of ordinary experience would have understandably looked upon her version of events with scepticism. S's genuine belief in the truth of her own account clearly belies an assumption at the heart of the adversarial trial system, namely that defendants who are fit to plead are also always capable of putting forward a defence which is in their own best interests.

F.13 The problem is complicated by the fact that an underlying personality disorder is highly likely to go unnoticed in the period leading up to the trial. This is so for two main reasons. Firstly, individuals tend to be reluctant to co-operate with efforts to assess their mental state. Dr F, a prison psychotherapist, commented in a report compiled while S was awaiting trial that:

[S] is keen to have some help. She is starting to look at her problems but she is apprehensive in disclosing things about herself for fear that they might be used against her … in court.

F.14 Secondly, personality disorders are latent to an extent that other "abnormalities of mind" are not. This further adds to the likelihood that they will go unnoticed. Only in-depth studies are likely to be able to establish whether or not they are likely to have had any bearing on the defendant's conduct. For this reason Dr B gave evidence to the Court of Appeal that defence counsel were unwise to have relied on the report she prepared with reference to a bail application to discount a defence of diminished responsibility.^[98] The preliminary reports were particularly inadequate to assess whether a defence of diminished responsibility was available because the defendant was denying any memory of, or responsibility for, the offence in question. Dr B, who examined S two days after her conviction, said that he could not offer a psychological or psychiatric explanation for the offence because she was denying that she was guilty.

F.15 Even after some time has passed, the exercise remained difficult. Seven years after the offence, when S had come to accept her responsibility for the killing, Dr M stated that:

[i]t must also be emphasized that in this assessment it was especially apparent that her characteristic coping strategy…is essentially one of denial! This made it all the more difficult…to explore her emotional state and thinking around the time of the index offence.

F.16 Thus a doctor unable to probe for a personality disorder in the limited examination time available prior to trial is likely to report, explicitly or by implication, that there is no relevant condition for the purposes of a diminished responsibility plea. Defence lawyers, in reliance on such preliminary reports, are less likely to pursue this avenue in the limited time available to prepare for trial. This is especially true when the alternative pleas being considered are incompatible with a diminished responsibility based argument.^[99]

F.17 The consequences of not having adequate notice taken of a personality disorder at trial were also reflected in S's sentence. In Dr B's opinion in her second report in 2004:

As part of her disorder she cannot accept responsibility for her behaviour and her projection of blame onto others acquires an almost delusional intensity. Such persons are said to be 'in denial' and psychologically unable to admit to wrongdoing or show appropriate remorse…. [Emphasis added.]

F.18 However, on recommending a tariff of 12 years to the Home Secretary, the trial judge had written:

I would have been minded to recommend less, had she been able to face up to what she had done, and in particular had she not been willing to allege in open court that her son could have committed murder.

On appeal

F.19 Admissibility at the appeal stage is not governed solely by whether or not the evidence would have been admissible (or indeed persuasive) at trial. Rather it is directed by the test set out in section 23 of the Criminal Appeal Act 1968 (the 1968 Act), of which the admissibility of the evidence at trial and its persuasiveness form only a part. Section 23 provides that:

(1) For the purposes of an appeal under this Part of this Act the Court of Appeal may, if they think it necessary or expedient in the interests of justice—

(a) order the production of any document, exhibit or other thing connected with the proceedings, the production of which appears to them to be necessary for the determination of the case;

(b) order any witness who would have been a compellable witness in the proceedings from which the appeal lies to attend for examination and be examined before the Court, whether or not he was called in those proceedings; and

(c) receive any evidence which was not adduced in the proceedings from which the appeal lies.

(2) The Court of Appeal shall, in considering whether to receive any evidence, have regard in particular to—

(a) whether the evidence appears to the Court to be capable of belief;

(b) whether it appears to the Court that the evidence may afford any ground for allowing the appeal;

(c) whether the evidence would have been admissible in the proceedings from which the appeal lies on an issue which is the subject of the appeal; and

(d) whether there is a reasonable explanation for the failure to adduce the evidence in those proceedings. [Emphasis added]

F.20 In S's case, when deciding whether it was "necessary or expedient in the interests of justice" to receive the new evidence, the Court had regard to the presumption that a defendant is only entitled to one trial. They noted the words of Lord Bingham CJ in Campbell, who said:

This Court has repeatedly underlined the necessity for defendants in criminal trials to advance their full defence before the jury and call any necessary evidence at that stage. It is not permissible to advance one defence before a jury and when that has failed, to devise a new defence, perhaps many years later, and then seek to raise the defence on appeal. ^[100]

F.21 To do so would be to "subvert the trial process".^[101] There is therefore a presumption, even where evidence would have been admissible (and indeed persuasive) at trial, that it would not be "in the interests of justice" to admit new evidence. However the Court recognised that, in exceptional circumstances, evidence should nevertheless be admitted. In S's case, the Court considered sections 23(2)(b) and 23(2)(d) to be at issue: whether the new evidence afforded a ground for allowing the appeal and whether there was a reasonable explanation for the failure to adduce the evidence at trial respectively.

F.22 In a technical sense, evidence of diminished responsibility was readily available to S's defence team. She had displayed symptoms of a personality disorder at the preliminary psychiatric examinations which took place before trial. However, her defence lawyers considered but actively rejected diminished responsibility as a "weak" line of argument. In such a situation, the words of Schiemann LJ in Weekes appear to apply:

[T]he defendant must put forward his whole case at trial and…it is not in the interests of justice to permit him to put forward his case with different evidence before different tribunals….^[102]

In sum, the defendant fell foul of the "two trials" principle described above.

F.23 Yet in practical terms there were several significant obstacles to adducing medical evidence to the degree of relevance and persuasiveness necessary to ground a partial defence of diminished responsibility at trial.^[103] It is possible to adopt a purposive approach to section 23(2)(d) and consider that these difficulties constituted a "reasonable explanation for the failure to adduce the evidence". However, the Court of Appeal believed that, in determining whether the test had been met, it was crucial to decide whether the S's amnesia about the circumstances of the killing were in truth a manifestation of her abnormality of mind. The answer to this would be indicative of whether there was a reasonable explanation for the defence not having been run at trial. The Court later noted that the critical feature about the case was that it had not been established that this was the case.

