Arnold H. Modell
We are familiar with the observation that the defences of the ego are variegated, but if one particular defence predominates it contributes to the structure of a specific character type (thus the predominance of repression is associated with hysteria, and isolation and intellectualization with the obsessive compulsive personality). A specific narcissistic defence against affects that we shall describe here may be looked upon in a similar fashion: it determines the ego structure of a variety of personality types. When it predominates it contributes to the formation of the personality structure that we associate with the narcissistic character disorder. To define a character type by means of a principal ego defence is in a sense an operational definition, for it is an observation made predominantly by means of the psychoanalytic process itself, and this implies that the defences in question not only characterize the mode of resistance to psychoanalysis but are also linked to the qualities of developing transference so that the nature of the transference itself can form the basis for nosology. I have suggested that borderline states can be defined by their characteristic transference (Modell, 1968), and Kohut (1971) has similarly defined the narcissistic character disorder by the absence of a transference neurosis and the development of a characteristic ‘idealizing and mirror' transference. So our discussion of a characteristic ego defence is necessarily associated with a discussion of the effect of ego structure upon the developing transference, and this will be dealt with in a later section.
The analyst first becomes aware of this particular narcissistic defence when he observes that there is a massive affect block in the opening phase of psychoanalysis. If other characteristics of an obsessive compulsive personality are present, this defence may be confused with that of isolation. Although there are subtle phenomenological differences that serve to contrast isolation and this narcissistic affect block, the clearest difference derives from the fact that the narcissistic defence is motivated by a fear of closeness to the object of the analyst. Isolation is a more purely intrapsychic defence, as Anna Freud (1936) has described, resulting from the fear of being overwhelmed by the intensity of affects. As I said, there are also phenomenological differences— when there is this narcissistic defence against affects one perceives after a while that there are not two people in the consulting room—the patient acts as if the analyst is not there and is talking to himself or behaves as if he is not there. The patient is ‘turned off' emotionally— he does not communicate feelings, for to do so would strengthen the object tie to the analyst.
The analyst's countertransference can usually be relied upon to differentiate the subtle differences in the phenomenology between the defence of isolation and the narcissistic defence against affects. In both there is an affect block but in the obsessive compulsive defence there is still a perception on the analyst’s part that the patient is relating to him. When the analyst is continuously in the presence of another person who does not seem to be interested in him, or indeed acts as if he was not there, he may experience this as an affront to his own narcissism and may accordingly become bored and sleepy (Kohut, 1971). Therefore our boredom and sleepiness should alert us to the possibility that this is in response to the patient’s narcissistic defence against affects. Though the analyst's withdrawal may be defensive, I do not believe that it is necessarily neurotic—it is a very human response to the patient's state of non-relatedness.
When this defence against affects is present, one can invariably observe that it is supported by a fantasy of grandiose self-sufficiency. This illusion of self-sufficiency is maintained in the face of a profoundly dependent relationship with the analyst and the analytic process; their state of non-relatedness is paradoxically associated with that of intense involvement. Grandiose and omnipotent fantasies permit these patients to maintain the illusion that they need nothing from others, that they can provide the source of their own emotional sustenance. Affects in this sense signify object seeking and object hunger, so that their absence supports the illusion that they seek nothing from the person of the analyst.
These people have an accurate endopsychic perception of their relationship to objects. They describe themselves as encased in a ‘plastic bubble’ or feel that they are really not ‘in the world’1—they arc in a cocoon: a cocoon provides sustenance for its occupant and protects it from the dangers of the environment; it is like a fortress which nothing leaves and nothing enters. A cocoon, no matter how well insulated, needs to be attached to something, and these people who may deny their dependent needs usually crave admiration.
