Salzman's commentary initially examines two theoretical considerations of the genesis of substance abuse. He elaborates on the hypothesis that changes in the family precede abuse and relates it to conceptual similarities with research on schizophrenic families. He points out that the proposed research that focuses on ongoing processes rather than causality also fails to identify factors leading to the selection of symptomatology. The author concludes with comments on his work in the area of psychoactive drugs and group functioning and suggests that the inclusion of the following methodologies in research design might be instructive: (a) the combining of self, group, and experimenter observations; (b) the comparison of family interactional patterns when the drug-abusing member is under chemical influence and when he is not; and (c) informal observations and discussions focused on the family's perception of the research as well as on task performance.Research that focuses on families with a substance abuser usually asks one of two questions. First, how does the presence of substance abuse in a family member generally affect the family functioningthat is, how does the family adapt to the substance abuse. The alternate question is, what is there about the family that has allowed substance abuse to get started and flourish. In either case, the general proposition is that the family process changed in relationship to the abuse, either as a partial cause of it or as a reaction to it. Let us examine each of these theoretical considerations.The first hypothesis is usually that drug abuse started in one member of the family and there was a change in the family process that is identifiable and researchable. It can be traced to the onset of the substance abuse in the member. The change in family process, which is the focus of the research, is the result of the drug abuse, the response of the family to the change of one member. Not infrequently, the change involves the extrusion of the abuser. Although this view of family response to substance abuse is not the favored family research focus, it may actually be one of the more frequent patterns of ghetto family involvement. According to several studies (2, 3, 4), inner city substance abuse follows the public health model of an epidemic and is due to a multiplicity of causes, mainly outside the family. In fact, since family membership may be more fluid in some ghetto-dwelling families, the family role may be rather small in the initiation and facilitation of ongoing abuse.By far the more common line of family research investigates the alternate hypothesis that drug abuse is in some way a necessary part of altered family process. Changes within the family to some extent precede the abuse, and even cause it. This is the general view of research that is summarized by the authors and represents their own view as well. They state, for example "We conceptualize the individual's substance abuse as intimately involved with the level of functioning of the family system." They also go on...