Banishing Ipse Dixit: The impact of Kumho Tire on forensic identification science

Washington and Lee Law Review, Summer 2000 by Saks, Michael J

I. Introduction

to the forensic identification sciences. Moreover, some forensic identification scientists looked for ways to evade Daubert scrutiny. The solution to this problem for those judges and for pseudo-scientists was to re-classify those fields as non-science.5 But Kumho Tire Co. v. Carmichael6 patched that hole, so to say, at least by its apparent terms.7 That is to say, hauling down the science flag and hosting the non-science flag does not exempt expert evidence from Daubert scrutiny.8

Together, Daubert and Kumho Tire do a remarkably clear job of commanding judges to property scrutinize fields, presumably including the forensic identification sciences, before admitting opinions from those fields' practitioners. But one can never underestimate the ingenuity of judges in finding ways to evade rules that tell them to do something that would lead to a result contrary to the one suggested by their intuitions. The post-Daubert, pre-Kumho Tire period was telling: Obeying the letter and spirit of Daubert would lead to significant exclusion of a type of evidence that the courts welcomed for most of the twentieth century. On the other hand, a ruling to admit these fields would be both a rejection of conventional science as the criterion for admission of empirical claims and a ruling in the teeth of repeated unanimous Supreme Court opinions declaring the conventional scientific method to be the touchstone for evaluating empirical claims of all kinds.

and to exclude expert opinions "connected to existing data only by the ipse dixit of the expert"?9

II. Defining the Fields

Let us be clear about the fields that we are discussing. There are two kinds of forensic science: On the one hand, there are normal applications of basic science. On the other hand, there is individualization science, or identification science. My focus will be on the latter, though it will not be my exclusive focus. Normal forensic science does things like determining what substance something is (e.g., what is that white powder?) or measuring the quantity of something (e.g., how much alcohol is in the murder victim's blood?). Forensic individualization sciences aim to connect a crime scene object or mark to the one and only source of that object or mark to the exclusion of all others in the world.

Examples of the forensic identification sciences include handwriting identification, fingerprints, firearms, toolmarks, bite marks, hair and fiber identification, tiras, footprints, and so on.10 This is a remarkable claim, especially considering how weak the theoretical and empirical bases of it are. Yet it is accepted widely in our culture as true. Question: How do the practitioners of these "sciences," or the public that has so long accepted the claims of expertise, know the claims to be true? The answer is ipse dixit (or ipse dixit's close cousin, "experience"). Once one appreciates the weakness of the bases of forensic identification science, one can better understand why the casualness of judges in admitting these fields creates a serious problem and how Kumho Tire, if obeyed by the lower courts, could bring about a revolution in the courts and in forensic science itself.11

III. The Scientific Status of These Fields

the techniques; instead, police investigators who sometimes were engaged in little more than a parody of science invented them.12 Other forensic sciences, what we might call the "normal forensic sciences" (e.g., forensic toxicology and forensic chemistry), borrow and apply principle from normal basic sciences such as physics, chemistry, and biology. Those applications have the benefit of basic research on which to build.

In seeking to estalish pinpoint linkages between crime scene evidence and known exemplars, the forensic identification sciences seek to accomplish something that no other field attempt to accomplish and about which no other field has developed any basic scientific knowledge. The forensic identification sciences have no basic science to undergird them. For most of their history, the forensic identification sciences had little or no academic or industrial infrastructure to provide them with knowledge, resources, or personnel. Instead, they invented themselves, and they exist on their own. They are an enterprise consisting of nearly all application and no science.

There is no systematic, rigorous, empirical research on which the forensic identification sciences' knowledge is built. If called upon to prove their claims, they have little or no data to marshal in their support. Instead, the forensic identification sciences point to a guild of mutually self-reassuring examiners who have come to believe in the truth of their claims, often sounding more like a faith-based religion than a data-based science.13

Forensic identification science examinations are overwhelmingly subjective affairs. Armed with no usable models and no base rate data, they must rely on impressions, subjective probability estimations, and intuition (termed "judgment" or "experience").15 Fingerprint identification experts have the advantage of large organized databases containing sets of previously collected and entered fingerprints. Their decisions, however, remain highly subjective.16 Indeed, the field requires experts to be doubly subjective: Not only must they reach a subjective judgment about the likelihood of a coincidental match, but they may not testify to an identification unless they believe that every other fingerprint expert's subjective judgment would render the same conclusion. Thus, fingerprint examiners must draw subjective impressions about other people's subjective impressions.