Ego Analysis Vs. Self Psychology


Kohut describes his transformation from what he calls a “drive psychologist” to a self psychologist as it happened in his treatment of Miss F. (The Analysis of the Self, 1971, pp. 283-295):

After a prolonged period of ignorance and misunderstanding during which I was inclined to argue with the patient about the correctness of my interpretations and to suspect the presence of stubborn, hidden resistances, I came to the crucial recognition that the patient demanded a specific response…and…completely rejected any other.

The specific response that Kohut decided Miss F. demanded was that he should stop making interpretations and that instead he should “respond empathically to her narcissistic display,” by “approval, mirroring, and echoing” (p. 287).

Now, as you know, Kohut did not mean that he just decided that Miss F. had this demand. He meant that he decided that this demand was genuine, that it did not conceal “the presence of stubborn, hidden resistances,” that she wasn’t bluffing (in the Alexander sense; see his case under “Id Analysis,” in Vignettes), and that he should comply. So he stopped arguing and he stopped making interpretations. He tried to satisfy Miss F.’s demand for an all-accepting alter ego.

This was a remarkable concession for any therapist to make, especially since Miss F.’s accusatory behavior was not at all designed to pull for rescue. Whenever Kohut made an interpretation, even of a very cautious kind, one that was hardly more than a reflection, Miss F. would explode, becoming “violently angry. ” She would “furiously” denounce him, accusing him of “undermining” her, and would cap this with the reproach “that with my remark I had destroyed everything she had built up; and that I was wrecking the analysis” (p. 286).

This was no invitation to empathy. It obviously must have felt to Kohut like “stubborn resistance,” as he said at first it did—as though Miss F. was determined to thwart and defeat him. Clearly, Kohut’s original response must have been to feel tested and to believe that he had to not give in to Miss F.’s flagrant provocations — that he had to call her bluff.

But then in his remarkable turnabout Kohut decided that his patient was right all along, that he had been undermining and wrecking the analysis. He saw that he was acting out the narcissistic countertransference. He hadn’t liked being pushed off the stage. But, more than that, he saw that what made him prone to act out the countertransference was not only his own dynamics but also was the drive model (or drive theory, in Greenberg & Mitchell’s terminology).

Originally, Kohut took the id analytic position that Miss F.’s anger was proof of the correctness of his interpretations. He later (1977, p. 90f) attacked the drive model’s justification of this position, as formulated by Hartmann, that interpretations loosen defenses, deneutralizing the aggressive drive-energy that had been bound up in them, and the patient responds with anger. Kohut opposed Hartmann’s formulation with his own conception that when a patient responds with anger to an interpretation, this means that the interpretation is unempathic.

But in the early stages of Miss F.’s analysis, when Kohut agreed with Hartmann, he had to see Miss F. as lashing out in a needlessly self-protective way. This would have made it hard for him to avoid feeling that she was a stubborn and grossly unappreciative adversary. When he came to think of her as acting out a “genuine” phase-specific developmental need he could then feel much more sympathetic and could indulge it with no felt loss of analytic authority. What Kohut’s self psychology and the other developmental ego psychologies have in common is, essentially, this revised conception of resistance. The message of the ego psychologies is that the patient isn’t always bluffing, and that much of what looks like resistance represents attempts to cope by the vulnerable ego. The recommendation is to suspend interpretation in the presence of resistance, potential or actual, at least as a resolvable parameter, and instead to offer some form of reassurance.


But why give up on interpretations? Can’t we correct for the drive fallacy and can’t we interpret without attacking the vulnerable ego? Must interpretations take the form of calling the patient’s bluff? That’s the question.

Going back to Miss F. , why would Kohut not have continued to make interpretations from his new perspective and when he was no longer at odds with her? We know he had concluded that she would reject any interpretation but how fair a test did he make of this conclusion?

