House of Lords
Publications on the Internet|
|Judgments - Regina v. Woollin
Lord Hoffmann Lord Hope of Craighead
(ON APPEAL FROM THE COURT OF APPEAL (CRIMINAL DIVISION))
I have read in draft the speeches prepared by my noble and learned friends, Lord Steyn and Lord Hope of Craighead. I too assented to this appeal being allowed, the conviction for murder quashed, the conviction for manslaughter substituted and the matter remitted to the Court of Appeal to pass sentence.
I have had the advantage of reading in draft the speeches prepared by my noble and learned friends, Lord Steyn and Lord Hope of Craighead. I agree with them, and I have nothing to add.
By an order made on 25 June 1998 your Lordships' House allowed this appeal; quashed the conviction of murder and substituted a conviction of manslaughter; and remitted the matter to the Court of Appeal to pass sentence. I now give my reasons for assenting to that course.
The case in a nutshell
The appellant lost his temper and threw his three-month-old son on to a hard surface. His son sustained a fractured skull and died. The appellant was charged with murder. The Crown did not contend that the appellant desired to kill his son or to cause him serious injury. The issue was whether the appellant nevertheless had the intention to cause serious harm. The appellant denied that he had any such intention. Subject to one qualification, the Recorder of Leeds summed up in accordance with the guidance given by Lord Lane, C.J. in Nedrick  1 W.L.R. 1025. The guidance of Lord Lane had been as follows (at 1028F):
But towards the end of his summing up the judge directed the jury that if they were satisfied that the appellant "must have realised and appreciated when he threw that child that there was a substantial risk that he would cause serious injury to it, then it would be open to you to find that he intended to cause injury to the child and you should convict him of murder." The jury found that the appellant had the necessary intention; they rejected a defence of provocation; and they convicted the appellant of murder. On appeal to the Court of Appeal (Criminal Division) the appellant's principal ground of appeal was that by directing the jury in terms of substantial risk the judge unacceptably enlarged the mental element of murder. The Court of Appeal rejected this ground of appeal and dismissed the appeal:  1 Crim.App.R. 97. Giving the judgment of the Court of Appeal Roch L.J. observed about Nedrick (107F) that:
The Court of Appeal certified the following questions as of general importance:
On appeal to your Lordships' House the terrain of the debate covered the correctness in law of the direction recommended by Lord Lane C.J. in Nedrick and, if that direction is sound, whether it should be used only in the limited category of cases envisaged by the Court of Appeal. And counsel for the appellant renewed his submission that by directing the jury in terms of substantial risk the judge illegitimately widened the mental element of murder.
The directions of the judge on the mental element
The facts of the case are fully set out in the careful judgment of Roch L.J:  1 Crim.App.R. 97. Given that the appeal is concerned with questions of law it is unnecessary at this stage to add to what I have already said about the state of the evidence when the judge came to sum up. But it is necessary to set out the judge's relevant directions of law with a brief explanation of the context and implications. The judge reminded the jury that the Crown did not allege an intention to kill. He accordingly concentrated on intention to do really serious bodily harm. He further reminded the jury that the Crown accepted that the defendant did not want to cause the child serious injuries. The judge then directed the jury as follows:
The first two questions identified by the judge appear in Lord Lane's guidance in Nedrick: at p. 1028B-D. The underlined passage is a classic direction in accordance with Nedrick: at p. 1028F.
After an overnight adjournment the judge continued his summing up. He returned to the mental element which had to be established in order to find the appellant guilty of murder. On this occasion the judge did not use the Nedrick direction. Instead the judge directed the jury as follows:
It is plain, and the Crown accepts, that a direction posing an issue as to appreciation of a "substantial risk" of causing serious injury is wider than a direction framed in terms of appreciation of a "virtual certainty (barring some unforeseen intervention)." If Lord Lane correctly stated the law in Nedrick, the judge's direction in terms of substantial risk was wrong. But the Crown argued, as I have indicated, that Nedrick was wrongly decided or, alternatively, that the principle as enunciated by Lord Lane does not apply to the present case.