Freud (1924, 1940) observed that when the ego is in conflict with the environment it attempts to resolve the conflict by means of denying a piece of reality. It does so at the expense of modification of its own structure. Freud employed the metaphor of a split in the ego—a failure of the ego’s synthetic functions. The mechanism of denial is thus in contrast to the mechanism of repression where the ego itself remains intact.2
As we have said earlier, a defensive alteration of the ego’s structure is necessarily linked to the phenomenon of transference: the ‘classical' neurotic—one who is capable of developing a transference neurosis— is one whose ego development is relatively unimpaired. In the ‘classical’ neurotic there is relatively good differentiation of psychic structures— conflicts are primarily internalized and not primarily between the individual and the environment. Further, in the ‘classical’ neurosis we assume a capacity to form a therapeutic alliance. Correspondingly, where there is ego impairment, this impairment will be reflected in a limitation of the ego's ability to form a therapeutic alliance, as is the case in the narcissistic character disorder.
The narcissistic defence against affects that we have been describing is similar to that of denial as it does involve a disturbance of the ego's relation to the environment with profound consequences to the ego’s structures.3 It differs, however, from denial in that it is a defence primarily against affects and not, as in denial, the perception of a particular piece of reality. But perhaps this narcissistic defence must be supported by a denial of reality—it may all be of one piece. Furthermore, this narcissistic defence against affects is similar to denial in that it can exist in a transitory form which is relatively benign or in a more permanent characterological form which is not benign. It is this characterological formation that we shall consider further.
We have observed that in our patients with narcissistic character disorders, there is probably a disturbance in the process of the development of the autonomy of the self vis-a-vis the child’s relationship to its mother. Although the specifics of the environmental trauma will vary, we have observed that the trauma results in the need to defend the sense of the separateness of the self against the intrusiveness of the mother.4 This is a confirmation of what Khan (1971) has described: ‘The ego of the child has prematurely and precociously brought the traumata of early childhood under its omnipotence and created an intrapsychic structure in the nature of an infantile neurosis which is false self organization.’
We have the impression that environmental trauma at the time when the child is developing a sense of self leads to the establishment of a precocious and vulnerable sense of autonomy. The environmental trauma may take the form of a disillusionment with the mother based on an accurate perception of the mother’s faulty judgement.5 For example, one patient who was in fact intellectually precocious perceived at the age of two or three that his mother was mad, although the extent of her madness was hidden and this fact was not acknowledged by her family or by her neighbours. The child, however, observed that his mother’s judgement was off and that he could not rely on her. Another patient perceived his mother at the same age to be flighty, childish and fatuous. The human child, as is true of all primates, is dependent upon the mother to protect it from the dangers of the environment so that the perception of the mother’s unreliability must have profound consequences.
We suggest that it leads in some patients to the formation of a precocious separation from the mother which is supported by fantasies of omnipotence. The perception that the mother's judgement is off may also be accompanied by the need to defend oneself from the mother’s excessive intrusiveness. For example, one patient's mother consistently undressed in front of him as if he were not a separate person and it did not really matter. In yet another patient, a woman, the mother did not respect the separateness and autonomy of the patient’s body— she successfully toilet trained my patient by inserting rectal suppositories from the age of six months and thereafter whenever it was thought to be needed.
Although the specific form of the environmental trauma may vary, we suggest that there is a common denominator in that these environmental traumata induce the formation of a precocious and premature sense of self—which retains its fragility and must be supported by omnipotent and grandiose fantasies.
Our description of a precocious structural formation resulting from environmental trauma forces us to clarify further the differences between this description and that of borderline states. For the borderline character is believed to be, at least in some measure, the result of environmental trauma that Winnicott has described as a failure of ‘good enough’ mothering. We must further remind ourselves that this narcissistic defence against affects that we associate with narcissistic character disorder may also exist as a defence in the borderline patient. We are reminded here of a very useful metaphor employed by Robert Knight (1954) and originally attributed to Freud, where the defensive structures of a given personality are likened to the troops of an army. An army’s defences may be strung out in an uneven fashion: troops may appear to be defending a forward position but the main body of forces may be well to the rear. The point of this metaphor is that ego arrests and regressions that form part of character are highly uneven, that defences are variegated in that our nosological distinctions are a description of where the main body of troops is (see Gedo & Goldberg, 1973, for a detailed discussion of this issue).