After he resolved his parameter and resumed interpreting with Miss F., Kohut (p. 292) says that first he pointed out “the intensity of her [narcissistic] demands and of her need for their fulfillment. ” He adds that this was an interpretation “which she resisted vigorously because she could no longer deny the presence of an extreme neediness in this area” (emphasis added). “She resisted…because she…could no longer deny” is a piece of analytic shorthand. This is to say that she redoubled her denials because, squirm though she might, she had been caught red-handed, just as in Alexander’s vignette. Kohut had called her bluff.

In combating Miss F.’s vigorous denial of her “infantile grandiosity” hadn’t Kohut gone back to his argumentative ways and his original conception of resistance as confirmation? What of his idea that interpretations that cause shock to the ego are unempathic?

Kohut does not discuss these questions or even acknowledge the issue, but there is no question that he pursued a line of interpretation that caused severe shock to the ego. His describes his “continuing revelation of her persistent infantile grandiosity and narcissism” as “slow, shame provoking, and anxious. ” Unlike Alexander, Kohut called his patient’s bluff gradually, but there is no doubt that, like Alexander, he thought that she had to go through this grinding agony because no one could stand having to recognize such infantile grandiosity.

Thus, when Kohut resumed the use of interpretation he still made interpretations from the side of the id, but where before he had been pointing out unacceptable drives he now was pointing out unacceptable needs.

As by now is familiar, the inevitability of shame in narcissistic patients is a key postulate in Kohut’s model, and he was one of those who assume that shameproneness indicates the lack of a superego (rather than a more warded-off or “nonmetabolized” superego). Thus, Kohut thought that “unneutralized exhibitionism” is inherently painful to the fragmented self and that this “painful narcissistic tension” is experienced as shame.

In other words, we are bound to feel shocked when having to recognize childish exhibitionism. So, as in drive theory, the problem is the inherent unacceptability of the drive itself. Kohut’s modification is the idea that the drive (exhibitionism/narcissism) is OK and even necessary, but that it needs to be “neutralized,” meaning expressed in more appropriate ways.

But, as I argued in my review of The Analysis of the Self (“Analysis Without Guilt” ), this still is classical id analytic thinking, since it misses the role of the superego. In the early theory drives clashed with “reality. ” It was not until near the end of Freud’s career that he recognized that drives clash with the superego (with internal objects or, if you will, negative self-talk). This is hard to see because our superegos are so much alike that we agree with patients that you shouldn’t be narcissistic.

This view dictates a different line of interpretation, in which Miss F.’s agony, rather than taken for granted (though titrated and empathized with), is itself the focus.


So Kohut was not in a position to make a fair test of his conclusion that Miss F. would not tolerate any kind of interpretation. Ego interpretations could only have been formulated if it were granted that Miss F.’s intense shame was evidence of a primitive superego—or superego precursor, if you prefer. Ego analysis might even better be called superego analysis. Although this would clear up some misconceptions, the term would be unfortunate because many patients are thought of as not having a superego, just as Kohut thought the narcissistic patient didn’t, and typically these are just the patients for whom ego (or superego) analysis is most called for, as appears to have been true for Miss F.

In effect, the superego is often on the surface of a problem and there is the paradox that the surface of a problem can be harder to see than the depths. In fact, I have already given you one example from the case of Miss F. that I am sure no one has noticed.

You remember in the statement from Kohut that I used to open this discussion, the one I have repeatedly referred back to. He said, to abridge it a bit: “After a prolonged period of ignorance and misunderstanding. . . I came to the crucial recognition that the patient demanded a specific response” (emphasis added). This is an interesting locution that everyone immediately grasps even though on the face of it it makes no sense.

If Miss F. was demanding a specific response—that Kohut stop interpreting and confine himself to “mirroring and echoing”—why would it be a “crucial recognition” for Kohut to see it, one that he came to only “after a prolonged period of ignorance and misunderstanding?” Either she was making this demand or she wasn’t.

