Self Psychology As A Bridge Between Existential-Humanistic Psychology and Psychoanalysis


STEPHAN A. TOBIN, Ph.D., is in the private practice of psychotherapy in Los Angeles and is an Assistant Clinical Professor of Psychology at UCLA. Originally trained in Freudian analyti­cally oriented therapy at the University of Michigan, he studied Gestalt therapy with Fritz Perls and Jim Simkin, was a cofounder and president of the Gestalt Therapy Institute of Los Angeles, and served as a trainer in the institute for many years. He also had training in neo-Reichian therapy and transpersonal psychology, and began a personal and professional exploration of self psychology approximately 10 years ago. He continues to teach and train therapists, and, although now practicing primarily as a self psychology therapist, continues to integrate aspects of Gestalt and neo-Reichian therapies into his work. He has written numerous articles on psychotherapy.


This article on psychoanalytic self psychology attempts to show how it incorporates the depth, complexity, thoroughness, and broad appli­cability of psychoanalysis with the philosophical and scientific pos­tulates of existential-humanistic psychology. Such major existential­ humanistic ideas as holism, phenomenology, ontology, field theory, free will, and an emphasis on values and ideals are shown to be fundamental to self psychology. The author also demonstrates how the concepts of transference and countertransference are redefined in a way that makes these invaluable aids to therapeutic work compatible with existential-humanistic precepts. Some important recent developmental research is brought into the discussion and shown to be relevant to the work with adult therapeutic clients. 

Finally, the author articulates some important advantages of self psychology over existential-humanistic psychotherapies as well as ways in which self psychology theorists could profit from reading important existential-humanistic writers.

In the late 1960s I began to be disenchanted with the orthodox Freudian analytic training I had received at the University of Michigan. Criticisms by Fritz Perls, Abraham Maslow, Carl Rogers, Rollo May, and other existential-humanistic psychologists helped me to articulate my dissatisfactions: Freudian psychoanalysis had too pessimistic a view of humans and their possibilities; it could be a deadening, overintellectualized form of therapy for both patient and therapist; it tended to create an authoritarian, simplistically "scientific," medically modeled therapist-client relationship de­ meaning to the client; it was often ineffective, resulting in little significant change in people who had had many years of analysis; it neglected the impact of the person of the therapist; it was elitist, both in terms of who did the therapy and who got the therapy; it promoted an objectification of clients rather than a view of them as unique human beings; and it subtly stressed an adaptation to mainstream, patriarchal, middle-class, socially acceptable values rather than stimulating the client's discovery of his or her own unique path in life.

As a reaction to these dissatisfactions, I began to move in the direction of existential-humanistic therapy, particularly Gestalt therapy. Gestalt therapy was an exhilarating breath of fresh air to me because of the emphasis on freedom of choice encouraged in clients, the creativity encouraged in the therapist, its physical and sensory emphasis, and the focus on expression of emotion rather than mere intellectual understanding. I had personal Gestalt therapy, studied with Fritz Perls and Jim Simkin, and eventually helped found the Gestalt Therapy Institute of Los Angeles, where I trained therapists for many years.

After a few years of practicing and teaching Gestalt, however, I also began to discern many limitations in this type of therapy. Some of these were stylistic, the result of the abrasive, confrontive, frequently destructive way Perls worked, a way that many other Gestaltists, myself included, emulated. In comparing Gestalt with many other types of existential therapies with which I had some acquaintance-for example, Rogerian, neo-Reichian, primal ther­apy, bioenergetics, rolfing, transactional analysis, and even trans-personal therapy-I began to realize that the limitations I saw in Gestalt therapy were in many existential-humanistic therapies: conceptual oversimplicity; a lack of sufficient consideration of individual differences; a lack of awareness of the tremendous influence of unconscious factors on behavior; an inadequate regard for transference issues and a systematic way of conceptualizing about transference, resulting in gross misunderstandings of the behavior of certain patients; a disregard of countertransference phenomena, frequently resulting in ineffective therapeutic experi­ences, at best, and destructive client experiences, at worst; a lack of understanding of just how deep-seated many clients' difficulties are, and how long and arduous is effective therapy with them; and the lack of consideration of developmental issues in understanding psychopathology.

The result of my growing awareness of the various inadequacies of the existential-humanistic therapies I explored left me in a serious dilemma: I much preferred the underlying philosophical assumptions of existential-humanistic psychology, but preferred the intellectual depth, sophistication, and precision of psychoanal­ysis. In my search for something more intellectually and personally satisfying, I began to find that there had been a great many developments in psychoanalysis since I had abandoned that theo­retical discipline in the sixties, and I started to study object relations theory. I found much to admire in the work of many of the English object relations theorists, such as Winnicott (1958, 1965), Guntrip (1969, 1971), Fairbairn (1952), and Balint (1968), who stressed the relationships between the human being and his or her caretakers in the formation of psychic structures from the beginning of life, rather than the Freudian emphasis on drives and an unfolding biology. They also adapted a more humanistic atti­tude toward therapy patients as persons. But they still seemed to me to lack a consistent, coherent metatheory of personality, devel­opment, and psychotherapy. Most of the American object relations theorists, such as Kernberg (1975, 1976) and Masterson (1976) were influenced greatly by Melanie Klein, and have retained too much of Freudian metatheory and Freudian philosophy to suit me. I began to realize that what I really wanted was a well worked out, theoretically sophisticated, consistent form of therapy that integrated psychoanalysis with existential-humanistic philosophy.

Of course, many writers within the broad and rather vague field of existential-humanistic psychology have attempted to integrate components of existential-humanism with aspects of psychoanal­ysis (e.g., Abraham Maslow, Rollo May, Medard Boss, James Bugental, Erich Fromm, and Irvin Yalom) and some psychoana­lytic theorists have incorporated existential-humanistic features into their theories of therapy and personality (e.g., Carl Jung, Alfred Adler, Karen Horney, and the previously mentioned English object relations theorists). Although I have learned a great deal of value from these theorists, none of their systems of personality and therapy have ever been very satisfying to me.

Finally, in the middle 1970s I discovered a new type of analytic therapy that did appear to have most of the qualities I was seeking: psychoanalytic self psychology, which, in its short life (about 17 years), has gone a long way toward achieving the philosophical and theoretical integration for which I have yearned.

In the rest of this article I present an outline of how I see self psychology combining the best of both existential-humanistic psy­chology and psychoanalysis.

A Brief History Of Self Psychology

Self psychology is largely the creation of Heinz Kohut, although there are many other creative theorists currently adding to and modifying self psychology theory and clinical practice. Some of these are Arnold Goldberg (1978, 1980), Robert Stolorow (Atwood & Stolorow, 1984; Stolorow, Brandschaft, & Atwood, 1987; Stolorow & Lachmann, 1980), Michael Basch (1980, 1983), Paul and Anna Ornstein (Ornstein & Ornstein, 1980, 1985), and Estelle and Morton Shane (Shane, 1985; Shane & Shane, 1980).

Kohut was a Viennese-born psychoanalyst who lived, practiced, and taught in Chicago after coming to the United States in 1940. Until 1971, when he published his first book, The Analysis of the Self, Kohut was known as an orthodox analyst, a favorite of Anna Freud's, and a leading light in the branch of Freudian analysis loosely called ego psychology. Starting with his 1971book, however, he began to create a theory and form of therapy that, although in­cluding the analytic basic assumption of the decisive importance of unconscious factors in human behavior, represented a major addition to psychoanalysis: increased understanding and im­proved treatment of those patients previously considered unana­lyzable, the narcissistic personality disorders. In The Analysis of the Self, he was still operating within the metatheoretical confines of Freudian theory, but in his second book, The Restoration of the Self, published in 1977, he offers an alternative perspective to emotional disorders that represents a basic departure from many of the seminal theoretical assumptions that Freud and almost all of his followers had held to be basic: the ubiquity of the Oedipus complex, the assumption of sexual and aggressive drives as the primary human motivational system, and the essential nature of structural conflict (between id, ego, and superego) in psychological problems. In this book Kohut created a new form of psychoanalysis, one that is better attuned to modern scientific postulates and that rests upon a metatheory that incorporates many existential-humanistic psychological principles.

Although there is no generally agreed-upon list of basic assump­tions of existential-humanistic psychology, these are the general attitudes and ideas I would include in such a list: (a) an emphasis on phenomenology rather than on physicalistic explanations of the nature of human beings; (b) holism rather than reductionism in conceptualizing about and working with patients; (c) a field theory approach to considering the therapeutic situation; (d) frequently, a greater emphasis on emotion in theory and practice than one finds in orthodox analysis (particularly stressed by Carl Rogers);

(e) a positive or at least neutral view of basic human nature; (f) accep­tance of free will as well as determinism as facts in human exis­tence; and (g) an approach in which therapists help their clients to define and determine their own values, life meanings, and life purposes, rather than imposing these purposes and meanings on them.

What are the distinguishing features of psychoanalytic thera­pies? I see these as a primary emphasis on unconscious determi­nants of human behavior; an implicit or explicit theory of development; a primary emphasis on exploration of the individual's past (as contrasted, for example, to the Gestalt notion of dealing with the past when the individual signals the presence of ''unfinished business"); the therapist's assumption of a posture of neutrality or nongratification in the therapy setting; attending to transference and countertransference issues; and viewing the therapeutic rela­tionship, at least partially, within a developmental framework.

It should be noted that Kohut, to my knowledge, never consid­ered himself to be an existential humanist. Stolorow (1976) and Kahn (1984) have pointed out many similarities between Kohut and Carl Rogers, but Kohut does not mention Rogers in any of his writings.

Although he was extremely interested in history and art, and he made some attempt to use his theories to understand people, such as Hitler, who have had an important historical impact (Kohut, 1985), he does not make reference in his writings to the great existential and humanistic thinkers. Whether or not he was directly influenced by these writers, however, he created a method of therapy and a theory of human personality that has the most important features of them. It is to these features that I now turn.

Existential-Humanistic Aspects Of Self Psychology

Although a complete exposition of self psychology is beyond the scope of this article, I shall attempt to discuss aspects of the theory and therapeutic methodology in considering each of the points I make.


Edmund Husserl is generally conceded to be the founder of phenomenology, and May (1958), Binswanger (163), and Boss (1963) are some of the psychologists who attempted to incorporate phenomenology into psychoanalysis and personality theory. I think that the existential philosophers and psychologists have pointed to what it is to be human and the problems facing human beings in general. But I do not think that they have managed to create complete enough systems of therapy or paid enough attention to developmental theory and research to give therapists an effective way of helping individuals with the existential difficulties they face. And as Atwood and Stolorow say:

We are in sympathy with such reformulations, insofar as their aim has been to free the phenomenological knowledge of psychoanalysis from its procrustean bed of mechanism and determinism. What has limited the success of so-called "existential analysis," however, has been its tendency uncritically to import into psychoanalytic theory philosophical concepts and categories not genuinely grounded in clinical observations . . . A truly psychoanalytic phenomenology re­sists the philosopher's temptation to define consciousness in univer­sal terms and instead seeks understanding of the phenomena tran­spiring in the specific intersubjective dialogue of the psychoanalytic situation. (Atwood & Stolorow, 1984)

Starting in the 1950s, when Kohut was still a respectable member of the psychoanalytic establishment, he was stating his belief that the subject matter for psychoanalytic science should be the inner life of human beings, and that this inner life can be approached, scientifically, only through the use of introspection and empathy (Kohut, 1959). He thus ruled out biological, sociolog­ical, and behavioral ways of obtaining data and attempted to free psychoanalysis from the materialistic, deterministic, and mecha­nistic concepts that were the legacy of Freud's adherence to nineteenth­ century biology. This methodological empathy/introspection maxim was later elaborated (Kohut, 1977) to rule out of self psychology some basic Freudian notions: the primacy of the instincts of aggres­sion and sexuality and the concept of psychic energy.

