INVOLUNTARY MANSLAUGHTER [1996] LC 237
Legislating the Criminal Code
Law Commision
Laid before Parliament by the Lord High Chancellor pursuant to section 3(2) of the Law Commissions Act 1965
CONTENTS
PART I: INTRODUCTION
1.1 – 1.9 Corporate manslaughter
1.10 – 1.21 The context in which this project is set
…
PART II: THE DIFFERENT WAYS OF COMMITTING ‘INVOLUNTARY MANSLAUGHTER’ UNDER THE PRESENT LAW
Introduction 2.1 – 2.2
Unlawful act manslaughter 2.3 – 2.7
Gross negligence manslaughter 2.8 – 2.16
…
Liability for omissions 2.22 – 2.25 19
‘Subjective’ recklessness 2.26 – 2.27 20
PART III: WHAT IS WRONG WITH THE PRESENT LAW?
Introduction 3.1
The breadth of the offence 3.2 – 3.4
Unlawful act manslaughter 3.5 – 3.6
Gross negligence manslaughter after Adomako 3.7 – 3.13
Liability for omissions 3.14 – 3.16
Conclusion 3.17
PART IV: THE MORAL BASIS OF CRIMINAL LIABILITY FOR UNINTENTIONALLY CAUSING DEATH
Introduction 4.1 – 4.3
Orthodox subjectivist theory 4.4 – 4.6
Subjectivist principles applied in the Law Commission’s recent work on offences against the person 4.7 – 4.9
Recklessly causing death 4.10 – 4.11
Criticisms of the subjectivist mens rea principle: can criminal liability based on inadvertence ever be justified? 4.12 – 4.16
What makes inadvertence culpable? 4.17 – 4.29
The conduct led to harmful consequences: ‘moral luck’ 4.30 – 4.33
The accused’s conduct was criminal in some way independent of the causing of death 4.34
The accused intended to cause some harm, or was aware of the risk of doing so, and/or it was foreseeable that her conduct created the risk of causing some harm 4.35 – 4.42
Conclusion 4.43 – 4.44
PART V: OUR PROPOSALS FOR REFORM
Introduction 5.1
One broad or several narrow offences? 5.2 – 5.5
Reckless killing 5.6
Unlawful act manslaughter 5.14 – 5.16
Killing by gross carelessness 5.17
…
Our final recommendation 5.24 – 5.34
Some examples 5.35 – 5.37
Omissions causing death 5.42 – 5.45
The maximum sentence 5.46 – 5.52
…
The Forfeiture Act 1982 5.70 – 5.76
PART VI: CORPORATE MANSLAUGHTER: THE PRESENT LAW
Introduction 6.1 – 6.4
Procedure 6.5 – 6.6
Interpretation provisions 6.7
…
The nature of the principle 6.27 – 6.34
Who are the controlling officers? 6.35 – 6.39
An indictment for manslaughter now lies against a corporation 6.40 – 6.48
The application to corporations of the substantive law of manslaughter 6.49 – 6.56
…
PART VIII: A NEW OFFENCE OF CORPORATE KILLING
Introduction 8.1 – 8.2
Foreseeability of the risk 8.3 – 8.4
Seriousness of the defendant’s conduct 8.5 – 8.7
Conduct of the defendant that causes death 8.8
Conduct of the defendant 8.9 – 8.35
Causation of death 8.36 – 8.39
The issue 8.40
…
Independent contractors and the proposed corporate offence 8.44
An illustration 8.45 – 8.50
Corporations 8.51 – 8.53
Unincorporated bodies 8.54 – 8.55
…
Consent to prosecution 8.63 – 8.66
Mode of trial 8.67
…
Compensation 8.71
Remedial action 8.72 – 8.76
Corporate liability for the individual offences 8.77
PART IX: SUMMARY OF OUR RECOMMENDATIONS
THE SCOPE AND STRUCTURE OF THIS REPORT
1.1 This report is concerned with the criminal liability of those who kill when they do not intend to cause death or serious injury. There are two conflicting schools of thought about the way in which the law should deal with such people. Some argue that society should always punish a person who causes terrible consequences to occur. Professor Hart puts the opposite view in these terms:
‘All civilised penal systems make liability to punishment for at any rate serious crime dependent not merely on the fact that the person to be punished has done the outward act of a crime, but on his having done it in a certain state or frame of mind or will.’ H L A Hart, Punishment and Responsibility: Essays in the Philosophy of Law (1968) p 114
In this report we consider what ‘frame of mind or will’ ought to be required if criminal liability is to be imposed for unintentional killing.
1.2 There are only two general homicide offences under the present law. The more serious of these, murder, requires proof of intention to kill or to cause serious injury, and the absence of such mitigating circumstances as the fact that the killer was provoked, or acted under diminished responsibility, or was the survivor of a suicide pact. Every other case of unlawful killing is included within the second homicide offence, manslaughter. This offence is, therefore, extremely broad. It ‘ranges in its gravity from the borders of murder right down to those of accidental death’ - Walker (1992) 13 Cr App R (S) 474 per Lord Lane CJ.
1.3 Although it is a single offence, manslaughter is commonly divided by lawyers into two separate categories, ‘voluntary’ and ‘involuntary’ manslaughter. The first of these describes cases where the accused intended to cause death or serious injury but is excused liability for murder because some mitigating factor may be present. In the present project we are concerned only with the second type, ‘involuntary’ manslaughter. This expression covers cases where there was no intention to kill or to cause serious injury, but where the law considers that the person who caused death was blameworthy in some other way.
1.4 Under the law as it stands at present, a person who unintentionally causes death is treated as sufficiently blameworthy to attract serious criminal sanctions in two cases. The first, known as ‘unlawful act manslaughter’, arises where the person who causes death was engaged in a criminal act which carried with it a risk of causing some, perhaps slight, injury to another person. The second type of involuntary manslaughter, ‘gross negligence manslaughter’, is harder to define. To put it very simply, the offence is committed by those who cause death through extreme carelessness.
1.5 In Part II we summarise the present law relating to both types of involuntary manslaughter, and in Part III we examine the contemporary problems they create. There are a number of minor problems in the form of uncertainties arising from the way in which the law has been formulated in particular cases. In addition to these uncertainties, however, there are two major problems. The first is that the present offence of manslaughter is too wide. This can cause problems both for judges on sentencing and for the public, who have difficulty in understanding the sentencing dilemma that faces a judge when an offence is so wide. It is in any event inappropriate that the same label should apply both to conduct on the borders of murder and to conduct on the borders of mere carelessness. The second major problem relates to unlawful act manslaughter: we consider that it is wrong in principle that a person should be convicted for causing death when the gravest risk apparently inherent in his conduct was the risk of causing some injury. This is a matter which we consider thoroughly in Part IV.
1.6 That Part is devoted to an exploration of the distinction between punishing a person for the consequences of his acts and punishing him for the state of mind in which he acted. The extent to which a person is responsible for the unintended consequences of his actions is, as we say there, one which has troubled philosophers for many years. There is no easy answer. However, it was important for us to come to a decision on this issue because we believe very strongly that the criminal law should rest, so far as is possible, on consistent, logical and principled foundations.
1.7 … We have eventually concluded that a person ought to be criminally liable for causing death only where he was aware that his conduct created a risk of causing death or serious injury to another, or where he was seriously at fault in failing to be aware of this risk. We believe that someone should only be blamed for failing to advert to such a risk if it would have been obvious to a reasonable person in his position, and he was himself capable of appreciating it at the material time.
1.8 In Part V we set out our detailed recommendations for a modern, codified law of involuntary manslaughter. In brief, we recommend the creation of two new offences in order to resolve the problems caused by the width of the present law. The more serious of the two offences, with a maximum penalty of life imprisonment, is called ‘reckless killing’. It would be committed by a person who unreasonably and consciously decides to run a risk of causing death or serious injury. The second new offence is called ‘killing by gross carelessness’. This would require proof of three matters.
First, that the defendant’s conduct involved an obvious risk of causing death or serious injury, of which he need not actually have been aware, as long as he was capable of appreciating it.
Secondly, that his conduct fell far below what could be expected of him in all the circumstances, or that he intended to cause some unlawful injury to another or was reckless whether he did so.
And, thirdly, that he caused death. We make no recommendation as to the maximum sentence for this offence, and if our recommendations are implemented it will be for others to determine what maximum is appropriate; but we have no reason to suppose that the maximum would be set at such a figure as to affect the levels of sentence currently imposed by the courts.
1.9 If our recommendations were implemented, English law would then possess, in effect, four degrees of general criminal homicide: murder, (voluntary) manslaughter, reckless killing and killing by gross carelessness. There would also be, as now, certain homicide offences aimed at specific situations, such as causing death by dangerous driving, infanticide and aiding and abetting suicide.
1.10 In this report we have decided to devote special attention to corporate liability for manslaughter, for three reasons.
First, as we will show, a number of recent cases have evoked demands for the use of the law of manslaughter following public disasters, and there appears to be a widespread feeling among the public that in such cases it would be wrong if the criminal law placed all the blame on junior employees who may be held individually responsible, and did not also fix responsibility in appropriate cases on their employers, who are operating, and profiting from, the service they provide to the public, and may be at least as culpable.
Second, we are conscious of the large number of people who die in factory and building site accidents and disasters each year: many of those deaths could and should have been prevented.
Third, there appear to have been only four prosecutions of a corporation for manslaughter in the history of English law, and only the last of these cases resulted in a conviction; significantly, this was a ‘one-man company’. It has been suggested that there are a number of outside factors which contribute to the low level of prosecutions brought against corporations for criminal offences generally.
1.11 To highlight the problems with the present law, it is helpful to refer to a series of recent disasters followed by inquiries which found corporate bodies at fault and meriting very serious criticisms. Perhaps surprisingly, no successful prosecution for manslaughter has been brought against any of the criticised parties.
1.12 On 18 November 1987 a fire of catastrophic proportions occurred in the King’s Cross underground station, claiming the lives of 31 people. In his report on the fire, Mr Desmond Fennell QC (as he then was) was critical of London Underground for not guarding against the unpredictability of the fire, and also because no one person was charged with overall responsibility.
1.13 In July 1988, the Piper Alpha oil platform disaster in the North Sea caused 167 deaths. In a public inquiry, conducted by Lord Cullen, which also served in effect as an inquest, serious criticism was directed at the platform operator, holding it responsible for the deaths.
1.14 On 12 December 1988, the Clapham rail crash caused 35 deaths and nearly 500 injuries when three rush-hour trains collided after a signal breakdown. In his report, Mr Anthony Hidden QC (as he then was) was very critical of British Rail, whose ‘concern for safety was permitted to co-exist with working practices which … were positively dangerous … the evidence showed the reality of [their] failure to carry that concern through into action’. Further, ‘the errors go much wider and higher in the organisation than merely to remain at the hands of those who were working that day’, and the report lists 16 serious relevant errors.
1.15 The reason for the absence of any conviction is probably the difficulty of mounting a manslaughter prosecution against a large-scale corporate defendant. This is illustrated by the prosecution following the tragedy which occurred on 6 March 1987, when the Herald of Free Enterprise, a roll-on roll-off car ferry, departed from Zeebrugge for Dover and shortly afterwards foundered with substantial loss of life. A judicial inquiry severely criticised P & O European Ferries (formerly Townsend Car Ferries Ltd). The jury at the inquest returned verdicts of unlawful killing in 187 cases, and eventually in June 1989 the DPP launched prosecutions against the company and seven individuals. But the trial collapsed after Turner J directed the jury to acquit the company and the five most senior individual defendants.
1.16 The outcome of this case provoked much criticism. The principal ground for the decision in relation to the case against the company was that, in order to convict the company of manslaughter, individual defendants who could be ‘identified’ with the company would have themselves to be guilty of manslaughter; since there was on the facts insufficient evidence to convict any such individual defendant, the case against the company also had to fail. This decision highlighted the major difficulty that has to be overcome before a company can be successfully prosecuted, namely that the relevant acts have to be committed by those ‘identified as the embodiment of the company itself’. This principle is usually called the identification doctrine.
1.17 The great difficulty arises in identifying the people who are the embodiment of the company. As one commentator has pointed out, one effect of the identification doctrine is that the more diffuse the company structure, and the more devolved the powers that are given to semi-autonomous managers, the easier it will be to avoid liability. Other critics have said that this point is of particular importance given the increasing tendency of many organisations to decentralise safety services in particular; they point out that it is in the interests of shrewd and unscrupulous management to do so. They also quote from a study38 which shows that companies sought to abrogate responsibility for the quality of their safety research by using contract laboratories, where the effects of fierce competition over price on the standard of safety checks could be said to be the responsibility of the laboratory itself. Another problem which was identified in the Zeebrugge inquiry was that no single individual had responsibility for safety matters. If responsibility for the development of safety monitoring is not vested in a particular group or individual, it becomes almost impossible to identify the ‘directing mind’ for whose shortcomings the company can be liable.
1.18 The problems that confront a prosecution for corporate manslaughter explain why there has only been one successful prosecution in England and Wales, and in that case against a small company. We have welcomed the opportunity to reconsider the principles of corporate liability in the light of the great obstacles now confronting those wishing to bring a prosecution; but we are also conscious of the need to ensure that companies are not unjustly convicted merely because they are in charge of an operation or a vessel on which there has been a disaster.
…
1.20 In Part VI, we set out our understanding of the way in which the present law on corporate liability has developed…
1.21 In Part VIII we set out the details of our new offence of corporate killing. Our main recommendations are as follows:
(1) There should be a special offence of corporate killing, broadly corresponding to the individual offence of killing by gross carelessness.
(2) Like the individual offence, the corporate offence should be committed only where the defendant’s conduct in causing the death falls far below what could reasonably be expected.
(3) Unlike the individual offence, the corporate offence should not require that the risk be obvious, or that the defendant be capable of appreciating the risk.
(4) For the purposes of the corporate offence, a death should be regarded as having been caused by the conduct of a corporation if it is caused by a failure, in the way in which the corporation’s activities are managed or organised, to ensure the health and safety of persons employed in or affected by those activities.
(5) For the purposes of the corporate offence, it should be possible for a management failure on the part of a corporation to be a cause of a person’s death even if the immediate cause is the act or omission of an individual.
(6) The corporate offence should be capable of commission by any corporation, however and wherever incorporated, other than a corporation
sole.49
(7) The corporate offence should not be capable of commission by an unincorporated body.
(8) The corporate offence should not be capable of commission by an individual, even as a secondary party.
(9) There should be liability for the corporate offence only if the injury that results in death is sustained in such a place that the English courts would have had jurisdiction over the offence had it been committed by an individual other than a British subject.
(10) There should be no requirement of consent to the bringing of private prosecutions for the corporate offence.
(11) The corporate offence should be triable only on indictment.
(12) Where a jury finds a defendant not guilty of any of the offences we recommend, it should be possible (subject to the overall discretion of the judge) for the jury to convict the defendant of an offence under section 2 or 3 of the Health and Safety at Work etc Act 1974.
(13) A court before which a corporation is convicted of the corporate offence should have power to order the corporation to take such steps, within such time, as the order specifies for remedying the failure in question and any matter which appears to the court to have resulted from the failure and been the cause or one of the causes of the death.56
(14) The ordinary principles of corporate liability should apply to the individual offences that we propose.
2.1 As we have observed, ‘involuntary manslaughter’ is the name given to those unintentional killings that are criminal at common law: causing death in the course of doing an unlawful act, and causing death by gross negligence or recklessness. ‘Involuntary manslaughter’ is not recognised as a separate crime in its own right: it is simply a label used to describe certain ways of committing the very broad common law crime of manslaughter.
…
2.3 The basis of this type of manslaughter is that the defendant caused the death of another by or in the course of performing an act which would have been unlawful whether or not death was caused. As Lord Parker CJ put it (Creamer [1966] 1 QB 72):
‘A man is guilty of involuntary manslaughter when he intends an unlawful act and one likely to do harm to the person and death results which was neither foreseen nor intended. It is the accident of death resulting which makes him guilty of manslaughter as opposed to some lesser offence.’
2.4 The alternative name of this type of crime, ‘constructive manslaughter’, draws attention to the fact that although the accused did not intend to cause serious harm or foresee the risk of doing so, and although an objective observer would not necessarily have predicted that serious harm would result, the accused’s responsibility for causing death is ‘constructed’ from her fault in committing a quite unconnected and possibly minor unlawful act. Because of this feature of the offence, the accused’s mental state is not assessed with reference to the death that she has accidentally caused, but only in relation to her unlawful act.
2.5 Over the years judges have tried in various ways to limit the scope of unlawful act manslaughter. Two ways in which they attempted to restrict liability were, first, by imposing stricter tests of causation than the test normally applied in criminal law, and, secondly, by requiring that the accused’s act must have been ‘directed at’ the deceased. Neither of these two approaches, however, has been consistently applied. A more lasting modification was the rule that the accused must have committed a crime of some sort in order to incur liability; at one time it was thought that the commission of a tort, if it caused death, was sufficient. In 1937 (Andrews v DPP [1937] AC 576) the House of Lords restricted the offence still further by holding that negligent acts, even those that were capable of constituting statutory criminal offences (such as dangerous driving), would not automatically be sufficient to found a conviction for manslaughter where death was caused. Instead, it became necessary to prove that the defendant’s negligence had been of a very high level. In such a case the prosecution would have to proceed under the second head of involuntary manslaughter, gross negligence manslaughter.
