A matter of opinion

Experts must engage with the law. Forensic biology expert Simon J Walsh notes that in mainstream international forensic journals, since 1990, only 0.9 per cent of articles examined legal issues (S. Walsh, (2004) 'Legal perceptions of forensic DNA profiling Part I: a review of the legal literature', Forensic Science International 155 (1) 51). Barrister Graham Cooke noted that one DNA expert did not know of the Criminal Procedure Rules and the duty they place on experts (G Cooke, 'Are all our experts and laboratories 'fit for purpose?'', Archbold News 2007). This seems inexcusable. The duties of expert witnesses, under the Criminal Procedure Rules, are now accessible through the internet.

Solicitor Mark Solon notes that, when assessed in law and procedure, many candidates, who had been experts for years, failed. This suggests they had been making the same mistake for years (Mark Solon, 'Experts: amateurs or accredited' 154 NLJ 7117 p.292). Real understanding must replace perceived understanding.

Solving the problem

Conversely, criminal practitioners should study forensics. Barristers are now educated in forensic accounting. One cannot help but be somewhat cynical that this particular discipline was identified for the potential financial reward it may attract. The House of Commons Science and Technology Committee noted that, in recent miscarriages, criticism had not been directed at the lawyers in those cases (House of Commons Science and Technology Committee, 'Forensic science on trial', Seventh Report of Session 2004-05 HC 96-I para.170). Given that counsel had asked inappropriate questions, in some cases training should be extended to forensic science.

A former DPP identified a further practical response, namely that an expert's qualifications should be confirmed (David Calvert-Smith QC, 'Medical evidence in criminal prosecutions', (2000) 68 Medico-Legal Journal 117-129). Evidence of this check should be retained on the solicitor's register of expert witnesses. This is helpful as Gene Morrison practised as a forensic expert for over 20 years with no instructing solicitor discovering he had purchased his qualifications.

In addition, credentials should be considered such as the exact area of expertise, qualifications, experience, evidence of continuing professional development, any adverse comments from advocates or judges and their standing within their profession.

Pre-trial negotiation may be used effectively. Experts may identify where they agree, where they disagree and why. Such clarity, early in proceedings, may help lawyers to advise their clients. Thought may also be given as to how to convey testimony clearly. Experts must, in turn, be aware that there is a debate as to whether images are instruments of logic, such as Wigmore maps, or instruments of rhetoric such as gruesome photographs (D Tait, 'Rethinking the role of the image in justice: visual evidence and science in the trial process', Law, Probability & Risk (OUP) 2007 6 311).

Pre-trial meetings may help eliminate error prior to trial. Erroneous thinking, such as inappropriate use of statistics, may be exposed pre-trial. Experts may only avail themselves of such opportunities, however, if they know how to make proceedings work for them. This further explains why experts should engage with the law.

The gladiatorial culture

Dr Pamplin, editor of the UK Register of Expert Witnesses, notes that 81 per cent of experts approved of pre-trial meetings:

'The provisions made for pre-trial meetings of experts will lead to a better, earlier assessment of expert evidence and will also save time by highlighting weak prosecution cases or forcing an early plea by the defendant. This view was supported by 90 per cent of the experts who responded to our survey. However, more than one expert reported that unless the current 'gladiatorial' culture in the criminal justice system can be changed, there is little chance of meetings of experts working in practice' (Pamplin, 'Underwhelming developments', 156 NLJ 7232 p.1082).

The 'gladiatorial' approach was identified by Mr Clements: 'I think I can count on the fingers of one hand the number of occasions on which'¦there has been agreement on the agenda.' Mr Clements argues that lawyers should be present to set the agenda and moderate behaviour as: 'I have had some bad experiences with my colleagues who just do not know how to behave.' (Clements, 'The changing role of the expert witness', Medico-Legal Journal 71 61 2003)

Mr Clement's view may, depending upon context, be incorrect. Experts should have received full instructions prior to pre-trial meetings. The expert should assume responsibility for preparation of their report (Paul Roberts, 'Science in the criminal process', at 479 (OUP)). Given the overriding objective of economy in Rule 1.1 of the Criminal Procedure Rules, lawyers may not always be present. Directions are less likely to indulge the idiosyncracies of experts who do not assume responsibility for preparing agenda, and reports, amicably.

Pre-trial openness may be achieved by removing legal professional privilege between a solicitor and an expert. Solicitor-advocate Julian Cohen astutely notes this is not prudent as it may increase costs by forcing solicitors to engage in lengthy non-minuted discussions, rather than recording information in a short letter (Julian Cohen, 'The Independent?', 145 NLJ 6704 p.1048).

