by Dr. Robert Saltzman, Ph.D.

While no one can agree on exactly what the self is, as adults we still have very real sense of self that permeates daily social experience. It rises in many forms. There is the sense of a self that is a single, distinct, integrated body; there is the agent of actions, the experiencer of feelings, the maker of intentions, the architect of plans. --Daniel Stern (1985, pp. 5-6)

Kohut The Freudian

In 1938, Freud, a Jew, was forced to flee Vienna for London. As Freud’s train left the station, a young medical student, standing on the platform, waved, and was favored by a tip of the hat from the great man whom he idolized but never had met. That was the first and last time that Heinz Kohut ever saw Freud, but later, as the president of the American Psychoanalytic Association, he delighted in recounting the story, which, according to Wolf (1996), Kohut interpreted as a symbolic passing of the torch.

Two years after Freud’s departure, Kohut left Vienna too, and, at the age of twenty-five, after a brief stay in London, emigrated to the United States, where he began a residency in clinical neurology at the University of Chicago. He was praised by his teachers as “a brilliant clinician, destined to become a star” (Siegel, 1996, p. 2), but Kohut, like Freud before him, turned his attention away from neurology in order to study the workings of the mind. Leaving the University of Chicago, he began work at the Chicago Institute for Psychoanalysis, where he trained in “classical Freudian theory, the prevailing psychoanalytic theory in the United States” (p. 2), and began a practice in the

traditional Freudian manner, modified by the influences of the American ego psychology school of Hartmann, Kris, and Lowenstein. This school was founded in the two basic Freudian principles of (1) transference: a displacement upon the analyst by the analysand of unconscious incestuous longings that play out the unresolved oedipal drama of childhood and (2) resistance: the unconscious process impeding the recognition of these desires. (Rowe & Mac Issac, 1995, p. 7)

In that way of working, as we have seen, emotional difficulties were understood as having their deepest roots in the two drives--the sexual drive that served to preserve the species, and the aggressive drive that served to preserve the individual self--coming into conflict with the prohibitions of the cultural surround, which prohibited free expression of the drives. Thus, the paradigm that Kohut was using in his clinical work was called the “drive-conflict” model, or simply “drive theory.”

In Freudian theory, the primary object for the expression of the sexual instinct is the parent of the opposite sex with whom the child desires an incestuous relationship, and the primary object for the expression of the aggressive instinct is the parent of the same sex, who is seen as a rival to be murdered. This is the famous oedipus complex. Any awareness of the oedipal desires creates profound anxiety, because the desires carry with them a fantasy of reprisals by the rivaled parent. For the male child, incestuous and murderous wishes produce a fantasy of castration by the father; the girl (already believing herself to be castrated, according to Freud) fears reprisal through abandonment by her mother. But since the same instincts that can produce these terrifying anxieties also are the wellsprings of action in the real world--the aggressive drive leading to action through assertiveness, and the reproductive drive leading to action through sexual striving--having to defend against expression of the drives produces inhibitions that interfere with living. As Kohut conducted his practice in the 1950s, his task, seen broadly, was to analyze the source of his patient’s anxieties so that those analysands could be freed of some of their inhibitions.

Further, psychoanalytic work aimed at helping the patient to discriminate between oedipal wishes, and more mature desires that carried with them the possibility of genuine satisfaction through effort in the real world. Freud had called need for such discernment “the reality principle,” and had asserted that

experiences of “optimal frustration” are responsible for the differentiation between a wish and reality. An optimal frustration is the period of delay a child experiences before a particular wish can be satisfied. Through the delay the child comes to realize that active steps must be taken in order to satisfy the wish. According to Kohut [who taught Freudian theory from 1958 until the late 1960s at the Chicago Institute for Psychoanalysis], Freud suggested that it is only through an optimal frustration, a frustration that is neither so intense as to be traumatic nor so minimal as to be insignificant, that wishes can be differentiated from reality. (Siegel, 1996, p. 27)

This point will be important later in looking at Kohut’s self psychology, because the idea that children (or analytic patients) learn through “optimal frustration” is central to his new conception.

The Move Away From Freudian Theory

Kohut was always sensitive to the problem of narcissism, probably because he struggled with his own pronounced narcissistic tendencies (Paul Ornstein, personal communication, August 29, 1997). Wolf, very much in the mode of self psychology, interpreted Kohut’s creativity and its final focus on narcissism as

a compensatory response to some early deprivations that had threatened the cohesiveness of his budding self. One major deprivation was the absence of his father during World War I. . . . The war had been a catastrophic interruption of his career as a concert pianist and he was unable to pursue his musical aims after he returned. One can easily imagine the father's depression and the son's disillusionment in the now returned father, who must have been a distantly admired hero during his military service.

Little Heinz was close to his mother and he remained so for many years. Yet certain remarks that he made at times left me with the impression that his mother was a somewhat distant woman who was overly involved with her social life, leaving Heinz in the care of servants and tutors. (1996, p. 14)

In an early paper, Kohut displayed his interest in the problem of narcissism, connecting its pathological manifestation to Freud’s concept of optimal frustration.

If the child is over-spoiled (not optimally frustrated), it retains an unusual amount of narcissism or omnipotence; and at the same time, because it lacks actual skills, feels inferior. Similarly, overly frustrating experiences . . . lead to retention of narcissistic omnipotence fantasies. (Kohut & Seitz 1960, cited in Siegel, 1996, p. 28)

But the narcissism to which Kohut referred was still, as in classical Freudian theory, a stage of human emotional development during which

libido is invested completely or mainly in the self, or the ego, or, more simply in the body. . . . [It] is considered normal in the very young; should it persist into adulthood it is usually classified as a neurosis and is generally characterized by a love of self that precedes, if not precludes, love of others. (Reber, 1985, p. 462)

For Freud, the normal developmental line--that which would not be neurotic--required that primary narcissism find later expression in “object libido,” first directed toward a homosexual object, then towards a heterosexual object, and finally, in its best development, culminating in abandonment of narcissism in favor of real object love (Siegel, 1996). Observe that this sequence of development implies that narcissism per se is pathological if it continues beyond infancy, that is, narcissism is something to be outgrown. Kohut had been teaching that point of view, and his 1960 paper on optimal frustration still presumed it, but in 1966 he abandoned the Freudian position, postulating that narcissism had a developmental line of its own--in fact two developmental lines--and that narcissistic development was normal, not pathological. Further, pointing to the contempt with which many psychoanalysts regarded “narcissists,” Kohut asserted that positing object love as a higher development than self love was judgmental and demeaning, putting the needs of society at large above those of the individual patient (Siegel, 1996).

Kohut (1966/1978a) hypothesized that the infant’s world was naturally blissful until inevitable failures of maternal care began to threaten that bliss. To defend against the destruction of its peaceful world, the infant created two new systems of “narcissistic perfection.” One, the “idealized parent imago,” attempted to protect the infant’s well-being by endowing an outside object with infinite power and goodness. The second, the “narcissistic self,” fantasized that everything good was contained within the infant, and everything bad was outside. Later, Kohut (1968/1978b) changed terminology: the “narcissistic self” became “the grandiose self.”

Each of the two forms of narcissism, grandiosity and idealization, followed its own course of development. The idealized parental imago--the image of a perfect other with whom one could totally merge, and who would be a source of endless strength, perfect kindness, and unlimited power--inevitably would be subjected to disappointing comparisons to the actual parent. If these were not too sudden or traumatic, that imago would be converted slowly into ideals. The child’s grandiose self--which Siegel (1996) compared to a comic book super-hero, able to accomplish anything without limit--desired to receive from important caregivers witness to its perfection and admiration of its magnificent powers. According to Kohut (1966/1978a), without such appreciative witnessing, without “the gleam in the mother’s eye,” the child’s narcissistic self could not properly mature; with adequate admiration, archaic grandiosity matured into realistic ambitions.

