Venn, R v

On 12 July 2002 in the Crown Court at St Albans, before His Honour Judge Findlay Baker QC and a jury, the appellant was convicted of 4 counts of indecent assault on a female. Counts 1 and 2 concerned a girl, LM. Counts 3 and 4 related to another girl, RB. On 13 September 2002 the appellant was sentenced by the same judge to 15 months’ imprisonment concurrent on each count. There were three grounds of appeal. The single judge refused leave in respect of ground 1. However, the appellant renewed his application for leave upon this ground, which we granted at the outset of the hearing.

Counts 1 and 2 were specimen counts in respect of allegations by LM that the appellant had indecently assaulted her between January 1999 and April 2001 when she was aged between 9 and 12 years old. The appellant was a close friend of the complainant’s family and he regularly visited her house where his own wife used to baby-sit. It was alleged that, on several such occasions, the appellant would rub the complainant’s breasts under her clothing and place his fingers up her vagina. The complainant subsequently revealed this to three school friends on 20 December 2001 when they were swapping secrets and the police were then informed. The appellant denied that any of the incidents took place.

Counts 3 and 4 were specific allegations made by RB in 2001, she then being 24 years of age. She first made the allegations when she was asked to be a character witness by the appellant in relation to the complaints of LM. RB and LM did not know each other and RB’s allegations went back to her childhood, some years before the alleged offences against LM when the appellant was also a family friend. Count 3 was an allegation of assault on 7 November 1992 at a fireworks party when RB was 3 months short of 15 years old. She alleged the appellant cupped his hands around her breasts over her coat for almost the entire length of the firework display. Count 4 occurred in May 1994 during the FA Cup Final, when she was just 16. The appellant entered the dining room and stood behind her. He then started to massage her breasts.

The appellant rejected entirely the allegations of LM under counts 1 and 2. He said no such thing had occurred and he did not know why she had made up the allegations. In respect of count 3, he had a clear recollection of the evening concerned. He said that he had held RB to steady her and prevent her slipping while they were standing on a slope watching the display. If his hand touched her breast it was an innocent and unconscious act and he was not aware of it. In respect of count 4, the appellant remembered being at the house during the Cup Final but he did not recall seeing the complainant or the power cut which she said had brought his fondling to an end. He denied any such conduct. Following his arrest he denied the offences in interview. Prior to the commencement of the trial the appellant’s counsel submitted that there should be separate trials for each of the two sets of allegations.

Because of the nature of the grounds of appeal it is necessary to recount the relevant evidence.

LM was aged 13 at the time of trial. In a video-recorded interview made on 28 December 2001 she stated the appellant was a close friend of the family. He had first assaulted her at her home when she was about 10 years old by placing his hand under her top and rubbing her breasts. He had later placed his hand down her trousers and placed his fingers up her vagina. Similar incidents happened on 5 or 6 more occasions in the play room or the bedroom. She did not say anything or resist because she was afraid he might do something else to her. The appellant did not hurt her but she thought what he did ‘was not very nice’. The incidents were not prolonged and occurred nearly every time he visited and found her alone. For this reason she tried to keep in the company of her brother. The incidents had ended in about April 2000. She did not tell her parents because she thought they would not believe her. She did not really understand what he was doing. In cross-examination she denied inventing the allegations in an attempt to compete with secrets told to her by her school-friend.

LM’s mother described how the appellant’s wife would babysit for her family. The appellant accompanied her because his own children were grown up and he did not need to stay at home. Occasionally he came alone. The two families were close and the appellant was treated as part of the complainant’s family. At the relevant time LM went to bed between 8.30pm and 9.00pm. On 2 or 3 occasions the appellant went upstairs on his own to say goodnight to the children. She described how LM had become withdrawn from 2001 onwards and given up pursuits such as riding, choir practice and going to school discos. By this time the appellant had moved away and only made occasional visits, but she would always seek to avoid the appellant when he visited. Immediately following LM’s video interview, she appeared dramatically to change back to her pre-2001 light-hearted demeanour and activities.

