The Law Commission

PART II

STATISTICS AND EMPIRICAL STUDIES

Introduction

2.1     The objective of this Part is to set out relevant statistical data which is currently available and to explain the nature and purpose of the other sources we are tapping into where our work remains in progress. The data, whether already available or currently being obtained, pertains both to the use of the current partial defences to murder and to the public perception of how, if at all, these defences accurately reflect society's needs. Accordingly, this Part will reflect that division. We consider first the use of the partial defences to murder.

Past and current use of the partial defences to murder

2.2     Several sources are relevant to this issue, some of which are complete and some of which comprise work we are undertaking and is yet to be completed. The earlier completed research and the relevant data emanating from it is described and set out immediately below. We then go on to describe the research and surveys currently being undertaken in connection with our project.

Published studies

Dell (1984)

2.3     In 1978 Susanne Dell began a research project to examine trends in the use of diminished responsibility as a partial defence to murder together with attendant sentencing trends. The results of her research were published in 1984.^[1] The research period spanned the fifteen years 1964 – 1979 although the main emphasis was on the years 1966 – 1977. The project only considered homicide convictions for males, not females. Whilst the main focus of the research was on sentencing trends, three of the statistical tables which Dell produced are of relevance for the present terms of reference.^[2] These are reproduced below.^[3]
2.4     Table 1 shows all convictions of males for homicide and is replicated from Home Office statistics. Tables 2 and 3 contain the specific findings made by Dell's research, using only a sample of murder convictions.^[4]

Table 1: Convictions for homicide by men

	Murder	Manslaughter by reason of diminished responsibility	Manslaughter by other means	Total number of homicides	Total number of homicides	Total number of homicides	Total number of homicides
	Number	%	Number	%	Number	%	Number
1964	44	32	31	23	59	44	134
1965	56	34	39	24	68	42	163
1966	70	34	44	21	94	45	208
1967	62	32	37	19	97	49	196
1968	73	35	33	16	102	49	208
1969	74	36	30	15	102	50	206
1970	94	37	55	21	108	42	257
1971	94	37	53	21	107	42	254
1972	76	30	57	23	119	47	252
1973	87	31	60	21	135	48	282
1974	108	34	63	20	150	47	321
1975	102	32	49	15	166	52	317
1976	104	30	70	20	172	50	346
1977	112	38	65	22	117	40	294
1978	101	35	65	23	119	42	285
1979	126	35	72	20	157	44	355

Table 2: Relationship between offender and victim

	1966-9	1970-3	1974-7	Average	Average	Average	Average
	No	%	No	%	No	%	%
Wife[5]	36	38	25	35	36	42	38
Son or daughter	11	11	10	14	4	5	10
Mother	3	3	1	1	6	7	4
Father	3	3	5	7	2	2	4
Stranger	14	15	7	10	11	13	13
Others	29	30	24	33	26	31	31
Totals	96	100	72	100	85	100	100

Table 3: Motives

	1966-9	1970-3	1974-7	Average	Average	Average	Average
	No	%	No	%	No	%	%
Amorous jealousy/possessiveness	15	16	10	14	18	21	17
Explosive reaction to quarrel, reprimand, etc.[6]	18	19	15	21	7	8	16
Sexual	8	8	4	6	3	4	6
Loss of control with child: baby battering	3	3	6	8	2	2	5
In course of theft, burglary, etc.	2	2	4	6	5	6	4
Killing of seriously ill family member	2	2	3	4	5	6	4
Other motives	6	6	3	4	5	6	5
Offender psychotic	36	38	21	29	33	39	35
Motivation unclear	6	6	6	8	7	8	8
Totals	96	100	72	100	85	100	100

Home Office Study (1989)

2.5     A slightly more recent set of statistics were gathered by the Home Office covering the period from 1982 to 1989 and included all convictions for domestic homicide in England and Wales. Usefully this data is subdivided into gender. However, the grouping of manslaughter convictions does not purport to distinguish between provocation and lack of intent as reasons for reducing murder to manslaughter. The tables are reproduced below.^[7]