By 1971, with the publication of Analysis of the Self, Kohut was prepared to discuss what would happen if the idealized parental image and the grandiose self did not properly mature. As we have seen, if the child’s idealized parental imago were gradually challenged by a series of disappointing, but tolerable experiences with the real parent, the idealization would be internalized by the child in the form of ideals. For example, the idealized imago of “daddy” as a flawlessly noble, entirely moral, perfectly fair guardian of truth, could mature into a love of justice in the abstract. However, if this line of development were disturbed, perhaps by a disappointment in the actual father that was too sudden and severe to be tolerated, the idealizing narcissism would not mature into ideals, but would remain compellingly active in the adult, who then might be forced to seek persons--rather than abstract ideals--to idealize for all of her or his life. One sees extreme examples of this pathology, for instance, in cult members. If the grandiose self were tolerably (optimally) disappointed--for example, mother sometimes praised the child, but sometimes was too busy to notice the child’s “greatness”--then the child’s needs for admiration could mature into the ambition to gain recognition for achievable accomplishments in the real world. However, if the disappointment were traumatic--perhaps, for instance, mother fell into a depression and never noticed the child--then, according to Kohut, there would be no line of development by which the child could imagine garnering sufficient appreciation through ordinary achievements, and she or he would retain an archaic grandiosity, going through life lacking self-esteem, perhaps feeling acutely sensitive to imagined slights, or perhaps demanding constantly to be a center of attention. In its extreme form, this is the so-called “borderline personality.”

Kohut (1971) conceived the successful maturation of the idealized parental imago as a process of “transmuting internalization.” This meant that the idealizations invested in the caregivers when gradually withdrawn would be absorbed into the psyche of the child where they would form “new structures that assume the psychological functions previously performed by the idealized object, devoid of the personal qualities of the object” (Siegel, 1996, p. 71). Note that Kohut, like Freud, needed to express his psychological understanding in terms of reified structures. Later, sensitive to criticism of this apparent hypostatization, he took pains to distance himself from Freud’s mental apparatus model, and to explain that he saw the structures only metaphorically.

When the self psychologist speaks of “psychic structure,” he is referring neither to the structures of a mental apparatus nor to the structures of any of the constituents of a mental apparatus but to the structure of the self. The structure of the self, in other words, is the theoretical correlate of those attributes of the self which, in their sum total, define this central concept of self psychology. While the notion of psychic structure is, like all theoretical constructions, no more than a tautology, it is still an invaluable aid to our thought and an indispensable tool when we communicate with one another. (1984, p. 99)

This question emerged earlier in my review of Freud’s notion of the unconscious. It is difficult to discuss psychic processes such as growth, development, and maturation--processes most of us would agree do somehow take place--without somehow imagining something, however ephemeral, that changes, but also somehow persists, that is, has “structure.” Let us bear in mind that psychology is an unusual discipline in that the object of study and the perceiver of that object are one and the same: the “self.” This identity of subject and object seems to impose the need for some reification if consideration of psychological processes is to go beyond addressing arcane metaphysical questions. That is, one may choose to dispense with philosophical fastidiousness in favor of establishing some workable representation of what might be happening intrapsychically. Since we probably must accept Freud’s notion that only a fraction of the psyche is accessible to us in any event, we will know that whatever image we make of it will be “wrong” no matter how we go about thinking psychologically. And this inescapable “wrongness” of any view of the psyche must inhere even irrespective of the poststructural understanding of the limitations of language, or the postmodern deconstruction of truth-claims. Since psyche is multivalent--a self-contradictory “complexio oppositorum,” as Jung (1969, p. 312) called it--under no circumstances could one view of it ever be “right.”

Akin to his understanding of the successful maturation of the idealized parental imago as a process of “transmuting internalization,” Kohut (1971) believed that the grandiose self also could be transmuted--recast from an expression of absurdly exaggerated demands for attention, and unobtainable fantasies of power, domination, and admiration, into a “self” capable of authentic enjoyment of ordinary life and its possibilities. But there was a difference. Transmutation of parental idealizations required some parental support--for example, some willingness on the part of the parent to accept the idealizations--but, if there were no traumatic, sudden disappointments in the idealized figure, the unavoidable, day-by-day comparison of the idealized imagoes with the actual parents would cause a more or less automatic, step-by-step withdrawal of the idealizations by means of their gradual metamorphosis into abstract ideals. On the other hand, transformation of archaic grandiosity into realistic self-esteem required active participation of the parents in acknowledging and supporting the child’s needs for admiration, and thus would demand parents who were comfortable with their own grandiosity. Siegel (1996) offered a perfect example of this “just-right” participation:

[A two year old] boy attempted to stand upright on the palm of his father's outstretched hand. The little boy climbed onto the outstretched palm and after a short struggle to find his balance, stood straight and tall. The father proudly proclaimed, “Champion of the World!!!” and the boy thrust his arms triumphantly above his head, beaming from ear to ear. Statue-like, they stood united until the boy, knowing just the right moment, turned and jumped into his father’s arms as they embraced in a cuddle of delight.

The game is a lovely example of a father’s participation in and literal support of his son's exhibitionistic narcissism. The father’s comfortable acceptance of his son’s grandiosity fosters the smooth integration of the boy's exhibitionism into the self. The father provides the psychological function of a mirror that validates, accepts and calmly reflects the boy’s grandiosity. The father’s participation in the game and his clear joy over his son’s expansiveness help modulate the intensity of the grandiose exuberance The father creates a safe playground in which the boy’s narcissism can exist unencumbered by shame, guilt, embarrassment, or overstimulation. (p. 87)

According to Kohut, a parent who easily performed the function of a mirror, as in Siegel’s example, or gracefully performed the other important function, that of being idealizable, would not be seen by the child as a person in his or her own right. Rather, that parent would be experienced as a part of the child’s own self, as an extension of the self, and Kohut (1971) named this capacity of serving as a psychological extension the “self-object function.” The person who fulfilled the function, he called a “self-object.” Later, he removed the hyphen, and created the term “selfobject”

to provide the idea that the function-providing object is not experienced as separate from the self. . . .

Psychological structures . . . are internalizations of the soothing, tension-regulating and adaptive functions that have previously been performed by selfobjects. They develop as a result of the gradual withdrawal of the narcissism invested in the old idealized objects and they continue to perform their psychological functions even in the absence of the selfobject. (Siegel, 1996, p. 72)

This view of early caregivers as selfobjects whose functions would

be internalized, thereby creating structures within the self, could not be reconciled with the Freudian conception in which parents were seen mainly as the objects of the drives and the source of drive gratification, and in which narcissism was seen as normal only in the very early stages of life, and pathological later. Kohut had tried to remain true to his Freudian roots--after all, he was President of the American Psychoanalytic Association, and his colleagues thought of him as “Mr. Psychoanalysis” (Wolf, 1996)--but his clinical experience, particularly his growing understanding of narcissism, was leading him further and further away from the idea that the “self” was produced by the intersection of innate drives with the outside world embodied in the parents. By 1977, he wrote,

It is possible to discern a self which, while it includes drives (and/or defenses) in its organization, has become a superordinated configuration whose significance transcends that of the sum of its parts. (1977, p. 97)

With this strong belief in a self “beyond knowing empirically, a supraordinate concept, transcending the sum of its parts, which has cohesiveness in space and continuity in time” (p. 177), Kohut had conceived of “an independent center of initiative . . . a central metapsychological concept” (Lee & Martin, 1991, p. 180). This was a radical departure from psychoanalytic theories that saw the “self” as a content of the “mind,” that is as a collection of self-representations. Kohut’s “self” functioned as a whole, as a unity: both as a unitary organizing principle, for example as the center of a continuity of experience--what Winnicott (1988) called “going on being”--and as a unitary agent of action, for example the infant learns motor control (Lee & Martin, 1991). Given that the infant’s milieu provided the requisites for proper emotional development, Kohut said, this “nuclear self,” would “establish an uninterrupted tension arc from basic ambitions, via basic talents and skills, toward basic ideals. This tension arc is the dynamic essence of the complete, nondefective self . . . whose establishment makes possible a creative-productive, fulfilling life” (1977, p. 4).