F.24 Consider the implications of the exercise the Court set for itself. Under section 23(2)(a), the Court of Appeal need only satisfy itself that the evidence is "capable of belief". The Court said that this criterion was plainly met. Yet the Court was now saying that, in order to determine whether there was a reasonable explanation for the failure to adduce the new psychiatric evidence at trial under section 23(2)(d), the Court had to decide whether it was persuaded by the thesis of Dr M and Dr B. In other words, because the persuasiveness of the appellant's fresh evidence was a prerequisite for its admissibility under section 23(2)(d), the Court felt compelled to go beyond deciding whether the evidence was capable of belief, and decide which set of evidence it actually preferred. The latter is an exercise ordinarily at the core of the role of the jury. Indeed, in giving its reasoning for preferring the evidence of Dr J on the one hand over that of Dr M and Dr B on the other, the Court could do little more than restate the opinions of Dr J.^[104]

F.25 Two principal problems emerge from this process. Both problems are peculiar to the situation where, because of an allegedly denial-inducing personality disorder, the persuasiveness of the new evidence itself determines whether there is a reasonable explanation for the failure to adduce evidence at trial. The first problem is that, as described above, the Court effectively takes on the role of the jury. As such the evidence is put to a higher standard of proof than that required of it under section 23(2)(a).

F.26 The second problem is that, when deciding on which set of evidence to prefer, the odds are stacked against acceptance of the evidence favourable to the appellant. As described in paragraphs F.13-F.16 above, there are likely to be earlier accounts of the appellant's mental state, suggesting that she has no relevant disorder, even if a later reports suggest the contrary. The very existence of competing reports disinclines the Court to accept the new evidence,^[105] not least because "[e]xpert witnesses, although inevitably varying in standing and experience are inter-changeable in a way in which factual witnesses are not."^[106] Further, when comparing the new psychiatric evidence to the other psychiatric evidence available, the Court considered it relevant that S was raising the defence of diminished responsibility for the first time years after the trial, and that the evidence sought to be admitted was both challenged and controversial.^[107]

F.27 The general evidential assumption (that evidence obtained long after the time of the offence is unpersuasive) belies the thesis that personality disorders and PTSD are only ever likely to emerge after a significant period. As a result, contemporaneous reports are in fact less likely to be representative of the defendant's mental state at the time of the offence. Nevertheless, the general principle might explain why the Court in S preferred the evidence of the single psychiatrist, Dr J, over that of the two psychiatrists whose mutually corroborating reports supported the case of the appellant.

THE EFFECT OF OUR PROPOSALS

Abolition of all partial defences^[108]

F.28 Although we are not making a provisional proposal to this effect, abolishing all partial defences (with mitigating factors going to sentence) would avoid some of the problems described above insofar as they stem from the adversarial nature of the current proceedings. In particular, defence teams would no longer have to choose between incompatible defences and so diminished responsibility would still be available for consideration after conviction. Indeed, because the extent of a personality disorder or PTSD is only likely to emerge once there has been time for reflection and further psychiatric examination, it is at this point that certain offenders would derive most benefit from consideration of the psychiatric evidence. Although such a system would ostensibly lead to a lengthier and costlier sentencing process, it would in turn reduce the need for appeals against conviction to the Court of Appeal such as that in the case discussed above.

F.29 There is, of course, a significant disadvantage to this approach: a killer whose mental responsibility was substantially impaired would nonetheless be labelled a "first degree murderer". However, as has been demonstrated above, certain offenders face formidable obstacles to obtaining an appropriate, lesser conviction under the present system. From a pragmatic perspective, it would be a significant improvement if such offenders were at least to receive the benefit of an appropriate tariff.

Provocation

F.30 In the above case, Dr M was of the opinion that S perceived the victim's objectively innocuous behaviour in the period immediately leading up to the attack as provocative because of her intoxication with a combination of drugs. He commented that:

[B]y the time he had shown what she perceived as deliberate defiance and aggravating insults she had lost her temper. She remembered thinking, "I can't stand it any more: he's doing this on purpose – he's gone too far!

F.31 Thus there was evidence of a loss of self-control. On the law of provocation at the time of trial, S would have further had to demonstrate, in essence, that her response was partially "excusable". (Although there were competing lines of authority as to the content of the provocation test at this time, the test was later to be formulated as one of 'excusability' by Lord Hoffmann in Smith (Morgan).^[109]) Following Holley^[110] she would have faced an additional hurdle in the form of the 'tightened' objective limb of the provocation test.

F.32 Our proposed reformulation of provocation^[111] is designed to address the paradigm case of domestic violence. It is notable that in S's case there was no evidence of a fear of serious violence or of gross provocation. Arguably the unavailability of the provocation defence is appropriate, because the victim's behaviour towards the defendant, although far from blameless over the longer term, was not in fact a major motivating factor in the offence.^[112]

Diminished responsibility^[113]

F.33 In S's case, Dr M noted that her "severe personality disorder" was due to "inherent causes as well as contributions from adverse aspects of her development…" He noted that:

Since adolescence she was virtually constantly intoxicated with a variety of substances…. She did not allow herself the opportunity of any periods of sustained sobriety and thus she failed to mature and develop through normal experience…there is much to indicate that with enforced abstinence [in prison] there has been a most impressive maturation of her personality.

F.34 Furthermore, Dr M drew attention to the psychiatric effects of the combination of years of sexual abuse suffered at the hands of the deceased and other men on the one hand, with a recent violent rape experienced by the defendant on the other hand. He was of the opinion that:

Undoubtedly, there must have been some generalising effects: thus she had become all the more sensitised over the inevitably abusive relationships she suffered from men…

F.35 The reformulated defence would require the jury to take into account all mental disorders, regardless of their origins. Here, the psychiatrist was of the opinion that environmental factors caused S to develop an abnormality of mental functioning in the months prior to the commission of the offence. On our reformulated defence, the resulting abnormality could be taken into account.

CONCLUSION

F.36 Under our proposals, S would have no defence of provocation available to her. We believe this is the right result, because her reduced culpability (if any) is best explained by reference to the abnormality of her mental functioning. To this end, our reformulated diminished responsibility defence would allow a wider range of contributory factors to be considered. It recognises that a killer's motivation is inevitably multifactorial, and allows for a more coherent, and therefore more believable, account of the contributing mental factors.^[114]

F.37 That said, however, the effectiveness of the defence remains dependent on the defendant's ability to admit to having perpetrated the killing with the requisite intent, which may well be absent at trial. We are of the opinion that reform of the substantive law can only go part way to ameliorating the particular difficulties faced by defendants in diminished responsibility cases such S, because any reformulated defence would continue to have to operate in an adversarial context. However we believe there remains scope for improved collection, presentation and use of psychiatric evidence in the courts within the adversarial system.^[115] We welcome suggestions as to how this may be achieved.

