Document Type
Book Chapter
Publication Date
10-2026
Abstract
As readers of this volume will be well aware, Federal Rule of Evidence (FRE) 702 and the Supreme Court’s famous 1993 opinion in Daubert v. Merrell Dow Pharmaceuticals require courts to assess the reliability of expert evidence before admitting it to the jury. Most states impose similar rules. These rules are commonly understood to require courts to perform a gatekeeping function to keep unreliable expert evidence away from the jury.
Concerningly, both empirical studies and anecdotal accounts suggest that judges apply this expert gatekeeping function differently to different types of litigants. Judges appear to require more reliability to admit evidence proffered by civil mass tort plaintiffs or criminal defendants and less reliability to admit evidence proffered by prosecutors. Put another way, litigants offering expert evidence do not get “the same thing, at the same time, and in the same place.”
This chapter makes two main contributions to our understanding of judicial bias in the application of the expert evidence rules. The first is diagnostic: While the expert evidence rules appear, superficially, to bind judicial discretion, in practice they do nothing of the kind. Contrary to the prevailing view, I contend that the Daubert standard as embodied in the Federal Rules of Evidence does not actually set a threshold reliability requirement for expert evidence. Instead, it creates a bureaucratic necessity for judges to assess reliability without dictating how much reliability should be required for admissibility. The rule masquerades as a duty that constrains judges by requiring them to keep junk science out of the courtroom. In reality, it does the opposite: it expands judicial discretion by creating a new rationale for exclusion without a mandate either to admit or to exclude in any given circumstance. Discretion, in turn, invites bias.
My second intervention is prescriptive. To reduce judicial bias in the application of expert evidence admissibility rules, I propose that courts should undertake party-blind review. This would require a two-stage process. Initially, parties would submit expert methods for review without revealing the results of applying those methods to case-specific facts. All party identifications would be redacted from these initial submissions and any opposing briefs. The judge would then rule party-blind on the admissibility of the expert method. Only if the judge decides that the method is sufficiently reliable to satisfy FRE 702 would the proffering party reveal its identity and submit the case-specific results for review.
Disciplines
Civil Procedure | Criminal Procedure | Evidence | Law
Recommended Citation
Rebecca Wexler,
Party-Blind Review: A Proposal To Reduce Judicial Bias In Expert Evidence Admissibility Rulings,
Critical Evidence, Bennett Capers, Jasmine E. Harris & Julia Simon-Kerr (Eds.), Cambridge University Press
(2026).
Available at:
https://scholarship.law.columbia.edu/faculty_scholarship/4896
Comments
This material has been published in "Critical Evidence", edited by Bennett Capers, Jasmine E. Harris and Julia Simon-Kerr. This version is free to view and download for private research and study only. Not for re-distribution or re-use. © Cambridge University Press.