Here is a view entirely different from that of Freud, who saw the ego as a collection of attitudes which emerged gradually from the intersection of the innate drives and the outside world, culminating in the unavoidable Sturm und Drang of the oedipal period which might either make or break the personality. Kohut stated explicitly that Freud’s theory of the oedipus complex had only limited explanatory power:

The classical theory of drives and objects explains a good deal about the child's oedipal experiences; par excellence it explains the child’s conflicts and in particular, the child’s guilt. But it falls short in providing an adequate framework for some of the most important experiences of man, those that relate to the development and vicissitudes of his self. To be explicit . . . these theories fail to do justice to the experiences that relate to the crucially important task of building and maintaining a cohesive nuclear self. (1977, pp. 223-224)

Kohut’s theory implied that the self was something meant to be, destined to occur, and that only a failure of necessary care could derail this process of development. In Freud, and later in Melanie Klein, one sees something very different. For Klein, the infant could be born already at risk of developmental failure, that is with an excess of aggression--the death instinct--that,

if left unchecked, would lead to self-annihilation. Freud felt that this destructive inner force became expressed through either sadism or self-directed masochism . . . Klein . . . [subscribed] to the notion of a death instinct and proposed that an inner struggle between the forces of life and death ultimately was projected onto the outer world. In Klein’s view, this development is responsible for the child’s early division of the world into good and bad. A large part of the death instinct, for example, is projected onto external objects. This leads to a world filled with malevolent and destructive figures, i.e., bad objects. (Cashdan, 1988, p. 6)

For Kohut, the situation was quite the opposite. Kohut did not view aggression as an “innate drive demanding discharge, but as a ‘disintegration product’ arising in reaction to an unresponsive environment inhibiting the development of the baby’s nuclear self” (White and Weiner, 1986, p. xx, italics mine), that is, aggression was not an innate drive, but a response to faulty parenting. Thus, Kohut’s schema was a parent blaming theory, in contrast to the baby blaming theories of Melanie Klein--who saw the parent as the target of the child’s frustration, envy, and hatred, and blamed later pathology on an excess of aggression in the newborn (Lionel Corbett, personal communication, December 2, 1995). This is the crux of Kohut’s departure from his earlier work as a Freudian analyst: drives do not make the self, that is, “man is not born to conflict . . . he is not of necessity in an adversary relationship with his parents or with his culture, and . . . his development is not determined or predetermined by drives or instincts as psychoanalysis has understood them” (Basch, 1984 p. 29).

Support For Kohut’s Move Away From Drive Theory

Much evidence from other disciplines sustains Kohut’s rejection of drive theory. For example, primate studies (Arnold, 1960) showed that shortages of food and water did not necessarily lead to increased aggression--often increased cooperation and sharing was the response to shortages. What led to aggression was the intrusion of an outsider into a group. Aggression from this vantage is not a basic drive--at least not in non-human primates--but a reaction to affective stimulation caused by a strange and threatening situation. The primate studies of Beach (cited by Basch, 1984) suggested that sexual behavior is not drive-based either. Her observations indicated that when a male was given unrestricted sexual access to a female, eventually he ceased to have intercourse with her, but this could not be rationalized as a case of the male having been drained of sexual “drive,” for when a new female was introduced to the male, he immediately had intercourse with her, suggesting, according to Beach, that sexual desire is not a matter of accumulated energy, but rather of the affective stimulation aroused by the presence of a suitable partner.

Basch (1984) opined that Freud’s drive-based conception of human psychology may have been, at any rate, less an inevitable conclusion of psychoanalytic research than an artifact of Freud’s zeal for coherent explanations of behavior.

Freud himself made it clear that his theories about drives and instinct were biological speculations that he adduced to explain his clinical findings. However. . . when Freud found them useful, the initial tentative nature of extra-analytic speculations tended to be forgotten, as if repeated usage had made these hypotheses more factual. ( p. 29)

In making this move from speculation to presumed fact, in converting “an idea which was at first a modus res considerandi, a way of regarding, [into a] form which strikes its imprint upon the insight,” as Hillman (1975, p. 144) put it, Freud may have been conflating instinct with affect. The studies of Silvan Tomkins (1995) suggest that affects are present in infants from the moment of birth--and, one might ask, if at that moment, why not a moment before, or a week before--and that this system of affects perhaps expresses the root level of the human personality, but that is not the same as saying that various instinctual drives are at the root level of the personality. According to Basch (1988), affects arise out of and express the reaction to various patterns of stimulation from the external environment, and are not drives, nor their expression, but “an advanced control mechanism that serves as an inner releasor for inherited and learned behavior patterns” (1984, p. 32). Differentiating affects from instinctual drives is important in making the move away from Freud’s drive-conflict psychology to Kohut’s self-selfobject psychology, because in Tomkins’ catalog of affects one finds a possible mechanism for the early communication between infant and caregiver that Kohut put forth as vital to the proper development of the nuclear self.

Tomkins’ work suggested that the infant is born with a mode of signaling to its caregivers whether its level of sensory stimulation is either within the optimal range or not. Later work went further. The infant observation studies of Tiffany Field (1985, cited in Basch, 1988) indicated that thirty-six hours after birth, long before they could have learned anything from experience, infants were affectively attuned to adults around them. When a trained model adopted various facial expressions--joy, sorrow, surprise--the infants responded in kind. Basch (1988) posited that such affective responses to the affective signals of caregivers would be laid down as memory traces by the infant, and that such engrams made up patterns of expectation which

are remobilized when sufficiently similar circumstances arise in the infant’s life. (Neither recall nor imagery is involved in this process. What is reactivated is the sensorimotor complex which has become associated with the autonomic response to stimulation that constitutes affect.) How a baby will react, whether to welcome or withdraw from a potential re-experience, depends a great deal on the nature of the affect that has come to be attached to the memory of similar previous occasions. In other words, the baby does not react essentially differently from the way he will when he gets older. One is much more likely to anticipate with interest, and to explore, those situations that have given pleasure and enhanced self-esteem rather than those that leave one tense with fear, anger, or shame. (p. 77)

For example, a baby manifesting what Tomkins (1995) called “distress-anguish” [crying, arched eyebrows, corners of the mouth turned down] is signaling that he or she is unable to organize an input of new stimuli into acceptable patterns. If the caregiver then responds by properly soothing the baby, and perhaps by naming the affective signal--”Oh, poor baby, you are frightened, aren’t you?”--the infant

begins to learn that his sense of self can be shaken by exposure to distressing new stimuli but that the affective arousal can be given a meaningful name [frightened] and that he can restore his sense of psychological balance when it is threatened in such situations by finding someone (a selfobject) who can understand his distress, help him organize what he has experienced, and make it meaningful. (Young, 1991, p. 77)

If Field’s observations were accurate, and if Basch was on target with his notion of the immediacy of affective memories and their power to influence subsequent behavior, one might conclude that, on the affective level at least, Kohut’s self-selfobject relations begin to form in the first hours of life.

Of course, affects and their associated behaviors are not all that an adult feels and does. To see the mouth of an hours old infant turn down in the form that we associate with the emotion “anger,” is not the same as saying the infant is feeling angry. The feelings and emotions of an adult are complicated. But it is not too great a stretch to imagine that the affective responses of the newborn are the beginning of line of development that will lead to adult feelings, emotions, and behaviors, and that this sequence of maturation of affects into feelings and then into emotions will be influenced by the affective communication between infant and caregiver that Field’s work suggests begins at birth.

Basch (1984, 1988) believed that feelings begin to emerge at around the age of eighteen months when affective reactions begin to intersect with the emerging sense of self. This accords with Kohut’s (1977) idea that with “average expectable care” a nuclear, bipolar self begins to be seen during the second year of life, “bipolar” meaning that this core self consists of two sectors, primary ambitions and primary ideals. Kohut believed that the child “has two chances to establish the firmness of this early bipolar self. The first opportunity is through an approving, confirming, mirroring relationship with a maternal selfobject; the second opportunity is through a relationship with an admired, idealized, usually ­paternal, selfobject” (Berkowitz, 1983, p. 128).