2.6 Another rule that judges have introduced relatively recently to limit the width of unlawful act manslaughter is the rule that the act that caused the death, in addition to being unlawful, must also have been ‘dangerous’, in the sense that ‘all sober and reasonable people would inevitably recognise [that it] must subject the other person to, at least, the risk of some harm resulting therefrom, albeit not serious harm’ (Church [1966] 1 QB 59; approved in Newbury [1977] AC 500). When applying this test, the ‘sober and reasonable person’ is accredited with any special knowledge that would have been available to the defendant, but no more (e.g. in Dawson (1985) 81 Cr App R 150 a petrol station attendant with a weak heart died of heart failure following the appellants’ attempted robbery of the station. In judging whether this act was ‘dangerous’, the Court of Appeal decided that the ‘sober and reasonable’ bystander could be assumed to know, like the appellants, that the gun used by them was a replica but that the victim might think that it was real, but not that the attendant had a weak heart). However, the reasonable observer will not have attributed to her any mistaken belief held by the accused (e.g. the belief held by the defendant in Ball [1989] Crim LR 730 that there was no risk created by loading his gun from the mixture of live and blank ammunition in his pocket and firing at the deceased).
2.7 It is unlawful act manslaughter, then, if D slaps V in the face, V loses her balance, falls to the ground and dies as a result of brain injury caused by hitting her head on the pavement; if D breaks into a house with intent to steal, and terrifies the occupant into a heart attack; or if D unlawfully carries a knife for self-defence, with which she accidentally stabs V. It is, of course, possible to think of numerous other examples.
2.8 Where a person causes death through extreme carelessness or incompetence, the law of gross negligence manslaughter is applied. Frequently the defendants in such cases are people carrying out jobs that require special skills or care, such as doctors, police or prison officers, ships’ captains or electricians, who fail to meet the standard which could be expected from them and cause death; however, an ordinary person who carries out a lawful activity, such as hunting or driving, without due caution, or who fails properly to look after a dependent person in her care, may be the subject of such a charge. The categories of unlawful act and gross negligence manslaughter are not mutually exclusive; for example, a defendant who unlawfully shoots at a trespasser may be guilty on both counts.
2.9 The early case-law indicated that to cause death by any lack of care whatsoever would amount to manslaughter. The development of the modern law can be traced to cases in the nineteenth century in which judges began to use the language of ‘gross negligence’. They were concerned to establish that a higher degree of fault ought to be necessary to incur criminal liability for manslaughter than that sufficient for civil liability for negligence.
2.10 In due course, in the case of Bateman the Court of Criminal Appeal held that gross negligence manslaughter involved the following elements:
(1) the defendant owed a duty to the deceased to take care;
(2) the defendant breached this duty;
(3) the breach caused the death of the deceased; and
(4) the defendant’s negligence was gross, that is, it showed such a disregard for the life and safety of others as to amount to a crime and deserve punishment.
This definition is circular – the jury should convict the accused of a crime if her behaviour was ‘criminal’ – and has been criticised on this ground. It is also uncertain and, because so much is left to the judgment of the jury, prone to inconsistent applications.
2.11 … Because judges found the terminology of ‘gross negligence’ unwieldy and difficult to explain to juries, they began to use the word ‘recklessness’ as a synonym, to describe a high degree of negligence. In other cases judges went further, and tried to give detailed definitions of recklessness (Stone and Dobinson [1977] QB 354). In doing so they succeeded, perhaps without intending to, in gradually changing the law that had been applied in Bateman. This culminated in the 1983 decision of the House of Lords in Seymour ([1983] 2 AC 493), which went some way towards removing the uncertainty that had previously characterised the law. However, this certainty was bought at the cost of widening the basis of liability and introducing a degree of rigidity into the way in which juries were directed.
2.12 In Seymour the House of Lords was concerned to identify the mental element required for ‘motor manslaughter’ – the short name used for convenience to describe gross negligence manslaughter when committed by the driver of a motor vehicle. In his speech, with which the other Law Lords agreed, Lord Roskill referred to a recent decision (Government of the USA v Jennings [1983] 1 AC 624) in which the House of Lords had held that the ingredients of motor manslaughter and of the statutory offence then in force of causing death by reckless driving were identical. He also referred to two decisions by the House in 1981 (Caldwell [1982] AC 341 and Lawrence [1982] AC 510), the combined effect of which was that, for the purposes of the offence of reckless driving, a person was reckless if:
(1) she did an act which in fact created an obvious and serious risk of injury to the person or substantial damage to property and
(2) when she did the act she either had not given any thought to the possibility of there being any such risk or had recognised that there was some risk involved and had nonetheless gone on to do it.
He concluded that, for motor manslaughter (and, by implication, for all cases of gross negligence manslaughter), the appropriate fault term was ‘recklessness’, and that this expression should bear the meaning ascribed to it in these 1981 decisions. This definition of recklessness is commonly described as ‘Caldwell recklessness’, after the leading case.
2.13 This judgment radically changed one aspect of the law of manslaughter. Under the Seymour rule, once the defendant had been shown by her conduct to have created an obvious and serious risk of causing physical injury to some other person, it was open to the jury to find her guilty whether her conduct was a result of mere inadvertence, conscious risk-taking or poor judgment. It was no longer open to a defendant to dispute guilt on the ground that her negligence had not been ‘gross’.
2.14 For a decade Seymour was applied fairly consistently by the courts, although in a few cases judges reverted to the previous law and language of gross negligence. This state of affairs was, however, recently ended by the decision of the House of Lords in Adomako ([1995] 1 AC 171). In this case… the accused, an anaesthetist, was acting as such during an eye operation which involved paralysing the patient. A tube became disconnected from the ventilator, the accused failed to notice the warning signs and the patient suffered a cardiac arrest and died. The House was asked to answer the following certified question:
‘in cases of manslaughter by criminal negligence not involving driving but involving a breach of duty is it a sufficient direction to the jury to adopt the gross negligence test set out in the Court of Appeal in [Prentice [1994] QB 304. This was the name under which Adomako was heard in the Court of Appeal] …, without reference to the test of … [Caldwell recklessness] or as adapted to the circumstances of the case?
2.15 Lord Mackay of Clashfern LC, who made the only substantial speech and with whom the other Law Lords agreed, disapproved the dictum of Lord Roskill in Seymour, and held that Bateman gross negligence was the appropriate test in manslaughter cases involving a breach of duty. He described the test for gross negligence manslaughter in the terms we set out in paragraph 3.8 below. In particular, he made it clear that it was a question of fact for the jury to determine whether the defendant’s breach of duty should be classified as gross negligence and therefore as a crime.
2.16 This decision resolved the principal uncertainty in the law – whether the test of Bateman gross negligence or of Caldwell recklessness should be applied. It also restored to the law the flexibility of the Bateman gross negligence test, which allowed the jury to consider the accused’s conduct in all the surrounding circumstances, and only punished her if her negligence was very serious. There are, however, still some remaining difficulties, which we consider in Part III below.
…
2.22 It is clearly established that the crime of involuntary manslaughter can be committed by omission, but only where the accused owes the deceased a duty to act. The circumstances in which a positive duty to act arises are uncertain, but we set out the common law position… in the paragraphs that follow.
2.23 There is no general rule in the criminal law imposing a duty to act. However, in the law of manslaughter a number of discrete cases have become established in which there is a duty to act; if the duty is neglected, and the person to whom it is owed dies, the person subject to the duty may be guilty of manslaughter. First, there is a duty to care for certain defined classes of helpless relatives: for example, spouses must take care of each other, and parents must look after their dependent children. A duty to act can also arise as a result of a contract (e.g. if an employer receives an employee or apprentice into her house, she is regarded as impliedly undertaking to provide the necessities of life if the other becomes ill), and a contractual duty can give rise to criminal liability if persons outside the contractual relationship, who are nonetheless likely to be injured by any failure to perform the contractual duty, are killed (e.g. Pittwood (1902) 19 TLR 37. A railway crossing gate-keeper had opened the gate to let a cart pass and then went off to his lunch, forgetting to shut it again, thereby allowing a haycart to cross the line and be struck by a train. He was convicted of manslaughter. It was argued on his behalf that he owed a duty only to his employers, the railway company, with whom he had contracted. Wright J held, however, that ‘there was gross and criminal negligence, as the man was paid to keep the gate shut and protect the public … . A man might incur criminal liability from a duty arising out of contract.’).
2.24 The most problematic instance of the duty to act arises where the accused has allegedly ‘undertaken’ to care for the deceased. During the second half of the nineteenth century the class of relationships capable of imposing criminal liability for omissions was extended to include voluntary undertakings, as where a person received into her house a young child or some other person who was unable to care for herself. The undertaking was expressly or impliedly given to a relative or to the previous custodian of the person received. In this century, however, the courts have extended this concept to cases where there has been no promise to care for the person received, by taking advantage of an ambiguity in the word ‘undertaking’, which can mean either a promise to do something or actually doing
it.
2.25 In one case, for example (Stone and Dobinson [1977] QB 354), the deceased, an elderly woman with anorexia nervosa, came to stay with her brother and his cohabitee, who were both of low intelligence, and subsequently starved herself to death. The Court of Appeal held that the question whether the couple owed a duty to care for the deceased was a question of fact for the jury, which was entitled to take into account the facts that she was a relative of one of the appellants, that she was occupying a room in his house, and that the other appellant had ‘undertaken’ the duty to care for her by trying to wash her and taking food to her.
2.26 Apart from unlawful act manslaughter and gross negligence manslaughter, there is one further way in which manslaughter may now be committed in the absence of intention to kill or cause serious injury. This arises when the accused is aware that her conduct involves a risk of causing death (or, probably, serious injury) and she unreasonably takes that risk. This combination of awareness of risk and unreasonable risk-taking is called ‘subjective’ recklessness (to distinguish it from ‘Caldwell’ recklessness, which has no requirement of awareness of the risk on the part of the accused. Also called Cunningham recklessness). Again, this type of mental state does not exclude liability for gross negligence or unlawful act manslaughter; a defendant may be guilty on all three counts.
2.27 Until ten years ago many cases of this type were treated as falling within the definition of murder. However, in a murder case in 1985 (Moloney [1985] AC 905) the House of Lords held that cases in which the defendant may have foreseen that death or really serious injury were highly probable to result from her act, without intending such consequences, would no longer constitute murder [see Woollin]. These cases must then have fallen, by default, into the scope of the offence of manslaughter. There is little or no separate authority, however, about this type of manslaughter, since such cases are dealt with in practice as cases of unlawful act manslaughter, and the accused’s awareness of the risk is taken into account only as an aggravating factor when it comes to sentencing.
3.1 We now turn to consider the problems created by the present law. The two major problems relate to the very wide range of conduct falling within the scope of involuntary manslaughter. As we explained in Part II, the offence encompasses:
first, cases involving conduct that falls only just short of murder, where the accused was aware of a risk of causing death or serious injury, although he did not intend to cause either;
second, cases where the accused is a professional person who makes a very serious mistake that results in death; and
third, cases where a relatively minor assault ends in death.
This leads to problems in sentencing and labelling, including the fundamental problem that many cases currently amounting to unlawful act manslaughter involve only minor fault on the part of the perpetrator, and therefore ought not, perhaps, to be described as manslaughter at all. There are also a number of more specific problems which we consider below.
THE BREADTH OF THE OFFENCE
3.2 The first problem, as we have just said, relates to the breadth of the conduct that is at present categorised as involuntary manslaughter. The width of the present offence can cause problems to judges on sentencing. As Lord Lane CJ remarked (Walker (1992) 13 Cr App R (S) 474, 476):
‘It is a truism to say that of all the crimes in the calendar, the crime of manslaughter faces the sentencing judge with the greatest problem, because manslaughter ranges in its gravity from the borders of murder right down to those of accidental death. It is never easy to strike exactly the right point at which to pitch the sentence.’
3.3 There is a strong argument in favour of defining criminal offences in terms of narrow bands of conduct, so that the judge can have the guidance of the jury on important factual questions, such as intention or awareness of risk. We agree with the notion that:
‘Questions of intention … involve the application of a test capable of precise definition (even though the task of drawing inferences from the evidence may be difficult). Gradations of culpability based on varying degrees of intention should, therefore, be incorporated into the definition of the offences, so that the issues can be contested with all that that implies in terms of the rules of procedure, evidence and quantum of proof.’ (D A Thomas, ‘Form and Function in Criminal Law’, in Peter Glazebrook (ed) Reshaping the Criminal Law (1978) p 28)
The same could be said of awareness of risk.
3.4 Another argument in favour of separate offences follows on from this point about sentencing. It is inappropriate that types of conduct that vary so widely in terms of fault should all carry the same descriptive label. The accused who sets fire to his house so that the council will re-house him, knowing that his wife and children are asleep inside and that they will almost certainly be killed or seriously injured, is blameworthy in a very different way from the electrician who causes death by miswiring an electrical appliance with a high degree of carelessness. It is arguable that the label ‘manslaughter’ is devalued, and the more serious forms of wrongdoing that it describes might come to be regarded as less serious, because it is also used to describe less heinous crimes. By the same token, juries might be reluctant to convict, for example, a highly incompetent doctor of manslaughter because of the perceived gravity of the offence.
3.5 The next problem with the present law also relates to the breadth of conduct falling within involuntary manslaughter. As we observed above (para 2.4), if a person commits a criminal act that carries a risk of causing some harm to another, and by chance he causes death, he will be guilty of unlawful act manslaughter. In some of these cases, the defendant would only have been guilty of a relatively trivial offence if death had not chanced to occur. For example, if D pushes V in a fight, and V staggers but does not fall, D will at most (if, say, V was bruised or injured in some other way) be guilty of causing actual bodily harm under section 47 of the Offences against the Person Act 1861, which carries a maximum sentence of five years’ imprisonment. If, however, V loses his balance and falls to the floor, knocking his head on the pavement and thereby sustaining fatal brain injuries, D will be guilty of manslaughter [which carries a maximum sentence of life imprisonment].
3.6 For reasons we explain in more detail below (Part IV), we consider that it is wrong in principle for the law to hold a person responsible for causing a result that he did not intend or foresee, and which would not even have been foreseeable by a reasonable person observing his conduct. Unlawful act manslaughter is therefore, we believe, unprincipled because it requires only that a foreseeable risk of causing some harm should have been inherent in the accused’s conduct, whereas he is convicted of actually causing death, and also to some extent punished for doing so (see, eg, Coleman (1991) 95 Cr App R 159 for the extent to which the causing of death is taken into account on sentencing in one class of unlawful act manslaughter).
3.7 There are also certain problems with the present law of gross negligence manslaughter, but they are much less fundamental than those considered above. As we explained in Part II, a recent House of Lords decision [i.e. Adomako] has resolved many of the problems that existed hitherto. In particular, the House decided that gross negligence (rather than Caldwell recklessness) is the appropriate test. There are, however, still some residual difficulties.
3.8 Lord Mackay of Clashfern LC, in describing the test for gross negligence manslaughter, said:
‘... the ordinary principles of the law of negligence apply to ascertain whether or not the defendant has been in breach of a duty of care towards the victim who has died. If such a breach of duty is established the next question is whether that breach of duty caused the death of the victim. If so, the jury must go on to consider whether that breach of duty should be characterised as gross negligence and therefore as a crime. This will depend on the seriousness of the breach of duty committed by the defendant in all the circumstances in which the defendant was placed when it occurred. The jury will have to consider whether the extent to which the defendant’s conduct departed from the proper standard of care incumbent upon him, involving as it must have done a risk of death to the patient, was such that it should be judged criminal.
3.9 The first problem with this test is that it is circular: the jury must be directed to convict the defendant of a crime if they think his conduct was ‘criminal’. In effect, this leaves a question of law to the jury, and, because juries do not give reasons for their decisions, it is impossible to tell what criteria will be applied in an individual case. This must lead to uncertainty in the law. The CPS has told us that prosecutors find it difficult to judge when to bring a prosecution, defendants have difficulty in deciding how to plead, and there is a danger that juries may bring in inconsistent verdicts on broadly similar evidence.
3.10 Other problems arise out of the Lord Chancellor’s use of the terminology of ‘duty of care’ and ‘negligence’, and his linkage of the civil and criminal law in his speech. The meanings of these words are not entirely clear in a criminal law context, nor is it clear to what extent they mean the same things in tort and in criminal law (Graham Virgo, ‘Reconstructing Manslaughter on Defective Foundations’ [1995] CLJ 14, said that ‘tortious duty of care can serve no useful function in this context and, anyway, the pragmatic approach to the concept which is adopted in the law of tort … is inappropriate in the criminal law’). Also, the ex turpi causa defence will not apply in a criminal context – see R v WACKER [2002] EWCA Crim 1944 in the Library.