Thus the precocious internalization of the self that we believe characterizes the narcissistic character disorder may be a facet of the personality organization of a borderline patient, but that is not where his major forces reside. In contrast to the patient with the narcissistic character disorder, the borderline patient evidences an intense object hunger. He is seeking objects and affects are communicated and, correspondingly, the countertransference affects are intense in the analyst. The borderline patient does not attempt to maintain an illusion of self-sufficiency—instead he invests the person of the analyst with the qualities of a transitional object placed between himself and a dangerous environment, so that his safety in the world depends on the presence of the therapist (Modell, 1968). Accordingly, the therapist experiences the patient's conflict with the environment directly, and he becomes the target of the patient's rage at reality. This is in marked contrast to the affect block experienced in the opening phases of analysis with a narcissistic character disorder. Affects are therefore the leading edge of this differential diagnosis (see also Modell, 1973). The intensity of affects in the borderline state and the non-relatedness of the narcissistic character disorder reflect differences in ego structure and it is this issue that we shall examine further.
A structural differentiation can be illustrated by reference to Gitelson’s (1962) distinction of the ‘open’ and ‘closed’ system, as observed in the opening phase of psychoanalysis. Gitelson quotes Anna Freud as follows: 'The infant and child do not present us with a “closed system”, and therefore it is necessary to pay more regard to the environment and less to the biological reasonableness and synthetic functions of the ego. The aim with children is to lift them to a secondary level of development through the use of the analyst as a new object.' Gitelson observes that 'we may say that this “secondary level of development” is an intrapsychic structure which in the end is, effectively, a “closed system” leading to relatively greater autonomy vis-a-vis the environment.’
The structural difference between the narcissistic character disorder and the borderline state is a consequence of the fact that the former has achieved some degree of secondary level development and presents us with a closed system, while the borderline patient remains in this more regressive open system. We believe, with Winnicott, that the borderline state may result from a failure of the maternal environment in the first and second years of life, i.e. a relative failure in ‘good enough’ mothering (Winnicott, 1960). This relative failure of an object relationship results in a miscarriage of the normal process of identification, a failure to take something in which is specifically a failure to identify with the pre-oedipal mother. As a result, a relative autonomy from the human environment has not been achieved and consequently there is persistent object hunger with the illusion that an object stands between the person and the dangers of the environment. This is analogous to a young child who makes executant use of its mother before identification has been achieved.
We suspect that the environmental trauma which may predispose to the development of narcissistic character disorder is less severe as compared to that which leads to a borderline state. A disturbance in the mother-child relationship may be more subtle. As we have described before, the mother may be reliable in the sense that she is there, but her judgement may be faulty and she may not respect the child’s separateness and autonomy. We have suggested that this type of trauma may be such as to induce a precocious but fragile development of the sense of self. In contrast to the borderline patient something is taken in. This precocious autonomy is based on what might be described as a primary identification with the pre-oedipal mother. It is as if the child states: 'I cannot trust my own mother; therefore I will become a better mother to myself.’ This internalization, however, does not lead to a sublimation of instinctual demands nor has true autonomy been achieved. The affect block that accompanies this structural deformation defends against the closeness of an object tie. There is a fear that this fragile sense of self will disintegrate if the object is permitted to intrude. And further, true autonomy from the object has not been achieved as infantile dependency persists. Thus we have the paradox of a state of non-relatedness, yet there is a sense of an enormous involvement in the analyst and the analytic process.
The affect block signifies a profound disturbance in the relationship of the self to the external world. It reflects a state of illusionary self-sufficiency. The sharing and communication of affects is object seeking.
By means of the illusion of self-sufficiency the individual is removed from the fear of closeness to objects, for he denies any instinctual demand made upon the object.