Kohut thought that even the so-called biological drives of sex and aggression were actually psychological, only having been called "drives" because people motivated by them experience a quality of drivenness, particularly if they block their direct expression. Even the object relation theorists, although stressing primary object-self relationships as the main determinants of psychological growth, still see impulses as the building blocks of relational configurations (Greenberg & Mitchell, 1983). Kohut, on the other hand, sees the relational configurations as primary, with the impulses of sex and aggression, when separated from the person's holistic existence, as secondary breakdown products of a fragmenting self.

Kohut was not opposed to using insights from biological fields in psychoanalysis, but felt that the constructs must be psycholog­ical, and he was explicitly critical of the tendency to overidealize the tools and methods of natural sciences while neglecting the tools of introspection and empathy.

In view of this theoretical emphasis on phenomenology, it should not be surprising that the self psychologist would also place great stress on phenomenology in the way he or she works in therapy. Rather than approaching treatment from the outside in, as do most analysts, the approach is from the inside out, that is, a continual attempt on the part of the therapist to put himself or herself into the client's psychic "shoes" so as to understand the client's subjec­tive world, particularly his or her emotions and developmental needs. This means that the therapist does not attempt to tell the client what unconscious motives and thoughts are "really" driving him or her, as a Kleinian would do, or to dig down into the patient's unconscious to unearth repressed motives and thoughts, the typi­cal Freudian approach. The self psychological approach is to at­ tempt to experience empathically, on a sustained basis, the inner life of the client, and to communicate this understanding to the client. This empathic process ultimately results in the client estab­lishing the kind of bond with the therapist that enables him or her to bring into the therapeutic relationship the yearnings (and fears of those yearnings) that were so traumatically disappointed in the past. This, in turn, allows the client eventually to reexperience the painful, frightening, and devastating events of the past.

This empathic process, also emphasized by Carl Rogers, was at first understood by Kohut to be solely an information-gathering technique, enabling the analyst to make the most accurate inter­pretation to the patient. This view of empathy was very different from the view of Rogers, who thought that the client's experiencing of the therapist's empathic stance was curative in its own right. In his last years, however, Kohut (1982) began to view empathy similarly to Rogers, saying "I must now, unfortunately, add that empathy per se, the mere presence of empathy, has also a benefi­cial, in a broad sense, a therapeutic effect-both in the clinical setting and in human life in general" (p. 397). (According to Whiten & Weiner l1986], he used the word "unfortunately" because of his conflict between his wish to be an objective scientist and his growing realization that "the empathic observer affects the very situation he or she is trying to understand" [p. 351).) Even with this change in his view of empathy, however, he still placed great emphasis on interpretation as a change agent.

Still a third way in which phenomenology (and empathy) enters into self psychology is in the way that the therapist approaches client perceptions that are drastically different from his or her own. Whereas a Freudian analyst might decide that the client is defen­sively distorting his or her perceptions of the world by, for example, the projection of unacceptable impulses, the self psychologist would attempt to understand the validity of the client's perception in the context of the client's world, both past and present, including the developmental level the client has reached. In fact, the ap­proach is never to evaluate the degree of truth, rationality, or morality of the client's way of being in the world, but to attempt to show the client that, given his or her life experiences and degree of awareness, he or she could not perceive the world and him- or herself any differently. As an example of the difference in approach between the sustained empathic approach of self psychology and the experience-distant approach of other forms of psychoanalysis, I shall mention a discussion I had with a group of therapists at a training institution where I gave a talk on self psychology. The discussion concerned a female patient who is married to a man who physically mistreats her. The formulation of the therapist was that this allegedly masochistic woman was unconsciously provoking her husband to hit her, her motive being to prove herself morally superior to him while, at the same time, projecting her unaccept­able aggressive impulses onto him. I have a number of objections to this "experience-distant" formulation.

First, it is a formulation not arrived at in an emphatic or introspective fashion, but through theoretical inference. Second, it makes the woman responsible for her husband's mistreatment of her. Third, because people who are in abusive relationships usually tolerate them because they feel so inadequate and worthless that they accept the abuse as justified, this interpretation, if presented to the patient, could make the patient feel even worse about herself, thus repeating the pattern of abuse to which the woman had become accustomed. Fourth, even as a hypothesis, not commu­nicated to the patient, it is a potentially dangerous formulation, because it could possibly lead to a particular prejudicial set in listening to the material the patient presented. The therapist then might be looking for evidence to support his or her hypothesis, rather than helping the patient learn why she puts up with the abuse, how the abuse feels, what it repeats from the past, and so on. I might add that, although I am more sympathetic to the woman's situation in this marriage, were I to treat the husband, I would take exactly the same empathic tack: to attempt to under­ stand and to help him understand what forces were motivating him to hit his wife, rather than judging him or trying to force him to stop doing it. My experience in similar circumstances has been that the abusive person is generally motivated by great feelings of narcissistic injury and a desire to right the wrong of the injury by striking out at the person by whom he or she has felt injured. But rather than impose this theory or expectation on the client, I would try to help the husband experience his own inner process.

Note that I am not attempting to take away the woman's respon­sibility for staying in the relationship, putting up with the abuse, and so on. I have found that people who tend to get themselves into a series of abusive relationships are frequently those whose par­ ents were both abusive almost from birth, or who had one abusive parent and one who was emotionally very withdrawn. Because any kind of contact, even hostile contact, is preferable to none, these people have learned to form relationships with abusive persons because their self-structures are organized around negative af­fects. Putting it another way, one could say that they attempt to achieve an external matching of an inner world of relations with hostile, destructive internal objects. I have also found that the only way out of this dilemma for these so-called masochistic persons is a long-standing relationship with a therapist to whom they can learn to relate in a positive way and whose positive attitudes toward them can ultimately be internalized.

Self Psychology is Holistic

The holistic attitude toward human beings and psychotherapy has been persuasively advanced by Harry Stack Sullivan (1953), Erik Erikson (1964), Rollo May (1958), Carl Rogers (1961), and Fritz Perls (1947; Perls, Hefferline, & Goodman, 1951), who got many of his ideas from Gestalt psychology and Kurt Goldstein's organismic theory of human personality. As Perls said:

The organism is a whole. As you can abstract the biochemical, behavioristic, experiental, etc. function and make one your specific sphere of interest, so you can approach the total organism from different aspects, provided that you realize that any change in any sphere produces a change in every other corresponding aspect. (Perls, 1969, pp. 167-168)

Self psychology is holistic in at least three ways: in its theoretical approach, in the conceptual framework of the personality theory and in the way the therapy is conducted. I shall discuss these aspects separately.

Theoretically, self psychology is consistently holistic rather than reductionistic in that the attempt is to discover meaningful, in­tegrative, expansive configurations in human behavior, not to analyze experience into disjunctive, abstract, experience-distant bits. Kohut took the same theoretical position about human beings that the Gestalt psychologists and the writers mentioned above take, that is, that one can understand people in their essence only by dealing with holistic, dynamic, integrative patterns. He felt that in attempting to reduce people theoretically to abstract processes, as did Freud, with such concepts as instinctual drives and mental apparatuses (ego, id, and superego), the essential experienced meanings of their functioning is lost.

Kohut felt that the basic human motive in life is the attempt to pre­ serve and enhance the self in its wholeness rather than merely reduc­ing the drive tension of instincts separate from whole functioning.

The main structural concept in self psychology is the self, and it is holistic in nature. The self concept was not, of course, invented by Kohut. Lecky (1945), Allport (1961), Adler (1927), Jung (1956), and Horney (1950) are just a few of the psychological writers who included the concept in their personality theories. The concept was not used in Freudian analysis until Hartmann (1964) pointed out that the concept of ego included phenomena that should be sub­sumed under a separate concept of self.

Among the defining properties of the self, in Kohut's initial definition (1966, 1971), is that it is the core of the personality and, experientially, what is identified by the person as being most uniquely him or her. The self is also a center of initiative and includes the person's awareness of him- or herself as existing in space and time.

Al so according to Kohut's earlier conception of self, it is partly conscious-therefore "experience-near"-and partly unconscious. In his later definition (1977, 1984), when he was essentially mak­ing the self a supraordinate concept at the center of the personality, the self is defined as bipolar, the individual's goals and ambitions being at one pole and his or her ideals and guides for living at the other pole. Between the two poles is a "tension-arc" of skills and talents with which the person attempts to perform the lifelong juggling act of meeting his or her goals while, at the same time, living in accord with his or her ideals, standards, and so on.

It is also important to note that Kohut stated, in response to criticisms of some vagueness in his definition of the self, and some differences between earlier and later definitions, that he did not feel that the self could ever be completely defined, because it is an empirically derived configuration and thus not completely reduc­ible to a definitive verbal description.

Stolorow, also a self psychologist, defines the self simply as the individual's organization of his or her self-experience in the world (Socarides & Stolorow, 1984). In fact, Stolorow makes a cogent case for rejecting some aspects of Kohut's definition of the bipolar self-concept because they are not accessible to introspection and empathy and the definition also separates structure from function (Stolorow, 1986)!

In a way similar to Martin Buber, Kohut saw the healthy, vital, cohesive self developing as a result of the individual's being treated and regarded as a whole person by those who serve the person as a selfobject. The selfobject, or person who fulfills selfobject func­tions for the individual, supports the individual's sense of continu­ity over time and in space. The selfobject concept will be discussed more fully in the next section.

Self psychology also exemplifies holism in the clinical setting. Most people who come for psychotherapy are suffering some deficit or defect in their self-structure, that is, in the ways in which they organize their experience of their worlds. This defect is usually in the vitality, cohesion, or harmony of the self-organization. What these people with self-defects need, then, to resume their psycho­ logical growth is a therapeutic experience that will help in the strengthening of the self-structure. As an attempt to meet this need, the self psychology therapist is continually responding to his or her client in a way that helps to confirm the client's essential wholeness and to explore what has transpired, in or out of the therapy, to create fragmentation, devitalization, or disharmony. Practically speaking, I am continually looking for the positive, integrative, holistic motives linked to my clients' behavior, even when that behavior itself is destructive to the person or others. An example of the holistic approach is a session with a client, Ms. J., who came in feeling very depressed because she had been cruel to her teen-aged daughter. Ms. J. was feeling much self-hatred and was ashamed to tell me what she had done. She was even feeling that her child-indeed, the whole world-would be better off if she herself were not alive.

Upon exploration, it turned out that Ms. J. had felt hurt and rejected by her daughter, who had compared her very negatively to the mother of a girlfriend. My client had become enraged, had struck her daughter, and then castigated her as being a "selfish, unappreciative, heartless brat." She had later felt very contrite and remorseful and had attempted to talk to her daughter, but her daughter had coldly rejected what were attempts, as she now realized, to obtain forgiveness from her. The daughter told Ms. J. that Ms. J. was "crazy," and that the daughter would be happy when she was old enough to get away from her. This interchange had utterly devastated the client, particularly because she fre­quently worries that she is psychotic as her mother was when she was a child.

Ms. J. was ashamed that she had felt so wounded by her daughter that she had lost control, which means to her that she is a very weak person. This, in turn, fills her with a feeling of hopelessness that, despite many years of therapy, she is still so vulnerable and fragile.

I helped Ms. J. to see that, given the lack of support for handling negative feelings she had gotten from her parents as a child, she had not developed the inner strength to bear rejections like the one she had sustained from her daughter. She then wailed ''Why, I have a little baby inside myself-and I'm the one that's supposed to be the mother! I even need my own daughter to be like a mother to me, to approve of me!" I said, "Of course you do. That baby part of you was treated so insensitively and with so much resentment that it feels very fragile and frightened and, at times, very hurt and enraged." I also pointed out that, when she referred to her ''baby­ part," she sounded full of contempt. She agreed that she does feel contempt for that baby and, as we explored this, we realized that her mother had probably communicated a similar feeling to her when she really was a baby, and was appropriately acting like one by trying to get nurturing, approval and acceptance from her mother.