3.11 …’negligence’ in the context of the crime of manslaughter probably means nothing more than ‘carelessness’: it does not carry the technical meaning that it has in the law of tort, where it depends on the existence of a duty of care owed and a breach of that duty. The Lord Chancellor said in Adomako that ‘the ordinary principles of the law of negligence apply to ascertain whether or not the defendant has been in breach of a duty of care towards the victim who has died’. This equation of the civil and criminal law concepts of negligence causes no problems where, as in Adomako itself, a death is caused by a badly performed positive act of the accused, because it is virtually certain that both tort and criminal law would hold that a duty was owed to the deceased not to injure him by a positive act.
3.12 It is possible, however, that the courts in future cases of omission might feel obliged to apply the decision in Adomako. If so, they would run into difficulties, because it is by no means certain that the scope of liability for negligent omissions is the same in criminal law as it is in tort. For example, in criminal law it would seem that once someone has voluntarily taken some steps to care for another, he may be liable if his care is not adequate and the other person dies (Stone and Dobinson [1977] QB 354). In tort, however, there is probably no liability if the defendant abandons an effort to care for someone and that person dies, unless he causes harm through his own incompetence (East Suffolk Rivers Catchment Board v Kent [1947] AC 74 is probably still authority on this point).
3.13 It is possible, therefore, that the decision in Adomako may have changed the criminal law in relation to liability for omissions, by equating it with the civil law of tort. This may have restricted the scope of the duty to act in criminal law, by implicitly overruling Stone and Dobinson; on the other hand, there may be cases where the law of tort imposes a more stringent duty to act than the criminal law had hitherto. The law on this subject is so unclear that it is difficult to tell whether the effect of Lord Mackay’s speech was indeed to change the law, and, if so, what the implications of this change might be. It is, however, clear that the terminology of ‘negligence’ and ‘duty of care’ is best avoided within the criminal law, because of the uncertainty and confusion that surround it.
(Simon Gardner, in ‘Manslaughter by Gross Negligence’ (1995) 111 LQR 22, 23, also criticises the use of this terminology: ‘The test [that there was a breach of a duty of care] probably originated in lawyers finding it helpful to conceive gross, criminal, negligence by contrasting it with ordinary, tortious, negligence. But since juries will be equally, if not more, unfamiliar with the latter, they will not be helped, and may even be confused, by being told to consider it’).
3.14 The final problem with the present law is also connected with liability for omissions. In Part II we set out the circumstances where, as far as we can tell, there is a duty to act, so that a person may be guilty of manslaughter if he fails to act and another person dies as a result. However, the present law is uncertain. This was demonstrated by the Criminal Law Revision Committee’s recommendation in 1980 that it ought not to be codified:
‘Most of us are of the opinion that the extent of the duty to act should be left undefined so that the courts can apply the common law to omissions. The main reason for this view is that the boundaries of the common law are not clearly marked and there would be difficulty in setting them out in statutory form.’
…
3.16 It is extremely unsatisfactory that the law should remain uncertain in this important area. We have considerable sympathy for the view expressed by Professor Glanville Williams on the CLRC’s recommendation against statutory formulation of the duty to act:
‘If the top lawyers in a Government committee find the law hard to state clearly, what hope have the Stones and Dobinsons of this world of ascertaining their legal position, in advance of prosecution, when they find themselves landed with a hunger-striking relative?’
Indeed, it is possible that the law in this area fails to meet the standard of certainty required by the European Convention on Human Rights.
3.17 In Part V below we set out our proposals for the reform of the law, which we hope will resolve many of the problems identified here – although some, we believe, cannot be solved in the context of this present project or at all. First, however, we consider the principles that, in our view, ought to underpin a modern law of involuntary manslaughter.
29 In Sunday Times v UK [1979] 2 EHRR 245, the European Court of Human Rights held that, before the state can impose coercive obligations by law on its citizens, the law must be formulated with sufficient certainty to enable the citizen to regulate his conduct and must also be ‘accessible’ to him, so that he has an adequate indication in advance of the rules which would be applied in any case.
4.1 In Parts II and III we examined the present law of involuntary manslaughter, and discussed the problems still inherent in it. This law has, of course, been developed by judges on a case by case basis. No single court has had the opportunity to consider the fundamental issues that underlie the whole of this area of the law. In this part of the report we turn to address the underlying principles, primarily the very basic, but also very difficult, question: when should a person be held criminally liable for unintentionally causing another person’s death?
4.2 The extent to which a person is responsible for the unintended results of her actions is a question that has puzzled legal philosophers for years (e.g. H L A Hart, Punishment and Responsibility: Essays in the Philosophy of Law (1968); R A Duff, Intention, Agency and Criminal Liability (1990); T Honore, ‘Responsibility and Luck’ (1988) 104 LQR 530; Bernard Williams, Moral Luck (1981); and Andrew Ashworth, ‘Taking the Consequences’ in S Shute, J Gardner and J Horder, Action and Value in the Criminal Law (1993)). We have had to consider this question again because we believe that it is essential that any new law of involuntary manslaughter that we may propose should be founded on just, coherent and logical principles.
4.3 We are also aware that our proposed new homicide offences ought, in time, to form part of a complete criminal code. It is evident that, in the interests of justice, logic and consistency, the same fundamental principles should, so far as possible, influence all the parts of this growing code. In this part, then, we begin by describing the philosophy that this Commission has traditionally applied in our criminal law reform work; it is known as ‘subjectivist legal theory’. We will then consider how this philosophy has shaped the work we have already done on the reform and codification of the law of offences against the person. Finally, we undertake a thorough examination of this philosophy, in order to decide whether it should apply in the present project.
4.4 The legal philosophy traditionally applied in mainstream English criminal law and by this Commission is known as ‘subjectivist theory’. It rests on the principle that moral guilt, and hence criminal liability, should be imposed only on people who can be said to have chosen to behave in a certain way or to cause or risk causing certain consequences. The roots of subjectivism lie in a liberal philosophy that regards individuals as autonomous beings, capable of choice, and each deserving of individual respect. It is called ‘subjectivism’ because of the significance it accords to the individual’s state of mind at the time of the prohibited conduct.
[This view can be contrasted with, e.g., ‘utilitarian theory’, which places emphasis on the social benefit to be derived from punishing a person (e.g. deterring others) rather than on the deserts of the individual offender himself.]
4.5 Three principles have been identified as inherent in this basis of liability.
The first of these is the ‘mens rea principle’, which imposes liability only for outcomes which were intended or knowingly risked by the alleged wrongdoer.
The second principle, the ‘belief principle’, judges a defendant according only to what she believed she was doing or risking.
Thirdly, according to the ‘principle of correspondence’, subjectivists insist that the fault element of a crime correspond to the conduct element; for example, if the conduct element is ‘causing serious injury’, the fault element ought to be ‘intention or recklessness as to causing serious injury’. This ensures that the defendant is punished only for causing a harm which she chose to risk or to bring about.
4.6 Subjectivist philosophy applies widely in the criminal law today. A man cannot be convicted of rape, for example, if he genuinely believed, albeit unreasonably, that his victim consented to sexual intercourse, because this belief would be incompatible with the intention to have intercourse with a woman without her consent, or recklessness as to that possibility, which are the mental states required for rape (DPP v Morgan [1976] AC 182).
Subjectivist principles applied in the Law Commission’s recent work on offences against the person
4.7 The principles just referred to were the basis of our law reform recommendations in our report on offences against the person (hereafter ‘Law Com No 218’) [Legislating the Criminal Code: Offences Against the Person and General Principles (1993) Law Com No 218. This was the first step towards our objective of producing a series of self-contained law reform Bills which, in time, could be combined into a single, unified criminal code], which was enthusiastically received. In it we recommended that the offences currently in force under sections 18, 20 and 47 of the Offences Against the Person Act 1861 should be replaced with three new offences, which were defined in the Criminal Law Bill included in the report as follows:
(1) A person is guilty of an offence if he intentionally causes serious injury to another.
(2) A person is guilty of an offence if he recklessly causes serious injury to another.
(3) A person is guilty of an offence if he intentionally or recklessly causes injury to another.
4.8 The Bill contains definitions of ‘intentionally’ and ‘recklessly’. The definition of ‘intentionally’ is not relevant here (except insofar as it requires a knowing decision to bring about a result). A person is defined as acting ‘recklessly’ in relation to a result if ‘he is aware of a risk that it will occur, and it is unreasonable, having regard to all the circumstances known to him, to take that risk’. The requirements of intention to cause injury, or awareness of the risk of doing so, in these proposed new offences are characteristic of orthodox subjectivist theory.
4.9 The main difference between the offences under the 1861 Act and those recommended in Law Com No 218 is that, for the proposed new offences, the accused’s mental state (that is, her intention or recklessness) corresponds to the harm for which she is punished. Under the new scheme, a person is only held responsible for causing serious injury if she intended to cause serious injury, or was reckless as to doing so. This contrasts with the present law, which allows a person to be convicted of causing serious injury, under section 20 of the 1861 Act, when all she intended or foresaw was the causing of some, perhaps minor, injury. Under the present law, it is not even necessary to show that a foreseeable risk of causing serious injury was inherent in her conduct. Similarly, under section 47, all that the accused has to foresee is any physical contact: the actual bodily harm for which she is convicted may be entirely unforeseeable.
4.10 Thus, in Law Com No 218 we recommended that a person ought to be held responsible for causing injury, or serious injury, when she intended to cause the harm in question, or was reckless as to doing so. We are quite certain that a person should, similarly, be held criminally responsible for causing death in circumstances where she unreasonably and knowingly runs a risk of causing death (or serious injury). Indeed – and we are sure that many people would agree with us – we consider this type of conscious risk-taking to be the most reprehensible form of unintentional homicide, on the very borders of murder (Until recently it was murder to cause death foreseeing that death or serious injury was a highly probable consequence of one’s conduct: Hyam v DPP [1975] AC 55). A person who sets fire to a house (in order, perhaps, to make an insurance claim, or to frighten a former lover) while others are asleep in it, thus causing death without intending to do so but knowing that there is a risk of doing so, is an example of this type of offender. The Criminal Law Revision Committee considered that involuntary manslaughter should be limited to this branch of the offence, because it was the only one in which there was a correspondence between the offender’s fault and the consequence of death (Fourteenth Report: Offences Against the Person (1980) Cmnd 7844, para 124).
4.11 The difficult question is whether, and in what circumstances, a person should be held criminally liable for causing death unintentionally when he was not aware that his conduct created such a risk. We consider this question in the following paragraphs.
CRITICISMS OF THE SUBJECTIVIST MENS REA PRINCIPLE: CAN CRIMINAL LIABILITY BASED ON INADVERTENCE EVER BE JUSTIFIED?
4.12 Orthodox subjectivist theory, then, requires the defendant to have been, at least, aware of the risk of causing the prohibited harm. However, there is a body of criticism, from very distinguished commentators, of the orthodox subjectivist mens rea principle. One ground of criticism is that it is based on a simplistic view of what constitutes knowledge or awareness of risk:
‘… while we do indeed sometimes make our knowledge of what we are doing explicit to ourselves in … silent mental reports, it is absurd to suggest that such knowledge can be actual only if it is made thus explicit. When I drive my car, my driving is guided by my (actual) knowledge of my car and of the context in which I am driving: but my driving is not accompanied by a constant silent monologue in which I tell myself what to do next, what the road conditions are, whether I am driving safely or not, and all the other facts of which I am certainly aware while I am driving. … The occurrence or the non-occurrence of certain explicit thoughts is irrelevant to whether I am actually aware of what I am doing: my actions can manifest my awareness even if no explicit thoughts about the relevant facts pass through my mind at the time.’ (R A Duff, Intention, Agency and Criminal Liability (1990) p 160. See also Alan Norrie, Crime, Reason and History (1993) ch 4.)
4.13 On this view of what constitutes a mental state, the contrast between awareness and lack of awareness of risk is not as stark as in conventional subjectivist accounts, and it is less clear why inadvertence ought not to be classified as mens rea in certain circumstances.
4.14 The main argument in favour of criminalising some forms of inadvertent risk-taking, however, is that in some circumstances a person is at fault in failing to consider the consequences that might be caused by her conduct. The example given by R A Duff is that of a bridegroom who misses his wedding because it slipped his mind when he was in the pub. An orthodox subjectivist would point to his lack of intention or awareness, and deem him consequently less culpable. The bride, however, would rightly condemn him, because it is plain from his conduct that he did not care, and this attitude is sufficient to make him blameworthy. Duff argues that this account retains a subjective element, because attitudes are subjective.
4.15 A similar argument was used by Lord Diplock in the famous case on criminal damage, Caldwell ([1982] AC 341):
‘If it had crossed his mind that there was a risk that someone’s property might be damaged but, because his mind was affected by rage or excitement or confused by drink, he did not appreciate the seriousness of the risk or trusted that good luck would prevent it happening, this state of mind would amount to malice in the restricted meaning placed upon that term by the Court of Appeal; whereas if, for any of these reasons, he did not even trouble to give his mind to the question whether there was any risk of damaging the property, this state of mind would not suffice to make him guilty of an offence under the Malicious Damage Act 1861. Neither state of mind seems to me to be less blameworthy than the other … .’
4.16 Professor Hart some years ago (Punishment and Responsibility (1968)) attacked the assumption that to allow criminal liability for negligence would be to set aside the requirement of mens rea as a precondition of punishment. His argument was that since ‘negligence’ implies a failure to do what ought to have been done, it is therefore more than inadvertence, it is culpable inadvertence:
‘Only a theory that mental operations like attending to, or thinking about, or examining a situation are somehow ‘either there or not there’, and so utterly outside our control, can lead to the theory that we are never responsible if, like the signalman who forgets to pull the signal, we fail to think or remember. ...
‘What is crucial is that those whom we punish should have had, when they acted, the normal capacities, physical and mental, for doing what the law requires and abstaining from what it forbids, and a fair opportunity to exercise these capacities. Where these capacities and opportunities are absent, as they are in different ways in the varied cases of accident, mistake, paralysis, reflex action, coercion, insanity etc, the moral protest is that it is morally wrong to punish because ‘he could not have helped it’ or ‘he could not have done otherwise’ or ‘he had no real choice’. But, as we have seen, there is no reason (unless we are to reject the whole business of responsibility and punishment) always to make this protest when someone who ‘just didn’t think’ is punished for carelessness. For in some cases at least we may say ‘he could have thought about what he was doing’ with just as much rational confidence as one can say of any intentional wrongdoing ‘he could have done otherwise’.
Professor Ashworth also concedes that negligence may be an appropriate standard for criminal liability where the harm risked was great, the risk obvious and the defendant had the capacity to take the required precautions (Principles of Criminal Law (1st ed 1991) pp 169–171; and see the 2nd ed (1995) pp 84–85).
4.17 In all the sources cited in paragraphs 4.12 – 4.16, the view is taken that it may be justifiable to impose criminal liability for the unforeseen consequences of a person’s acts, at any rate where the harm risked is great and the actor’s failure to advert to this risk is culpable. We are persuaded by this reasoning. In the following paragraphs, therefore, we consider the criteria by which culpable inadvertence should be judged if it is to attract the sanctions of the criminal law when death results.
4.18 The first criterion of culpability upon which we must insist is that the harm to which the accused failed to advert was at least foreseeable, if not strikingly foreseeable or obvious (L H Leigh in ‘Liability for Negligence: A Lordly Legacy?’ (1995) 58 MLR 457, 465, says: ‘It is clear that there must be an obvious and serious risk of physical injury or damage to property before liability for inadvertence can arise.’). If the accused is an ordinary person, she cannot be blamed for failing to take notice of a risk if it would not have been apparent to an average person in her position, because the criminal law cannot require an exceptional standard of perception or awareness from her. If the accused held herself out as an expert of some kind, however, a higher standard can be expected from her; if she is a doctor, for example, she will be at fault if she fails to advert to a risk that would have been obvious to the average doctor in her position.
4.19 As a matter of strict principle, the accused ought only to be held liable for causing death if the risk to which she culpably failed to advert was a risk of death (In Adomako, the Lord Chancellor expressed the test in terms of a risk of death). In practice, however, there is a very thin line between behaviour that risks serious injury and behaviour that risks death, because it is frequently a matter of chance, depending on such factors as the availability of medical treatment, whether serious injury leads to death. Admittedly it is possible for conduct to involve a risk of serious injury (such as a broken limb) though not a risk of death; but intention to cause serious injury constitutes the mens rea of murder although the actus reus is the causing of death, and we see no compelling reason to distinguish between murder and manslaughter in this respect. We consider, therefore, that it would not be wrong in principle if a person were to be held responsible for causing death through failing to advert to a clear risk of causing death or serious injury – subject of course to a second criterion, to which we now turn.
