PART XVI SEVERANCE OF COUNTS OR INFORMATIONS
16.1 The present law governing the severance of counts or informations properly joined is set out at paragraphs 2.91 – 2.95 above. Counts may be joined if they "are founded on the same facts", or if they "form or are a part of a series of offences of the same or a similar character".[1] In Barrell and Wilson,[2] the Court of Appeal said that
the phrase "founded on the same facts" does not mean that for charges to be properly joined in the same indictment, the facts in relation to the respective charges must be identical in substance or virtually contemporaneous. The test is whether the charges have a common factual origin.[3]
16.2 A judge has a discretion to order that properly-joined counts be tried separately.[4]
THE RESPONSE ON CONSULTATION
The logic of this approach is very powerful but, as we have seen, it is inconsistent with general practice. We do not feel that we can make any provisional recommendations on this point, though the present practice does appear to be open to criticism on grounds of logic. We ask whether the present rules in respect of joinder of charges are adequate, or whether the courts should sever charges where prejudicial evidence is not inter-admissible between different charges, especially in sex cases.[5]
16.5 We have considered the following options:
(1) Leave the law as it is, governed by Christou.[6]
(2) Provide that where the evidence on one count is admissible on another count because it falls within clause 2, 7 or 8 of the draft Bill there is a presumption in favour of joinder and a discretion to sever (as now); but if evidence on one count is not admissible on another count, then the counts must be severed.
(3) Provide that if the evidence on one count is admissible on another count then there is a presumption in favour of joinder and a discretion to sever (as now); but if evidence on one count is not admissible on another count, and the defence so request, there is a presumption that the counts be severed unless the court is satisfied that the defendant can receive a fair trial without severance and it is otherwise desirable that the counts are heard together.
Joinder and fairness
16.6 There are perceived to be a number of advantages in joining counts or informations. Joinder is said to be in the interests of justice because: it makes for speedier justice with the result that it is more likely that memories will be accurate; successive trials enable the defendant and the witnesses to "rehearse" their evidence; and the fact-finders will be able to see the "complete picture" of the alleged offending.[7] It is also said that delay should be avoided because it might cause distress to witnesses and defendants, particularly those who are vulnerable. For example, one respondent complained that automatic severance of charges might lead to "appalling delays" in cases where children are complainants in cases of alleged sexual assault and "defeat the 'fast-track' system which all are agreed is necessary in such cases".
16.7 Against these arguments it may be said that, as we noted in the consultation paper, joint trials are not necessarily the most efficient option:[8] if charges are severed, the Crown will usually seek to have the most serious charges tried first. Depending on the outcome of this, guilty pleas may be offered for the remaining charges or the charges may be withdrawn. With regard to conservation of judicial resources Dawson argues in his article that joint trials may well place more of a strain upon such resources, because such trials are less likely to begin on time, will take longer to complete and are at greater risk of defence objections, leading to an appeal.
Option 1
16.9 The difficulty with leaving the law as it stands is put thus, in Professor Birch's commentary on Wrench:[9]
… the connection which justifies joinder is more slender than that which permits admissibility. For the purpose of joinder, it was held to be enough that the offences were generally similar in the sense that they both showed a "sexual interest in young female children". Thus, if the answer to the question concerning admissibility is yes, the jury will be told that they may draw inferences from the apparent connection. If the answer is no, then because the offences are heard together the evidence will be before the jury anyway and, though they will be told not to draw the inference, they cannot be prevented from doing so. There is something profoundly perverse about the latter outcome.
16.10 Since DPP v P evidence of bad character is permissible in a wider range of cases than it was: "striking similarity" is no longer needed. It is therefore quite likely that if the facts of some of the authorities on severance recurred now, the severance question would not be argued, because evidence on one count would be ruled sufficiently probative on another to be admissible on it.[10] The difficulty described by Professor Birch may not, therefore, be quite as acute as it has been, but it still exists. The difficulty arises in two types of case. The first is where the similarity of the facts in one count is enough to have raised the question whether that evidence on that count is admissible on other counts in the same indictment but where that application has been rejected. The second is where one count on the indictment is of a sexual nature, whether or not the other counts on the indictment are of such a nature. Certain cases, indeed most of the really difficult ones, combine both elements.
