T, R. v

A footwear-mark examiner may give an evaluative opinion founded on experience where there is a sufficiently reliable basis for it. The examiner may sometimes be confined to saying that footwear could have made a mark. A more definite opinion requires factors within the examiner’s expertise, such as an unusual pattern or size.

Current footwear-distribution data cannot support mathematical likelihood ratios or formulae. Such calculations create an appearance of precision without a reliable statistical foundation. Any evaluative opinion must be transparently explained in the expert’s report, identify its factors and data, and be presented as subjective expert judgment rather than as scientific precision.

The appellant appealed against a Crown Court conviction. Footwear marks found at the scene of a murder were compared with Nike trainers recovered from the appellant’s home. The trial expert gave evidence of moderate scientific support for the proposition that the trainers made the marks.

After conviction, fresh examination of the expert’s working papers revealed that he had used a likelihood-ratio formula and statistical assumptions which had not been disclosed to the parties, the judge or the jury. The appeal considered the admissibility and proper presentation of evaluative footwear-mark evidence, and whether the undisclosed methodology rendered the conviction unsafe under section 23 of the Criminal Appeal Act 1968.

1. Appeal allowed; conviction quashed. The undisclosed calculations and assumptions could reasonably have affected the jury’s verdict. The footwear evidence was not transparently presented, and the court could not regard the conviction as safe.

2. Expert evidence is admissible only where it has a sufficiently reliable scientific basis. That threshold is not confined to evidence supported by a statistical database. An expert may give an evaluative opinion founded on experience where there is another sufficiently reliable basis, subject to testing at trial: Reed & Reed.

3. A footwear examiner may compare pattern, size, mould detail, wear and damage, and may take account of the relevant dates. In some cases the examiner can go no further than saying that the footwear could or could not have made the mark. In others, factors such as an unusual pattern or size may justify a more definite evaluative opinion. The opinion must remain confined to matters within the examiner’s expertise and must not evaluate the other evidence in the case.

4. There was no sufficiently reliable statistical basis for calculating a likelihood ratio from footwear pattern, size, wear and damage. Footwear distribution and use are affected by changing and unmeasured variables. A mathematical formula would therefore give an illusory appearance of objective probability. The court followed the warnings in R v Dennis Adams [1996] 2 Cr App R 467, R v Doheny [1997] 1 Cr App R 369 and R v Adams (No 2) [1998] 1 Cr App R 377 against using Bayes theorem or similar methods outside fields with a firm statistical foundation.

5. Where an examiner gives an opinion beyond could have made, the report must disclose the factors, data and method used. It should make clear that the conclusion is subjective and based on experience. The word scientific should not be used if it suggests a precision and objectivity which the field does not possess. Compliance with Part 33 of the Criminal Procedure Rules permits reliability issues to be resolved at a pre-trial hearing.

6. Applying Pendleton [2001] 1 WLR 72, the fresh evidence might reasonably have affected the jury’s decision. The appeal was therefore allowed and the conviction quashed.

• Court of Appeal (Criminal Division): Allowed the appeal and quashed the conviction: [2010] EWCA Crim 2439.
• Crown Court: Convicted the appellant. No citation for the Crown Court decision is stated in the judgment.