4.20 The second criterion of culpability which we consider to be essential is that the accused herself would have been capable of perceiving the risk in question, had she directed her mind to it (This criterion was also required by Professors Hart and Ashworth (see para 4.16 above); by L H Leigh, ‘Liability for Inadvertence: A Lordly Legacy?’ (1995) 58 MLR 457, 467; and, we think, implicitly by Lord Diplock on a true reading of Caldwell: see J Parry, Offences Against Property (1989) paras 6.27 – 6.30). Since the fault of the accused lies in her failure to consider a risk, she cannot be punished for this failure if the risk in question would never have been apparent to her, no matter how hard she thought about the potential consequences of her conduct. If this criterion is not insisted upon, the accused will, in essence, be punished for being less intelligent, mature or capable than the average person.
4.21 This is what happened in the criminal damage case, Elliott v C ([1983] 1 WLR 939). The defendant in this case was a 14 year old girl of low intelligence, who entered a garden shed at 5 am, having been out all night. She poured white spirit on the floor and threw matches on it, thus setting fire to the shed. The magistrates found that, in view of her age and understanding, her lack of experience of inflammable spirit and the fact that she must have been exhausted, the risk that the shed and its contents would be destroyed as a result of her actions would not have been obvious to her or appreciated by her, even if she had given any thought to it. However, the Divisional Court held that she ought to be convicted anyway, because the risk would have been obvious to a reasonable person in her position, and because at the time of starting the fire she had given no thought to the possibility of there being a risk of destroying the shed. According to the court’s interpretation of the relevant House of Lords authority (Caldwell [1982] AC 341), this was enough for a conviction of criminal damage.
4.22 We consider the position taken in Elliott v C to be highly unsatisfactory (indeed, Robert Goff LJ in the Divisional Court made it clear that he reached this decision with great reluctance and only because he was bound by the authority of Caldwell). It is hardly to be supposed that a blind person will be held reckless for inadvertently creating a risk which would have been obvious to a person with the power of sight; yet a child of 14 was held in this case to be reckless as a result of thoughtlessly creating a risk which she was incapable of appreciating, although it would have been obvious to an adult. A person cannot be said to be morally at fault in failing to advert to a risk if she lacked the capacity to do so.
4.23 If the criteria in paragraphs 4.17 – 4.22 are satisfied, we consider that it is appropriate to impose liability for inadvertently causing harm in cases where the harm risked is very serious. Where a person embarks on a course of conduct which inherently involves a risk of causing death or serious injury to another, society is justified in requiring a higher standard of care from her than from someone whose conduct involves a lesser risk or no risk at all.
(L H Leigh, ‘Liability for Inadvertence: A Lordly Legacy?’ (1995) 58 MLR 457, 467, argues:
‘Whatever be the basis of punishment, a body of rules which required us all to be careful in all aspects of our daily lives on pain of punishment would seem totalitarian. It would seem an extreme assertion of the right to punish in order to uphold social values. It would also seem too diffused to meet the exigencies of any educative theory of punishment.
‘It may well be that social awareness in respect of certain discrete dangers can be achieved by osmosis in the generation of which the status of certain instances of gross want of care can play a part.
‘No doubt also, in certain environments, this may even work at the level of consciousness. No soldier is left in doubt that it is dangerous and wrong to point a rifle at someone whom he does not mean to kill.
‘The same immediacy of perception of danger could not be said to be present in respect of a wide range of activities and situations which we face in our daily lives. Whether or not one believes that punishment can be justified on educative grounds, it must surely be admitted that at most it can only apply in particular situations of obvious and grave danger which are singled out as presenting obvious risks. Whether or not one believes that it is morally right to punish those whose social attitudes appear to evidence a contempt for accepted social values, no such justification could be advanced in respect of all cases of inadvertence causing harm.’)
J L Austin made this point graphically when he wrote ‘We may plead that we trod on the snail inadvertently: but not on the baby – you ought to look where you’re putting your great feet’ (‘A Plea for Excuses’ in Proceedings of the Aristotelian Society, New Series, vol 57 (1956–57)).
4.24 The criminal law has traditionally drawn a distinction between cases where death is caused by culpable inadvertence and cases where less serious forms of injury are caused, by imposing criminal liability for gross negligence manslaughter but not for causing non-fatal injury through gross negligence. It is more difficult to distinguish cases where the actor risks causing death or serious injury, but fortuitously causes only serious injury. Such cases, however, fall outside the scope of this project.
4.25 It may be helpful at this stage if we attempt to apply these principles to a few concrete examples.
4.26 Example 1: D is an anaesthetist who causes her patient V’s death because she fails to notice that a ventilation tube has become disconnected and that V has turned blue.
• would fall within our criteria of culpability if expert evidence showed that the risk of V’s death or serious injury would have been obvious to a competent anaesthetist in D’s position.
4.27 Example 2: D, an adult of average intelligence, in the course of a fight hits V over the head with a spanner. In the heat of the moment, D does not realise that death or serious injury may result; but the blow cracks V’s skull and causes her death.
• would fall within our criteria of culpability, because his conduct created an obvious risk of causing death or serious injury, which he was capable of appreciating, and which he ought to have considered before acting.
4.28 Example 3: D, in the course of a fight, slaps V once across the face. V loses her balance and falls to the floor, cracks her skull, and dies.
• would not necessarily fall within our criteria, because, arguably, there is not an obvious risk of causing death or serious injury inherent in her conduct.
4.29 If it is thought that the accused in this last example, or in any other case which does not meet our criteria, ought to be held liable for causing death, it must be on the basis that his conduct was culpable although his failure to advert to the risk of death or serious injury was not. We now consider some reasons why this might arguably be so; and, in the case of each such reason, whether it justifies convicting the defendant in respect of a death which he cannot be blamed for having failed to foresee.
4.30 It is arguable that a person is morally responsible for all the consequences that flow from her conduct, and that when these consequences are harmful it is appropriate for the law to hold her liable for them as well (such reasoning is, however, more usual in the context of the law of tort than in relation to criminal liability). A philosophical justification for this view is provided by ‘moral luck’ arguments, which hold that ethics are not separable from luck. It has been suggested that luck influences most aspects of our lives – including the capacities and personalities we are born with – and that, in the course of a lifetime and throughout a community, good and bad luck are fairly evenly distributed. Individuals cannot prevent the outcomes of their actions being influenced by luck, and to the extent that their actions impinge on others in a harmful way, they are inevitably judged by others on those outcomes. For example, if a child runs from behind a car into the path of a van which is being entirely properly driven, and the child is knocked down and killed, it is commonly said that the death of the child will be on the conscience of the driver, although everyone accepts that the death was ‘an accident.’
4.31 Professor Bernard Williams discusses the regret felt by an agent who was unintentionally responsible for the causing of harm, which is frequently accompanied by a desire, however illogical, to make reparation (Moral Luck (1981) ch 2:
‘What degree of such feeling is appropriate, and what attempts at reparative action or substitutes for it, are questions for particular cases, and that there is room in the area for irrational and self-punitive excess, no one is likely to deny. But equally it would be a kind of insanity never to experience sentiments of this kind towards anyone, and it would be an insane concept of rationality which insisted that a rational person never would. To insist on such a conception of rationality, moreover, would, apart from other kinds of absurdity, suggest a large falsehood: that we might, if we conducted ourselves clear-headedly enough, entirely detach ourselves from the unintentional aspects of our actions, relegating their costs to, so to speak, the insurance fund, and yet still retain our identity and character as agents.’
4.32 This argument is echoed by Professor Honore (Responsibility and Luck (1988) 104 LQR 530):
‘… outcome-allocation is crucial to our identity as persons … . If actions and outcomes were not ascribed to us on the basis of our bodily movements and their mutual accompaniments, we could have no continuing history or character.’
4.33 Because in everyday life the consequences of our decisions are attributed to us in a variety of ways, it is therefore argued that a person should be held morally and legally responsible (at least in theory) for all the harmful outcomes of her actions. There are a number of reasons why we do not find this argument persuasive.
First, the consequences of a person’s actions are not the only factors that are relevant to her identity: the moral judgments which the individual and others make about her responsibility for consequences are also important.
Secondly, and perhaps more importantly, just because judgments based on outcome-allocation do occur in everyday life, this does not mean that they ought to do so. If it is thought that the popular allocation of blame is illogical or unfair in some way, it is even more illogical and unfair to compound the effect of luck by giving it legal significance. While an argument might be made for this in the civil law, where reparation and loss allocation are in issue, it is difficult to extend it to the criminal law, which is concerned with public censure and punishment, and where moral culpability ought, in our view, to be the deciding factor.
The accused’s conduct was criminal in some way independent of the causing of death
4.34 It might be argued that if a person embarks on a train of behaviour which is contrary to the criminal law, she should take the consequences if death ensues. An example will clearly illustrate why we reject this argument:
D steals a cake from a shop, and feeds it to her friend, V. Unknown to either of them, the cake contains nuts, to which V is allergic, and V dies. D ought to be punished for stealing the cake, but V’s death cannot be said to be her responsibility any more than it would if she had purchased the cake quite properly. It seems extremely harsh automatically to hold D responsible for all the unforeseeable consequences simply because theft is a crime.
The accused intended to cause some harm, or was aware of the risk of doing so, and/or it was foreseeable that her conduct created the risk of causing some harm
4.35 As we saw in Part II, the present law provides that a person is guilty of ‘unlawful act manslaughter’ if she causes death by committing an act which is a crime in itself, and which carries a foreseeable risk of causing some injury to another person. A person who commits a relatively minor assault which unexpectedly causes death is thus guilty of manslaughter (D in the example at para 4.28 above).
4.36 Our respondents were divided on the question whether this type of manslaughter, or something very close to it, should continue to exist. A number of different reasons were given. For example, the Law Society… was of the view that ‘those who commit crimes involving, albeit slight, violence should take the consequences if the results turn out to be more catastrophic than they expected.’
Mr Justice Rix agreed:
‘It seems to me that once a person undertakes a violent act he sets himself deliberately … on a road which is not only seriously anti-social, … but potentially leading to calamitous results. … [H]e has deliberately embarked on an act of criminal violence, which it is, or ought to be, well known, leads to incalculable consequences.’
It is interesting that Rix J described the accused’s culpability in terms of failing to advert to a ‘well known’ risk of causing serious harm: this is similar to the first criterion upon which we insisted in our discussion of ‘culpable inadvertence’ above (para 4.18).
4.37 Dr John Gardner (a fellow of Brasenose College, Oxford), in his response, argued that the starting point in assessing criminal liability ought to be what the actor did and the consequences of her action:
‘The first question, in all cases of culpability, is ‘what did the defendant do?’, the answer to which will be some concrete action with results and circumstances incorporated into it already, e.g. ‘kill’ … . Then we must ask, naturally, to what extent the culpability is mitigated or moderated by the conditions under which the act was performed, including the accidental nature of the result etc. … It is not ‘why does the mere fact that someone happens to die add to one’s crime, or make a major crime out of an otherwise venal act?’, but rather, ‘how does the mere fact that one kills accidentally serve to mitigate or otherwise intercede in the wrongness of killing?’
4.38 He did not, however, maintain that all killings should be subject to criminal sanction.
First, he argued, the defendant must be culpable in some way, even if this culpability does not extend to the causing of death.
Secondly, principles of justice and the rule of law require that the killer must have some forewarning that her act will incur some criminal liability. On his view, then, unlawful act manslaughter is, in principle, perfectly acceptable:
‘… since the act was plainly dangerous, culpability is not eliminated, and this was still a wrongful killing. Then it is asked ‘what protections are required to make sure that the defendant has not been taken totally unawares by the law?’ – to which the answer is that the act must have been criminal under some other heading as well as dangerous, so as to put the defendant on legal notice.
(See also Jeremy Horder ‘A Critique of the Correspondence Principle in Criminal Law’ [1995] Crim LR 759, where it is argued that ‘the fact that I deliberately wrong V arguably changes my normative position vis-a-vis the risk of adverse consequences of that wrongdoing to V.’)
4.39 Unless one accepts moral luck arguments (paras 4.30 – 4.33), it is not clear why a person ought to be held criminally responsible for causing death if death or serious injury were the unforeseeable consequences of her conduct, just because she foresaw, or it was foreseeable, that some harm would result. Surely a person who, for example, pushes roughly into a queue is morally to blame for the foreseeable consequences of her actions – that a few people might get jostled, possibly even lightly bruised, and that people might get annoyed – but not for causing a death if, quite unexpectedly, she sets in train a series of events which leads to such an outcome. We consider that the criminal law should properly be concerned with questions of moral culpability, and we do not think that an accused who is culpable for causing some harm is sufficiently blameworthy to be held liable for the unforeseeable consequence of death.
4.40 One final argument in favour of recommending that a person ought to be liable for causing death, even if death or serious injury were not foreseeable consequences of her action, would be that this would be necessary for the protection of the public. This argument was considered by the Royal Commission on Capital Punishment ((1949–53) Report, Cmd 8932, paras 77–111) which, in 1953, recommended the abolition of the doctrine of constructive malice (i.e. it was murder if a person killed another, even quite accidentally, while committing a felony or while resisting an officer of justice) in murder:
‘We think it would be generally agreed that any liability for constructive crime offends against modern feeling, and that any departure from a subjective test of criminal liability can be justified, if at all, only if it is clearly established that it is essential for the protection of the public.’
4.41 The Royal Commission concluded that the public would be adequately protected by the existence of other criminal offences – principally, it has to be said, manslaughter.
4.42 Since, in the cases here under discussion (e.g. the example at para 4.28), the risk of causing death or serious injury was neither foreseen by the accused, nor foreseeable by her, it is difficult to see what deterrent effect would be achieved by imposing criminal liability for causing death which would not be achieved equally by imposing liability for the appropriate non-fatal offence.
4.43 In conclusion, we consider, as a matter of principle, that the criminal law ought to hold a person responsible for unintentionally causing death only in the following circumstances:
(1) when she unreasonably and advertently takes a risk of causing death or serious injury; or
(2) when she unreasonably and inadvertently takes a risk of causing death or serious injury, where her failure to advert to the risk is culpable because
(a) the risk is obviously foreseeable, and
(b) she has the capacity to advert to the risk.
4.44 We now turn to describe how these fundamental policy decisions, together with the views expressed on consultation, have influenced our detailed proposals for the reform of the law.
5.1 In Part IV we considered the circumstances in which a person ought to be held criminally liable for the unintentional causing of death, and concluded that this should be the case only where there was an obvious risk of causing death or serious injury, which he was capable of appreciating. In this part we consider in detail how this decision about liability should be reflected in the composition of individual offences…
5.2 As we have observed, involuntary manslaughter is an exceptionally broad category of offence. It seems to us to be inappropriate that types of conduct which vary widely in terms of fault should all carry the same descriptive label. Furthermore, its width can cause problems to the judge on sentencing, because he is unable to receive the jury’s guidance on matters that are crucial to the severity of the penalty deserved, such as the accused’s foresight of the risk of causing death.
5.3 For these reasons, we recommend the creation of two different offences of unintentional killing, based on differing fault elements, rather than one single, broad offence. We adopted a similar approach in the context of non-fatal offences in Law Com No 218, where we created a hierarchy of offences, graded by reference both to the seriousness of the injury caused and to the accused’s mental state. (Recommendation 1)
5.4 There might be some disadvantages in having separate offences.
First, there might be a danger of court time being wasted in legal argument as to where the exact borders of each offence lay. We do not believe, however, that this danger would be too great, since the offences that we recommend are defined in terms of easily understood degrees of fault.
Secondly, there might be a danger that if the prosecution is provided with a choice of separate offences, it might undercharge or accept pleas to lesser offences than would be appropriate. This is a potential problem wherever a hierarchy of offences is created, but a single very wide offence carries with it what we believe to be the much greater dangers to which we refer in paragraph 5.2 above.
5.5 The prosecution would not be disadvantaged by the creation of several different offences if it was unclear at the start of the trial whether, for example, the accused was aware of a risk of death or whether he displayed culpable inadvertence towards it, because it would always be possible to charge the separate offences in the alternative. If, for some reason, this procedure was not followed, the rules on alternative verdicts that we propose (in paras 5.57 – 5.60 below) would mean that the jury could convict of a less serious manslaughter offence on an indictment charging a more serious offence, even if the lesser offence was not specifically charged. Similarly, we propose later in this part that both of the new offences ought to be available as alternative verdicts on a charge of murder (paras 5.53 – 5.55 below).
5.6 The first of our proposed new offences… is ‘reckless killing’. This offence shares the same concept of ‘recklessness’ as the non-fatal offences in Law Com No 218 (para 4.7 above), and is drafted as follows:
A person who by his conduct causes the death of another is guilty of reckless killing if –
(a) he is aware of a risk that his conduct will cause death or serious injury; and
(b) it is unreasonable for him to take that risk having regard to the circumstances as he knows or believes them to be.
…
5.12 Our proposals would mean that, where both the defendant and the deceased knew that the defendant’s conduct involved a risk of death or serious injury to the deceased, but the deceased nevertheless consented to it – for example, where it consisted in the carrying out of a surgical operation – the defendant would be guilty of reckless killing if it was unreasonable of him to take that risk.
