Harcourt Chambers1 Garden CourtCoram ChambersNYASHind CourtDNA Legalimage of 4 Paper Buildings logoGarden Courtsite by Zehuti

Home > Articles > 2013 archive

Expert Evidence - Another Chapter in a Continuing Story

David Bedingfield, barrister at 4 Paper Buildings, sets the latest reforms relating to expert evidence in their historical context and considers the dilemmas which expert evidence has long presented to the courts.

David Bedingfield, barrister, 4 Paper Buildings

Are we witnessing a sea change in the treatment and consideration of expert evidence? The most obvious aspect of the shift, of course, is the new imposition by the Family Division of the requirement that a trial judge find an expert's evidence to be "necessary" before permitting instruction.

But an examination of the history of the use of expert evidence shows that the reforms that have been implemented over the past several years, including the Civil Procedure Rules 1998 and the new PD 25 of the Family Procedure Rules 2010, are merely part of a century-long struggle by courts to provide an answer to the problems posed by expert evidence. This new rule is, I submit, simply a further chapter in this on-going story.

The expert paradox
The expert, as we all know, is expected to give an opinion about the most significant issues in a case. 1 A paradox underlies the use of all expert evidence: the reason an expert is required is that the decision-maker lacks the expertise of the expert and requires that expert's help. How is that same decision-maker also competent to judge the content of the expert's evidence? How is the decision-maker to choose, for example between two competing experts, each using different methodologies beyond the ken of any non-specialist? 2

Judges, and Parliament, have been seeking an answer to that question since at least the late 18th century. One of the responses always canvassed is this: cede the decision to a specialist panel of experts. Courts, however, have been jealous of their jurisdiction to decide disputes, and (it is submitted) for good reason: courts in the United Kingdom have deservedly developed a reputation for fairness and for careful consideration of the competing interests involved in any dispute.  Courts have also now developed a considerable body of law, giving to litigants and their advisors at least some reasonable certainty that like cases will be treated alike.

This move to restrict expert evidence must therefore be understood as part of a century-long struggle by the judiciary to maintain its jurisdiction, and to deal with the ever-increasing complexity presented by disputes involving scientific or medical evidence.

The history of the expert
First, the principles:

1) all litigation can be reduced to core disputes;

2) all disputes can be reduced to competing stories, and each story-teller seeks to tell the decision-maker a narrative that convinces the decision-maker to decide the dispute in that party's favour;

3) an expert, unlike any other witness, can give an opinion about the competing narratives on offer.

A court must treat an expert with some deference, because a judge is only a specialist lawyer, and is not a specialist in whatever scientific (or other) point is in dispute.  The expert, on the other hand, is considered an expert precisely because he or she has an expertise the judge does not have.

As Tal Golan tells us in his magisterial history of expert evidence, the "creation story" for most historians looking at the question of experts in the courtroom is the case of Folkes v Chadd [1782] 3 Doug KB 157, where Chief Justice Gould (and a jury) had before them the classic civil dispute that cries out for "experts" to tell us what to do. 3

 The case concerned Wells Harbour, a natural harbour surrounded by low-lying salt marshes. The marshland surrounding the harbour was fertile, and several local farmers began in the late 1760s to drain the marshland for agricultural use. The harbour at the same time began silting up, threatening the shipping industry that was growing in an around Wells Harbour. The question, of course, was the cause of the silting: was it the farmers' new embankments, or would the silting have occurred in any event? The harbour commissioners claimed the embankments caused the silting. The farmers denied this. The harbour commissioners sued the farmers, and the battle lines were drawn.

Both parties were well-funded, and both sides therefore commissioned experts. The harbour commissioners relied on anecdotal and experience-based experts. Those witnesses set out in factual terms the rapid deterioration of the harbour after the embankments were built. Civil engineers also gave evidence for the commissioners, based on the factual background given by earlier witnesses, and offered the conclusion (which, of course, was the ultimate issue in the case) that the embankments had caused the deterioration in the harbour.

The landowners brought in the scientist John Smeaton, a member of the Royal Society. Smeaton sought to apply "scientific principles," and gave the opinion that natural causes predominated—the harbour's silting process was inevitable.