Marian Tolpin (1983) pointed out that Kohut later expanded this bipolar schema of the self into a tripolar one, by adding another kind of selfobject relatedness which he called “twinship,” but which she preferred to call “the partnering self.”

The basic psychological constituents of the nuclear self are: (1) the grandiose exhibitionistic self--the child saying, as it were, “Look, ma, no hands!” and looking to his parents for their “gleam” and return smile; (2) the idealizing self--the child looking up to and reaching up to the parents, to be lifted up, calmed down, and perked up so that he can recover and begin again with some enthusiasm to look around to see where the action is; (3) the “partnering” self--the child who wants to hold hands with the parents, to be with them (like the [small boy] who peed when [his father did]). (p. 120)

Inevitably, Kohut’s tri-partite scheme was expanded further--Stern (1985) recognized seven different forms of selfobject relationships--but the root idea perdures: each infant is born with a fragmentary self which gains firmness, cohesiveness, definition, and stability through interaction with caregivers who respond empathically to the infant’s needs. With sufficient “empathic attunement,” the infant gradually internalizes the functions of the caregivers, but inadequate attunement engenders developmental arrests in one or more of the sectors of the self, leaving structural deficits that demand compensation in ways which might have been acceptable in the child, but which are not phase appropriate for an adult. For example, an infant who lacked proper mirroring (did not see enough of the gleam of interest and approval in its mother’s eye) might grow into an adult who seeks the approval of others in ways and in situations which are inexpedient, destructive, or dangerous. Depending on how archaic the developmental lacunae, and how numerous, the adult person might experience only occasional, isolated sensitivities, as most of us do, or could be enmeshed in the throes of a severe narcissistic disorder characterized by a self that will not cohere, the “fragmenting self” that requires constant stimulation (affective arousal) merely to preserve its feeling of being at all. Such cases come to the attention of therapists through a variety of manifestations: depression, rage, anxiety, addictions, perverse behavior, extreme sensitivity to slights, obsessive thinking, sexual acting out, etc. (Kohut 1977, 1984).

Self Psychology in Clinical Practice

Fortunately, according to Kohut (1977), there exists in the human psyche an inherent impetus to go forward in one’s development which makes therapeutic intervention possible even in serious disorders of the self. Such intervention is a complex matter, but to provide a short recapitulation: the analyst, “tunes in” empathically to the patient through “vicarious introspection . . . the capacity to think and feel oneself into the inner life of another person” (Kohut, 1984, p. 82). This empathic attunement provokes the formation of transferences that take on the character of the patient’s developmental deficits. That is, upon feeling the attunement of the analyst, the patient begins to demand from the analyst what he or she needed, but was not given, in infancy and childhood. Thus, a patient whose infantile grandiosity was not sufficiently mirrored--seen and supported--might want to be listened to unceasingly and with unflagging interest. The analyst, even if he or she attempts to provide all that the patient demands, inevitably falls short, provoking negative reactions on the part of the patient. The analyst inevitably falls short, because even if the she or he has arrived at a more or less accurate understanding of what the patient feels, and even if the analyst acknowledges that the patient has a right to those feelings, there is no way that such archaic needs could ever be totally satisfied within the constraints of the analytic situation. The patient’s reactions to such empathic failures may then be interpreted with regard to the patient’s developmental history, that is, as related to the patient’s traumatic (intolerable) disappointments in parental selfobjects. According to Kohut, just as the inevitable but tolerable disappointments in parental empathy were occasions for incorporation of the parent’s selfobject functions, such cycles of therapeutic “failure” and interpretation of the patient’s disappointment will be opportunities for the patient to incorporate the selfobject functions of the analyst by means of transmuting internalization. (1979, 1984). Thus, in this conception, psychoanalysis no longer is a process of “making the unconscious conscious” or “where id was, there ego shall be” (Freud 1933, cited by Kohut, 1984, p. 103), but rather the analyst’s specific understanding of the patient’s need, and the analyst’s acknowledgement that such a need “is legitimate . . . both as the revival of an old unfulfilled need and as the manifestation of a universal need for selfobjects that persists throughout life” (p. 103).

Kohut’s schema attributed a teleology to the unfoldment of the self. He rejected the drive-conflict theories of Freud, because he came to believe, like C. G. Jung, that the self was something mysterious and unknowable: “[only its] introspectively or empathically perceived psychological manifestations are open to us” (1977, p. 311). Like Jung, Kohut ascribed to the self an intrinsic kismet: ”the nuclear self. . .a structure that, once it has been established, has, from the beginning, a destiny, a potential life curve” (1978c, p. 594). For Kohut, the fulfillment of this destiny was the main event in a life. What Jung perhaps would have called “individuation,” Kohut called being true to one’s “innermost design,” as in this passage:

I believe that there is, later in life, a specific point that can be seen as crucially significant--a point in the life curve of the self at which a final crucial test determines whether the previous development had failed or had succeeded. . . . But I am inclined to put the pivotal point even later--to late middle age when, nearing the ultimate decline, we ask ourselves whether we have been true to our innermost design. This is the time of utmost hopelessness for some, of utter lethargy, of that depression without guilt and self-directed aggression, which overtakes those who feel that they have failed and cannot remedy the failure in the time and with the energies still at their disposal. (1977, p. 241)

Confirmation of Kohut’s Theories in Direct Infant Observation

Viewing each individual life as having an “innermost design” opened Kohut to charges of being “metaphysical” and “unscientific,” as did his insistence on vicarious introspection as the basic method of an analyst’s becoming immersed in the patient’s material, which “lies outside the realm of causal sequences [that can be accounted for by] our present means of logical or psychological explanation“ (1971, p. 302). His ideas have been deprecated as mystical--”[he] opens the door to a strange and enchanted land of make believe” (Hanley & Mason, 1976, cited by Corbett, 1989)--and disparaged as speculations lacking convincing evidence (Summers, 1994). About his indictment as a mystic, one can say little, for such judgments depend upon where along the logical spectrum the judge is situated. From my standpoint, the scientist, even if she or he claims to understand the limitations of that vantage, usually misses seeing something of the enchantment in life, and the seer, no matter how acute, often cannot avoid at least some reification of what might better be treated as metaphorical. But the second objection to Kohut’s theories, that they lack substance because Kohut elaborated his system from material gleaned in his analyses of adult patients, and on subsequent reconstruction of what simply conjectured had occurred in infancy and early childhood--not on direct observation of infants--perhaps may be addressed by looking for confirmation in the work of some theorists who were “baby watchers.”

John Bowlby, although he trained with Joan Riviere, a Kleinian analyst, and was supervised by Klein herself, eventually rejected the Kleinian notion that the wellspring of emotional development was the group of endogenous fantasies stemming from the death instinct and its conflict with libidinal drives, and went on to propose that actual family experiences, not internal fantasies, were uppermost in importance (Bretherton, 1994). Bowlby’s voluminous observations of children and interviews with their parents convinced him that

if mental development is to proceed smoothly, it would appear necessary for the undifferentiated psyche to be exposed . . . to the influence of the psychic organizer--the mother. . . . She orients him in space and time, provides his environment, permits the satisfaction of some impulses, restricts others. . . . Gradually he learns these arts himself. (1951, p. 53, cited by Bretherton, 1994)

This conception seems to clearly to evoke Kohut’s parental selfobject and its step-by-step internalization. Bowlby’s statement that,

The joint task of the therapist and client is to understand the origins of the client’s dysfunctional internal working models of self and attachment figures. . . .The therapist can be most helpful by serving as a reliable, secure base from which an individual can begin the arduous task of exploring and reworking his or her internal working models (p. 454, italics mine.),

supports Kohut’s notion of the therapist’s fulfilling selfobject functions for the patient, fostering their incremental incorporation.