5.13 …[W]e recommend the creation of a new offence of reckless killing, which would be committed if:
(1) a person by his or her conduct causes the death of another;
(2) he or she is aware of a risk that his or her conduct will cause death or serious injury; and
(3) it is unreasonable for him or her to take that risk, having regard to the circumstances as he or she knows or believes them to be. (Recommendation 2)
5.14 In Part IV we concluded that, as a matter of principle, the criminal law is justified in holding a person liable for causing a death, which he neither intended nor foresaw, only in cases where he should have adverted to a risk of causing death or serious injury which was inherent in his conduct. A person is, we believe, at fault in failing to advert to such a risk only if it would have been obvious to a reasonable person in his position, and if he himself was capable of appreciating it (paras 4.17 – 4.42). As we explained above (paras 3.5 – 3.6), the form of manslaughter known as ‘unlawful act’ or ‘constructive’ manslaughter does not meet these criteria. It follows that this form of the offence is in our view inconsistent with the principles that we believe ought to govern criminal liability.
5.15 We are conscious that to many people this conclusion will seem to be at odds with common sense. The instinct to blame a person for what he has actually done, rather than for the aspects of his conduct that are blameworthy, is a powerful one, and we can sympathise with those who believe that this should be the basis of criminal liability for homicide. Indeed, to some extent it is reflected in our proposals, since we recommend the retention of a sharp distinction between those cases where death results and those where it does not: only in the former case would there be liability for gross carelessness, as distinct from recklessness. What we cannot accept is the proposition that, whenever death has resulted, and the person causing it can fairly be held responsible for the injury (however minor) that caused it, it is automatically fair to hold him responsible for the death. We believe that the law should be founded on principle rather than instinct; and we believe that liability for unlawful act manslaughter is unjustifiable in principle.
5.16 For this reason we recommend the abolition of unlawful act manslaughter in its present form. This would not of course mean that all those who would be convicted under the present law of unlawful act manslaughter would escape criminal liability altogether. The overwhelming majority of such cases would fall within one of the offences that we do propose: even if the defendant was not aware of the risk of death or serious injury (in which case he would be guilty of reckless killing) it would usually be possible to say that that risk was obvious and that he should have been aware of it – in which case he would be guilty of the offence of killing by gross carelessness that we propose below. In the minority of cases where this is not so, he could be prosecuted for the appropriate non-fatal offence. (Recommendation 3)
5.17 The second new offence which we recommend ought to be created is ‘killing by gross carelessness’…
…
5.24 …We believe that, so far as is possible, this new offence answers … many of the problems inherent in the present gross negligence offence (see paras 3.7 – 3.13 above).
5.25 The new offence is, to a certain extent, modelled on the test of ‘dangerousness’ in road traffic offences (Road Traffic Act 1988, s 2A(1), inserted by Road Traffic Act 1991, s 1, provides: ... a person is to be regarded as driving dangerously if ...(a) the way he drives falls far below what would be expected of a competent and careful driver, and (b) it would be obvious to a competent and careful driver that driving in that way would be dangerous.).
This is a test with which lawyers, the courts and the public are now familiar. Our researches have not been able to discover any criticism of the way in which the ‘dangerousness’ test in the Road Traffic Act operates in practice. Like the road traffic offences, the new offence is targeted at the person whose conduct falls far below that which could be expected from him, in the face of a risk which would have been obvious to a reasonable person in his position. It avoids reliance on the troubled concepts of ‘negligence’ and ‘duty of care’ (see paras 3.7 – 3.13 above).
…
5.27 It is an important element of the new offence that the risk of death or serious injury would have been obvious to a reasonable person in the accused’s position. ‘Obvious’ in this context means ‘immediately apparent’, ‘striking’ or ‘glaring’; we believe that a person cannot be blamed for failing to notice a risk if it would not have been obvious to a reasonable person in his place. We chose the word ‘obvious’ rather than, for example, ‘foreseeable’ because we think that the former is more generous to the defendant, and thus closer to the concept of culpable inadvertence discussed in Part IV (paras 4.12 – 4.42). Also, it is a word which, we believe, juries will readily understand…
5.28 It will be a question of fact for the jury in every case whether the risk that the accused’s conduct would cause death or serious injury would have been obvious to a reasonable person in his position. When considering this element, it must attribute to ‘the reasonable person’ any relevant facts within the knowledge of the accused at the time in question.
Thus if, for example, the accused broke into the house of an elderly person, and it is proved that he knew that his victim had a weak heart, this knowledge will be attributed to the reasonable person, and the jury may decide that it would have been obvious to such a person that the accused’s conduct carried a risk of causing death or serious injury to the victim.
Similarly, if the accused held himself out to possess any special skill or experience, the reasonable person will be credited with this. If, therefore, the accused was a surgeon carrying out an operation, and a risk of causing death or serious injury to the patient would have been obvious to a reasonable surgeon in his position, this element of the offence will be satisfied.
5.29 The next element of killing by gross carelessness is that the accused must have been capable of appreciating the risk at the material time. We explained why this is a necessary precondition of culpable inadvertence in Part IV (paras 4.20 – 4.22 above).
For the purposes of this new offence, it is immaterial whether the accused was not capable of appreciating the risk because of a permanent disability, such as blindness or low intelligence, or because he was temporarily tired or ill etc.
However, we would draw attention to the fact that the law at present allows the actus reus to be treated as a continuous course of conduct (see Thabo Meli [1954] 1 All ER 373; Le Brun [1992] QB 61; Fagan v Metropolitan Police Commissioner [1969] 1 QB 439; Miller [1983] 2 AC 161), so that if at any time during the actus reus the accused had the requisite capacity to appreciate material risks, he would be excluded from the protection of the clause.
An example of this type of defendant is the motorist who continues driving when he knows that he is very tired and eventually swerves into the opposite lane and causes an accident. It is possible that at the time of the accident he was so exhausted that he lacked the capacity to appreciate the risk inherent in swerving. However, if at some point during the time he was driving he had the capacity to realise that driving when very tired involves an obvious risk of causing death or serious injury, he could still fall within the scope of the new offence.
5.30 The Bill applies to the two new offences the present law relating to the effect of intoxication on criminal liability. Under the present law, where an allegation of recklessness (that is, awareness of risk) has to be proved, and the defendant was not aware of the risk in question because he was voluntarily intoxicated, the jury should be asked to consider whether he would have been aware of the risk had he been sober, and convict him if the answer to this question is yes. This principle will, of course, apply to the offence of reckless killing.
We know of no authority that applies the general rule to a requirement of capacity to recognise a risk, as distinct from a requirement that the risk be actually recognised, but we are confident that if the point arose the rule would be applied to the former kind of requirement too. Thus, if a person charged with killing by gross carelessness was not capable of appreciating the risk at the time in question because he was voluntarily intoxicated, through drink or drugs, the jury will have to disregard this and decide whether he would have had the relevant capacity had he been sober.
5.31 Finally, it must be proved either:
(i) that the accused’s conduct fell far below what could reasonably be expected of him in the circumstances, or
(ii) that he intended by his conduct to cause some injury or was aware of, and unreasonably took, the risk that it might do so.
This element of the new offence is intended to catch only the very worst cases in which a person inadvertently causes death, as is appropriate for a serious homicide offence.
5.32 The first of the two alternative ways of satisfying this element of the offence is similar to the test of ‘dangerousness’ in the offence of causing death by dangerous driving: the accused’s conduct must fall far below what could be expected from him. This formulation is intended to avoid the circularity of the Adomako formulation, although it would still leave a large degree of judgment to the jury, and this might lead to inconsistent verdicts being entered in different cases based on similar facts. We can see no way around this problem, without attempting to define the offence in such rigid and detailed terms that it would be unworkable. The jury are required to consider the accused’s conduct ‘in all the circumstances’. It could therefore consider, for example, the pressures and conditions under which he acted or failed to act.
5.33 However, it is not necessary to prove that the accused’s conduct fell far below the required standard if it can be shown that he intended to cause some injury to another, or was aware of a risk of doing so, which he unreasonably took.
We included this provision because…the elements of unlawful act manslaughter are easier to explain to juries, and simpler for them to understand, than gross negligence manslaughter; similarly, we think it will be easier for juries to decide whether the defendant acted intentionally or recklessly in respect of some injury than whether his conduct fell far below what could reasonably be expected.
This fact would not of course justify an alternative form of the offence if there were any danger of that alternative catching a defendant whose conduct is not seriously culpable; but we consider that a person who intentionally or recklessly causes some injury, thereby creating a risk of causing death or serious injury which ought to have been obvious to him, will always be seriously culpable.
(Unless of course he is justified in inflicting injury or taking the risk that he may do so, e.g. on grounds of self-defence. In that case he would be guilty of no offence if the only injury caused were the injury that he intends to cause or is aware that he may cause, and this alternative form of the gross carelessness offence would therefore not apply. The prosecution might nevertheless seek a conviction of the gross carelessness offence on the ground that, in the light of the obvious risk of death or serious injury, his conduct nevertheless fell far below what could reasonably be expected.)
Indeed it is hard to imagine circumstances in which this requirement would be satisfied but that of conduct falling far below the required standard would not. In other words, the alternative adds little or nothing to the reach of the offence; it serves only to simplify it for the jury, by dispensing with the need to consider a question which will almost inevitably be academic.
5.34 For all these reasons, we recommend the creation of a new offence of killing by gross carelessness, which would be committed if:
(1) a person by his or her conduct causes the death of another;
(2) a risk that his or her conduct will cause death or serious injury
would be obvious to a reasonable person in his or her position;
(3) he or she is capable of appreciating that risk at the material time; and
(4) either
(a) his or her conduct falls far below what can reasonably be expected of him or her in the circumstances, or
(b) he or she intends by his or her conduct to cause some injury, or is aware of, and unreasonably takes, the risk that it may do so, and the conduct causing (or intended to cause) the injury constitutes an offence. (Recommendation 4)
5.35 The following examples will illustrate how the new offence would operate in practice.
5.36 Example 1: D, a climbing instructor, took a group of inexperienced climbers out with inadequate equipment in very bad weather. They got trapped and one of them died.
In order to convict D, the jury would have to answer ‘yes’ to all the following questions:
(1) Would it have been obvious to a reasonable climbing instructor in D’s place that taking a group of inexperienced climbers out in the prevailing conditions would create a risk of causing death or serious injury to one of them?
(2) Was D capable of appreciating this risk? and
(3) Did his conduct fall far below what could reasonably be expected of him in all the circumstances?
5.37 Example 2: D caused V’s death by punching him in the head, not realising that serious injury might result; the impact of the blow caused a blood clot in the brain.
The jury would have to decide whether it would have been obvious to a reasonable person in D’s position that punching V as hard as he did would create a risk of causing death or serious injury, and whether D was capable of appreciating the risk at the time in question (unless he was incapable due to voluntary intoxication - see para 5.30 above). If the answer to both of these questions is ‘yes’, and if it is satisfied that D intended to cause some injury to V, or was reckless as to doing so, the jury must convict. If not, the accused may be convicted of the appropriate non-fatal offence in the alternative (see para 5.59 below).
…
5.42 As we observed in Part III, the law that governs the circumstances in which a positive duty to act arises, so as to impose criminal liability for death caused by omission, is very uncertain, particularly in relation to ‘voluntary undertakings’. The policy behind the present law has been the subject of severe criticism from academic lawyers (eg, Professor Andrew Ashworth, ‘The Scope of Criminal Liability for Omissions’ (1989) 105 LQR 424, and Professor Glanville Williams, ‘Criminal Omissions – the Conventional View’ (1991) 107 LQR 86: both attack the present position, although they have very different views on what should replace it)…
…
5.45 … [W]e recommend that the duty to act continue to be governed by the common law for the purposes of involuntary manslaughter for the time being. (Recommendation 5)
5.46 We have considered what should be the maximum sentence available on conviction of the offence we recommend. In the case of reckless killing, the answer is obvious: since the offence is intended to cover the most serious forms of involuntary manslaughter, it is clear that the maximum sentence must be the maximum currently available on a conviction of manslaughter – namely life imprisonment.
5.47 The appropriate maximum for our proposed offence of killing by gross carelessness is much harder to determine. Certainly it should, in our opinion, be a determinate sentence rather than life, because we regard the offence as less serious than that of reckless killing. That is not to say that there will not be some cases of killing by gross carelessness which are more serious than some cases of reckless killing. What we mean, by describing the offence of killing by gross carelessness as less serious, is that the worst examples of reckless killing (falling short of murder) will inevitably be more serious than the worst examples of killing by gross carelessness (falling short of reckless killing). There may be good reason to punish a person who fails to appreciate a risk of death which he ought to have appreciated, but we do not think that such a person could ever be as culpable as (for example) the terrorist who leaves a bomb in a public place, knowing that people may well be killed.
5.48 Similarly we think that killing by gross carelessness is (in this sense) a less serious offence, and should therefore be punishable with a lesser maximum sentence, than our proposed offence of intentionally causing serious injury, which we recommended should carry life imprisonment.
We are conscious that to many people this will seem paradoxical, since it would mean that an offence of causing non-fatal injury would be more serious than one of causing death. We recognise the strength of the concern to which this may give rise. However, we cannot meet this concern without abandoning the principle for which we have argued in Part IV above – that the actual outcome of the defendant’s conduct, though inevitably of great importance to his liability, ought to carry less weight than its culpability.
The fact that a person has caused death must clearly be a major factor in determining what offence, if any, he has committed, and our recommendations would make no change in that respect. They would mean, for example, that a person who through gross carelessness causes death would be guilty of a serious offence, whereas a person who is equally careless but causes non-fatal injury only, or no injury, would in general be guilty of no offence at all (i.e in the absence of a specific offence appropriate to the particular activity in question). It is the fact of death that justifies the imposition of – and under our proposals would continue to incur – liability for carelessness falling short of mens rea.
5.49 However, the principle by which we have been guided in formulating our recommendations is that the fact of death, though undeniably of great significance, cannot be as significant, for the purposes of criminal liability, as the moral culpability of the defendant’s conduct; and that the culpability of his conduct depends primarily on what consequences he intended to cause by it, what consequences he was aware it might cause, and what consequences he should have been aware it might cause. A person who intends to cause serious injury is clearly more culpable, and should be sentenced more severely, than one who intends to cause only minor injury. If the latter causes not just minor injury but death, that is a factor that ought to be reflected both in the offence and in the sentence; but in our view it ought still to be given less weight than the fact that what he intended was minor injury rather than serious injury, and non-fatal injury rather than death. From this perspective, the intentional causing of serious injury can indeed be more serious than the unintentional causing of death, even where the defendant intends to cause some injury; and a fortiori where he does not.
5.50 Therefore we conclude that the maximum sentence for killing by gross carelessness should be a determinate one. The question of how long it should be, however, we find much harder to answer. It is conventional for the most serious offences that do not carry life imprisonment to be punishable with 14 years’ imprisonment. It is true that determinate sentences of more than 14 years are sometimes imposed in the worst cases of involuntary manslaughter. Sentences of 18 years have been upheld for killing in the course of robbery, and of 15 years for manslaughter by arson; and killing in the course of rape might justify a comparable sentence. However, it seems highly probable that most such cases would fall within our proposed offence of reckless killing, and if so charged could therefore be punished with life imprisonment. In addition, robbery, arson and rape are themselves punishable with life imprisonment, so that there would be no question of the judge being unable to impose an appropriate overall sentence in such a case.
The real issue is whether it should be open to the judge to impose a sentence of (say) 15 years for the killing alone, as distinct from 15 years for the robbery, arson or rape and (say) 10 years concurrent for an unintentional killing in the course of it. Many would say that a sentence of the latter kind would be an affront to common sense, and that is a view with which we can sympathise. On the other hand we also see force in the view that the maximum sentence for an offence ought to be set at a level appropriate for the worst imaginable case of that offence, but should not attempt to reflect the gravity of all the other offences that might conceivably be committed at the same time.
…
5.52 …We therefore make no recommendation as to the maximum sentence for the offence of killing by gross carelessness.
5.53 The present law on alternative verdicts is to be found in section 6(2) of the Criminal Law Act 1967. This provides that a person found not guilty of murder on an indictment for murder may be convicted of, inter alia, manslaughter. It is noteworthy that it is a matter for the judge’s discretion whether he directs the jury about the option of finding the accused guilty of an alternative offence: he is under no obligation to do so.
This is important because in some cases the accused might be unfairly prejudiced by the suggestion, at a late stage in the trial, that he might be guilty of another offence which he has not had the opportunity to counter in the course of his defence. If, however, the possibility that the accused is guilty only of a lesser offence has fairly arisen on the evidence, and if directing the jury about it will not unnecessarily complicate the case, then the judge should, in the interests of justice, leave the alternative to them.
5.54 This judicial discretion will, perhaps, be even more important in relation to the new offence of killing by gross carelessness, if, as we recommend, it is able to stand as an alternative to murder. This is because in many murder trials evidence will not automatically have been presented on the elements of this new offence.