One advantage of the structural approach is that it may help to unify seemingly disparate clinical observations. For I believe that the false self described by Winnicott and the narcissistic disorder outlined by Kohut refer to similar if not identical clinical types. I would add also that the earlier description of the ‘as if' personality by Helene Deutsch (1942) bears a close familial resemblance. In comparing these various clinical syndromes we believe that it may be clarifying to start with a consideration of a specific defence which is linked to a disturbance in the sense of self and which may exist as an admixture with other character defences, notably the obsessive compulsive’s defence of isolation (Khan, 1971). I am suggesting that our description of this defence against affects may be covering the same clinical ground as Winnicott’s description of the false self (Myerson has also addressed himself to this issue). Winnicott, however, intended his false self description to be that of a syndrome rather than of a specific defence of the ego. Winnicott’s description of the false self is primarily a description of a disturbance of affects (Winnicott, I960). Winnicott understood that the experiencing and sharing of feelings helps to organize an early sense of self; and alternatively, that one can keep oneself hidden by one’s failure to share genuine feeling. Where there is a false self, what is shared and displayed is essentially false because it is based on compliance. In psychoanalysis one learns to use one’s countertransference perceptions to separate true and false affects (Modell, 1973). One learns, e.g. that anger may be feigned, that the patient’s tears do not move us, that transference feelings may not be genuine but intended as a manipulation. We know, from Winnicott’s description and from our own observations, that these people are not only unable to communicate genuine feeling to others but are cut off from an internal experience of their own affects, a similar lack of genuine emotional experience was observed by Helene Deutsch who described how patients show ‘a completely passive attitude to the environment with a highly plastic readiness to pick up signals from the outer world and to mould oneself and one's behaviour accordingly’ (p. 265). Further we know that in the presence of such compliance there is massive unconscious defiance. There is the implicit attitude that ‘if you are fooled into accepting the feeling I display as the truth, you are contemptible.’ And the analyst’s failure to be fooled is reassuring.
Kohut, in contrast to Winnicott, does not place his clinical observations in an object relation context. I believe it is these theoretical differences that may have obscured the fact that the false self syndrome and the narcissistic personality disorder may designate identical clinical phenomena. Whether we consider the ‘false self to be a syndrome or to be a defence may be related to a quantitative structural issue. The well differentiated structures of the classical neurotic capable of a transference neurosis is in a sense idealized. For all of us carry with us vestiges of ego arrests. The narcissistic defence that we have been describing may be observed in a transitory form in otherwise mature personalities, and in the analysis of a classical neurotic who is capable of a transference neurosis it is not uncommon to observe at times affect blocking based upon fantasies of self-sufficiency. Similarly, I have noted that vestiges of the transitional object relationship may also exist in a mature love relationship where there is acceptance of the separateness of the object. If the quality of the transference can be used as the basis of our nosology this would imply a quantitative factor—that the structural deformation must be substantial. It is the introduction of this quantitative factor that allows us to think of the structural deformation accompanying the narcissistic defence against affects as occupying either a partial sector of the personality, i.e. as a defence, or as constituting a character syndrome.
I have suggested that theoretical differences may have obscured the recognition that similar clinical phenomena were being observed, i.e. that Winnicott’s false self syndrome and Kohut’s description of narcissistic character disorder may describe identical clinical phenomena. However, when we consider the question of the therapeutic alliance in narcissistic character disorders—a concept related to problems of psychoanalytic technique as well as an hypothesis concerning how psychoanalysis obtains its results—we move into an area of greater controversy. Although we may be able to agree on the description of the clinical phenomena, there is little agreement regarding the inferences that can be drawn from these observations, i.e. the question of the analysability of narcissistic character disorders.
