Smith, R. v (Rev 1)

Where credible fresh evidence creates a real and substantial dispute about evidence central to a conviction, the appellate court must ask whether it might reasonably have affected the jury’s decision. Under the test in [2002] 1 WLR 72, a conviction is unsafe if a jury might reasonably have accepted the fresh evidence and consequently lacked surety of guilt. Strong remaining circumstantial evidence does not cure that defect where it has material weaknesses.

The court also made obiter observations that contested fingerprint evidence should be obtained, recorded and presented transparently, with access to competent independent expertise and robust quality standards.

The appellant, the deceased’s next-door neighbour, was convicted of murder at the Crown Court at Nottingham and sentenced to life imprisonment with a minimum term of 30 years. The prosecution case included circumstantial evidence concerning the appellant’s finances, dealings with the deceased, opportunity, accounts of his movements and footwear marks.

A blood-marked fingerprint on a door handle was treated at trial as the appellant’s. The appellant’s principal fingerprint expert was not called after her qualifications were challenged. After conviction, the court received fresh evidence from two further fingerprint experts who disputed the identification. The central issue was whether that evidence might reasonably have affected the jury’s verdict and therefore rendered the conviction unsafe.

Disposition

1. The appeal was allowed and the conviction for murder was quashed.
2. The Crown accepted that the conditions in section 23 of the Criminal Appeal Act 1968 permitted the court to receive the fresh fingerprint evidence. Applying R v Pendleton, [2002] 1 WLR 72, the question was whether the evidence might reasonably have affected the jury’s decision to convict.
3. The fresh expert evidence disclosed a real and substantial dispute about whether the blood-marked print on the door handle was the appellant’s. The jury could reasonably have accepted the analysis of the new defence experts. If it did, the fingerprint evidence would no longer identify the appellant as the person who left a print likely to have been made by the killer.
4. There was a strong circumstantial case, but it had weaknesses. The crime scene had been compromised, no blood traces were found in the appellant’s house or car despite the likelihood that the killer was covered in blood, and the prosecution timing case was open to challenge. A jury might therefore have concluded that the Crown had not made it sure of guilt.
5. The court added that, had it needed further support for that conclusion, it would have considered the disadvantage caused by the absence of the appellant’s principal fingerprint expert, the inadequate pre-trial expert reports, and the presentation of the fingerprint evidence in a form which did not realistically enable the jury to resolve the expert dispute.

Obiter observations

The court said that its concerns about fingerprint practice were not material to its decision. It nevertheless stressed the importance of independent fingerprint expertise, judicial rather than police determination of expert competence, detailed contemporaneous records, transparent comparison material and effective presentation of disputed evidence. It invited examination of common quality standards through a robust and accountable system.

• Court of Appeal (Criminal Division) — by leave of the single judge, received fresh fingerprint evidence and quashed the murder conviction.
• Crown Court at Nottingham — convicted the appellant of murder on 12 December 2008 and imposed life imprisonment with a minimum term of 30 years.