Abstract

Abstract Anxiety about the deliberative abilities of ordinary citizens, feared to be too easily influenced by the powers of rhetoric, has accompanied democracy since its birth. This anxiety is reflected in critiques of the American jury system. This article examines efforts in the middle of the twentieth century to rationalize jury decision making through the use of mathematical probability. These efforts—one a trial in which a prosecutor used dubious statistics to help convict a couple of a robbery and the other a call for juries to use formulas for assessing the likelihood of guilt—reflect a desire to simultaneously harness and contain the dangerousness of rhetoric. More significantly, proposals to mathematize jury decision making individualize deliberation and privilege expert over everyday knowledge, signaling a threat not only to this important feature of American democracy but also to the ability of citizens to deliberate collectively in debates increasingly dominated by statistical reasoning. Acknowledgments Many thanks to the editor and the two anonymous reviewers of this article, as well as to Greg Goodale, Bonnie Jefferson, David Kellogg, Chuck Morris, Elizabeth Shea, and Patricia Sullivan for helpfully commenting on drafts of this article at various stages. Thanks to Alyson Wilson and Greg Wilson for conversations about subjective probability; any remaining misconceptions in this article are entirely my own. I am grateful to Hugh Baxter, Ellen Cushman, Cassandra Jackson, and John Schaeffer for conceptual help, and to Greg Clark, Kathleen Kelly, Marina Leslie, Bernadette Longo, John Monberg, Terese Guinsatao Monberg, Jeff Strobel, and Christine Wolff for encouragement. Notes 1Interestingly, although Tribe was a Harvard law professor at the time he challenged Finkelstein and Fairley's proposal, he had been a law clerk on the California Supreme Court when People v. Collins was heard. An undergraduate math major, Tribe had drafted much of the opinion but had left the court months before it was issued. He did not disclose his role in the Collins opinion until 2004. See Fisher. 2As of this writing, Tribe's article has been cited in 84 state and federal cases. See, for example, United States v. Massey, 594 F.2d 676 (1979), which quotes extensively from Tribe in support of reversing the conviction of a man convicted of robbery based on probability data about hair samples. A more recent case, United States v. Veysey, 334 F.3d 600 (2003), contains an extensive discussion of legal commentary, including Tribe's 1971 article, but rules in favor of admitting statistical evidence. 3In the federal court system, jury verdicts must be unanimous in civil trials unless otherwise agreed to by the litigating parties and must always be unanimous in criminal trials. See the Federal Rules of Civil Procedure and the Federal Rules of Criminal Procedure. In all state courts, jury verdicts must be unanimous in criminal trials. The number required in civil trials varies by state. See United States Department of Justice 233.