Jury unanimity in homicide

Criminal Law Review

Volumn , Issue APR., 2001, Pages 283-300

  Taylor, Richard ^a  

^a UNIVERSITY OF CENTRAL LANCASHIRE   (United Kingdom)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 27844498852     PISSN: 0011135X     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (65)

1. Agbim (1979) Crim.L.R. 171.
2. "Satisfying the Jury" (1988) Crim.L.R. 335.
3. Jones, The Times February 17, 1999;
4. Carr (2000) 2 Cr.App.R. 149,
5. Boreman (2000) 1 All E.R. 307. The issue is also arising more frequently in relation to other offences see R v. D (2000) All E.R.(D) 1066 (indecent assault).
6. Referred to as "a Brown direction" based on Brown (1983) 79 Cr.App.R. 115 where Eveleigh L.J. said, "In a case such as that with which we are now dealing, the following principles apply: 1. Each ingredient of the offence must be proved to the satisfaction of each and every member of the jury (subject to the majority direction). 2. However, where a number of matters are specified in the charge as together constituting one ingredient of the offence, and any one of them is capable of doing so, then it is enough to establish the ingredient that any one of them is proved; but (because of the first principle above) any such matter must be proved to the satisfaction of the whole jury. The jury should be directed accordingly, and it should be made clear to them as well that they should all be satisfied that the statement upon which they are agreed was an inducement as alleged."
7. Above n.3.
8. i.e. that a Brown direction should have been given.
9. See the penultimate paragraph of the transcript.
10. Similar arguments would apply to diminished responsibility except of course that the burden of proof would be different and the proposition that it is manslaughter rather than murder is expressly provided for in the statute (s.2(3) of the Homicide Act 1957) rather than being a matter of common law. It may be objected that the term not-murder manslaughter is unnecessary since we already have a coterminous expression "voluntary manslaughter". However, not-murder manslaughter brings out much better the important point, which is partially obscured by "voluntary manslaughter", that these types of manslaughter are cases where the ingredients of murder are proved but the offence is "not-murder" because of a partial excuse which, inter alia, avoids the mandatory sentence for murder and confers a sentencing discretion.
11. The possibility that the jury may base a verdict of manslaughter on provocation even though they have not satisfied themselves that D had the mens rea for murder may conveniently be called the provocation error possibility.
12. Even this may be open to question - provocation requires that D has lost self-control. The loss of self-control may be so great that D was completely oblivious to the consequences of his actions and so did not even have the mens rea for an unlawful act.
13. Although this disagreement does not matter as far as the manslaughter verdict is concerned it may matter in terms of whether a verdict of not guilty of murder can be returned since the jury cannot say that they are agreed that the prosecution have not proved intent nor can they say that they are agreed that the prosecution have not disproved provocation - the trial judge would therefore be justified in ordering a retrial but may equally accept a verdict of manslaughter even without a verdict of not guilty of murder. See Saunders (1988) A.C. 148.
14. But see the provocation error possibility.
15. Although the Court of Appeal ruled in Jones that juries do not have to be asked or, if asked, do not have to answer, whether they arrived at manslaughter via route a) or route c) as identified here, this does create a difficulty for the judge in sentencing. It also covers up the question of whether there is unanimity of the jury as to the reason for not-murder (see n.11 above) and therefore of whether the judge ought to be considering a retrial of the murder charge rather than merely accepting a conviction of manslaughter. This was not a live issue in Jones since it was already a retrial but on a first trial it may be more pertinent to ask the jury whether they were unanimous on one of the grounds for it not being murder - see Smith, n.2 above, on unanimity in relation to defences.
16. See n.9 above.
17. Subject to the provocation error possibility.
18. Or between various types of voluntary manslaughter if one wishes to stick to the traditional terminology which however does not bring out quite as clearly the point being sought to be emphasised here.
19. Or "involuntary".
20. Subjectively reckless manslaughter is in theory a third possible type of basic manslaughter - see Smith & Hogan, Criminal Law (9th ed., 1999) - but never seems to be the basis of a change of manslaughter rather than murder since a) until relatively recently it was quite capable of being murder under Hyam or, prior to that, was obscured by constructive malice and b) even following Nedrick and Woollin, the prosecution would charge murder if there was a chance that the jury would infer or "find" a wish to kill etc. or foresight of virtual certainty of death etc. If murder was inappropriate or difficult to prove, the prosecution would simply fall back on alleging and proving unlawful act manslaughter (or alternatively gross negligence) which is much easier to prove than subjective recklessness as to death or grievous bodily harm.
