Abstract
We are writing in support of Jay Greenberg’s thesis, advanced in his paper “Therapeutic Action and the Analyst’s Responsibility” (JAPA 63/1), that analysts are unable to determine whether their interventions are “correct” or effective. Over the years, many analysts have grappled with this problem. The historical lack of rebuttal substantiates his view and underlines the importance of assessing the fundamental implications of this position for psychoanalysis.
Greenberg’s thesis is implied in the statement “I am not assuming that one way of understanding the dynamics of this exchange is more ‘correct’ than another; nor do I have any information that would lead me to believe that one intervention is more analytically or therapeutically effective . . .” (p. 25). He reinforces his assertion by quoting Rachel Blass’s discussion, in her commentary on the paper, of Civitarese’s treatment of a patient; there she suggests that while we “cannot provide certain proof regarding the truth of a theory, what [discussion] can provide is a deeper and more precise understanding of one’s own theory and those of others” (p. 60; emphasis added). “At the risk of repeating myself,” Greenberg responds, “this is precisely my point” (p. 95).
Greenberg’s assertion that our inability to judge whether an intervention can be judged either “correct” or effective has important implications, including the fundamental, complex questions of whether psychoanalysis is a science, and whether we lack the capacity to evaluate analytic competence. We will substantiate his position by referring to earlier analysts who have made the same assertion.
Rubovitz-Seitz (1998) notes that the problem of the reliability of interpretation did not surface clearly until Glover (1952) recognized that there is “no effective control of conclusions based on interpretation, [and this fact] is the Achilles heel of psychoanalytic research” (p. 405). Rapaport, too, asserted that “there is [as of now] no established canon [in psychoanalysis] for the interpretation of clinical observations” (1960, p. 113). Responding to that concern, a group of psychoanalysts in Chicago, including T. M. French, W. C. Lewis, J. G. Kopecs, G. H. Pollock, F. P. Robbins, L. B. Shapiro, R. M. Whitman, and the aforementioned Rubovitz-Seitz undertook a systematic investigation of that problem (Seitz 1966); they reported that despite working together for over three years, and employing various amounts and kinds of clinical data, they were never able to reach satisfactory agreement on the blind interpretation of the same case material. Other investigators who have documented the reliability problem include Sklansky et al. (1966), Weber, Elinson, and Moss (1966), Thomä et al. (1976), Fisher and Greenberg (1977), Werman (1979), Runyon (1981), Spence (1982), DeWitt et al. (1983), Peterfreund (1983), Rosenbaum and Muroff (1984), Fosshage and Loew (1987), and Bernardi (1989). Aron (1999) sounded a similar note, declaring that it is not possible to determine whether a given interpretation or intervention is “correct,” because other analysts and supervisors will propose a different interpretation or intervention.
Thus, this substantial body of earlier analytic opinion supports the validity of Greenberg’s thesis. It has fundamental implications for the nature of psychoanalysis and of psychoanalytic treatment, implications in need of exploration.