APPENDIX G

BACKGROUND INFORMATION ABOUT

MURDER AND HOMICIDE

INTRODUCTION

G.1 There is no doubt that homicide, and murder in particular, looms large in public and media debates about crime. For example, on 9^th November 2005, in its search facility www.bbc.co.uk listed 11,292 entries for news items under 'murder' and 943 under 'manslaughter'. That easily exceeds the combined total for the number of news items on robbery (2,237), rape (2,588) and burglary (1,084).

G.2 The devastating impact that murder, and manslaughter, can have on victims' families has led to the establishment of flourishing support groups for these 'secondary' victims. Important examples are Support after Murder and Manslaughter (SAMM), and the Victims of Crime Trust. Rape aside, no other crime has warranted quite this level of sustained, specialised attention amongst voluntary groups.

G.3 The importance of murder is also reflected in the way it is investigated, prosecuted and tried.

G.4 Within its Specialist Crime Directorate, the Metropolitan Police has, for example, a Homicide Command dedicated to the investigation of homicides. It is divided into three units (West; South; East), each unit being led by a Detective Chief Inspector, with 33 dedicated staff. The units are supported by, for example, a Murder Review Group, Coroner's officers, and pathologists.

G.5 Turning to prosecution, murder cases are usually handled by the Crown Prosecution Service in the area in which the case arises. The Chief Prosecutor in the area must be notified of the case, so that he or she can either deal with it personally, or ensure that someone else with sufficient experience and seniority deals with it. However, cases involving special difficulty or controversy, such as terrorist cases, suspicious deaths in custody, or possible euthanasia cases will be referred to CPS Headquarters Casework Directorate.

G.6 On request, the CPS will meet the family of someone killed through criminal activity, to explain the decisions taken about prosecution. Any decision to drop or to substantially alter a charge is communicated to the victim's family, alongwith reasons for the decision.

G.7 Murder cases may only be tried by specially chosen and trained judges. They are `ticketed' to try murder cases, following the recommendation of the presiding judge on the relevant circuit. In England and Wales, there are 86 circuit judges, out of 628 in total, who are ticketed to try murder cases.

G.8 Murder is a rare crime in England and Wales. In 2003-2004, there were 1,109,017 violent crimes recorded by the police, but only 0.1% of these involved homicide (853 deaths). Further, only about a third of these homicides result in convictions for murder (277 such convictions in 2003). The rest range from causing death by dangerous driving to manslaughter.

IS SOCIETY NOW SOFTER, OR TOUGHER ON MURDERERS?

G.9 Homicide has, in common with some other forms of violent crime, been on the increase over the last forty years. In 1965, the year that the death penalty for murder was finally abolished, 58 people went to prison under a mandatory life sentence. By 2003, this figure had risen to 277. The numbers of those going to prison for murder roughly doubled between the late 1960s and the late 1970s, and doubled again between the late 1970s and the mid 1990s, since when the rate of increase has slowed. It should not be forgotten that there are also in the region of 80-90 convictions for attempted murder in any given year. Many of these may have been cases in which the victim only survived by luck or the skill of the surgeon.

G.10 In recent times, governments have become ever tougher on convicted murderers, in terms of the time that they spend in custody. When most people conjure an image of an England and Wales with the death penalty for murder, they probably think back to notorious mid-twentieth century cases, such as those that ended in the executions of Ruth Ellis, Derek Bentley, and others. They thus come to imagine a society that kept the number of murders lower by punishing that crime with the ultimate penalty. In fact, the death penalty was at that time as often as not turned into one of life imprisonment through the exercise of the prerogative of mercy (this was done in 90% of cases involving women offenders).

G.11 Furthermore, less well known is the fact that, at that time, once the penalty of death had been turned into one of life imprisonment, offenders were treated with surprising leniency.

G.12 As pointed out above, in 1965, the year of the abolition of the death penalty, 58 murderers were given the mandatory life penalty. 25 of these offenders - nearly half - had been released from prison on licence within a ten-year period from the date of conviction. 22 more had been released on licence 10-15 years after conviction. Only seven served more than fifteen years behind bars.

G.13 Contrast that statistic with the figures twenty years later. In 1985, of the 173 murderers who received the mandatory penalty, only 6 served less than 10 years behind bars. By way of contrast, 56 served more than 10 years before release, and the rest had not been released by 2002 (when these figures were compiled).

G.14 Moreover, that period (the late 1960s to the late 1980s), saw a steady fall in the number of cases in which the Parole Board was prepared to recommend release. Parole was recommended in 50% of all lifer cases in 1968, but was only recommended in roughly 20% of mandatory lifer cases in 2003/4.

G.15 By the year 2003, there were nearly 2,700 prisoners, in all, serving mandatory life sentences for murder in England and Wales, whereas there had been only 133 lifers in 1957. The increase is mainly due to an increasingly less indulgent answer given by all branches of government - Parliament, the executive, and the judiciary -to the question of how long the custodial element of a life sentence should be.

G.16 There are still a small number of murderers who are released early (at or about the 10 year mark). We conducted a small study of a tranche of those convicted between 1994 and 1996, whose tariffs were set at ten years or fewer, to see what the circumstances may have been that led to their early release (see Appendix E).

G.17 Predominantly, cases in which early release occurred involved some element of provocation from the victim, albeit provocation of a gravity insufficient to persuade the jury to acquit of murder and convict of manslaughter instead. After provocation, the most common mitigating factors were evidence of mental disorder, and evidence that the offender was in some way seeking to defend him or herself from attack but went too far in killing the victim. A significant number of such cases involved a defendant who was immature for their age at the time of the offence, even if they had no specific defence to murder.

G.18 The abolition of the death penalty means that the length of time in custody under a life sentence has, for many people, effectively become the only benchmark for judging how seriously the crime is being taken. In that regard, as indicated above, there can be no denying that all branches of government have become progressively more punitive in their treatment of murderers.

G.19 Previously, the mere fact that the death penalty had to be passed in almost all cases came to symbolise how seriously the crime was taken. The actual length of time spent in custody, when the death penalty was commuted, seems to have taken second place. The abandonment of that older standpoint must be a welcome development, whatever one's view on the mandatory penalty, or on the length of time that should be served in prison for murder.

WHO TYPICALLY COMMITS, AND IS A VICTIM OF, MURDER?

G.20 The typical perpetrator of homicide is white (78% of cases), male (90% of cases), and between 21 and 40 years of age (63% of cases). He will have killed a spouse, lover, relative, offspring or acquaintance (70% of cases), using either a sharp or blunt instrument (50% of cases).

