The Queen v Crawford

An appellate court reviewing a trial judge’s findings based on witness credibility and reliability must recognise the trial judge’s evidential advantage. It should not reverse those findings merely because it would assess the evidence differently. Intervention requires an error of principle or material showing that the judge was plainly wrong. A trial judge need not resolve every issue raised at trial, only those necessary to decide guilt or innocence. Evidence bearing on the truthfulness of other witnesses need not undermine the accuracy of a principal eyewitness unless it has a logical connection with that accuracy. The absence of a defendant’s DNA or fingerprints may support the defence but is not determinative. Inferences about DNA deposition or retention require an evidential basis; speculation is insufficient.

The respondent was convicted by a judge sitting without a jury of possessing an unlicensed firearm. The prosecution case depended substantially on a police officer’s evidence that, during a foot chase, he saw the respondent throw a handgun into nearby bushes, where a matching gun was later found.

The Court of Appeal of the Cayman Islands quashed the conviction. It considered that the trial judge had failed to address discrepancies between photographs of the gun and had misunderstood the significance of the absence of the respondent’s DNA. The Crown appealed to the Privy Council. The central issues were the proper appellate approach to findings based on witness assessment, the relevance of the photographic discrepancies to the eyewitness evidence, and the proper treatment of the DNA evidence.

1. Appeal allowed; conviction restored. Lord Hughes, delivering the judgment of the Board, advised that the Crown’s appeal should be allowed. The Court of Appeal had not identified an error of principle warranting interference with the trial judge’s verdict.
2. An appellate court reviewing primary factual findings based on witnesses whom the trial judge saw and heard operates at a real disadvantage. Unless the trial judge failed to use that advantage, palpably misused it, or was plainly wrong on material evidence, the appellate court should not reverse merely because it would evaluate credibility, reliability or probabilities differently. This applies to judge-alone criminal trials, subject to the criminal standard of proof.
3. A trial court is not required to make findings on every question raised at trial. It must resolve matters necessary to decide the issue before it. Here, the relevant question was whether the officer accurately saw the respondent throw the gun, not whether the gun had later been moved or the scene evidence improved. Any possible untruthfulness by officers involved in the later search did not logically affect the accuracy of the earlier observation.
4. The trial judge had addressed the photographic discrepancy, the possibility of evidential improvement and the risk that the eyewitness had assumed he saw a gun because of the earlier complaint and the respondent’s history. He was entitled to assess the witness’s accuracy independently and to rely on supporting circumstances, including the respondent’s statement after arrest and the discovery of a gun where the officer said the object had landed. No corroboration was legally required.
5. The direction that a person handling an object may leave no recoverable DNA or fingerprints, and that their absence was supportive but not determinative, was correct. The presence of another person’s DNA did not exclude handling by the respondent, since successive people could handle the gun. The suggested greater likelihood of DNA retention because the respondent’s hand was hot or his grasp firm had no evidential foundation and was speculation. The judge was not required to repeat his correct analysis in his concluding paragraph.

• Privy Council: allowed the Crown’s appeal and restored the conviction: [2015] UKPC 44.
• Court of Appeal of the Cayman Islands: quashed the respondent’s conviction, holding that the trial judge had mishandled the photographic discrepancy and the absence of the respondent’s DNA.