To imagine a language is to imagine a form oflife.
Wittgenstein-Philosophical InvestigationsMuch can be commended in the excellent papers by Duncan (1992) and Held (1992). For example, we strongly support Duncan's suggestion that relationship variables such as empathy, warmth, and acceptance are central to therapy. Held holds out promise for commonalities, clarity, and organization to assist us through the chaotic diversity and the atmosphere of crisis that have beset our field as the systems era disintegrates.There can be little doubt, as Duncan (1992) implies, that the most overlooked and yet most consistently found variables relating to therapy outcome are often lumped together under the rubric oftherapist relationship variables. Patterson (1984) suggests that relationship variables such as warmth and genuineness are more related to what happens in therapy than all other variables combined. The role of interpretation, suggestion, reeducation, instruction, specific interventions, and other therapist-induced intrusions pale, according to Patterson, when compared to the role and importance of relationship variables. Patterson (1984) also suggests (as Duncan implies) that somehow the relationship in therapy extends to the broader interpersonal context and that this is what makes for system change outside the therapy context. The specific definition, application, and role of these important issues and variables, however, are quite unclear in the therapy literature.Held (1992) reminds us that the philosophical implications of our theories are central to their clinical application. Her attempts to find common characteristics across the broad spectrum of family therapies rest heavily on her use of the concepts of realism and antirealism. It appears to us that the convergence in practice posited by Held may have more to do with the futility of therapy than it does with any other single issue. We would agree with her apparent conclusion. The distinction and import of this realist/ antirealist dimension collapses in the enactment ofthe dramas oflife and psychotherapy. It truly makes little practical difference in terms of what happens. As Duncan (1992) also implies, it is the import of the meaning that counts in therapy not the truth or realism ofthe meaning. Held (1992) reduces the importance of therapeutic and relationship variables to this realist/antirealist dimension. In accomplishing this reduction, she finds congruence and commonality across disparate positions.However, more importantly for us, her definition of antirealism is close to that commonly accepted by family therapists as radical constructivism and second-order cybernetics. This constructivist position is widely held in the family therapy community. It has serious limitations and constraints. It is a cognitive view that describes meaning as existing only in the head of a single person. It is a theoretical stance that seems very close to the earlier perspective of symbolic interaction (Mead, 1934). The symbolic interactionist perspective considers perso...