Everyone knows that Kohut means she wasn’t. He was using another piece of analytic shorthand, which in its general form reads: “She really was making this demand.” We know it means she really wasn’t, that she was denying it.

So when you read Kohut’s statement you didn’t think, “Hey, wait a minute! What does he mean she ‘demanded’? If it took him years to see it, how demanding could she have been? ” You didn’t think that because we all are used to this elision; it’s the legacy of id analysis.

We know that Kohut means that he figured out why Miss F. had her tantrums. She got upset when this demand wasn’t met and she calmed down when it was. The fact that she couldn’t “demand” this specific response is treated as a technicality.

Similarly, Kohut treated Miss F.’s defense against making any demands as, to paraphrase Alexander, a resistance that is diminished by the verbalization of what the patient is resisting, that is, in other words, as largely irrelevant. Thus, at this point you probably have the picture of Miss F. as a complaining, demanding, angry, and provocative woman whose stormy outbursts kept Kohut at bay and, as some classical analysists think, may even have caused him to give up the practice of psychoanalysis to become a psychotherapist. But that’s how she was “really” acting, meaning that she wasn’t literally acting that way all that much (cf. how she was “really” demanding).

In actuality Miss F.’s narcissistic demand was so well hidden that Kohut was not only unaware of it “for extended phases of the analysis,” as we know, but had quite an opposite impression. This will surprise you. During this period, Miss F.’s hours had “the appearance of a well-moving self-analysis” (p. 285), broken only by her angry responses when Kohut made an interpretation. Miss F. also was active professionally, had many friends, and “had had several love relationships and some serious suitors”—all by the age of twenty-five.

Miss F.’s narcissism was so well hidden it generated another countertransference symptom in Kohut. In this phase of the analysis he was “struggling with boredom and inattentiveness,” which “was puzzling since the patient dealt with object-directed preoccupations, inside and outside the analytic situation, and present as well as past” (p. 285). Kohut says that he finally realized that what he had been reacting to was Miss F.’s “infantile grandiosity and exhibitionism,” but that it had been “covered for a long time by a display of independence and self-sufficiency” (p. 293, emphasis added).

Miss F.’s independence and self-sufficiency, although clearly impressive, is taken as merely a “cover” and a “display” by Kohut, and he says no more about it. Id analysis is characterized by this treatment of defense as if it is merely the manifest content and this treatment of the warded-off content as if it is what is really going on. For Kohut to dismiss Miss F.’s defense as a cover and a display is to say that she really is narcissistic. It is just like Kohut’s saying that Miss F. is demanding when she can’t be demanding. But if we take her defense seriously we need to say that she can’t be narcissistic, that is, not directly, not wholeheartedly, not with full awareness.

Obviously, from this ego analytic view one would not think that Miss F.’s narcissism needed to be indulged rather than analyzed. Her intense shame about it is, of course, not taken as inevitable. It becomes the focus of analysis. This means she needs interpretations rather than that an ego psychology is needed to protect her from interpretations.


It remains to show how the events Kohut took to be the direct expression of Miss F.’s unneutralized narcissism could be seen as indications that she was struggling with a condemnatory superego. How she was trying to solve the problem of how to get any narcissistic gratification in view of her unconscious need to prove otherwise.

Here I need to bring in one more piece of information about the case. Kohut reported that Miss F. would typically carry the hour like a “well moving self analysis,” but that “at approximately the midpoint of the sessions, she would suddenly get violently angry at me for being silent and would reproach me for not giving her any support” (p. 286).

It was only later in the hours, when Kohut did say something, that “she would again get violently angry. ” (“Violently angry,” by the way, is an instance of id analytic hype, designed to show how deep the feeling is; it is unlikely that Kohut means anything more by this than that Miss F. shouted at him. In other words, she wasn’t violently angry. )

Kohut believed that the abruptness of her reaction (p. 286) and the “high-pitched tone of her voice” (p. 288) revealed “the archaic nature of her need” (p. 286). This was convincing to Kohut, so much so that he considered it to be the most telling diagnostic clue in the case. What makes his diagnosis less convincing to the reader is the fact that he does not consider the more obvious, ego analytic, explanation, even just to dismiss it.

