There is apparently little difference in the effectiveness of different kinds of psychotherapy. One explanation is that common factors cut across them. The major schools do not much attend to common factors, however, and it may be that outcome equivalence is due as much to common neglect of common factors as to their use. Five common factors are discussed: the therapeutic relationship, expectations, confronting problems, mastery, and attribution of outcome. Linear conceptions of causality seem to contribute to the problem of selective neglect of common factors, and it is suggested that reciprocal interaction may constitute a more viable way of understanding therapeutic processes. Although technical eclecticism and therapy integration have been proposed as potentially integrating the common factors, eclecticism lacks theoretical coherence, and viable integrations have yet to be offered, although statistical methods for testing them (causal modeling) may exist. To suggest what a productive integration could look like, a clinical example is connected metaphorically to a developmental model of attachment. Strategies for further research are also suggested.