Psychotherapy 2012, Vol. 49, No. 3, 347–348
© 2012 American Psychological Association 0033-3204/12/$12.00 DOI: 10.1037/a0029115
AUTHORS’ RESPONSE
Parallel Process: When Does a Parallel Occurrence Carry Meaning? Terence J. G. Tracey, Cynthia E. Glidden-Tracey, and Jamie Bludworth Arizona State University Watkins (“Some thoughts about parallel process and psychotherapy supervision: When is a parallel just a parallel?” Psychotherapy, 2012, pp. 344 –346) has commented on our paper on the parallel process in supervision. He takes exception to our interpersonal definition of parallel process and implies that it is not capturing real parallel process. We argue for the theoretical and methodological merits of our definition as well as the validity support demonstrated. Keywords: parallel process, supervision, interpersonal behavior, interpersonal circumplex
the psychodynamic literature as still espousing both views. In addition, models beyond the psychodynamic that incorporate parallel processes and adopt a more totalistic view, include the developmental models (e.g., Loganbill, Hardy, & Delworth, 1982; Stoltenberg & Delworth, 1987) and the interactional/interpersonal models (e.g., (Kell & Mueller, 1966; Kiesler, 1992; Mueller & Kell, 1972). We argued in our paper that the endorsement of parallel process was very common across theoretical lines but that there is little definitional consensus. Part of the problem lies in the term “parallel process” which is concise, but far from precise. The vagueness of the terminology seems especially problematic when comparing “parallel processes” with “process parallels,” as Watkins does to make his point that what we studied may indeed be co-occurring events, but not necessarily relevant to the most important tasks of therapy supervision. His question, from his title, is whether some observed and presumed “parallel processes” in supervision triads are simply “naturally occurring process parallels” (p. 345). He further indicates that it is crucial to be able to distinguish these more trivial parallels from those “particular types of parallel process . . . that we most wish to explore in the therapy/supervision interface.” We would place the emphasis on identifying parallels that carry significant meaning in supervision and aid in understanding on why and how to use those identified parallel processes to promote effective supervision and psychotherapy. Given this lack of definitional consensus across the supervision literature, and our endorsement of interactional/interpersonal theories, we used these, especially the predictable dynamics of dominance/submission and affiliation/hostility, as the basis of our definition of parallel process. So our focus may not match the one favored by Watkins emphasizing unconscious processing, but that does not disqualify the viability of our interpersonal emphasis. At issue for Watkins is our focus on parallel processes that include behavior that may not be unconsciously determined. Many parallels between the client–therapist and trainee–supervisor interactions can occur simply due to the similarity of roles. The client comes to the therapist for help and the trainee comes to the supervisor for help. While the help processes involved in each are different, the hierarchical nature of the roles is similar, and this can
We had two reactions to Watkins’ (this issue, pp. 344 –346) comments on our paper “Are there parallel processes in psychotherapy supervision? An empirical examination.” (Tracey, Bludworth, & Glidden-Tracey, this issue, pp. 330 –343). First, and more visceral, was “Gee, someone actually read it.” One of the difficulties of conducting research is that once a manuscript is published, there is little feedback of any sort. So it is common to think that the manuscript fell into a black hole. Consequently, we are gratified that indeed at least one person read and took time to comment on our work. After the gratitude of the first reaction wore off, a second reaction arose, questioning the first. Watkins lauded our methodology but raised issues with our definitions of parallel process in supervision. He presents his comment as complementary to our article. We appreciate the compliment regarding our study’s method, but are less sure about the extent to which Watkins’ comments complement our conceptual and operational definitions. Watkins (this issue) questions our usage of what he seems to consider a totalistic definition of parallel process. As he notes, the definition of parallel process started from the psychoanalytic literature and focused on the unconscious determinants of behavior and how these are carried into supervision. As such, parallel process has been conceptually tied to transference and countertransference. There are various views of transference, counter transference, and parallel process in psychoanalytic and psychodynamic models. One view focuses only on those behaviors that arise from unconscious motivations, whereas another view includes those definitions (the totalistic view) that focus on all behavior and reactions by both participants. Although the reviewer implies that this is a dead debate and that the forces against the totalistic view have prevailed, at least in psychodynamic circles, this is not decided. While it is beyond the scope of this commentary, we view
Terence J. G. Tracey, Cynthia E. Glidden-Tracey, and Jamie Bludworth, Counseling and Counseling Psychology, Arizona State University. Correspondence concerning this article should be addressed to Terence J. G. Tracey, Arizona State University, 446 Payne Hall, mc-0811, Tempe, AZ 85287-0811. E-mail:
[email protected] 347
348
TRACEY, GLIDDEN-TRACEY, AND BLUDWORTH