The premises of the appeal
The first premise of any examination of the issues raised by this appeal is that it is at present settled law that a defendant may be convicted of murder if it is established (1) that he had an intent to kill or (2) that he had an intent to cause really serious bodily injury: Reg. v. Cunningham  A.C. 566. In regard to (2) the intent does not correspond to the harm which resulted, i.e. the causing of death. It is a species of constructive crime: see my speech in Reg. v. Powell (Anthony)  3 W.L.R. 959, 966D-967C and Lord Mustill's concurring observations, at p. 963H. This feature of the law of murder may have contributed to the problems which courts have experienced with mens rea in murder. But, unless the House of Lords or Parliament have occasion to revisit this point, the sufficiency of an intent to cause serious harm is the basic assumption upon which any analysis must proceed. Secondly, I approach the issues arising on this appeal on the basis that it does not follow that "intent" necessarily has precisely the same meaning in every context in the criminal law. The focus of the present appeal is the crime of murder.
The context of the decision in Nedrick
My Lords, since the early sixties the House has on a number of occasions considered the mens rea required to establish murder. It would be right to acknowledge that none of these decisions satisfactorily settled the law. In Director of Public Prosecutions v. Smith  A.C. 290 the defendant tried to avoid arrest and killed a policeman by driving off with the policeman clinging to the car. The House ruled (1) that the defendant committed murder because death or grievous bodily harm was foreseen by him as a "likely" result of his act and (2) that he was deemed to have foreseen the risk a reasonable person in his position would have foreseen. There was widespread and severe criticism of the second part of the decision in Smith. In retrospect it is now clear the criminal law was set on a wrong course. By section 8 of the Criminal Justice Act 1967 Parliament reversed the effect of Smith. Since then one thing at least has been clear: the mental element of murder is concerned with the subjective question of what was in the mind of the man accused of murder. In Reg. v. Hyam  A.C. 55 the House of Lords had an opportunity to consider what state of mind, apart from the case where a defendant acts with the purpose of killing or causing serious injury, may be sufficient to constitute the necessary intention. The defendant had burnt down the house of her rival in love, thereby killing her children. The judge directed the jury to convict the defendant of murder if she knew that it was highly probable that her act would cause death or serious bodily harm. The jury convicted her of murder. The House upheld the conviction by a majority of three to two. But the Law Lords constituting the majority gave different reasons: one adopted the "highly probable" test; another thought a test of probability was sufficient; and a third thought it was sufficient if the defendant realised there was "a serious risk." The law of murder was in a state of disarray. The decision in Hyam was not only criticised by academic writers but was badly received in the profession. The next opportunity for the House of Lords to examine the mental element of murder came in Reg. v. Moloney  A.C. 905. The clear effect of Moloney was to narrow down the broad approach to mens rea adopted in Hyam. In the leading judgment Lord Bridge of Harwich observed in Moloney with the approval of all the Law Lords (at 925H):
Lord Bridge paraphrased this idea in terms of "moral certainty" (at 926F). In the result the House adopted a narrower test of what may constitute intention which is similar to the "virtual certainty" test in Nedrick: see also the answer to the certified question in Moloney, pp. 908D and 929H. It is true that Lord Bridge said that in the "rare cases" in which it might be necessary to direct a jury by reference to foresight of consequences it would be sufficient to place the following two questions before the jury (at 929G):
It seems clear that Lord Bridge used "natural consequence" as implicitly conveying the concept of a high probability. But the guidance did not make that clear. The suggested direction soon caused practical difficulties. The problems caused by the guidance arose a year later in acute form in Reg. v. Hancock and Shankland  A.C. 455. Two miners on strike had pushed a concrete block from a bridge onto a three-lane highway on which a miner was being taken to work by taxi. The concrete block hit the taxi and killed the driver. The defendants were charged with murder. The defendants said that they merely intended to block the road and to frighten the non-striking miner. Following the guidance in Moloney the judge directed the jury to ask themselves: "Was death or serious injury a natural consequence of what was done? Did a defendant foresee that consequence as a natural consequence?" The jury convicted the defendants of murder. The Court of Appeal held that the Moloney guidelines, and the judge's direction in terms of those guidelines, were defective and potentially misleading. The conviction of murder was quashed. There was an appeal to the House of Lords. In the only speech Lord Scarman accepted that the Moloney guidelines were misleading since they omitted any reference to probability. Lord Scarman observed (at 473F):
In the Court of Appeal in Hancock Lord Lane had formulated guidelines for the assistance of the juries: 473G. Lord Scarman was not persuaded that guidelines were desirable. The House did not, however, rule out guidelines but emphasized that they should be sparingly used.