Table 4: Domestic homicide verdicts – female defendants

	1982	1983	1984	1985	1986	1987	1988	1989	Total
Murder	3	2	1	4	1	7	4	5	27
Section 2 Manslaughter	3	10	4	3	5	2	4	5	36
Other Manslaughter	8	4	11	8	9	9	10	11	70
Lesser Offence	...	...	...	...	3	...	...	1	4
Acquitted / unfit to plead	4	5	6	4	4	6	7	4	40
Total	18	21	22	19	22	24	25	26	177

Table 5: Domestic homicide verdicts – male defendants

	1982	1983	1984	1985	1986	1987	1988	1989	Total
Murder	34	28	29	29	43	26	38	51	278
Section 2 Manslaughter	44	31	31	31	33	22	20	27	239
Other Manslaughter	27	27	26	33	33	29	30	20	225
Lesser Offence	...	2	...	...	3	...	1	3	6
Acquitted / unfit to plead	5	1	9	5	6	6	3	2	37
Total	110	89	95	98	115	83	92	103	785

Home Office Statistics: Crime in England and Wales 2001/2002, Supplementary Volume (January 2003)

2.6     In July 2002 the Home Office published its first report in a new annual series entitled Crime in England and Wales. The supplementary volume published in January 2003 provides further detailed information gathered from police-recorded crimes and the British Crime Survey as well as other sources of crime data. The relevant information has been extracted and reproduced below.
2.7     In 2001/02 there were 858 deaths initially recorded by the police as homicides in England and Wales. Of these, 26 deaths were no longer recorded as homicides leaving a total of 832 (582 male victims and 250 female victims). 72 per cent of female victims, in comparison to 40 per cent of male victims knew the main or only suspect at the time of the offence. Of particular importance is the significant difference between the percentage of homicides committed by male partners/ex-partners compared to that committed by female partners/ex-partners.

Table 6: Percentage of homicide victims killed by partner/ex-partner as currently recorded^[8]

	91	92	93	94	95	96	97	97/ 98	98/ 99	99/ 00	00/ 01	01/ 02
Male Victims	8	9	5	9	9	6	8	8	6	5	5	5
Female Victims	43	45	42	38	42	45	46	48	35	38	44	46

Crime in England and Wales 2002/2003 (July 2003)

2.8     This is the second publication by the Home Office in the Crime in England and Wales series. It too brings together statistics from police-recorded crimes and the British Crime Survey. "Domestic violence" in this Home Office paper is defined as "all violent incidents, except mugging, which involve partners, ex-partners, household members or other relatives".^[9]
2.9     Consistent with the results above, domestic violence is the only category of violence where the risks for women (73 per cent) were higher than for men (27 per cent). By contrast, the risk of violence being perpetrated by strangers was substantially higher for male victims than for female victims.

Table 7: Number of violent incidents against men and women according to British Crime Survey typology of violence (2002/03)

	All (thousands)	% of all incidents	% against men within violence type	% against women within violence type
All Violence	2, 781[10]	100	62	38
Domestic	501	18	27	73
Mugging	388	14	59	41
Stranger	949	34	83	17
Acquaintance	942	34	60	40

Ongoing studies

Nuffield Foundation Project relating to defences of diminished responsibility and provocation

2.10     This project relating to the defences of diminished responsibility and provocation (and those of insanity and cases in which the defendant claims unfitness to plead) is being led by Professor Mackay. It involves extracting data from a sample of cases over a five year period from 1997 to 2001. The sample contains all Crown Court cases in which a defence of diminished responsibility, provocation or insanity was raised or where the defendant claimed unfitness to plead. Amongst the data collected the researchers have access to suitably anonymised psychiatric reports, which will reveal whether there has been a section 2 diagnosis and the type of mental illness diagnosed. It should also show where there has been disagreement between experts as to the diagnosis. The results of Professor Mackay's research are due to become available in early 2004. We are most grateful to the Nuffield Foundation and to Professor Mackay for making this research available to us.