Margaret Mahler, an analyst who devoted years to direct observation of infants and toddlers with serious pathology, as well as to a parallel program of observation of normal children, formulated a theory of development which appears to support Kohut’s in many of its essentials, albeit the economy of drives remained important in her conception. In Mahler’s separation-individuation theory, the mother, as primary love object, is at the center of development, and her “empathic” response to the child is crucial for enabling the child to develop through the various cycles of separation-individuation in appropriate ways (Delany, 1994). “In addition to being a source of drive gratification, the parent functions as a temporary auxiliary ego, an organizing influence, a source of the developing inner sense of self, and an object for internalization and identification” (Settlage, 1994, p. 21, italics mine.). This language sounds like Kohut’s selfobject functions. Mahler’s “refueling” (1975)--the nonverbal cues of support that a mother gives to her baby as it struggles to assert itself in the separation phase-- seems close to Tolpin’s view of the idealizing self object function (“perked up so that he can recover and begin again with some enthusiasm to look around”), cited supra. Mahler’s idea of the importance of the empathic parent for proper early development finds parallels in Kohut’s work, and Mahler seems particularly close to Kohut in postulating that borderline pathology finds its genesis in early developmental arrests--for her, failure in the “rapprochement” subphase of separation-individuation in the sixteen to twenty-five month old child (Mahler, 1975)-- and in her notion that “remobilization of previously arrested development is an integral feature and powerful therapeutic force within the psychoanalytic process” (Levine, 1994, p. 123).

Stern’s (1985) inferences about the subjective life of infants, drawn from his own exhaustive infant studies and from the observational data of others, support many of Kohut’s formulations both implicitly and explicitly. For example, Stern said,

When the development of an infant is observed with a clinical eye one sees almost no pathology. . . . Instead there are characteristic patterns and some variant patterns, but there is very little basis for believing that any deviations from the norm are going to result in later pathology. When there are deviations, it is the relationship with the caregivers and not the infant alone that appears deviant. (p. 186, italics mine.)

This strongly supports Kohut’s central teleonomy: given an adequate self-selfobject milieu, the self tends toward proper development.

Kohut believed that affect and experience unite to produce internalized, persistent imagoes, which change and mature with continuing experience, but remain forever necessary to the sense of being oneself--the selfobjects. In Stern’s nomenclature these became ”Representations of Interactions that have been Generalized” (RIGs), which Stern said, “constitute a basic unit for the representation of the core self” (1985, p. 98). If the RIG implicated an actual person, according to Stern, then an activation of the RIG (which could ensue from the repetition of even a small part of its original contents) could produce an “evoked companion,” the feeling of the inner presence of such a person, no longer dependent on his or her physical existence, which seems very close to Kohut’s selfobject.

In his discussion of “core-relatedness,” Stern (1985) did not mention Kohut, but Stern’s observations produced a schema delineating ways in which inappropriate interactions with caregivers might give rise to the states of being that self psychologists call the “understimulated self,” and the “overstimulated self.” In his treatment of “intersubjective relatedness,” Stern drew distinctions among “non-attunement,” “selective attunement,” and “misattunement,” demonstrating their origins in the characters of the parents, and explaining the variety of their possible influences on the developing self. One feels on reading Stern’s anecdotes that the bones of Kohut’s theories suddenly acquire living flesh, and that the typical environmental/developmental milieux that self psychologists deduce from reconstructions of the backgrounds of adult patients find substantial incarnation in the particular infants and their parents named as the subjects of Stern’s researches.

The developmental theories of Donald Winnicott sound strikingly similar to those of Kohut. Since Winnicott, psychiatrist at Paddington Green Children’s Hospital for forty years, stated that he personally had taken 20,000 case histories, and that, “direct clinical observations of babies . . . have indeed been the main basis for everything that I have built into theory” (Winnicott, 1971, cited by Davis & Wallbridge, 1981), such similarities do much to provide corroboration in infant observation for the ideas of self psychology. According to Bacal,

We must assume that Kohut had some familiarity with Winnicott’s ideas, but it is perhaps for these reasons [their unsystematized presentation and sometimes difficult exposition] that he chose not to build on them, despite their evident similarity to his own. (1989, p. 259)

In fact, we may do more than assume that Kohut was familiar with the ideas of Winnicott. Kohut (1977) stated explicitly that he had read Winnicott, but that he chose not to cite Winnicott’s work because,

I did not set myself the task of integrating the results of my work with the results of the work of others . . . that had been formulated within a vague, ambiguous, or shifting theoretical framework. I felt that to undertake such a task at this point was not only inadvisable but that it would indeed pose an insurmountable obstacle on the way toward my objectives. (p. xxi)

The background here, which Kohut would not discuss even with his closest friends, is that he had been diagnosed in 1971 with leukemia, and knew that he was dying (Paul Ornstein, personal communication, September 1, 1996). When he said that “my focus is not on scholarly completeness” (Kohut, 1977, p. xx), he meant that he simply did not have time for anything but hastening to communicate his own ideas before dying. Nevertheless, Kohut was well aware of Winnicott, and, it seems to me, that Winnicott’s central ideas, drawn, not from Kohut’s procedure of “empathic immersion in the transference” of adult patients (1977, p. xxii), but from Winnicott’s observations of infants and their parents, greatly support and enliven Kohut’s theoretical posture.

Kohut’s “optimal developmental ambience” (1984), seems perfectly embodied in this passage from Winnicott: “The good enough mother. . . is one who makes active adaptation to the infant’s needs, an active adaptation that gradually lessens, according to the infant’s growing ability to account for failure of adaptation and to tolerate the results of frustration” (1971, p. 10). If the mother were not “good enough,” and therapeutic intervention were required, Winnicott believed that,

The children in our care, in so far as they have a need for therapy, are going through phases in which they go back and experience again . . . the early relationships which were not satisfactory in their past history. We are able to identify with them as the mother identifies herself with her infant, temporarily but fully. (1965b, p. 16)

This seems parallel to Kohut’s view of the analyst’s empathically supplying necessary selfobject functions until they are absorbed by transmuting internalization.

One of Kohut’s most important departures from his early Freudian orthodoxy was his declaration that castration anxiety, central to the previous psychoanalytic canon, was not

a feature of the oedipal phase of the healthy child of healthy parents. . . . The healthy child of healthy parents enters the oedipal phase joyfully. The joy he experiences is due not only to the fact that he himself responds with pride to a developmental achievement, that is, to a new and expanding capacity for affection and assertiveness, but also to the fact that this achievement elicits a glow of empathic joy and pride from the side of the oedipal-phase selfobjects. (1984, p. 14, italics mine)

This revisioning of the oedipal period as a phase of development that will not be pathogenic unless impaired previous development and a deficient family milieu has left the child without healthy self-selfobject relations--Kohut said elsewhere that, “The particular set of conflicts called the Oedipus complex are not the primary cause of psychopathology but its result” (1984, p. 53)--closely resembles Winnicott’s (1988) statement that,

The Oedipus Complex is . . . a description of an achievement of health. Ill-health belongs not to the Oedipus Complex, but to the repression of ideas and inhibitions that follow from . . . conflict. Happy and healthy is the boy who reaches [the oedipal stage] in his emotional and physical development when the family is intact, and who can be seen through the awkward situations . . . by his own two parents that he knows well, parents who tolerate ideas, and whose inter-relationship is sound enough so that they do not fear the strain of loyalties brought about by the child’s loves and hates. (p. 50, italics mine)

Self psychology, which views the selfobject experience as depending partly on the capacity for illusion and partly on the personal qualities of the person fulfilling the selfobject functions, finds a close analog in Winnicott’s (1971) concept of the “transitional object.” Winnicott said that creativity and the ability for real relatedness depended greatly on the illusory space first occupied by the transitional object remaining open throughout life. Bacal (1989), speaking of self psychology, said that “throughout life, the personal meaning of the interpersonal relationship, or object relationship, always depends to some extent upon illusion” (p. 268).