DC Hines gave evidence of LM’s distress when he visited her home before she gave the video interview. (The defence sought to exclude this evidence and its admission by the judge formed the subject of ground 3 of the appeal.)

The conduct of the appellant came to light after LM told her school friends on 20 December 2001 following a visit to her home by the appellant and his wife. Three of the complainants’ friends gave evidence of note passing following a music lesson in which they divulged secrets. S, a friend of LM, gave evidence that LM passed her a note to the effect that she wanted to tell her a secret at lunch. LM then told her “Its my uncle John, when he comes to baby-sit us. He comes into my room and starts touching me. He fingers me, puts his fingers up me and touches me up here”, indicating her breast. LM said she had been too scared to tell anyone previously. Her friends asked if she was joking because it was not very funny if she was, and LM then burst into tears. She said she had tried to push the appellant off on one occasion, but he pushed her down and she was frightened he would hurt her. Two other friends gave a similar account of that episode.

RB described how, in childhood she became a good friend of the appellant’s daughter. She had never met LM or her family. She only learned of them when she was asked to be a character witness for the appellant and this led her to make her own complaint to the police. The first incident (count 3) had occurred 9 years earlier at a firework display in St Albans on 7 November 1992 when the complainant was almost 15. While she was standing on a slope the appellant, who was standing at a higher level, put his arms around her from behind and cupped her breasts in his hands over her coat. He did this near the beginning of the display and continued till the end – about 30 minutes. She felt awkward and thought it might have been an accident. She only realised it was not after a second incident.

The second incident (Count 4) occurred on the day of an FA Cup Final in May 1994 when she was just over 16. The appellant had been invited to watch the match and was getting a beer from the kitchen. He entered the dining room where the complainant was sitting doing her homework and stood behind her, placing both hands on her front. He began massaging her breasts while she was talking about her homework. She felt uncomfortable and unable to escape because she was seated. She was unsure how long the incident lasted but estimated a minute. It ended when there was a power cut and she used this as an excuse to go to the cupboard and sort it out. When power was restored, the appellant went back to watch television with her father. She did not consent to the activity though she did nothing to signify either consent or lack of consent. Afterwards, she told her mother, but her mother did not take it seriously, putting it down to the appellant’s ‘touchy-feely’ manner. RB then realised the earlier incident was not an accident and felt she should have told her mother so. However, she ‘did not want to rock the boat’. She decided to complain to the police when the appellant approached her in respect of the complaint by LM.

The mother of RB also gave evidence. She said she had been out during the Cup Final. After the appellant and his wife left her daughter had approached her, looking uncomfortable and telling her of the incident. She considered the best course of action would be to ensure that she never left RB alone with the appellant again.

Dealing with counts 1 and 2, the appellant said he was friendly with LM until she was about 9 years old when he started going to Devon. After that he had little conversation with her. He last recalled babysitting when she was present in the middle of 1998. In cross-examination he accepted he was with the complainant on occasions after her tenth birthday, but said he was never alone with her in the playroom and only once in the bedroom. There was no need for him to go upstairs and say goodnight after she was 10 because she had always been up and about. He denied indecently assaulting her in either the playroom or the bedroom. He said he had gone to her bedroom on one occasion to look at her computer because her mother was concerned she might be concealing things on it.

As to counts 3 and 4, the appellant agreed that he was friendly with the family of RB and had gone to the firework display (count 3). He said that her account was untrue. He had noticed she could not keep her balance on the slope. He was standing higher and put his left hand under her arm to steady her. If his hand touched her breast through her coat it was not deliberate and he was not conscious of it. He said he did not deliberately hold or cup his hands around her breast. As to count 4, he recalled going to the complainant’s house to watch the football. He had no recollection of any power cut or of seeing the complainant in the house. He denied the assault had occurred or that he had forgotten about it until reminded by her complaint. What she said was untrue. He knew of no reason why either complainant should hold any grudge against him. He had decided to approach family and friends for support after the complaints brought by LM In particular, he and his wife had decided to approach RB because she could confirm that nothing had happened to her. His wife telephoned RB’s mother, only to be told of the allegation when RB was 12, which he immediately denied.

