How can a forensic result be a “decision”? A critical analysis of ongoing reforms of forensic reporting formats for federal examiners

Résumé 0

The decade since the publication of the 2009 National Research Council report on forensic science has seen the increasing use of a new word to describe forensic results. What were once called “facts,” “determinations,” “conclusions,” or “opinions,” are increasingly described as “decisions.” Prior to 2009, however, the term “decision” was rarely used to describe forensic results. Lay audiences, such as lawyers, might be forgiven for perceiving this as a surprising turn. In its plain English meaning, a “decision” would seem to be a strange word choice to describe the outcome of a scientific analysis, given its connotation of choice and preference. In this Article, we trace the recent history of the term “decision” in forensic analysis. We simply and clearly explain the scientific fields of “decision theory” and “decision analysis” and their application to forensic science. We then analyze the Department of Justice (DOJ) Uniform Language for Testimony and Reporting (ULTR) documents that use the term. We argue that these documents fail to articulate coherent frameworks for reporting forensic results. The Article identifies what we perceive to be some key stumbling blocks to developing such frameworks. These include a reluctance to observe decision theory principles, a reluctance to cohere with sound probabilistic principles, and a reluctance to conform to particular logical concepts associated with these theories, such as proper scoring rules. The Article elucidates each of these perceived stumbling blocks and proposes a way to move forward to more defensible reporting frameworks. Finally, we explain what the use of the term “decision” could accomplish for forensic science and what an appropriate deployment of the term would require.