In response to:

The Patient Is Always Right from the December 20, 1984 issue

To the Editors:

It is a pleasure to read Janet Malcolm’s lucid and engaging prose reviewing two volumes on “Analysis of Transference” by Merton Gill [NYR, December 20, 1984].

However, as a member of the profession involved, I wonder why of all the books on psychoanalysis published, important and trivial, wise and foolish, profound and simplistic, pretentious and modest, original and derivative, insightful and pedestrian, that particular two volume set should have been chosen for review. While it is a serious attempt to deal with an important issue it is not trailblazing, revolutionary or especially noteworthy, or addressed to the laity. It is a technical work and meant for the profession alone.

I should not have troubled to comment about the review if it were not for the fact that Ms. Malcolm reprints transcripts of fragments of what purport to be analytic sessions. A non-analytic reader encountering these transcripts could be forgiven for regarding them as silly, and the analysts who participated in them as stupid, petty and pretentious. These are caricatures of analysis, presented as the real thing.

Ms. Malcolm asks the crucial question: “How representative of the profession at large are the analysts in Gill’s book?” If she had checked with any of the really competent analysts whom she knows personally, she would have heard that these excerpts are absurd examples of a procedure that is analysis in name only.

Transference interpretation is indeed central to analytic technique, but it is not all of analytic technique. Moreover the analysis of the transference is not usefully pursued by hectoring the patient with petty arguments. One does not offer the patient an interpretation that he will be unable to accept. The transference interpretation will be more acceptable if it can be presented against a background of other material which is less challenging to the patient, and with which it is clearly congruent. For example, “We have some evidence from the sessions of the last several days that your affection for some of the men in your life is reinforced by the remorse that follows attempts to hurt them. Do you think your friendliness to me today has a similar background?” Such a formulation does not accuse or demean the patient. It shows him or her a general pattern of which the transference is an exemplar, and offers him or her the opportunity to reject the interpretation or modify it. Such rejections or modifications may be merely defensive or incorrect, or, not infrequently, they may be entirely correct.

Analysts’ reluctance to record sessions is based upon their refusal to do so without the patient’s consent, and their anticipation that even if the patient consents, the knowledge that the session will be played back and read by a wider audience exerts a chilling and distorting effect on the procedure. For many years it has been the practice of analysts, for research or educational purposes, to report sessions to colleagues from their notes. I am surprised that Ms. Malcolm, who has read widely and well in psychoanalytic literature, characterizes the usual clinical history given there as “self-congratulatory.” In some cases they are that and transparently false. In most cases the author attempts to render an accurate portrayal of a fragment of an analysis to illustrate a clinical or theoretical point. Authentic material has the ring of correctness, while falsification usually betrays itself.

By choosing this kind of material to review, Ms. Malcolm is able to conclude again that psychoanalysis is what she has previously termed “The Impossible Profession.”

Mortimer Ostow, M.D.

Riverdale, New York

To the Editors:

There are several trenchant observations in Janet Malcolm’s review of Merton Gill’s Analysis of Transference, and Gill and Hoffman’s companion set of case illustrations. One such gem is her crisply worded judgment that “the self-congratulatory clinical histories in the analytic literature cannot be accepted as evidence of anything beyond the writers’ self-regard.” Nevertheless, I believe that Ms. Malcolm has fundamentally misunderstood Gill’s point. She appears to believe that Gill is advocating a friendly, conversational stance as the therapeutic instrument. That is not Gill’s point at all.

Gill’s focus is on the centrality of the transference in psychoanalytic treatment. That focus does not in any way imply that the analyst be “fraternal and conversational” with the patient and thereby to woo patients from their neuroses, although tact and responsiveness are surely non-specific contributions to the effectiveness of any psychotherapy, as empirical data suggest. Rather, Gill’s recommendation is to explore the patient’s resistance to an awareness of the transference. This exploration requires that the analyst attend to how the patient avoids talking about his experience of the analytic situation. In large measure, then, the analytic work consists in alertness to the patient’s subtle and latent references to the transference and thereby to bring into the open the patient’s resistances to confronting those ideas and feelings about the relationship. This process may require interrupting the patient’s associations, clarifying what the patient may have only hinted at, and even extended dialogue, in order to dissect the relative contributions to the transference of analyst and patient. Although these contributions will vary from case to case, Gill notes that the analyst who looks for them with an open mind will see not only the patient’s contribution but is likely to be astonished at how much greater his own role may seem to be than he or she had anticipated. Inasmuch as analytic progress occurs within the arena of the transference, a view few analysts would disagree with, unexamined components of the transference will impede that progress by permitting neurotic behavior to persist, unremarked on and even undetected in the present situation of the treatment.


