Although various social groupings--aggregates, dyads, groups, and organizations--have been the subject of much past study, researchers have encountered an absence of a systematic theory to explain behavior within marriage and family units. This paper examines the diversity present in research concerning marriage and family units and suggests that these differences stem from the false assumption that "the family," as commonly perceived, represents a heuristic unit of analysis. Criteria based on observations of behavior in other social groupings can be used to predict the differences in interactions which occur between families. The implications of this paradigm for conflict resolution within kinship groups are also examined. (KS)