Legal Affairs

Current Issue


printer friendly
email this article
letter to the editor

space space space

September|October 2004
Continental Divide By Jeffrey Rosen
Newtonians and Charlatans By Daniel J. Kevles
The Other Detainees By Serena Hoy

Newtonians and Charlatans

Is the scientist on the stand spouting quackery? Two new books weigh in on how expert testimony has bedeviled courts for centuries.

By Daniel J. Kevles

SCIENCE IS UBIQUITOUS IN THE CONTEMPORARY LEGAL SYSTEM, a frequently contentious presence in criminal and civil actions ranging from murder to malpractice, from mass tort litigation to the government's regulatory proceedings. When scientific witnesses offer contradictory data and conflicting conclusions, they prompt complaints that the courts are being swamped with "junk science." Perhaps. But a good deal of scientific evidence is based on established science—for example, the chemical composition of poisons or the atomic spectra of paints—which is difficult to contest. There is also a hefty amount of testimony about what I'll call "claimant science"—science that is unsettled, in process, and debated out of the courts as well as in them. DNA fingerprinting, for example, has now been settled, but it was controversial in the late 1980s.

As Tal Golan demonstrates in Laws of Men and Laws of Nature, an original and deeply researched book, discontent with scientific expertise has been a feature of the Anglo-American legal process since the late 18th century, when scientists started moving more frequently into the witness box. The conflicts increasingly involved claimant science. One such dispute led to the founding doctrine for determining who can serve as a scientific expert. The rule was pronounced in 1782 by Lord Mansfield, chief justice of the King's Bench in England, in what is commonly known as the Wells Harbor case.

Wells Harbor, which lies on the northern coast of Norfolk, England, was silting up. The board of commissioners for the harbor argued that the cause was a privately constructed embankment to reclaim land for farming. They sued to compel the owners of the land to cut down the embankment. In their defense, the owners argued that the degradation of the harbor was nature's doing. Distinguished scientific experts were enlisted to argue each side.

On the side of the commissioners were four senior civil engineers, skilled, practical, and unconcerned with general laws of nature. They took direct measurements of water flow in the harbor in order to demonstrate that it had been impaired by the embankment. The principal scientist on the side of the landowners was John Smeaton, a civil engineer, too, and a specialist in harbors, but also a member of the Royal Society, and well-versed in Newtonian theory. He accounted for the history of the harbor in terms of the dynamic flow of tidal waters and the deposition of sediment. The latter ultimately raised the marsh land to such a height that the backwater scouring of the channels steadily declined, choking the harbor with silt.

The commission's lawyer, George Hardinge, contended that Smeaton should not be permitted to speak to the jury. Hardinge insisted that, since Smeaton's analysis rested on general laws, it would not provide the jury with a factual basis for judgment. "I cannot believe that when the question is, whether a defect arises from a natural or an artificial cause, the opinions of men of science are not to be received," Lord Mansfield wrote in deciding the appeal. "The cause of the decay of the harbor is also a matter of science, and still more so whether the removal of the bank can be beneficial. On this such men as Mr. Smeaton alone can judge."

Commentators have long argued that Mansfield's opinion stands for the proposition that any kind of science should be allowed into the courts. Golan, a historian at the University of California, San Diego, has probed the records of the Wells Harbor case to provide a convincing reinterpretation. He argues that the issue for Mansfield was whose expertise counted: Only that of men of tested experience, such as the four engineers engaged by the commission? Or also that of the Newtonian philosophers who possessed and deployed recondite knowledge of the laws of nature? Mansfield accepted Smeaton's views, Golan explains, because they represented claimant science that was verging on becoming the establishment view. Smeaton exemplified "the growing importance of natural philosophy as a worthwhile, even useful pursuit." It was as such, in Mansfield's judgment, that the scientist's testimony was worthy of admission to the courts.

MANSFIELD HAD DISREGARDED SMEATON'S OBVIOUS PARTISANSHIP, preferring to associate him with the kind of disinterested authorities he himself was in the habit of consulting. Golan notes that, at the time, men of science identified themselves and were seen by the community as honorable, unbiased gentlemen, honest with each other as well as about nature. Scientists in the witness box, however, steadily diverged from Mansfield's ideal. They were already hired guns. The more claimant science entered the courts—which it did in suits involving chemistry, patents, medicine, pollution, and murder—the more scientists were presented with opportunities for emolument. By the mid-19th century, their testimony, a seeming cacophony of contradictions, raised doubts about their integrity and their science in the eyes of the legal profession and the public. In 1862, an editorial in the popular magazine Quarterly Review noted that judges were increasingly of the view that scientific testimony, "which ought to be the most decisive and convincing of them all, is of all the most suspicious and unsatisfactory."