Finally, Kohut’s idea that early development progressed by means of optimal frustration seems perfectly described by Winnicott here:

It could be said that at the beginning the mother must adapt almost exactly to the infant’s needs in order that the infant personality shall develop without distortion. She is able to fail in her adaptation, however, and to fail increasingly, and this is because the infant’s mind and the infant’s intellectual processes are able to account for and so to allow for failures in adaptation. In this way the mind is allied to the mother and takes over part of her function. (1965a, p. 7)

The Making of an Apostate

Hindsight is always 20-20 vision. After reviewing some of the corroboration for Kohut’s ideas in the work of independent researchers and theoreticians, it seems clear that Kohut’s picture of human emotional development, if not correct--let us recall that since psyche is multivalent, and self-contradictory, no view of it could ever be “correct”--at least seems useful and reasonable. Today, thousands of therapists use the ideas of self psychology routinely in clinical practice, and the Jungian analytic community in particular has been profoundly influenced by Kohut’s work (Corbett, 1989). But in the 1960s, the move away from classical psychoanalysis was not easy. Kohut was among the aristocracy in orthodox Freudian society. He was President of the American Psychoanalytic Association. He was Vice-President of the International Psychoanalytic Association, and was expected to be the next President. Kohut also was famous at the Chicago Psychoanalytic Institute for teaching the best Freudian theory courses, and for writing the most engaging papers (Wolf, 1996). But he was about to abandon the theoretical basis of a brilliant career, and to slide into apostasy by creating a theory that cut the heart--the oedipal, drive-conflict concept-- out of Freud’s system. His creation of self psychology would cause friends to abandon him, and colleagues to keep their distance (Ernest Wolf, personal communication, September 1, 1997). A Kohut lecture would be condemned as not sufficiently eulogistic of Freud, and shortly after, he would be ousted from the Psychoanalytic Education Council of the Chicago Institute by a vote of its other members who would conspire by telephone to remove him, wounding him deeply (Paul Tolpin, personal communication, September 9, 1997). What could induce such a difficult and costly iconoclasm?

In the first place, many of Kohut’s patients had told him that his oedipal interpretations missed the mark, and he found himself having increasing difficulty in ascribing their objections to “resistance.” As he put this in his final book, How Does Analysis Cure, published posthumously,

If there is one lesson that 1 have learned during my life as an analyst, it is the lesson that what my patients tell me is likely to be true--that many times when I believed that I was right and my patients were wrong, it turned out, though often only after a prolonged search, that my rightness was superficial whereas their rightness was profound. (1984, pp. 93-94)

Secondly, his experience as President of the American Psychoanalytic Association (1964-1965) had been deeply discouraging. Ernest Wolf, a colleague, collaborator, and close friend of his, remembers Kohut remarking bitterly, and often, on the flagrant narcissism, and the self-serving political maneuvering of previously respected colleagues. According to Wolf, Kohut already had the basis for his ideas about narcissism--”by 1966 he had the makings of the new paradigm”-- but his experience of the narcissistic behavior of his colleagues may be what finally moved him to make them public (personal communication, September 1, 1997). Paul Ornstein, another close colleague and collaborator, agrees. As Ornstein remembers events, Kohut’s first paper emphasizing the importance of narcissism and its treatment was delivered as Kohut’s Presidential Address to the Association, delivered just as his term was ending (personal communication, September 1, 1997).

Further, Kohut struggled constantly with his own narcissistic proclivities. Ernest Wolf remembers that Kohut was cautious about choosing those with whom he would speak about his theories,

not because he was haughty as some people believed, but because he was vulnerable, and needed constant mirroring. He was not arrogant, but vulnerable. In those days I would get to my office early, usually before seven when expecting an eight o’clock patient. Often around seven the phone would ring, and it would be Heinz, feeling lonely and needing to chat. It was that kind of narcissism. Not the arrogant kind of self-overestimation. (personal communication, September 1, 1997).

Again, Paul Ornstein agrees with Wolf. According to Ornstein,

He couldn’t tolerate stupidity easily, and was selective about who he would invite to his home for evenings of conversation. Then he was less interested in our thoughts then in having us listen to his. Yes, in that sense he was narcissistic. Except that he had the most interesting thoughts. My perception is that some of them were awed by what he had to say, and then resented it.

Some people described him as distant, uncaring about others, highly narcissistic. And of course, in that sense, narcissistic he was. Self absorbed he was. But that never interfered with his relationship with me and with Anna [Anna Ornstein, Paul’s wife]. He was extremely sensitive and vulnerable, so he protected himself. (personal communication, August 29,


The Two Analyses of Mr. Z

In 1979, Kohut published a paper, The two analyses of Mr. Z., describing a pair of analyses that he had conducted with the same patient. The first analysis, of four and a half years duration, with meetings five times per week, took place while Kohut still practiced in the classical Freudian tradition. Mr. Z was a graduate student in his mid-twenties who presented to Kohut complaining of an inability to form relationships with women, and with a list of somatic complaints such as sweaty palms, irregular heartbeats, and a feeling of uncomfortable fullness in his stomach. The second analysis, begun after a hiatus of five years, that is, almost ten years after the beginning of the first, was carried out using Kohut’s new ideas about developmental deficits in the structure of the self, narcissistic transferences, and the possibility of cure by means of selfobject experiences within the analysis. This second analysis also lasted for four and a half years with meetings five times a week. Taken together, the two analyses provided a way of comparing the clinical effects of the two differing theoretical perspectives.

The case history is complex--could any analysis of nine years duration, five days a week not be complex?--but briefly: while Mr. Z was a young child, his father became seriously ill, and while hospitalized fell in love with his nurse, eventually leaving home to live with her. While his father was away, Mr. Z, then about four years old, slept in his bed. When his father came back home again, a year and a half later, Mr. Z moved out of his father’s bed, and began sleeping at the foot of his mother’s bed, where he repeatedly witnessed their sexual intercourse. At that time, Mr. Z began masturbation, involving fantasies of being sold as a slave to a woman who used him to clean up her feces and urine, urinated into his mouth, and in various other bizarre ways humiliated him. Later, around age eleven, Mr. Z, had a homosexual experience with a camp counselor whom he idealized, and in whose presence he had always felt calmer and stronger.

As an adult, Mr. Z continued compulsive masturbation--his only sexual outlet--still accompanied by masochistic fantasies of servitude and humiliation. He had little social life, except for evenings out with his mother. But these embarrassing facts only emerged later in the analysis. At first, Mr. Z presented himself to Kohut as grandiose, arrogant, and superior. In the classical Freudian mode, Kohut interpreted that supercilious attitude as an expression of Mr. Z’s imagined oedipal victory, when, at age three his father went to the hospital, leaving Mr. Z’s mother alone with him. In that Freudian scenario, hidden behind the repression barrier lay castration anxiety and depression due to what really had been an oedipal defeat: the father’s coming back. After four and a half years of making such interpretations, Kohut ended the work, pointing to a decrease in Mr. Z’s masturbation and accompanying masochistic fantasies, and to Mr. Z’s moving out of his mother’s home and initiating some sexual experiences with women as indications of success in the work. Also, Kohut noted that Mr. Z’s narcissistic demands had moderated, and that finally he was able to accept Kohut’s interpretations of those demands as defenses against his repressed fear of castration. Near the end of the analysis, Mr. Z. had the following dream:

[Mr. Z] was in a house, at the inner side of a door which was a crack open. Outside was the father, loaded with gift-wrapped packages, wanting to enter, The patient was intensely frightened and attempted to close the door in order to keep the father out. (Kohut 1979, pp, 407-8)

From the same classical outlook that had determined his interpretations of his patient’s narcissism, Kohut interpreted this dream as expressing Mr. Z’s ambivalence towards his father, marked by fear of castration due to Mr. Z’s having been living alone with his mother. Soon after, the analysis ended, time well spent in Kohut’s estimation.

Five years later, Mr. Z. was back. He still suffered, it seemed, from lack of success with women, and what sexual liaisons he was able to arrange still needed to be accompanied by humiliating masochistic fantasies. He felt unhappy at work, and found life in general unsatisfactory. There was a delay before Kohut could begin working with Mr. Z again, but after a couple of initial appointments, used just to plan for the subsequent work, the patient began to feel better. Kohut, using his new theoretical lens, pictured this sudden improvement in mood as the result of Mr. Z’s having established an idealization of Kohut, similar to the idealization of the camp counselor during adolescence. An initial dream in this second analysis, the image of a powerful figure of a man who evoked in Mr. Z associations to his father, to the camp counselor, and to Kohut himself, supported Kohut’s idea of an idealizing transference.

