By Andrew Roberts
The question of how and by whom the reliability of expert testimony should be evaluated is problematic. For many, the prevailing approach in Australia is a cause for concern. However, the recent case of Honeysett v The Queen  HCA 29 presented the High Court with an opportunity to grasp the nettle. In that case, the issue was whether and how a jury could be assisted in comparing the image of the accused taken at a police station to an armed robber captured on a CCTV image whose head was covered by a hood by an anatomy professor pointing out similarities in the images and the absence of any differences.
In criminal trials witnesses are generally prohibited from expressing opinions on matters that are to be determined by the jury. Witnesses are generally expected to testify only to the facts. The drawing of inferences from those facts is the exclusive province of the jury. Opinions offered by witnesses are excluded because they are superfluous. The prohibition is subject, however, to significant exceptions, one of which permits an opinion to be expressed by a witness who possesses ‘specialised knowledge’ — provided that the opinion is ‘wholly or substantially based’ on that knowledge. Such witnesses are allowed to express opinions because the jury is thought to lack the knowledge and experience that would enable it to draw rational and reliable inferences. In such circumstances, the expert’s opinions are required to ensure that verdict returned by the jury is the product of sound reasoning.
Why reliability matters
What then of the issue of reliability? If the justification for allowing experts to offer opinions is that the jury lacks the competence required to draw the inferences drawn by the expert, is the idea that evaluation of the reliability of expert opinion should be left to the jury plausible?
There would seem to be some incoherence in an approach that on the one hand permits experts to offer opinions on account of the jury’s lack of competence in the subject-matter, and on the other trusts the jury to discount opinions that are insufficiently reliable (an issue that was acknowledged implicitly by the English Court of Appeal in R v Henderson  EWCA 1269). Nevertheless, the provisions of the Uniform Evidence Law governing the reception of expert evidence have been interpreted in a way that leaves the issue of reliability to the jury — a matter to be taken into account when determining how much weight ought to be placed on an expert’s testimony. In this respect the uniform evidence legislation is identical to the common law in England and Wales. The state of the law in that jurisdiction is considered by many — including members of the judiciary (see R v Henderson at ) — to be unsatisfactory, and the Law Commission has published a draft bill which if enacted would require trial judges to determine as a condition of admissibility whether expert evidence is sufficiently reliable.
It might be pointed out in defence of the approach in uniform evidence jurisdictions that experts’ testimony will be subject to cross-examination, during which matters that might have a bearing on the reliability of the methods used by the expert and the opinions that he or she expresses can be explored. But accepting the view that cross-examination provides an effective means of testing expert testimony requires us to be satisfied that those conducting cross-examination have sufficient understanding to engage with the expert on methodological and interpretative issues; that such engagement will expose any shortcomings; and that in cases of revelation the nature and extent of the problems with the expert’s evidence will be recognised and understood by the jury. In its critique of the English common law, the Law Commission observed (at [1.21]) that ‘there is a basis for believing that, where evidence of questionable reliability is admitted, it is not effectively challenged in cross-examination’. This view was formed in part in light of the submission of the UK Register of Expert Witnesses, which reported that the experience of its members was that ‘cross-examining advocates tend not to probe, test or challenge the underlying basis of an expert’s opinion evidence but instead adopt the simpler approach of trying to undermine the expert’s credibility’. If this is right, it suggests that the issue of reliability ought to be attended to at the point in the process at which a party seeks to introduce expert evidence. In other words, one response to concerns over the effectiveness of cross-examination as a means of exposing weaknesses in expert evidence, and the ability of juries to understand and take account of them, might be to require trial judges to undertake an inquiry into the reliability of an expert’s testimony in determining whether a party should be able to introduce it. The English Law Commission proposed a legislative scheme (not yet enacted) that would impose a gate-keeping duty on trial judges and lead to the exclusion of expert evidence that was not ‘sufficiently reliable’.
In Australia’s uniform evidence legislation, the provision that provides an exception to the general prohibition on opinion evidence has been interpreted in a way that precludes consideration of the issue of reliability. The relevant provision is s 79 which permits such evidence to be given by a person who has ‘specialised knowledge which is based on the person’s training study or experience’ and the evidence is based wholly or mainly on that knowledge. In R v Tang  NSWCCA 167, it was said that ‘the focus of attention [in determining admissibility] must be on the words “specialised knowledge”, not on the introduction of an extraneous idea such as “reliability”’. In other cases, fleeting references can be found to a requirement that the body of knowledge on which expert testimony is to be based on a reliable body of knowledge. But no attempt has been made to explain how the issue of reliability should be approached, and the courts have steered well clear of the idea that a formal and structured inquiry into reliability such as that developed by the US Supreme Court in Daubert v Merrell Dow Pharmaceuticals Inc, 509 US 579 (1993)) or proposed by the English Law Commission, is required by s 79.
