A Conversation with Lester Luborsky

GA: I see the latest in the series of beautiful Basic Books is out. How did you get the idea for developing an improved method for understanding countertransference?

LL: I'm glad you started there, because the history of the idea is not well known, and for a long time it seemed like the search for the Holy Grail. Like the rest of us, I have been relying on the concept of transference clinically for many years. Ever since Freud put forward the concept 100 years ago, our whole field has relied on the concept for making formulations of transference and then using them for interpretations. Why, then, should I have spent time trying to improve the concept by making an operational measure of it? Why try to fix it if it works?

The answer is that the concept doesn't work in one practical sense. Even expert clinicians provide formulations that are difficult to reconcile with those of other expert clinicians. Philip Seitz first began to show this in 1966. His conclusion came after he had spent several years with his research group of analysts at the Chicago Psychoanalytic Society; the analysts could not agree on such concepts even when judging the same sessions. Actually this conclusion, whether it was for or against agreement, would have been unwarranted because the Chicago group did not use a method capable of proving agreement or disagreement. The same finding on unreliability of clinical formulations was reported by Kathy De Witt and her colleagues (1983) in San Francisco.

GA: How did you get started?

LL: I got into this research many years ago armed only with the hope that greater agreement could be attained when analysts were provided with guidelines to supplement and focus their clinical skill in conceptualizing transference. In the 1950s, at the Menninger Foundation with Robert Wallerstein and Lewis Robbins, I applied a system for describing the transference pattern of each patient in the Menninger Psychotherapy Research Project. We thought that the system worked pretty well, but we were not as aware, as we are now, of the difficulties of achieving consensus. For about four years in the early 1970s, I worked on the same goal with the Analytic Research Group at the Institute of the Pennsylvania Hospital with a group that included Homer Curtis, Sydney Pulver, Henry Bachrach, Perry Ottenberg, Richard Lower, and Philip Escoll. The experience in our group was the same as that at Menningers. We did not succeed in fashioning a guided system for clinical judgment of the transference pattern. It was only in 1975 that I began to succeed by means of a method that I should have tried at the start—retracing my own clinical operations in making a general formulation of the central relationship pattern in a series of sessions. In the 1980s, Paul Crits-Christoph and I together did many of the essential reliability and validity studies of that measure.

GA: Would you describe the formulation system.

LL: The steps in the formulation system grew out of catching and observing the mental processes I used in making a formulation. In collecting these self-observations, I noticed first that I was attending most keenly to the patient's narrative accounts of interactions with the therapist and with other people. Second, I was particularly attentive to the most recurrent aspects of these narratives of interactions. I came to realize that I was paying most attention to three categories within these interactions: what the patient wanted from the other people, how the other people reacted, and how the patient reacted to their reactions. After trying these and other categories, I settled on these three as the most routinely evident and serviceable for inferring the transference pattern.

After a couple of years of practice in identifying these categories, with my colleagues doing the same, it became clear that the Core Conflictual Relationship Theme (CCRT) method was ready to be born. It also became clearer as time went on and we did more studies that the CCRT looks much like Freud's 1912 “transference template.” So the CCRT provides clinicians with guidelines that are similar to those that experienced clinicians use in the usual inference processes in formulating transference patterns.

GA: Nothing is as persuasive as a concrete clinical example of the application. Would you give us one.

LL: I just happen to have one ready. It's one of the two I presented before the Seattle Psychoanalytic Society. One was an analytic case, and the other a psychoanalytic psychotherapy case. This is the psychotherapy case.

He is Mr. Howard, an 18-year-old in his first year of college whose main complaints were anxiety states, pain in his penis, guilt and difficulty in being assertive. He started the third session by saying, “I'm very anxious now. I want to tell you more about me.” At this point the patient started recalling past and present interactions with other people. As a sample of how the CCRT works, I'll tell you the first three of the six that he told in this session (these are highly condensed):

Narrative #1 was about mother (when he was about age 12): “I asked mother a question about sex. Mother said, ‘You're too old for that now,’ and my sister laughed. I was upset.”

Narrative #2 was also about mother (when he was about age 10): “I remember my mother stopped letting me get in bed with my parents at night when it was cold. It was very upsetting to me.”

Narrative #3 was about the therapist: “I'm now very tense and I'm getting a headache and suddenly I'm unable to say anything to you. There are moments when guys get together when they have nothing to say to each other; there is no such thing as complete rapport.”