What could be done? Victorian scientists seeking to establish themselves as an authoritative professional community were discomfited by such charges. The aristocrats among them deplored the mercenary undertakings of a growing number of their colleagues. A much larger group proposed to reform scientific witnessing by making the experts servants of the court, independent of the parties to a suit. But the practice of hiring partisan experts had come to stay. The reform proposals ran too counter to the adversarial system and the right to a judgment by one's peers guaranteed by the jury system. In England, the problems arising from dueling scientific witnesses were resolved instead through judicial discretion. As Golan points out, a British judge could filter expert scientific testimony, giving guidance to juries about which expert's views were the more authoritative and even ordering a trial without a jury if he thought the complexities of the case warranted it. Subjected to these restraints, the problem of conflicting expert testimony began subsiding in the English courts by the late 19th century.

In the United States, however, the problem was worsening. By way of illustration, Golan offers a vignette of the nation's expanding community of microscopists. This group of scientists struggled through self-regulation to limit what their members could say in murder trials about the identification of blood. The issue was whether human blood could be distinguished from animal blood under a microscope by the appearance and size of their red corpuscles. Most microscopists held that the blood of different mammals could not be differentiated, and so were reluctant to testify when a defendant's life might be at stake. They could distinguish mammalian blood from that of reptiles, birds, or fish, ï┐Żbut "to say more than this," the president of the American Microscopical Society declared to the membership in 1892, is "to imperil human life, and . . . to make scientific experts more dangerous to society than the very criminals they are called upon to convict."

Yet self-regulation was not uniformly practiced. Eager to reap the rewards of testifying, some microscopists continued to claim that their techniques could distinguish between human and animal blood. Such tests made their way into many trials through the latter half of the 19th century. American judges may not have been seduced by the claims, but, unlike their British brethren, they had no way to keep them out. In about three-fourths of the states, legislative prohibitions or voluntary agreements by the courts prevented judges from charging juries on questions of fact. Under these circumstances, Golan writes, the only means judges had to influence the jury's reception of expert testimony was to control its form—for example, by allowing responses only to hypothetical questions, rather than to ones about the evidence at hand. That approach worked well with X-rays, which were the subject of scientific testimony within a year of their discovery.

In general, however, the exposure of the courts to problematic science was compounded after the turn of the 20th century by the proliferation of new disciplines such as experimental psychology. Unlike the physical sciences, the field was regarded as intellectually suspect, marked by unfounded generalizations, inherent contradictions, and the absence of a coherent body of theory. Hugo Münsterberg, a practitioner of the beginning science, tried to convince American courts to turn to psychological experts to assess the veracity of testimony. Their method, developed in Europe, implausibly sought to test the memory of witnesses by assessing their associations and measuring their reaction times. John Henry Wigmore, dean of the Northwestern University School of Law and his generation's leading authority on evidence, soundly attacked Münsterberg's ideas, contending in part that psychology was not yet even a claimant science.

Rebuffed, a number of psychologists turned to measuring the physiological changes provoked by the emotions to which offering testimony could give rise. This turn led William Marston, a Harvard graduate who had worked in Münsterberg's laboratory, to devise a means for detecting lies by monitoring a witness's systolic blood pressure. The test seemed highly reliable. In 1922, Marston volunteered it to the defense in the murder trial of James Alphonso Frye in Washington, D.C. The trial judge refused to admit Marston's lie detector on a technicality. The real problem, Golan argues, was the impertinence of claiming that a machine could determine truth better than the legal process could. In 1923, the Court of Appeal of the District of Columbia upheld the trial judge's decision, ruling the lie detector inadmissible on the grounds that it had not yet won "general acceptance in the particular field in which it belongs."

General acceptance among whom, Golan rightly asks, and how was the answer to be determined? The court's cursory opinion in Frye cited no authority and provided no clue. Yet despite the flimsiness, its rationale for excluding scientific testimony came to dominate American legal thought. Golan argues that the Frye rule appealed to judges because it empowered them to determine not only whether an expert could testify but also, contrary to Lord Mansfield, the level of his expertise.

