A suggested structure for a science of relationships

Abstract An attempt is made to sketch the structure of a science, in the sense of an ordered body of knowledge, of interpersonal relationships. The special properties of relationships are discussed. Relationships exist over time, involve properties beyond those of the constituent interactions, include both behavior and subjective experience, and exist in a social context. It is suggested that a science of human relationships will require three overlapping phases: (1) Description, listing 10 categories of dimensions potentially useful for ordering the data; (2) Principles of dynamics, which may concern external influences on the dyad, interchanges between the participants, or intrapsychic processes; and (3) The specification of the limits of applicability of those principles. In general, a balance must be found between the search for generalisations and recognition of the complexity of human relationships.