The problem facing the Court of Appeal in Nedrick
In Hancock Lord Scarman did not express disagreement with the test of foresight of a probability which is "little short of overwhelming" as enunciated in Moloney. Lord Scarman also did not express disagreement with the law underlying Lord Lane's model direction in Hancock which was based on a defendant having "appreciated that what he did was highly likely to cause death or really serious bodily injury." Lord Scarman merely said that model directions were generally undesirable. Moreover, Lord Scarman thought that where explanation is required the jury should be directed as to the relevance of probability without expressly stating the matter in terms of any particular level of probability. The manner in which trial judges were to direct juries was left unclear. Moreover, in practice juries sometimes ask probing questions which cannot easily be ignored by trial judges. For example, imagine that in a case such as Hancock the jury sent a note to the judge to the following effect:
One may alter the question by substituting "highly probable" for "probable". Or one may imagine the jury asking whether a foresight of a "substantial risk" that the defendant's act would cause serious injury was enough. What is the judge to say to the jury? Hancock does not rule out an answer by the judge but it certainly does not explain how such questions are to be answered. It is well known that judges were sometimes advised to deflect such questions by the statement that "intention" is an ordinary word in the English language. That is surely an unhelpful response to what may be a sensible question. In these circumstances it is not altogether surprising that in Nedrick the Court of Appeal felt compelled to provide a model direction for the assistance of trial judges.
In Nedrick the appellant poured paraffin through the front door of a house and set it alight. In the fire a child died. The facts were remarkably similar to those in Hyam. The trial judge in Nedrick framed his direction in terms of foresight of a high probability that the act would result in serious bodily injury. Lord Lane observed (at 1028C-F):
While I have thought it right to give the full text of Lord Lane's observations, it is obvious that the underlined passage contains the critical direction. The effect of the critical direction is that a result foreseen as virtually certain is an intended result.
The direct attack on Nedrick
It is now possible to consider the Crown's direct challenge to the correctness of Nedrick. First, the Crown argued that Nedrick prevents the jury from considering all the evidence in the case relevant to intention. The argument is that this is contrary to the provisions of section 8 of the Criminal Justice Act 1967. This provision reads as follows:
Paragraph (a) is an instruction to the judge and is not relevant to the issues on this appeal. The Crown's argument relied on paragraph (b) which is concerned with the function of the jury. It is no more than a legislative instruction that in considering their findings on intention or foresight the jury must take into account all relevant evidence: see Edward Griew, States of Mind, Presumptions and Inferences, essay in Criminal Law: Essays in Honour of J. C. Smith, ed. Peter Smith, (1987), 68 at 76-77. Nedrick is undoubtedly concerned with the mental element which is sufficient for murder. So, for that matter, in their different ways were Smith, Hyam, Moloney and Hancock. But, as Lord Lane emphasised in the last sentence of Nedrick, "The decision is one for the jury to be reached upon a consideration of all the evidence." Nedrick does not prevent a jury from considering all the evidence: it merely stated what state of mind (in the absence of a purpose to kill or to cause serious harm) is sufficient for murder. I would therefore reject the Crown's first argument.