Yet they were not empowered enough. By the early 1990s, highly questionable science—if not altogether junk—was said to be flooding the courts, especially in mass tort and personal injury cases. For example, so-called experts testified that physical trauma causes cancers, that trace chemicals in the environment stimulate a multiplex of diseases and disorders, and that birthing difficulties give rise to cerebral palsy. In the 1993 case Daubert v. Merrell Dow, in which the lower court rejected a claim that the anti-nausea drug Benedictin produced birth defects, the Supreme Court went beyond the Frye rule to lay down criteria by which trial judges could assess the reliability of scientific knowledge. Judges were to ask, Was the knowledge testable? Had it been subjected to peer review? How frequently did it yield errors? Was it well standardized in its applications?

Faithfully applied, the court's ruling may be helping to staunch the flow of questionable science. But it can't solve the problem on its own. Much claimant science is likely to pass through the Daubert gate, leaving unresolved the problem of expecting lay juries to sort through technical conflicts between experts. That ongoing and pervasive challenge may compel Americans, like our English counterparts, to allow judges to guide juries on the merits of scientific witnessing. That would mean less rigid adherence to the principle of trial by jury, but the trade-off might be worth it.

IN HIS BOOK LABORATORY OF JUSTICE, David Faigman takes the Supreme Court to task for persistently failing to inquire into the merits of the scientific evidence figuring in the cases before it. "For 200-plus years the Court has demonstrated remarkable insensitivity to empirical questions," he contends. "Its factual jurisprudence is slapdash, sloppy, and, too often, supercilious."

Faigman, a professor of law at the University of California, Hastings, reaches this conclusion by reviewing the encounters with science that have been integral to the court's evolving approach in key areas of constitutional law: equal protection, due process, freedom of religion, and the right to privacy. Much of the book traces familiar territory. But if his subject matter is well known, Faigman ably deploys it to make the point that the Supreme Court should engage more actively with science. He effectively spotlights cases in which the court inadequately scrutinized the science before it, or burdened a scientific conclusion with greater constitutional weight than was wise or warranted.

In the 1990 case Employment Division of Human Resources of Oregon v. Smith, the court upheld an Oregon drug control law that barred Alfred Smith from taking peyote during a Native American religious ceremony. In reaching this conclusion, Justice Sandra Day O'Connor, siding with the majority, gave credence to Oregon's claim that allowing Smith to take peyote would undermine the state's capacity to fight a war on drugs and would possibly harm Smith and other peyote users. Faigman notes that O'Connor could have inquired as to whether any empirical studies supported the state's rationales, but she did not. In dissent, Justice Harry Blackmun noted that a number of states, as well as the federal government, were letting Native Americans use peyote without any evident adverse effect on law enforcement or the health of the users.

Turning to the social science behind Brown v. Board of Education and to Roe v. Wade, Faigman argues, as others have before, that the court's decisions in both cases rested on findings that were at best inconclusive. On its own, Kenneth Clark's research into the impact of segregated schools on the personality development of black children could not justify ending segregation. And determining the point at which a pregnancy becomes viable could not on its own support Roe's trimester approach to allowing abortion. In these difficult and thorny cases, maintains Faigman, the court "used science to disguise a policy choice." Yet as Faigman acknowledges, Blackmun delved into the science in Roe, drawing on his experience as counsel for the Mayo Clinic when he spent a summer enmeshed in the study of fetal viability. This is the sort of inquiry that Faigman prescribes, yet he faults the inferences the justice drew, and reasonably so. The underlying problem is that inquiries such as Blackmun's seek to navigate the conflicts in claimant science that Golan's history of expert witnessing reveals. Compounding the difficulty, much of the expertise in play in Supreme Court cases concerns social science, an area of frequently unsettled knowledge. Unsettled science today may become established science tomorrow, or it may fall by the wayside. In either case, the claimant science of the moment offers a hazardous foundation on which to erect principles of constitutional law.

Daniel J. Kevles is the Stanley Woodward Professor of History at Yale. He periodically teaches at Yale Law School. His past works include The Baltimore Case and he is currently working on a book about the history of intellectual property in living organisms.

printer friendly email this article letter to the editor reprint premissions
space space space

Contact Us