The Best Insurance against Miscarriages of Justice Caused by Junk Science: An Admissibility Test That Is Scientifically and Legally Sound

Abstract

This article is a contribution to a forthcoming symposium on wrongful convictions. In some cases, a wrongful conviction is virtually unavoidable. The empirical data demonstrating the invalidity of the expert's technique or theory may become available only after the conviction. However, in other cases, the courts can minimize the risk of a wrongful conviction by applying a rigorous admissibility test. Part I of this article considers -- and then rejects -- two possible admissibility tests. One is an extreme version of the relevancy test allowing the admission of testimony based on a technique or theory so long as one qualified expert vouches for the technique or theory. The second is the traditional FRYE general acceptance test. Both tests provide inadequate assurance against wrongful conviction because they demand nothing more than ipse dixit -- either that of an individual expert or the collective ipse dixit of a substantial majority of the specialists in a field. Both approaches represent the antithesis of scientific methodology. Part II of the article proposes a refined version of the DAUBERT reliability test. After stating the proposed test, the article dissects the test and explains each test component. Part II argues that the proposed test reflects sound scientific methodology and represents a synthesis of the governing statutes and cases. Finally, Part III explores both the utility of the proposed test and its limitations. As Part III establishes, the satisfaction of the proposed test guarantees neither the admissibility of the expert's testimony nor its legal sufficiency to support a conviction. Most importantly, Part III underscores that the use of a sound admissibility test does not preclude the possibility that later scientific research will discredit the technique or theory relied on at the time of the earlier trial. Vigilance against wrongful convictions must be a both/and proposition: We must not only apply an exacting test to assess the technique or theory at the time of the initial proffer of the evidence, but we also have to revise the postconviction relief statutes to correct miscarriages of justices that are exposed only by later scientific investigation.