R v BEF

Bad-character evidence must show a propensity to commit offences of the kind charged, not merely a general propensity to commit sexual offences. A conviction involving a mature adult does not, simply because the victim may be described as young or vulnerable, establish a propensity to abuse a young child. The court must assess similarities and differences in context, including age, relationship, setting, course of conduct and timing. Subsequent conduct must be capable of showing a propensity at the earlier time. Under the Criminal Justice Act 2003, limited probative value combined with potent prejudice requires exclusion where admission would undermine fairness.

The appellant was convicted in the Crown Court at Snaresbrook of sexual assaults and rapes against his stepdaughter, committed when she was aged seven to ten. The prosecution sought to adduce evidence of his later convictions for raping a mature adult woman. The trial judge admitted the evidence through the bad-character propensity gateway and declined to exclude it on fairness grounds. The appellant appealed, arguing that the convictions did not demonstrate a propensity to commit sexual offences against a child and that their admission rendered the trial unfair. The central issues were the probative significance of the similarities and differences between the offending, and whether the convictions were safe despite the admission.

1. Appeal allowed. The convictions were quashed and a retrial on all counts was ordered.
2. Under sections 101(1)(d) and 103 of the Criminal Justice Act 2003, the relevant issue was whether the proposed evidence showed a propensity to commit offences of the kind charged: sexual abuse and rape of a young child. A general propensity to commit sexual offences against females was insufficient.
3. The inquiry was fact-sensitive. The court had to assess similarities and differences overall, including the victims’ age and maturity, their relationship with the defendant, the setting, whether the conduct was spontaneous or part of a course of abuse, intoxication and the lapse of time. Vulnerability, abuse of power, oral sex and risk of discovery were not distinctive features on these facts and carried little weight. The court expressed no concluded general rule about the psychological causes of offending against adults and children.
4. The court applied the three questions identified in Hanson [2005] EWCA Crim 824. The convictions showed, at most, a general propensity to commit serious sexual offences against adult females. They did not make it appreciably more likely that the appellant had committed the particular offences alleged against the child. The analogy with Miller [2010] EWCA Crim 1578 was not sustainable because the factual differences were materially greater.
5. Under section 101(3), the damaging effect on the defence case was not by itself enough to require exclusion. However, the convictions had limited probative value and a very potent prejudicial effect. The trial judge had not conducted the necessary balancing exercise, so their admission was insupportable. The appellate court also applied the safety approach in Renda [2006] 1 WLR 2948. The summing-up did not cure the prejudice because it failed to identify the probative features or explain how the jury should approach the later offending.
6. The Court of Appeal could not conclude that the convictions would necessarily have been the same without the inadmissible evidence. The appellant was to be re-arraigned on a fresh indictment within two months and retried before a different judge. A reporting restriction was continued under section 4(2) of the Contempt of Court Act 1981 until the retrial.

1. Crown Court at Snaresbrook: The appellant was convicted of five counts of sexual assault on a child under 13 and four counts of rape of a child under 13. Evidence of later rape convictions was admitted.
2. Court of Appeal (Criminal Division): The appeal was allowed, the convictions were quashed and a retrial was ordered: [2023] EWCA Crim 1362.