Judges have a special gatekeeper role, as set down by the Supreme Court in 1993 following the Daubert case, in which a company was sued for the damage its breast implants had done to women. The four factors that judges should consider when evaluating testimony on scientific matters from expert witnesses are: the testability, the error rate, the degree of acceptance by the relevant scientific community, and whether the results had been peer-reviewed and published. In a new decision this March the court said that judges may disallow testimony from engineers that does not meet the criteria laid down for scientists. They accepted an engineer's testimony and so supported a claim against a manufacturer whose tyre had had a blow-out - the resulting accident killed a child. It is often difficult for judges to assess evidence on the basis of the Daubert criteria when the experts giving their views on the reconstruction of an accident would neither publish their work nor have it peer-reviewed. But help may be forthcoming.
The American Association for the Advancement of Science, with support from several charitable institutions, has launched a five-year pilot project to supply judges with a list of experts to provide advice in complicated cases. The aim is to avoid the legal confusion when expert witnesses hired by each side dispute each other's evidence. The Internet could also help judges to find people respected in their field, though since the information on the Net is not peer-reviewed, it should be treated with extreme caution.
Trial lawyers seem, not surprisingly, to be much less enthusiastic; they protest that a change could tip the balance of the current adversarial system, and that putting a stamp of approval on a witness could overawe judges and juries. They apparently see a case being decided more by rhetoric than by reliable evidence. For scientists the work may not be rewarding, but it is clearly their public duty to help.
In Britain the adversarial system persists, though suggestions are being made that resemble the American proposals, namely that the court might insist on just one expert witness, instead of one from each side. And the courts do recognise that there are problems when dealing with science and technology; in special cases, an expert may sit alongside the judge to explain the technicalities.
The law needs to come to terms with the fact that science is not an adversarial system, and the most reliable scientific views are arrived at by consensus. The courts should be suspicious of maverick scientists; with rare exceptions, they are simply wrong. The core of science is remarkably reliable and it takes exceptional scientific and technical skills to make the right judgments in disputes. There are, however, people who are able to do so, and their skills should be used.
The writer is professor of biology as applied to medicine at University College LondonReuse content