Where there is a precociously developed and fragile sense of self defended by fantasies of omnipotence and self-sufficiency, a therapeutic alliance is not possible. Zetzel (1956, 1958) has linked the capacity to form a therapeutic alliance to the ego’s capacity to form mature object relations. It is assumed that in order to form a good therapeutic alliance there must be a sufficiently mature sector of the patient’s personality for him to perceive the analyst as a separate person and to separate the transference neurosis from his perception of the analyst as a unique and separate individual. This concept is based on the earlier observations of Sterba (1934) and Bibring (1937) of the split in the ego that enables the relatively mature portion of the patient’s ego to identify with the analyst as a ‘real’ person and to share with the analyst the common goals of the therapeutic aims of psychoanalysis. This capacity to form a therapeutic alliance has been equated by some to be synonymous with analysability (see also Greenson, 1965).
To state it simply: the capacity to form a therapeutic alliance assumes some capacity to accept the separateness and the reality of the analyst (the reality of the analyst refers to his qualities both as an analyst and as a person). As we have described earlier, al least in the opening phase of psychoanalysis with narcissistic character disorders, there is the illusion that there is only one person in the consulting room—either the analyst is not there or the patient is not there. What substitutes for a therapeutic alliance is a magical belief that to be in the presence of the idealized analyst will effect a change—they will acquire the idealized characteristics not by means of an active identification but by means of a magical process. As one patient described it, choosing the right analyst is like joining the proper club—one derives a sense of identity by means of contiguity.
We know that alongside the idealization of the analyst is the very opposite—an intense derogation. What the analyst says is negated and dismissed, as if his words were worthless. I have described the narcissistic defence as a cocoon which nothing leaves and nothing enters. The analyst experiences the frustration that his clarifications and interpretations have not been heard: nothing seems to be getting through.
With a relative absence of a therapeutic alliance, we cannot count on the patient's ability to identify with the observing ego of the analyst. Indeed an identification is warded off defensively for fear of being intruded upon, of being unduly influenced or being taken over. I believe Winnicott (1969) has this problem in mind when he describes: ‘For instance, it is only in recent years that I have become able to wait and wait for the natural evolutional transference arising out of the patient’s growing trust in the psychoanalytic technique and setting, and to avoid breaking up this natural process by making interpretations' (page 86). Interpretations are perceived by the patient as a threat to his autonomy. Envy of the analyst's knowledge is also a factor in rejecting his interpretations. It is not, however, the predominant issue as is the case in patients with a pronounced negative therapeutic reaction (Modell, 1965). There the analyst’s interpretations are taken in but they lead to a worsening of the patient's condition.
After several years of analytic labour, an embryonic therapeutic alliance may develop with these patients, but they may never establish an analytic setting where the analyst’s communications are received with suspended independent judgment to be worked over, to be either accepted or rejected.
There is indeed a question as to whether the therapeutic progress which may occur in spite of the absence of the therapeutic alliance is the result of the analyst’s interpretations or may be the result of processes that are preverbal. Anna Freud has expressed her scepticism concerning the therapeutic results of an analysis that relies predominantly on preverbal processes. She states (1969),
What strikes the observer first is the change in the type of psychic material with which the analysis is dealing. Instead of exploring the disharmonies between the various agencies within the structured personality, the analyst is concerned with the events which lead from the chaotic, undifferentiated stale towards the initial building up of a psychic structure. This means going beyond the area of intrapsychic conflict, which has always been the legitimate target for psychoanalysis, and into the darker area of an interaction between innate endowment and environmental influence. The implied aim is to undo or to counteract the impact of the very forces on which the rudiments of personality are based (P. 38).
