This is the appeal against conviction, with the limited leave of the full court, of H, who on 18 May 2005 in the crown court at Preston before HHJ Russell QC and a jury was convicted on 22 out of 23 counts of sexual abuse against three stepsons and was sentenced to a total of 15 years imprisonment.

Counts 1 to 5 concerned the appellant’s stepson to whom I will refer as D, the son of a woman with whom the appellant had had a brief relationship. There were two counts of indecent assault on a male person and three counts of indecency with a child. The five counts spanned the period from July 1987 to July 1994, when D was between the ages of 3 and 10: but from the evidence given at trial the more likely dating of the offences which the jury found had occurred as represented by these specimen counts was during the three or four years from 1989 to 1992 or 1993, when D was 5 to 8 or 9 years old (the earlier counts had covered a three year period). D had been 3 when the relationship between the appellant and his mother started, but about 5 when the family moved to an address at which the abuse was stated to have occurred, on occasions when his mother was out at bingo. On one such occasion D said that the appellant had put on a photographic video, masturbated him and then made D masturbate him. That kind of incident was said to have happened regularly each month. Once, when they were out in a car together, the appellant made D masturbate him until just before he ejaculated into tissues. The jury convicted on counts 1, 2 and 5 by a majority of 10 to 2, on count 3 by a majority of 11 to 1, and on count 4 unanimously.

Counts 6 to 17 concerned the second complainant, to whom I will refer as C, who was the elder of two full brothers, the sons of the appellant’s subsequent partner (to whom I will refer as P). They were half-brothers to a sister, the daughter of the appellant and P. Count 12, a count of indecency with a child was the only count on which the jury were unable to return a verdict. The other counts were 3 counts of indecent assault on a male person, 5 counts of indecency with a child, and 3 counts of attempted rape. C gave evidence of abuse by the appellant for about six years, 1998 to 2004, when he was between the ages of 9 and 15. When he was about 9 he was in the attic when the appellant pulled down his lower clothing, fondled his genitalia and carried out an act of oral sex on him. Another specific incident took place in the bathroom, when the appellant stood him on the toilet seat and carried out an act of oral sex. He would also regularly masturbate C and get C to masturbate him. From the age of 13, he made C carry out oral sex on him. He often did these things when sharing a bed with C. On one occasion he mentioned that he missed doing things to D. On a number of occasions the appellant had attempted to penetrate C anally. The last occasion of abuse occurred about only one month prior to C being interviewed by the police in May 2004. He said that on one occasion (“just once”) he was present when the appellant had sexually abused his brother, but he had also elsewhere in his evidence related at least two incidents of abuse to his brother at which he had been present. He told his mother about these matters in 2003, but she did not believe him. The jury convicted on counts 6 and 10 by a majority of 11 to 1, on counts 8 and 9 on a majority of 10 to 2, and on counts 7, 11, 13, 14, and counts 15-17 (the attempted rape counts) unanimously.

C’s brother M was the third complainant. He was almost exactly one year younger and in his case the abuse took place for about 4 years between about 1999 and 2003, when he was 9 to 13 years old. The first occasion was in bed one night when the appellant woke him, pulled his shorts off, played with his genitalia and masturbated him. The last occasion was late in 2003 when he made M masturbate him in the living room. His case concerned counts 18 to 23, which comprised four counts of indecent assault on a male person and 2 counts of indecency with a child. He gave evidence that the appellant regularly played with his penis, masturbated him when they shared a bath together, and would masturbate himself in M’s presence. M said that the appellant had abused both him and his brother together “sometimes” and also said that was “six or seven times”. He also said that he had complained to their mother in 2003, but she did not believe him. He said that he had not discussed the case with D, and had only learned of what had happened to D on the day of his interview. The jury convicted on all five counts which concerned M unanimously.

That was the evidence of abuse which came from the three boys.