Judges' reports in murder cases

2.11     We have undertaken a survey to collect relevant anonymised details from the judges' reports which are prepared in all cases where the mandatory sentence of life imprisonment has been imposed following a conviction for murder. Approximately 450 such reports were collated, covering the period from October 2000 to June 2003. The material is currently being processed for analysis.
2.12     This survey is intended to provide information on matters relevant to our terms of reference. The use of the existing partial defences to murder will be analysed with particular reference to whether, and to what extent, gender affects which defences are raised, and whether it affects the likelihood of multiple defences being raised. The survey will also seek to elicit information on which defences are pleaded by which gender in cases of killing of a current or former partner.
2.13     Whether there is any correlation between the defence pleaded and the method of killing or the relationship of the accused to the deceased is also to be examined. Further questions probe the issues of the use of psychiatric reports in murder cases and any correlation between the context of the offence and the tariff recommended.

Crown Prosecution Service internal review of homicide

2.14     The Crown Prosecution Service is also currently undertaking, at our request, an internal review of their homicide caseload. The sample they are surveying comprises approximately 300 cases spanning the period from July 2000 to June 2003. The survey will be considered in tandem with our judges' reports survey. The CPS survey is of those cases where one or both of the partial defences have been raised, whether the plea was successful or not. The CPS is collating data on issues such as the relationship of the accused to the deceased, their respective gender, method of killing, context of the crime, defences pleaded and psychiatric or other medical evidence adduced.
2.15     Completion of the Crown Prosecution Service work is anticipated during the consultation period. We are most grateful to the CPS for undertaking this work on our behalf.

Public perceptions of current homicide law

2.16     Considerable research in this field has been carried out by Professor Barry Mitchell of Coventry University. Professor Mitchell has also conducted a study into the role of psychiatrists in cases of diminished responsibility, focusing on the perspective and concerns of the medical practitioners involved.^[11] That research is of importance to the present project, in particular diminished responsibility, and we refer to it in Part VII.^[12]

Published studies

2.17     Professor Mitchell, in collaboration with Social and Community Planning Research, conducted a quantitative study in 1998 with the "primary aim of testing the law's assumption that it has public support for the way in which it deals with homicides".^[13] The research sample was designed to be representative of the key demographic variables in the English and Welsh populations. There were 822 participants. Each respondent was presented with eight hypothetical scenarios of homicide, including situations of battered spouses^[14] and self-defence.^[15] The participants were asked to rank the seriousness of each scenario, giving reasons. They were also asked to indicate what they felt to be the appropriate sentence in the most extreme cases and how their perception of the crime and the appropriate sentence would alter if certain variations were made to the scenarios posed.
2.18     The research revealed that the personal culpability of the defendant was a key factor in the public's evaluation of homicides. Particular factors which were influential in assessing personal culpability were premeditation, planning, the vulnerability of the victim, justifiable self-preservation and the extent of the defendant's fault.^[16] Professor Mitchell notes that the law differentiates between gravity in homicides solely on the basis of the killer's culpability. He also notes, however, that "one dimension of culpability which the law largely seeks to avoid when framing offences and identifying justifications and excuses concerns the killer's motive and yet this very factor featured quite prominently in respondents' assessments".^[17] Professor Mitchell raises the question whether any law reform would be able adequately to take account of motives without being open to abuse.^[18] Similar observations can be made in relation to matters of wider social policy such as "mercy killing" and situations where the moral imperative would be to act to preserve human life, but where there is no law requiring the person to do so.
2.19     In addition, the participants were asked to place the different scenarios into offence categories. The purpose was to see whether "the public recognise variations in the gravity of homicides and, if they do, whether they feel such variations should be reflected in convictions for different offences such as murder and manslaughter, or some alternative form of formal labelling".^[19] Interestingly, in relation to the terms of reference of this project, the battered spouse scenario was placed by approximately one quarter of participants in the same category as "mercy killing",^[20] which was in itself widely considered to be the least culpable of the scenarios.^[21]
2.20     13 per cent of participants gave the lowest available score to those scenarios where the killing had been perpetrated in self-defence or self-preservation. No participants used the word "murder" to describe either the battered spouse or self-defence scenarios. Approximately 15 per cent considered these examples to be "killings", and the same proportion described the self-defence scenario as "manslaughter".^[22] 21 per cent classified the self-defence scenario as "self-preservation", but only 14 per cent made the same assessment for the battered spouse scenario.^[23] The "necessity" scenario elicited a similar percentage of categorisations as self-preservation.^[24]
2.21     The response to the killing by the battered spouse was wide ranging. It was classified as planned or premeditated by 12 per cent of participants. Factors cited by participants as affecting their evaluation of the relative gravity of the offence included the presence of alternative courses of action, premeditation and, by way of contrast, the persistent abuse suffered as well as elements of self-preservation.^[25]
2.22     In 2000 Professor Mitchell carried out a small follow-up study with 33 of the original participants. The purpose was stated to be to "enhance our understanding of public opinion on how the criminal justice system should respond to homicide, and in particular to obtain more data on the extent to which there is public support for fundamental principles such as individual autonomy, fair labelling, and correspondence".^[26]
2.23     The participants were referred back to the same scenarios used in the initial study. They were asked to comment on and explain their observations pertaining to gravity and the existence and effect of any aggravating factors. A number of new issues were also introduced. Of particular relevance for the present project were the questions concerning the treatment of mentally ill killers by the criminal justice system.^[27] The unanimous response was that the killer should not even be subjected to criminal prosecution if it could be proved that he would not have committed the fatal act but for his mental illness.^[28] There was, however, a consistent opinion that a mentally ill killer must be taken "out of circulation" in order to protect society. Four of the participants expressed sympathy for the killer.
2.24     Professor Mitchell concludes; "the follow up survey provided encouraging evidence that, given the opportunity to consider the matter, the public present neither knee-jerk nor especially punitive views about the away in which the criminal justice system should deal with homicide … . [T]here were real signs of attempts to give proportionate and discriminating responses, of efforts to focus not solely on the loss of life but to take account of the circumstances generally, including the personal culpability of the killer… . [T]here is little doubt that they share the view that distinctions in gravity should be made according to the offender's personal culpability".^[29]

