Causation, Statistical Evidence, and Toxic Torts

Abstract

In this paper, I present a puzzle about how courts react to purely statistical evidence and my own tentative approach to solving it. The basic puzzle is that while there are a number of contexts in which statistical evidence is rejected as a foundation for liability, there are others such as toxic torts in which such evidence is thought to be sufficient. While a number of attempts have been made to explain why statistical evidence is unacceptable in a variety of contexts, significantly fewer have brought to light cases in which such evidence is commonplace. Through an examination of toxic torts, I show that it is untenable to claim that as a general matter, courts will not ground findings of liability in statistical evidence. I then put forward a more nuanced view, according to which findings of liability can be justified when grounded in direct—but not indirect—statistical evidence.