The first issue, which Anna Freud raised by implication, is the comparison of the results obtained from a psychoanalysis of a patient capable of forming a transference neurosis and a therapeutic alliance to the result of the psychoanalysis of a narcissistic character disorder. There is also a separate issue: what is it in the psychoanalytic process itself that facilitates growth and maturation and promotes the differentiation of psychic structure in the narcissistic character? My own observations lead me to believe that there is a possibility of significant growth in the psychoanalysis of narcissistic character disorders but that the narcissistic character may not achieve a full resolution of his neurosis as can sometimes be achieved with the analysis of a transference neurosis. We suspect that the healing forces may be of a different order in patients with structural deformation; the healing may be facilitated by the analytic setting itself serving as a holding environment (in Win-nicott’s sense). There is gratification implicit in the constancy and reliability of the analyst's judgment as well as the analyst’s capacity to perceive the patient’s unique identity. The analyst, unlike the patient, is able to maintain the patient's identity to focus over time. This is the mirroring function that Spitz (1965) observed in the mother’s response to the child’s smile. This is the capacity to perceive the patient as a ‘thou’ (in Buber’s terms). Lest I be misunderstood, I am not suggesting the introduction of special parameters of activity in Ferenczi's sense. But I am suggesting, as others have, that there is gratification in the object tie to the analyst without the need for the analyst to demonstrate his affection or friendliness.6
The healing process in the analytic setting may move silently so that the analyst may experience the feeling during this phase that nothing seems to be happening. I would suggest that during this period the patient may be experiencing a silent structural growth that may correspond to the idealizing transference described by Kohut. It is a phase prior to the development of the therapeutic alliance.
This idealizing phase gives way to a more negative transference relationship and this negativity itself may support the patient’s growing sense of individuation (Winnicott, 1969). When there is growing perception of the analyst as a separate person, there may be concurrently a vestigal capacity to form a therapeutic alliance and fragments of the transference neurosis may begin to appear.
Although we acknowledge that the psychoanalysis of narcissistic character disorders can lead to significant therapeutic gains, such analyses may prove to be interminable if the gains do not also result in the establishment of a transference neurosis and therapeutic alliance.
A narcissistic defence against affects, unlike isolation, is a defence against an object relationship. Object relations are strengthened by the sharing of genuine affects so that the failure to share feelings or the presentation of false feelings creates distance between the self and other objects. The defence is similar to that of denial in that it entails a modification of the ego’s own structure. We have suggested that this modification consists of a precocious but fragile establishment of a sense of self.
The defence may occupy a sector of the personality or reflect a more massive structural arrest. When there is this structural arrest, we believe that this narcissistic defence forms the basis for the narcissistic character disorder described by Kohut and the false self of Winnicott.
This precocious sense of self leading to an illusion of self-sufficiency may also be found in other disorders, including the borderline patient, but the borderline patient, in contrast, suffers from a failure of internalization which leads to object hunger in contrast to the denial of object need of the narcissistic disorder. We suspect that the environmental trauma that may contribute to the narcissistic disorder is less severe as compared to the borderline states and may consist of the mother’s failure to accept the child’s separateness and autonomy, resulting in a fear of the mother’s intrusiveness.
The fear of the maternal object’s intrusiveness contributes to the relative inability to form a therapeutic alliance in the psychoanalysis of narcissistic character disorders. The analyst’s interpretations are experienced as dangerous, not necessarily because of their content but due to the fear of the analyst’s intrusive influence. Our understanding of the means of effecting therapeutic change must be modified in patients with narcissistic character disorders for, in contrast to the ‘classical’ neurotic, analytic progress is not obtained by means of interpreting the transference neurosis in the context of a working or therapeutic alliance. Although we acknowledge that the psychoanalysis of narcissistic disorders can lead to significant therapeutic gains, such analyses may prove to be interminable if the gains do not also result in the establishment of a transference neurosis and therapeutic alliance.
1 Volkan (1971) described a patient who felt encased in an iron bail and used his fantasies as a transitional object.
2 Kohut (1971) has described this distinction between repression and denial using the terms horizontal and vertical split, where the vertical split corresponds to Freud’s description of denial.
3 In an earlier discussion of denial I have tried to connect the mechanisms of denial to a failure to accept the separateness of objects (Modell, 1961).