21. At the least, and possibly of murder depending on the scale of the surgery. Suppose for example it was an amputation which would raise the question of whether an intention to amputate a diseased limb can constitute an intent to cause (grievous bodily) harm.
22. Dicta by Swinton Thomas L.J. in Cawthorne (1996) 2 Cr.App.R.(S.) 445 at 450-451 may be thought to go against this argument. "There are grave dangers in asking juries how they have reached a particular verdict. For example, they may not all have reached it by precisely the same route." However, that case was an appeal against sentence rather than against conviction and was concerned with the issue of whether the jury should be asked for the basis of their verdict for the purposes of sentencing and unanimity was not the issue being addressed as can be seen from the fact that Brown was not cited. Paradoxically, Swinton Thomas L.J. went on to refer approvingly to the practice of asking the jury for the basis of their verdict if it was a choice between provocation and diminished responsibility but not if it were a choice between lack of intent (viz. unlawful act) provocation and gross negligence. The thrust of the present argument is that, whatever might be the requirements for sentencing purposes, in terms of jury unanimity on guilt, the choice between diminished responsibility and provocation is less important (these being not-murder reasons) whereas the choice between gross negligence and unlawful act is crucial, these being separate grounds of liability as to at least one of which the jury must be unanimous. (Cf. Judge Clarke's suggestion, below at p. 303 at (para. 13), that there should be two separate counts of manslaughter where both unlawful act and gross negligence are in issue.) In contrast, as between diminished responsibility and provocation, if half of the jury think D has proved diminished responsibility and the other half think the prosecution have not disproved provocation, whilst a murder verdict can clearly not be returned, manslaughter may be since the lack of unanimity as to the not-murder reason does not prevent (presumed) unanimity on manslaughter at the very least on the basis of unlawful act.
23. See La Forest J. in Thatcher, below n.31.
24. (1981) 2 All E.R. 617.
25. (1988) Crim.L.R. 335 at 344.
26. Hence the possibility of D being not guilty on both of the occasions would have been ruled out by the whole jury.
27. Hence the possibility of D being not guilty on both of the bases has not been ruled out by the whole jury.
28. (1997) 1 Cr.App.R. 1,
29. (1996) Crim.L.R. 722.
30. "Then" is used here in the sense of "therefore". It is not simply a question of either/or, the fact of "not a)" is the premise from which one deduces it must be b).
31. Independent alternatives are normally taken to exist where the probability of each one does not affect the probability of the other. One can always argue about whether this is fully satisfied on a given set of facts but it can be assumed for the purposes of the hypothetical even though in reality e.g. lack of consent may increase the chances that the surgeon may operate hurriedly and therefore grossly negligently.
32. As will be seen, mutual exclusivity is merely one type or example of non-independence.
33. This, it will be seen, is a result of them being both exhaustive and mutually exclusive and therefore "complementary" which, as noted by Robertson & Vignaux (1993) 12 O.J.L.S. 457 at 464 involves that "if the plausibility of A increases, the plausibility of 'not A' must decrease by the same amount and vice versa". Thus complementarity is the opposite of full independence.
34. (1987) 39 D.L.R. (4th) 275 at 313, cited in Robertson & Vignaux (above) as a correct application of the rules of probability which they demonstrate to be the "logical rules for thinking about facts in legal cases".
35. In Thatcher above, Lamer C.J. quoted with approval an article by Peter Mckinnon 51 CR (3d) 134 at 137 where he wrote "If, as their deliberations progress, the jurors though not in agreement in their assessment of the probable type of involvement, are satisfied beyond a reasonable doubt that the only possible types of involvement establish guilt, they should convict."
36. Therefore they will be fully causally interdependent and "complementary".
37. The absence of a third exculpatory explanation is what makes mutually exclusive possibilities fully causally interdependent. Thus mutual exclusivity (an essentially negative concept) means if a) then not b), and if b), then not a), but it leaves open the possibility of c). However, if the (exculpatory) possibility of c) (and d, e, f . . . etc.) is ruled out as well then full causal interdependence (a more positive concept) is established which means if not a), then b) and if not b) then a).
