In Roundup case, the science will go on trial first
Judge’s approach is latest try to resolve ongoing legal debate
In a San Francisco courtroom this month, a jury will be asked to weigh a complicated question: Did Roundup weedkiller cause a man’s cancer?
The jurors will assess the credibility of competing studies that delve into cell mutations, cancer epidemiology and genotoxicity. They’ll hear evidence purporting to show why California resident Edwin Hardeman’s exposure to Roundup was dangerous, and other analyses arguing it was perfectly safe.
But unlike in a prior trial brought against the herbicide’s maker, Bayer AG, the jurors won’t simultaneously hear allegations that the company hid dangers about its product from the public. Instead, they’ll take part in an unusual split trial focused first on the science, and then, only if they find the plaintiff’s claims valid, on the question of negligence.
A federal judge approved a request by the company for this slimmed-down tri- al, over the objection of the plaintiff’s lawyers, to let the jury evaluate the alleged dangers of Roundup without what he called the significant distraction of attacks on the company’s behavior.
The approach is the latest attempt by courts to resolve a long-running debate over how to ensure the fairest decisions in cases concerning complicated science. The results could influence hundreds of similar Roundup cases—and guide judges in other cases that hinge on science.
Courts have grappled for decades to find the right balance of what scientific evidence jurors can see. On one hand, the threshold for admissible scientific evidence needs to be low enough to encourage consumers to bring legitimate grievances against makers of potentially harmful products. On the other hand, it needs to be high enough to prevent floods of frivolous lawsuits based on faulty science or bogus claims.
Judges have typically played a central role in the debate by screening expert witnesses at the outset of trial. But the
ROUNDUP continued on B4
level of expertise of the judges themselves varies, as does the leeway afforded them to decide what gets in or not.
At the federal level, judges follow a uniform standard for how to allow in expert witnesses and scientific evidence. States, however, have diverged in their approaches. New Jersey last year tightened expert-witness standards to help block what corporate defendants often deem “junk science.” Florida’s highest court, meanwhile, in October reversed a legislative proposal that would have aligned the state with the generally more stringent federal threshold.
Further complicating the issue: Science is often unsettled within the medical community, and the root causes of a disease are often unknown.
“If the science is not junk, but at a stage of knowledge that’s truly indeterminate, what do we do about that?” said Alexander Lemann, a professor at Marquette University Law School. “It’s a problem that’s not solved by having the judge be a more aggressive gatekeeper.”
In the Roundup case, around 9,300 home gardeners, landscapers and agricultural workers have sued Bayer in the U.S., claiming the weedkiller causes non-Hodgkin lymphoma and other cancers. Bayer inherited the potential liability through its acquisition last summer of seed and pesticide maker Monsanto Co.
The U.S. Environmental Protection Agency has said for decades that glyphosate, the primary chemical in Roundup, is not likely to be carcinogenic to humans. In 2015, the International Agency for Research on Cancer, a World Health Organization branch, designated glyphosate as “probably carcinogenic to humans.” That kicked off a wave of litigation and regulatory scrutiny around the world.
Bayer calls the IARC designation an outlier that contradicts more than 800 studies submitted to the EPA and other regulators. The company contends glyphosate is an essential farming tool that has been deemed safe for more than 40 years.
In August, the first Roundup cancer trial resulted in a $289.2 million verdict for a former school groundskeeper who blamed his non-Hodgkin lymphoma on repeated exposure to two Bayer weedkillers. A California state court judge in October cut the verdict to $78.5 million, which the company is appealing.
Over 17 days of trial, jurors learned the three prongs of science that the plaintiff’s lawyers say proved their case: animal studies, epidemiological studies of human populations, and studies showing how glyphosate affects cellular processes.
Day after day, each scientific point came with a counterpoint.
The plaintiff’s lawyers discussed a 2010 study of 20 mice painted with a glyphosate formulation that caused 40% to develop tumors. Bayer lawyers called the study flawed and said research on rodents isn’t directly applicable to humans.
Bayer showed a study tracking farm workers over time that concluded there was no associ- ation between non-Hodgkin lymphoma and glyphosate exposure. A plaintiff’s expert said the study suffered from measurement errors.
Each side says it wins on the science, which will be tested at the trial starting Feb. 25 before U.S. District Judge Vince Chhabria in San Francisco. If jurors find Roundup caused Mr. Hardeman’s non-Hodgkin lymphoma, a second phase will question whether Bayer acted with negligence.
Judge Chhabria has been vetting experts for the case using the Daubert standard, named for a 1993 U.S. Supreme Court case. The standard, employed in all federal courts and many states, requires judges to consider the soundness of scientific methods employed by experts as well as whether the expert is qualified and the evidence is relevant. Daubert replaced the Frye standard, still used in some states, which looks at whether the scientific evidence has gained general acceptance in its field.
The Daubert method assumes judges are capable of understanding scientific methods. But sometimes even the smartest judges are out of their league, said David Faigman, the dean of University of California Hastings College of the Law.
He and other academics have proposed the idea of litigants hiring neutral experts to present opinions to the jury or sending expert reports out for peer review. Mr. Faigman even launched the peer-review concept as a company. But outside a few scattered cases, such alternatives haven’t gained traction.
Mr. Faigman said judges often say they’re persuaded by a plaintiff ’s report explaining why a product is dangerous— only to change their thinking after seeing the defense’s report.
“And they have no ability to distinguish who’s correct,” he said, adding that as a result, “judges punt, and turn it over to a jury to decide.”
Kevin Burke, a Minnesota state court judge and past president of the American Judges Association, said he believes reforms are needed to help juries better understand science but dismisses critics who say judges and juries are so ill-equipped “we need to throw the whole system out.”
Gary Kitahata, a juror in last summer’s Roundup trial, said he “determined our role was not to deal with the nitty-gritty of the science” but rather to do a “broader-scale analysis.” Equally important, the 65-yearold municipal adviser said, were emails purporting to show Monsanto’s resistance to conducting more scientific studies, information that may not be included in the science-only phase of the trial.
Monsanto said the record demonstrates the products have been extensively studied and that the company “has been a leader in supporting research regarding their safety.”
Debates surrounding scientific evidence grip courts around the country. New Jersey had long been known as a favorable place to bring product-liability lawsuits against the state’s many pharmaceutical companies. Last year, the state’s highest court tightened standards, directing judges to consider both the methodology used by experts and the underlying data.
The ruling came in long-running litigation claiming acne drug Accutane causes Crohn’s disease. The court effectively ended the case by concluding that the plaintiffs’ experts “deviated from core scientific principles and strayed from their own claimed methodology in order to reach their conclusions.”
Other cases have fizzled after science didn’t hold up in court, including claims that birthcontrol device Mirena caused hypertension, that morningsickness pill Bendectin led to birth defects, and litigation aiming to tie silicone breast implants to autoimmune diseases.
Other product-liability cases have led to big payouts when the science backing plaintiffs’ claims is affirmed, like against painkiller Vioxx, cigarettes and asbestos products.
In Judge Chhabria’s courtroom last month, the plaintiff’s attorneys argued the science can’t be viewed in isolation from Monsanto’s conduct and that splitting the trial is unheard-of in major product-liability cases.
Attorneys for Bayer and Mr. Hardeman later tussled over what to include in the first phase of the trial. Would internal emails allegedly showing Monsanto influencing a scientist’s research count? Should the opinions of regulators stay out? Judge Chhabria conceded parsing out just the science isn’t clean-cut, saying: “It’s going to be difficult to draw the line on some of these questions.”