The growing number of wrongful convictions exposed over the past two-and-a-half decades, and the research that points to a few recurring types of flawed evidence in those cases, raise questions about the effectiveness of the rules of evidence and the constitutional admissibility standards that are designed to guard against unreliable evidence. Drawing on emerging empirical data, this Article concludes that the system can and should be adjusted to do a better job of guarding against undue reliance on flawed evidence. The Article first considers the role of reliability screening as a constitutional concern. The wrongful convictions data identify what might be called _suspect evidentiary categoriesÓ _ a few types of evidence (eyewitness identifications, confessions, forensic science, and snitch testimony) that are both recurring features of wrongful convictions and not otherwise susceptible to correction through traditional trial mechanisms and that, therefore, can and should be subjected to heightened scrutiny for reliability under the Due Process Clause.
Recognizing, however, that the Supreme Court is moving away from using constitutional doctrine to screen for reliability, this Article considers other mechanisms for better ensuring reliable evidence and accurate trial outcomes. First, current trends in Supreme Court jurisprudence suggest a due process framework that focuses upstream of the trial process on regulating the police and prosecutorial conduct that generates some of the most suspect trial evidence. Second, the Article assesses new applications of non-constitutional evidence law that offer promise for filling the void in reliability review of such suspect types of evidence. Finally, the Article considers remedies in addition to exclusion that might aid in the enterprise of mitigating the harm from flawed evidence. Principal among these are broader use of expert witnesses and jury instructions to educate fact finders about the counter-intuitive but scientifically established qualities of these categories of suspect evidence. And because courts have proven reluctant to apply reliability-based exclusionary rules rigorously, the Article concludes by exploring partial exclusion _ excluding the most objectionable parts of the evidence while permitting other parts _ as a remedy that courts might be more likely to actually enforce.