O'Dowd v R

When deciding whether to admit disputed bad-character allegations under section 101(3) of the Criminal Justice Act 2003, the judge must assess their cumulative effect on trial fairness. The assessment must address the likely length and complexity of satellite issues, the burden on the jury, and whether evidence can fairly be selected, truncated or controlled by case-management directions.

The number of allegations is not decisive. Their nature, complexity, and the time needed to investigate contested allegations determine whether the trial risks losing its proper focus. A trial made excessively long and complex by such evidence may render convictions unsafe.

O'Dowd v R was an appeal against convictions at the Central Criminal Court before HH Judge Hawkins QC. The appellant was convicted of false imprisonment, threats to kill, rape, sexual assault and poisoning arising from allegations made by SS.

Before trial, the judge admitted evidence from three other women who alleged that the appellant had raped them. One allegation had resulted in an acquittal, one in a conviction, and proceedings on the third had been stayed for abuse of process. Each allegation was disputed. The trial lasted about six and a half months.

The central issue was whether admitting all the disputed bad-character evidence under section 101(3) of the Criminal Justice Act 2003, and the resulting satellite litigation, made the convictions unsafe.

1. Appeal allowed. The convictions were unsafe and could not stand.

2. Section 101 creates a two-stage process. Once evidence passes a gateway in section 101(1), section 101(3) requires its exclusion if admission would adversely affect the fairness of proceedings to the extent that it ought not be admitted. The fairness assessment requires the judge to look forward and evaluate the practical consequences of admitting disputed evidence before the trial begins.

3. The trial judge had considered each allegation carefully, and there was a good case for placing some similar-fact evidence before the jury. However, he had not adequately considered the cumulative effect of admitting three separate, contested allegations. Two concerned conduct 17 and 22 years before the index allegations. Each required detailed scrutiny and additional witnesses, creating three substantial satellite trials.

4. The court stressed that the risk depends not simply on the number of allegations, but on their nature, complexity and the time needed to prove them fairly. The judge should have considered using his case-management powers to impose a timetable, truncate the evidence fairly, or require the Crown to select its strongest allegation.

5. Bad-character evidence occupied 16 of the 42 evidential sitting days and about one-third of the summing-up. Combined with interruptions caused by the appellant’s health and changes of representation, the six-and-a-half-month trial made it very difficult for the jury to retain its focus on the charged offences.

6. The judge’s directions on the weaknesses in the bad-character evidence did not, by themselves, make the convictions unsafe. Ordinarily, however, a judge should draw together the cumulative strengths and weaknesses of such evidence in directions tailored to the case, rather than leaving them dispersed through a narrative review.

• Court of Appeal (Criminal Division): Allowed the appeal against conviction and held that the verdicts were unsafe.
• Central Criminal Court: The appellant was convicted before HH Judge Hawkins QC and sentenced to life imprisonment on three counts, with concurrent sentences on the remaining counts.