There are several consequences of Gill’s prescription to explore vigorously the resistances to the awareness of the transference, and Ms. Malcolm focussed on two of them. The first is that Gill explicitly rejects the idea that transference is a delusion. Gill’s view of reality is clearly a relativistic or perspectival one. Whether or not the analyst had actually been late for the appointment with the patient is not at issue for Gill. What is at issue is the meaning of the analyst’s lateness to the patient, and patients usually resist exploring such meanings; they prefer to accept those events at face value. To hold that the analyst was in fact late and that the patient had constructed something very personal out of the lateness is not a “contradiction,” as Ms. Malcolm asserts. Gill’s view accepts the reality of the analyst’s lateness as well as the fact that several (not merely one) interpretations about its meaning to the patient are possible. This may seem paradoxical, but it is not a contradiction.

A second consequence of Gill’s emphasis on the primacy both of the transference and of the patient’s resistance to its awareness, occurs in the presentation of the clinical examples. Those vignettes are meant simply to illustrate the thesis of the monograph; they are not meant to justify the thesis and surely not to prove it. It is wrong of Ms. Malcolm to attribute flaws to the principles rather than to the analysts’ too enthusiastic application of them. Further, Gill and Hoffman are commenting on the performance of a skill, as any mentor would comment on the performance of violin playing or of batting a baseball; performance is never perfect.

Therefore to take the illustrations of Gill’s technical recommendations as a test of either the adequacy or the “depth” of the analytic work is mistaken. Indeed one may effectively argue that the unanalyzed resistances to the awareness of the transference may well have the effect of preventing the patient from confronting the “deeper” conflicts of regressive experiences and the more emotional aspects of the transference. The analyst who tells his patient, as did the one whose analytic work Ms. Malcolm seems to admire, “I take the idea about knocking all the books off the wall as if you wanted to knock my penis off,” is not, as Ms. Malcolm suggests, being a perfectly work-manlike analyst. He is doing what has been called wild analysis, i.e., bypassing the present relationship and the experience of that relationship in favor of formulaic pronouncements concerning psychosexuality. Such interprerative statements actually illustrate an unanalytic, superficial approach that Freud had warned against in his paper On Wild Psychoanalysis.

Yet Ms. Malcolm has made a telling point: We do not know whether Gill’s ideas would increase the effectiveness of psychoanalytic treatment, because empirical scrutiny of the actual psychoanalytic process is scarce, but there are a few exceptions, and these monographs represent one of those exceptions, since, in addition to its theoretical contributions, it proposes a system for recording and investigating the psychoanalytic process.

Philip S. Holzman, Ph.D.

Harvard University

Cambridge, Massachusetts

To the Editors:

In her review of December 20, Janet Malcolm chides Merton Gill for ideas that are “commendably logical, admirably fair-minded, but utterly unpsychoanalytic.” By what criteria and according to whose standards? She admits that there is no archive of recorded cases because so few analysts tape record their sessions, but in some undefined way, she seems to know what is psychoanalytic and what is not. If the analyst is part lawyer, part oracle, and part hairdresser, as she claims, then does orthodoxy stem from an equal mixture of all three, or from some more complicated formula known only to Freud (and perhaps a few privileged insiders)?

So long as argument in the field rests largely on appeals to authority and an adversarial (“I’m right and he’s wrong”) approach, there is little hope for clarity or understanding. To bring the clinical happening out of the closet (or consulting room) and into the light of day is one of the reasons that led Gill and Hoffman to publish and gloss their transcripts. Study of these and other dialogues will help to show when the psychoanalyst is conducting the treatment and when he gives way to the oracle, the lawyer, or the hairdresser. Malcolm’s new mythology may represent an advance over id, ego, and superego, but the facts are probably even more to the point—and likely to be more surprising. At the moment, however, they are in short supply and the resulting vacuum allows every analyst—and every critic—to create the world of treatment in his or her own image. The importance of Gill and Hoffman’s contribution is that it gives us a chance to see inside the house of psychoanalysis and begin exchanging hearsay for knowledge and mythology for a floor plan.


