An Overview of the Treatment of Severe Narcissistic Pathology


An Overview of the Treatment of Severe Narcissistic Pathology



This chapter is an overview of the pathology of severe narcissistic personality disorders, the common features of their transference developments in psychoanalysis and transference-focused psychotherapy (TFP), as well as differences in their clinical presentations and the corresponding implications for the technical approaches. It represents an expansion of both the theoretical framework and the clinical field explored in earlier contributions (Kernberg 2004, 2007) and reflects the technical developments of my approach to these patients, particularly in the interpretation of specific constellations of transference developments and the detailed interpretive integration of external reality into transference analysis. In this chapter, I present the various narcissistic syndromes that I will describe in order of increasing severity, beginning with narcissistic transferences at a high, stable level of functioning; next considering narcissistic transferences at a fluctuating, borderline level; then considering extreme nondepressive suicidality and self-destructiveness; and finally turning to the antisocial dimension.

The integrated clinical-theoretical frame that I have constructed over several decades draws on the work of a number of authors who have described different constellations of severe narcissistic pathology and their clinical manifestations. As I outline the development of my own conceptual model, I will place my own thinking in relation to these authors’ contributions and show how I have linked aspects of their work together with my own to form an integrated theoretical framework. I will also place this integrative model in relation to work of other authors who have contributed significantly to the delineation of clinical constellations of severe narcissistic pathology.

This work represents the experience of the Personality Disorders Institute of Weill Cornell Medical College in recent years and is an effort to integrate conceptually clinical syndromes, transference developments, and therapeutic approaches from a common theoretical perspective. It is based on the Personality Disorders Institute’s experiences in carrying out both standard psychoanalysis and TFP with a broad spectrum of narcissistic patients, and exploring their commonalities and differences (Clarkin et al. 2006).

Our basic assumptions have evolved in the context of our experiences in treating a range of narcissistic pathology as follows: the so-called thin-skinned patients described by Rosenfeld (1987), patients with the syndrome of malignant narcissism described in earlier work (Kernberg 2007), chronically suicidal and self-mutilating patients with narcissistic pathology (Kernberg 2004), and patients with the “dead mother” syndrome described by André Green (1993b).

Before spelling out basic commonalities and specific differences of various constellations, I wish to summarize briefly a theoretical frame arrived at in the context of this work and applied to the understanding of the relationships of these syndromes. This frame should be considered a tentative hypothesis to be explored further regarding its validity, and the outline that follows should be considered as a tentative working document.

Our theoretical model is linked to a general hypothesis regarding the relationship of narcissism and aggression, and we believe that our clinical examples provide support for this hypothesis. Our views of basic Freudian drive theory are similar to the later formulations of André Green (Green 2007; Kernberg 2009). Green suggested that the gradual disappearance of references to narcissism after Freud’s formulation of the dual drive theory reflected his awareness that the destructive nature of self-directed aggression, what Green has called the “narcissism of death,” coincides with the death drive. In other words, the dual drive theory of libido and death implies that both libido and aggression are invested in the self (thus constituting elements of narcissism) and also invested in object relations (thus determining the profound struggle between love and hatred that lies at the bottom of object relations) as well as in the constitution of the self. Under normal circumstances, the shift from the schizoid/paranoid into the depressive position guarantees the dominance of libidinal investment in the self as well as in the relations with significant others. Under conditions of predominance of aggression, this may be reflected in a fixation at a level of primitive defensive operations and the corresponding identity diffusion typical of borderline personality organization. Alternatively, aggression may be condensed within the structure of a defensive, pathological grandiose self, constituting the basis of the most severe constellations of pathological narcissism.

Pathological narcissism is always characterized by the crystallization of a pathological grandiose self, but libidinal dominance in such a pathological structure gives origin to the better-functioning narcissistic personalities with effective defenses against the direct manifestation of aggression in their world of object relations. In contrast, the most severe cases of narcissistic personalities present a dominant infiltration of the pathological grandiose self with aggression. This may be reflected in the syndrome of malignant narcissism, in which aggression is still mostly directed against internalized object relations reflected in conflicts with the external world. In the most severe cases, aggression is directed not only against all internal relations with significant others but against the very self. Here we find the radical de-objectalization of the “dead mother” syndrome (discussed later in this chapter) and the severely self-mutilating narcissistic personalities.