4 Kohut (1971) recognizes the central importance of the tear of intrusiveness but does not see the process as object related. He states: '... the central anxiety encountered in the analysis of narcissistic personality disorders is not castration anxiety but the fear of dedifferentiating intrusion of the narcissistic structures and their energies into the ego’ (p. 152).
5 Rycroft (1955) has suggested that idealization is a response to actual object loss linked to the disappointment with a frustrating object. Kohut (1971) stresses the trauma that ensues from the mother's lack of empathy (p.46).
6 I have suggested in another paper that this process may have instinctual backing from the instincts that observe object ties in contrast to the more directly sexual instinct seeking immediate discharge (Modell, 1974).
Bibring, E. (1937). [Contribution to the] Symposium on the theory of the therapeutic results of psychoanalysis, Int. J. Psycho-Anal. 18, 170-189.
Deutsch, H. (1942). Some forms of emotional disturbance and their relationship to schizophrenia. In Neuroses and Character Types. New York: Int. Univ. Press, 1965.
Freud, A. (1936). The Ego and the Mechanisms of Defence. New York: Int. Univ. Press, rev. ed. 1967.
Freud, A. (1969). Difficulties in the Path of Psychoanalysis. New York: Int. Univ. Press.
Freud, S. (1924). The loss of reality in neurosis and psychosis. S.E. 19.
Freud, S. (1940). Splitting of the ego in the process of defence. S.E. 23.
Gedo, J. & Goldberg, A. (1973). Models of the Mind. Chicago: Univ, of Chicago Press.
Gitelson, M. (1962). The first phase of psychoanalysis. (Contribution to a symposium on ‘the curative factors in psychoanalysis’.) hit. J. Psycho-Anal. 43, 194-205.
Grecnson, R. R. (1965). The working alliance and the transference neurosis. Psy-chottnal. Q. 34, 155-181.
Khan, M. M. R. (1971). infantile neuroses as a false-self organization. Psychoanai. Q. 40, 245-263.
Knight, R. (1954). Borderline stales. In Psychoanalytic Psychiatry and Psychology. New York: Int. Univ. Press.
Kohut, H.(197I). The Analysis of the Self. New York: Int. Univ. Press.
Modell, A. H. (1961). Denial and the sense of separateness. J. Am. psychoanai. Ass. 9, 533-547.
Modell, A. H. (1965). On having the right to a life: an aspect of the superego's development, Int. J. Psycho-Anal. 46, 323 -331.
Modell, A. H. (1968). Object Love and Reality. New York: Int. Univ. Press.
Modell, A. H. (1973). Affects and psychoanalytic knowledge. The Annual of Psychoanalysis, vol. I. New York: Quadrangle Books.
Modell, A. H. (1975). The ego and the id—fifty years later, hit. J. Psycho-Anal. 56,5768.
Myerson, P. G. (1975). The false self condition. (Unpublished manuscript.)
Rycroft, C. (1955). Two notes on idealization, illusion and disillusion as normal and abnormal psychological processes, hit. J. Psycho-Anal. 36,81-87.
Spitz, R. (1965). The First Year of Life. New York: Int. Univ. Press.
Sterba, R. (1934). The fate of Ihe ego in analytic therapy, hit. J. Psycho-Anal. 15, 117126.
Volkan, V. D. (1973). Transitional fantasies in the analysis of a narcissistic personality.
J. Am. psychoanai. Ass. 21,351-376.
Winnicott, D. W. (I960). The theory of the parent-infant relationship. In The Matura-tional Processes and the Facilitating Environment. New York: Int. Univ. Press, 1965.
Winnicott, D. W. (1969). The use of an objecl and relating through identifications, (n Playing and Reality. New York: Basic Books, 1971.
Zetzel. E. (1956). The concept of transference. In The Capacity far Emotional Growth.
New York: Int. Univ. Press. 1970,
Zetzel. E. (1958). Therapeutic alliance in the analysis of hysteria. In The Capacity for Emotional Growth. New York: Int. Univ. Press, 1970.