In attempting to determine if Hazelwood had consumed alcohol, the authors applied speech analysis techniques to five samples of his voice that had been culled from audio tapes provided by the National Transportation Safety Board at various times before and after the accident. All taped communications contained the words "Exxon Valdez" and this phrase was the focus of quantitative analysis. Hazelwood's articulation showed he was having difficulty with the fine motor control needed to produce certain sounds. It took Hazelwood about 50 percent longer to say "Exxon Valdez" at the time of the accident than the day before. In addition, Hazelwood's voice pitch was dramatically lower in the samples recorded around the time of the accident. Because lowered pitch and increased variability are characteristic of alcohol-impaired speech, this evidence suggested that Hazelwood had been consuming alcohol at the time his ship ran aground. Although acoustic-phonetic differences in the speech samples of Hazelwood are consistent with the findings of controlled studies on the effects of alcohol on speech, they do not prove that Hazelwood was drunk. Other explanations for changed speech patterns must also be evaluated and discounted, such as the physical state of the speaker and the mechanical condition of recordings. Following a discussion of legal standards for admitting novel evidence, including the Frye test and relevancy tests, the authors conclude that acoustic analysis should be admissible under the emerging relevancy test for scientific evidence. 116 footnotes and 6 tables