Ms. J. began to feel less depressed and to understand why she reacted to her daughter in the way she had. She also saw that, in a way, her reaction was constructive: Instead of just passively taking her daughter's abuse, she had reacted more aggressively than usual. We had been working on the reasons she feared reacting angrily to others, including her daughter: Her mother had been very punitive when Ms. J. had gotten angry at her and Ms. J. was afraid of being at all like her frequently furious, out-of-control mother. I also pointed out, when she said she was no better as a parent than her mother, that her daughter had at least felt free to be critical of her, whereas she had not dared do that openly to her own mother. I also expressed understanding of the pain and guilt she felt when she realized the pain she had inflicted on her own daughter. After all, she had suffered in the same way. She even began to understand in subsequent sessions that her daughter's coldness was merely a way of getting back at her instead of taking the very frightening risk of further expressing the hurt and humil­iation that underlay her anger.

Note that I was not condoning her abusive treatment of her daughter, just responding holistically to all of her. I certainly did not need to tell her that expressing her feelings in that way is not good for her daughter. She knows that very well, and feels ashamed and frightened of her destructive behavior. I trust that, when she develops more understanding and acceptance of herself, she will no longer react that way to her daughter. I think that she will also be more accepting of her despised ''baby" needs and find appropri­ate ways of meeting those needs.

Self Psychology is a Field Theory

Self psychologists emphasize the human's embeddedness in a "field of intersubjectivity" (Stolorow et al., 1983) all through life. Thus the human infant's psychological growth is viewed not as being primarily determined by an unfolding biological program, as in Freud's stages of psychosexual maturation, but as the outcome of an infant/caretaker field, in which the approving, soothing, confirming, and accepting functions of the caretaker are essential to the development of a resilient self. Kohut coined the term "selfobject" to refer to an object (another person) who is related to by the person only in terms of the phase-appropriate needs of his or her developing self. There is little or no recognition of the separateness of the object and it is not viewed as having its own center of initiative.

Winnicott (1965), Guntrip (1969), and Fairbairn (1952) have also stressed the importance of the caretaker/infant field in the formation of psychic structure, but did not distinguish between object and selfobject needs.

Although the genetic endowment of the infant was not empha­sized by Kohut, Demos (1984), Stern (1984), and other current developmental researchers and theorists have stressed the impor­tance of "fit" between the infant and his or her selfobjects.

One of Kohut's more famous sayings is that human beings need empathic selfobjects as much for psychological survival as they need oxygen for physiological survival. (This theoretical emphasis is also being supported by a great deal of recent developmental research. For an excellent summary and an exciting theoretical synthesis of the research, see Stern, 1984.)

The caretaker, or selfobject, fulfills selfobject functions for the infant and child, that is, acting so as to maintain the existence, cohesion, and vitality of the infant's self when the infant cannot provide these functions for him- or herself. According to Kohut, the infant expects to control the parental selfobject in as total a way as the human adult expects to be able to control his or her own body, and the selfobject in infancy is not even experienced as separate from the self. But there is quite a bit of evidence (Stern, 1984) that, contrary to Freud and to Mahler, Pine, and Bergman (1975), there is never a stage of life when the distinction between self and others is blurred. This notion of a phase in which the infant perceives a fused self/selfobject unit is not, however, essential to self psychology theory; the essential criterion is whether or not the other is needed for the maintenance or enhancement of the self.

Selfobject needs do not exist only in childhood, to be outgrown in adulthood; Kohut felt that even adults who were lucky enough to have "good enough" parenting still have the need for affirming and idealized selfobjects throughout life, albeit in a more mature, complex, abstract way than the child or the self-disturbed adult.

In those adults who have not had adequate parenting, the self-structure is usually lacking in vitality or cohesion, or does not function in a holistic, harmonious fashion. Such people are contin­ually looking to others to fulfill their archaic selfobject needs. Examples are persons who need constant attention and approval from others, and those who attempt to shore up their essential feelings of emptiness and weakness by identifying with, blindly idealizing, and following charismatic leaders.

One place many people attempt to get these selfobject needs met is in the therapy setting. Typically, however, these clients also have great, usually unconscious, resistance to allowing their archaic selfobject needs (for perfect mirroring or merger) to surface in therapy. Although the needs are very powerful, the unconscious fear is that, were they to be expressed to the therapist, the latter would react in the same disastrous ways the parents reacted to the client in childhood. Stolorow et al. (1987) conceptualize a ''bipolar transference configuration" in therapy. They believe that much patient behavior in therapy can be viewed as oscillating between hopes that the therapist will gratify the patient's needs for mirror­ ing and/or merger, at one pole, to resistance to expressing these selfobject needs because of a fear of a repetition of traumatic injury by the therapist, at the other pole.

It should be noted that the therapist does not actually provide these selfobject gratifications, beyond empathically communicated understanding, but helps the client to see what these needs are and helps him or her to begin to find ways of meeting them in life. No one is capable of meeting these needs in the same way that "good enough" parents can with their children, and, even the therapist's degree of understanding of the client is limited. There­ fore, the client experiences numerous nontraumatic disappoint­ments in the therapeutic situation, disappointments that the ther­apist must deal with in an empathic, systematic way. There are a number of steps involved in this process, which Kohut called "optimal frustration," and other self psychologists (e.g., Bacal, 1985) have termed "optimal responsiveness": (1) The therapist senses that a narcissistic injury (self-fragmenting, self-devitalizing or self-deharmonizing) has occurred. Sometimes the signs of the injury are very subtle; for example, the client may seem suddenly a bit withdrawn in the therapy hour or may uncharacteristically show up late for a session. At other times the signs are very dramatic; for example, the client threatens to quit therapy or becomes very depressed or enraged. (2) The therapist tries to ascertain what may have happened in the therapeutic field to trigger the injury; it may have been an incorrect interpretation, the therapist's momentary inattention in a particular session, or the therapist's announcement that he or she is going on vacation or raising the therapy fee. (3) Once the triggering event is found and communicated to the client, the therapist then attempts to find out what the event meant to the client in the present. For example, the client may feel that the incorrect interpretation by the previously idealized therapist means he or she is completely incompetent and unable to help the client, leaving him or her once more completely alone in a world perceived as hostile and uncomprehending. (4) The next step is connecting the event to representative childhood experiences in which traumatic selfobject failures oc­curred. (5) The final event is a reestablishment of the self-selfobject bond between the therapist and the client, which generally results automatically upon the successful fulfillment of the first four steps.

As a result of numerous such transactions in therapy, the client is able to resume his or her developmental growth. Eventually, the process results in the client being satisfied with less concrete, less frequent, more mature ways of achieving selfobject gratification in the world than when he or she entered therapy.

Kohut was well aware of Heisenberg's work in physics and felt that just as the physicist can never assume that the phenomena he or she is observing is unaffected by the observational process, the therapist can never consider himself or herself as a purely "objective" observer. Freud's stance, vis-a-vis his patients, was what Kohut termed a "truth" morality, in which the analyst ap­proaches the patient with the detachment of a surgeon involved in an operation. In a manner similar to the existential criticism of Western "subject-object split" (e.g., May, 1958), Kohut criticized this idealization of objectivity, which attempts to maintain a rigid barrier between the perceived and the perceiver, and is especially inappropriate in the therapeutic situation, particularly with those persons who suffered precisely from the experience of being emotionally and empathically cut off from their caretakers as children. He stressed that taking an aloof, objective stance with such patients reinforces their experience of themselves as being alone, unworthy of understanding, and needing to adapt to the reality of others in order to survive.

It is true, of course, that many modern psychoanalysts do not believe that the therapist should be a "blank screen" with patients. There is even evidence that Freud himself did not function this way. There is still a sizable minority, however, who do subscribe to this view of the therapist's proper stance vis-a-vis clients (Malcom, 1981).

Note that Kohut did not mean that the therapist should express feelings to clients, or be inappropriately self-revealing, indiscriminately supportive, or confrontive. Just as the empathic or under­ standing caretaker knows instinctively what his or her child needs for the furthering of psychological and emotional growth at various stages, not infantalizing the child or pushing him or her beyond the limits of which he or she is capable, so the self psychology therapist attempts to ascertain just how much and what kind of support is needed at various stages of the therapy. Thus theory, such as the Oedipus complex or even self psychology concepts, is not regarded as more "true" than the phenomenological reality of the particular patient. It is, of course, true that one never can be completely free of the impact of theory, even if implicit, on what one observes in the therapy situation. It is important, however, for the therapist to be able to suspend theory so as to listen empathic­ ally to the client. Freud himself took this view (Freud, 1938). It was only by suspending his knowledge of the usual way analysts conceived of narcissistic problems that enabled Kohut to formulate an earlier version of self psychology (Kohut, 1971).

What Kohut advocated, is, I believe, very close to the I-Thou attitude toward others that Martin Buber (1958) discussed. In fact, Kohut's definition of the empathic attitude is almost identical to Buber's definition of I-Thou: "the capacity to think and feel oneself into the inner life of another person . . . while simultaneously retaining the stance of an observer" (Kohut, 1984, p. 82). Compare this to Buber's description of what he viewed as the correct therapist attitude toward clients:

The psychotherapist, like the educator, must stand again and again not merely at his own pole in the bipolar relations, but also with the strength of present realization at the other pole, and experience the effect of his own action. . . . Healing, like educating, is only possible to the one who lives over against the other, and yet is detached. (Buber, 1958, p. 133)

It is important to note that Buber defined the concept of empathy in a different way than Kohut defined it, that is, as a state in which the individual excludes awareness of his or her own internal state while experiencing the internal state of the other (Buber, 1965).

"Inclusion" is the word he used to describe what Kohut means by "empathy."

In describing how the therapist should treat clients, Kohut felt that the therapist must take a position of "ordinary human responsiveness," feeling and reacting to the patient as a caring, respectful human being, rather than attempting to be an emotionally neutral, observing blank screen. Although self psychology analysts, being mostly respectable, middle-class professionals who would still like acceptance from mainstream psychoanalysis, (except for those who seem to relish their roles as psychoanalytic revolutionaries) would not use the term, one could describe the therapeutic attitude of the self psychology therapist as one of love. This is not the possessive, self-gratification of object love, but a kind of deeply respectful, empathic, nonpitying caring in which the individual client is valued and responded to above and beyond the therapist's morality, theories, and narcissistic needs. This is an attitude similar to Rogers's "unconditional positive regard" (1961) and what spiritual teachers have termed "heart" or "fourth chakra" love. Once again, Martin Buber's concept of the I-Thou or dialogic attitude is also applicable (Friedman, 1985).

This approach to other human beings is, of course, a kind of ideal that perhaps no one, outside of a Buddha or a Jesus, actually ever reaches and sustains for long. Self psychology therapists empha­size the understanding and working through of inevitable em­pathic failures, caused either by the therapist's lack of understand­ ing, unresolved countertransference issues, or unavoidable reality issues (e.g., the therapist's taking a vacation) that result in the "optimally frustrating" or, perhaps more appropriately, the "opti­mally responsive" process (Bacal, 1985) described above.

Kohut's belief in the importance of the therapist studying other aspects of human knowledge-for example, history-is an indication that he saw the "field" including much more than the therapist and the client within the consultation room; it includes the broader society within which the patient has lived, the history of that society, and so on.