Ongoing studies

2.25     As part of the Law Commission's current project, Professor Mitchell is undertaking a more specifically focused research study on public attitudes towards pleas of provocation, diminished responsibility and excessive use of force in self-defence. The research sample will include approximately 40 individuals of varied backgrounds. Again participants will be presented with a series of scenarios and will be asked to comment on their relative gravity and the effect on their view where the facts are adjusted. Participants will also be asked to place the scenarios into groups marking respectively the least and most serious as well as those where the participant believes that no offence has been committed. They will also be asked how this grouping of scenarios into offences would be affected by there being a mandatory life sentence for the most serious group.
2.26     Whilst the scale of this study is such that it could not be described as a national survey of public opinion, it represents the detailed response of a group which reflects a broad spectrum of individual characteristics, together with those of secondary victims. It will, in our view, provide a valuable indication not only of the range of views about the way that the criminal justice system currently deals with these homicides, but also about the reforms that people would like to see implemented.
2.27     We hope to receive the results of Professor Mitchell's research in early 2004.

Note 1    S Dell, Murder into Manslaughter: The diminished responsibility defence in practice (1984).     [Back]

Note 2    Ibid, Appendix: Tables and Figure, Table1.2, Table 2.6 and Table 2.7.    [Back]

Note 3    Table 1.2 is reproduced as Table 1, Table 2.6 is reproduced as Table 2 and Table 2.7 is reproduced as Table 3.    [Back]

Note 4    According to Dell, “In deciding how to sample the population, one of the problems was that the total number of diminished responsibility convictions was in the early years considerably smaller than in the later years… . Because of this … it was decided to … sample the first [period] at a higher rate than the other two. … [t]he sampling procedure was to select two-thirds of the men convicted in the early years [1966-9], and one third of those convicted in the middle [1970-3] and late [1974-7] years”. S Dell, Murder into Manslaughter: The diminished responsibility defence in practice (1984) p 4.    [Back]

Note 5    This statistic includes cohabitee (9 cases), mistress (4) and girlfriend (2). In the 18 cases where there was more than one victim, the first is shown.    [Back]