The matters had come to light with a complaint to the police by D, as he explained in his evidence. That had occurred in May 2004, when he was 19. He was 20 at the time of trial. In May 2004 he had been staying with the appellant for some months together with his pregnant girlfriend. He accepted that for some 10 years there had been no abuse, and that he had often stayed with the appellant in his home or in his caravan even after the relationship with his mother had ended. He had told his girlfriend about the abuse, because he was concerned for his child to be. She had encouraged him to bring it into the open. They moved out on the day he made his complaint to the police (his witness statement was dated 23 May 2004). On the next day, 24 May, the two younger complainants had given their evidence on video. D’s girlfriend was not a witness.

The two brothers had previously spoken of their treatment to their mother, in the summer of 2003. She gave evidence about this complaint. She said that the children were mixed up and angry, and had returned (from visiting the appellant) and had complained to her of the sexual abuse at his hands. They said it had been going on for some time. She said: “I was shocked at first I didn’t know what to think. I spoke to a friend. I was scared to deal with it. I did believe them but didn’t want to believe them, and I did not complain to the police.” However, she did raise it with the appellant a few days later. He denied it and brushed it off, and asked if the children were still coming for the weekend. She continued to allow the children (and other siblings) to visit and stay with the appellant. The judge said this of her evidence: “You may think some of her evidence was designed to justify herself and was not really very much assistance in the case, but it is important that you should be reminded of some of what she said.”

The appellant gave evidence denying the allegations, as he had done when arrested and interviewed. He said the three complainants were lying. The only reason he could think of for D to lie was that they had argued about a motorbike about a week or more before his arrest (on 27 May 2004). D had been staying with him at the time with D’s girlfriend. After the argument D and his girlfriend had left and not returned. He said that even after his relationship with D’s mother had broken down, D had continued to visit him and to stay every weekend, and that when D left school at 17 he had lived permanently with him, right up to the time of D’s complaint to the police. As for the brothers, he said he had no idea why they had made up the allegations other than spiteful jealousy of D and the brothers’ sister. He agreed that P had confronted him about a year prior to his arrest: he reacted with shock and anger. He offered to take the children to the police station to have tests, but she refused. Thereafter the children had continued to visit and stay overnight with him, as before, until the time of his arrest.

There are four grounds of appeal for which the full court has given leave, and granted a considerable extension of time (some five years). On behalf of the appellant Mr Barlow has varied the order in which he has presented the four grounds. In his written material he gave greatest emphasis to his first ground, but in his oral submissions he left that for last. In those oral submissions he presented his grounds in the following order. First, the judge had misdirected the jury in failing to tell them that the brothers’ complaint to their mother was not independent evidence. Secondly, the judge had misdirected the jury in not expressly telling them that the evidence of the three complainants was not cross-admissible across the various counts. Thirdly, the judge had failed to direct the jury appropriately about the dangers of conscious or subconscious contamination of the evidence of D. And fourthly, the judge had failed to tell the jury of the dangers, at any rate in the case of D, who was speaking of historic abuse going back to his very early childhood, of delay and false childhood memories. In his written material, Mr Barlow had presented these four grounds, which we will here number in the order in which they were addressed in our court, as above, by promoting his ultimately fourth ground to first place.

Mr Barlow submitted that the judge failed to assist the jury with the status of the evidence of the brothers’ complaint to their mother. They should have been warned that that evidence was not independent of the complainants themselves. That had been the standard direction under the common law, prior to the Criminal Law Act 2003, see R v. Islam [1999] 1 Cr App R 22; and it remained necessary in terms of section 120 of the new Act, see R v. AA [2007] EWCA Crim 1779. In Islam it was said that juries should be directed as to the limited significance of recent complaint since it did not come from an independent source and could not constitute independent confirmation of a complainant’s testimony. In AA, in the era of section 120 of the new Act, which allowed evidence of previous consistent statements to be admissible as evidence of their truth, Lord Justice Laws said this:

Laws LJ went on to explain that in the circumstances of that case the judge’s failure to direct the jury accordingly made the conviction unsafe (at para 18). In that case the complaint in question was of a single offence of rape, made promptly the next day to the complainant’s friend, after the complainant and the defendant had spent the previous evening together, and the defence was consent. But there had been no report to the police for another seven weeks. The judge had agreed in discussions with counsel to direct the jury that the statement of recent complaint did not come from an independent source; but he failed in the event to do so. On the other hand he did direct the jury about the potential of the evidence of recent complaint to assist the jury in deciding where the truth lay: “It is for you to decide whether it is consistent, whether it is accurate and it makes [AB] more credible”. Laws LJ observed:

In the present case, the judge gave no direction on this subject at all. He merely introduced the mother’s evidence in the way we have outlined above, namely that the jury might think she was a somewhat unsatisfactory witness but it was important that they be reminded of some of what she said. And he then reminded them inter alia of her sons’ complaint to her in 2003, which took 7 lines out of several pages concerned with her evidence. No details were given of the complaint, only that “They said he’d done things to them sexually and they’d had enough, they told me of some of the things he had done”. The ramifications of her evidence which were explored somewhat further, however, were her reactions, and the fact that she continued to allow the children to visit and stay with the appellant. This evidence was therefore two-edged so far as Crown and appellant were concerned: the mother did not really believe the children, at any rate on the appellant’s understanding of her evidence (it was somewhat equivocal) and on the brothers’ understanding of her reaction to them, as was shown by the fact that she did not go to the police but continued to allow the children to stay with the appellant. It would seem, moreover, that counsel at trial (not counsel on the appeal) did not require an Islam type direction. It may be said moreover that it must have been obvious that the evidence of complaint was not independent of the children. It was their complaint. Nor did it have the immediacy and impact of truly “recent” complaint; and it was not a case where the conduct was acknowledged and the sole issue was consent. In the circumstances of AA, however, against the background of a materially delayed complaint to the police, an all but immediate complaint to a friend following a single act of what was either non-consensual rape or consensual intercourse was of special significance. The allegations here, however, were of an ongoing course of conduct, both before and after the complaint, and the appellant relied on the evidence of the complaint for his own purposes as being part of the (consistent) fabrication of which he complained. For all these reasons the situation in the present trial was very different from that in AA.

AA is not itself reported or noted in Archbold. It can be contrasted with R v. Amrani [2011] EWCA Crim 1517, where AA was cited and Islam was again relied upon. Again the case concerned a single act of rape, this time within an acknowledged relationship, and again the defence was consent. Giving the judgment of this court, Pill LJ observed that “Much of what was said in Islam has been superseded by the enactment of section 120” (at para 13). He referred to the amended JSB directions as of May 2009 (at para 17) and said: “While the commentary dealing with section 120(4) and (7) does not expressly require the independence direction, the illustrations given recommend its inclusion”. He then quoted from the summing up to demonstrate that the judge had made clear that the primary task of the jury was to decide whether the complainant or the appellant was telling the truth (at para 18). He also observed (at para 19) that “It must have been obvious that the complainant’s own evidence of complaints, whether in the witness box or to another, was not evidence independent of her.” Pill LJ concluded as follows:

In our judgment, much of the same may be said in this case, and more. It may be inferred that the judge was not asked by defence (or Crown) counsel for an independence direction, and much must be allowed for the feel of the case which the trial judge and counsel would have had. In circumstances where it was the defence rather than, or as much as, the prosecution which was relying on the mother’s evidence, there was a danger in overcomplicating matters. The defence case, of course, was not that what occurred had been consensual, but that nothing had occurred and that the boys were fabricating wickedly. As for the real issue in the case, the judge’s directions were fair, helpful and to the point. It was clear from the trial as a whole and his directions that the jury had to decide where the truth was to be found, in the complainants’ allegations or in the defence, but that the burden lay on the prosecution. He gave clear direction as to difficulties that historic abuse might present for a defendant and warned the jury of a danger of real prejudice to a defendant. He asked the jury to consider “why these matters did not come to light sooner”, and whether that was a reflection on the reliability of the complainants or arose from the conduct of the defendant. In saying that, he was emphasising that the reliability of the complainants was the critical question. Above all, he stressed this:

In these circumstances, we would be inclined to the conclusion that there was no misdirection at all. If, however, by the standards of AA and Amrani there was some lack of a direction about the mother’s evidence of complaints not coming from an independent source, we would nevertheless conclude, as this court did in Amrani, although we would consider this a stronger case, that the verdicts of the jury were not rendered unsafe by virtue of that misdirection alone.