It should be clear by now that this approach means that psycho­ therapy is primarily seen as a real relationship, with the therapist having to consider continually his or her effect on his or her client. This is in striking contrast to Freud's notion of the analyst as an objective observer of the client. Of course this objective-observer stance is not rigidly adhered to by most psychoanalysts today except for those orthodox Freudians described by Malcom (1981). Even Freud actually behaved in a much more human way with many patients than he encouraged his followers to do.

Although Kohut at first rejected the idea that the therapist provides a corrective emotional experience for the client, his final papers and books show that he had accepted this aspect of therapy (Kohut, 1984). Hirsch (1987) places Kohut among those therapeutic theorists who view the therapist's role as one of participant­ observer.

When the field theoretical emphasis of self psychology is taken to its logical extreme, it is incorrect even to diagnose a patient in a certain way, for example, as a ''borderline state,'' because the diagnostic description does not specify the context within which the client's pathology is manifest. Thus a patient might display borderline symptomatology with one therapist, very different symptomatology with another. Addressing the issue of borderline and psychotic diagnosis, Kohut said:

The borderline [diagnosis], with the psychoses, is a relative one. It depends not only on the observed field, that is, on the patient and his pathology, but also on the ability of the therapist to extend his empathy to the patient. Insofar as you can truly build a bridge of empathy to a person, to that extent he is not psychotic. You have not cured him from his psychosis; he isn't psychotic. Once you are with him and have built this bridge, he has ceased to be psychotic. I am calmly treating people now who are delusional. And it doesn't particularly frighten me, for I understand what's going on. The delusions are in response to things felt about me and the world. They become a psychologically meaningful way of expressing states. ( Kohut, 1985)

Stolorow et al. (1987) take a similar position, even suggesting that the category of "borderline state" is not a distinct pathological entity.

It is important to note that this field emphasis does not make self psychology therapy an interpersonal therapy per se, a term that describes many existential-humanistic approaches. The therapist's main focus is still on the client's intrapsychic processes. It is incumbent, on the therapist, however, if one takes a self psychological perspective, to consider his or her own personality as well as the client's behavior in the therapeutic dyad. The result is that there are few proscriptions or prescriptions of therapist behaviornin the therapeutic situation. Whether or not the therapist answers a client's question or grants a client's request for information or a hug, or accepts a gift from a client, depends on the personal predilections of the therapist, his or her assessment of the meaning of the request, and his or her assessment of the developmental stage of the client's self/selfobject relations.

Other types of therapy have, of course, recognized the importance of looking at the therapist's contributions to the therapeutic environment. But Kohut, to my knowledge, is the first psychoanalyst to bring field theory into psychoanalysis and to tie the different stages of the therapeutic relationship to developmental processes with his understanding of the therapist's self-maintaining functions for the client.

An example of adherence to the field theoretical aspect of self psychology is the way I dealt with a patient who had formed a rapid idealizing selfobject transference to me. This idealized bond, which consisted of her seeing me as omnipotently powerful and omnisciently wise, had enabled her to relinquish the many hypochondriacal fears with which she had entered therapy. She was dealing more effectively with her husband and had been able to obtain a part-time job, which she had felt she needed for quite some time but which she felt terrified of attempting because of her very low self-esteem.

She came to a session 10 minutes late-which was not usual for her-and said that she had been depressed all week and that she could not tie the depression to anything specific in her life. She sounded withdrawn and distant from me and again started to complain about her fears of AIDS, cancer, and heart disease. I asked if anything had happened in our last session that might have triggered the depression. She at first denied this, then admit­ ted that there was one thing: my suggestion that she seemed to want to appear "lusty." (She was being quite seductive in her behavior toward me at the time.) As we explored this, it became clear that my comment had been devastating to her, not so much because she resented this description of her (she had been raised a strict Catholic and by a mother who was intolerant of any kind of sensuality, let alone lust), but because it meant to her that I really did not understand her after all and was not at all powerful and wise.

We traced back the antecedents of her reaction to incidents during childhood in which she had been castigated by her parents for any attempt to achieve sensual pleasure. She had had an aunt who did encourage her sensual enjoyment, but the mother had disapproved of this aunt and her "wicked" ways. She also remem­bered how lonely and misunderstood she had felt all through her childhood.

This exploration resulted in the restoration of our bond, a deepening in her understanding of herself and a cessation of the hypochondriacal fears.

Some therapists (e.g., Kernberg, 1975) would encourage the therapist to interpret this idealized transference as a projection of a pathological grandiose self to the therapist. Rather than the field theoretical approach I took, in which this woman's reaction was seen as the result of the interaction between her narcissistically organized subjective world and my (momentarily) uncomprehend­ing subjective world, these therapists would probably view her reaction as an inevitable inability to maintain this projected gran­diosity in the face of my mistaken comment. This formulation could lead to an interpretation geared to the underlying feelings of inadequacy covered up by the projected grandiosity and would, I believe, have inflicted an additional injury on the client.

The self psychological view is that the grandiosity, whether owned by the patient or attributed to others, is a normal phase of development at which the client is fixated. The procedure, then, is to allow the patient to maintain the fantasy of his or her grandios­ity or the therapist's omnipotence, and to deal interpretively with the inevitable disappointment when the patient's or therapist's frailties become inescapably clear.

View of Affect

Another holistic and field theoretical aspect of self psychology is its view of affect as playing a primary role in the structuralization of self-experience (Basch, 1984; Socarides & Stolorow, 1984). Most psychoanalytic theorists have subscribed to the James-Lange theory of emotion: The human being is seen as first being aware in infancy only of a global visceral response, which is experienced as an affect only when later given a cognitive interpretation or label.

Freud generally accepted this view, seeing affect as a derivative of drive, the individual experiencing affect only when the ability to reflect is achieved. Although many analytic theorists, such as Fenichel (1945), have stressed the importance of affect-accompanied insight rather than mere intellectual insight in therapy, orthodox Freudian psychoanalysts have not, in my opinion, sufficiently appreciated the great importance of affect in human motivation, and have not known how to help many of their clients (e.g., obsessive character disorders) work through their resistances to the experiencing and expression of affect. This is one of the reasons why many alternative forms of therapy-for example, neo-Reichian, bioenergetics, and Gestalt therapy-have arisen, which have at­ tempted to devise more direct methods of increasing the client's ability to experience affect.

I have found, however, that many of the body therapy ap­proaches have erred in the other direction, that is, stressing affect expression at the expense of cognition. It is clear that both are important (Heider, 1974).

Interestingly, Virginia Demos, an important current develop­ mental researcher and theorist, has resuscitated the theory of emotion proposed by Sylvan Tomkins, a writer who has been a contributor to humanistic psychology literature, and used this theory as a basis for some important developmental research on the essential role of emotion in shaping human behavior (Demos, 1985). It was the view of Tomkins (1962, 1963) that affects are biologi­cally inherited programs that control facial muscle responses; and autonomic, blood flow, respiratory, and vocal responses. He saw the human being as inheriting the potential to feel and express a number of discrete emotions. This view is being confirmed by a growing number of researchers (e.g., Ekman, 1972; Izard, 1968, 1971; Gaensbauer, 1982).

Demos (1985) and Socarides and Stolorow (1984) take the position that emotion is essential to the structuralization of the self, which makes emotion much more important in development than previously recognized, and much more important in therapy. Ac­ cording to Socarides and Stolorow, the caretaker or selfobject helps the infant and child to differentiate between different affects, to synthesize affectively discrepant experiences, to tolerate the experiencing of affects (particularly negative affects such as fear, anger, and sadness) and to learn to use them as self-signals, and to desomatize and cognitively articulate affects. Note that this posi­tion once again puts self psychology squarely into the field theory camp, with the self-selfobject field being held responsible for the child's growing integration of affect into the self-organization, rather than seeing the organization of the self as strictly up to the individual or his or her unfolding biology.

The individual who seeks therapy has typically had experiences with caretakers that made it necessary to block off awareness and expression of some or even all affects. As Demos ( 1984) has pointed out, the expression of a negative affect by a child can elicit behavior from the caregiver that results in the child experiencing an even worse affect, for example, the child expresses anger, the parent hits the child, and the child then experiences pain and fear (of the expression of anger). When the individual has had many such experiences, the emotion is repressed or disavowed and it eventu­ally is no longer available to the individual. When circumstances are such that the defense against the emotion no longer works, the individual is apt to feel overwhelmed by emotion, for example, exploding with uncontrolled rage. The therapist's job, then, is to help the individual work through the resistances against the affects, to learn to allow him or herself to experience the affects, to integrate affective experience with a positive self-concept, and to then learn to express them in appropriate situations and in appropriate ways. This usually must first happen within the therapeutic relationship, that is, in the context of the transference.

The strong emphasis on affect also tends to help the self psychologist avoid the overintellectualism of orthodox analysis. In fact, the empathic process, the cornerstone of self psychology therapy, primarily involves the therapist's attempts to understand what the client is feeling, rather than merely what he or she is thinking.

I use the term "affect" rather than "emotion" because Stern (1984) has pointed to a whole area of affect that is different from emotion, the vitality affects. These are

elusive qualities [which are] better captured by dynamic, kinetic terms such as "surging," "fading away," "fleeting," "explosive," "crescendo," "decrescendo," ''bursting," "drawn out," and so on. These qualities of experience are most certainly sensible . . . and of great daily, even momentary, importance. It is these feelings that will be elicited by changes in motivational states, appetites, and tension.

The different forms of feeling elicited by these vital processes impinge on the organism most of the time. We are never without their presence, whether or not we are conscious of them, while "regular" affects [the emotions] come and go. (Stern, 1984, p. 54)

Later on, ''They concern how a behavior, any behavior, all behavior is performed, not what behavior is performed" (Stern, 1984, p. 157).

For some examples of mothers' responses to their infants' vitality affects, Stern gives the following:

A nine-month-old girl becomes very excited about a toy and reaches for it. As she grabs it, she lets out an exuberant "aaaah!" and looks at her mother. Her mother looks back, scrunches up her shoulders, and performs a terrific shimmy with her upper body, like a go-go dancer. The shimmy lasts only about as long as her daughter's "aaaah!" but is equally excited, joyful and intense.

A nine-month-old boy bangs his hand on a soft toy, at first in some anger but gradually with pleasure, exuberance, and humor. He sets up a steady rhythm. Mother falls into his rhythm and says, "kaaaaa­ bam, kaaaaa-bam," the "bam" falling on the stroke and the "kaaaaa" riding with the preparatory upswing and the suspenseful holding of his arm aloft before it falls. (Stern, 1984, p. 140)

I think a great many instances in which patients do not improve in therapy are due to a poor fit between the patient's style of expressing vitality affects and the therapist's style. As an example of the kind of problem that can arise, consider an emotionally constricted patient who took a long time to learn to allow himself to feel. He had had a very depressed mother, a woman who could not respond with joy, pleasure, and energetic excitement to her lively, inquisitive, bright child. He came in for a session excitedly telling me that he had spent an enjoyable weekend travelling with his girlfriend. I responded with a smile and an affirming sound to his telling me of the weekend, but he soon lost his excitement and even seemed a bit depressed. Upon exploration, he said, as he has many times in the past, that he does not feel I am really interested in him and excited about his accomplishments and growth. I was puzzled, because I had responded positively to him, and I asked him if he had noticed my pleased response. He said he had but it had not made any impact on him because it was so "lukewarm" and he even thought that I did not mean it. I then realized that, even though I had been genuinely pleased, the intensity of my response did not match his. When I suggested that this might account for his disappointment, he agreed, and we were then able to tie the incident to the devastating effect on him of the lack of response of his mother to exciting events he had related to her in childhood. Incidentally, if I had responded much more excitedly to his telling me about his weekend than he had told it to me, he probably would also have felt a lack of attunement by me and would still have had a negative emotional reaction. Although neither self psychologists nor other types of therapists have begun yet to focus widely on this category of affect, I feel confident it will come to be viewed of great importance in the development of the individual and in the empathic response of therapists to clients.

