Although therapeutic thinking and practice have entered their second century, most practitioners remain largely uncertain as to what data to trust, including "what works" and, "why it works." If anything, practitioners' reliance on ever-increasing numbers of theoretical models and either vague or contradictory research findings as primary guidelines to their practice has added to the confusion. What can the practitioner trust? This paper puts forward the "dangerous proposal" that it is of primary value for clinical professionals to place their trust on an often overlooked variable: the clients' statements about their therapy and their relationship to and with their therapist. It is argued that through the clarification of this variable will emerge a better understanding of the nonspecific variables of the therapeutic relationship itself and, as significantly, of the encompassing interrelational realms of discourse evident in all therapeutic models. OUTCOME VARIABLES: PSYCHOTHERAPISTS' ASSUMPTIONS In a recent article by Scott Miller, various significant quandaries regarding psychotherapists' assumptions about significant outcome variables were brought out (1). Among these were the following: 1. The most successful treatments of depression were those that employed psychotherapeutic interventions alone. Indeed, it has been argued that "[m]ore effective therapists have a psychological rather than biological orientation in their treatment approach" (Blatt et al., quoted in 1, p. 40). 2. Combined treatments employing medication and psychotherapy were less effective than those treatments that only employed psychotherapy (Beutler, in [1]). 3. While 96% of studies funded by drug companies showed favorable results for the treatment of various psychological disorders via medication, only 37 % of studies not funded by drug companies showed similar results (1). Miller concluded that "[o]ver the decades, generations of therapists have come to suspect that it isn't so much what they do-what theory, what model, what technique or even what medication-that helps people, but who they are and who their clients are, as well as the idiosyncratic personal fit between themselves and the people who come to see them" (1, p. 40). OUTCOME VARIBLES: CLIENTS' STATEMENTS In line with this conclusion, numerous studies have indicated "that the quality and strength of the therapeutic relationship was the primary determinant of successful outcome across treatments-including medication" (1, p. 41). But what is it about the therapeutic alliance itself that might determine what are the significant variables that will increase the likelihood of positive outcomes? In their important text, Handbook of Psychotherapy Integration, Norcross and Goldfried emphasized various qualitative factors that are common to the major therapies as being far more significant than any specific factors that have been identified with any particular model or theory of psychotherapy (2). Of most interest to this discussion are those studies that focused upon clients' statements concerning the effectiveness of treatment. Clients most frequently attribute successful treatment outcomes to the personal qualities of their therapist. Specifically, the therapist's ability and willingness to understand and accept them emerge as the most significant variables across numerous studies (2). David Howe's invaluable text, On Being a Client, confirms the above conclusion (3). Further, Howe lists three key factors in clients' perspectives regarding what they found to be necessary factors for the success of their therapy. The first, and by far the most widespread, response emphasized the clients' experience of having the opportunity to talk in a structured setting. The fact that clients could sit in a room and talk for a particular period of time was perceived by them as being by far the most valuable and beneficial therapeutic variable. …