5.55 Indeed, murder cases in which it will be appropriate for the judge to leave this charge to the jury may occur relatively rarely; we believe, however, that such cases will be sufficiently frequent to justify providing for them in the legislation. For example, such a power might be useful in a case where D pours petrol through the letter box of his wife’s lover, V, and sets it alight, killing V in the fire. In such a case it might be difficult to be sure, until the evidence has been given, whether D:
(1) intended to kill or seriously injure V, or
(2) intended only to frighten V, but was aware that his conduct would create a risk of killing or seriously injuring him, or
(3) intended only to frighten V, but ought to have adverted to the risks created by his conduct. In such a case it would be convenient for the judge to have the power to leave alternative allegations of reckless killing and killing by gross carelessness to the jury, in the event that they find D not guilty of murder, provided he considers this to be in the interests of justice.
We therefore recommend that both of the new homicide offences should be available as alternative verdicts to murder. (Recommendation 6)
…
5.70 A common law principle, based on public policy, debars a person who has unlawfully killed another from acquiring a benefit in consequence of the killing,and therefore from taking any benefit under the victim’s will or intestacy.
5.71 The Forfeiture Act 1982 (‘the Act’) empowers the court, except where a person stands convicted of murder, to grant discretionary relief from this principle, to which the Act refers as the ‘forfeiture rule’. Where the court determines that the rule would apply to preclude a person who has unlawfully killed another from acquiring any interest in property, the court may make an order modifying the rule. The Act provides that the court should not make an order unless satisfied that, having regard to the conduct of the offender and of the deceased and to such other circumstances as appear material, the justice of the case requires the effect of the forfeiture rule to be so modified.
5.72 The forfeiture rule and the Act were considered in Re K (deceased) ([1985] Ch 85 (Vinelott J), [1986] Ch 180 (CA)), in which a wife who had been subjected to repeated violence at the hands of her husband killed him when a shotgun with which she was threatening him went off. She was convicted of manslaughter and placed on probation for two years. Vinelott J (whose decision was upheld by the Court of Appeal) held that the forfeiture rule applied, but granted relief under the Act. He took into account (among other things) the fact that the applicant had been a loyal wife who had suffered grave violence.
5.73 We have considered whether the Act should be amended in consequence of the proposed replacement of involuntary manslaughter with the two offences of reckless killing and of killing by gross carelessness.
5.74 As in the case of murder, a person cannot be convicted of reckless killing without having in mind the consequences of his actions. In murder, he must intend to kill or cause serious personal injury; in reckless killing he must be aware of the risk that his conduct will have one of those consequences. It seems arguable therefore that the considerations which prompted the exclusion of murder from the Act are applicable to reckless killing as well. …
5.75 Whatever the merits of the argument… we take the view that they cannot prevail against the following consideration. It seems from authorities such as Re K (deceased) that at present the court may exercise its discretionary power to grant relief against the forfeiture rule in circumstances where, had the proposed offence of reckless killing been available, the person seeking relief might have been convicted of that offence. In the absence of consultation on this issue we do not consider that it is properly open to us to make a recommendation that would remove from a person seeking relief, a right which he has at present, although no doubt discretion would only be exercised in favour of such a person in wholly exceptional circumstances.
5.76 We accordingly make no recommendation for the amendment of the Forfeiture Act 1982.
PART VI - CORPORATE MANSLAUGHTER: THE PRESENT LAW
6.1 Although a corporation is a separate legal person, it has no physical existence; and it can therefore act only through individuals who are its servants or agents. At one time it was thought that, for procedural reasons, a corporation was not indictable. Those procedural reasons have long since ceased to apply; and, as a matter of substantive law, two main techniques have been developed for attributing to a corporation the acts and states of mind of individuals it employs.
6.2 It is convenient to call the first technique ‘identification’. It originated in the 1940s and the concept has been developed in later cases. Broadly, under this doctrine those who control the corporation are treated, for the purpose of criminal liability, as embodying the corporation: the acts and states of mind of those who control a company are in law those of the company itself.
6.3 The second technique, of vicarious liability, emerged much earlier, in the nineteenth century. According to this doctrine a company is vicariously liable for the acts of any employee wherever an individual employer would be so liable.
6.4 Where neither technique applies, a corporation is not criminally liable. As Lord Hoffmann emphasised in a recent Privy Council decision, reference to a company ‘as such’ might suggest that there is something out there called the company of which one can meaningfully say that it can or cannot do something. There is in fact no such thing as the company as such…only the applicable rules. To say that a company cannot do something means only that there is no one whose doing of that act would, under the rules of attribution, count as an act of the company (Meridian Global Funds Management Asia Ltd v The Securities Commission [1995] 2 AC 500).
6.5 Many years ago it was thought that a corporation could not be indicted for any offence. However, Criminal Justice Act 1925, s 33(3) now provides:
‘On arraignment of a corporation, the corporation may enter in writing by its representative a plea of guilty or not guilty, and if either the corporation does not appear by a representative or, though it does so appear, fails to enter as aforesaid any plea, the court shall order a plea of not guilty to be entered, and the trial shall proceed as though the corporation had duly entered a plea of not guilty.’
…
6.7 …[T]he Interpretation Act 1978 defines ‘person’ as including ‘a body of persons corporate’ (s 5, Sch 1); and, in relation to such bodies, the definition applies to any Act ‘whenever passed relating to an offence punishable on indictment or on summary conviction’ (s 22, Sch 2, para 4(5)).
…
6.27 … In summary, the governing principle is that those who control or manage the affairs of a company are regarded as embodying the company itself. The introduction of this principle enabled criminal liability to be imposed on a corporation, whether as perpetrator or accomplice, for virtually any offence, notwithstanding that mens rea was required, and without having to rely on statutory construction.
6.28 The origin of the principle lies in a civil case (Lennard’s Carrying Co Ltd v Asiatic Petroleum Co Ltd [1915] AC 705) in which it was held that, for the purpose of a statute referring to ‘actual fault or privity’, the privity of the company’s manager was the privity of the company itself. In an oft-cited passage, Viscount Haldane LC said:
‘[A] corporation is an abstraction. It has no mind of its own any more than it has a body; its active and directing will must consequently be sought in the person of somebody who for some purposes may be called an agent, but who is really the directing mind and will of the corporation, the very ego and centre of the personality of the corporation.’
6.29 In 1957, Denning LJ, in another well-known passage (subsequently cited with approval and explained in Tesco Supermarkets Ltd v Nattrass [1972] AC 153),41 said:
‘A company in many ways may be likened to a human body. It has a brain and a nerve centre which controls what it does. It also has hands which hold the tools and act in accordance with directions from the centre. Some of the people in the company are mere servants and agents who are nothing more than hands to do the work and cannot be said to represent the mind or will. Others are directors and managers who represent the directing mind and will of the company, and control what it does. The state of mind of these managers is the state of mind of the company and is treated by the law as such.’ (H L Bolton (Engineering) Co Ltd v T J Graham & Sons Ltd [1957] 1 QB 159)
6.30 In the first of the 1944 cases, DPP v Kent and Sussex Contractors Ltd [1944] 1 KB 146, a company was charged with offences contrary to the Defence (General) Regulations 1939, of making use of a document (signed by the transport manager of the company) which was false in a material particular, with intent to deceive; and of making a statement (in the document) which it knew to be false in a material particular. The magistrates found that the servants of the company knew that the statement
was false, and used the document with intent to deceive, but they held that the company could not itself be guilty of the offences charged because it was not possible to impute the required mens rea to the company. The Divisional Court disagreed. Lord Caldecote CJ explained how a company can form a criminal intent:
‘I think that a great deal of [counsel for the company]’s argument on the question whether there can be imputed to a company the knowledge or intent of the officers of the company falls to the ground, because although the directors or general manager of a company are its agents, they are something more. A company is incapable of acting or speaking or even of thinking except in so far as its officers have acted, spoken or thought … . In the present case the first charge against the company was of doing something with intent to deceive, and the second was that of making a statement which the company knew to be false in a material particular. Once the ingredients of the offences are stated in that way it is unnecessary, in my view, to inquire whether it is proved that the company’s officers acted on its behalf. The officers are the company for this purpose … .’
6.31 Later that year, in ICR Haulage Ltd [1944] KB 551, a company was held indictable for common law conspiracy to defraud, another offence requiring mens rea to which vicarious liability could not apply. The corporation was not held responsible on the basis of liability for the acts of its agents; instead it was regarded as having committed the acts personally. DPP v Kent and Sussex Contractors Ltd [1944] KB 146 was treated as authority for the proposition that a state of mind could be attributed to a company. The last of the trio of 1944 cases was Moore v Bresler [1944] 2 All ER 515, which followed the two earlier decisions and is considered below.
6.32 This principle was developed thereafter on a case by case basis. The leading authority is the decision of the House of Lords in Tesco Supermarkets Ltd v Nattrass [1972] AC 153, in which a company was charged with an offence under the Trade Descriptions Act 1968. It invoked the defence of due diligence provided by the Act, and argued that the commission of the offence was due to the act or default of ‘another person’ – namely, the branch manager, who had failed to supervise the assistant who actually committed the offence. The defence was held to be available, on the ground that the branch manager was not part of the ‘mind’ of the company. The principle of identification, and the clear distinction between it and the doctrine of vicarious liability, were described by Lord Reid:
[A corporation] must act through living persons, though not always one or the same person. Then the person who acts is not speaking or acting for the company. He is acting as the company and his mind which directs his acts is the mind of the company. There is no question of the company being vicariously liable … . He is an embodiment of the company … and his mind is the mind of the company. If it is a guilty mind then that guilt is the guilt of the company.’
6.33 Referring to the distinction drawn by Lord Denning between the ‘brains and nerve centre’ of a company and its hands, Lord Diplock expressed the view that Lord Denning’s ‘vivid metaphor’ was not to be taken as authority for extending the class of persons whose acts were treated as those of the company itself beyond those entitled under its articles of association to exercise the company’s powers.
6.34 The distinction between vicarious liability and the liability of corporations under the identification principle was also emphasised more recently, in R v HM Coroner for East Kent, ex p Spooner (1989) 88 Cr App R 10. Bingham LJ said in that case:
‘It is important to bear in mind an important distinction. A company may be vicariously liable for the negligent acts and omissions of its servants and agents, but for a company to be criminally liable for manslaughter … it is required that the mens rea and actus reus of manslaughter should be established not against those who acted for or in the name of the company but against those who were to be identified as the embodiment of the company itself.’
6.35 Although Lord Denning’s dictum cited in paragraph 6.29 above was approved by the majority in the House of Lords in the Tesco case, the speeches showed variations in the detailed application of the test. Lord Reid said that a company may be held criminally liable for the acts only of ‘the board of directors, the managing director and perhaps other superior officers of a company [who] carry out the functions of management and speak and act as the company … .’
Viscount Dilhorne, on the other hand, said that a company should only be identified with a person ‘who is in actual control of the operations of a company or of part of them and who is not responsible to another person in the company for the manner in which he discharges his duties in the sense of being under his orders.’
Lord Diplock thought that the question was to be answered by ‘identifying those natural persons who by the memorandum and articles of association or as a result of action taken by the directors or by the company in general meeting pursuant to the articles are entrusted with the exercise of the powers of the company.’
Lord Pearson, too, thought that the constitution of the particular company should be taken into account.
6.36 The tests outlined above would, if applied strictly, produce rather different results. Viscount Dilhorne’s test would appear to be stricter than the others, since there are very few people in a company who are not responsible to others for the manner in which they discharge their duties. However, the general principle is clear: the courts must attempt to identify the ‘directing mind and will’ of the corporation, the process of such identification being a matter of law.
6.37 It is noteworthy that under the principles enunciated in the Tesco case a branch manager was not regarded as a controlling officer. Lord Pearson explained in that case:
‘In the present case the company has some hundreds of retail shops, and it would be far from reasonable to say that every one of its shop managers is the same person as the company … Supervision of the details of operations is not normally a function of higher management; it is normally carried out by employees at the level of foreman, chargehands, overlookers, floor managers and ‘shop’ managers (in the factory sense of ‘shop’).’
6.38 Although there is little direct authority on the matter, it would seem right in principle that the person who is identified with the corporation renders it liable only so long as she acts within the scope of her office. However, this requirement does not mean that the corporation’s liability is necessarily excluded where the activities in question are contrary to its interests. In Moore v Bresler Ltd [1944] 2 All ER 515, for instance, the respondent company was convicted of making false tax returns. The returns were actually made by the secretary of the company and the general manager of the branch concerned, and were designed to conceal their own fraudulent sale of company property. The court held that:
‘The sales undoubtedly were fraudulent, but they were sales made with the authority of the respondent company by these two men as agents for the respondent company … . These two men were important officials of the company, and when they made statements and rendered returns … they were clearly making those statements and giving those returns as officers of the company … . Their acts, therefore, … were the acts of the company.
6.39 It is not clear whether the principle of identification can apply to a director or official whose appointment is invalid. Dicta by Lord Diplock in the Tesco case suggest that it would not: he emphasised that ‘the obvious and only place’ to look in deciding whose acts are to be identified with the corporation is the constitution of the corporation, its articles and memorandum of association. This emphasis on the formal structure of the company would rule out anyone not validly appointed under the relevant Companies Act.
An indictment for manslaughter now lies against a corporation
…
6.42 In 1987 the decision of a coroner (who had held that a corporation could not be indicted for manslaughter) was challenged in an application for judicial review. The issue was not fully argued, but Bingham LJ saw no reason in principle why such a charge could not be established and ‘was tentatively of opinion’ that an indictment would lie (HM Coroner for East Kent, ex p Spooner (1989) 88 Cr App R 10).
6.43 In 1990 the same question was argued in depth in P & O European Ferries (Dover) Ltd (1991) 93 Cr App R 72. In that case Turner J comprehensively reviewed the authorities (including some in other jurisdictions) and concluded that an indictment for manslaughter would lie today against a corporation. Although this ruling has not yet been considered at appellate level, it is plainly of great persuasive authority.
6.44 Turner J outlined the development of corporate criminal liability. He pointed out that statements in works such as Coke, Hale, Blackstone and Stephen (which defined homicide as a killing by a human being) were not exclusive. Rather, they ‘reflected the historical fact that, at the dates when these definitions originated, the concept of criminal liability of a corporation, just as their very existence, was not within the contemplation of the courts or the writers of [those] legal treatises.’
…
6.47 Rejecting the argument… that a corporation could not, as a matter of substantive law, be indicted for manslaughter, Turner J considered in detail the subsequent authorities that had introduced and developed the principle of identification. That principle had transformed corporate liability since, by ‘identifying’ the corporation with the state of mind and actions of one of its controlling officers, it became possible to impute mens rea to a corporation and so to convict it of an offence requiring a mental element. Turner J concluded his summary of the English authorities as follows:
‘Since the nineteenth century there has been a huge increase in the numbers and activities of corporations … . A clear case can be made for imputing to such corporations social duties including the duty not to offend all relevant parts of the criminal law. By tracing the history of the cases decided by the English Courts over the period of the last 150 years, it can be seen how first tentatively and finally confidently the Courts have been able to ascribe to corporations a ‘mind’ which is generally one of the essential ingredients of common law and statutory offences. …
‘Once a state of mind could be effectively attributed to a corporation, all that remained was to determine the means by which that state of mind could be ascertained and imputed to a non-natural person. That done, the obstacle to the acceptance of general criminal liability of a corporation was overcome. …
‘[T]here is nothing essentially incongruous in the notion that a corporation should be guilty of the offence of unlawful killing. … [W]here a corporation, through the controlling mind of one of its agents, does an act which fulfils the prerequisites of the crime of manslaughter, it is properly indictable for the crime of manslaughter.
6.48 The first conviction of a company of manslaughter in English legal history took place in 1994, in Kite and OLL Ltd Winchester Crown Court, 8 December 1994, unreported. Since the company was a one-man concern whose ‘directing mind’ was plainly its managing director, the company’s liability was established automatically by his conviction.
6.49 The prosecution against P & O European Ferries (Dover) Ltd was terminated when Turner J directed the jury that, as a matter of law, there was no evidence upon which they could properly convict six of the eight defendants, including the company, of manslaughter. The principal ground for this decision in relation to the case against the company, was that, in order to convict it of manslaughter, one of the individual defendants who could be ‘identified’ with the company would have himself to be guilty of manslaughter. Since there was insufficient evidence on which to convict any of those individual defendants, the case against the company had to fail.
6.50 In coming to this conclusion Turner J ruled against the adoption into English criminal law of the ‘principle of aggregation’. This principle would have enabled the faults of a number of different individuals, none of whose faults would individually have amounted to the mental element of manslaughter, to be aggregated, so that in their totality they might have amounted to such a high degree of fault that the company could have been convicted of manslaughter. Because of the rejection of the ‘aggregation’ approach, the company could only be convicted if an individual who ‘could properly be said to have been acting as the embodiment of the company’ was also guilty.