Donald P. Spence

Rutgers Medical School

Piscataway, New Jersey

Janet Malcolm replies:

Mortimer Ostow rebukes me for not consulting “any of the really competent analysts whom [I know] personally” on the question of how representative the analysts in Gill and Hoffman’s book are. I suppose it was my skepticism that kept me from such consultation. Although I like to think that the analysts I know are “really competent” (or even competent), I will never be sure unless they reveal their work, as the analysts in Gill and Hoffman’s book have done.

Some of Dr. Ostow’s outrage about the Gill and Hoffman sessions may derive from his unfamiliarity with transcripts of tape-recordings. It takes a while to get one’s ear in, to not be put off by the disjointedness, banality, and triviality of the dialogue—to, as it were, make the imaginative leap back from the stuttering text of the transcript to the illusory narrativity of spoken language. If Dr. Ostow had heard the tapes instead of reading the transcripts, he might have found the sessions less off-putting. In any case, his contemptuous dismissal of the Gill and Hoffman sessions would have more weight if it had been accompanied by a transcript of a session that he considered characteristic of good analysis.

I think that analysts overestimate the difficulty of recording sessions on tape; some of them seem almost phobic about it. Surely one day analysts will look back on today’s debate over whether to tape record or not as they now look back on the controversy that polarized the profession in the late Forties and early Fifties over whether to change from linen head-napkins to paper ones. (See the Brickman Report, 1975.)

Philip Holzman’s enthusiastic account of Gill’s recommendations on technique provides a useful and helpful coda to my unenthusiastic one. But I must question Professor Holzman’s notion that by “wild” analysis Freud meant analysis that “bypasses the present relationship and the experience of that relationship in favor of formulaic pronouncements concerning psychosexuality.” By “wild” analysis Freud simply meant analysis that was clumsily and tactlessly conducted, and that failed to properly prepare the patient for the inevitable psychosexual interpretations to come. In the ” ‘Wild’ Analysis” paper, Freud warned analysts not against an “unanalytic, superficial approach” but against being too deeply analytic too soon. “Psychoanalytic intervention…absolutely requires a fairly long period of contact with the patient,” Freud wrote, and went on, “Attempts to ‘rush’ him at first consultation, by brusquely telling him the secrets which have been discovered by the physician, are technically objectionable.”

Freud put this even better in his paper “On Beginning the Treatment,” where he condemns “any line of behavior which would lead us to give the patient a translation of his symptoms as soon as we have guessed it ourselves, or would even lead us to regard it as a special triumph to fling these ‘solutions’ in his face at the first interview.” He goes on to say, “What self-complacency and thoughtlessness must be possessed by anyone who can, on the shortest acquaintance, inform a stranger who is entirely ignorant of all the tenets of psycho-analysis, that he is attached to his mother by incestuous ties, that he harbours secret wishes for the death of his wife whom he appears to love, that he conceals an intention to betray his superiors, and so on.”

Although I hardly “admire” the analyst who brusquely told his patient that she wanted to unman him, I did not—and do not—consider him a “wild” analyst, to a large extent because of the sense I received from the session that there had been a long, ongoing discussion between this patient and analyst about her “penis envy,” and that his startling comment “I take the idea about knocking all the books off the wall as if you wanted to knock my penis off” was not nearly as startling to the patient as it is to the reader. As an analysis develops, the participants forge a kind of personal language between them (if this language contains psychoanalytic jargon, these days it is more apt to be a contribution of the savvy patient than of the laconic analyst); one of the drawbacks (of which Gill and Hoffman are well aware) of presenting isolated sessions is that they give an imprecise and often misleading idea of the development of this language. However, it occurs to me—against my own case—that since Gill and Hoffman probably heard the earlier sessions of the analysis under discussion, their harsh judgment of the analyst is probably justified, and my more lenient view is not.

I am sorry that Donald Spence found my review authoritarian and adversarial. It was an attempt (inevitably unsuccessful) to solve some of the koans of psychoanalytic therapy, and to pay tribute to Dr. Gill and Dr. Hoffman’s ideas by arguing with them.

This Issue

February 28, 1985