A Positive View of Humans

Freud's metatheory, as existential-humanistic psychologists such as Rogers ( 1961) have repeatedly pointed out, is an essentially pessimistic view of human beings, implying that the most we can hope for in life is control or sublimation of the human instincts of destructiveness and sexuality, which, in their natural state, press for immediate gratification.

In contrast, one finds a more positive view of humans and their potential for growth in self psychology. First, Kohut saw the human infant as having a nuclear self from birth and as being "perfect" in its own right, not as the Freudian undifferentiated blob who develops a sense of self only with the growth, through conflict between the pleasure principle and environmental demands, of the ability to think.

Second, self psychology also implies a more positive view of human nature than Freudian theory in its stress on the self and its empathic selfobject as the primary psychological configuration while, at the same time, rejecting Freud's view of biological drives of aggression and infantile sexuality as the human being's basic motivators. As Basch (1984) points out, there is a great deal of evidence from the fields of biology, ethology, and the research on sexuality by Frank Beach, that sex and aggression simply do not function as basic drives. Kohut saw hostility or aggression, as I said earlier, not as a basic instinct, but as the human being's reaction to a traumatic blow to the sense of self. And when sexuality has the driven, split-off quality seen in certain disorders, it is considered the result of a fragmentation of the self, not a pure expression of a sexual drive.

Third, because the maintenance of the self-structure is seen as the primary motive of the individual, the patient's feelings, behavior, and attitudes are viewed in a much more accepting way than they tended to be viewed by Freud. This is true even when the overt behavior is destructive to other aspects of the individual's exis­tence. The example I gave earlier of my approach with the patient who had been physically violent toward her daughter is an example of this attitude.

The prevailing view of many therapists toward their patients has been that they feel, believe, and act in the disturbed ways they do because they are avoiding or resisting or denying something else.

For example, the dependent person is seen as avoiding the fears of being separate and independent; the obsessive-compulsive is viewed as repressing oedipal strivings. The therapist's job, then, in this view, is to make the appropriate interpretation at the appropriate time so as to push or lead the patient to a realization of the more mature strivings that are being avoided. What this view of the patient implies is that the therapist knows what the truth is, not the patient; the patient is irrational, manipulative, or immature in avoiding the truth; and the patient must change in the direction the therapist views as necessary in order to become healthy.

The self psychological view is that patients are practically always trying to make the best of things with what they know of themselves and the world and that, in many instances, they are just looking for a chance to connect with someone (a selfobject) in order to fill in the missing aspects of the self and resume their psychological growth as human beings. This creates a subtle but profound switch in the therapist's attitude toward clients that has a very powerful effect on them. They sense an attitude of trust, respect, and acceptance from the therapist that, for some clients, is a necessary ingredient in their growth in therapy. Carl Rogers (1961) has also stressed the importance of the therapist's self­ awareness and positive attitude toward the client.

Of course, many people who come for therapy believe that their problems in living are caused by their badness, cowardice, or laziness, so they think they actually need to be pushed, disciplined, and treated harshly by their therapists in order to get better.

It should be noted that, although the therapist's positive atti­tude toward the client and self-awareness are considered necessary, they are not sufficient. Particularly in working with clients who have severe defects in the self-structure, such as borderlines and narcissistic personality disorders, it is also necessary for the therapist to have a comprehensive knowledge of theory, childhood development, and a deep awareness of himself or herself, particu­larly with respect to narcissistic countertransference issues.

One knotty problem is that therapists are frequently unaware that they even have significant narcissistic issues. I firmly believe that therapists who intend working with self-disordered clients need intensive, analytically oriented therapy or analysis with practitioners who are themselves knowledgeable about narcissistic issues.

People with narcissistic personality disorders or borderline pa­thology, as well as other self-disordered patients, sometimes treat therapists as if they are objects, on this earth only to minister to the client's needs, and this attitude can be a severe strain on the therapist's own narcissistic equilibrium. Other patients behave in ways that, on the surface, appear to be self-destructive. If the therapist takes the attitude that the client is still doing the best he or she can to preserve the integrity of the self, I believe the therapist will always find that behavior that appears self-destructive has a positive aim.

For example, a patient who had had some minor surgery on her foot came to my office very upset because she had had to park fairly far from my office. She was very much out of breath and her foot was hurting her. She was angry at me, which was frightening to her, and furious with herself for "wasting" quite a bit of time trying to find a place to park closer to my office. Her fury was intensified because she had worn stylish but physically unsupportive little black pumps instead of the therapeutic boot the surgeon had told her to wear. After initially helping her to understand and express her intense feelings, I suggested that, in the immediate future, at least until her foot healed, she park in my parking place in back and that I park on the street during her weekly session with me. She refused this offer, but did not know why she felt she had to. I then offered to bring her car around to the back of my building at the end of the session, but she again refused. With a different theoretical orientation, I could have interpreted this as a masochistic wish on her part, or as an attempt to deny her physical infirmity or to not allow herself to be dependent on me. I had already learned in the past that any kind of interpretation that implied she had done something "wrong'' was emotionally devastating to her, because her self-esteem was always in a precarious state. I knew that this vulnerability to narcissistic injury would be particularly true in that session because she was already feeling a great deal of self-hatred. As we explored why she did not want to allow me to help her, it became clear that to do so would be humiliating to her because it would be a clear-cut sign of her relative infirmity and "inferiority" to me. She had other, more serious physical difficulties and began to discuss the shame she felt in other situations when she could not do things that other people could do easily. I also suggested that she must have found the boot very unbecoming and that it was more important to her to look good when she came to see me than to take the best care of her foot. She also agreed to this interpretation, and soon began to calm down and to feel better. She could also then express her irritation with me for having moved my office to a location in which parking is much more difficult.

It is interesting to note that in a subsequent session, when she had to park at a 20-minute meter, I suggested that I go out to put money in the meter when the 20 minutes was up, and she proudly said that she could now allow me to do this. I believe it was my respectful attitude in the earlier session toward her seemingly self-destructive behavior that enabled her to view this later situation differently.

I should note that I am not denying that people frequently do destructive things to themselves and others. This particular patient and I learned, at an advanced point in her therapy, that she tended to accentuate her pain and minimize pleasure in her daily life. This was because she got positive attention from others, particularly me, when she was in pain, and, because she was frightened of allowing herself pleasure, found the experiencing of pain preferable to the unbearable inner emptiness she was continually defending against. She even once described her pain as "sweet." But knowing the kind of family she had grown up in and the kind of life she had led made it clear to me that the deliberate creation of pain was preferable to the experiencing of the emptiness inside; thus the pain had a positive meaning in her life.

An Emphasis on Life-Long Interdependence

Kohut had a fundamental disagreement with the Freudian view (which is the prevalent Western cultural view) that psychological growth consists of a linear movement from symbiosis and dependency to separation and independence from others. This view has achieved scientific respectability through the separation-individuation theory of Margaret Mahler (Mahler et al., 1975).

Kohut's view, which is being confirmed by some recent develop­ mental research (e.g., Demos, 1985; Stern, 1984) is that the human being is not only born with much more competence, wholeness, and social relatedness than was previously realized (except by many mothers!), but never outgrows the need for self-selfobject relation­ ships in adulthood. Kohut (1984) also felt that the emphasis upon autonomy in American society is

part of a supraordinated moral system in scientific disguise, per­fectly attuned to the prevailing values of Western civilization which elevate the capacity to be self-sufficient and shun the inclination to become dependent. ( p. 208)

People do, of course, become more differentiated and mature in the kinds of self-selfobject relationships they form, but they never stop needing approval, recognition, acceptance, and to identify with some valued ideals or group that transcends their own individual existence. This emphasis of Kohut's is very much in line with feminist critiques of male-dominated personality theories, which stress traditional "male" values of competition, independence, and separateness at the expense of "female" values of relation­ ship, cooperation, and interdependence (Gilligan, 1982). (It should be noted that Kohut's analysis of Western culture does not include the insights of feminism; he does not indicate that it is mostly men who are likely to espouse the values of independence, while es­ chewing dependence, and sometimes appears to be advocating the view that the rightful role of women is to serve selfobject needs of men!)

I also believe that this emphasis on interdependence is a much needed outlook in a world that is rapidly being destroyed by those who do not see the essential interdependence of all life. This is another example of the field-theoretical nature of self psychology.

Cultural Issues, Ontology, and Infant Development

As Rollo May (1958) has pointed out, the philosophical explora­tion of existential issues has been taking place at least since the time of Socrates. The major studies have been done, however, in the past 120 years, since Kierkegaard first published his strong objections to the arid technical rationalism he saw in Western culture.

Some of this exploration of existential issues has concerned the problems of human existence that affect all human beings: the necessity of dealing in some way with the knowledge of the finite­ ness of our lives; the incomplete control we have over existence; our limited but still considerable ability to act; the very nature of being (ontology); our difficulty in feeling connected to other human beings; and the responsibility we have toward the world because of our great power to affect the world, both in good and evil ways. Some of the same philosophers, and other writers as well, have examined these same existential issues as they affect modern human beings, particularly Western human life (e.g., Marx, Nietzsche, May, Kierkegaard, Sartre): the social fragmentation, alienation, rootlessness, despair, meaninglessness, emptiness, and tremendous destructiveness that characterize much of modern Western life.

Christopher Lasch (1979, 1984) has presented a very cogent analysis of current existential issues in political, economic, and psychological terms, using the insights of object relations theory. He sees the major crisis today as the inability of human beings to find meaning in their lives because modern life offers no satisfactory ways of resolving the universal trauma of separation, which begins at birth and continues throughout life. He says,

Selfhood presents itself, at first, as a painful separation from the surrounding environment, and this original experience of over­ whelming loss becomes the basis of all subsequent experiences of alienation, of historical myths of a lost golden age, and of the myth of the primary fall from grace, which finds its way into so many religions. (Lasch, 1984, pp. 163-164)

It should be noted that Lasch is defining the self as a self in separation, similarly to the way it is defined by Mahler et al. ( 1975). He says that previously the fact of separation was successfully dealt with first by accepting it, and second by certain creative and meaningful enterprises to which humans could devote themselves (e.g., religion, art, and work). But work, prior to the great industrialization of the world's economies, was more satisfying because people could take pride in and see concretely the results of their labor. Now the emphasis is on assembly line work, the hierarchical corporate structure, and service jobs, which usually do not give the satisfaction people presumably once got from their work. In addi­tion, families have become increasingly isolated and fragmented because of the fact that jobs are performed away from home, so children are raised with limited contact with one or even both parents. The typical home does not include grandparents, older siblings, servants, and other people who formerly lived in or near the home and helped to take care of the young. He also points to the great stress on mass consumption in highly industrialized societies, which leads people to the addictive illusion that the acquisition of products advertised in the mass media will lead to a sense of satisfaction.

Lasch sees modern art, music, mystical religions, and other aspects of modern life, which are designed to deal with the dilemma of separation, as empty attempts "to restore the illusion of oneness without any acknowledgment of an intervening experience of separation" (p. 164). ·

But, as I pointed out in the previous section of this article, the self according to Kohut, recent developmental research, and femi­nist psychology, is essentially a self in relation (Surrey, 1984). Therefore, we never really achieve or should hope to achieve complete separation or independence. In his theoretical treatment of separation, Lasch makes the same mistake made by psychoanalytic theory, including the object relations theories on which Lasch relies in his analysis.