This analysis focused on Mr. Z’s mother, who recently had become psychotic with paranoid delusions. Mr. Z remembered that in childhood his mother had seen him not as a person in his own right, but as an appendage. Like Portnoy’s mother in Philip Roth’s Portnoy’s Complaint, a novel, interestingly enough, about a compulsive masturbator, Mr. Z’s mother would examine his feces, and search his skin for blackheads. He was allowed no privacy; she would enter his room, unannounced, whenever she liked. In Kohut’s terms, Mr. Z had been forced into a merger with his mother. In this second analysis, Mr. Z came to comprehend that his previous view of his mother as a devoted, loving parent had been one-sided, and that his mother’s love was far from unconditional. Her “love” required that Mr. Z submit entirely to her demands, giving up any hope of autonomy. She gave her love only on condition that he remain for her a permanent selfobject whose presence was necessary for her to sustain her own fragmenting self, necessary, that is, for her to avoid breaking through into the psychotic core of her borderline personality. That latent psychosis, against which she always had used her merger with Mr. Z to defend, finally had emerged in her paranoid delusions when, during the course of the first analysis, Mr. Z finally had left her house for a place of his own. Kohut was surprised that almost nothing of that outre parent-child relationship had emerged in the first analysis. He searched for the reason for that flaw in the first analysis, finally imputing it to his having forced upon his patient the,

convictions of a classical analyst who saw the material that the patient presented in terms of infantile drives and of conflicts about them, and of agencies of a mental apparatus either clashing or cooperating with each other. (Kohut 1979, p. 423).

And Kohut blamed himself for having become for his patient,

a replica of the mother's hidden psychosis, of a distorted outlook on the world to which he had adjusted in childhood, which he had readily accepted as reality--an attitude of compliance and acceptance that he now reinstated with regard to me and to the seemingly unshakable convictions that I held. (p. 423)

Thus, for Kohut, the second analysis of Mr. Z stood as a major watershed. Having understood that classical analysis forced something upon the patient, there was no going back; he would have to come out of the closet with his self psychology. As he expressed this in his final writing:

This patient, as I saw clearly in the second analysis, must have confronted me, as many other analysands throughout my professional life undoubtedly had, with the recognition that his needs were primary and real and not defensive. But whereas formerly . . . I had silently made my compromises with established theory and technique, I was this time in greater need of keeping those forces in check which demanded that I acknowledge what I saw and how I saw it. These forces whispered to me that I should not only stand up for my new insights in the clinical situation but also . . . raise the findings of individual psychology to more general levels, by expressing them in a carefully chosen, well-defined terminology, and by communicating them to the broad scientific community. (1984, p. 88)

Although it would cost Kohut several important friends, and cause colleagues to keep him at a distance, and although he struggled for a time at least to maintain Freud’s language, if not his metapsychology (Ernest Wolf, personal communication, September 1, 1997), the cat would have to be let out of the bag.

As the second analysis of Mr. Z continued, Kohut concluded that the homosexual relationship with the camp counselor had not been about sex qua sex; rather it had been the expression of Mr. Z’s need for a meaningful relationship with a strong male figure. Affiliation with his own father could not have provided that relationship, because, in early childhood, his father had almost always been away, and even after his return, Mr. Z saw him as a weak man, kowtowing to a domineering, demanding wife. As the latter awareness developed, Mr. Z suddenly understood why his father had wanted to leave home for another woman, whereupon he recovered a forgotten memory of a ski trip with his father at age nine. On the ski trip, Mr. Z had seen his father in a different light. His father, an excellent skier, had seemed at ease in the world. Also, there was a woman present who had what seemed to be a sexual relationship with his father. Perhaps she was the nurse for whom he earlier had left his wife. Mr. Z remembered having kept those details secret from his mother, as if he and his father had shared a special masculine understanding. In Kohut’s view, the recovery of this memory indicated a revival, within the analysis, of Mr. Z’s childhood need for a man he could idealize.

In this analysis, Kohut no longer construed Mr. Z’s having begun masturbation after witnessing his parents intercourse as an expression of incestuous fantasies. This time, Kohut explained the masturbation as Mr. Z’s attempts to regulate the overstimulation of his undeveloped self occasioned by his mother’s demanding that Mr. Z merge into her adult experience, thereby precluding his own normal experience which would have furnished the phase appropriate, gentler stimulation of a child’s increasing autonomy. And Kohut now perceived Mr. Z’s masochistic fantasies not as manifestations of unconscious guilt for his oedipal fantasies, but as symbols of domination by his unstable, demanding mother.

As this analysis drew to a close, Mr. Z reintroduced the earlier dream of being frightened when his father appeared at the door with gifts. This time, Kohut provided a different interpretation, no longer construing the dream as a reference to fear of castration. From the viewpoint of his new self psychology, and in the light of the recovered memories of the ski trip, Kohut understood the dream as expressing a yearning for the father and the gifts he carried--masculinity and psychological strength--gifts which the child’s enfeebled self, due its enmeshment as a permanent selfobject for his mother, had not been able to receive. In this view, closing the door was not an attempt to protect himself from a castrating father figure, but rather an effort to modulate the excitement he had felt at the return of his long-absent father.

In the classical view, according to Kohut (1979), the psyche of Mr. Z could be pictured as having a horizontal split (Freud’s repression barrier). Above the horizontal split was Mr. Z’s overt grandiosity and arrogance due to his imaginary oedipal victory over his father (his father’s departure). Below the horizontal split lay Mr. Z’s castration anxiety and depression due to his actual oedipal defeat (his father’s return). In Kohut’s new paradigm, the psyche of Mr. Z had a vertical split. To one side of the split lay Mr. Z’s overt arrogance and superiority “on the basis of persisting merger with the . . . idealized mother [who] confirms the patient’s superiority over father provided patient remains an appendage of her” (p. 446). On the other side of the vertical split, lay another part of Mr. Z’s psyche, this part containing a repression barrier à la Freud. Above the repression barrier, that is consciously, Mr. Z experienced impaired self-esteem, depression, and masochism. And this is what was split off (vertically) from his arrogance; in other words, when he was feeling arrogant, he was protected against his depression and low self-regard. Below the repression barrier was not castration anxiety as in the Freudian understanding, but rage against his domineering mother, and self-assertive male sexuality with which he could not get in touch.

The difference between these two schemata is striking. For the Freudian Kohut, grandiosity was the outcome of an imagined oedipal victory, but this was a grandiosity founded on a shaky footing, since below the repression barrier lay the anxiety of paternal reprisals through castration. For Kohut the self psychologist, Mr. Z’s grandiosity had nothing to do with the vicissitudes of the oedipal situation. Rather it was a defense against the depression and low self-esteem occasioned by the patient’s lack of proper mirroring by his mother. It was a defense against a failure of normal development in the “grandiose sector” of the self due to mother never having appreciated Mr. Z for himself, but only as a selfobject serving her needs, a reversal of the appropriate parent-child relationship. As long as his defensive grandiosity was dominant, Mr. Z could not properly make the idealizing connection to a male figure which he needed to be healed. But as Kohut properly understood and explained Mr. Z’s arrogance, Mr. Z was free to idealize Kohut, and, through that idealization, slowly to incorporate the selfobject analyst--to make the functions that Kohut was carrying out for him become part of him. Later, Kohut learned that the second analysis really had been helpful: Mr. Z was happily married, and was regarded as outstandingly successful in his chosen career.