Common Transference Features Reflecting the Pathological Grandiose Self

Although the clinical characteristics of narcissistic patients may vary widely depending on the degree of severity of the pathology and the regressive features of these patients, there are certain common transference developments that remain constant. They include the activation of a dominant transference relationship between a grandiose, entitled, superior self and a depreciated object that reflects the patient’s pathological grandiose self, and a complementary, devalued, inferior, paralyzed counterpart that is usually projected onto the therapist but sometimes enacted by the patient himself or herself. This devalued object corresponds to the patient’s own devalued, dissociated, projected, or regressed infantile self. In contrast to the typical activation of dissociated internalized relations between aspects of the patient’s infantile self and his or her internalized dissociated object representations (i.e., self representation–object representation dyads), here the dyads were constituted by the relation between the pathological grandiose self and devalued self representations.

This constellation of grandiose and devalued self, however, may appear in different forms, depending on various associated characterological features. These features determine the types of narcissistic personality disorders described in the literature as the “thick-skinned” and “thin-skinned” narcissists, the “syndrome of arrogance,” chronic nondepressive suicidality and parasuicidality, and narcissistic pathology with antisocial features.

In addition, patients in the narcissistic spectrum cannot depend on the therapist or analyst. His or her interventions are discarded, ignored, or distrustfully examined for anything “new” that the patient thinks he or she has not heard before. The patient does not feel understood or helped by the therapist’s manifestation of interest and concern, and is unable to explore with curiosity what the analyst’s comments may evoke in his or her mind. The patient talks either to himself or herself in the presence of the therapist or to the therapist to influence him or her in the direction the patient desires. As mentioned in earlier work (Kernberg 1984), this affects the countertransference reaction of the therapist in that it conveys the impression that the therapist really is alone in the room.

These two features (i.e., the activation of this particular self–self relationship in the transference and the patient’s incapacity to depend on the therapist) differentiate narcissistic patients at all levels of severity of illness from the usual type of primitive, dissociated object relations activated in the psychoanalytic approach to patients with borderline personality organization. These narcissistic transferences, when dominant, require lengthy and consistent working through. What follows are particular forms these transferences take in different patient constellations and technical approaches to them that we have found helpful. In describing these constellations in a sequence of degrees of clinical severity, the emergence of predominant infiltration of the grandiose self with aggression in the severe realm of this pathology and the gradually predominant direction of this aggression against the self in the most severe cases will become apparent.


Standard psychoanalysis is generally indicated for narcissistic patients functioning at a relatively high level if some symptomatic difficulties motivate them to enter treatment.

Typical Transference Development

The dominant experience of the analyst with this type of narcissistic patient, initially, is as if there was no transference. In fact, the transference is between the pathological grandiose self that denies all dependency and an unimportant outsider, who might be useful as a source of admiration but is also potentially dangerous. The analyst, in the patient’s mind, might replicate the patient’s grandiosity, devaluing the patient in the process, or, under the effect of the patient’s implicit devaluation of him or her, crumble and give the patient a sense of wasted time, emptiness, and disappointment in the treatment. It is as if the patient were defending himself or herself against the double danger of either being depreciated by somebody who acts superior to the patient, or wasting his or her time and money with a worthless therapist. Efforts to control the analyst—who should be as good as the patient but not better, because this would evoke envy or, worse, devaluation—characterize this transference. This dominant transference may remain stubbornly unchangeable over an extended time.

Narcissistic patients may perceive the treatment as “cognitive learning.” They may be curious about interpretations as knowledge that they must learn and incorporate in order to not have any further need of the analyst and that typically, once absorbed, tends to be unconsciously devalued, with no felt need for further exploration by the patient. One patient carefully repeated my interpretations to him, “checking” their correctness or questionable nature. The same patient repeatedly experienced questions emerging in his or her mind about whether this particular session or a segment of a session was “good” or “useless.” These patients are forced to carry out omnipotent control of the analyst to keep him or her within an acceptable range of value for them, and behind this control and distancing are powerful defenses against profound envy and resentment of having to depend on what the analyst presumably has to give and the patient needs.