G.21 Whilst these percentages are almost all very similar to those, for example, in the USA, there is one notable exception. In the USA, 67% of all murders (in 2003) involved the use of firearms, compared with a figure of between roughly 5% and 10% in recent years in England and Wales.

G.22 The killing will typically have been of a single victim (98% of cases), and will probably have taken place while the offender was in a violent rage, or during a quarrel (54% of cases). In 72% of cases, the offender will have a criminal record, probably with a violent or sexual element to the crime(s) (50% of cases).

G.23 The typical victim of homicide is also white (78% of cases), and male (68% of cases), although the proportion of female victims (32% of cases) is much higher than the proportion of female perpetrators (10% of cases). The victim may well be a little older, on average, than the perpetrator of homicide. The victim is over 30 years of age in 62% of cases. Only 2% of perpetrators are over 60 years of age, whereas 12% of victims are over that age.

G.24 It would not be wrong, then, to think of homicide as an event typically arising out of quarrels or feuds between adult men in the prime of life, that have led to a loss of temper, probably on both sides, and hence to the use of violence.

G.25 We must be careful in drawing conclusions, however, because these figures relate to 'homicide', meaning principally murder and manslaughter, and are not confined to murder cases.

G.26 There are also further nuances in the statistics. One is the much higher proportion of women charged with murder where the victim was their partner (35% of women defendants), as compared with the proportion of men in that situation (19% of male defendants).

WHEN PEOPLE ARE CHARGED WITH MURDER, WHAT DO THEY TYPICALLY PLEAD?

G.27 At least half of all murder trials are taken up with claims that the police have got the wrong person. In another 10-15% of cases, there is a guilty plea, or no apparent defence.

G.28 When it comes to defences currently related to the substantive law, our earlier research divided the figures between men and women defendants (the figures below are first those for men, followed by those for women).

G.29 The most popular specific defence is lack of intent (25%; 29% of cases). It will be important to bear this in mind, if the mental element in murder is changed. After 'lack of intent', the most popular plea is provocation (18% in both cases), followed by self-defence (14%; 10% of cases) and diminished responsibility (8%; 14% of cases).

APPENDIX H ALTERNATIVE VERDICTS IN HOMICIDE CASES

H.1 What is the position where a jury is divided on its verdict in a homicide trial? This may occur either because some wish to convict of murder and some of a lesser form of homicide, or because they all opt for a lesser form of homicide, but for different reasons. It is an issue on which the law must speak with clarity if, following reform of the law of murder and homicide, more verdicts to choose between become available to the jury.

H.2 At present, the issue may arise when some members of the jury wish to convict of murder and some of manslaughter, or because they all opt for manslaughter but for different reasons. This is, then, the context in which our discussion is placed.

DIFFERENT FORMS OF MANSLAUGHTER

H.3 This topic is discussed extensively by Professor Richard Taylor,^[116] who analyses three Court of Appeal cases.^[117] H.4 The general principle appears to be that it does not matter that the jury disagrees on some of the facts, provided that the facts on which they do agree are sufficient to justify a verdict of guilty. These facts can include a set of alternatives as long as it would be possible for a person to believe that the alternatives are exhaustive and that the members of the jury are unanimous (or believe by the required majority) that this is so.

H.5 Suppose that the jury members agree that the defendant deliberately stabbed the victim, and that the victim died as a result.^[118] Half of them believe that the defendant never intended to inflict a serious injury, but that there was no provocation for the attack. The other half believes that the defendant did intend serious injury, but was provoked. Thus, the first half would convict of involuntary manslaughter of the "unlawful act" variety. The second half would convict of voluntary manslaughter on the ground of provocation.

H.6 In such a case they can arrive at a unanimous verdict of manslaughter, because the facts on which they agree (deliberate stabbing and consequent death) are in themselves sufficient for that verdict, without going into the questions of intent to inflict serious harm and of provocation. They would not differ over a finding of involuntary manslaughter. They would only differ on their reasons for not finding murder.

H.7 Suppose, now, that the victim had marks of two different types of injury, and the circumstances are such that the defendant must have inflicted both of them, but it is uncertain which was the cause of death. Again the jury may be evenly divided between the two explanations, but can return a unanimous verdict of manslaughter, because they are unanimously of the opinion that the defendant must have caused the death by an unlawful act one way or the other.

H.8 Suppose, however, that the circumstances are such that the defendant must have inflicted one or other of the two injuries found on the victim, but not necessarily both. Suppose that half the jury believes that the defendant inflicted only the first injury, and that that was the cause of death. The other half believes that the defendant inflicted only the second injury and that that was the cause of death. In this case we have two completely inconsistent narratives, even though each, taken by itself, would result in a finding of guilty of manslaughter by unlawful act. They do not add up to an agreed certainty, as the questions of who inflicted the injury and of which injury caused the death are logically independent of each other. There is simply no evidential basis on which a person could conclude "It is possible that the defendant inflicted injury A and that the victim died of injury A, or that the defendant inflicted injury B and that the victim died of injury B, but not that the defendant inflicted injury A and that the victim died of injury B or vice versa".

H.9 This means that it is usually improper to convict where half the jury believes that there is an unlawful act (or provocation) and the other half believes that there is gross negligence. Suppose that it is certain that the defendant died as the result of a negligently performed operation. Half the jury believes that the victim never consented to the operation but that the negligence was not gross. The other half believes that the victim consented but that the negligence was gross. Again these do not add up to a certainty of manslaughter, because the two issues are logically independent: there is no basis to conclude "I think the operation was done without consent, but if there was consent then it was grossly negligent".

H.10 There could however be a case in which the defendant performed two acts, one unlawful and the other grossly negligent, and the jury only differs on which caused the death.^[119] Here the jury can convict of manslaughter, as both the unlawfulness and the gross negligence have been found unanimously, and one can logically conclude that one or other must have caused the death, so that one way or the other the defendant must have been to blame.

H.11 In summary it is not sufficient that there is unanimity of outcome. The jury must agree (unanimously or by the required majority) on a set of propositions, which may include one or more sets of exhaustive alternatives. However, this set of propositions must be capable of being consistently held by a single individual, and must be sufficient to establish the offence.

H.12 The above discussion concerns the sort of direction that ought to be given to the jury if they of their own motion explain their division of opinion and seek guidance about what verdicts are available to them. Further questions are:

(1) how far the court should give directions in this sort of detail even if the jury does not request them;

(2) whether a unanimous verdict of manslaughter can be impugned on the ground that the court did not ask the jury by which of the possible alternative routes they arrived at it.