6.51 In reaching his decision about the individual defendants, Turner J applied what was, in the period between Seymour [1983] 2 AC 493 and Prentice [1993] 3 WLR 927, thought to be the ruling law for manslaughter, the recklessness test of Caldwell and Seymour (see paras 2.12 – 2.13 above). He said:
‘Before any of these defendants … could be convicted ..., it was necessary for the prosecution to prove as against each such defendant not just one or more of the failures alleged against them in the indictment, but that – and this is the nub of the present situation – such failures were the result of recklessness in each defendant, in the now legally approved sense that they either gave no thought to an obvious and serious risk that the vessel would sail with her bow doors open, when trimmed by the head, and capsize, in circumstances unknown to shipboard management, or, alternatively, that if thought or consideration to that risk was given, each defendant, nevertheless, went on to run it.’
6.52 There was insufficient prosecution evidence to justify a finding that the risk of the vessel putting to sea with her bow doors open was ‘obvious’ within the Caldwell/Lawrence definition. The appropriate test of ‘obviousness’ in this case was
‘what the hypothetically prudent master or mariner or whosoever would have perceived as obvious and serious.’
This formulation was not disputed by the prosecution, and it was undoubtedly the correct approach to take since an ordinary person, with no experience of shipping, could not be expected to perceive this possibility as an obvious risk in an unfamiliar and complex system.
6.53 Turner J rejected the prosecution argument that the test should operate in a similar way to the test of foreseeability employed in cases of civil negligence, so as to allow the jury to infer that the risk of the ship sailing with her bow doors open was obvious from the very fact that the safety system in place was defective and that this defect had allowed that eventuality to occur. Referring to Andrews [1937] AC 576 (para 2.5 above), he emphasised that recklessness in manslaughter was intended to be more culpable than ordinary civil negligence: the criterion of reasonable foreseeability of the risk was not appropriate. Instead, it was necessary to show that the risk was ‘obvious’ in the sense that it would actually have occurred to a reasonably prudent person in the position of the defendant. What was required was
‘some evidence upon which the jury, being properly directed, can find that the particular defendant failed to observe that which was ‘obvious and serious’, which words themselves convey a meaning that the defendant’s perception of the existence of risk was seriously deficient when compared to that of a reasonably prudent person engaged in the same kind of activity as that of the defendant whose conduct is being called into question.’
6.54 The prosecution evidence did not go far enough on this issue. It consisted of the testimony of a number of ships’ masters who were, or had been, in the employment of the defendant company, who all said that it had not occurred to them that any risk existed, let alone that it was an obvious one. This evidence alone would not have been fatal. Indeed, it might even have advanced the prosecution case against the defendant company, since it supported the allegation that no-one in the company had given any thought to the risk, within the first limb of Caldwell recklessness. However, the prosecution was not able to prove through the testimony of witnesses from outside the defendant company that the risk was ‘obvious’. Turner J referred to the evidence of witnesses from other shipping lines as to the practice adopted on various of their ships:
‘I do not understand that the statements of any of these witnesses condescend to criticism of the system employed by the defendants in this case as one which created an obvious and serious risk, except to the extent that any legitimate deduction may be made from the fact that they took precautions other than those employed by any of these defendants.’
6.55 For these reasons the prosecution against the ferry company failed, despite the findings of a judicial inquiry, in the Sheen Report (MV Herald of Free Enterprise: Report of the Court No 8074, Department of Transport (1987)), that all concerned in management must be regarded as sharing responsibility for the failure of management and that from top to bottom the body corporate was infected with the disease of sloppiness.
Even if Turner J… had approached the issue of individual liability on the basis of gross negligence rather than of Caldwell recklessness, it seems likely that he would have reached the same conclusion. The dominant test remained the test set out in Bateman (1925) 19 Cr App R 8, 14 (para 2.10 above), of doing something which no reasonably skilled doctor would have done. On this approach, based as it is on the practices of the relevant profession or industry, it would have been difficult to prove that the mode of operation of this ship, although not that of other companies, fell seriously below prevailing standards.
6.56 Evidence of the type adduced before Turner J would also present difficulties to the prosecution even if, in the case of a corporate defendant, it were possible to apply some version of the aggregation approach, and to look more widely, and not merely at the responsibility of individuals. The fact that none of the witnesses saw the method of operating the vessel as creating an obvious and serious risk of disaster might be thought to suggest that the company’s attitude and method of organisation, which had been so seriously criticised by the Sheen inquiry, were not unique within the industry.
…
8.1 … In this Part we consider … how the corporate offence should therefore be defined. …
8.2 For the offence of killing by gross carelessness, it must be proved:
(1) that the defendant’s conduct caused the death,
(2) that the risk of death or serious injury would have been obvious to a reasonable person in her position, and that she was capable of appreciating that risk, and
(3) that her conduct fell far below what could reasonably be expected of her in the circumstances.
8.3 In our view, the second of these requirements cannot appropriately be applied to corporations, which, as Lord Hoffmann has recently emphasised (Meridian Global Funds Management Asia Ltd v The Securities Commission [1995] 3 WLR 413), are only metaphysical entities. To hypothesise a human being who could be in the same position as the corporation is a logical impossibility, and it would therefore be meaningless to enquire, as in the offence of killing by gross carelessness, whether the risk would have been ‘obvious’ to such a person. Moreover, corporations have no ‘capacity’, in the sense in which we use that term in this report in relation to an individual, so that it would be equally impossible to enquire whether the defendant corporation had the capacity to appreciate the risk. It is also, in our view, unnecessary. In judging the conduct of an individual defendant, the law must in fairness take account of such personal characteristics as may make it harder for her to appreciate risks that another person would appreciate; but the same considerations scarcely apply to a corporate defendant.
8.4 We have therefore concluded that the foreseeability of the risk, either to a hypothetical person in the defendant’s position or to the defendant itself, should not be included in the definition of the corporate offence. This will not prevent juries from finding (in general terms) that the risk was, or should have been, obvious to any individual or group of individuals within the company who were or should have been responsible for taking safety measures, in deciding whether the company’s conduct fell below the required standard. Nor would we wish to discourage the jury from approaching its task in that way. We are simply concerned, in formulating the new offence, to remove the legal requirement under the present law to identify individuals within the company whose conduct is to be attributed to the company itself.
8.5 On the other hand we see no reason why the third requirement for the individual offence, that the defendant’s conduct must have fallen far below what could reasonably be expected of her in the circumstances, should not apply equally to the corporate offence. This approach… is based on our view that the offence ought to be one of last resort, available only when all the other sanctions that already exist seem inappropriate or inadequate, and that, therefore, the negligence in question must have been very serious.
8.6 We have therefore concluded, for the same reasons, that the new corporate offence should be committed only where the defendant’s conduct fell far below what could reasonably be expected of it ‘in the circumstances’. In our view, it would be neither practicable nor desirable to specify in legislation what those ‘circumstances’ should or should not include: in every case it would be for the jury to decide whether the corporation’s conduct fell within that description. In many cases this would involve the jury in balancing such matters as the likelihood and possible extent of the harm arising from the way in which the company conducted its operations against the social utility of its activities and the cost and practicability of taking steps to eliminate or reduce the risk of death or serious personal injury.
8.7 The jury might also think it right to take account of the extent (if any) to which the defendant corporation’s conduct diverged from practices generally regarded as acceptable within the trade or industry in question. This could not be conclusive, since the fact that a given practice is common does not in itself mean that the observance of that practice cannot fall far below what can reasonably be expected; but it might well be highly relevant. The weight to be attached to it, if any, would be a matter for the jury.
8.8 Of the three requirements for the individual offence of killing by gross carelessness, therefore, we envisage that the second (namely the obviousness of the risk, and the defendant’s capacity to appreciate it) should be discarded for the purposes of the corporate offence, whereas the third (namely that the defendant’s conduct in causing the death should have fallen far below what could reasonably be expected) should be retained. It remains to be determined what should be done about the first, namely that the defendant’s conduct should have caused the death.
Obviously that requirement must be retained in some form; equally obviously (in the light of the difficulties that we have explored in determining whether particular conduct can, under the present law, be regarded as the conduct of a company and not merely of its human agents), it must be adapted for the purposes of the corporate defendant. There are two aspects to this requirement:
first, the defendant must have acted, or omitted to act, in a particular way; and
second, the death must have resulted from that act or omission.
In the case of an individual defendant it is rarely necessary to distinguish these two aspects: once the facts are known, there is no difficulty in distinguishing the defendant’s conduct from someone else’s. In the case of a corporate defendant, however, this distinction is problematic. Since we have rejected the option of attributing to the corporation everything done (or not done) by its agents, we must find a way of identifying that conduct which can properly be attributed to it.
The question is: in what circumstances can it properly be said, not merely that the conduct of a corporation’s agents has caused a death, but that the conduct of the corporation itself has done so?
8.9 In answering this question we have not had to start with an entirely clean slate. In the first place we have borne in mind the analogy of the identification principle laid down in Tesco Supermarkets v Nattrass [1972] AC 153 (above, paras 6.32 – 6.33, 6.35 – 6.39), which distinguishes between those agents of a company that qualify as its ‘controlling minds’ and those that do not.
…[W]e do not think that this principle is in itself sufficient for the imposition of corporate liability in every case of homicide where such liability would be justified; but the main reason for this is that the principle requires the prosecution to identify one or more ‘controlling minds’ who are themselves guilty of a homicide offence.
The distinction drawn in the Tesco case between things done in the management and organisation of the company on the one hand, and things done at a purely operational level on the other, seems to us to encapsulate the nature of the distinction that we need to draw. The difference between our approach and the identification principle is that we think the distinction should be drawn in terms of the kind of conduct that can incur liability, rather than the status of the person or persons responsible for it.
8.10 Secondly, we have drawn on the law governing an employer’s common law obligation to take care for the safety of employees, and one aspect of that obligation in particular – namely, the employer’s duty to provide a safe system of work. This obligation is personal to the employer and is quite distinct from any vicarious liability that may arise in respect of injury caused to an employee by a fellow employee in the course of their employment. A breach of this obligation is not just negligence for which the employer is (vicariously) responsible: it is the employer’s own negligence.
The distinction thus corresponds to the distinction that we seek to draw, in the case of a corporate employer, between the conduct of the corporation and the conduct of its employees alone; and it is because of this analogy that we have taken this obligation as a starting-point in defining the kind of conduct that we propose as an element of the new corporate offence. In effect, we propose to use it as a model for the duty of every corporation to all those (not just employees) who may be affected by the corporation’s activities.
…
8.20 Under our proposals, individuals within the company could be concurrently liable, in respect of an incident for which the company was liable, for the offence of killing by gross carelessness; and, whether or not they were so liable, their conduct might be relevant to the corporate offence as part of the circumstances surrounding that offence. For the purpose of the corporate offence and by contrast with the present law, however, there would be no need to identify the controlling officers of the company. The question would be whether there had been a management failure, rather than, as at present, whether there was blameworthy conduct on the part of any individual or group of individuals which should be attributed to the company.
8.21 To take a simple hypothetical example, if a lorry driver employed by a company causes death by dangerous driving in the course of the company’s business, this act would not of itself involve a management failure so as to incur corporate liability; nor would the company be vicariously liable for the driver’s negligence. The company might be liable, however, if the incident occurred because the driver was overtired at the material time in consequence of a requirement to work excessively long hours, or because she consistently worked very long hours in her desire to earn overtime, and the company had no adequate system of monitoring to ensure that this did not happen.
…
8.23 We accept that there will be some cases in which the jury will have to draw a somewhat fine line between an employee’s ‘casual’ negligence and a management failure. Such cases abound in the field of employer’s liability. …
…
8.34 …[T]here is no reason in principle why a ‘management failure’ within the meaning of the proposed offence should not consist in a failure to provide safe premises or equipment, or competent staff. Nor do we suggest that the offence should be defined as hinging on whether the corporation’s civil liability for the death would be personal or vicarious: in the context of criminal trials, such a test would be unworkable…
…[W]e believe that the distinction between ‘management failure’ and operational negligence is an appropriate way of differentiating, in the context of involuntary homicide, between the conduct of a corporation and the conduct of its employees alone. Moreover, we would emphasise that a corporation would be liable only in extremely limited circumstances, namely where its conduct fell far below what could reasonably be expected of it in the circumstances. The offence would be confined to cases of very serious negligence.
8.35 We therefore recommend:
(1) that there should be a special offence of corporate killing, broadly corresponding to the individual offence of killing by gross carelessness;
(2) that (like the individual offence) the corporate offence should be committed only where the defendant’s conduct in causing the death falls far below what could reasonably be expected;
(3) that (unlike the individual offence) the corporate offence should not require that the risk be obvious, or that the defendant be capable of appreciating the risk; and
(4) that, for the purposes of the corporate offence, a death should be regarded as having been caused by the conduct of a corporation if it is caused by a failure, in the way in which the corporation’s activities are managed or organised, to ensure the health and safety of persons employed in or affected by those activities. (Recommendation 11)
8.36 Our proposed concept of ‘management failure’ is an attempt to define what, for the purposes of a corporate counterpart to the individual offence of killing by gross carelessness, can fairly be regarded as unacceptably dangerous conduct by a corporation. But it must of course be proved, as in the individual offence, that the defendant’s conduct (which, in the present context, means the management failure) caused the death.
To a large extent this will involve the application of the ordinary principles of causation, as in any other homicide offence. If, for example, the jury are not satisfied beyond reasonable doubt that the death would not have occurred had it not been for the management failure, the offence will not be proved.
Even if the death would not otherwise have occurred, it will be open to the jury to conclude that the ‘chain of causation’ was broken by some unforeseeable act or event, and that the management failure was not itself a cause of the death but merely part of the events leading up to it. If, for example, the management failure consisted of a failure to ensure that some potentially dangerous operation was properly supervised, a jury would be unlikely to conclude that this failure caused the death if the immediate cause was a deliberate act by an employee rather than a merely careless one – even if that act would probably not have occurred had a supervisor been present.
8.37 However, we think that the scope for any defence of a ‘break in the chain of causation’ should be very limited. In many, perhaps most, cases it will be the operational negligence of one or more of the company’s employees that is most closely connected in point of time with the death.
For example, the immediate cause of the death might be the failure of an employee, through lack of attention, to give a signal which she was employed to give. Indeed, depending on the circumstances, the employee in question may personally be guilty of our proposed offence of killing by gross carelessness. It does not, in our view, follow that the employee’s conduct should in itself absolve the corporation from liability, because the management failure may have consisted in a failure to take precautions against the very kind of error that in fact occurred.
If a company chooses to organise its operations as if all its employees were paragons of efficiency and prudence, and they are not, the company is at fault; if an employee then displays human fallibility, and death results, the company cannot be permitted to deny responsibility for the death on the ground that the employee was to blame. The company’s fault lies in its failure to anticipate the foreseeable negligence of its employee, and any consequence of such negligence should therefore be treated as a consequence of the company’s fault.
8.38 It is not clear how far the ordinary law of causation takes account of this reasoning. As Professor Ashworth (Principles of Criminal Law (2nd ed 1995)) has explained:
‘[T]he principle of individual autonomy presumes that, where an individual who is neither mentally disordered nor an infant has made a sufficient causal contribution to an occurrence, it is inappropriate to trace the causation any further. This is taken to justify not only picking out D’s conduct from other possible causes and regarding that conduct as operating on a ‘stage already set’, but also declining to look behind D’s conduct for other persons who might be said to have contributed to D acting as he or she did.’
8.39 In our view, therefore, there is a danger that, without more, the application of the ordinary rules of causation would in many cases result in a management failure being treated as a ‘stage already set’, and hence not linked in law to the death.
In our view the legislation should include an express provision to the effect that in this kind of situation the management failure may be a cause of the death, even if the immediate cause is the act or omission of an individual. Whether in all the circumstances the management failure is a cause of the death, in spite of the intervening act or omission of an individual, will be a matter for the common sense of the jury.
We recommend that, for the purposes of the corporate offence, it should be possible for a management failure on the part of a corporation to be a cause of a person’s death even if the immediate cause is the act or omission of an individual. (Recommendation 12)
INDEPENDENT CONTRACTORS
The issue
8.40 A corporation may employ an independent contractor to carry out work in a variety of situations. One who engages an independent contractor is not normally liable to others for the negligence of that contractor; and an employer’s duty of care in tort does not render her liable to her employees for injury sustained through the negligence of her contractor, save in exceptional circumstances ((There are several exceptions. They include the case in which the employer fails to coordinate the activities of subcontractors (McArdle v Andmac Roofing Co [1967] 1 WLR 356). Another exception arises where the employer exercises control over the contractor’s operations, in the sense that she can tell the contractor’s employees what to do and what safety precautions to adopt, or where she exercises joint or partial control over them in that respect (see, eg, Associated Octel Co Ltd [1994] 4 All ER 1051)).
