The concept of automatism as a consequence of severe alcoholic intoxication received much public and media attention as a result of the decision of the Supreme Court of Canada in the case of Daviault v. R. and decisions in several other courts subsequently. As a result of public indignation over the decision in Daviault v. R., the Minister of Justice introduced legislation to the House of Commons to eliminate intoxicated automatism as a possible defense in such crimes as assault and sexual assault. This legislation eventually came before the Standing Committee on Justice and Legal Affairs, which was required to decide whether or not to recommend this proposed change in the Criminal Code. In its deliberations, the committee was required to make a number of explicit or implicit decisions about whether or not such automatism really exists, whether it should or should not be available as a defense, and, if so, under what circumstances it should be available. The present paper was originally prepared with the hope of facilitating the committee's decisions by setting out as clearly as possible the scientific evidence concerning automatism and how this evidence impinges upon the legal and social issues. For this purpose, the paper reviews briefly the concept of automatism as it appeared to be understood in the courts at that time, how the term is used in medicine and in psychiatry, and the evidence concerning the possible or actual occurrence of automatism as a phenomenon associated with intoxication by alcohol and other drugs. Automatism as understood by the courts It is beyond the scope of this paper to review in detail the use of the term automatism throughout the legal and judicial system and the legal literature. However, to point out the problems for which scientific evidence is needed, it may be informative to examine the apparent meaning of the term as used in four recent and much-publicized judgments. 1. In Daviault v. R. (Supreme Court of Canada), the majority decision referred to extreme intoxication akin to automatism or insanity. The testimony of an expert witness used the French term amnesie-automatisme, which the Court interpreted as blackout. The accused was presumed to have a blood alcohol level (BAL) of 400-600 mg per 100 ml. 2. In R. v. Sullivan (Court of Appeal for British Columbia), the accused's BAL was estimated as 160-200 mg per 100 ml. One expert testified that the accused was probably in a of automatism due to both organic and psychological causes, and thus was unable to form criminal intent. A second expert witness was of the view that the accused was in a dissociated state at the time of the offense and therefore unlikely to have had any intent to commit the crime. 3. In R. v. Blair (Alberta Court of Queen's Bench), the expert witnesses stated that automatism was directly related to the height of the BAL, being virtually 100 certain at a BAL of 400 mg per 100 ml and 75 probable at a BAL of 200 mg per 100 ml. In this case, one expert estimated the accused's BAL as 550-750 mg per 100 ml. 4. In R. v. Dinh-tri Nguyen (Ontario Court [Provincial Division]), testimony was presented in the trial concerning the fact that the accused was a known crack addict and that he was in a highly abnormal and mentally deranged at the time of the offense. However, no expert evidence was brought before the court with respect to the actions of cocaine and its mental effects on the user. The judge concluded that he could not admit the defense of automatism because he had no evidence on this phenomenon and no understanding of its precise nature. These four cases raise a number of questions that should be resolvable by the consensus of scientific evidence. The fourth case makes clear that the courts are in fact dependent upon the expert testimony offered during trials, and that there is no clear legal concept of automatism that is distinct from, or independent of, the scientific (or allegedly scientific) concepts advanced by the expert witnesses. …