From the viewpoint of analytic self psychology, the mechanism of Mr. Z’s healing was not primarily self-knowledge--not Freud’s “where id was there ego shall be”--but the gradual accumulation of self structure through transmuting internalization of the functions at first provided by the psychoanalyst. Kohut (1984) later said that such healing took place through a three step process, the first two steps of which were 1) analysis of the patient’s defenses, and 2) unfolding of the transferences,

while the third step, the essential one because it defines the aim and the result of the cure--is the opening of a path of empathy between self and selfobject, specifically, the establishment of empathic in-tuneness between self and selfobject on mature adult levels. This new channel of empathy permanently takes the place of the formerly repressed or split-off archaic narcissistic relationship; it supplants the bondage that had formerly tied the archaic self to the archaic selfobject. ( p. 65)

In clinical practice, step one would involve the analyst’s becoming aware of the characteristic defenses of the patient. In Mr. Z’s case, the defenses were arrogance, superiority, and superciliousness. In another patient the defenses might be entirely different, perhaps even opposite: self-deprecation and meekness, for instance. With this awareness, the analyst would have understood the defenses, which she or he then could explain to the patient. The explanation would not necessarily be--in fact would probably not be--a technical one. For example, the analyst might say, “You seem to need to feel superior to me.” Being understood in this way would provoke step two, the formation of transferences in the patient. These transferences, in Kohut’s conception, would take on the character of the developmental deficits in the patient’s self structure. For example, if the patient had not received proper mirroring of entirely natural early grandiosity, the patient might require mirroring by the analyst. Such mirroring might take the form of the analyst’s noticing and supporting the patient’s accomplishments, or perhaps simply listening to the patient with real attention. But the patient’s needs, expressed within the transference, would never be entirely satisfied. That would provoke reactions or complaints from the patient. Then, those needs--in this example, the needs for mirroring--could be explained. For example, the analyst might say, “When I understood what you were telling me, I noticed that you seemed to relax. You seemed calm and contented. It must have felt good to be heard and understood. But when I misunderstood you, you became upset.” With the transference explained, step three, the opening of a path of empathic communication on a mature level, could ensue. In other words, the analyst and the patient together could understand and empathize with the patient’s situation in a way not possible when the patient was confined within the structure of defenses, and this communication in itself would be healing. Instead of remaining trapped within a deficient self, tied to early self objects no longer appropriate to adult life, the patient would be internalizing the communication style of an intact self, which then would become part of the patient’s self, as it might have in childhood with sufficient empathic care.

Was Kohut Mr. Z?

This one case demonstrates both:1) the clinical/theoretical reasons why Kohut needed to abandon drive-conflict theory, and 2) many of the theoretical assumptions of his new self psychology, foremost of which is that when an impaired sense of self is strengthened, problems abate. On that basis alone, Mr. Z’s case is ideal material for understanding the evolution of Kohut’s ideas, but there is more. Earlier, we saw that Kohut’s interest in narcissism was not just theoretical, but personal. Recall that Ernest Wolf said Kohut “was vulnerable, and needed constant mirroring” (personal communication, September 1, 1997), and that Paul Ornstein said that, “people described him as distant, uncaring about others, highly narcissistic. And of course, in that sense, narcissistic he was. . . . he was extremely sensitive and vulnerable, so he protected himself” (personal communication, August 29, 1997). Since his death, it has been rumored that Kohut probably was Mr. Z. My conversations with Kohut’s contemporaries lend credence to that conjecture.

According to Paul Stepansky, who collaborated in the posthumous publication of Kohut’s final book, “Betty [his wife], and Tom [his son] confirmed to me that Kohut was Mr. Z” (personal communication, December 9, 1996). Ernest Wolf, Kohut’s close friend and collaborator, agreed.

For one thing, there are many parallels between Mr. Z’s case and Kohut’s own biography. Like Mr. Z, Kohut’s father was absent [he was away at war], and after returning was chronically depressed. Also, Kohut was not allowed to go to school with other children, but was educated at home by a tutor who may have been the camp counselor in the Mr. Z case history. Anyway, I long ago accepted that Mr. Z was Kohut’s self-analysis, even before the controversy. Freud did it all the time. I never even needed to ask him.

Paul Tolpin, who along with Michael Basch, John Gedo, Arnold Goldberg, David Marcus, and Paul Ornstein, attended chapter-by-chapter readings of Kohut’s first book in progress, and who earlier had been Kohut’s analysand, had this to say:

I think it’s highly likely that he was. In fact years ago before anybody was talking about it, I realized from things he had said about himself that it did sound like him. He had let things drop that sounded so much like Mr. Z. He had an analysis with Ruth Eissler in Chicago and hated it, and the only other analysis he had after that was with himself. He spoke about being a lonely child, and how important his tutor was to him. That tutor was like the camp counselor in Mr. Z.

My wife [Marian Tolpin], and Ernie [Wolf], and I got together one night and somehow it all came out that each of the three of us already had been thinking that it was a possibility. There were so many points of convergence between his story and Mr. Z’s. He may have put together parts of himself with parts of somebody else. It hard to think of Kohut tasting his own feces which is part of that case too. . . . I guess we don’t like to think of that in our leaders. They should be healthy people who understand the ill.

Some people feel that it was improper to use himself and to say that he had discovered that in other people, but Freud also created things right out of his own development. Like when he peed in the pot in his father’s bedroom, and the father said “You’ll never grow up to be a decent man,” and that became castration anxiety. I don’t know that men do have that kind of castration anxiety. It’s possible, but it doesn’t come up in patients of mine who focus on that specifically.

Whether all the details are correct or whether he intensified the case to get across the desperateness of this poor man in order to demonstrate his theory, I think he was Mr. Z. (Paul Tolpin, personal communication, September 10, 1997).

Paul Ornstein, who besides being a friend, colleague, and member, along with Tolpin, of the early group, edited and wrote the introductions for volumes of Kohut’s papers, could not confirm that Kohut was Mr. Z, but thought it likely.

Many analysts before, including Freud and Anna Freud, have presented self-analyses as if they came from patients. When someone finds certain constellations, genetic or dynamic, as a result of self-analysis, he is bound to see it also in other people. Also, Kohut had a case from a German analyst that he wanted to use in Restoration of the Self. At the last minute, the German analyst changed his mind, and refused to let Kohut publish the case, so Kohut used Mr. Z. But Kohut burned his letters before his death, so now we will never know.

In any case, for me it is irrelevant. The speculation is that the first analysis was Kohut’s analysis with Ruth Eissler in Chicago, and the second, his own, but what is important to me is not whether Kohut was Mr. Z--although he may have been--but extracting the essence of the meaning of the comparison of the two analyses. I think that the meaning of the nature of the interpretation of the dream in the first and the second is the clinching aspect of what sort of changes he arrived at in his theory. That is, rather than seeing the dream as a defense against oedipal feelings, he felt at that moment that it was a self protection against overstimulation.

And yet, the need for the gift from the father--because that is a theory of his also, how to have masculine substance in yourself--and the recapturing of the relationship with the father that time at the ski resort, seemed to describe how he saw psychoanalysis and what it can do, and what the transference is all about. So, very sketchily, the dream is what he saw as describing the essential difference in the two analyses. Also, the other major point related to it was that rather than seeing what Mr. Z presented as defensive, seeing it as expressing what he experienced. And that is a major difference. (personal communication, August 29 1997)

Ornstein went on to say that,

If a theory is going to be good, it has to come from within, from the authenticity and validity at the personal level. The analysts who live on are the ones whose contribution definitely occurred through self-analysis.

On that point, Ernest Wolf told me that,

The ideas of a certain theorist--I won’t say who it is, but you can probably guess [I could not guess]--were received this way by the other analysts: “Ah, yes. That sounds just like many of my patients.” But Kohut’s ideas were received like this: “Ah, yes. That reminds me of myself. Do you see the difference in that? Kohut’s ideas were alive because they came from within, from deep self-analysis. (personal communication, September 1, 1997)

My days of conversations with Kohut’s contemporaries produced in me a deepened compassion for Kohut, a respect for his courage, and a heightened esteem for his theories of psychological development and psychotherapeutic practice which, in my view, as much as any important work of art, clearly speak to our human generality while, at the same time, expressing the deepest personal struggles of their inventor.