Mr Barlow submits that the judge failed to give an adequate direction concerning the non cross-admissibility of the evidence of each of the three complainants either for supporting each other’s testimony or even for supporting his own evidence on other counts in which he was the complainant. For this purpose Mr Barlow relied on R v. Dye [2003] EWCA Crim 2424, [2004] 1 Cr App R 19, a case prior to the introduction of the bad character reforms of the Criminal Justice Act 2003. That concerned a case of rape and other sexual abuse against two step-daughters. The issue in the appeal was whether the judge there had failed to direct the jury in sufficiently clear terms to consider the allegations of each complainant entirely separately.

Nelson J, giving the judgment of this court, concluded thus:

This court considered that an adequate direction “sufficiently clear to bring home to the jury the absolute need for them to treat the evidence in relation to each set of allegations separately” had not been given (at para 27) and that therefore the convictions were unsafe.

Mr Barlow submitted that the same should apply here. He pointed out that although the judge gave a more or less standard direction about treating each count separately, he did not give the additional direction required by Nelson J.

What the judge did was as follows. He gave a strong direction to treat each count separately. He said:

Mr Barlow complains that this was an invitation to the jury to cross-apply the evidence of each of the brothers to counts involving the other brother: and that this misdirection was compounded by the fact that the judge gave the jury no assistance in identifying which of the counts before them might have been the count concerned with the possibly sole incident in which both brothers had been involved at the same time. If the judge was not able to identify the incident for the jury, then the judge should have directed the jury to ignore all such evidence.

We disagree. We consider that the effect of the judge’s direction about the brothers was accurate (he accurately used the expression “In some instances”, but he later reminded the jury that C had also said it was “just once”), was directed to that limited evidence of the brothers where they had spoken of witnessing the other being abused, and, together with the strong passage about considering each count and the evidence on each separately, amounted to an additional indication to the jury that, with that limited exception, the evidence from each complainant related only to the matters of which each complainant spoke. Moreover, consistently with that approach, the judge kept his summary of the evidence of each complainant entirely separate. But he also pointed out where there was any inconsistency between the brothers in their accounts of events at which either said both had been present.

As it happens, this trial took place in May 2005, very shortly after the coming into effect of the bad character reforms of the 2003 Act. We are told by both counsel (neither of whom we repeat were involved in the trial) that it is believed that there was some application to admit bad character under section 101 of that Act, which failed. However, the detail of that is a mystery. There is no ruling available. Mr Barlow’s advice on appeal dated 11 September 2010 refers to having seen a ruling on bad character, but no such ruling appears to exist. We have no first-hand information. There are no documents of any kind regarding such an application.

Under the 2003 Act, the common law regarding bad character has been replaced by the statutory provisions. R v. Chopra [2006] EWCA Crim 2133, [2007] 1 Cr App R 16 said that cross-admissibility depended on coming within one of the gateways of section 101, although not necessarily as a matter of propensity: see also R v Freeman; R v Crawford [2008] EWCA Crim 1863, [2009] 1 Cr App R 11. The new statute has enabled a common sense approach to be taken to the concept that each similar complaint makes each other similar complaint the more likely. The reality is that independent people do not make false allegations of a like nature against the same person, in the absence of collusion or contamination of their evidence.

In these circumstances, it is instructive to see how R v. D has fared since its date, and/or since the arrival of the 2003 Act when there has been no application under section 101 to make the evidence on one count cross-admissible on another.

In R v. Lloyd [2005] EWCA Crim 1111 there was an appeal against conviction on ten counts of indecent assault. There were four complainants at trial. The judge directed the jury to consider the evidence in relation to each count separately, but he also appears on at least three occasions to have invited the jury to consider the degrees of similarity between some features of the evidence relating to individual complainants (a feature absent from the present case). The court quashed the convictions. Treacy J, giving the judgment of this court, having cited para 26 of the judgment of Nelson J in Dye, said this:

In R v. F [2005] EWCA Crim 3217 there were two complainants. The appellant had been convicted on three counts of rape and one of indecent assault. The judgment of this court was given by Scott Baker LJ. He also cited para 26 of Nelson J’s judgment in Dye, but was unwilling to follow it in his case. He considered that the directions concerning separate consideration, and the separate treatment given to the evidence of each complainant, sufficed. He referred to R v. Musquera [1999] Crim L R 857 (which had also been relied on in Dye and Lloyd), where Potter LJ had said this:

R v. Wall [2005] EWCA Crim 3251 concerned four complainants and ten counts of indecent assault. The submission made to our court based on Dye (and Lloyd) was again made there. The appeal was dismissed. The Recorder of Cardiff gave the judgment of the court. Having cited from the judgment of Nelson J in Dye at paras 25/26, he continued (at para 14):

The learned Recorder next considered a submission that in the absence of a Dye type direction, the jury may have wrongly regarded the evidence as cross admissible in respect of each set of allegations and so rely upon what amounts to no more than evidence of propensity as evidence of guilt. However, he concluded that that was unlikely. He said (at para 16):

Finally, in R v. Mackay [2010] EWCA Crim 167, in the era of the 2003 Act, the Dye argument was again advanced, but failed. This was a rather different kind of case, where the defendant had been convicted of separate woundings on different victims over a month apart. The defence case to one wounding was that it was not him; and on the other wounding, that he was assaulted by the victim but did not retaliate. An application of severance was refused. There was, it seems, no section 101 ruling. The direction on separate consideration went further than in the other cases considered (and in our case) in that the judge had expressly said that the jury should not use the evidence on one count in deciding guilt on the other count. However, he had continued by saying that the jury might consider the credibility of a witness as impacting on the whole of his evidence. As to that, Pill LJ remarked (at para 21):

We consider that in the light of this jurisprudence as a whole the suggested direction in Dye has not fared well as a ground of appeal. It has never been part of a standard JSB direction. No complaint could be made about the giving of such a direction, but we believe that Scott Baker LJ was right to say that it is not required as a rule of law. Everything depends on the directions and facts of a particular case, and the danger that the jury might seek to use the evidence of one complainant as evidence of his guilt on counts concerned only with another complainant.

The present case, however, is not materially different from F and Wall. Lloyd is of no assistance because there the trial judge did on at least three separate occasions invite the jury to consider degrees of similarity. Moreover, the foundation of the direction in Dye appears to have proceeded on a misunderstanding and extension of what was said by Potter LJ in Musquera. In our judgment the directions in this case were sufficient, and the whole course of the summing up only served to reinforce the direction of separate treatment. In particular, it was only where the evidence of the two brothers went expressly in support of the appellant’s abuse of the other that the judge, in suitably cautious terms, permitted the jury to regard the evidence of one as bearing on the evidence of the other. We do not regard the jury as having been in danger of using the evidence of the complainants as cross admissible in general, or as evidence of propensity. Of course, they had to form a view of the credibility of each of the witnesses whom they heard. That, however, they performed their duty of loyally giving separate treatment to each count is demonstrated by their different majorities throughout the counts, as well as by their failure to come to a verdict on count 12.

As it was, Dye was not originally in the forefront of Mr Barlow’s oral submissions. He presented them, as had his skeleton for the appeal, without reference to Dye (although it had figured in his original advice). It was only when we asked Ms Bracewell QC, who appeared for the Crown, if she knew of any authority bearing on Mr Barlow’s submissions, that, late in the day Mr Barlow referred us to Dye. In the event, we regard that caution, if caution it was, as being justified.

We therefore conclude that there was no material misdirection in this case and we dismiss this ground of appeal.

We would simply add this, although we do not rely on it for the purpose of our decision. Even if we had considered that there had been a misdirection, the jurisprudence shows that it is matter of degree. Especially in circumstances in which no satisfactory account could be given to us about any application or ruling under section 101 of the 2003 Act, we would have concluded that the convictions are not unsafe: in the light of the jurisprudence regarding section 101 as it has developed in this context (see Chopra etc), none of which would have been available to the judge within a few weeks of the bad character provisions coming into effect, we are satisfied that it would have been entirely appropriate in this case for the gateway provisions of section 101(1)(d) to have been operated so as to make the complainants’ evidence cross-admissible. We consider that we are entitled to take that into account in considering safety, whatever the judge might have ruled.