lead to some parallel processes. Watkins notes that this focus on total definition can thus include some simple aspects of this parallelism. As such, he implies that we are not really studying parallel process at all but more simple parallels of interaction. As evidence, he offers some quotes differentiating simple parallelism from more substantive parallel process. All of these quotes derive from the unconscious determinants model of parallel process, casting doubt on other possible interpretations. Clearly, we and Watkins have different ideas about what parallels of process are worth most attention in supervision. However, Watkins does not provide any structure or guidance explaining how to use and validate the unconscious determinants definition of parallel process to permit better differentiation of substantive parallel processes from simple parallelism. The issues then are not unconscious versus totalistic interpretations of parallel process, but are instead (1) what theoretically derived definitions of parallel process differentiate meaningful processes from those that are more coincidental? and (2) does any one proposed definition do a better job of accounting for data variation than proposed alternatives? Watkins states that a definition of parallel process “where focus, clarity, consistency, operationality, and specificity are our guiding definitional desiderata” (p. 345) is needed. We agree whole-heartedly. We believe our definition meets these criteria. It stems from a strong theoretical base, that is, interpersonal models, and it has explicit clear, consistent, operational, and specific definitions of observable behavior and prediction of outcomes of interactions. Moreover, our definition is, to date, the only one that does meet these criteria. The merits of any other definition remain to be demonstrated. Beyond presenting no alternative model or definition, Watkins fails to perceive what was found in the study. The study found clear parallel process with respect to both dominance and affiliation. As noted earlier, one might expect some parallelism given the hierarchical nature of the interaction. The client asks the therapist for help as the therapist asks the supervisor. However, what was found were patterns of interaction for each supervision session segment that matched the immediately preceding therapy session with the client under discussion more that the patterns exhibited by the same supervisor or by the same supervisee taken randomly across the multiple supervision triads in which each participant was involved. So the hierarchical nature of the interaction (i.e., the relative roles and the basis for Watkins’ simple parallelism) was evident in every randomly drawn comparison, yet the individual paired sessions demonstrated greater parallel process than these representative random drawings. So even when there are common parallelisms and typical personal styles taken into account, a clear pattern of parallel process occurred in terms of interpersonal behaviors marked by dominance/submission and affiliation/ hostility. This can occur only if both participants are altering their behavior in the hypothesized manner. So the parallelism found was much more prominent and interactive than the implied simple parallelisms posited by the reviewer. Further evidence for the value of the parallel process definition we used, however, is given by the pattern of parallel process that was found to be related to positive outcomes in the manner predicted by theory. Were the pattern simply one of trivial parallelisms as posited by Watkins, then there should be no relation to outcome. The relation of the observed and tested parallel process to actual client outcome argues strongly against a claim of trivial
patterns. The distinctive pattern of the therapist bringing the client interaction up into supervision and the supervisor altering this pattern and then the therapist bringing it back down into the therapy process is the crux of parallel process as posited by interpersonal theorists (e.g., Kell & Mueller, 1966). Hence, the definition we adopted also gets good criterion validity support. It is related to client outcome in the manner posited by theory. In conclusion, we appreciate Watkins’ feedback and critique of our research study, and we agree that distinguishing the meaningful and useful aspects of parallel processes in supervision from less significant parallel occurrences is a crucial component of empirical investigation of supervision processes and outcomes. However, we focus on somewhat different interpretations of what makes a triadic supervision pattern meaningful in terms of parallel processes. Clearly, he has one view of parallel process that rests on unconscious determinants, and he argues that our definition does not capture these. Watkins points out the value of sorting out conscious, preconscious, and/or unconscious parallel processes of supervision apart from parallels of process that are less relevant. As we have argued, we adopt an interpersonal view of parallel process. We see our definition as including unconscious motivations but also more. The issue is the merit of this greater inclusion. We have not seen any definitions or operationalization of parallel process (or transference and counter transference) that meet the guidelines Watkins specified. It would be good to see these and study their merits relative to our more interpersonal one. But our study found strong support for the interpersonal view and supported the conclusion that what we studied were not trivial parallels. Quite the contrary, we found strong support for the presence and value of an interpersonal conception of parallel process in supervision. Regardless of our points of disagreement with Watkins, we wish to thank him for noticing and commenting on our work. Such communication about research is too rare.
References Kell, B. L., & Mueller, W. J. (1966). Impact and change: A study of counseling relationships. Englewood Cliffs, NJ: Prentice Hall. Kiesler, D. J. (1983). The 1982 interpersonal circle: A taxonomy for complementarity in human transactions. Psychological Review, 90, 185– 214. doi:10.1037/0033-295X.90.3.185 Kiesler, D. J. (1992). Interpersonal circle inventories: Pantheoretical applications to psychotherapy research and practice. Journal of Psychotherapy Integration, 2, 77–99. Loganbill, C. R., Hardy, E. V., & Delworth, U. (1982). Clinical supervision: A conceptual model. The Counseling Psychologist, 10, 3– 42. doi:10.1177/0011000082101002 Mueller, W. J., & Kell, B. L. (1972). Coping with conflict: Supervising counselors and psychotherapists. New York, NY: Appleton-CenturyCrofts. Stoltenberg, C. D., & Delworth, U. (1987). Developmental supervision: A training model for counselors and psychotherapists. San Francisco, CA: Jossey-Bass. Tracey, T. J. G., Bludworth, J., & Glidden-Tracey, C. E. (2011). Are there parallel processes in psychotherapy supervision? An empirical examination. Psychotherapy, 49, 330 –343. doi:10.1037/a0026246 Watkins, C. E., Jr. (2012). Some thoughts about parallel process and psychotherapy supervision: When is a parallel just a parallel? Psychotherapy, 49, 344 –346.
Received May 16, 2012 Accepted May 21, 2012 䡲