In other words, must the court actively look for ambiguity, either in summing up or following the verdict?

H.13 In Jones it was held that, where a verdict of manslaughter is returned as an alternative to murder, and therefore the only issue can be provocation versus lack of intent to kill or do serious harm, the court is under no obligation to enquire about the basis of the verdict. It may do so if it wishes as an aid to sentencing, but only if the judge warned the jury in the course of summing up that he or she intended to do so.

H.14 It remains uncertain what the position would be in a case like that in paragraph

H.8 or paragraph H.9 above, where the two alternative possible bases for a manslaughter verdict do not add up to a certainty. It seems to us that:

(1) the judge in summing up should explain the two alternatives carefully, and direct that the jury should not convict unless it can agree unanimously (or by the required majority) on one of them; but

(2) following such a verdict, there is no obligation on the judge to ask which of the two bases the jury convicted on; though once more, he or she may do so as an aid to sentencing provided that due warning has been given in the summing up.

H.15 As stated, problems are more likely to arise when the jury is divided between unlawful act (or provocation) manslaughter and gross negligence than when it is divided between two forms of unlawful act, or between unlawful act and provocation. This however is a matter of degree, and it is possible to devise exceptional cases on both sides of the line. The only reason is that instances of unlawful act or provocation manslaughter are likely to have more factual features in common with each other than with instances of negligent manslaughter: it is not because negligent manslaughter is in any way treated as a different offence.

H.16 For as long as manslaughter continues to be treated as a single offence which may take various factual forms, it seems unlikely that there is anything unsatisfactory in the law as explained above. Any complexities follow from the logic of the factual situation rather than from irrationalities in the law as such.

MURDER AND MANSLAUGHTER

H.17 A more difficult situation arises when the jury is divided between those who wish to convict of murder and those who wish to convict of manslaughter. Again the position may vary according to whether the manslaughter alternative takes the form of provocation or "unlawful act" on the one hand or gross negligence on the other.

Provocation/unlawful act

H.18 Logic would indicate that, in this situation, the jury should be able to convict of manslaughter. All are agreed that there was an unlawful act causing death: the only issue is whether there is an element present that would make it murder. There are two possible counter-arguments to this, one of substance and one of procedure.

Does manslaughter exclude murder?

H.19 The argument of substance is that there is no unanimous finding of manslaughter, because it is part of the definition of manslaughter that it does not amount to murder. It is quite true that textbooks often seem to talk in this way, but this is for the sake of brevity, to distinguish murder from mere manslaughter.

H.20 If the issue of provocation arises, and provocation is not disproved beyond reasonable doubt, the jury can certainly return a verdict of manslaughter without actual proof of provocation. Similarly if the issue of intent to cause death or serious bodily harm, arises, and that intent is not proved beyond reasonable doubt, the jury can certainly return a verdict of manslaughter, without the need for intent to be disproved. Thus "not-murder" is descriptive of the kind of situation in which manslaughter verdicts are needed, but it is not a formal ingredient of the offence of manslaughter to be proved beyond reasonable doubt.

H.21 In other words a verdict of manslaughter does not require certainty that the killing did not amount to murder. It would be absurd if the jury had to acquit a defendant entirely because it was not sure whether to convict of murder or manslaughter. Murder implies manslaughter, in the same way that causing grievous bodily harm with intent to do grievous bodily harm, contrary to section 18 of the Offences Against the Person Act 1861, implies the malicious infliction of grievous bodily harm, contrary to section 20 of the Act.

Separate verdict on murder?

H.22 The argument of procedure is that a jury divided between two opinions held with certainty is not the same as a jury in doubt. A unanimous verdict (or the required majority) is required for an acquittal as much as for a conviction. Thus to allow the jury to convict of manslaughter in this situation is wrong because it lets the defendant off the hook on the charge of murder. In fact the jury was divided on the question of murder, and the defendant ought to be re-tried.

H.23 The difficulty with this proposition is that, if the defendant is re-tried and the second jury is also divided between murder and manslaughter, the defendant will then go free on all charges. Requiring a second trial instead of accepting a manslaughter verdict looks like a bet of "double or quits".

H.24 One possibility would be for the jury (at the first trial) to return a verdict of manslaughter and at the same time report that they were unable to agree on murder. It is true that there would then be a major problem about whether it would be proper to order a re-trial on the issue of murder alone, or whether, although murder and manslaughter are technically separate offences, this would in effect amount to trying the defendant twice for the same crime. As against this, there can be no doubt but that such a verdict would be available if the indictment contained a count of manslaughter in addition to one of murder, though this would give rise to the identical problem about re-trials.

H.25 Logically it should be possible for the jury to return such a verdict (guilty of manslaughter; disagreed on murder) even if the indictment is for murder alone, as every charge of murder necessarily contains a charge of manslaughter. This, however, is only as a matter of theoretical law. In practice the courts insist on unanimity of outcome.

H.26 Judge Clarke^[120][121] points out that a similar problem arises in the converse situation, where the jury has unanimously rejected murder but is divided between manslaughter and acquittal. Here too, if the jury simply reported disagreement the defendant would be re-tried, and possibly convicted of murder by the second jury. This could be avoided if the indictment contained a manslaughter charge and the jury acquitted of murder and reported disagreement on manslaughter. He suggests that, in any case where on the facts this kind of disagreement seems likely, the indictment should be amended to add a manslaughter charge. Such a solution could be adapted to the present problem.

H.27 Returning to the case where the jury is divided between murder and manslaughter, three situations are possible:

(1) The members who wish to convict of murder are persuaded by the others to back down, on the reasoning that they will not get a verdict of murder anyway and that a verdict of manslaughter is better than a re-trial: the division of opinion then never comes to light.

(2) The jury reports that they are unable to agree on a verdict, and there is a re-trial on everything.

(3) Judge Clarke's suggestion (or rather its mirror image) is adopted, a count of manslaughter is added to the indictment, and the jury convicts on that and reports disagreement on murder.

H.28 If this last suggestion is followed, the court will then have to decide whether it is proper to order a re-trial on the issue of murder. If there is a re-trial, presumably the first court will defer sentencing on the manslaughter count until the murder trial is over, to avoid double sentencing.

Recklessness/gross negligence

H.29 In some cases there is a clear choice between murder and negligent manslaughter: for example, where the defendant left a well uncovered by night and the jury is divided on whether this was purely reckless or set as a deliberate trap. Here it would be possible for one person to conclude that this action is so unusual in itself that the defendant must either have intended harm or not have cared. The situation will then be the same as in the unlawful act cases. It ought theoretically to be possible to return a verdict of manslaughter while reporting disagreement on murder, but this may require the addition of a manslaughter count.