The trial judge, Chief Justice Gould, excluded Smeaton's evidence. In the justice's view, Smeaton's opinion was not based upon enough specific observations and direct evidence, and therefore was nothing more than speculation the jury should not be permitted to hear. Lord Mansfield and the Court of Appeal, however, allowed the landowners' appeal, holding that the jury should hear Smeaton's evidence.  In Mansfield's view, Chief Justice Gould, when he excluded Smeaton's evidence, took from the jury the real issue to be decided: between the two competing scientific theories, which theory is more likely than not correct?

Lord Mansfield's view that the trial judge should not play an overly restrictive gatekeeping function therefore became the common-law rule. 4 The use of experts became, if not commonplace, at least not rare, certainly not rare in the more well-funded actions in the QBD. A party in a civil or criminal matter was permitted to present to a jury expert evidence that presumed to answer the ultimate issue before the jury, though of course the "ultimate issue rule" was employed to force experts to tip-toe around actually responding to the question the jury had to answer. 5 Golan sets out several 19th century examples: Was a new sugar-refining process prone to explode? Did a copper-smelting operation create acid rain and destroy a season of crops? Did pollution from alum works contaminate well water? As Golan shows, in each case the plaintiff's experts gave one conclusion, the defendants' experts gave another. The trial judges did not seek to examine prior to the trial the scientific methodologies used by the competing experts, but simply left all of that to cross examination before the jury.

The expert in the criminal courts
The criminal courts also saw the rise of the expert after Mansfield's judgment, but as Professor Katherine Watson has shown (See "Medical and Chemical Expertise in English Trials for Criminal Poisoning, 1750-1914"; Cambridge Journal of Medical History, July 2006, pp 373-390), it would appear there were few opportunities for juries in criminal cases to hear "battles of the experts" prior to the mid-1830s.  Professor Watson shows why: in several cases where defendants had allegedly poisoned the victim by arsenic or other toxic chemicals, the only evidence before the jury was from a poorly trained apothecary or surgeon who had only a basic knowledge of chemistry and toxicology—which of course were fields in their infancy in any event. Many times, even in the absence of any contrary expert evidence, the jury simply wasn't convinced, even though it would appear from all accounts that in fact the victim had been poisoned.

Police and the Home Office therefore sought to instruct better, more sophisticated experts. As Professor Watson tells us, by the 1830s the Home Office had settled on a few well respected and trusted experts in the most difficult cases.

In 1831, the Society of Apothecaries mandated that candidates for its diploma attend lectures in medical jurisprudence. Guy's Hospital established a lectureship in medical jurisprudence the same year. The young surgeon Alfred Swaine Taylor (1806-1880) was given the post, and, as Watson tells us, for the next 40 years Taylor worked tirelessly to establish a profession of forensic toxicologists. His textbooks influenced generations of scientists.

Other toxicologists also emerged during the 1830s, including, in Bristol, William Herapath (1796-1868). Herapath was largely self-taught, but developed such a reputation for brilliance he was regularly consulted as a professional analyst. He became one of the founders of the Bristol Medical School, which opened in 1832. He lectured on chemistry and toxicology there until 1867.

Herapath's appearance in 1835 at the trial of Mary Ann Burdock, where he successfully demonstrated the presence of arsenic in a body that had been buried for 14 months, marked a first in English forensic analysis. It led to a career for Herapath as an expert witness that was second only to Taylor's.
The call for reform
The professionalization of the expert witness in both civil and criminal cases in the 1840s and 1850s led (inevitably, some would say) to the charge that some experts would say whatever they were being paid to say. Golan cites newspaper stories from the 1840s that revealed how juries were "bewildered, perplexed, and left in despair as to knowing how to decide" when presented with competing expert evidence. Judges were frustrated and "disgusted at the partisanship" of the witnesses. (P.62)

By the 1860s, Golan shows us, judges and commentators were pressing for reform. Golan sets out a series of lectures given in 1860 at the Royal Society of Arts in England that sound stunningly modern; in fact, the reforms suggested in 1860 have been implemented, one by one, in the last century and a half. Some are being implemented today. What were they?