This dominant relationship between the pathological grandiose self and an outsider to be controlled so that he or she does not become a replica of the devalued part of the self-concept may oscillate into its opposite (i.e., periods of feelings of inferiority, failure, and humiliation on the part of the patient, and fantasies of the analyst’s grandiosity and contemptuous superiority now may dominate the picture). Upon systematic analysis of this transference, the relationship between the pathological grandiose self and the devalued self is gradually dismantled into its components, which are idealized representations of self and others that are enacted and/or projected by the patient.

As the pathological grandiose structure is dismantled, we see subtle shifts toward transferences that seem more object related, while they have a more primitive and predominantly paranoid character as the patient projects grandiose and threatening aspects of significant others of his or her past onto the analyst. One patient, a mental health professional whom I was treating in a relatively small town where the entire mental health community was familiar with one another, was spreading derogatory comments about me to a few colleagues. A few weeks later, he heard these same comments repeated by others, became frightened, and confessed them in a session. On exploration of this behavior, it turned out that he had been acting out an identification with his mother, whose chronic feelings of inferiority had motivated her to spread derogatory gossip regarding friends within her social circle. The patient then became afraid that I might be discussing his case with other colleagues.

The task now becomes to explore the gradual development of periods of rupture of the pathological grandiose self into these component internalized part-object relations and the correspondent unconscious conflicts involved. The treatment, over time, becomes more like that of ordinary borderline patients, in terms of split-off activation of idealized and persecutory relationships. Gradually, the underlying conflicts beneath the defensive structure of the pathological grandiose self may emerge, typically as intense primitive aggression linked to feelings of envy, a condensation of pregenital and oedipal conflicts, and the activation of traumatic early experiences against which the pathological grandiose self had become a major defensive structure.

Thick-Skinned Narcissists

There are other cases, however, where even careful, consistent interpretation of the defensive function of the grandiose self does not lead to a gradual shift into its internalized ideal self and ideal object components. Instead, severe splitting between periods of unremitting grandiosity and brief, dissociated, devastating experiences of self-devaluation, depression, and suicidal tendency requires very long term interpretive dismantling of a sadistically infiltrated grandiose self. The “thick-skinned” narcissistic personalities described by Rosenfeld (1987) belong to this group of patients.

These are patients who generally function quite well in their social and work life, but who are so bereft of an internal world of object relationships that they seem to be completely closed to the possibility of fantasy and daydreaming. They live in the concrete reality of their interactions with the analyst, which do not lead to any development of fantasy, desire, fear, or deeper conflict. The patients themselves may state freely that they see no reason why they should develop any particular emotional reaction to a relationship that, after all, follows the principles of a commercial contract in which “one person gets paid to take care of the problems of the other.” A patient assured me that although I seemed like a nice person, if I died suddenly, he would have no particular feeling about it.

The patient’s efforts to carry out free association may be severely distorted by his or her unwavering need to keep the analytic situation under control. Free associations are so well structured and orderly in sequence that they reveal an intentional effort to direct the analyst’s attention in a certain direction, preconsciously planned by the patient. Or the patient’s “checking” on the meaning of what he or she is saying while communicating it to the analyst may provide the patient’s discourse with an empty, intellectualized quality.

I have found it helpful to analyze the defensive function of this way of associating by focusing on the patient’s reactions to the analyst’s interpretive interventions. The patient may simply ignore what the analyst says, resuming his or her monologue after a respectful brief silence, or examine carefully all the implications of the analyst’s comments, speculate on their meaning, express agreement or disagreement, or dismantle the analyst’s statement altogether—anything, really, to avoid being impacted by the analyst in a way that is not totally under the patient’s control in an attempt to avoid a humiliating indication of his or her dependency on, and therefore inferiority to, the analyst. The interpretation of this constellation may allow the patient to become aware of his or her need to control the analytic relationship implied in the style of carrying out free association.

An additional helpful technical approach to these patients may be to analyze in great detail their difficulties outside the transference situation, the details of their difficulties in work, in intimate sexual relations, in their social life, and with their families. It is usually not difficult for patients, by sharply focusing on the areas of daily conflicts in their lives, to gradually open up their understanding of aspects of their interactions that are motivated by emotional pressures and deeper fears or desires, which can then be linked to similar manifestations that in subtle ways emerge in the transference. The “microscopic” analysis of extratransferential relations allows for gradual exploration of the transference itself. For example, behind the indifference toward their partners, one may detect the emergence of envy of the other person’s capacity for love and friendship and the freedom to have an interesting daily world of experiences from which the patient is excluded.