…
8.44 We believe that there is no need to make specific provision in the present context in relation to the employment of a contractor by the company. In every case it will be for the jury to determine:
(1) whether a death of which the immediate cause was the conduct of a contractor employed by the company was attributable, at least in part, to a management failure on the part of the company, and
(2) if so, whether that failure amounted to conduct falling far below what could reasonably be expected of the company in the circumstances…
8.45 We will now show how we envisage the new offence would operate by reference to the 1987 Zeebrugge ferry disaster, which involved the ‘roll-on roll-off’ passenger and freight ferry, Herald of Free Enterprise (see paras 6.49 – 6.56 above). The ferry set sail from Zeebrugge inner harbour and capsized four minutes after crossing the outer mole, with the loss of 150 passengers and 38 crew members.
The immediate cause of the capsize was that the ferry had set sail with her inner and outer bow doors open. The responsibility for shutting the doors lay with the assistant bosun, who had fallen asleep in his cabin, thereby missing the ‘Harbour Stations’ call and failing to shut the doors.
The Chief Officer was under a duty as loading officer of the G deck to ensure that the bow doors were closed, but he interpreted this as a duty to ensure that the assistant bosun was at the controls.
Subsequently, the report of the inquiry by Mr Justice Sheen into the disaster (‘the Sheen Report’) said of the Chief Officer’s failure to ensure that the doors were closed that, of all the many faults which combined to lead directly or indirectly to this tragic disaster, his was the most immediate.
The Chief Officer could in theory have remained on the G deck until the doors were closed before going to his harbour station on the bridge. However, although this would have taken less than three minutes, loading officers always felt under such pressure to leave the berth immediately that this was not done.
8.46 The Master of the ferry on the day in question was responsible for the safety of the ship and those on board. The inquiry therefore found that in setting out to sea with the doors open he was responsible for the loss of the ship. The Master, however, had followed the system approved by the Senior Master, and no reference was made in the company’s ‘Ship’s Standing Orders’ to the closing of the doors. Moreover, this was not the first occasion on which the company’s ships had gone to sea with doors open, and the management had not acted upon reports of the earlier incidents.
8.47 The Senior Master’s functions included the function of acting as co-ordinator between all the Masters who commanded the Herald and their officers, in order to achieve uniformity in the practices adopted on board by the different crews. He failed to enforce such orders as had been issued, and also failed to issue orders relating to the closing of the bow doors on G deck. The Sheen Report found that he ‘should have introduced a fail-safe system’.
8.48 The criticism in the Sheen Report did not stop with those on board the ship:
‘[F]ull investigation into the circumstances of the disaster leads inexorably to the conclusion that the underlying or cardinal faults lay higher up in the Company [than the Master, the Chief Officer, the assistant bosun and the Senior Master]. The Board of Directors did not appreciate their responsibility for the safe management of their ships. They did not apply their minds to the question: What orders should be given for the safety of our ships? The directors did not have any proper comprehension of what their duties were.
‘There appears to have been a lack of thought about the way in which the Herald ought to have been organised for the Dover/Zeebrugge run. All concerned in management, from the members of the Board of Directors down to the junior superintendents, were guilty of fault in that all must be regarded as sharing responsibility for the failure of management. From top to bottom the body corporate was infected with the disease of sloppiness. … The failure on the part of the shore management to give proper and clear directions was a contributory cause of the disaster.’
8.49 As we explained above (paras 6.49 – 6.56 above), the prosecution against P & O European Ferries (Dover) Ltd ultimately failed. The judge directed the jury that, as a matter of law, there was no evidence upon which they could properly convict six of the eight defendants, including the company, of manslaughter.
The principal ground for this decision in relation to the case against the company, was that, in order to convict it of manslaughter, one of the individual defendants who could be ‘identified’ with the company would have himself to be guilty of manslaughter. Since there was insufficient evidence on which to convict any of those individual defendants, the case against the company had to fail.
8.50 If circumstances such as these were to occur again, we think it would probably be open to a jury to conclude that, even if the immediate cause of the deaths was the conduct of the assistant bosun, the Chief Officer or both, another of the causes was the failure of the company to devise a safe system for the operation of its ferries; and that that failure fell far below what could reasonably have been expected. In these circumstances the company could be convicted of our proposed new offence.
POTENTIAL DEFENDANTS
Corporations
8.51 We consider that the new offence should extend to all corporations, irrespective of the legal means by which they are incorporated. This would include not only those incorporated under a general public Act (such as the Companies Act 1985) but also those incorporated at common law (such as the Corporation of London), by royal charter (such as the BBC, and most universities), by private or local Act (such as certain public utility companies) or special public Act (including a number of organisations in the public sector). Most of these corporations have no shareholders and are not run with a view to profit, but we do not regard this as a reason for exempting them from the rules applicable to other corporations.
8.52 We also think that the offence should extend to corporations incorporated abroad. If a death results from the mismanagement of a company, we see no reason why the company’s liability should be affected by the place where it happens to have been incorporated, any more than the liability of an individual (for things done in England and Wales) is affected by her nationality. We do not propose, in general, that the offence should be committed where the fatal injury occurs outside England and Wales; but this is a question of the offence’s territorial extent. It does not follow that foreign corporations should be immune from prosecution in respect of fatal accidents that do occur in England and Wales.
8.53 However, we propose that corporations sole should be excluded. A corporation sole is a corporation constituted in a single person in right of some office or function, which grants that person a special legal capacity to act in certain ways: examples are government ministers and archbishops. The corporation sole is in reality a legal device for differentiating between an office-holder’s personal capacity and her capacity qua holder of that office for the time being. It is expressly excluded from the definition of a corporation in section 740 of the Companies Act 1985, and we exclude it from our proposed corporate offence as well. We recommend that the offence of corporate killing should be capable of commission by any corporation, however and wherever incorporated, other than a corporation sole. (Recommendation 13)
8.54 We have considered whether the proposed new offence should apply to partnerships, trusts (such as hospital trusts) and other unincorporated bodies. Many such organisations are for practical purposes indistinguishable from corporations, and it is arguable that their liability for fatal accidents should be the same. However, we have concluded that it would be inappropriate for us to recommend such an extension of the offence at the present time. Under the existing law the individuals who comprise an unincorporated body may be criminally liable for manslaughter, as for any other offence; and, by contrast with the law relating to corporations, the question of attributing the conduct of individuals to the body itself does not arise. In this respect the law will be unaffected by the replacement of manslaughter with the offences in the draft Bill of reckless killing and killing by gross carelessness.
8.55 It would clearly be wrong to extend the offence to all unincorporated bodies, because there are many such bodies (for example, a partnership of two individuals, employing no-one) that would be unfairly disadvantaged by being charged with the corporate offence (which does not require foreseeability - see para 8.4 above) rather than that of killing by gross carelessness (which does). Any extension of the offence beyond incorporated bodies would therefore raise intractable problems as to the kinds of unincorporated body that ought and ought not to be included. But there has been no consultation on any proposal to this effect, either in the consultation paper or in any other form. We think it would be wise to await experience of the operation of our proposed corporate offence, in the context of the kind of organisation for which it is primarily designed – namely the commercial corporation – before considering whether to extend it further.
8.56 …We recommend that the offence of corporate killing should not be capable of commission by an unincorporated body. (Recommendation 14)
…
8.63 We are very conscious of the strength of feeling understandably engendered by fatal accidents, and of how much pressure there can be for a prosecution. At present it is initially for the Crown Prosecution Service (‘CPS’) to decide whether there is sufficient evidence to offer a realistic prospect of a conviction, and (if so) whether the public interest requires a prosecution; but if the CPS decides not to prosecute, on either ground, a private individual (such as a relative of the deceased) may either seek judicial review of the decision or bring a private prosecution.
[An example is] the sinking of the pleasure cruiser Marchioness on the River Thames on 20 August 1989, where judicial review was sought, without success, of the DPP’s decision to charge only an offence under s 32 of the Merchant Shipping Act 1988 against the master of the Bowbelle. An application for judicial review would succeed only in limited circumstances, for example, where the DPP was shown to have acted in bad faith or to have failed to apply the Code for Crown Prosecutors as in R v DPP, ex p C [1995] 1 Cr App R 136.
In R v Bow Street Stipendiary Magistrate, ex p South Coast Shipping Co Ltd (1993) 96 Cr App R 405 it was held that the DPP’s decision not to bring manslaughter charges in respect of the Marchioness disaster did not preclude the bringing of a private prosecution, subject to the right of the DPP to intervene in the proceedings and discontinue them.
The CPS has power to take over a private prosecution and discontinue it (Prosecution of Offences Act 1985, ss 6(2)), but will not necessarily think it appropriate to do so merely because it decided not to institute proceedings itself. A decision to discontinue may be open to judicial review (Turner v DPP (1979) 68 Cr App R 70).
Private prosecutions are also controlled to some extent by the magistrates’ court… [T]he court can decline to issue a summons if the proceedings appear to be vexatious; but such a refusal can itself be challenged by judicial review…
8.64 We have considered whether these procedures are a sufficient safeguard against the risk of private prosecutions for the corporate offence in cases where the CPS’s decision not to prosecute is entirely justified. The effect of our proposed offence would be to make it easier to secure a conviction against a company whose operations have caused a death. It might therefore be argued that, if the evidence would be less likely to be held insufficient, it must also be less likely that proceedings would be brought on insufficient evidence.
Moreover, the incidence of vexatious proceedings for manslaughter does not at present seem to be unduly high. However, this may be largely a consequence of the financial risk involved in bringing private proceedings that may result in an acquittal and an order for costs; the easier the offence is to prove, the smaller that risk will be perceived to be, and the more likely it is that private proceedings will be brought. And a proportion of those proceedings will undoubtedly be in cases that are clearly inappropriate for prosecution, even under the less restrictive rules that we propose.
8.65 We are aware that the definition of the offence we propose is in broad terms and relies to an unusual degree on the judgment of the jury. There will therefore be many cases where, although a jury would be unlikely to convict, it cannot be said that no reasonable jury could convict. In these cases the courts would have no power to prevent a private prosecution from going ahead (unless the proceedings appeared to be an abuse of the process of the court, which would be unlikely if there were a prima facie case), and it would be up to the CPS to intervene and discontinue the proceedings on the ground that there is no ‘realistic prospect of a conviction’ – in other words, that an acquittal is a more likely outcome than a conviction. In such a case the CPS will not begin or continue a prosecution: the question whether the public interest requires a prosecution does not arise.
8.66 However, the right of a private individual to bring criminal proceedings, subject to the usual controls, is in our view an important one which should not be lightly set aside. Indeed, in a sense it is precisely the kind of case with which we are here concerned, where the public pressure for a prosecution is likely to be at its greatest, that that right is most important: it is in the most serious cases, such as homicide, that a decision not to prosecute is most likely to be challenged. It would in our view be perverse to remove the right to bring a private prosecution in the very case where it is most likely to be invoked.
We recommend that there should be no requirement of consent to the bringing of private prosecutions for the corporate offence. (Recommendation 17)
8.67 …
We recommend that the offence of corporate killing should be triable only on indictment. (Recommendation 18)
…
THE COURT’S POWERS ON CONVICTION
8.71 The court would have its ordinary powers to order compensation.
8.76 [W]e recommend that:
(1) a court before which a corporation is convicted of corporate killing should have power to order the corporation to take such steps, within such time, as the order specifies for remedying the failure in question and any matter which appears to the court to have resulted from the failure and been the cause or one of the causes of the death;
(2) the power to make such an order should arise only on an application by the prosecution (or the Health and Safety Executive or any other body or person designated for this purpose by the Secretary of State, either generally or in relation to the case in question) specifying the terms of the proposed order; and
(3) any such order should be on such terms (whether those proposed or others) as the court considers appropriate having regard to any representations made, and any evidence adduced, by the prosecution (or any other body or person applying for such an order) or on behalf of the corporation. (Recommendation 20)
8.77 We recommended above (para 8.56) that there should be no question of individual liability for the corporate offence, because that offence is intended as a practical device to ensure that corporations cannot escape liability for killing by gross carelessness merely because their decision-making structures are large and complex.
It does not follow, in our view, that there should be no corporate liability for the offences we have (for convenience) referred to as the individual offences. The existence of the corporate offence would normally make it unnecessary for the prosecution to charge a corporation with reckless killing or killing by gross carelessness, and thus undertake the burden of showing that a ‘controlling mind’ of the corporation was guilty of the offence charged: even if no such person could be identified, the corporation could still be convicted of a homicide offence if the death were caused by a management failure of the requisite gravity.
But, just because it would not normally be necessary to charge the corporation with an individual offence, it does not follow that it would never be appropriate; still less does it follow that it should not be possible. There may be the occasional case where, although under the identification principle the conduct of the individual responsible is the conduct of the company, it is arguable that that conduct does not amount to a management failure. Even where this is not the case… we see no reason why it should not continue to be possible for the company to be convicted of the same offence as the individual responsible.
We recommend that the ordinary principles of corporate liability should apply to the individual offences that we propose. (Recommendation 21)
1. We recommend the creation of two different offences of unintentional killing, based on differing fault elements, rather than one single, broad offence.
2. We recommend the creation of a new offence of reckless killing, which would be committed if:
(1) a person by his or her conduct causes the death of another;
(2) he or she is aware of a risk that his or her conduct will cause death or serious injury; and
(3) it is unreasonable for him or her to take that risk, having regard to the circumstances as he or she knows or believes them to be.
3. We recommend the abolition of unlawful act manslaughter in its present
form.
4. We recommend the creation of a new offence of killing by gross carelessness, which would be committed if:
(1) a person by his or her conduct causes the death of another;
(2) a risk that his or her conduct will cause death or serious injury would
be obvious to a reasonable person in his or her position;
(3) he or she is capable of appreciating that risk at the material time; and
(4) either
(a) his or her conduct falls far below what can reasonably be expected of him or her in the circumstances, or
(b) he or she intends by his or her conduct to cause some injury, or is aware of, and unreasonably takes, the risk that it may do so, and the conduct causing (or intended to cause) the injury constitutes an offence.
5. We recommend that the duty to act continue to be governed by the common law for the purposes of involuntary manslaughter for the time being.
6. We recommend that both of the new homicide offences should be available as alternative verdicts to murder.
7. We recommend that the long established practice, that where there is a possibility on a count of murder of the jury returning a verdict of manslaughter, a separate count of manslaughter is not added to the indictment, be abandoned.
8. We recommend that the question whether any other offence may constitute an alternative on a charge of reckless killing or killing by gross carelessness should be governed by the general provisions of section 6(3) of the Criminal Law Act 1967.
9. We recommend that killing by gross carelessness should be an alternative to a charge of reckless killing.
10. We recommend that no change should be made to the existing offences of causing death by bad driving, and that it should also be possible, where appropriate, to prosecute such cases as reckless killing or killing by gross carelessness.
CORPORATE MANSLAUGHTER
11. We recommend:
(1) that there should be a special offence of corporate killing, broadly corresponding to the individual offence of killing by gross carelessness;
(2) that (like the individual offence) the corporate offence should be committed only where the defendant’s conduct in causing the death falls far below what could reasonably be expected;
(3) that (unlike the individual offence) the corporate offence should not require that the risk be obvious, or that the defendant be capable of appreciating the risk; and
(4) that, for the purposes of the corporate offence, a death should be regarded as having been caused by the conduct of a corporation if it is caused by a failure, in the way in which the corporation’s activities are managed or organised, to ensure the health and safety of persons employed in or affected by those activities.
12. We recommend that, for the purposes of the corporate offence, it should be possible for a management failure on the part of a corporation to be a cause of a person’s death even if the immediate cause is the act or omission of an individual.
13. We recommend that the offence of corporate killing should be capable of commission by any corporation, however and wherever incorporated, other than a corporation sole.
14. We recommend that the offence of corporate killing should not be capable of commission by an unincorporated body.
15. We recommend that the offence of corporate killing should not be capable of commission by an individual, even as a secondary party.
16. We recommend that there should be liability for the corporate offence only if the injury that results in the death is sustained in such a place that the English courts would have had jurisdiction over the offence had it been committed by an individual other than a British subject.
17. We recommend that there should be no requirement of consent to the bringing of private prosecutions for the corporate offence.
18. We recommend that the offence of corporate killing should be triable only on indictment.
19. We recommend that, where the jury finds a defendant not guilty of any of the offences we recommend, it should be possible (subject to the overall discretion of the judge) for the jury to convict the defendant of an offence under section 2 of 3 of the Health and Safety at Work etc Act 1974.
20. We recommend that
(1) a court before which a corporation is convicted of corporate killing should have power to order the corporation to take such steps, within such time, as the order specifies for remedying the failure in question and any matter which appears to the court to have resulted from the failure and been the cause or one of the causes of the death;
(2) the power to make such an order should arise only on an application by the prosecution (or the Health and Safety Executive or any other body or person designated for this purpose by the Secretary of State, either generally or in relation to the case in question) specifying the terms of the proposed order; and
(3) any such order should be on such terms (whether those proposed or others) as the court considers appropriate having regard to any representations made, and any evidence adduced, by the prosecution (or any other body or person applying for such an order) or on behalf of the corporation.
21. We recommend that the ordinary principles of corporate liability should apply to the individual offences that we propose.
…