H.30 In other cases, however, the grounds for finding murder and the grounds for finding gross negligence will be so different that it would be impossible for the same person to believe that, while neither is certain in itself, one or the other must have occurred. The situation is then analogous to that of the negligent operation (paragraph H.9 above), and the jury will have to report disagreement, necessitating a re-trial.

Conclusion

H.31 The problem is not serious enough to require a statutory solution, unless it turns out that the same thing often happens in cases other than homicide where the jury is divided between a serious offence and a lesser offence contained within it. In the case of wounding and grievous bodily harm, Judge Clarke reports that the usual practice is to charge both the section 18 and the section 20 offence. We are not aware of a practice where a jury convicts of the section 20 offence and reports disagreement on the section 18 offence: that is, whether the court and the prosecution must accept the situation or there can be a re-trial on the section 18 offence alone. There may be many other examples of lesser offences contained within more serious offences where the same difficulty could occur.

THE EFFECT OF CREATING MORE HOMICIDE OFFENCES

H.32 A key question is whether this problem would be aggravated if more homicide offences were created as a result of proposals such as those in the main body of this Consultation Paper. Depending on how the new offences are defined, the problem could occur in two different forms:

(1) In some cases, two offences may "nest". That is, the more serious offence is defined as the less serious offence plus some aggravating circumstance or state of mind, and every instance of the more serious offence is by definition also an instance of the less serious offence. In that case, the position where the jury is divided between the two offences will be the same as that for murder and manslaughter, as described above. The desired position is that, whether or not the lesser offence is specifically charged, the jury should be able to convict of the lesser offence and report disagreement on the more serious offence.

(2) In other cases, two offences may exist side by side, in such a way that neither implies the other, even if there is some overlap. An example would be if unlawful act manslaughter and negligent manslaughter were made into separate offences. In such a case, even if it is logically possible on the facts to conclude that the defendant must have committed one or the other, it will be impossible to convict of either if the jury is split or undecided between the two.

H.33 The basic point is that it is not sufficient that the jury agrees on a single narrative, and that any alternatives within that narrative are logically exhaustive. The alternatives must also support the same nominal offence. That is, there must be unanimity of outcome as well as unanimity of narrative.

H.34 For this reason, proliferation of offences should be avoided so far as possible, and where there are several related offences they should be designed to nest. For example if it is desired to separate out negligent manslaughter, it should be defined as any culpable killing where the mental element is at least negligence: it should therefore include all instances of unlawful act manslaughter, provoked manslaughter and murder.

H.35 Another possibility would be to enact a general provision that, in defined cases where a court is uncertain whether a defendant has committed greater offence A or lesser offence B, but is certain that he or she must have committed one of them, it may convict the defendant of offence B. This would cover three categories of case:

(1) Where offence A contains the same basic ingredients as offence B, but is differentiated from it by some aggravating circumstance;

(2) Where offence A contains the same factual ingredients as offence B, but is differentiated from it by some more culpable state of mind;

(3) Where offence A is a substantive offence and offence B is that of being an accessory or some other inchoate offence connected to A.

H.36 This last suggestion involves a fiction that offence B has been committed when on the facts it might well not have been. It might also prove extremely difficult to draft. Some variant of it might well be needed for category (3), but that is a separate project. For categories (1) and (2), it would be better to seek this result by careful definition of the offences.

APPENDIX I

PERSONS AND ORGANISATIONS

I.1 We would like to record our gratitude to:

(1) Professor Barry Mitchell, Coventry University, for making available his report on a public survey of murder and mandatory sentencing in homicides;

(2) Professor Claire Finkelstein, University of Pennsylvania Law School; Professor Winifred H Holland, University of Western Ontario; Professor Ian Leader-Elliott, University of Adelaide; Ms Antje Pedain, Magdalene College, University of Cambridge; Professor John Spencer QC, Selwyn College, University of Cambridge; and Dr Victor Tadros of the University of Edinburgh, for providing us with studies of the law of other jurisdictions;

(3) the High Court and Circuit judges who completed a question paper on options for reform of the fault element of murder;

(4) Nicolette Movick and Simon Jeal of the Crown Prosecution Service for their help in enabling us to carry out a survey of Crown Prosecutors;

(5) the Crown Prosecutors who took the time and trouble to complete the questionnaire "Murder Project: Prosecution Decisions in Homicide Cases";

(6) the National Offender and Management Lifer Review and Recall Team for their assistance while we accessed files held by the Team;

(7) Harriet Wistrich of Birnberg Peirce Solicitors who enabled us to complete the case study in Appendix F.

I.2 In addition, during the months leading up to the publication of the Consultation Paper, the Law Commission's Criminal Law Team prepared a considerable number of draft papers. These were circulated to a number of individuals and organisations for comment. Their responses have been invaluable in the preparation of this paper. We thank:

(1) Age Concern;

(2) His Honour Judge Broderick;

(3) Dr Madelyn Hicks, Consultant Psychiatrist and Honorary Lecturer at King's College, University of London;

(4) Justice for Women;

(5) Kids Company;

(6) Professor Ronnie Mackay, De Montfort University;

(7) Dr B Mahendra, Barrister and Consultant Psychiatrist;

(8) Nacro;

(9) Professor Alan Norrie, King's College, University of London;

(10) The NSPCC;

(11) Reverend Oliver O'Donovan, FBA, Regius Professor of Moral and Pastoral Theology and Canon of Christ Church, University of Oxford;

(12) Rights of Women;

(13) Professor A P Simester, University of Nottingham;

(14) Professor G R Sullivan, University of Durham;

(15) Victim Support;

(16) Dr Eileen Vizard, Consultant Child and Adolescent Psychiatrist and Honorary Senior Lecturer of University College London;

(17) Professor William Wilson, Queen Mary, University of London.

