As a chapter of Caroline Keenan's and my book Child Abuse Law and Policy across Boundaries (OUP 2007) is devoted to the reliability of expert evidence relating to child abuse, this is an area of particular interest to me.
I fully endorse the recommendations in this consultation paper. In our book we compare the approaches of several jurisdictions and argue that a more systematic analytical approach should be taken to the admissibility of expert evidence. So I agree that the Law Commission's preference for a Daubert-style test and guidelines for the trial court should be adopted. However I think it would be more practicable and efficient to have them incorporated in a Statutory Instrument or Practice Direction rather than legislation, making them easier to amend after experience with them.
That said, I submit that the risk to the criminal justice system is not only the admission of bogus evidence (such as Dr Colin Paterson's 'temporary brittle bone syndrome'); a more endemic risk is sloppy and incompetent practice and unfounded and speculative pieces of testimony. The Law Commission's admissibility test (¶ 6.26) would not intercept the opinion evidence of Prof Roy Meadow nor of Dr David Southwell (whose erasure from the medical practice register for serious professional misconduct was upheld by the High Court last week: Southall v General Medical Council [2009] EWHC 1155 (Admin)). Both of these paediatricians were internationally recognised experts in diagnosing physical child abuse, and had testified many times before the English criminal and family courts, so their biases and their opinions which ventured into territory in which they were not qualified, such as statistical probability, were more difficult for child protection agencies, counsel and the courts to identify.
Like another respondent, I would commend to the Law Commission all four volumes of The Hon Stephen T Goudge, Commissioner, Report of the Inquiry into Paediatric Forensic Pathology in Ontario (1 Oct 2008). Like Dr Meadow and Dr Southwell, Dr Charles Smith, the Director of the Ontario Paediatric Forensic Pathology Unit, was a highly plausible and persuasive witness who inspired deference in professional colleagues, child protection agencies and criminal justice practitioners and judges; like them he was strongly predisposed to 'think dirty' and find that a child had died of non-accidental injuries; like them he was the cause of miscarriages of justice.
Mr Justice Goudge stressed that the trial judge's role as gatekeeper in determining whether the expert scientific evidence has sufficient threshold reliability to be considered by the trier of fact should not be confined to so-called 'novel science'. He recommended that both forensic pathology and the criminal justice system should be vigilant about the following aspects of a forensic pathologist's opinion:
1. whether the substance of the opinion and the language in which it is expressed as susceptible to varied meanings or otherwise does not elucidate the pathology issues at stake;
2. whether the level of confidence or certainty that the expert has in the opinion is accurately expressed;
3. whether the opinion addresses other explanations for the pathology findings;
4. whether the opinion is in an area of controversy within the forensic pathology community;
5. whether all or part of the opinion falls outside the pathologist's area of expertise;
6. whether the opinion is based, in whole or in part, on non-pathology information provided to the pathologist;
7. whether the opinion relies, in whole or in part, on other expert opinions provided to the pathologist; and
8. whether the opinion includes the facts and the reasoning process relied on to form the opinion.
Obviously these criteria are applicable to all areas of opinion evidence. So I would urge that the guidelines which the Law Commission proposes for experience-based expert evidence (¶ 6.35) should be applied to all expert witnesses, even those testifying in areas of recognised empirically-based research.
Because of these concerns, I am disappointed that the Law Commission seems to have rejected the option of compulsory accreditation or registration for expert witnesses (¶ 6.75), on the basis that it would be impracticable and and unduly costly and would not bring any additional benefit in terms of quality assurance. Having called many expert witnesses during my own civil practice as a barrister, I know that the scrutiny which is required of expert testimony is very costly for the parties. So I believe that the cost of compulsory accreditation need to be balanced against the cost to the criminal justice system of either conducting a case-by-case review of every expert's qualifications and scope of expertise, or of letting unreliable evidence be heard by the jury. Obviously accreditation would not relieve the trial judge of the obligation to ensure that the expert's opinion was relevant to the facts in the issue and based on the factual evidence in the case, but it would provide a basis of identifying, not just bogus science and bogus scientists, but also the parameters of expertise of particular professional witnesses. So I would hope that accreditation will receive further consideration in your eventual recommendations, with a more detailed analysis of the costs and benefits.
I agree that the same evidentiary reliability test should be applied to defence evidence as to prosecution evidence (¶ 6.57), as no party is entitled to have taken patently unreliable evidence admitted. I also, with some hesitation, agree to the recommendations regarding court-appointed assessors. I am hesitant because my own experience as counsel of tribunals being assisted by their own experts has not been a good one, as it was exceptionally difficult to challenge the experts' expertise and in camera advice to the tribunals. So I strongly agree that it is essential that the parties have an opportunity to question the assessor and the basis of his or her views (¶ 6.68). Finally, I agree that the question of evidentiary reliability should as a matter of practice be determined before the jury sworn, to ensure that the court is not pressed to make quick decisions for logistical reasons associated with delay for the jury, witnesses and the courthouse. The factual matrix for the opinion should be evident from the expert report itself, and so a preliminary pre-trial ruling should be feasible. In situations, however, where the evidence called in the case turns out not to support that opinion, then I believe that a voir dire might be necessary to (re)consider the issue of admissibility. I also strongly support the availability of an interlocutory appeal, so that a trial is not held unnecessarily where the prosecution's case depends upon an expert’s evidence which prima facie is unreliable.
Laura Hoyano
Fellow & Tutor in Law, Wadham College, Oxford