Why a Conviction Should Not Be Based on a Single Piece of Evidence: A Proposal for Reform
52 Pages Posted: 1 Aug 2010
Date Written: 2007
This article illustrates a serious flaw in the conventional legal approach enabling a conviction based solely on one piece of evidence. This flaw derives from a cognitive illusion referred to as “the fallacy of the transposed conditional.” People might assume a low error rate in evidence only leads to a small percentage of wrongful convictions. We show that, counterintuitively, even a very low error rate might lead to a wrongful conviction in most cases where the conviction is based on a single piece of evidence. Case law has indicated some awareness of this fallacy, primarily when considering the random match probability for DNA evidence. However, there is almost no awareness of the significance of this fallacy in assessing other types of evidence not considered probabilistic or of the significance of laboratory errors in DNA testing. We show that mistakes do happen with all key types of evidence: fingerprints, DNA, confessions, and eyewitness testimony. We then demonstrate the tremendous impact that even a small probability of error has on our confidence in a conviction. In the end, we propose legislative reform that would make it impossible to convict someone on the basis of any single piece of evidence linking him to a criminal offense.
Suggested Citation: Suggested Citation