The indifferent patient I referred to above bitterly resented his wife talking animatedly for hours with her friends by phone, comparing it with his own restrictions and limitations in such interesting interchange with acquaintances. Similarly, behind the admiration and sexual excitement involved in their transitory infatuations, one may discover and highlight for the patient resentment of an exciting, teasing, and withholding other, and its repetition of past similar experiences with significant objects of childhood and infancy. The development of negative therapeutic reactions in the sessions, reflecting unconscious wishes to devalue the treatment following moments of envious recognition of the therapist’s capacity to help and be interested in them, offers another entrance into this transference development. Above all, these patients constantly need to compare themselves with others. The fluctuations between triumphant superiority and anxious fear of being devalued are dominant issues in the extratransferential relations to be explored, providing a bridge to the later exploration of similar issues in the transference.


In our research projects studying the treatment of borderline personality disorders with TFP, we found a significant number of borderline patients with predominant narcissistic transference developments, and became more able to identify patients with narcissistic personality disorders functioning at an overt borderline level as well as patients whose symptoms centered on inordinate arrogance and aggressiveness and patients with the syndrome of malignant narcissism (Clarkin et al. 2006; Kernberg 2004, 2007).

Patients with severe clinical syndromes may present indications for psychoanalysis or psychoanalytic psychotherapy, but when they appear as descriptively “borderline”—that is, with a general chaos of behavior patterns and breakdown in social life, work, love, and sex—they may optimally be treated with TFP (Clarkin et al. 2006). This is particularly true when they present prognostically negative features such as the infiltration of the pathological grandiose self with severe, ego-syntonic aggression toward others or self, chronic suicidal behavior, and, particularly, antisocial behavior.

The contribution of aggression directed against others or self, severe paranoid traits, and antisocial behavior constitutes the syndrome of malignant narcissism (see Chapter 11, “The Differential Diagnosis of Antisocial Behavior”). Although malignant narcissism is at the boundary of treatability, TFP may be indicated if a clear framework and structure for the conduct of the treatment can be established and maintained.

Within this most severe group of narcissistic patients we find other typical developments in the transference that correspond to the thin-skinned narcissistic personalities (Rosenfeld 1987), the syndrome of arrogance (Bion 1967), or an almost psychotic level of social functioning by patients who have zero tolerance for any triangulation (Britton 2004). Some of these patients present a severe, chronic self-destructiveness with a significant risk of suicide (Kernberg 2007). Indications for specific treatments and complications in the treatment, prognosis, and technique are affected by the variables of the degree of superego integration, antisocial behavior, paranoid tendencies, the ego-syntonic nature of aggression, and chronic manifest self-destructive suicidal and parasuicidal behavior.

Thin-Skinned Narcissists

Thin-skinned narcissists are more severely regressed and often fail in analytic treatments but respond well to TFP. They present both severe infiltration of the pathological grandiose self with aggression and a structural weakness of the pathological grandiose self, so that shifts from states of arrogance, superiority, and contemptuous feelings about the analyst to severe feelings of inferiority, humiliation, depression, self-accusation, and suicidal tendencies may occur frequently and rapidly. Such shifts, sometimes, are motivated by minor sources of triumph or defeat, and their hypersensitivity to any experienced criticism, real or fantasized. Their clinical presentation evinces severe characterological depression or chronic dysthymic reactions, suicidal tendencies, and marked identity diffusion—despite the pathological grandiose self, reflected in absence of goals, uncertainty, and confusion about their life direction and relations. Ego-syntonic sadistic features are both expressed and projected onto the therapist, who is frequently perceived as a seductive, scheming, and dishonest persecutor. These are patients with rapid changes in transference developments, extreme frustration for not being able to totally control the therapist in terms of his or her thinking or behavior, and angry outbursts of total devaluation of the therapist and wishes to interrupt the treatment. Their contemptuous attitude may take the form of accusing the therapist of not understanding anything, projecting onto the therapist the confusion in their own relationships with significant others, and heightened paranoid tendencies in the transference.