The judge made various rulings in the course of the case in the absence of the jury which give rise to the three grounds of appeal. The first was a ruling on the application of the defence to sever counts 1 and 2 from counts 3 and 4 which application was refused.

On 8 July 2002 the judge gave a preliminary ruling on severance, indicating his refusal to grant the application. He said there was no doubt the original joinder of the 4 counts was proper in view of their similarities, notwithstanding the difference in time. He said the issue was whether or not to sever the two sets of complaints and order separate trials to avoid the risk of a jury improperly using one set of allegations to buttress the other. His reasons were that (1) the Crown was entitled to rebut suggestions that the appellant had made innocent contact in both cases. He acknowledged the appellant’s submission that in respect of LM the defence was not one of innocent or accidental contact but a complete denial. However, he said that the jury might construe an assertion of innocent contact from certain of the appellant’s responses in interview, in which he had spoken of LM frequently sitting on his lap, even when she was a bit old for it and of having innocently put his arms round her when playing at her computer. (2) The judge said that, quite apart from the fact that the prosecution relied upon a ‘similar fact’ argument, there was a connection between the two complaints in that the second complaint was made as a result of the appellant approaching the complainant in respect of the first complaint. This also made it desirable to hear the two sets of allegations together.

On the next day the judge indicated that he had given further thought to the argument of the prosecution on the question of ‘similar facts’ and his mind was still open on that question. Nonetheless it did not change his overall conclusion on joinder; he simply mentioned it to make counsel aware of the slight change of position. Later that day he gave his reasoned ruling.

Having set out the nature of the complaints the judge stated, as the defence had conceded, that the counts had been properly joined in the first place. He said that while there was not such similarity in the modus operandi in relation to the two girls that it amounted to evidence of ‘similar facts’, the offences were nonetheless of a similar character as required for joinder by Rule 9 of the Indictment Rules in that the girls concerned were both the children of close family friends and in each case it was alleged that the defendant took advantage of opportunities which were offered by what may properly be described as family occasions, babysitting, social occasions or the like, rendering his alleged misconduct on each occasion an abuse of his relationship with the girls’ respective families.

He said that severance was a more difficult issue. He acknowledged the potential for prejudice arising from the fact that the jury might regard each set of allegations as more likely to be well founded because of the fact that the other allegations were made, and he quoted a passage from the statement of the mother of RB which indicated her own reaction that she had only taken her own daughter’s complaint seriously after she had heard of the allegations by LM. However, the judge observed:

“On the other hand, the jury will be able to make a more objective assessment of the facts than the witnesses, and they will receive directions both about the need for objectivity and the need for separate and independent consideration of the different charges. It is certainly my experience that juries take such directions very seriously, and it is not, in my judgment, far-fetched to say that in this kind of case their approach can often be seen reflected in the nature and pattern of their verdict.”

The judge accepted also that, if the two sets of allegations were tried separately, the manner in which RB’s allegations came to light could be withheld from the jury without significant effect either to the Crown or to the defence case.

He then proceeded to deal with the contention for the Crown that the defendant’s response to the separate allegations at interview justified introduction of ‘similar fact’ evidence by way of rebuttal. The judge indicated that, having been in favour of that contention when he first ruled on severance, he had subsequently indicated that he would require further argument on the question of ‘similar facts’ in the course of the trial before he made up his mind. Finally, he stated:

“I can at least say this: that even if this is a case in which reliance can be placed on ‘similar fact’ evidence, it is and will remain a case in which careful directions will have to be given to the jury to explain what are and what are not permissible lines of reasoning on their part. That will require further consideration when all the evidence has been presented. I shall give it that further consideration and I shall give counsel an opportunity at a later stage to address me when they have considered with me what those directions should be.”