In the second place the Crown submitted that Nedrick is in conflict with the decision of the House in Hancock. Counsel argued that in order "to bring some coherence to the process of determining intention Lord Lane specified a minimum level of foresight, namely virtual certainty". But that is not in conflict with the decision in Hancock which, apart from disapproving Lord Bridge's "natural consequence" model direction, approved Moloney in all other respects. And in Moloney Lord Bridge said that if a person foresees the probability of a consequence as little short of overwhelming, this "will suffice to establish the necessary intent." Nor did the House in Hancock rule out the framing of model directions by the Court of Appeal for the assistance of trial judges. I would therefore reject the argument that the guidance given in Nedrick was in conflict with the decision of the House in Hancock.
The Crown did not argue that as a matter of policy foresight of a virtual certainty is too narrow a test in murder. Subject to minor qualifications, the decision in Nedrick, was widely welcomed by distinguished academic writers: see J.C. Smith (1986) Crim.L.R. 742-744; Glanville Williams, The Mens Rea for Murder: Leave It Alone, 105 (1989) L.Q.R. 387; J. R. Spencer,  C.L.J. 366-367; Andrew Ashworth, Principles of Criminal Law, 2nd ed. (1995), p. 172. It is also of interest that it is very similar to the threshold of being aware "that it will occur in the ordinary course of events" in the Law Commission's draft Criminal Code: compare also J.C. Smith, A Note on Intention (1990) Cr. L.R. 85, at 86. Moreover, over a period of twelve years since Nedrick the test of foresight of virtual certainty has apparently caused no practical difficulties. It is simple and clear. It is true that it may exclude a conviction of murder in the often cited terrorist example where a member of the bomb disposal team is killed. In such a case it may realistically be said that the terrorist did not foresee the killing of a member of the bomb disposal team as a virtual certainty. That may be a consequence of not framing the principle in terms of risk taking. Such cases ought to cause no substantial difficulty since immediately below murder there is available a verdict of manslaughter which may attract in the discretion of the court a life sentence. In any event, as Lord Lane eloquently argued in a debate in the House of Lords, to frame a principle for particular difficulties regarding terrorism "would produce corresponding injustices which would be very hard to eradicate": Hansard (H.L. Debates), 6 November 1989, col. 480. I am satisfied that the Nedrick test, which was squarely based on the decision of the House in Moloney, is pitched at the right level of foresight.
The argument that Nedrick has limited application.
The Court of Appeal held that the phrase "a virtual certainty" should be confined to cases where the evidence of intent is limited to admitted actions of the accused and the consequences of those actions. It is not obligatory where there is other evidence to consider. The Crown's alternative submission on the appeal was to the same effect. This distinction would introduce yet another complication into a branch of the criminal law where simplicity is of supreme importance. The distinction is dependent on the vagaries of the evidence in particular cases. Moreover, a jury may reject the other evidence to which the Court of Appeal refers. And in preparing his summing up a judge could not ignore this possibility. If the Court of Appeal's view is right, it might compel a judge to pose different tests depending on what evidence the jury accepts. For my part, and with the greatest respect, I have to say that this distinction would be likely to produce great practical difficulties. But, most importantly, the distinction is not based on any principled view regarding the mental element in murder. Contrary to the view of the Court of Appeal, I would also hold that section 8(b) of the Act of 1967 does not compel such a result.
In my view the ruling of the Court of Appeal was wrong. It may be appropriate to give a direction in accordance with Nedrick in any case in which the defendant may not have desired the result of his act. But I accept the trial judge is best placed to decide what direction is required by the circumstances of the case.
The disposal of the present appeal
It follows that judge should not have departed from the Nedrick direction. By using the phrase "substantial risk" the judge blurred the line between intention and recklessness, and hence between murder and manslaughter. The misdirection enlarged the scope of the mental element required for murder. It was a material misdirection. At one stage it was argued that the earlier correct direction "cured" the subsequent incorrect direction. A misdirection cannot by any means always be cured by the fact that the judge at an earlier or later stage gave a correct direction. After all, how is a jury to choose between a correct and an incorrect direction on a point of law? If a misdirection is to be corrected, it must be done in the plainest terms: Archbold, Criminal Pleading, Evidence & Practice, 1998, para. 4-374.