Often these are patients who have been severely traumatized by physical or sexual abuse or gross neglect in their infancy or early childhood. Their unconscious tendency to reactivate these traumatizations in the transference both conveys important information about the past and poses difficulties because of their high potential for acting out. These are the patients for whom John Steiner (1993) recommended to interpret “in the projection,” clarifying their experiences and views of the analyst rather than interpreting what is being projected into the analyst. This corresponds to an extremely effective technical approach of TFP whereby the therapist points out, at every opportunity, the kind of relationship that the patient’s experience is activating in the transference, with attention to how the same relationship tends to get activated, again and again, with role reversals. In the role reversal, the patient experiences himself or herself in the role of the internal object representation, a role that he or she had previously projected onto the therapist while remaining identified with an aspect of his or her self. Now, however, the self-aspect is projected into the object, the patient can experience what he or she had previously projected into the object, and the full nature of what had been projected may become available to the patient’s subjective experience.

For example, a patient thought that the therapist showed a sarcastic and depreciatory behavior toward him. Enraged, he became extremely critical and derogatory toward the therapist, treating the therapist as totally worthless, dishonest, and not understanding. Again, 10 minutes later, the patient had a sense that he was being treated in exactly the same way by the therapist. By highlighting the patient’s grandiose, devaluing experience with a perceived worthless or despicable therapist, without attempting to directly explain this as a projection, but showing that it was a reversal of the patient’s experience of having been attacked by the therapist in similar superior and derogatory ways that preceded it, the role reversals could be interpreted and the interpretation of projective identification thus completed. Feelings of having been attacked motivated the counterattack by the patient, and the counterattack motivated fears that the therapist, in turn, would take revenge in the same way, so that vicious cycles were generated by which the same relationship got enacted, again and again, with such role reversals.

The patient’s contemptuous, derogatory attitude toward the therapist could gradually be interpreted as a protection against the activation of an opposite role relationship. It also became easier to interpret brief periods of idealized relations with the therapist as a combination of actual recognition of something good going on in the relation and an effort, on the patient’s part, to protect himself against the bad and frightening experiences at those other moments of alternating, mutual devaluation. In our experience, a consistent interpretation of the split transferences, with particular consideration of the involvement of an easily fragmented pathological grandiose self, has been very helpful with thin-skinned narcissistic patients. At the same time, whenever what occurs in the transference seems to be a practical repetition of past experiences that the patient has been conscious of at different moments, this lends itself to a complete interpretation of the genetic aspects of the patient’s conflicts that have become an emotional reality rather than a defensive intellectualization of the patient’s remembered or reconstituted past.

Rosenfeld (1987) suggested that thin-skinned narcissistic patients who had experienced severe traumatizations in their past might be retraumatized by the analyst’s interpretations of their aggressive conflicts in the transference. In our experience, the clarification of the dominant transference “in the projection,” without full interpretation of the projective identification at that point, permits completion of the interpretation once the reciprocal activation of self and object representation in the transference has taken place. This TFP-derived technique permits the therapist to carry out a systematic interpretive approach to extremely negative transferences without the patient’s experiencing this interpretation as an attack. Here, I agree with Steiner (2008), who questions Rosenfeld’s excessive caution in interpreting negative transference developments in these patients.

The Syndrome of Arrogance

Bion (1967) has described the syndrome of arrogance in severely regressed patients, consisting of 1) an openly aggressive and extremely arrogant behavior toward the therapist; 2) incapacity for any cognitive reflection, so that the patient appears to be “pseudo stupid”; and 3) an inordinate curiosity about the therapist rather than one’s self. Bion describes the projection of these characteristics into the analyst and proposes as an essential dynamic of this situation: the activation of the frustrated baby’s rage at an impatient mother who would talk to him or her without really understanding the baby and “stupidly” expects the baby to respond as verbally as she addresses the baby. It implies the projective identification of a destructive, sadistic object interfering with the communicative aspects of projective identification. The analyst’s containing this projection of an extremely destructive internal object geared to eliminate all verbal communication would be the key to managing this situation. In our experience, it reflects the chronic enactment in the sessions of primitive hatred and envy, but with the characteristic that the aggression is acted out without the patient’s having any self-reflective awareness of it. The arrogant behavior expresses his or her combative aggressive needs: the need to not demonstrate any capacity for cognitive communication with the therapist, the desperate effort not to have to acquire any awareness of the significance of his own behavior, and the need to control the therapist to avoid the projected and feared aggression to return in the form of counterattacks from the therapist.

