Atkins & Anor v R

Factual and Procedural Background

The appeals concern the appropriate manner in which an expert in facial photograph comparison may express his conclusions. The expert provided evidence comparing the face of an offender caught on indistinct CCTV footage with that of the Defendant. Neither the Crown nor the expert claimed a positive identification, and the defence did not contest the admissibility of the expert's evidence regarding facial similarities. The contention was that the expert should not have expressed the level of support the similarities lent to the allegation that the man in the footage was the Defendant, using a sliding scale of support ranging from "lends no support" to "lends powerful support".

On the night of 5th to 6th November 2006, a team of three men committed two armed robberies in the west London area. The first victim, an elderly widow, was attacked and bludgeoned to death during a robbery at her home. The robbers then proceeded to a second house where they attempted a violent burglary but were repulsed. The Defendants, brothers, were convicted of murder, two aggravated burglaries, wounding with intent, and a firearms offence.

Evidence established one robber was a man who had died while on remand. The Defendants were known to have been in company with him shortly before and after the offences. Both Defendants denied involvement, each presenting an alibi. CCTV footage from the second robbery showed one of the robbers briefly unmasked, and this footage was examined by the facial comparison expert.

The expert, a medical artist with over 40 years' experience and about 20 years specialising in facial comparison, used various techniques including morphological examination, measurements, and superimposition to compare the CCTV image to photographs of the Defendants and others. He eliminated all but the Defendant based on these comparisons and identified multiple facial similarities. He acknowledged limitations such as image quality and the absence of a statistical database for facial features. The expert expressed his opinion using a sliding scale of support, placing the evidence between "lends support" and "lends strong support" to the proposition that the man in the footage was the Defendant.

The trial judge admitted the expert's evidence except for the part expressing the level of support, which was challenged by the defence. The judge carefully directed the jury on the limitations of the evidence and the subjective nature of the expert's opinion. Both Defendants were sentenced to life imprisonment with a minimum term of 35 years, and they sought leave to appeal against the sentence length.

Legal Issues Presented

1. Whether a facial photograph comparison expert may express an opinion on the level of support the similarities lend to the identification of a person, using a sliding scale of expressions, in the absence of a statistical database.
2. Whether such an expression of opinion risks misleading the jury by suggesting a scientific or statistical basis where none exists.
3. The admissibility and proper scope of expert evidence in facial comparison cases.
4. The appropriateness of the sentence length imposed on the Defendants.

Arguments of the Parties

Appellant's Arguments

• The expert's use of a sliding scale to express the level of support for identification was impermissible due to the lack of a statistical database underpinning such expressions.
• Such expressions risk conveying a false impression of scientific certainty and may mislead the jury into overvaluing the evidence.
• The absence of objective measures means the expert should be confined to identifying similarities or dissimilarities without expressing a degree of support.

Respondent's Arguments

• The expert's opinion based on extensive experience and comparison techniques is admissible and valuable to the jury.
• Prohibiting expression of opinion on significance would leave the jury with raw data and no guidance, risking unsafe conclusions.
• The sliding scale is a conventional and acceptable way to communicate subjective expert judgment, provided the jury is properly directed on its limitations.

Table of Precedents Cited

Precedent	Rule or Principle Cited For	Application by the Court
R v Gray [2003] EWCA Crim 1001	Caution regarding facial imaging expert evidence and the absence of a statistical database; expert opinion should not extend beyond identifying similarities.	The court noted Gray expressed strong caution but did not consider all relevant authority; the current court doubted Gray would have restricted expert opinion as strictly.
Clarke [1995] 2 Cr App R 425	General admissibility of facial comparison expert evidence.	Supported the principle that facial comparison evidence is admissible.
Briddick [2001] EWCA Crim 984	General admissibility of facial comparison expert evidence.	Reinforced Clarke's ruling on admissibility.
Attorney-General's Reference 2 of 2002 [2003] 1 Cr App R 321	Expert opinion on image comparison is admissible; jury must have access to images.	Confirmed expert facial comparison evidence can be given to assist the jury.
R v Stockwell (1993) 97 Cr App R 260	Expert status and admissibility of opinion on ultimate issue in facial comparison evidence.	Rejected challenges to expert's qualifications and admissibility of opinion on identification issue.
Hookway (unreported 1998)	Facial mapping evidence can be left to the jury even when expressed as "very powerful support".	Confirmed admissibility of expert opinion expressed on a scale despite absence of database.
R v Gardner [2004] EWCA Crim 1639	Expert may give opinion on degree of probability of images being the same; no rule prohibits going beyond listing similarities.	Directly supports the admissibility of expressed expert opinion on significance of facial comparison findings.
R v Tang [2006] NSWCCA 167	Discussion on admissibility of sliding scale expressions in photographic comparison; lack of scientific basis for sliding scale.	Distinguished between admissible facial comparison evidence and inadmissible posture comparison; did not conclusively rule out use of sliding scale for facial comparison.
Luttrell [2004] EWCA Crim 1344	Expert evidence admissible even if not verifiable by scientific discipline.	Supported principle that expertise can justify admissibility despite lack of scientific verification.
O'Brien [2006] EWCA Crim 1741	Sentencing principles regarding totality of offences.	Supported the judge's approach to sentence length reflecting combined offences.

Court's Reasoning and Analysis

The court carefully analysed the role and limits of expert evidence in facial photograph comparison cases. It acknowledged the absence of a statistical database recording the frequency of facial features in the population but rejected the argument that this absence precludes an expert from expressing an informed opinion on the significance of similarities found. The court emphasized that an expert's judgment, based on extensive study and experience, provides valuable guidance to the jury, which would otherwise be left to interpret raw data without assistance.

The court reviewed relevant authority, distinguishing the present case from earlier cautionary rulings such as in R v Gray, which lacked full argument and relevant precedents. It upheld the principle from R v Gardner that experts may express opinions on the degree of probability that images depict the same person.

The court recognized the risk that sliding scale expressions might be misunderstood as scientifically precise but concluded that, with proper judicial directions emphasizing the subjective nature of such opinions and the lack of a statistical basis, their use is permissible. The court stressed the importance of judicial summing up to clarify these points to the jury.

Further, the court noted that excluding such opinions entirely would impair the ability of parties to challenge or support facial comparison evidence effectively. It also noted that similar sliding scale expressions are commonly used in other forensic comparisons, often without comprehensive databases.

Regarding sentencing, the court found the minimum term of 35 years justified by the gravity of the offences, including murder with aggravating factors and serious violent offences against the second household, reflecting the totality of criminal conduct.

Holding and Implications

The appeals against conviction are dismissed and the applications for leave to appeal against sentence are refused.

The court held that a facial photograph comparison expert, properly qualified and experienced, may express an informed opinion on the significance of similarities using a conventional sliding scale of expressions, even in the absence of a statistical database. This opinion must be clearly presented as subjective and non-statistical to the jury. The ruling affirms the admissibility of such expert evidence and clarifies the limits and necessary judicial directions to safeguard fairness.

The decision does not establish new precedent restricting expert opinion but rather confirms and refines existing principles, balancing the utility of expert judgment with the need to avoid misleading the jury. The sentence imposed on the Defendants was upheld as appropriate given the seriousness and totality of the offences.