Honeysett presented the High Court with an opportunity to revisit these issues. The appellant had been convicted of an armed robbery carried out at a suburban hotel. The event was recorded by CCTV cameras, which depicted one of the robbers holding a distinctive pink-handled hammer. That person’s head was shrouded in a white pillowcase or T-Shirt covering the head and face and leaving only the eyes exposed. The only other part of the persons anatomy revealed in the images was a small area of skin where there was a gap between the sleeves of a garment and gloves that were being worn. The prosecution case was that the hooded robber depicted in the images was the appellant. At trial, it relied on circumstantial evidence. DNA, which matched the appellant’s DNA was found on a white T-Shirt discovered in a car that was thought to be the getaway vehicle. Matching DNA was also discovered on the handle of the hammer used in the robbery, along with DNA belonging to an unidentified person.
The prosecution also relied on the evidence of an expert who had examined images of the robber and the appellant. Notwithstanding that almost all of the robber’s body was covered, based on his examination of the images the expert claimed to have identified various points of anatomical similarity and no dissimilarities between the appellant and the robber. These conclusions were based entirely on the expert’s perception. No form of anthropometric measurement was used, as the images were not taken from the same angle and body-positioning across the images was not the same. His observation of the images, he explained, did not differ from the kind of observation that could be carried out by a lay observer, save for the fact that his training and study in the field of anatomy meant that he had a better understanding of the shape and proportions of the human body. His observations concerning the robber depicted in the CCTV images were that:
He is an adult male of ectomorphic (thin, ‘skinny’) body build. His shoulders are approximately the same width as his hips. His body height is medium compared to other persons, and to familiar objects (eg doorways) visible in the images from the [offence]. He carries himself very straight, so that his hips are standing forward while his back has a very clearly visible lumbar lordosis (the small of his back is bent forward) overhung by the shoulder area. Although the offender covers his head and face with a cloth (what looks like a T-shirt) … the knitted fabric is elastic and adheres closely to the vault of his skull (= braincase). This shows that his hair is short and does not distort the layout of the fabric. The shape of the head is clearly dolichocephalic (= long head, elongated oval when viewed from the top) as opposed to brachycephalic (= short head, nearly spherical). The offender is right-handed in his actions. … Although most of the body of the offender is covered by clothing, head wrap and gloves, an area of naked skin above his wrist (between the glove and the sleeve) in images … is visible and can be compared to the skin colour of a female hotel employee on the same images.
In respect of the appellant he noted that:
[he] is an adult male of ectomorphic (= slim) body buil[d]. His hips and shoulders are of approximately the same width. His stance is very straight with well marked lumbar lordosis and pelvis shifted forward. His skull vault is dolichocephalic when viewed from the top. Comparison of lateral (side) and front views of his head also indicates the head … is long but narrow. His skin is dark, darker than that of persons of European extraction, but not ‘black’. … He is right-handed — uses his right hand to sign documents.
Other experts involved in proceedings raised concerns about the distorting effect of the lenses used in cameras. The accuracy of the images produced by a camera will depend on the physical properties of the lens. Lenses with differing properties will distort in different ways the appearance of persons depicted in images. Unless the properties of lenses in cameras used to produce two images are known, the reliability of conclusions concerning anatomical similarity of the persons depicted in those images will be compromised. The prosecution expert’s response to these criticisms of his methods was to state that he took the possibility of distortion into account in reaching his conclusions, but did not explain the way in which he did this. The expert expressed the view that the robber had short hair, as did the appellant, and the same shaped skull as the appellant. The expert had previously been challenged on his ability to discern the anatomical features of a head covered in fabric in another case (R v Morgan  NSWCCA 257) in which he gave evidence similar to that which he provided in Honeysett. On that occasion he explained that ‘he had worked with the garment industry some years before, and had learned that in the “construction of clothing, so-called hang of garments is extremely important and it is related directly to body shape and posture”’. He further explained that ‘the force of gravity spreads the cloth on the body in a way that is inescapable because the force of gravity always acts and therefore certain aspects of body shape can be deduced from the study of images of cloth to body’.
Honeysett’s first appeal
Honeysett was convicted of robbery and subsequently appealed. Among the grounds of the appeal was that the expert’s evidence should not have been admitted. It was claimed that it did not meet the conditions set out in s 79 of the Evidence Act 1995 (NSW), that is to say, his testimony was not based on ‘specialised knowledge’. The Court was satisfied however, that the expert, on the basis of knowledge acquired through training and experience
offered an expertly trained eye for observation of anatomical features, an expert capacity to compare a feature in one set of images with the cognate feature in another set of images and to describe them … as similar or dissimilar, an expert capacity to position anatomical features on a categorical scale accepted in the profession as a useful means of categorising like as opposed to unlike anatomical presentations of a particular feature and a common language for the descriptions.
The challenge to admissibility was renewed in the High Court.
