Twenty male alcoholic volunteers were randomly assigned to an emetic aversion conditioning group (n -7), a shock aversion conditioning group (n = 7), or a control group (n = 6) receiving no experimental treatment. All subjects received standard inpatient alcoholism treatment in addition to any aversion therapy. All subjects participated in psychophysiological assessment sessions and taste-test sessions before and after conditioning. Dependent measures across attitudinal, behavioral, and psychophysiological response systems all suggested that only emetic subjects acquired aversions to alcohol. Results are discussed with respect to their theoretical and clinical significance.Alcohol aversion therapy is one of the oldest behavior therapy techniques, the first reports appearing in the literature in the early 1940s (e.g., Thimann, 1943;Voegtlin, 1940). The unconditioned stimulus (UCS) in these early studies was, with only one exception (Kantorovich, 1929), an emetic. Although early investigators such as Voegtlin (1940) viewed the pairing of alcohol with an emetic as Pavlovian conditioning, such a conditioning procedure would now be labeled taste aversion conditioning (Garcia, Hankins, & Rusiniak, 1974;Seligman, 1970).Early clinical reports of the use of emetic aversion therapy for alcoholics were usually quite positive (e.g.