Lasch does attempt to deal with the issue of narcissism, which is a crucial psychological factor in current cultural problems. He again relies on the Freudian notion of narcissism, as developed by Kernberg, which postulates that narcissism in the adult is pathological and primitive, a regressive attempt to compensate for underlying feelings of impotence and inadequacy by the creation of a pathological grandiose self. In contrast, Kohut's depiction of narcissism, as I pointed out earlier, is that it is universal and that no one, no matter how psychologically mature, ever outgrows the need for mirroring of one's goals and intentions, and for identifica­tion with something powerful and meaningful outside oneself. The difference between the narcissism of the immature adult, on the one hand, and the child and immature adult, on the other, is in the complexity, differentiation, and frequency of the needed narcissistic gratification.

A combination of Kohut's conception of narcissism, plus the devel­opmental theory of Daniel Sternwill, I believe, lead to a more complete analysis of what ails Western culture, and will provide a much-needed dimension to the ontological issues with which Kierkegaard, Nietzsche, Heidegger, Sartre, and other existential philosophers have dealt. This is because self psychology starts with the primary self-selfobject field, stresses the need for selfobject ties throughout life for everyone, no matter how mature and well-integrated, and can account for the rage, despair, and destructiveness that result when attempts to achieve selfobject ties fail.

An example of an analysis congruent with Self Psychological theory is Alice Miller's discussion (1983) of the extremely negative childhood experiences of Adolph Hitler, the other Nazi leaders, and the German people and how these experiences led to the terrible atrocities perpetrated by Nazi Germany. There were, of course, many other causes: the humiliation Germany suffered because of the Treaty of Versailles after World War I, the latent anti-Semitism that has existed in Europe for centuries, the misery and deprivation the German people experienced during the Great Depression, the greed of the heads of the great German corporations and their fear of Communism, and so on. But understanding of the Nazi phenomenon is incomplete without a thorough examination of the widespread rigidity, brutality, and emotional constriction that generally characterized the child-rearing practices in Germany, and the specific childhood experiences of leading Nazis.

With respect to ontological issues of human beings, I now turn to a brief discussion of Daniel Stern's work. Stern does not identify himself as a self psychologist, but his work is certainly compatible with self psychology and is already being used to buttress the theoretical scaffolding of self psychology (Lichtenberg, 1987; Terman, 1986).

Stern has postulated four senses of self, all of which start in infancy, but continue to develop throughout the life cycle and exist concurrently. These are the sense of an emergent self (awareness of the process of the core self in formation); the core self (awareness of the self and others as coherent, physical entities); a subjective self (awareness of the self and others as having inner, psychological lives, which can be shared); and a verbal self.

The second, the sense of a core self, and the third, the sense of a subjective self, are the senses of self that are most relevant to the ontological issues that face all human beings. The sense of a core self, which starts to form at around two months of age, consists of the individual's experience of him or herself as being a single, coherent, bounded physical entity, as being an agent (who can initiate events, determine their direction, end them, and so on), and who has distinct affective experiences and a unique history (memory). At the same time this sense of self is forming, the infant begins to experience others as also being single, coherent physical entities, and

the subjective social world is altered and interpersonal experience operates in a different domain, a domain of core-related ness. This developmental transformation of creation occurs somewhere be­ tween the second and sixth months of life, when infants sense that they and mother are quite separate physically, are different agents, have distinct affective experiences, and have separate histories. (Stern, 1984, p. 27, italics in original)

The sense of a subjective self, which starts to form between the seventh and ninth months of life, enables the infant to realize that he or she has a mind, with internal subjective states (but not yet verbal!), that others also have minds, with subjective internal states, and that a possibility for intersubjectivity (that is, sharing of internal states) between infant and caretaker now exists. This is the domain of intersubjective relatedness.

It is very important to note that both of these senses of self go on out of awareness and are not registered verbally. Stern (1984) says, "In fact, the experience of intersubjective relatedness, like that of core-relatedness, can only be alluded to; it cannot really be described (although poets can evoke it)" (p. 27). And, I would add, so can certain existential philosophers. Thus I believe that the onto­ logical issues dealt with by the great philosophers relate to these two senses of self: core and subjective. Without firm senses of core and subjective self, one is apt to feel powerless to act, to have a sense of unreality about one's very existence (the lack of the "I am" experience; May, 1958), to feel disconnected from one's past, to feel alone emotionally, and misunderstood by others, and so on. I also believe that philosophers such as Nietzsche and Kierkegaard, who showed signs of eccentricity and emotional illness, were acutely aware of these problems precisely because they had deficits in these senses of self. (They certainly did not have severe problems in the sense of a verbal self, Stern's fourth sense of self; they communicated very well, at least in their writings.)

My attempt to tie their ontological sensitivity to childhood developmental factors is not in any way a denial of the relevance of their insights to general Western cultural issues. It is usually the case that a great artist or thinker becomes aware of certain issues precisely because of an acute sensitivity to existential dilemmas other people face but cannot articulate. Nietzsche had a weak, emotionally disturbed father who died when Nietzsche was very young. He was raised by domineering women-an older sister, his mother, and a grandmother- and was very dependent during his whole life on his sister, who has been described as a manipula­tive, conniving person (Hollydale, 1965). He was unaccepting of his own feelings of weakness and inadequacy and wrote with contempt of weak, needy people. He never married and had only one brief romantic involvement, which was broken up by his sister. He suffered from migraine headaches, depression, great feelings of loneliness, and other psychosomatic ailments his whole life. He stressed core self issues in his writings-the need for power and the fragmentation of the modern self-and was an extreme non­ conformist. During the last 10 years of his life he was delusional, believing he was God. He thus attempted to shore up a shaky core by endowing himself with ultimate power.

Kierkegaard's writings focused more on emotional alienation in his writings and he appears to have had more difficulty with his sense of a subjective self, because the need to be understood and emotionally connected to others is essential to this sense of self.

When any attempt is made to explain the profound insights of Nietzsche and Kierkegaard in terms of their life experiences and psychological dynamics, their admirers object, as if such explanations mean discounting these insights. Thus May (1958) wants to attribute Nietzsche's mental illness to his "intensity of self­ consciousness" (p. 29). I think Nietzsche was able to articulate issues shared by all people-core self issues-because of his great self-awareness, his articulateness, and the difficulties he had with his sense of existence.

Why is it, then, that so many people have been able to identify with the core and subjective self issues that Nietzsche and Kierkegaard wrote about? The reason is because these self senses are, as I said previously, not sequentially formed stages that are solidified once and for all during a normal infancy. Rather, they are lifelong agendas. Stern says,

Because all senses of the self, once formed, remain active, growing, subjective processes throughout life, any one of them is vulnerable to deformations occurring at any life point. Similarly, since life issues such as autonomy or control are issues for the life span, they too are vulnerable at any life point. (p. 260)

Thus a person who has a relatively coherent sense of a core self may be faced with overwhelming difficulties around freedom of movement and a sense of coherence later than infancy.

Stern gives the example of a patient whose major complaints in therapy centered around feelings of weakness, an inability to initiate actions and to cope with life by herself. These current complaints centered around a feeling of paralysis in her law career. Her difficulties seemed to stem from ages 8-10 when she was bedridden with rheumatic fever. She had been ordered to stay in bed, not even to walk to the window to look out. Even if she tried to go somewhere, she became too fatigued physically. For anything physical to happen-to go upstairs or downstairs, to get a book, to open a window-she had to wait for someone else to come and do it. Stern says that her adult problems in feeling coherent and experiencing herself as having a sense of power and autonomy seemed to stem from an inability to work through the emotional effects of the powerful impingements on her sense of core self during her illness.

Anyone who has had to deal frequently with huge traffic jams when going to and from work, or has lived through a large earth­ quake, knows how blockage of free movement and disruptions of groundedness can threaten one's sense of self!

An example of someone who had problems working through threats to subjective self sense is a middle-aged woman who was extremely upset because the man she had lived with for 10 years had suddenly, without any notice, left her for his ex-wife. A year after he left she was still so upset that she was practically para­lyzed in her life. She had been a vivacious, articulate, active person with many interests, including painting, and many friends. She was now avoiding her friends, found it unbearable to go near her easel, and was continually thinking of her ex-lover with rage and grief. She was also hating herself for continuing to dwell on these negative feelings and for being unable to resume her life.

As we explored her difficulties, it became clear that what she was so upset about was not so much the fact of his leaving, but how he left: suddenly, with no prior discussion about the inner conflicts she realized he must have been experiencing for a long time before his departure. The way he left cast a retroactive pall over the whole relationship. "I don't think he ever really loved me. I don't even think he ever knew me. If he did, he would have come and talked with me about his feelings toward his ex-wife. I would have let him go if he had told me that would be best for him."

The "narrative point of origin" of this woman's difficulty, the key therapeutic metaphor, was a sudden and traumatic disruption in her life that had occurred during childhood, and the inability of her parents and older siblings to understand her emotional turmoil and to help her deal with it.

Thus Kohut and Stern can add a much-needed dimension to the political, social, historical, and economic analyses of our grave planetary problems by deepening our understanding of the devastating effects on the self of the widespread lack of grounding, stability, and emotional connectedness. Their insights can also show us how these blows to the self produce emptiness, despair, and rage; and lead to restitutive attempts of self-aggrandizement at the expense of the environment and other peoples, and other kinds of violence to oneself and others.

Kohut also felt that the survival of humanity will come only with an increased emphasis on empathy instead of on the truth-and­ reality focus so prevalent today. I do not think that the importance of empathy-and the lack of it-can be overstated in today's world. Only people who do not identify with the victims of a nuclear attack can speak dispassionately of "limited" nuclear war. Only those who do not allow themselves to imagine what it is like to have a starving child in a Third-World country can speak rationally of political and military strategies that ignore the oppression and grinding poverty there.

Self Psychology Combines Determinism and Free Will

Kohut believed that his work with adult patients showed that the development of the self is largely determined by transactions between the primary caretaker and the infant/young child, and that studying the vicissitudes of these transactions could explain a great deal of adult behavior. This view is a deterministic one and different from that held by humanistic psychologists, such as Rogers and Perls, who have tended to deemphasize the importance of the past. Although as a Gestalt therapist I used to see the past as important only when focused upon as the repository of "unfinished business," I now realize that the client's view of the therapeutic relationship, particularly in narcissistic and borderline disorders, just does not make sense unless the individual's past relationships with primary caretakers are considered.

Because of the emphasis in Freudian theory on biological no­tions of a preprogrammed unfolding of psychosexual stages and of instinctual drives, it has been difficult for orthodox analysts to conceive of the possibility of free will. Kohut is also unusual among analytic theorists, however, in that he saw free will as just as much a fact as determinism in psychological functioning because humans can experience freedom of choice introspectively. His very definition of the self implies free will in that, whatever its history of development, it can become independent from its past, is oriented toward the future, and is free to pursue its own creatively chosen aims. In the person with a cohesive, resilient, vital, and harmonious self, the self becomes an independent center of initiative, a unity that tries to follow its own unique course. Indeed, a diagnostic sign of a cohesive, harmonious, vital self-structure is the individual's awareness of, and striving toward, realistic goals that are in accord with a sound system of values and ideals. Thus humans are seen in self psychology as motivated by the future as well as by the past.

It should also be noted that the self can become caught up in conflict and lose its independence; therefore, the evolution of the self is a lifelong process, not a stage that is achieved and then maintained forever (Horner, personal communication, 1987). In this regard, Stern's work (1984), described in the previous section, is relevant here too.


Although Freud's concept of the superego certainly stressed the importance of internalized (usually introjected) moral precepts and ideals in human functioning, Kohut made the presence of firm, coherent, mature ideals part of the basic strength or resiliency of the individual. These ideals are part of the idealizing pole of the bipolar self. Significantly, he does not specify which values he thinks people should have-that, he believed, was up to the individual. He did believe, however, that these values and ideals are important ingredients in a cohesive, resilient self, as important as the individual's ambitions, which are part of the other pole of the self.