I have found that the best way to manage this syndrome, which presents in some of the most severe types of narcissistic regression, is a combination of interpretation and maintaining firm boundaries of the treatment situation. The therapist needs to set very clear limits regarding how far the patient may express his or her aggression verbally, without attacking the therapist, or the office, physically or invading the therapist’s space outside the sessions, so that the attacks are limited to an extent that can be contained in the therapeutic sessions. At the same time, the therapist’s focus needs to be on the patient’s intolerance to recognize the sadistic pleasure that he or she obtains from the aggressive behavior. The defense against the fear of recognition of the pleasure in his sadistic behavior, when it is overcome, permits the patient to accept that pleasure without fear of retaliation or guilt. By the same token, this acceptance tends to reduce the intensity of the aggression, opening the possibility of studying the origins of that reaction in the transference.

One of our patients cut the plants in her therapist’s office and abused him verbally in public spaces, but eventually she was able to acknowledge and explore her pleasure in these attacks as an unconscious identification with a sadistic aunt who had dominated her childhood and had been extremely physically abusive. The decomposition of the pathological grandiose self into its component internalized object relations—in this case, the activation of the sadistic aunt identification—signals the resolution of the structure of the pathological grandiose self during transference interpretation.

There are patients who function at a much higher level, with much better control of their behavior inside and outside the sessions, but in whom the attitudes of contempt, devaluation, and depreciatory competition with the therapist go with periods of self-contempt, despair, and suicidality under the effects of this self-contempt. These patients usually have strong paranoid tendencies and tend to justify their contemptuous behavior through intellectual debates with the therapist that are expressed in combatively arrogant attitudes. This same contempt creates severe conflicts at work, in the social sphere, and in intimate relations. With these higher-functioning patients, the contemptuous behavior and arrogance is less intense and overwhelming than in the regressed arrogant patients; nonetheless, it is very clear to the analyst that he or she is being treated contemptuously and clear to the patient that he or she is doing that, so that, in the short run, this is open to transference exploration. However, the extended duration of unrelenting contempt may undermine the analyst’s positive disposition toward the patient, which is one of the unconscious objectives of this transference development: both revenge against hated parental images and, at bottom, a desperate effort to still maintain a good relationship with the analyst and not be abandoned, despite this behavior.

One such patient came to us after a series of previous analysts, all of whom she ridiculed to other persons outside the treatment. Upon settling to work with a member of our group, she was highly contemptuous of him as well and bad-mouthed him for quite some time before this behavior became evident and could be explored in the transference. Another patient secretively bought a complete set of CDs to train therapists in cognitive-behavioral psychotherapy, cheerfully arguing with the therapist over weeks about the limitations of his analytically oriented approach. The working through of these issues in the transference and the consistent interpretation of all the implications and features of the patient’s arrogance and contempt may resolve the problem, but at a significant cost of the analyst’s working through the corresponding countertransference reactions. Over time, these patients manage to significantly challenge the analyst’s self-regard and confidence in his or her work.

In fact, the tolerance of the countertransference may become quite a central issue in such cases. When patients involve third parties (e.g., complaining to relatives and to other therapists, requesting consultations to protest about the way they are being treated), it is particularly difficult to maintain a technically neutral stance. It may become necessary in some cases to establish limits to the patient’s behavior, geared to permit the indispensable sense of security—physical, emotional, professional, legal—the therapist needs to maintain his or her position of concerned objectivity. The therapist’s capacity to evaluate what enactments and projective identifications are activated in the transference/countertransference developments needs to be protected.

At all levels of narcissistic pathology, the mechanism of omnipotent control represents an important unconscious effort on the patient’s part to prevent change to occur—to “freeze” the therapeutic situation—as a crucial attempt to protect the patient’s identification with his or her pathological grandiose self. At “higher” levels of narcissistic pathology, this may take the form of idiosyncratic convictions such as political ideologies or highly personal thought systems, even in persons with otherwise excellent reality testing. At the borderline level of narcissistic pathology examined here, these personal thought systems may take on a quasi-delusional quality and may serve to protect the patient’s conviction of his or her intellectual superiority. One patient was convinced that any love for a woman signified an inferiorizing weakness; another patient was convinced that he was the greatest artist in his field and that all experiences to the contrary were engineered by fellow artists who envied him. These beliefs had the effect of preventing the therapy from having any impact on the patient and protecting the grandiosity. The development of “incompatible realities” in the transference, and the technique used under this condition that has been described in earlier work (Kernberg 2004, 2007), may become an important aspect of transference analysis of these cases.

