Have pity, please, on jurors called upon to absorb complex scientific and medical information in the courtroom. In 2007, when the manufacturers of the artificial sweetener Equal challenged a competitor’s advertising slogan that “Splenda is made from sugar, so it tastes like sugar,” the jurors had to hear expert testimony from chemists and biologists detailing the molecular structure of sucralose and sucrose, the chemical derivation of household sweeteners, and the results of fluorescence spectroscopy and synchrotron studies. In the end, the warring parties reached an out-of-court settlement. But after their courtroom ordeal, the jurors’ morning muffins must have tasted a bit less sweet.
Scientific evidence has long played a prominent role in patent cases, but increasingly scientific and medical experts are weighing in on criminal trials and in litigation involving product liability and commercial disputes. New areas of legal specialization have sprouted up as a result: scientists who have moved to the practice of law to handle technologically complex cases and the creation of science and technology divisions within law firms. Leonard, Street and Deinard recently formed one such group in Minneapolis. “We formed it because we’re seeing a lot more science appear in cases, cutting across new boundaries,” says Jim Bullard, who co-leads the practice. “Now science can crop up in employment and breach of contract cases, almost anywhere.” In lawsuits involving the termination of scientific employees, for instance, employers might have to present evidence showing that the work of the employee was substandard. “So we put together a resource in our firm for all of our lawyers to use. We need to tell our clients’ science stories effectively,” Bullard says.
Typically, people feel more persuaded by emotionally moving stories than by scientific evidence, especially when the science is confusing or in conflict. During the silicone breast implant litigation of the 1990s, jurors frequently handed down verdicts in favor of plaintiffs who alleged that the implants had harmed them, despite the lack of statistical or epidemiological evidence that any harm had actually taken place. Why? Because the plaintiffs clearly were hurting, although nobody knew why. When the science is unclear, jurors go elsewhere to formulate their thoughts. Alluring junk science sometimes fills that void. How can judges and juries without specialized technological knowledge possibly interpret the scientific evidence that litters their path? How can they tell the difference between good and junk science, especially when those boundaries are blurring, particularly in the pages of the nation’s press?
The astronomer Carl Sagan described pseudoscientific efforts as those that “use the methods and findings of science, while in fact they are faithless to its nature—often because they are based on insufficient evidence or because they ignore clues that point the other way. They ripple with gullibility. With the uninformed cooperation (and often the cynical connivance) of newspapers, magazines, book publishers, radio, television, movie producers, and the like, such ideas are easily and widely available. Far more difficult to come by are the alternative, more challenging, and even more dazzling findings of science.”
Sagan’s definition suggests that unreliable science should be easy to spot with just a little diligence. Such is not the case. Intelligent people who want to do a good job in the jury box succumb to junk science all the time. In the early 20th century, countless defendants were convicted of serious crimes on the basis of microscopic hair analysis, which government studies debunked as worthless during the 1970s. (Government investigations of the hair technology began when it became clear that analysis experts could not agree on criteria for comparing hair found at crime scenes with the hair of suspects; still, most people convicted in part on the basis of this evidence remained in prison for years afterward.)
Hair analysis and other forms of junk science flourish because they satisfy an innate human tendency called “hindsight bias.” When we know that someone has been the victim of a crime or has fallen sick after using a particular product, we tend to allow our knowledge of the outcome to affect our attempts to understand its cause. In the process, we overlook evidence and facts that may not agree with our perception of a cause-and-effect connection. Junk science, with its questionable methodology, false interpretations and unsupported conclusions, can help us build the trail of evidence that leads to our conclusion. “I’ve seen junk science come out of highly credentialed experts, ranging from partial truths, selective use of data and facts, misleading graphical representations, and distorting calculations by manipulating the input,” says Dan Colton, co-leader of Leonard, Street and Deinard’s science group. “That’s what makes dealing with junk science so difficult.” In the silicone breast implant cases of the 1990s, for example, some junk science experts gave testimony postulating a connection, unsupported by clinical trial data, between silicone and the overstimulation of the immune system, which they said destroyed the health of many women.
Dispensers of questionable science—who often look and sound no different from purveyors of proven science—benefit from the elevated status given to scientific evidence compared with human testimony. So when some jurors saw that some women with silicone breast implants had contracted mysterious autoimmune syndromes, they concluded that the implants were the cause despite sound science to the contrary. The junk scientists had human suffering on their side, while the legitimate scientists could only less persuasively hold that the women’s anguish was unrelated to their implants. (In 2001, the Food and Drug Administration ended its decade-long moratorium on the use of the implants, although several manufacturers had already gone bankrupt following product liability decisions against them.)
Like other forms of bad evidence, however, it is possible to demolish junk science in the courtroom. Opposing counsel can cast light on data the experts are ignoring and expose the limitations of their findings. “One fun thing about these cases is that junk experts can be pinned down. They have to agree with you on the fundamental principles of science,” says Jim Bullard. “If you can back them into that corner, they will end up agreeing with you. And when you get them into that corner, they’re caught.”