16.11 In the case of Christou,[11] C was charged with sexual offences against two younger female cousins. All three had been living in the same household at the time of the offences. There was sufficient connection between the offences for them to be properly joined (they formed "part of a series of offences of the same or a similar character" as required by rule 9 of the Indictment Rules 1971) but were not cross-admissible. Their Lordships held that the counts need not be severed. The law was stated by Lord Taylor of Gosforth in the leading speech in the following terms:
Lord Lane CJ in the quoted passage, [in Cannan][12] refrained from specifying the factors a judge should consider when "taking into account all things he should". They will vary from case to case, but the essential criterion is the achievement of a fair resolution of the issues. That requires fairness to the accused but also to the prosecution and those involved in it. Some, but by no means an exhaustive list, of the factors which may need to be considered are: – how discrete or inter-related are the facts giving rise to the counts; the impact of ordering two or more trials on the defendant and his family, on the victims and their families and on press publicity; and importantly, whether directions the judge can give to the jury will suffice to secure a fair trial if the counts are tried together. In regard to that last factor, jury trials are conducted on the basis that the judge's directions of law are to be applied faithfully. Experience shows … that juries, where counts are jointly tried, do follow the judge's directions and consider the counts separately.
Approaching the question of severance as indicated above, judges will often consider it right to order separate trials. But I reject the argument that either generally or in respect of any class of case, the judge must so order …[13]
Option 2
16.13 The logic which leads to option 2 is strong: if the evidence is either too prejudicial, and/or insufficiently probative for the court to consider that it is nevertheless fair to allow the jury or magistrates to take account of it, it is hard to see what would justify running the risk of prejudice by allowing the evidence to be heard, although not admissible. As Roskill LJ put it,[14]
It is the duty of the trial Judge to see that so far as is humanly possible an accused person charged with more than one offence is not put in danger of conviction upon any one offence by reason of evidence being given which is not admissible in relation to that one offence.
16.15 One kind of case which option 2 would affect is illustrated by the facts of Christou itself. Another example of counts properly joined but which would have to be severed under this option is Cannan.[15] Once the trial judge had decided the evidence relating to six counts regarding three women was not cross-admissible, the counts would have had to be severed into three separate trials. Consider also the facts of D.[16] D was charged with various sexual offences against two boys, ND and KC. Counts 1 – 6 concerned ND, over the period 1978 – 1984. Counts 7 and 8 concerned KC and allegedly took place in 1985. The two boys did not know each other, but it seems they and the appellant were members of the same judo club. The defence alleged collusion, but the allegation did not seem to have much weight in it. The connecting factors between the two sets of offences were the similarity of the conduct alleged, and the judo club nexus. It was expressly noted in the report that, not only was evidence relating to one boy not admissible on the counts concerning the other boy, but "the prosecution did not suggest or conduct the case on the basis that the evidence of ND was in any way supportive of KC's evidence or probative of the case of KC against the defendant" and vice versa. Yet the eight counts were heard together. The question of severance is not referred to in the report at all. In such a case, option 2 would result in two separate trials.[17] No argument for cross-admissibility was raised. Thus the assumption must be that the evidence of offences alleged regarding KC was not said to be admissible on the charges concerning ND. If option 2 were adopted then the court would have no option but to sever the two sets of allegations notwithstanding that it may be that under the rules for joinder of counts the evidence of one set of counts is an integral part of the allegations of the other set of counts, and so the two sets should, by those lights, be heard together.
Option 3
THE RECOMMENDATION
16.21 We recommend that where
(1) a defendant is charged with more than one offence,
(2) evidence of the defendant's bad character is admissible on one of the offences charged but inadmissible on another, and
(3) the defendant applies for the offences to be tried separately,
that application should be granted unless the court is satisfied that the defendant can receive a fair trial.[18]
16.22 The same rules are to apply in the magistrates' courts and in the Crown Court.
16.23 These rules are not intended to prejudice any other power to sever which a court may have.
Special rules for cases involving sexual offences?
