The language of family therapy : what we say we do and what we actually do in therapy

Systemic family therapy, as a variant of the "talking cure", has developed its conceptual base during the second half of the twentieth century. Its founding fathers and mothers made a conceptual break with psychoanalysis, and this theoretical distinction has until recently been well established. Contemporary theorists have shown an interest in narrative metaphors and sought to situate systemic therapy within the terms of postmodernist and specifically social constructionist discourses. By this fact a challenge is presented to the researcher who wishes to subject to scrutiny the theoretical claims made for this form of human activity: how to rigorously evaluate theoretical propositions whilst employing a methodology that is congruent with the assumptive base of family therapy. The present study represents an attempt at taking up this challenge. Family therapy sessions are videotaped, transcribed and subjected to a discursive anlalysis. The method is in tune with social constructionist premises and allows for a meaningful analysis of such contemporary theoretical preoccupations as the therapeutic relationship, power, gender, culture and the injunction to place the self of the therapist within the system. The actual enactment of these theoretical premises is examined and the conditions for the successful accomplishment of discursive, and hence therapeutic, goals is explored. A finding emerges that cannot be adequately accounted for within a post-foundationalist epistemology of socially and culturally-situated talk: consistent individual differences in the positions taken by interactants. In order to explain this finding it has been found necessary to insert an ontology of subjectivity within social constructionist explanatory frameworks. A nonrational, non-unitary version of the individual is constructed that bears more than a passing resemblance to the psychoanalytic subject. Consideration is given to the implications of these findings for future research