That is, however, not the end of the matter. For my part, I have given anxious consideration to the observation of the Court of Appeal that, if the judge had used the phrase "a virtual certainty," the verdict would have been the same. In this case there was no suggestion of any other ill-treatment of the child. It would also be putting matters too high to say that on the evidence before the jury it was an open-and-shut case of murder rather than manslaughter. In my view the conviction of murder is unsafe. The conviction of murder must be quashed.
The status of Nedrick
In my view Lord Lane's judgment in Nedrick provided valuable assistance to trial judges. The model direction is by now a tried-and-tested formula. Trial judges ought to continue to use it. On matters of detail I have three observations, which can best be understood if I set out again the relevant part of Lord Lane's judgment. It was as follows:
First, I am persuaded by the speech of my noble and learned friend, Lord Hope of Craighead, that it is unlikely, if ever, to be helpful to direct the jury in terms of the two questions set out in (A). I agree that these questions may detract from the clarity of the critical direction in (B). Secondly, in their writings previously cited Glanville Williams, J.C. Smith and Andrew Ashworth observed that the use of the words "to infer" in (B) may detract from the clarity of the model direction. I agree. I would substitute the words "to find." Thirdly, the first sentence of (C) does not form part of the model direction. But it would always be right for the judge to say, as Lord Lane put it, that the decision is for the jury upon a consideration of all the evidence in the case.
The certified questions
Given my conclusions the certified questions fall away.
I have had the advantage of reading in draft the speeches prepared by my noble and learned friends, Lord Steyn and Lord Hope of Craighead. For the reasons which they give, I agree that this appeal should be allowed, the conviction for murder quashed, the conviction for manslaughter substituted and the matter remitted to the Court of Appeal to pass sentence.
LORD HOPE OF CRAIGHEAD
I have had the advantage of reading in draft the speech which has been prepared by my noble and learned friend, Lord Steyn. I agree with it, and I wish to add only these brief comments.
I attach great importance to the search for a direction which is both clear and simple. It should be expressed in as few words as possible. That is essential if it is to be intelligible. A jury cannot be expected to absorb and apply a direction which attempts to deal with every situation which might conceivably arise. I think that the Nedrick direction, which is (B) in Lord Steyn's analysis, fulfils this requirement admirably. But the substitution of the word "find" for "infer" is an improvement, in the interests of clarity, and I also would make this change to it. However I regard the questions in (A), which are derived from Lord Scarman's speech in Reg. v. Hancock and Shankland  A.C. 455, 473F, as detracting from the clarity of the critical direction. I would prefer to say therefore that it is unlikely, if ever, to be helpful to tell the jury that they should ask themselves these questions. I think that it would be better to give them the critical direction, and then to tell them that the decision was theirs upon a consideration of all the evidence.
As for the terrorist example, I think that Lord Mustill's observations in Attorney-General's Reference (No. 3 of 1994)  A.C. 245, 261D-F are also relevant. In that passage he gave as an example of "indiscriminate malice," which belongs to the category of deliberate murder where the defendant consciously intended to kill the victim, the example of the terrorist who hides a bomb in an aircraft. As he explained, the intention is aimed at the class of potential victims of which the actual victim forms part, even although the identity of the ultimate victim is not yet fixed at the start when the intent is combined with the actus reus which ultimately causes the explosion. The answer to the question whether those who attempt to dispose of the bomb are within that class will depend on the circumstances. All that needs to be said is that it may not be necessary in every such case to rely on the alternative verdict of manslaughter.
Lords Parliament Commons Search Contact Us Index
|© Parliamentary copyright 1998||Prepared 22 July 1998|