I.3 We thank the members of the Advisory Group to the Criminal Law Team for making themselves available for meetings and for commenting on the draft papers referred to in paragraph I.2 above:

(1) Commander David Armand, Metropolitan Police;

(2) Professor Andrew Ashworth QC, All Souls College, University of Oxford;

(3) Lord Justice Auld;

(4) Mr Alphege Bell, 2-4 Tudor Street Chambers;

(5) Mr Colin Chapman, Crown Prosecution Service;

(6) Professor Ian Dennis, University College London;

(7) Professor Nigel Eastman, Barrister and Forensic Psychiatrist, St George's Hospital, London;

(8) Mr Anthony Edwards, T V Edwards Solicitors;

(9) Ms Deborah Grice, Home Office;

(10) Mr Bruce Holder QC, 6 King's Bench Walk;

(11) Mr Jaswant Narwal, Serious Crimes Unit, Old Bailey;

(12) Professor David Ormerod, University of Leeds;

(13) Mrs Nicky Padfield, Fitzwilliam College, University of Cambridge;

(14) Mr David Perry, 6 King's Bench Walk;

(15) Mr Justice Pitchers;

(16) Mr Edward Rees QC, Doughty Street Chambers;

(17) His Honour Judge Jeremy Roberts QC;

(18) Professor John Spencer QC, Selwyn College, University of Cambridge.

I.4 We are grateful to the following for writing to us with suggestions for changes to the law:

(1) David Martin;

(2) Peter Waugh;

(3) M J Devaney;

(4) Professor Katherine O'Donovan, Queen Mary, University of London.

Ý
Ü  

Note 1   I should like to formally record my thanks for the assistance and advice received in carrying out this survey. Natcen in London, and especially Lucy Dillon, offered some very constructive advice on the methodology, and in particular the way in which the issues were examined in the group meetings. PH Research in Oldham, especially Janet Ralphs, provided considerable help in setting up and organising the group meetings which ran very smoothly. Of course, the participants in the groups gave up their time to consider the issues and express their opinions. Finally, my employer, Coventry University has enabled me to organise my time in order to undertake this survey.    [Back]

Note 2   One person was unable to attend the second meeting.    [Back]

Note 3   This is apparent from previous public surveys carried out by the author; see, for example, Barry Mitchell, “Public Perceptions of Homicide and Criminal Justice”, British Journal of Criminology (1998) 38(3) 453-472.    [Back]

Note 4   Those who rejected the suggestion that killing someone in a vulnerable occupation did so on the ground that all human life should be treated as of equal value.    [Back]

Note 5   At some stage in either or both meetings, each group spent some time emphasising the importance of premeditation in the sense of an intent to kill. Some participants in groups 3 and 4 in particular talked of cases where the offender set out armed to kill as deserving a mandatory sentence.    [Back]

Note 6   At least two participants in the third group cited murder in the course of rape as especially serious.    [Back]

Note 7   An interesting analysis of the Russian-roulette killer’s culpability can be found in William Wilson, “Murder and the Structure of Homicide”, in Andrew Ashworth & Barry Mitchell (eds) Rethinking English Homicide Law (2000) Oxford University Press, at 41, 42.    [Back]

Note 8   See, for example, “Murder, as every practitioner of the law knows, though often described as one of the utmost heinousness, is not in fact necessarily so, but consists in a hole bundle of offences of vastly differing degrees of culpability, ranging from brutal, cynical and repeated offences like the so-called Moors murders to the almost venial, if objectively immoral “mercy killing” of a beloved partner”, R v Howe [1987] 2 WLR 568, 581 per Lord Hailsham.    [Back]

Note 9   It was clear from the group discussions that although the issue was not addressed in depth, participants favoured some separation of offences – e.g. “I’d call that manslaughter, not murder”.    [Back]

Note 10   In a survey carried out between August and October 2003, the author reported that 39 of 62 (62.9%) of respondents would not advocate mandatory sentencing in murder cases; see Law Commission Law Com No 290, Partial Defences to Murder (2004) London: TSO, Appendix C, para 68.    [Back]

Note 11   In each case a room was hired so that the participants would not be disturbed but could focus their attention on the issues.    [Back]

Note 12   It was felt that this would provide sufficient opportunity or participants to reflect on the first discussions whilst simultaneously maintaining continuity between the two meetings.    [Back]

Note 13   It was felt that two hours was the maximum time that participants could reasonably be expected to engage in each discussion. Inevitably, the amount of time available effectively limited the range of issues which could be addressed.    [Back]

Note 14   Definition of current law: “Murder may be committed when a killing stemmed from either an intention to kill or from an intention to inflict grievous bodily harm.”    [Back]

Note 15   A-G for Jersey v Holley [2005] UKPC 23, [2005] 3 WLR 29. See also Mohammed [2005] EWCA Crim 1880.    [Back]

Note 16   See inter alia Broadly v HMA 1991 SCCR 416.    [Back]

Note 17   Drury v HMA 2001 SCCR 583, 591 per Lord Rodger.    [Back]

Note 18   MacDonald’s Criminal Law (5th ed) 89 and subject to judicial approval on countless occasions. The word “wickedly” in square brackets was, however, inserted by Drury v HMA. 2001 SCCR 583.    [Back]

Note 19   Cawthorne v HMA 1968 JC 32.    [Back]

Note 20   Ibid, 38 per Lord Cameron.    [Back]

Note 21   Ibid, where no criticism was made on appeal.    [Back]

Note 22   Arthur v HMA 2002 SCCR 796.    [Back]

Note 23   See Halliday v HMA 1998 SCCR 509, where D’s decision to wash his clothes rather than call an ambulance after a vicious attack was found to be evidence of indifference. Arguably, if evidence of indifference is relevant then, by logic, the lack of such evidence must be too.    [Back]

Note 24   McKinnon v HMA 2003 SCCR 224.    [Back]

Note 25   Brown v HMA 1993 SSCR 382.    [Back]

Note 26   Drury v HMA 2001 SCCR 583.    [Back]

Note 27   HMA v Galbraith 2001 SCCR 551.    [Back]

Note 28   Brennan v HMA 1977 JC 38. It appears that this is for policy reasons.    [Back]

Note 29   Code Penal, art 132-18.    [Back]

Note 30   Ibid, 132-29. However, CP art 132-30 stipulates that this option is not available if the defendant has been given a custodial sentence in the last 5 years.    [Back]

Note 31   I> Ibid, art 221-1.    [Back]

Note 32   Ibid, arts 221-2 and 221-4.    [Back]

Note 33   Ibid, art 221-3.    [Back]

Note 34   Ibid, art 122-5.    [Back]

Note 35   Ibid, art 122-6.    [Back]

Note 36   Ibid, art 122-1.    [Back]

Note 37   German Penal Code, s.212.    [Back]

Note 38   Ibid, s 227.    [Back]

Note 39   Ibid, s 211(2).    [Back]

Note 40   Ibid, s 211(1).    [Back]

Note 41   Ibid, s 212.    [Back]

Note 42   Ibid, s 213.    [Back]

Note 43   Ibid, s 216.    [Back]

Note 44   Ibid, s 20.    [Back]