Greater consultation is needed between solicitor and expert for pre-trial meetings to work. Expert witnesses complain that judges set unreasonable time limits. A judge noted this may be because lawyers do not know the time frame it will take the expert to prepare their report. Lawyers should ask this question of an expert who, in turn, should be aware of the need to provide such timelines.

Legal duty at the pre-trial meeting

At these pre-trial meetings experts must be aware of their legal duty. The Ikarian Reefer [1993] 2 Lloyds Rep 68 notes that: 'Expert evidence presented to the court should be, and should be seen to be, the independent product of the expert uninfluenced as to form or content by the exigencies of litigation.'

With respect the above statement does not sum up the duties of the expert. If it were not for litigation the expert would not be preparing a report at all. The report should be 'influenced' and geared towards the litigation. To fail in this regard may be to introduce irrelevant evidence. Experts should, however, not be partisan. This may, in turn, help to break down some of the barriers encountered by Mr Clements. Certainly, an expert's fees should not be contingent upon any outcome.

Julian Cohen notes that while lawyers become concerned when experts make concessions, it is actually the expert who refuses to make sensible concessions that damages the client most. The expert's credibility may be severely dented by such refusals. (Julian Cohen, 'The Independent?', 145 NLJ 6704 p.1048). Naturally, concessions pre-trial, not perceived by the jury, do the least harm to credibility in front of the jury.

At the pre-trial meetings experts must be alert to The Ikarian Reef's assertion that: 'Facts or assumptions upon which the opinion was based should be stated.' Again, this may be qualified. To state all underlying assumptions would be a task of infinite regress. Drawing attention to the key assumptions will, however, be exceptionally useful when experts meet to discuss their testimony.

Larger databases and greater research

In terms of the trial itself there are a number of potentially vacuous cross-examination techniques which experts should become aware of. One approach to examining experts is to question about the desirability of a larger database and greater research. Such an approach was adopted in Dallagher [2003] 1 Cr. App. R. 12 and Professor Vanezis conceded both points.

The difficulty is that this is a truism. It is inevitable that every discipline would benefit from greater research and larger databases yet this, in itself, does little to negate the discipline. Re-examination should simply serve to show the truism involved in this line of examination.

The same is true in respect of an expert witness conceding that they may be mistaken. Once again the degree of error may be clarified on re-examination as error is possible in every situation. The cross-examination extends only so far.

An oft-used question, with experts who frequently testify for the prosecution, is to ask the expert to identify how many times they have testified for the defence. This then permits assertions of potential bias. The expert needs to be alert to this. It is a useful, but limited, piece of evidence.

Laboratory standards

One prudent line of questioning is to check that experts have followed internal laboratory standards. Cooke notes that one DNA expert failed to draw attention to a peak which arguably showed a conclusion different to the expert's conclusion. Laboratory standards dictated that notes, relating to interpretation, be made on separate paper. When asked why the expert had not satisfied this internal quality requirement the expert stated 'to save paper'.

Equally, the laboratories' quality should be assessed. Graham Cooke notes that every scientist in one laboratory committed the prosecutor's fallacy.

It must be borne in mind whether the defence have had the opportunity to test the sample or whether this was depleted by prosecution experts during initial investigation. This may, in turn, affect weight.

Probability and statistics

Correct use of probability, and statistics, is a double-edged sword. On one hand it shows there is uncertainty to the jury. On the other hand the very act of attributing a probability to an outcome may suggest a certainty that in reality simply does not exist. The probability may not be able to factor in all variables.

In Dallagher Mr Van Der Lugt 'was sure' (at 199) and 'absolutely convinced' (at 200) that the earprint matched the defendant. Professor Vanezis thought it 'highly likely' (at 199). This was the case even though they were merely assuming ears were different and acknowledged that there was a 'real possibility' (at 199) that two separate ears may produce similar prints. This led Dr Champod to express concern about the terms in which the expert expresses opinions (at 200).

The court noted that experts had to express opinions in certain terms. This reasoning seems bizarre. The reason why a probabilistic testimony was not possible was because there was, effectively, no database on the frequency of given patterns of ear prints in the population. If there was a larger database a probability match could be calculated. Probability could not be calculated simply because there was no evidence to compare it against. It was the very absence of any supporting evidence that allowed the expert to be certain. Perversely, the greater the evidence base, the less certain the testimony.

The Law Commission is currently reviewing the rules on opinion evidence., Professor Dawid's (professor of statistics at Cambridge University) opinion on Clark (Sally Clark [2003] 2 FCR 447) provides succinct proof that questions of probability, if relevant, should be addressed by suitably qualified mathematicians. These measures are practical in nature, and do not require changes to the rules per se.