An attempt to limit the opportunity? — The Crown’s revisionary account
The appeal presented the High Court with an opportunity to tackle a range of issues relating to the reception of expert evidence. The significance of its decision both for future practice (the practice of legal actors and of those engaged in forensic science) and our understanding of the law would depend on which issues it chose to focus. The Crown conceded that the only specialised knowledge that supported the expert’s evidence was his knowledge of anatomy. The Crown’s claim was that his evidence was not ‘body mapping’ evidence. Rather it was an account of the anatomical characteristics visible in each of the images, adduced to prove no more than that the appellant could not be excluded as the offender. The Crown also suggested that the witness had not examined the images over any substantial period.
What might prompt such concessions? One view is that they render the case something of a sacrificial offering. Although it made it more likely that the appeal would be upheld, the effect of the concessions, if accepted, would be to narrow the basis on which the decision might be made, and ensure that the prevailing approach to the admissibility of expert evidence — an approach that enables the prosecution to rely on ‘scientific’ opinions that have not been subject to validation studies that might establish how accurate and reliable they are — remains largely, if not wholly, intact. If the expert’s evidence were to be viewed as the Crown suggested it ought to be viewed — as an account of anatomical features rather than comparison evidence — the judgment would leave untouched the law relating not only to the admissibility of ‘body mapping’ evidence, but for all forms of comparison evidence in which reliance is placed on a ‘match’, including ‘facial mapping’ evidence, tool mark and shoeprint analysis, and fingerprint evidence. Were the claim that the expert had not examined the images for any significant period to be accepted, there would be no reason for the Court to consider the law concerning admissibility of evidence given by a witness, who through study of material available to a jury is said to acquire expertise ad hoc in that material. The concession that the expert’s field of study was anatomy, invited the Court to focus attention on the credentials of the particular witness, rather than issues affecting the admissibility of forensic comparison evidence generally. The various submissions and concessions would seem to increase the likelihood of the appeal being upheld, but hold out the prospect of a decision that would have few implications for the routine use of incriminating comparison (and expert evidence more generally) in criminal trials.
An opportunity passed up
The revisionary nature of the Crown’s account of the evidence given by the expert during the appeal hearing in the NSW Court of Criminal Appeal was not lost on the High Court. In the previous appeal, the expert had not limited his testimony to the provision of independent — in the sense that there was no cross-referencing of features identified in one image to those identified in the other — descriptive accounts of the features visible in each. He had concluded that the appellant and the robber shared eight anatomical features and that he was not able to discern any differences in the features visible in the two images. This was clearly comparison evidence, adduced for the purposes of establishing identity. Unfortunately, this issue was not pursued by the High Court. Instead it chose to decide the appeal on the basis that the expert’s testimony regarding the shape of the heads of those depicted in the images was not based wholly or substantially on his specialised knowledge. The reasoning that led to this conclusion, set out over 3 short paragraphs of the judgment, is not entirely clear. It was suggested that the expert’s evidence amounted to no more than his ‘subjective impression of what he saw when he looked at the images’. The expert’s evidence, it was said, ‘gave the unwarranted appearance of science to the prosecution case’, and the prosecutor could have invited the jury to examine the images and find the similarities themselves. In other words the expert was giving evidence that did not call for specialised knowledge. The jury were sufficiently well-placed to draw the relevant inferences.
But the problem with this suggestion is that given that the CCTV image of the robber depicted someone whose head was almost entirely covered with a loose fitting hood it would be reasonable to assume that the jury would not have been able to do what the expert claimed to have done. Curiously, no reference is made to this aspect of the expert’s evidence in the part of the judgment in which the Court sets out its reasons. Many might reasonably think (in common with other experts who gave evidence in the various proceedings) that the expert’s opinion was nothing more than guesswork. It is difficult to avoid the conclusion that the High Court was intent on avoiding discussion of reliability as a condition of the admissibility of expert evidence under s 79.
Although it had been said that the appeal in Honeysett would provide the High Court with an opportunity to define ‘specialised knowledge’ in a way that would ensure that trial judges would be attentive to issues of reliability in admissibility decisions, it is an opportunity that has been passed up. The decision is likely to be inconsequential. It has nothing to say about the admissibility of comparison evidence, about ad hoc expertise, or reliability as a condition of the admissibility of expert evidence under s 79 generally — aspects of the law that have been subject to cogent criticism and remain a matter of deep concern. With respect to the reception of ‘scientific’ evidence based on techniques and methods that could be tested but have not been, it seems likely that it will be business as usual in criminal trials.
Andrew Roberts is a Senior Lecturer at Melbourne Law School
AGLC3 Citation: Andrew Roberts, ‘Expert Evidence and Unreliability in the High Court: Honeysett v The Queen’ on Opinions on High (3 September 2014) <https://blogs.unimelb.edu.au/opinionsonhigh/2014/09/03/roberts-honeysett>.