1) Get rid of the jury. Concerned about the ignorance of the fact-finder and the paradox of demanding that a non-scientist decide questions of science? Take the decision from the lay fact-finder and give it to a professional. As Golan shows us, the supposed impossibility of jurors understanding difficult scientific evidence played a large role in the eventual elimination of the jury from civil cases in the United Kingdom. 6

2) Give the court, not the parties, the sole right to commission an expert report. The scientific community since early Victorian times sought to address the problem presented by the spectre of the "hired gun" expert who would say whatever he was paid to say by giving to the court (rather than to either party) the sole power to appoint an expert in the case. Michigan and Rhode Island, in the US, passed Acts in the early 1900s mandating court-appointed experts. Golan discusses proposals from bar associations to establish official lists of experts in the mid-nineteenth century. The Children Act 1989 (and the Family Proceedings Rules 1991) provided that only if the trial judge agreed could experts be instructed. (The test to be applied with regard to whether experts were required was from 1991 until 31 January 2013 simply whether the expert evidence would be helpful to the decision-maker. That changed, of course, in February 2013.) And of course in 1998 civil judges in England and Wales were given control of the instruction of experts—subject, of course, to Article 6 of the ECHR and basic common-law fairness.

3) Scrutinise carefully the methodology of any proposed expert. Courts should not be permitted to rely on expert evidence unless it was deemed "reliable" by the medical or scientific community. (In other words, Lord Mansfield got it wrong when he said that in the end the jury should in most cases be permitted to hear whatever expert the parties submitted for their consideration.) The United Kingdom Law Commission, in March, 2011, essentially sought to restate this principle in its recommendations regarding the use of experts in criminal jury matters.  It is clear that the new rule of "necessity" imposed by the Family Division follows from this view: that heretofore courts have not scrutinised carefully enough the actually expertise of the expert. Judges, in this view, must be experts themselves, at least about the narrow issue involved in the dispute before them. The Law Commission proposed that in criminal cases there be made available to the judges an expert panel with the sole purpose of commenting upon the proposed "reliability" of any expert's report in dispute. 8

4) Finally, and as a last resort, give the decision to a panel of experts.  It is no secret that the coalition government considered very carefully whether applications under Part IV of the Children Act 1989 should be taken away from the judiciary and placed in front of a tribunal, much like a Mental Health Tribunal or a Special Educational Needs Tribunal, with no funding for legal aid for the parties. That, at present, seems no longer to be the case, but it is right that Parliament will certainly do something to stop the rise in expenditure caused by lengthy care proceedings unless the judiciary imposes its own cost-cutting measures.

150 years later, a call for more reform
How, then, should we consider the recent changes imposed upon litigants in family proceedings? Readers will of course know that the Family Procedure Rules 2010 were amended to require, as of 31 January 2013, that judges first find that an expert is "necessary" prior to giving approval for the instruction of that expert. Two panels of the Court of Appeal have now addressed the meaning of "necessary", and it is clear that the test makes it likely that the use of experts in family proceedings will decrease. 9

Again, however, the changes must be viewed in the context of several developments over the last two decades. First, it was not assumed that litigation under the Children Act 1989 would become as expert driven as it has become. It was assumed by Parliament that care proceedings under Part IV of the Act would require no more than 12 weeks, primarily because it was assumed that social services departments would have fully assessed the child prior to issuing proceedings. Courts in most cases, in other words, would simply rubber-stamp the local authority's decision. That has not proved correct, for two reasons: 1) local authorities were in many cases staffed by incompetent social workers who were perceived as being unable to assess the sometimes complex family dynamics presented in difficult cases under Part IV of the Children Act 1989; 2) litigants in cases under Part IV of the Children Act 1989 were fully funded by legal aid, and courts routinely gave litigants permission to instruct experts who might contradict the local authority's conclusions. Cases therefore became lengthier and more expensive as experts were brought in to assess and re-assess the parents and those who might potentially parent the child. This expense has grown exponentially since implementation of the Children Act 1989 in October, 1991.

The second development in the last several years, of course, is that public funding for litigation is being radically reduced. There is a rationing, of sorts, that has now been put in place by a judiciary that is seeking to preserve, as far as possible, a system that is  facing a twenty to thirty per cent reduction in its funding.