The Crown sought to adduce evidence from DC Hines of LM’s reaction when he spoke to her at her home about her complaint. It was his evidence that, after she repeated the allegation, “she immediately broke down in tears and wept uncontrollably”. The judge stated that he agreed that, if the purpose of the evidence was that the distress exhibited by LM supported the veracity of the account she had given, it should be excluded. However he went on to rule as follows:

“The purpose is this: part of the Crown’s evidence is, expressed very generally, to the effect that the complainant, in the period leading up to the breaking of the news of her complaint, had behaved uncharacteristically, become uninterested in things that a girl of her age might expect to be interested in, and had given up a number of activities. However, immediately after she had given a video-taped interview to the police, she changed and reverted to the affectionate normal girl that her parents had previously known, and this evidence coming very shortly before the video-taping of the interview is a part of the earlier uncharacteristic behaviour.

For that purpose I can see that it is relevant and also I conclude that it is admissible, though it will be necessary for me to direct the jury carefully as to its value overall in this trial which I may add is probably rather less than might appear from the amount of time that has necessarily been devoted to determining whether it should be put before the jury.”

The judge’s summing up to the jury was careful, conscientious and fair. On the question of similarity, he first of all directed the jury about the necessity to give separate consideration to each count and in particular stated:

“I need in this case to direct you that you cannot say, if you find the defendant guilty on one count, oh well, that makes it more likely he is guilty of the next one, or the one after, or whatever. That is an impermissible line of reasoning. Except in one very particular respect you must not allow the evidence relating to one count to influence you in your decision as to the defendant’s innocence or guilt on any of the other counts in the indictment.”

A little later he referred to the exception in this way:

“It is to do with the two complainants, [LM] and [RB], making similar accusations against the defendant, and the question that you are entitled to ask is this: is it reasonably possible that [LM] and [RB] could independently and falsely, that is by lying, make similar accusations against the defendant? This is a legitimate question in relation to the allegations where the defence is that these witnesses have lied, and that is counts 1 and 2, the [LM] allegations, and count 4, the final allegation of [RB].

In count 3, which I exclude from this, the fireworks allegation, the defence is not simply that nothing happened, but that, if anything did, it happened by inadvertence.”

He continued:

“When you consider this question, is it reasonably possible that the two people independently making similar allegations could both be lying, you need also to look at the degree of similarity between the allegations which are made. I suppose it is obvious that the greater the degree of similarity the more likely it is that the independent witnesses were telling the truth, that is the witnesses who were independent from one another because you might think it a remarkable coincidence that they hit upon the same or similar lies especially if those extend to matters of detail.

Correspondingly – and you have got to look at this as well – the less the degree of similarity the less weight can properly be given to that evidence. So look at this, each girl said – and it is not disputed – that the defendant was a family friend; each girl made a complaint of being assaulted in her home; each girl said that the assaults occurred when she was a child broadly speaking, that is similarity.

Also take into account that, although both were children, there is a considerable age range between 10 and 16, and although there is a common accusation of feeling on the breast, the [LM] accusations over feeling of the breast underneath clothing and the [RB] allegations were over clothing. The [LM] allegations also involved something different, digital penetration of the vagina, and bear in mind too that these accusations from one girl and from the other were separated in point of time by a number of years. Those are the sort of matters that you will no doubt take into account when considering the question of similarity.”

On the question distress the judge said as follows:

“ … the main issue in this case is a simple one: have these girls in substance told you the truth about what happened? The defence is, in substance, no they have not, they have been lying. That makes it necessary for you to be particularly careful about evidence of distress. You should not in this case say to yourselves that evidence of distress makes it more likely that the account a distressed girl has given at the time she was distressed is true. It is not a line of reasoning which you should adopt.