Yet a long line of court decisions has established that scientific experts are subject, if challenged, to a preliminary screening that might prevent jurors from ever hearing their testimony. In Frye v. United States, a 1923 murder case that hinged upon the reliability of an early lie detection device, the U.S. Supreme Court ruled that scientific evidence should be admitted only if the expert’s opinion was generally accepted in the scientific community. (The Court ended up disallowing the results of the lie detection test.) Fellow scientists, not judges, determined the acceptability of expert testimony. Thus, reliable and relevant testimony could be excluded if the scientific community lagged behind in recognizing the scientific techniques being challenged. Minnesota courts have adopted and still use the Frye standard of gatekeeping.
In 1993, the Supreme Court heard the case of Daubert v. Merrell Dow Pharmaceuticals, which was brought by people who asserted that a drug called Bendectin caused birth defects. The Court’s ruling replaced the Frye criteria with a new set of standards that required federal judges to ensure that the opinions of challenged experts withstood peer review, had been widely accepted, had undergone testing and carried reasonable rates of error. Judges, not scientists, decided whether to exclude the testimony. A later Supreme Court decision expanded the ability of judges to apply the Daubert standards to other kinds of expert testimony.
How do a Daubert hearing and the resulting judicial gatekeeping work? And how effective are they? In 2007, a good example played out in the Minneapolis courtroom of U.S. District Judge Michael Davis. His task was to decide upon the admissibility under Daubert standards of the opinions of experts called to testify in the Baycol product liability litigation, a five-year-old dispute involving allegations that the cholesterol-lowering drug Baycol caused a damaging muscle disease called rhabdomyloysis in more than 1,000 people. During his 13 years on the federal bench, Judge Davis had previously made a half-dozen Daubert decisions. “You review the expert findings, and the adverse findings, and then take a look at the peer review on the subject,” Davis says of the procedure. “It’s a difficult process, but that’s part of being a judge.”
Judge Davis believes that he and his peers on the bench are able to handle the scientific and technical demands of controlling the gate for expert witnesses. “Being generalists, we have a lot of items come before us that we’re not experts on,” he says. “We do the background reading given to us by attorneys, which includes medical journals, medical studies, treatises, medical dictionaries and whatever else the lawyers make sure we have to make our determination. We do the best job we possibly can.”
In this case, Judge Davis’ determination resulted in an 89-page opinion that reviewed the acceptability of testimony from 13 expert witnesses. On challenges from the defendants, Bayer Corporation and Bayer AG, he excluded all or part of the testimony of 12 of them. Judge Davis admitted the testimony of one expert witness was challenged by the plaintiffs. “We clearly have to have some gatekeeping in the courtroom,” says Bucky Zimmerman, a counsel for the plaintiffs. “The problem is that issues presented in court are often way ahead of science, because science takes years to come to a decision in which reasonable scientists using consistent methodology will agree.”
Martyn Smith, Ph.D., a professor of toxicology in the School of Public Health at the University of California, Berkeley, part of whose testimony Judge Davis excluded from the Baycol litigation, does not believe that jurors’ access to good science was helped by Judge Davis’ decision or by most Daubert rulings in general. “Clearly there are people who put out fraudulent theories and crazy stuff in the courtroom, but that happens on both sides,” Smith says. “Whether a judge is a good gatekeeper for that, I don’t think so. Most judges I know got into the law because they didn’t want to do science.”
Judge Davis disagreed with the scientific reliability of Smith’s opinions on the toxicity of Baycol and the significance of animal studies. He based part of that decision on the contrary research of K. Sandy Pang, a professor of pharmacology and toxicology at the University of Toronto. “Judges weigh the evidence before them, and who am I to challenge them on whether they have the right to form an opinion?” asks Pang, whose Baycol testimony was her first experience with an American court. “They do the best they can to make an informed decision. How can jurors in the system form an opinion on who is right or wrong? I don’t feel I can criticize that process.”
But Smith, whose testimony in another lawsuit was excluded in a previous Daubert decision in New York, believes that the process can prevent legitimate science from appearing in the federal courts. “For someone at Berkeley or Harvard or some place like that, it’s kind of crazy for judges to say they know more about science than you do,” he says. “I don’t think anyone should act as a gatekeeper. Frauds are always found out, and one side can mount a challenging argument. This is what we do in science all the time—people come in with crazy theories; sometimes they’re right, sometimes they’re wrong. They are tested, and we go forward. To ask judges to make a decision about exclusion is unfair—they should be there to maintain an impartial presence in court.”
This is an infrequently discussed conflict between science and the law. Science accepts the introduction of novel ideas, as long as there is plenty of time to test them. Over the course of years, those notions will thrive or wither. But our legal system demands more immediate determinations of right and wrong. It cannot wait decades for the slow and costly peer-reviewed advances dictated by the scientific method.
Most of the science excluded nowadays is more untested than trashy. In addition to Martyn Smith, the list of experts excluded in Judge Michael Davis’ Daubert ruling included highly regarded academics and researchers whose opinions could be valuable to jurors despite their nonconformity with the Daubert standards. Perhaps the 21st century and its absorption of technology into every kind of legal case from dog bites to drug injuries demands another examination of the role of speculative science. “What does junk science mean?” asks Zimmerman. “Junk bonds are risky, but that doesn’t mean we shouldn’t own them. I don’t want us to connect junk science with worthless science. It is undoubtedly a harder hurdle to get a case before a jury now than it was before Daubert.” L&P
Back to Articles