38. Even if logically "either a or if not then b" reasoning is permissible in the exhaustive but non-mutually exclusive type of cases, it may be that the dangers involved (e.g. too ready an assumption that the inculpatory alternatives are exhaustive) may outweigh any slight legitimate increase in probative force and the either/or reasoning should not be put to the jury. See the caution expressed by La Forest J. in Thatcher at p.314 where he said, "There may however be cases where the interrelationship between competing Crown theories and evidence adduced in relation thereto will not justify a verdict of guilt. In each case therefore, it will be for the trial judge, having regard to the nature of the offence, the theories of the parties and the totality of the evidence, to realistically assess the possibility that the evidence will be used improperly and to direct the jury accordingly." (emphasis added)
39. Crucially, the 60% probability of not b) does not at the same time amount to a 60% probability of a) because a) and b) are not complementary (or exhaustive).
40. 0.4 chance of not a) multiplied by 0.6 chance of not b).
41. And thus only a 76% chance that he was guilty of manslaughter on one or other basis which includes a 24% chance that he was guilty on both bases (0.6 chance of a) multiplied by 0.4 chance of b)).
42. (1992) Crim.L.R. 880 at 882.
43. Assuming the two conflicting witnesses constitute the principal evidence against D.
44. Not implausible inferences may be that one or both of the witnesses is trying to frame D or that D has been mistaken by both witnesses for a third party. If the jury do draw either of these inferences there is a third exculpatory explanation of the facts and the "if not a), then b)" reasoning should not apply and given the difficulty of ensuring that the jury do not fall into this trap it would be safest in the inconsistent witness example not to leave the "if not a), then b)" reasoning to the jury.
45. (2000) 2 Cr.App.R. 149;
46. (2000) Crim.L.R. 193.
47. So the Thabo Meli principle could not apply.
48. (2000) 1 All E.R. 307.
49. ibid. at 316h.
50. "a) Some or all (of the jury) could have found that all three defendants were involved in causing the injuries and that these were an operating cause of death. (b) Some or all could have found that the injuries were not an operating cause of death, but they were sure that all three defendants had taken part in starting the fire, which was agreed to be an operating cause of death. (c) Some or all could have found that all three defendants were involved in causing the injuries and that these were an operating cause and that all three had deliberately set the fire (another operating cause). (d) Some or all could have found that all three had caused the injuries and these were an operating cause and that only two of them had deliberately set the fire".
51. This is unconvincing since the jury were never told that they could not convict him on this ground which was not withdrawn from the jury, so we really don't know that they did not convict him on this ground.
52. As in Jones the court was dealing with a case that had already gone to a retrial once.
53. Above.
54. (1997) 1 Cr.App.R. 14 (an alleged case of affray based on separate sequences of conduct) and see also now R v. D (2000) All E.R. (D) 1066 (variety of alleged acts of indecent assault).
55. This could be taken to mean each juror agrees "one allegation is proven" even though they differ as to which particular allegation. See More (1987) 86 Cr.App.R. 234, 239.
56. See the discussion of Jones above.
57. See the discussion of the hypothetical at p. 288 above.
58. See the discussion of Carr above.
59. Or indeed any other offence.
60. This may be the choice between principal and accessory as in Giannetto, or between two different causes of death as in the A-G's Ref case.
61. As in Giannetto.
62. And hence as a result of mutual exclusivity and exhaustiveness are complementary, i.e. fully causally interdependent.
63. As in the case of our hypothetical or as in Boreman.
64. If the time has come for a majority direction, the words "at least ten of you" can be substituted for the word "all".
65. Since this article was completed, the Court of Appeal, in Kyte, 2001 All E.R. (D) 54 (Jan), has considered another case where jury unanimity in homicide was required and quashed a conviction for murder where two alternative routes to murder were left to the jury and they had not been directed that they had to be unanimous as to one of them. A crucial factor was that the case was left to the jury on the basis that the appellant could only be guilty under the second route if she was present at the scene of the second potential cause of death. Since it was possible that she was not so present, the decision that the jury had to be unanimous on the location of the cause of death (and her presence at that location) seems clearly correct. On the other hand, if the jury had been sure that she was present, aiding and abetting, on both occasions, their being divided as to which act caused death would, it is submitted, have not then been a problem since in that case she could have been convicted on an if not a) then b) basis, the possibility of there being a cause of death without her participation having been ruled out by the whole jury (see note 24 above).