I've gone over these narratives with many clinicians, who typically infer that the patient wanted physical and psychological closeness with his mother (and with the therapist) and that when this closeness was abruptly cut off, the symptoms of anxiety, self-devaluation, difficulty in being assertive, and distance made their appearance. That formulation essentially is the CCRT. It has in it his wish to be close to his mother and not be cut off from closeness; the response from the other person, which is to deny closeness; and the responses of the self, which are anxiety states, self-devaluation, difficulty in being assertive, and distance. In other words, the CCRT follows the clinical method, but it does so by guiding the clinician to be sure to include the wishes, responses from others, and responses of self.

By getting the clinician to infer according to these guidelines, we obtain a reliable measure of transference. When these guidelines are not followed and when clinicians follow their own style, it is more difficult to determine how much clinicians can agree with each other.

GA: Tell us more about how the CCRT can be used clinically.

LL: Understanding Transference contains a chapter on the clinical uses of the CCRT. To sum it up: In the everyday clinical practice of psychoanalysis and psychotherapy, clinicians need to have a convenient supplementary set of guides to clinical inference about the transference pattern. This idea can be put more strongly in terms of the opportunity the guidelines provide for a disciplined approach that helps in getting a grasp of the central conflicts of the patient, and it also has great potential in teaching and research.

The method in the book meets these needs. When the transference potential becomes more activated, as it did at the time of Mr. Howard's third narrative, the therapist is prepared to interpret one level of the transference. At that point the therapist said, “You feel that you will not get from me what you wish, and so you are feeling more anxious now and you have nothing to say.” This interpretation clearly was informed by knowledge of the recurrent components within the narratives of the interactions with others and with the self. These recurrent components make up the CCRT; the CCRT, in turn, is used as a basis for interpretations. The interpretations can be considered as accurate to the degree that they converge with the formulation of the transference, as in the operational definition adopted by Paul Crits-Christoph and our group in one of the chapters of the book.

There are other clinical uses of training in the CCRT. As illustrated in this example, (1) the therapist was helped in the timing of the interpretations because the CCRT formulation was already apparent to him from the first two narratives. When the transference became activated in the third narrative (the one about the therapist), he was ready to provide the interpretation. It is also clear from this example that (2) the CCRT guidelines helped in locating the conflictual context of the symptom. The patient's main symptoms, including the anxiety, were activated in the third narrative. The context for the increase in the anxiety was the arousal of his wish to be close to the therapist, but the wish was followed by the expectation that it would be frustrated. The wish and the negative expectations of responses precipitated the anxiety symptoms on this and other occasions.

GA: Do you find that practicing analysts have difficulty understanding and applying the CCRT method in their work?

LL: There has been considerable interest in the CCRT method, and some analysts in the United States and elsewhere have begun to apply it. But some have found difficulties in applying the method in practice.

GA:

GA: That the CCRT based on manifest dreams is congruent with the CCRT using waking narratives (as shown in Carol Popp's chapter in the new book on transference) may imply that the CCRT partly reflects conscious material.

LL: It does, but as you say, only in part. As I have pointed out, it has conscious, as well as unconscious, components. The material of a session, including the dreams, helps to identify the unconscious levels of meaning of the CCRT, as the example of Mr. Howard illustrates.

GA: How does the CCRT method articulate with the work of others doing research into the psychoanalytic process? Do the different methods reinforce each other?

LL: Yes, ever since the CCRT method was started in 1976 there has been a huge increase in research on the topic of developing transference measures. There are now 15 of them, including the CCRT. One of them is well known to you, the Gill and Hoffman (1982) measure called “The Patient's Experience of the Relationship with the Therapist.” There are now half a dozen studies of single cases that show some commonality across measures, as described in the Luborsky and Crits-Christoph book.

GA: How would you sum up the importance of the CCRT work for our field?

LL: One of the main discoveries that came with its use is that analysts and

other clinicians can agree on their transference formulations, but only when they follow certain common guidelines for making inferences about the transference pattern. When they follow their own inclinations about the level of abstraction and the components of their transference formulations, however, it is difficult to tell if they agree or not.

A second major benefit from the CCRT work is that with the new method it is possible to examine the evidence for each of Freud's many observations about his transference concept. One of the contributions of the book is that I reviewed all that Freud wrote on transference, and I brought together his 22 observations in one place (in chapter 19). One by one each of Freud's observations is compared with evidence from the CCRT. So far we have been able to make a translation of his observations into CCRT terms for 17 of the 22 observations.

For nine of these observations we have already carried out studies that showed a good correspondence between Freud's observations and the CCRT evidence. There is now enough evidence to show that the CCRT and the transference template are best thought of as being on the same conceptual level and as having much in common. More precisely, a clinician's transference formulation is the clinician's estimate of transference; a clinician's formulation in CCRT terms is a version of the concept.