IN SPITE OF the recently acquired status of alcoholism as a medically recognized disease, its cause remains unknown and its course and prognosis are uncomfortably variable. It is generally assumed, however, that there is such a thing as an alcoholism syndrome, a cluster of traits, conditions, or behaviors which are widely, though perhaps imperfectly, shared by the bulk of alcoholics. Although there are many definitions of alcoholism, those with the widest currency, such as that of the World Health Organization,1 overlap to a considerable extent; some of their salient features include (1) unusual alcohol-directed behavior (such as excessive, frequent consumption), (2) interference with smooth social and economic functioning, (3) impairment of satisfactory psychological functioning, and (4) deterioration of physical health. As a preliminary investigation in the clinical portion of our alcoholism research program, it seemed important to test the underlying implication that these traits form a cluster of