You heard also about [LM]’s distress … when she spoke to police when they visited her home shortly before she went and gave the video-taped interview. You heard about that for an essentially different reason. It is part of a different area of the case which I will now deal with. Part of the Crown’s case was that [LM]’s behaviour from 2001 onwards until these matters came to light was not her natural behaviour. That by itself would prove nothing, the Crown I think accepts that, but there is evidence, and it comes from [the mother of LM] principally, that immediately after [LM] had given her video-taped interview she reverted to her normal self, and the Crown relies upon the fact that the changeover occurred at the time of the video-taped interview. What Mr Trimmer, the prosecutor, says about it is this, that that change occurring at that time tends to show that [LM]’s account at interview may be relied upon …

Mr Trimmer relies upon the change at that particular time and, indeed, its suddenness as giving some support to the Crown’s case. When you are looking at the suddenness of it you have as part of it the Crown’s evidence about her demeanour when she divulged her account to her friends or to the police. I will say in relation to that, that you will have to be sure that the distress of which a number of people spoke was true and not feigned before you can take it into account in the context which I have just outlined. In the end, members of the jury, what weight, if any, you give to this evidence is going to depend upon your evaluation of [LM’s mother’s] evidence in particular, taking account though of points made by Mr Wade about the date of onset of the unnatural behaviour, and I suppose your evaluation of the significance of the distress if you find that to be of any significance. I have spent a little while dealing with this matter, members of the jury, perhaps more than it really deserves in the overall context of this case, because the point is perhaps you may feel a relatively small one within the trial.”

The grounds of appeal are three in number. Ground 1 complains of the judge’s reasoning in refusing to sever counts 1 and 2 from counts 3 and 4 on the grounds that, in the absence of such severance, the appellant would suffer ‘inescapable and unfair’ prejudice by reason of the likelihood that, however directed, the jury would treat the complaints as mutually supportive.

Criticism is directed at the reasoning of the judge in the course of his rulings. It is pointed out that, in his preliminary ruling on 8 July, he put his decision upon two bases. First, that there were ‘similar facts’ (which it is accepted, following DPP v P [1991] 93 Crim App R 267, was a shorthand for identifiable common or related features of probative value going beyond mere coincidence) upon which the prosecution were entitled to rely in order to rebut a defence of accidental or otherwise innocent contact; second, that the complaint of RB arose as a consequence of the revelation of the first complaint (see paragraph 16 above). However, when the judge gave his full reasons, he stated that he would require further argument upon the first point in the course of the trial before finally making up his mind.

Later on, the judge indicated that he would allow the Crown to rely on ‘similar fact’ evidence. However, as is clear from his summing-up, he did not do so on the basis he had indicated earlier i.e. that it was admissible in rebuttal of the appellant’s defence of innocent contact, but that, in relation to the defence on counts 1, 2 and 4, the defence was that each of the complainants was lying.

We accept the submission of Mr Wade that the fact that the second complaint arose as a consequence of the first was not a factor which should have influenced the judge on the question of severance. It was irrelevant in relation to the complaint by LM and, in giving his reasons, the judge made clear that he accepted that the manner in which RB’s allegations came to light could be withheld from the jury on some formula to be agreed between prosecution and defence without any significant adverse effect on the Crown case or the defence case. However, we say no more on that aspect because, at the outset of the appeal, Mr Wade indicated he made his submissions on the basis that ground 1 of the appeal would stand or fall on the question of whether or not the judge was correct to regard the allegations of each complainant as having sufficient similarity properly to fall within the principle enunciated in DPP v P [1991] 2 AC 447. We therefore proceed to consider ground 2.

The question of the admissibility of ‘similar fact’ evidence depends upon the degree of its relevance. If it does no more than suggest propensity i.e. that the defendant is the sort of person who might commit the offence charged, it is inadmissible. If on the other hand it goes further than that and can be shown to be relevant to, and probative of, a particular issue in the case, then it is admissible provided its probative value outweighs its prejudicial effect. In DPP v P [1991] 2 AC 447, this area of the law was freed from the notion of a litmus, or all-purpose, test of ‘striking similarity’ as the touchstone of admissibility. In that case, Lord Mackay LC observed that it is:

“ … not appropriate to single out striking similarity as an essential element in every case …. The essential feature of evidence which is to be admitted is that its probative force in support of the allegation that an accused person committed a crime is sufficiently great to make it just to admit the evidence, not withstanding that it is prejudicial to the accused intending to show that he was guilty of another crime …

Once the principle is recognised, that what has to be assessed is the probative force of the evidence in question, the infinite variety of circumstances in which the question arises, demonstrates that there is no single manner in which this can be achieved. Whether the evidence has sufficient probative value to outweigh its prejudicial effect must in each case be a question of degree …” (460D-461A)

“Where the identity of the perpetrator is an issue, and evidence of this kind is important in that connection, obviously something in the nature of what has been called in the course of the argument a signature or other special feature will be necessary. To transpose this requirement to other situations where the question is whether a crime has been committed, rather than who did commit it, is to impose an unnecessary and improper restriction upon the application of the principle.” (462 F-G)

As a result of those passages, it was, until the decision in R v John W [1998] 2 Crim App R 289 generally thought that the ‘striking similarity’ or ‘signature’ test would continue to apply in all cases where identification was in issue. However, in John W it was made clear that, even where identity is an issue, the degree of similarity required will vary according to the other evidence in the case (see pp 300G – 303G). The court extracted from the various strands of Lord Mackay’s speech a test which it stated in the context of a case of identity as follows:

“Evidence tending to show that a defendant has committed an offence charged in count A may be used to reach a verdict on count B and vice versa, if:

the circumstances of both offences (as the jury would be entitled to find them) are such as to provide sufficient probative support for the conclusion that the defendant committed both offences, and it would therefore be fair for the evidence to be used in this way notwithstanding the prejudicial effect of so doing.”

The court added:

“To decide whether the evidence provides sufficient probative support so that it is fair to use it in this way, may require careful analysis and trial judges would be wise to seek the support of counsel before closing speeches are made, if not before.”

While the court was at pains to stress the context of identity in which it propounded the test we have quoted, the test enunciated seems to us to be one susceptible of broad application to any issue in respect of which so-called ‘similar facts’ are admissible.

In R v Musquera [1999] Crim LR 857 at 858 this court observed in general terms that, while the decision in DPP v P had eliminated the necessity to identify a striking similarity, it was still necessary to invoke some identifiable common feature or features constituting a significant connection and going beyond mere propensity or coincidence. It has to be observed that broad tests of the kind propounded in John W and Musquera do little to provide an easy guide to admissibility from case to case. That is perhaps inevitable, bearing in mind the infinite variety of factual situations which may be involved and the fact that the prosecution may legitimately seek to rely upon ‘similar facts’ in a variety of different ‘issue’ situations. The classic examples are (1) the question of identity, in which respect the Crown Court Bench Book 1 at 20.1 and 20.2 gives two specimen directions, (2) where mistake, accident or innocent association is in issue: see Archbold 2003 at 13-16 to 13-21, (3) where the defence is based on an assertion that two or more complainants are lying or mistaken; see Archbold 13-22 to 13-26(e) and direction 20.3. In all these cases, the nature of the identifiable common feature or features which may constitute a significant connection is bound to depend upon the context and on circumstances which cannot be prescribed. Where, as in this case, the prosecution witnesses are alleged to have made up their stories in a situation where collusion or cross-contamination can be discounted, the existence of common features in the nature or context of the separate offences which are the subject of complaint may, whether separately or cumulatively, be more readily regarded as non-coincidental and therefore probative on the issue of lies than would be the case if identity were the issue. That is because, in a case of this kind, the similar facts relied on are the making of similar allegations and not the events which are described in the allegations: see the Commentary of Professor Sir John Smith at [1999] Crim LR 859. As observed in R v Ryder [1994] 98 Crim App R 242 at 250

“The rationale of similar fact evidence is that two or more people do not make up or mistakenly make similar allegations against the same person independently of each other.”