And the third (and to my mind equally important) development is this: judges who hear these cases are better trained and more specialised tribunals than in the past. These judges are increasingly not helped by certain kinds of "expert" evidence.  Perhaps the most obvious examples would be evidence from psychologists regarding either risk assessment of dangerous parents or regarding the likely causes of bad parenting by parents with a "personality disorder". Another is the assessment by a "drug and alcohol" specialist regarding the risk of someone resuming a dependency on controlled substances or alcohol.

The evidence for this development is of course nothing more than anecdotal, but there are studies now available (and widely read by judges) that at the very least present a troubling picture. A report produced by the University of Central Lancashire (with funds provided by the Family Justice Council) was published in February, 2012, and identified a number of current criticisms of psychological reports: 10

• Psychological evidence has been presented as scientific fact in cases where the evidence amounted to no more than speculation and conjecture.

• The theories on which the reports were based were often untested.

• Evidence is too often based on concepts not accepted generally in the field, concepts that have not been (and many times cannot be) demonstrated empirically.

• Psychometric evidence has been submitted as scientific fact when it does not meet the criteria.

• There is an overuse of jargon by psychologists, with no explanation of the terms used.

• In too many cases, few attempts are made by the psychologists to evaluate the credibility of the source of information; allegations are therefore often reported as facts.

• Risk assessments focus primarily on unstructured clinical and actuarial approaches, rather than on structured clinical assessment, with or without actuarial anchoring.

The report is required reading. Almost 20 per cent of the psychologists who submitted reports were not qualified, in the view of the assessors, to provide a psychological opinion. Nearly all of the witnesses were not maintaining clinical practices, but were instead full-time expert witnesses. There was an over reliance on psychometrics. Some of the psychologists used defunct or out of date assessment techniques. Too many of the reports made comments about mental health when there was clearly no expert background in that area. Two-thirds of the reports were rated as "poor" or "very poor". 11

In response to this study (and no doubt prompted by cost-cutting concerns as well), the coalition government on 6 May 2013 announced consultation regarding "new national standards to raise the quality of experts used in family courts".  (The consultation ended in July.) The government wants to make certain that only "qualified, experienced and recognised professionals" provide evidence. According to the government, "[f]or too long there has been an increasing trend in England and Wales for expert witnesses to provide unnecessary and costly evidence—in the form of further written statements, clarifications and additional court appearances." Family Justice Minister Lord McNally, in his statement announcing the consultation, relied upon the independent Family Justice Review by David Norgrove, noting that Norgrove and his team had identified "weaknesses in the quality of evidence" given in family proceedings. (The announcement also noted, and revealingly, I contend, that some £52 million had been spent in the 12 months prior to October 2011 on expert reports in family proceedings.)

It has to be accepted that in part the decision to restrict the instruction of experts to cases where it is "necessary" rather than simply desirable or helpful is budget-driven and represents a rationing of scarce resources. But it is also undeniable that experienced family judges (in other words, the High Court and the appellate judiciary) see case after case where psychologists or other experts spend 40 pages to re-state the blindingly obvious.  The point of experts is to bring expertise to the case that is not ordinarily available to the decision-maker. It is clearly the view of the President of the Family Division, and the Court of Appeal, that in too many cases trial judges do not focus sufficiently on the question of whether the supposed expert will really bring expertise that is not already available to a well-trained judge.

The question to be answered by the Court of Appeal or the Supreme Court, of course, will come in the hard case: for example, a non-accidental injury shaking case where there will be clear disagreements between distinguished experts. A court deciding this type of case, it is submitted, must be given the discretion to determine that both experts are "necessary". In other words, decisions regarding these life and death cases must not be driven solely by the demands of the Chancellor of the Exchequer.



1 I avoid the use of the term "ultimate issue," but it is now accepted—after some 200 years of debate—that in fact experts regularly give their opinion about the ultimate issues to be decided by the jury.
2 This was famously pointed out by Judge Learned Hand in his 1901 article for the Harvard Law Review (perhaps the most cited article about evidence in all of legal literature): How can we expect jurors to decide between experts when the jurors' ignorance is the premise for allowing the expert to testify in the first place?  (See Learned Hand, Historical and Practical Considerations Regarding Expert Testimony 15 Harv L Rev 40, 54 (1901).)