Intolerance of Triangulation

An extreme form of omnipotent control may evolve as part of the intolerance of triangulation (Britton 2004) in very regressed narcissistic patients. Intolerance of triangulation refers to a particularly severe distortion of internalized object relations, a regression within which the patient cannot tolerate any thoughts that are different from his or her own. The role of the therapist is to confirm the view of the patient and to assure the patient of the reality and stability of that shared experience. Any contribution from the therapist that is at variance with the patient’s thinking is destabilizing to the patient’s grandiose self and carries malignant implications. At bottom, it is the search for a perfect symbiotic relationship within a dyad that does not tolerate the disruption by a third, excluded object and represents the fragile omnipotence of a pathological grandiose self that attempts to maintain absolute control over experienced reality. Here, archaic oedipal conflicts (i.e., the intolerance of the relation between the parental couple from which the infant feels excluded) surface in the transference as the envious resentment of the therapist’s relation to an internal object of his or her own (his or her independent thinking, theory, or reflection about what is going on in the relation with the patient), making different perspectives intolerable to the patient.

Efforts by the therapist to bring in views different from that of the patient are experienced either as a total abandonment and rejection or as a sadistic intrusion, an aggressive effort to control the patient’s mind. This situation, originally described by Britton (2004), may even be observed in patients functioning at a relatively high level within the spectrum of narcissistic pathology, in which it is revealed in very subtle ways of rejection of the therapist’s independent thinking by highly sophisticated maneuvers that reassert the patient’s initial views and force the therapist into a temporary emotional retreat. But this condition emerges as well with extremely regressed narcissistic patients functioning on an overt borderline level, in which the patient’s experience of reality has such an extremely distorted quality that it is close to psychotic.

In this latter case, the patient may be totally convinced of the realistic nature of an emotional experience, within which the patient’s behavior may have been extremely inappropriate and practically psychotic in the context of its social surround. Here, any effort of the therapist to probe reality testing may be experienced as an attack, bringing about a rageful effort by the patient to shake off the intrusion.

One patient maintained an almost delusional conviction that a man, who was obviously exploiting her and showed his indifference at every turn, was in love with her. In turn, she treated the therapist who was trying to confront her with her illusions as if he had a totally unrealistic view of reality. This woman functioned remarkably well in other areas of her life. Another patient with severely overt borderline functioning created such a disruption at a family funeral that family members escorted him away from the grave. In the subsequent sessions, he raged against his family’s callous indifference to his intense suffering and would become incensed by any effort by the therapist to tactfully question whether, under the circumstances, the expression of his mourning had acquired characteristics that were, in fact, quite problematic and socially inappropriate.

A technical approach that has been helpful with these conditions is enormous patience and consistency in pointing out to the patient the fact that any view that is different from his or her own creates an intense pain, as if the patient’s thinking or very capacity to deal with reality were questioned, so that the patient has to protect himself or herself against such a dangerous assault. To the patient, it seems to be a case of the therapist trying to drive the patient mad. The analyst needs to spell out gradually the nature of the danger that is creating panic in the patient: the fear of a total disqualification of his or her capacity to think, of total abandonment and loneliness, and of the patient’s own intense enraged reaction to this dangerous situation, along with his or her fantasies of the sadistic intention of an imaginary outsider—who seems to be committed to destroying the safety of the patient’s earlier experience—and the therapist’s collusion with such an enemy. By its absence, in these cases the situation described points to the importance of the existence of a “three-person psychology,” which constitutes the basis of a therapeutic relationship, and the resolution of this severe regression marks an important improvement for these patients.

The concept of the three-person psychology refers to the consideration of the therapeutic relationship as determined by, at a minimum, the transference, the countertransference, and the analyst/therapist’s position as an “excluded third party”—that is, that part of the analyst’s personality able to explore the transference/countertransference relationship without being immersed in it (Kernberg 2012). It is a reflection of an internal split in the analyst’s ego in relating to the patient that gets “obliterated” again and again in countertransference enactments and in the developments of projective counter-identifications (Grinberg 1979). At a deeper, symbolic level, the three-person psychology refers to the oedipal structuralization of the analytic relationship and the potential for intolerance of the oedipal situation in regressive, symbiotic transferences. At this symbolic level, the patient becomes the excluded third party, the infant excluded from the relationship of the parental couple. This is the case of the patients described by Britton (2004).