16.24 The risk of prejudice is particularly great when the counts concern offences which are particularly likely to revolt the fact-finders, such as certain sexual offences. We noted[19] the comments made by Lord Cross and Lord Wilberforce in Boardman[20] that it was simply asking too much of a jury to perform the "mental gymnastics" necessary to enable them to put out of their minds, when considering each count, that other people were making similar allegations against the defendant.
16.25 There is some support in the authorities for a special rule favouring severance in cases of multiple sexual offences, the rationale being that special prejudice attaches to sexual allegations, but it was confirmed in Christou that no special rule exists.[21] Most respondents did not favour special rules. Professor McEwan, on the other hand, favoured change to make it more likely that sexual allegations would be heard together. In one of the cases she cites, Maloney,[22] the trial judge was looking for "striking similarity", which of course a judge has not been obliged to find since DPP v P and so such a case would not now be repeated.[23] Professor McEwan's concerns are very real ones, but their target is really the rules on admissibility. Where evidence is not relevant or sufficiently relevant, but is prejudicial, our view is that the fairness of the defendant's trial must be the paramount consideration, and our recommendation makes it so.
Note 1 Indictment Rules SI 1971 No 1253, r 9. [Back] Note 2 (1979) 69 Cr App R 250, 252-253. [Back] Note 3 But see Lockley and Sainsbury [1997] Crim LR 455. [Back] Note 4 Indictments Act 1915, s 5(3). The position is the same in the magistrates’ courts. See paras 2.93 – 2.95 above. [Back] Note 7 A discussion and criticism of a number of these perceived advantages can be found in the article by R Dawson, “Joint Trials of Defendants in Criminal Cases: An Analysis of Efficiencies and Prejudices” (1979) 77 Mich LR 13. [Back] Note 9 [1996] Crim LR 265, 267. [Back] Note 10 Eg, Flack [1969] 1 WLR 937 where D was charged with sexual offences against his three sisters. It was obviously a close-run thing as to whether the evidence of one sister was admissible on counts relating to the others, as the judge changed his mind during the course of the trial. [Back] Note 12 (1991) 92 Cr App R 16. [Back] Note 13 [1997] AC 117, 129. [Back] Note 14 Blackstock (1980) 70 Cr App R 34, 37. [Back] Note 15 (1991) 92 Cr App R 16. This concerned allegations relating to three victims. The only factors which connected the victims was evidence tending to show in each case that the defendant was the attacker, and the attacks were of a similar kind, involving abduction or attempted abduction. In one case that evidence consisted of DNA, in another it was identification evidence, and in the third it was circumstantial evidence. [Back] Note 16 Unreported. No 98/6262/W2. Convictions on three of the counts were quashed on the ground of delay and a misdirection on a specific finding as to one complainant’s age. [Back] Note 17 Tickner [1992] Crim LR 44 is a similar case: the defendant was deputy headmaster and the complainants were four pupils at his school. The counts were not severed. [Back] Note 18 This recommendation is given effect by cl 12 of the draft Bill. [Back] Note 19 Para 10.109 of the consultation paper. [Back] Note 20 [1975] AC 421, 459. [Back] Note 21 See, eg, Lord Goddard CJ in Sims [1946] KB 531, 536; Lord Cross in Boardman [1975] AC 459; De Jesus (1986) 61 ALJR 1 and B [1989] 2 Qd R 343, cited with approval in Kemp [1997] 1 Qd R 383, 396. See Roderick Munday, “Vaguely similar facts and severance of counts” (1996) 160 JP 663 – 668. [Back] Note 22 Dispatches, “Getting away with Rape” Channel 4, February 16, 1994; cited in Professor McEwan’s article “Law Commission Dodges the Nettles in Consultation Paper No. 141” [1997] Crim LR 93, 96. [Back] Note 23 Looking at the facts of Z [2000] 3 WLR 117, in which the prosecution would have succeeded at first instance in adducing evidence of other alleged rapes admitted had it not been for the Sambasivam difficulty (described at para 11.62 above), it is evident that the argument is based on coincidence reasoning: it defies belief that it is just an unlucky coincidence that this number of women claim that this man dispensed with their consent to intercourse. The prosecution surely should have succeeded in Maloney on the same basis. [Back]