3 See Golan, Laws of Men and Laws of Nature: the history of Scientific Expert Evidence in England and America; Cambridge: Harvard University Press (2004).

4 As Golan shows us, however, it is overly simplistic to say that Mansfield LJ knowingly created the current regime of experts countering experts.  That instead was the fault of a gradual accretion of adversarialism in the 19th century and increasing control of evidence by the parties in the face of a complacent judiciary. (See P50-51.)

5 See  Re: A and BC Chewing Co Ltd (1968) 1 QB 159; Cross on Evidence, 7th ed, p. 501; See also R v Stockwell (1993)  97 Cr App R 260, The Times, 11 March 1993; ". . . . if there ever was such a prohibition [against an expert giving an opinion on the ultimate issue] it has long been hallowed in the breach."

6 In 1846, Parliament created the County Courts, where judges were able to decide both issues of fact and law.  Many litigants preferred the cheaper and quicker option of a civil trial in the County Courts before a judge rather than the expensive trial before a jury in the Queen's Bench Division, and that became the preferred option. In 1854 a change in the rules in the QBD gave litigants in civil cases the option of choosing a judge rather than a jury, and many, if not most, chose to have a judge.  In 1933, Parliament provided in the Administration of Justice (Miscellaneous Provisions) Act 1933 that juries should be empanelled in cases involving allegations of fraud, libel, slander, malicious prosecution, false imprisonment, seduction, and breach of promise of marriage, and that judges in the QBD had a discretion whether to empanel juries in other civil cases: it quickly became almost impossible to convince a judge to empanel a jury in any case where scientific or other expert evidence was likely to be heard  (In 1981, Parliament in the Senior Courts Act 1981, s. 59, removed seduction and breach of promise of marriage from juries.)  In 1966, it was held that juries did not have the ability to assess damages in personal injury cases.  See Ward v James [1966] 1 QB 271, CA, per Lord Denning. Again, the primary reason it was believed juries could not perform this task was because of the prevalence of expert evidence regarding the impact of the injury on the claimant, and the likely disagreement of the medical experts regarding both causation and damages.  In 1990, a claimant in the London Underground fire at King's Cross sought to have a jury hear his case. The court refused to do so, citing Denning LJ's reasoning in Ward v James. See Singh v London Underground, The Independent, 26 April 1990.

7 The Law Commission was chaired by Munby P, currently the President of the Family Division. The Commission examined several miscarriages of justice that had occurred after poorly reasoned and shoddy expert evidence had been relied upon by judges and juries, and concluded that trial judges should play a more active "gatekeeping" role with regard to expert evidence. In particular, trial judges should apply a new reliability-based admissibility test for expert opinion evidence. The opinion evidence of an expert should be admissible only if the court is satisfied that it is sufficiently reliable to be admitted. To determine whether the evidence is sufficiently reliable, the court must apply the following test: 1) is the evidence predicated on sound principles, techniques and assumptions;  2) have the principles, techniques and functions been properly applied to the facts of the case; 3) is the evidence supported by those principles, techniques and assumptions?  See Law Commission Report No 190: "Expert evidence in Criminal Proceedings in England and Wales".

8 That proposal of course brings with it its own problems: will the parties have an opportunity to cross examine the panel when a decision goes against a party? Will the parties have access to all of the work of the panel?

9 See In the Matter of TG (A Child) [2013] EWCA Civ 5Re H-L (A Child) [2013] EWCA Civ 655.

10 The Central Lancashire researchers sought to undertake a small, largely qualitative study to explore the quality of expert psychological reports submitted in family proceedings. Using recognized experts in the field, the study sought to judge quality by considering first the qualification of the expert giving evidence, and second by developing and applying a framework of quality measures. The quality measures were developed by drawing on established experts in the field.

11 For an example of judicial disquiet about expert reports, see Pauffley J in the case of IA (A Child) [2013] EWHC 2499 (Fam) where she criticised an expert who produced a lengthy paediatric overview when asked only to produce a psychological report.