This paper traces the evolution of expert scientific witness qualification standards in U.S. federal courts, beginning with the 1923 Frye decision's general acceptance standard through the landmark 1993 Daubert ruling and its four-part reliability test, to the 1999 Carmichael decision that finally reconciled the competing frameworks. Special attention is given to how each standard affected the admissibility of forensic fire investigation testimony, where practical field experience and sensory observation play central roles that no single recognized scientific discipline fully encompasses. The paper argues that the post-Frye/Daubert evolution ultimately benefits fire investigators by reducing unjust exclusions of relevant evidence while simultaneously demanding higher standards of methodology and record-keeping.
The paper demonstrates effective use of case-law synthesis: rather than treating Frye, Daubert, and Carmichael as isolated rulings, it traces a single doctrinal thread — the foundational qualification of expert witnesses — across 76 years of jurisprudence, showing how each decision responded to the limitations of its predecessor. This technique is essential in legal and forensic writing where understanding precedent relationships is more important than analyzing any single decision in isolation.
The paper opens with the Frye standard (1923), establishing the baseline. It then introduces Daubert (1993) as an expansion of that baseline, followed by a section applying both standards critically to the specific field of fire investigation. The conclusion covers Carmichael (1999) and its dual practical implications for fire investigators. This four-part structure moves from general legal doctrine to specific applied consequences, a classic funnel pattern well-suited to forensic and legal analysis papers.
The 1923 U.S. Supreme Court Frye decision established the criteria used by courts to determine the foundational qualification of proffered scientific expert witnesses in federal cases. Specifically, Frye introduced the general acceptance standard necessary to qualify the particular field of science supporting the intended testimony of expert witnesses (Cleary 2007). According to the Frye criterion, in order for expert witness testimony to be admissible, the underlying field of science corresponding to the intended testimony must be "generally accepted" within the scientific community.
The Frye standard was widely adopted in state courts and persisted for 70 years, resulting in the exclusion of polygraph evidence and voice stress analysis (VSA) data. The significance of consensus in the scientific community remained the most widely accepted standard for the qualification of expert witnesses until 1993, when the U.S. Supreme Court introduced a four-part test in deciding Daubert, finally resolving the tension between the Frye criteria and the wording of Federal Rule 702, enacted by Congress in 1975 (Burnette 2008).
Whereas Frye required that the field of science involved be "generally accepted" in the scientific community, Daubert (and subsequent cases) expanded the possible foundation for the qualification of expert witnesses to include practical and vocational "experience" without necessitating any specific generally accepted field of science (Burnette 2008). In principle, Daubert significantly improved the prospect of introducing forensic fire investigators' testimony at trial, precisely because of the role that practical experience and intuition play in arson investigations, in which no two cases are ever identical.
The Daubert criteria consisted of two levels of inquiry. The first emphasized (1) the relevance and (2) the reliability of proposed expert testimony. The second level of scrutiny pertained to four criteria for determining reliability (Friedman 2005). Under Daubert, the reliability of potential scientific testimony presented by expert witnesses was determined by: (1) the objective, controlled testability of the principle; (2) its prior publication in peer-reviewed publications; (3) an ascertainable error rate; and (4) the degree of scientific acceptance within the community of the scientific discipline involved (McCormick 2007).
You’re 39% through this paper. Sign up to read the remaining 2 sections.
Sign Up Now — Instant Access Already a member? Log inAlways verify citation format against your institution’s current style guide requirements.