Note 45   Ibid, s 21.    [Back]

Note 46   Ibid, s 49.    [Back]

Note 47   Ibid, s 33.    [Back]

Note 48   Ibid, s 35.    [Back]

Note 49   Hereafter “the Code”.    [Back]

Note 50   The “Charter” being the Canadian Charter of Rights and Freedoms.    [Back]

Note 51   Code, s 222(5).    [Back]

Note 52   Code, s 229.    [Back]

Note 53   Martineau [1990] 2 SCR 633.    [Back]

Note 54   Latimer [2001] 1 SCR 3.    [Back]

Note 55   Dove (2004) 187 CCC (3d) 506 (BCCA).    [Back]

Note 56   Code, s 745.    [Back]

Note 57   Code, s 231.    [Back]

Note 58   Code, s 232.    [Back]

Note 59   Thibert [1996] 1 SCR 37.    [Back]

Note 60   Code, s 16.    [Back]

Note 61   Cooper [1980] 1 SCR 1149.    [Back]

Note 62   Ibid.    [Back]

Note 63   Chaulk [1990] 3 SCR 1303.    [Back]

Note 64   Code, s 672.54.    [Back]

Note 65   Stone [1999] 2 SCR 290.    [Back]

Note 66   Section 17.    [Back]

Note 67   Paquette [1977] 2 SCR 189.    [Back]

Note 68   This process has begun, Fraser [2002] NSJ No 400 (Prov Ct) being a lower court authority striking down the part of the list of excluded offences containing robbery.    [Back]

Note 69   The Code s 33 reverses the decision in Daviault [1994] 3 SCR 63.    [Back]

Note 70   See Pintar (1996) 110 CCC (3d) 402 (Ont CA).    [Back]

Note 71   Lavallee [1990] 1 SCR 852.    [Back]

Note 72   Contrast Perrault v R [1971] SCR 196 and Robinson [1996] SCR 683.    [Back]

Note 73   Canadian Penal Code (“CPC”), s 187.    [Back]

Note 74   Ibid, s 194.    [Back]

Note 75   New York Penal Law (“NYPL”), s 125.00.    [Back]

Note 76   CPC, s 188.    [Back]

Note 77   Ibid, s 188.    [Back]

Note 78   Californian Jury Instructions, s 8.11.    [Back]

Note 79   CPC, s 189.    [Back]

Note 80   Ibid, s 190.    [Back]

Note 81   NYPL, s 125.    [Back]

Note 82   Ibid, 15.05.    [Back]

Note 83   Although it is arguable that burglary with intent to kill ought not form the basis of felony murder due to the doctrine of merger.    [Back]

Note 84   NYPL, s 125.27.    [Back]

Note 85   CPC, s 189.5.    [Back]

Note 86   Ibid, s 196.    [Back]

Note 87   This will only reduce the charge to voluntary manslaughter, ibid, s 192.    [Back]

Note 88   Ibid, s 197.    [Back]

Note 89   NYPL, s 125.25.    [Back]

Note 90   Crabbe (1985) 156 CLR 464.    [Back]

Note 91   Boughey (1986) 161 CLR 10.    [Back]

Note 92   The separating of conjoined twins seemingly being a narrow exception to this rule: State of Queensland v Nolan [2001] QSC 174.    [Back]

Note 93   Attention should be drawn to the fact that these files did not contain the sentencing judge’s comments and contained some information that came to light post-sentencing. Therefore, studying these files was not necessarily guaranteed to present the same image of the case that the sentencing judge had.    [Back]

Note 94   We are grateful to Harriet Wistrich of Birnberg Peirce Solicitors for access to the case papers when preparing the study.    [Back]

Note 95   For which see the well known facts of Ahluwalia [1992] 4 All ER 889, in which the defendant was severely abused by her husband over a period of several years. She killed him by setting fire to his bedclothes while he was asleep.    [Back]

Note 96   A “confession and avoidance” plea.    [Back]

Note 97   Indeed, the Court of Appeal overturned the murder conviction of the defendant in Devaney [2005] EWCA Crim 944 and substituted one of manslaughter by reason of diminished responsibility, which derived from a personality disorder. It is notable that, in that case, the psychiatric evidence presented in favour of the appellant was uncontested by the Crown.    [Back]

Note 98   See paras F.7-F.8 above.    [Back]

Note 99   See para F.7 above.    [Back]

Note 100   See para F.5 above.    [Back]

Note 101   [1997] 1 Cr App R 199, 204.    [Back]

Note 102   Steven Jones [1997] 1 Cr App R 86, 93, per Lord Bingham CJ.    [Back]

Note 103   [1999] Cr App R 520, 522.    [Back]

Note 104   Described at para F.12-F.15 above.    [Back]

Note 105   Compare paras F.9 and F.10 above.    [Back]

Note 106   Cf Devaney [2005] EWCA Crim 944 in which the uncontested new evidence grounded a substituted verdict of manslaughter on the grounds of diminished responsibility.    [Back]

Note 107   Steven Jones [1997] 1 Cr App R 86, 93, per Lord Bingham CJ.    [Back]

Note 108   The Court’s thinking echoed the words of Lord Taylor CJ in Ahluwalia [1993] Cr App R 133, 142: “…if there is no evidence to support diminished responsibility at the time of trial, this court would view any wholly retrospective medical evidence obtained long after the trial with considerable scepticism.”    [Back]

Note 109   See Part 6.    [Back]

Note 110   [2001] 1 AC 146.    [Back]

Note 111   [2005] UKPC 23; [2005] 3 WLR 29.    [Back]

Note 112   Set out in Part 6.    [Back]

Note 113   Note that a plea of provocation, however formulated, would still be predicated on acceptance of responsibility for death coupled with the requisite intent.    [Back]

Note 114   Our proposed revision of the partial defence of diminished responsibility is also set out in Part 6.    [Back]

Note 115   Note the comment of Dr Madelyn Hicks, quoted at para 6.40 of this paper.    [Back]

Note 116   See the discussion at paras 6.99-6.116 of Part 6.    [Back]

Note 117   “Jury Unanimity in Homicide”, [2001] Crim L R 283.    [Back]

Note 118   Jones, The Times 17 February 1999; Carr [2000] 2 Cr App R 149; Boreham [2000] 1 All ER 307.    [Back]

Note 119   Jones, above.    [Back]

Note 120   Such as A-G’s Reference (No 4 of 1980) (1981) 2 All E R 617.    [Back]

Note 121   Clarke, “Jury Unanimity – A Practitioner’s Problem”, [2001] Crim LR 301.    [Back]