Note 6    Other than in a setting of amorous jealousy.    [Back]

Note 7    Quoted in Written Answer, Hansard (HC) 17 October 1991, vol 196, cols 190-192. See S Bandalli, “Provocation from the Home Office” [1992] Crim LR 716. Bandalli explains, in n 3, “whilst the term ‘domestic homicide’ is used, in fact the category includes lover, former lover, … spouse’s lover and lover’s spouse or other associate”.    [Back]

Note 8    As at 2 October 2003.    [Back]

Note 9    Research, Development and Statistics Directorate, Crime in England and Wales 2002/2003 (2003) 76.    [Back]

Note 10    This total is the actual and not the rounded total which is achieved by adding the totals from the sub-categories. The rounded total is 2,780.     [Back]

Note 11    B Mitchell, “Putting diminished responsibility into practice: a forensic psychiatric perspective” (1997) 8 Journal of Forensic Psychiatry 620.     [Back]

Note 12    See paras 7.66 – 7.68.    [Back]

Note 13    B Mitchell, “Public Perceptions of Homicide and Criminal Justice” (1998) 38 British Journal of Criminology 453 at p 456.    [Back]

Note 14    The text given to the participants read: “A woman had been physically and sexually abused by her husband for three years. He came home one evening and started hitting her again. She felt that she couldn’t stand any more abuse, so she waited until her husband was sleeping, then hit him over the head with a saucepan, killing him”, B Mitchell, “Public Perceptions of Homicide and Criminal Justice” (1998) 38 British Journal of Criminology 453 at p 457.    [Back]

Note 15    The text given to the participants read: “Two men were having a heated argument at work which developed into a fight. One of them picked up a screwdriver and lunged at the other. Fearing that he would otherwise be stabbed, the unarmed man grabbed a spanner, and in self-defence he hit the other man over the head with it, killing him”, B Mitchell, “Public Perceptions of Homicide and Criminal Justice” (1998) 38 British Journal of Criminology 453 at p 457.    [Back]

Note 16    B Mitchell “Public Perceptions of Homicide and Criminal Justice” (1998) 38 British Journal of Criminology 453 at pp 467–468.    [Back]

Note 17    Ibid, at p 468.    [Back]

Note 18    Ibid.    [Back]

Note 19    Ibid, at p 465.    [Back]

Note 20    26% of the sample associated the two scenarios. B Mitchell, “Public Perceptions of Homicide and Criminal Justice” (1998) 38 British Journal of Criminology 453 at p 466.    [Back]

Note 21    49% of the sample attributed the lowest rating to mercy killing. B Mitchell, “Public Perceptions of Homicide and Criminal Justice” (1998) 38 British Journal of Criminology 453 at p 464.    [Back]

Note 22    B Mitchell, “Public Perceptions of Homicide and Criminal Justice” (1998) 38 British Journal of Criminology 453 at p 466.    [Back]

Note 23    Ibid.    [Back]

Note 24    The text given to the participants read: “Two mountain climbers were roped together. One of them slipped and fell. The other tried to hold on to the rocks for both of them, but he knew that if he did not cut the rope, they would both die. To save himself, he cut himself loose, knowing that the other climber would fall to his death.” B Mitchell, “Public Perceptions of Homicide and Criminal Justice” (1998) 38 British Journal of Criminology 453 at p 457. 20% of respondents categorised this as self-preservation.     [Back]

Note 25    B Mitchell, “Public Perceptions of Homicide and Criminal Justice” (1998) 38 British Journal of Criminology 453 at p 460.    [Back]

Note 26    B Mitchell, “Further Evidence of the Relationship Between Legal and Public Opinion on the Law of Homicide” [2000] Criminal Law Review 814 at p 815.    [Back]

Note 27    Ibid, at p 817.    [Back]

Note 28    Ibid, at p 820.    [Back]

Note 29    B Mitchell, “Further Evidence of The Relationship Between Legal and Public Opinion on the Law of Homicide” [2000] Criminal Law Review 814 at pp 825-826. (emphasis added)    [Back]