The phenomenological experience of metaphor in psychotherapy

This study was designed to investigate the nature of phenomenological experience at the time metaphors are produced in psychotherapy. It attempted to develop a methodology which could compare internal experience with externally observable behavior thereby creating a bridge between the internal and external worlds of the subject. It also attempted to design a method whereby the effect of an intervention could be studied. In order to accomplish these goals a phenomenological inquiry was developed and used in conjunction with observations of the psychotherapy process to create a non-reductionist approach to the study of human experience.

The subjects were seven experienced and respected psychotherapists. The methodology consisted of obtaining tape recordings of psychotherapy sessions from each of the seven psychotherapists. When a psychotherapist submitted a tape recording which met the sample criterion (having two novel metaphors, two frozen metaphors and two examples of literal language produced by the therapist) he was interviewed about his internal experience during each of the language samples. In addition to language ratings, raters determined whether there was an emotional expression and whether the language was first introduced by the therapist or the patient. When both patient and therapist were congruent on these variables, these processes were labelled affective and linguistic concordance respectively.

The study found few relationships between language categories rated by the Barlow et al. (1971) rating method and phenomenological experience as reported by therapists. It did find discrepancies between data presented by therapists and raters, particularly regarding the classification of language. These discrepancies showed that neither the therapist nor rater alone could accurately determine what had occurred during the session. As a result of these findings, it was concluded that research which does not include phenomenological inquiry and observational data is frequently inaccurate, lacking important data which could contribute to reliability, validity and finally, clinical relevance. Concordance measures were considered important to studying the psychotherapy process and Winnicott's (1971) concept of collaborative "play." Finally, the study was considered to be a preliminary step towards establishing a bridge between the internal and external worlds of the subject, in the present case, the therapist.