Mr Barlow submits that the judge erred in failing to give a direction to the jury about the possibility of conscious or subconscious contamination of D by reason of the complaints of the two brothers. He said that the jury should have been directed as follows:

He submitted that such a direction was necessary in a case where the abuse of which D complained was historic rather than current, having stopped some 11 years before the appellant’s arrest, but where the other complainants were complaining of abuse which continued until not long before that arrest; and where there was evidence that the brothers knew D and met and spoke with him. Although unrelated, they were all step-sons of the appellant and visited and stayed with him.

In our judgment, however, there was no need for such a direction. There was no evidence of any contamination between D and the brothers. On the contrary, it was D’s complaint to the police which brought the complaints of the brothers to light and it was only at the time of their video interviews that they learned of what had happened to D. The reason why D complained when he did, under the influence of his pregnant girlfriend, and not because of any suggestive information from the brothers, was evidence in the case. There was no evidence that the brothers’ complaints, or any conversations they might have held with him, had influenced D into making his complaints, which were recorded by the police in his witness statement of 23 May 2004, before the brothers were contacted and gave their evidence via video interviews. On the contrary, there was evidence in the contrary direction, because C said in cross-examination that D had told him (not the other way around), after the allegations had all come to light, that he had also suffered abuse from the appellant (but he did not go into the detail of it). Before that he, C, had not had an inkling of it (even though the appellant had mentioned to him that he missed doing it to D). The brothers’ complaint to their mother in 2003 was limited to her. In these circumstances Ms Bracewell submitted that it was not for the judge to start hares running. We agree.

This ground of appeal, which had started life as Mr Barlow’s primary ground, dealt with in his advice and skeleton at greater length than any of the others, was ultimately advanced as little more than a makeweight.

Mr Barlow submitted that the judge ought to have warned the jury about the dangers and unreliability of purported memories of early childhood. This ground was again directed to the case of D, where on one view of the evidence the earliest recollection took him back to the age of 3/4 years old. Mr Barlow referred to expert evidence of the unreliability of detailed early childhood memories which had been proffered from a Professor Conway: see R v. JH and TG (deceased) [2005] EWCA Crim 1828. That evidence was to the effect that there is childhood amnesia until about the age of 6 or 7, before which childhood memory is disjointed and patchy, so that detailed recollection should be regarded as unreliable.

However, such evidence has come to be regarded as unsatisfactory in itself: see R v. Jonathan CWS; R v. Malcolm W [2006] EWCA Crim 1404; R v. E [2009] EWCA Crim 1370. At most this controversial evidence, now sceptically regarded, could in any event relate only to counts 1 and 2 on the indictment (which, however, because they stretched over three years took D up to his sixth birthday). Moreover, D did not purport to remember early matters in any suspicious detail. Mr Barlow suggested that the jury should have been warned that even an honest and apparently credible witness, speaking of his extreme childhood, may be mistaken and then led astray by false recollection: as though this was the constant experience of the courts, as a sort of analogy to a Turnbull identification direction. Mr Barlow accepted that there was no authority to support this submission, but said that it was time there was one. We disagree. The difficulties of recollection of our early childhood are familiar to us all: although perhaps it is only those who have suffered abuse at an early age who can really understand the extent to which the abuse may be known even if the details of the surrounding circumstances are not. In any event the judge did warn the jury, in more traditional terms, of the problem of delay, the danger of prejudice to a defendant, and that this must be in the jury’s mind when deciding whether the prosecution had made them sure of the defendant’s guilt. He also cautioned them that the passage of time “may play tricks on memories”; and asked them to “Look at all of the evidence fairly and apply your collective knowledge of life in deciding where the truth lies”. In our judgment, these were entirely satisfactory directions.

After dealing with these four grounds separately, we also stood back and asked ourselves whether any combination of concerns deriving from any of them should cumulatively lead us to consider that these convictions are unsafe. We concluded that it did not.

These then are the reserved reasons why we announced our decision on 29 July 2011, shortly after the hearing of the appeal, that we were dismissing it.