The more obvious, ego analytic, explanation is that when, after Miss F. had been accomplishing her well-moving self-analysis, Kohut was silent—or said anything—she felt slighted. Consider Kohut’s well-known description of the narcissistic patient as someone who “feels he controls and possesses [the analyst] with a self-evident certainty that is akin to the adult’s experience of his control over his own body and mind” (1971, p. 90). In other words, he feels he has the analyst’s rapt attention, the very thing the patient wants (but also can’t want).

This is, of course, a familiar phenomenon and it is easy to imagine that when Miss F. was acting independent and self-sufficient and as though she needed so little from him that he felt bored, she was caught up in a narcissistic fantasy—an experience of his total and undivided attention and approval. Then when he was expected to respond and didn’t, or when he said something that indicated less than total approval, she felt slighted—as if he “had destroyed everything she had built up” (in other words, had destroyed the fantasy). At that point she could accuse Kohut of “undermining her” and even of “not giving her any support” (p. 286). Now she could have a narcissistic claim. She may well have felt cheated of the appreciation she felt she had earned by acting so independent and self-sufficient, by not making any demands.

Here is another one of those key distinctions. Kohut took her intensity to mean the intensity of her need—that it was archaic, meaning traumatically unsatisfied in childhood. Nowadays patients will gratefully agree with such (an object relations) interpretation. It even is puzzling why Kohut didn’t get that kind of response rather than distress. Probably it is a matter of tone and timing, and modern day therapists are much less ambivalent about being openly sympathetic. But, is it not also familiar that when a narcissistic or borderline patient has this kind of tantrum, it also conveys feeling unentitled to the need? After all, such an exaggerated complaint—that by making some inadequate or somewhat disapproving comment, Kohut had destroyed the whole analysis—seems to directly indicate that Miss F. did not feel at all confident of the validity of her complaint.

Grandiose patients may be more ashamed than anyone else of their need for ordinary approval and recognition. Their experience of totally possessing the analyst’s attention is highly precarious and easily and invariably yields to the experience of being totally possessed by the analyst’s need for attention, although they are no better able to tolerate this experience of the analyst’s narcissism.

Interpreting from the side of the ego would mean presenting to Miss F. her difficulty being narcissistic and how she suffers from having to be so independent and self-sufficient. It means showing how, despite all her independence and selfsufficiency, she is always in danger of feeling childish if she has any needs for approval or recognition. This sort of interpretation is familiar enough, of course. The key conceptual operation is to couch the case in terms that make this sort of interpretation relevant.

Interpreting from the side of the ego would also mean showing Miss F. how much trouble she is having organizing her complaints about the analyst and how she really can’t enjoy complaining. This would expose how she has trouble believing that Kohut really is destroying what she has built up and really is wrecking the analysis and, finally, how she has trouble imagining that the analyst has ordinary narcissistic needs, even ordinary shame about them.



Alexander, F. (1935) The problem of psychoanalytic technique.
Psychoanalytic Quarterly. Also in The Scope of Psychoanalysis: Selected Papers of Franz Alexander, Basic Books: NY, 1961.
Alexander, F. (1956) Two forms of regression and their therapeutic implications
Psychoanalytic Quarterly, (25). Also in: Selected Papers.
Fenichel, O. (1941) Problems of Psychoanalytic Technique. Psychoanalytic Quarterly, NY.
Freud, S. (1937) Analysis terminable and interminable. Standard Edition, XXIII, London: Hogarth, 1964.
Kohut, H. (1971) The Analysis of the Self. NY: International Universities Press.
Kohut, H. (1977) The Restoration of the Self. NY: International Universities Press.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s