The basis of the judge’s view that this was a case where ‘similar fact’ evidence was admissible is clear from his summing-up, namely in relation to the allegation that each of the complainants was lying. The features of similarity were that in each case the defendant was a close family friend, each girl complained of an assault in her own home in the course of a visit and each made a similar accusation of a squeezing or feeling of the breast, save that the interference with LM went further on occasions than with RB, in that it also involved digital penetration. The engineering of the opportunity, the initiation of the interference, the nature of the assault and the breach of trust were all similar in character. Nonetheless the judge was rightly careful to direct the jury that the weight to be given to such matters was entirely for them and to be balanced against countervailing points of distinction.

The complaint of Mr Wade in this case was advanced before us along the following lines. He recognised that, in making a judgment of the balance to be struck between relevance and prejudice, the criteria for admissibility may vary, depending upon the issue to which such evidence relates. He accepted that a ‘striking’ feature which may yet be required in various situations where identity is in issue, but may not be necessary where the issue is one of witness credibility, since the relevance of ‘similar fact’ evidence in the latter case is a means of assessing whether to accept or reject the possibility of the defendant being coincidentally the victim of two false complaints. He conceded that, in such a case, the inherent unlikelihood that (in the absence of collusion) the same person would find himself falsely accused by different individuals is enhanced, and the probative value increased, by the number (or unusual nature) of the identifiable common features relied on.

That said however, Mr Wade points out that, as a preliminary to making his ruling on severance, the judge had observed that there was not such similarity in the ‘modus operandi’ allegedly adopted in relation to each complainant as would justify ‘similar fact’ evidence being relied upon. He submits that the judge was right when he so stated and that there was no justification for his change of view as represented and set out in his summing-up.

Mr Wade submits that the common features identified by the judge were not distinctive but ‘run-of-the-mill’, and that they were outweighed by the points of distinction which could be made between the two sets of offences. In particular, the two sets of complaints involve pre- and post-pubescent girls, the massaging of the breasts was under clothing in one case and over clothing in another, and there was digital penetration in one case only.

Finally, Mr Wade complained that there was no specific warning to the jury against using the evidence as proof of propensity, and no specific direction to the jury that the purpose of the evidence was as a means of assessing the reliability of each complainant. In those circumstances, he submits the jury were likely to be led to the conclusion that the appellant was guilty because he had a propensity to commit offences of that type.

We can say at once that we reject the last two criticisms of the summing-up. There was a clear warning in respect of propensity: see paragraph 23 above. Equally, it was made clear that the purpose of considering questions of similarity went to the question whether or not the two complainants were lying: see paragraph 24 above.

Similarly, as the passage of the transcript quoted at paragraph 25 above makes clear, the judge carefully put the points of similarity and dissimilarity between the two complainants to the jury for their consideration.

The question which remains is whether the features identified to the jury were nonetheless so ‘run-of-the-mill’ as to fail to amount to a sufficient connection or relationship for the purpose of the rule as stated in DPP v P. It is our view that, in the context of this case and on the issue whether the complainants were lying in respect of their non-collusive complaints, the judge cannot be faulted for admitting the evidence. While the ‘similar facts’ rule, as explained and diluted by DPP v P and subsequent decisions, remains a rule of law as to admissibility rather than a matter of discretion (c.f. the question of fairness to the defendant arising under s. 78 of the Police and Criminal Evidence Act 1984), its now more broad-ranging and contextual nature undoubtedly offers judges a greater degree of latitude in assessing whether a combination of features, not striking in themselves, amount to a sufficient connection for the purposes of the rule. We consider that this was a borderline case which called for careful consideration by the judge in conducting the balancing exercise between probative force and prejudice. It plainly received such consideration and we are not disposed to overturn it, particularly given the careful direction which the jury received upon the issue.