For example, one client of mine, a handsome, charming, intelli­gent man, was a successful lawyer. He worked in an area of law that he saw, however, as meaningless, and his view of the worth of his work made it impossible for him to feel proud of his success. His lack of firm values and ideals also made it impossible for him to devote himself to something else that he would see as of greater value. In an attempt to fill an inner void, he spent much of his life frantically trying to fill himself up with drugs, dating a succession of beautiful women (none of whom could he let himself love), and doing dangerous things like racing cars. None of these activities gave him any real pleasure or satisfaction.

At the opposite extreme was a self-sacrificing minister who had a rigid set of ideals that excluded any possibility of striving for personal goals or gratification. He was forced into therapy, which he considered an "indulgence," only because he had had a severe depressive episode after having extended himself beyond his en­ durance while doing missionary work.

In contrast to these two men was another patient-also a successful lawyer-who was able to support himself with his high intelligence and his analytic abilities to the point that he had created a fairly good practice, but not nearly as lucrative as that of the attorney discussed above. He worked, however, within an area of the law that he felt was socially very important. Thus, although he was actually not as financially successful as the other attorney, the presence of firm ideals and values enabled him to feel much more fulfilled.

Emphasis in Helping Client Develop Core Program Rather than Adapt

Kohut (1982) believed that psychoanalytic theorists such as Franz Alexander and Heinz Hartmann, with concepts such as "drive," "dependency" and "adaptation," have made psychoanalysis "less of a science and more of a moral system, and psychoanalysis as therapy . . . less of a scientific procedure . . . and more an educa­tional procedure . . . toward which the patient is led" (p. 399). In this he was very much in agreement with Rogers (1951), who made the same critique of the usual attitude of therapists toward clients. Kohut wrote that "knowledge values and independence value have been the leading values of the psychoanalyst, and . . . . They have guided him toward selective perception and selective action within the psychological field." The unacknowledged and unquestioned presence of these values have interfered with "the analyst's ability to allow his analysands to develop in accordance with their own nuclear programme and destiny" (Kohut, 1982, p. 399).

Advantages Of Self Psychology Over Existential-Humanistic Therapies

If Kohut merely mirrored many of the ideas and beliefs of existential-humanism, there would be little advantage to his the­ory. However, I believe there are many advantages to self psychol­ogy, advantages that originally stimulated me to attempt to inte­grate self psychology with Gestalt therapy (Tobin, 1982).

Theory of Development

I think that a relatively complete theoretical system, such as self psychology, with its theory of development, has a distinct advantage over those systems that do not incorporate a theory of development and an emphasis on learning about adult patients from developmental observations of infants and young children. Gestalt therapy as well as other forms of existential-humanistic therapy with which I am familiar, suffer from not having an adequate theory of human development. Without such a theory, based on observations and research, much of the behavior we observe in our offices does not make sense. The narcissistically disturbed person, with his or her seemingly unreasonable, insatia­ble needs for approval, love, and acceptance and an unwillingness to respond to the therapist as a real, separate person is puzzling to the therapist who is unaware of the primitive, archaic selfobject needs of such people. In fact, much of the subtle hostility with which therapists typically describe and treat "difficult" patients is due to the narcissistic injuries they themselves suffer from such patients.

A Theory of Missing Psychological Structure

Rogers and other existential-humanistic theorists seem to as­sume the presence of a cohesive self in which psychological prob­lems are due to conflicts between different subselves (as in Gestalt therapy's topdog-underdog conflicts) or to the disavowal of aspects of the self that do not fit into the self-concept (as in person-centered therapy). Kohut sees many patients as suffering from a lack of a cohesive self and evolved a method of treatment that heals, strengthens, and vitalizes the self. In addition, because of his developmental understanding and his emphasis on transference (particularly selfobject transference) and countertransference (particularly narcissistic countertransference), the self psycholo­gist is less apt to make the mistakes that are frequently made by therapists whose theories do not give the understanding that would enable them to deal effectively with the self-disordered patient. Also, when the self psychologist does unwittingly inflict a narcissistic injury on the patient, such a therapist is theoretically and personally prepared to discover what he or she did to cause this wound, to tie this experience to genetic material and to own his or her error. This process, referred to as "optimal responsiveness" (Bacal, 1985) is the single most healing aspect of self psychology.


Existential-humanistic psychologists have tended to eschew diagnosis because of the negative therapeutic effects on patients sometimes caused by labeling them. This is particularly true when the therapist treats the patient as a sick person in need of the ministrations of the superior, objective, healthy therapist. Al­ though the use of diagnosis is frequently destructive, I believe that the egalitarian, humane atmosphere created in self psychology therapy decreases the possible negative effects of the formation of diagnoses. I also believe, on the other hand, that diagnosis is sometimes absolutely necessary, for example, when distinguishing between certain narcissistic personality disorders and borderline conditions. It should be noted, however, that at least one self psychology theorist, Robert Stolorow (Stolorow et al., 1987), is questioning the validity of the borderline diagnosis as a separate, distinct category. It could very well be that self psychologists will create their own diagnostic system, one not based on the assump­tions of Freudian theory.

Emphasis on Considering the Past

As I said earlier, many existential-humanistic theorists have downplayed the importance of considering the past in psychother­apy. Part of the reason for this attitude was an understandable rebellion against what frequently turned into, in Freudian analy­sis, an interminable digging into the past and neglect of the individual's present strengths. Even those theorists who did work with the past, such as Fritz Perls, regarded it as "unfinished business" that must be worked through before the person can live totally in the present. Kohut had a different perspective; he felt that many therapy patients' self-structures are weak precisely because they feel cut off from their pasts. Thus growth consists, at least partially, in helping them reconnect with their pasts.

Kohut also contrasts this view of the past to the orthodox Freudian view of the past, using as an example the attempts of the narrator in Marcel Proust's Remembrance of Things Past to reunite his fragmented self-structure by recalling childhood memories. Kohut (1977) says,

The Proustian recovery of childhood memories constitutes a psycho­ logical achievement significantly different from the filling in of infantile amnesia, which as Freud taught us, is the precondition for the solution of structural conflicts and thus for the cure of a psycho­ neurosis. The Proustian recovery of the past is in the service of healing the discontinuity of the self. (pp. 181-182)

Emphasis on Transference and Countertransference

Existential-humanists, perhaps because they believe the rela­tionship between therapist and client itself is the essential, cura­tive factor, have tended to ignore or downplay transferential and countertransferential aspects of the relationship. Labeling some­ thing as transference can imply that the client's behavior toward the therapist is unrealistic and only due to the client's distorted view, and that the therapist's view is the correct one. Self psychology's conceptualization of transference avoids this view. Instead of seeing transference as the client's distorted perception of the therapist, the reasons for the client's perception- his or her internal "reality"-are explored and dealt with empathically, and the resonances from the past are explored. Transference is thus seen as the patient's way of organizing the world, including the "world" of the consulting room, rather than as a distortion of reality (Stolorow & Lachmann, 1984/85).

Method of Training and Therapy

In preparing for this article and for a symposium on Carl Rogers I reread some sections of Rogers's Client-Centered Therapy and On Becoming a Person, books I had not read for many years. I was amazed at the extent to which he captured many of the essential features of self psychology as long ago as 1951: field theory, phenomenology, the empathic stance in therapy, holism, and so on. The little that I saw of Rogers's actual work-for example, his interview of "Gloria" in the series of three therapeutic interviews Everett Shostrom filmed-revealed him as an outstanding psychothera­pist who exemplified, in his own person, the existential-humanistic therapeutic attitude.

However, I do not believe he was able to provide a deep enough theory of personality development and psychotherapeutic change that would enable trainees to achieve his own high degree of effectiveness. I have already spelled out the reasons for this lack of a theory of transference and countertransference: insufficient understanding of the importance of exploring the patient's past, particularly in those with personality disorders; and incomplete conceptualization of the self-selfobject configurations that make it possible for a person to take the humane, caring, accepting, affirm­ ing attitude toward clients Rogers appeared able to take instinct­ively. Most of us are unable to feel "unconditional positive regard" for many of our narcissistically disturbed clients without hard work on understanding our own narcissistic issues. For that kind of understanding, I believe that self psychology theory and therapy will prove to be very helpful.


All too often one gets the impression from reading Kohut that his theoretical and philosophical ideas sprang spontaneously from his own mind and were not based on the ideas of others. For example, Winnicott's concept of the "subjective object" is very similar to Kohut's concept of the "selfobject," but Kohut makes no mention of Winnicott as a contributor to his own theories. Recognizing this, he attempts to explain this neglect as due to his aim, at least in his 1977 book, The Restoration of the Self, of "an attempt to struggle toward greater clarity" rather than "scholarly completeness" (Kohut, 1977, p. xx). Similarly, Rogers has made extremely impor­tant contributions to the use of empathy in therapy and, although Kohut earlier defined empathy differently from Rogers in his last article (Kohut, 1982), he seemed to be taking an identical position (Chessick, 1985), yet there is no mention of Rogers. Rollo May (1958) 30 years ago gave a profound description of the compartmentaliza­tion, inner breakdown, subject-object split (field theory), and reliance on "technical reason" that characterize modern Western hu­man beings. In fact, May shows that many of these ideas go back to Nietzsche and Kierkegaard, who lived and wrote in the nineteenth century! But one looks in vain throughout Kohut's works for references to Rogers, Maslow, May, Nietzsche, Kierkegaard, and so on.

This is not simply a matter of giving credit where credit is due. Rather, many of the existential-humanistic ideas that underlie self psychology's theory and way of working have been stated more succinctly, systematically, and in greater depth by these earlier existentialists and humanistic psychologists, and it would be worthwhile for self psychology theorists to study them and bring their ideas into self psychology theory.

One self psychologist who has been doing some of that is Robert Stolorow who, in collaboration with George Atwood, has written two books in which he brings in discussions of existential phenom­ enology (Atwood & Stolorow, 1984: Stolorow & Atwood, 1979). Stolorow has also written a brief but cogent article on the similar­ities between Kohut's and Rogers's techniques, therapeutic pro­cesses, and notions of the ideal outcomes of therapy (Stolorow, 1976). Perhaps Stolorow knew of Rogers's work because he is a clinical psychologist, whereas most of the self psychology analysts practicing and writing today are physicians.

Another problem I see in self psychology is that in their attempt to create a consistent psychoanalytic theory on intersubjectivity, Kohut and the other self psychology theorists do not sufficiently address the interpersonal. Stolorow, in his attempts to be theoret­ically consistent, for example, claims that selfobjects are not people, but are functions performed by others that have an effect on psycho­ logical structure (1986). Guntrip's criticism of object-relations the­ory could therefore also apply to self psychology:

It does now have a truly psychodynamic theory of the development of the individual ego in personal relations; but not of the complex fact of the personal relationship itself between two egos . . . the theory has not yet properly conceptualized Buber's [mutual] "I-Thou" relation, two persons being both ego and object to each other at the same time, and in such a way that their reality as persons becomes, as it develops in the relationship, what neither of them would have become apart from the relationship. This is what happens in good marriages and friendships . . . It raises the ques­tion Winnicott (1967) says psychoanalysis has not yet faced: "What is life about, apart from illness?" (Guntrip, 1969, p. 389)

Existential-humanistic psychologists such as Rogers (1961), Yalom (1980), and Bugental (1965, 1987), who have addressed themselves more fully to interpersonal relationships, could bring something important to self psychology.


