Scientific Uncertainty and Causation in Tort Law
Geistfeld, Mark, Vanderbilt Law Review
Tort cases involving scientific uncertainty frequently present courts with a difficult causation issue. In the paradigmatic case, the available scientific evidence indicates that a substance might be hazardous, but does not establish that the substance is hazardous.1
When presented with such evidence, courts must decide whether the plaintiff has adequately proven that her injury was tortiously caused by the substance.
This causal issue potentially arises whenever we do not fully understand how a substance interacts with the body and produces an adverse health outcome. We do not, for example, adequately understand the etiology of cancer.2 To assess whether a substance may cause injuries with unknown etiology, we observe health outcomes in populations of animals exposed to large amounts of the substance, study the biochemical effects of the substance on cells, organs, and embryos, and compare the substance's chemical composition to other known health hazards.3 Though informative, these studies usually cannot determine whether the substance is hazardous. That determination typically requires a large-scale study comparing the incidence of adverse health outcomes in groups of exposed and non-exposed individuals, or comparing the incidence of exposure across injured and healthy groups. These epidemiological studies are expensive, time-consuming, and require that a large number of people be exposed to the substance. Without such study, however, the hazardous properties of the substance often cannot be established with existing scientific methods. Consequently, substances are often introduced into the environment before there is conclusive scientific evidence regarding their health hazards. How should this scientific uncertainty affect the tort rights of an individual who has been exposed to such a substance and has the type of injury, such as cancer, that is plausibly attributable to the substance in light of the available scientific evidence?
After the United States Supreme Court's decision in Daubert v. Merrell Dow Pharmaceuticals, Inc.,4 an increasing number of courts have held that causation must be established by epidemiological evidence showing that a population of individuals exposed to the substance faced at least twice the risk of suffering the injury in question.5 According to these courts, such epidemiological studies are the only reliable evidence showing that the substance more likely than not caused the plaintiffs injury.6 Numerous commentators criticize this evidentiary requirement, arguing that it is inconsistent with fundamental tort principles, particularly when applied to substances that have not been subjected to epidemiological study.7 According to these critics and others, the lack of conclusive scientific evidence, and the unfairness of placing the burden of factual uncertainty on plaintiffs, require the adoption of special rules, such as placing the burden on a defendant manufacturer to prove that its product is not hazardous.8
The vast majority of potentially hazardous substances have not been subjected to epidemiological study,9 creating an evidentiary gap of potential concern to the tort system.10 In some important contexts involving evidentiary gaps, application of ordinary rules would undermine tort norms, so the tort system has adopted special rules to redress the inequity.ll An important question, then, is whether cases of scientific uncertainty involve the type of problem that justifies a special evidentiary rule for establishing causation.
As Part I shows, the critics of the epidemiological evidentiary requirement have not justified a special rule for cases of scientific uncertainty. Part II then explains why there is no need to adopt new evidentiary rules in the products liability context. In the paradigmatic case involving cancer, for example, tort liability can be imposed on product sellers for not adequately warning about the scientific evidence suggesting that the product is carcinogenic. …