From Law wiki, the wiki for law research

[1995] 1 AC 596 (HL). The defendant was tried and convicted of multiple counts of sexual offences against his step-daughters. His appeal was primarily on the basis that the complainants had colluded in their evidence and that, therefore, following Dpp v boardman (1975) and Dpp v p (1991) their evidence should not have been admitted together. To allow the jury to hear the evidence of both complainants in each indictment was to prejudice the jury beyond what the value of the evidence justified.

It was accepted that the evidence of the complainants was cross-admissible under the 'striking similarity' principle of Boardman and earlier cases; what was at issue was whether the judge should have excluded it because there was a strong possibility that it had been fabricated.

The House of lords ruled that, in general, it was not the role of the judge to determine whether evidence had been fabricated in collusion between the complainants. This was the jury's job. Therefore a question of admissibility should not turn on whether or not there was collusion, but only on the probative value of the evidence if it were accepted as true.

The problem with this ruling should be obvious. While the correctness of the argument that the job of determining whether testimony is to be believed falls to the jury cannot be disputed, in ruling on admissibility the judge must have regard to the credibility of the testimony. If this were not the case, then evidence of criminality offered by one complainant would always be admissible against the other. Why? Because such evidence is conclusive proof of the defendant's guilt, and no judge can logically conclude that evidence that proves the defendant guilty has insufficient probative value. The consequence of R, therefore, is to remove all judicial discretion where evidence is put forward for cross-admissibility between indictments.

See evidence of disposition for discussion.