In this article I have attempted to discuss the strengths and weaknesses of orthodox psychoanalysis and existential-humanistic psychology, to show how psychoanalytic self psychology incorpo­rates many of the strengths of each while avoiding many of the weaknesses, and to give some flavor of the philosophical assump­tions that underlie self psychology. I have also made an attempt to point to some lacks in self psychology that I believe existential­ humanistic psychology could rectify. I tried to integrate some of the theoretical concepts with the philosophical assumptions, but, be­ cause of the limited scope of the article, was unable to bring in more than some of the very basic concepts, such as the self, the selfobject, empathy, and optimal responsiveness. I want to emphasize that I have discussed these concepts in a very cursory way. There are also a number of other important theoretical and practical aspects of self psychology that I did not even mention: the concept of trans­ muting internalization, selfobject transference, selfobject counter­ transference, and so on.

Like any deep, expansive, systematic body of knowledge, self psychology is very complex and not easy to master. Those readers who would like to explore self psychology more fully would do well to study one or more of the excellent basic texts that have been written: White and Weiner (1986), Elson (1986), Chessick (1985), or Stolorow and Lachmann (1980). An excellent recent book, which modifies and expands on self psychology, is Stolorow et al. (1987). Kohut's writings, although difficult reading for some, are indis­pensable; the best introduction is The Restoration of the Self (Kohut, 1977). An excellent outline of self psychology theory is by Kohut and Wolf (1978).

An understanding that is both affective as well as cognitive is, of course, even more difficult to arrive at. For that, I believe personal therapy is necessary.


Adler, A. (1927). The practice and theory of individual psychology. New York: Harcourt.

Allport, G. W. (1961). Pattern and growth in personality. New York: Holt, Rinehart & Winston.

Atwood, G. E., & Stolorow, R. D. (1984). Structures of subjectivity: Exploration in psychoanalytic phenomenology.

Hillsdale, NJ: Analytic Press. Bacal, H. A. ( 1985). Optimal responsiveness and the therapeutic process. In A. Goldlberg (Ed.), Progress in self psychology (Vol. 1; pp. 202-227). New York: Guilford.

Balint, M. (1968). The basic fault. London: Tavistock.

Basch, M. (1980). Doing psychotherapy. New York: Basic Books.

Basch, M. (1983). Empathic understanding: A review of the concept and some theoretical considerations. Journal of the American Psychoanalytic Association, 31, 101-126.

Basch, M. (1984). Selfobject and selfobject transference: Theoretical implications. In A. Goldberg & P. Stepansky (Eds.), Kohut's legacy: Contributions to self psychology. Hillsdale, NJ: Analytic Press.

Binswanger, L. (1963). Being-in-the-world. New York: Basic Books.

Boss, M. (1963). Psychoanalysis and daseinanalysis. New York: Basic Books.

Buber, M. ( 1958). I and thou. New York: Scribner's.

Buber, M. (1965). The knowledge of man. New York: Harper & Row.

Bugental, J.F.T. (1965). The search for authenticity: An existential-analytic approach to psychotherapy. New York: Holt, Rinehart & Winston.

Bugental, J.F.T. (1987). The art of the psychotherapist. New York: Norton. Chessick, R. D. (1985). Psychology of the self and the treatment of narcis­sism. Northvale, NJ: Jason Aronson.

Demos, E. V. (1984). Empathy and affect: Reflections on infant experience. In J. Lichtenberg, M. Bernstein, & D. Silver (Eds.), Empathy (Vol. 2). Hillsdale, NJ: Analytic Press.

Demos, E. V. (October, 1985). Affect and the development of the self- A new frontier. Paper presented at the Self Psychology Conference, New York.

Ekman, P. (1972). Universal and cultural differences in facial expression of emotion. Nebraska Symposium on Motivation, 19, 207-283.

Elson, M. ( 1986). Self psychology in clinical social work . New York: Norton. Erikson, E. (1964). Childhood and society. New York: Norton.

Fairbairn, W.R.D. (1952). An object relations theory of the personality. New York: Basic Books.

Fenichel, 0. (1945). The psychoanalytic theory of neuroses. New York: Norton.

Freud, S. (1938). An outline of psycho-analysis. London: Hogarth.

Friedman, M. (1985). Healing through meeting and the problematic of mutuality. Journal of Humanistic Psychology, 25, 7-40.

Gaensbauer, T. (1982). The differentiation of discrete affects. Psychoana­ lytic Study of the Child, 37, 29-66.

Gilligan, C. (1982). In a different voice. Cambridge, MA: Harvard Univer­sity Press.

Goldberg, A. (Ed.). (1978). The psychology of the self: A casebook. New York: International  Universities  Press.

Goldberg, A (Ed.). (1980). Advances in self psychology. New York: Interna­tional Universities Press.

Greenberg, J. R., & Mitchell, S. A. (1983). Object relations in psychoana­lytic theory. Cambridge, MA: Harvard University Press.

Guntrip, H. ( 1969). Schizoid phenomena, object relations and the self New York: Basic Books.

Guntrip, H. (1971). Psychoanalytic theory, therapy and the self New York: Basic Books.

Hartmann, H. (1964). Essays on ego psychology. New York: International Universities Press.

Heider, J. (1974). Catharsis in human potential encounter. Journal  of Humanistic Psychology, 14( 4), 27-47.

Hirsch, I. (1987). Varying modes of analytic participation. Journal of the American Academy of Psychoanalysis, 15, 205-222.

Hollydale, R. J. (1965). Nietzsche. London: Ark Paperbacks.

Horney, K. (1950). Neurosis and human growth. New York: Norton.

Izard, C. E. (1968). The emotions and emotion constructs in personality and cultural research. In Handbook of Modern Personality Theory. Chicago: Aldine.

Izard, C. E. (1971). The face of emotions. New York: Appleton-Century­ Crofts.

Jung, C. G. (1956). Two essays on analytical psychology. New York: Meridian. Kahn, E. (1984). Heinz Kohut and Carl Rogers: A timely comparison. American Psychologist, 40, 893-904.

Kernberg, 0. (1975). Borderline conditions and pathological narcissism. New York: Jason Aronson.

Kernberg, 0. (1976). Object relations theory and clinical psychoanalysis. New York: Jason Aronson.

Kohut, H. (1959). Introspection, empathy and psychoanalysis. Journal of the American Psychoanalytic Association, 7, 459-483.

Kohut, H. (1966). Forms and transformations of narcissism. Journal of the American Psychoanalytic Association, 14, 243-272.

Kohut, H. (1971). The analysis of the self. New York: International Univer­sities Press.

Kohut, H. (1977). The restoration of the self. New York: International Universities  Press.

Kohut, H. (1982). Introspection, empathy, and the semi-circle of mental health. International Journal of Psychoanalysis, 63, 395-407.

Kohut, H. (1984). How does analysis cure? Chicago: University of Chicago Press.

Kohut, H. ( 1985). Self psychology and the humanities: Reflections on a new psychoanalytic approach. New York: Norton.

Kohut, H., & Wolf, E. (1978). The disorders of the self and their treatment: An outline. International Journal of Psychoanalysis, 59, 413-425.

Lasch, C. (1979). The culture of narcissism. New York: Norton. Lasch, C. (1984). The minimal self New York: Norton.

Lecky, P. (1945). Self-consistency. New York: Island Press.

Lichtenberg, J. D. (1987). Infant studies and clinical work with adults. Psychoanalytic Inquiry, 7, 311-330.

Mahler, M. S., Pine, F., & Bergman, A. (1975). The psychological birth of the human infant. New York: Basic Books.

Malcom, J. 0981). Psychoanalysis: The impossible profession. New York: Knopf.

Masterson, J. (1976). Psychotherapy of the borderline adult: A developmen­tal approach. New York: Brunner/Maze!.

May, R. (1958). The origins and significance of the existential movement in psychology. InR. May, E. Angel, & H. F. Ellenberger (Eds.), Existence: A new dimension in psychiatry and psychology (pp. 3-36). New York: Basic Books.

Miller, A. ( 1983). For your own good: Hidden cruelty in child-rearing and the roots of violence. New York: Farrar, Straus, Giroux.

Ornstein, P. H., & Ornstein, A. (1980). Formulating interpretations in clinical psychoanalysis. International Journal of Psychoanalysis, 61, 203-211.

Ornstein, P. H., & Ornstein, A. ( 1985). Clinical understanding and explain­ ing: The empathic vantage point. In A. Goldberg (Ed.), Progress in self psychology (Vol. 1). New York: Guilford.

Perls, F. (1947). Ego, hunger and aggression: The beginning of Gestalt therapy. New York: Random House.

Perls, F. ( 1969). In and out the garbage pail. New York: Bantam.

Perls, F., Hefferline, R., & Goodman, P. (1951). Gestalt therapy: Excitement and growth in the human personality. New York: Julian Press.

Rogers, C. R. (1951). Client-centered therapy. Boston: Houghton Miffiin. Rogers, C. R. (1961). On becoming a person: A therapist's view of psycho­ therapy. Boston: Houghton Mifflin.

Shane, M., & Shane, E. (1980). Psychoanalytic developmental theories of the self: An integration. In A. Goldberg (Ed.), Advances in self psychology. New York: International Universities Press.

Shane, M. (1985). Change and integration in psychoanalytic developmental theory. In C. F. Settlage & R. Brockbank (Eds.), New ideas in psychoanalysis. Hillsdale, NJ: Analytic Press.

Socarides, D. D., & Stolorow, R. D. (1984). Affects and selfobjects. Annual of Psychoanalysis, 12, 105-119.

Stern, D. N. ( 1984). The interpersonal world of the infant. New York: Basic Books.

Stolorow, R. D. (1976). Psychoanalytic reflections on client-centered ther­ apy in the light of modern conceptions of narcissism. Psychotherapy: Theory, Research and Practice, 13, 26-29.

Stolorow, R. D. (1986). Critical reflections on the theory of self psychology: An inside view. Psychoanalytic Inquiry, 6, 387-402.

Stolorow, R. D., & Atwood, G. E. (1979). Faces in a cloud: Subjectivity in personality theory. New York: Jason Aronson.

Stolorow, R. D., Brandschaft, B., & Atwood, G. (1983). Intersubjectivity in psychoanalytic treatment, with special reference to archaic states. Bulletin of the Menninger Clinic, 47, 117-128.

Stolorow, R. D., Brandchaft, B., & Atwood, G. E. (1987). Psychoanalytic treatment: An intersubjective approach. Hillsdale, NJ: Analytic Press. Stolorow, R. D., & Lachmann, F. (1980). Psychoanalysis of developmental arrests: Theory and treatment. New York: International Universities Press.

Stolorow, R. D., & Lachmann, F. (1984/1985). Transference: The future of an illusion. Annual of Psychoanalysis 12 I 13, 19-37.

Sullivan, H. S. (1953). The interpersonal theory of psychiatry. New York: Norton.

Surrey, J. (1984). The self-in-relation: A theory of women's development. Wellesley, MA: Stone Center for Developmental Services and Studies at Wellesley College.

Terman, D. M. (October, 1986). Optimum frustration, structuralization, and the therapeutic process: A model whose time has passed. Paper presented at a conference on Clinical Advances in Self Psychology, San Diego, CA.

Tobin, S. A. (1982). Self disorders, Gestalt therapy and self psychology. Gestalt Journal, 5, 3-44.

Tomkins, S. (1962). Affect, imagery, consciousness: Vol. 1. The positive affects. New York: Springer.

Tomkins, S. (1963). Af fect, imagery, consciousness: Vol 2. The negative affects. New York: Springer.

White, M. T., & Weiner, M. B. (1986). The theory and practice of self psychology. New York: Brunner/Mazel.

Winnicott, D. W. (1958). Through paediatrics to psycho-analysis. London: Hogarth.

Winnicott, D. W. (1965). The maturation processes and the facilitating environment. London: Hogarth.

Winnicott, D. W. (1977). The location of cultural experience. International Journal of Psychoanalysis, 48,27-44.

Yalom, I. (1980). Existential psychotherapy. New York: Basic Books.