We would only add under this ground that we do not attach importance to the judge’s observation at the outset of his ruling on severance that there was not such similarity in the ‘modus operandi’ in relation to the two complainants that it amounted to evidence of ‘similar facts’. It is clear to us that, at that point, he was restricting his observation to the physical molestation involved in each case and not having regard to the several common features we have identified at paragraph 36 above. In any event, as the judge made clear at the end of that ruling, he had not made up his mind on the question of ‘similar facts’ at that stage, and that he would require further argument in the course of the trial (see paragraph 21 above). While the question of the admissibility of ‘similar fact’ evidence in relation to various counts joined in the indictment is always an important one to be considered and will frequently govern the outcome of an application to sever, it is not necessarily decisive: see R v Christou [1997] AC 117. However, as we have already indicated, Mr Wade’s position has been that the outcome of ground 1 of his appeal depends on the success or failure of his submissions on ground 2. We reject his submissions on ground 2, and ground 1 therefore also fails.

Historically, the rules concerning the admissibility of evidence as to the distress of a complainant were developed in the context of the requirement for corroboration in relation to sexual offences, which requirement has since been abolished. However, the observations as to the circumstances in which evidence as to distress may carry probative weight in such cases remain valid: see for instance R v Redpath [1962] 47 Crim App R 319 and R v Chauhan [1981] 73 Crim App R 232 which make clear that such evidence should carry no weight if it is only part and parcel of the making of a complaint. However, it may properly be afforded weight if the complainant is unaware of being observed, and if the distress is exhibited at the time of, or shortly after, the offence itself, in circumstances which appear to implicate the accused. It was on those grounds that the defence opposed the admission of the police evidence as to LM’s distress when making her complaint, and it appears that it was accepted by the prosecution that, characterised simply as evidence of distress, the evidence was not admissible (see paragraph 22 above). However, the prosecution sought to justify its admission on a different ground, which the judge accepted as a basis for its admission (ibid). That ground was that it was linked with, and integral to, the evidence of the mother as to the untypically depressed demeanour of LM over a long period which was dramatically reversed following her revelation and complaint as to the conduct of the appellant. It was submitted that such evidence of demeanour was in principle admissible as supportive of her account and demonstrating that it had not been recently invented.

It appears that, before the judge, and indeed before us, Mr Wade accepted that in principle such evidence of demeanour would be admissible if it might reasonably be linked with the earlier sexual abuse, although he did not accept that it could be so connected in this case.

It seems to us, as it seemed to the judge, that if the general evidence of demeanour was admissible (see next below), then the evidence of the carthartic distress at interview was likewise admissible as pin-pointing the return to normality and supporting the attribution of the earlier depressed demeanour to the abuse which had occurred.

The question whether long-term demeanour, as opposed to short-term distress is admissible was considered by this court in R v Keast [1998] Crim LR 748. It was observed in that case that the distress of a complainant shortly after or at the time of an allegation of rape or indecent assault could be admitted to show consistency with the description of the incident given by the victim, but it could not be regarded as confirming the victim’s story from an independent source. As to long-term demeanour however, the court stated that, where the sexual abuse was alleged to have taken place over a period of time it would be difficult to attribute evidence of the victim’s demeanour to that sexual abuse, although it might be admitted for the purpose of showing that the victim’s account had not been recently invented. The court added that to allow evidence of demeanour to be given merely to show consistency or inconsistency with the complainant’s account obscured the fact that, unless there were some concrete basis for regarding the demeanour and state of mind described by the witnesses as confirming or disproving that sexual abuse had occurred, it could not assist a jury bringing their common sense to bear on who was telling the truth. The court nonetheless held that the admission of the evidence of demeanour did not render the appellant’s conviction unsafe because the judge had clearly directed the jury that the evidence concerning demeanour in no way confirmed what she had said.

In view of the uncertainties involved in establishing a link between LM’s demeanour and the earlier abuse, we consider that, as in Keast, rather than leaving it to the jury to decide whether, on the basis of the mother’s evidence, any significance could be attached to LM’s behaviour over the relevant period, the judge should have excluded it from their consideration. Having admitted it, however, it is plain that he gave the jury a careful direction, which ended with a clear indication that they should attach little importance to such evidence.

Having considered the case carefully, we do not doubt the safety of the appellant’s conviction and the appeal is therefore dismissed.