Severe Narcissistic Suicidality

It usually is not difficult to differentiate the chronic suicidal and parasuicidal behavior of patients with severe personality disorders without dominant narcissistic features from that of narcissistic personalities functioning on an overt borderline level. Nondepressive suicidal behavior of patients with borderline personality disorders usually is impulsive, an equivalent or symptom of an acute affect storm, related to a frustrating, enraging, or traumatizing experience, or an effort to influence or control a close family relative or love/hate object. Conversely, chronic suicidal or parasuicidal behavior of patients with narcissistic personality disorders evolves slowly and in a determined way over a period of weeks or months and is prepared and carried out in what impresses an observer as a cool and deliberate plot, quite often in the context of surface behavior that is seemingly friendly and relaxed. Patients with this extreme degree of severity of narcissistic pathology differ from the constellation in whom suicidal behavior is part of severely disturbed, aggressive, and shifting moods with strong depressive features. Here, in contrast, suicidal and parasuicidal behavior—even severe self-mutilation—punctuate a generally stable, seemingly normal behavior.

From a psychodynamic and transferential perspective, this behavior reflects a deep and consistent aggressive devaluation of the external world, a radical devaluation of significant others and the self, a “negative narcissism,” in Green’s (1993b) terms, with a patient’s profound sense of superiority derived from overcoming all feelings of fear of pain and death, all feelings of needs involving others, and a sense of omnipotence by controlling one’s own death as a final, absolute power and freedom. This general transference disposition takes many different forms. The therapist’s capacity, within his or her countertransference reaction, to empathize with this terrible psychological reality of the patient, with that part of the patient identified with a self-murderous grandiose self, may become a crucial aspect of transference interpretation. The therapist is a natural enemy of that internal object, and a major question arises as to whether the therapist has any ally in the patient’s mind: Is there some way to contact the patient’s oppressed, tenuous wish to survive? Highlighting this potential internal conflict in the patient’s mind as it becomes activated in the transference, a true struggle between the death drive and the wish to live, is a major therapeutic task in these cases.

For example, one of our patients chronically ingested rat poison to try to kill herself. Despite careful searches, carried out during her hospitalization at our service, it was not possible to find the source for obtaining the poisonous substance. The patient denied her continuous ingestion of the poison, while the serum prothrombin gradually increased over days. She had a history of severe internal hemorrhages that required extensive diagnostic and therapeutic interventions, and, in the middle of all this, she maintained an apparently calm and almost cheerful attitude that belied the extreme gravity of her condition.

Another patient carried out severe self-mutilations, cutting tendons, which led to the loss of fingers, and attempted to set herself on fire, creating a near catastrophe that threatened the life of numerous people living in her apartment building. Sometimes her self-destructive attempts were serious parasuicidal behaviors, such as profound self-cutting that occurred in the middle of what seemed perfectly relaxed and adaptive behavior by the patient. Still another patient went out with her sister to go shopping, spent what appeared to be a very pleasant afternoon with her, and, after getting back home, retired to her bedroom. Her sister approached her to share with her some of the objects they had bought and found her sister with multiple cuts on both arms, bleeding heavily, and requiring major interventions at an emergency service.

There are patients who deny their suicidal intentions and preparations and convey a sense of triumph in their capacity to shock the unsuspecting therapist with their behavior. Other patients may talk freely about their suicidality, while implying that it is beyond their control and that for the time being they are not in touch with that part of themselves that wishes them to be dead. The implicit attack on family and on the therapist in such behaviors often remains unconscious, but at times is accompanied by a sense of sadistic satisfaction and triumph. At the same time, patients may unconsciously bring about situations in which their suicidal behavior would seem to reflect some neglect or insufficient alertness on the part of the therapist, and families may become enraged at what they perceive as the therapist’s failure to contain or prevent the patient’s behavior. Some patients triumphantly point out that their behavior is not only not under their own control but not under the therapist’s either and thus reflects the therapist’s impotence and incompetence. We have all seen cases in which family members unwittingly collude with the patient to blame the therapist and then move on to find another. The patient, in the role of a “serial killer” of therapists, thus experiences an unconscious sense of triumph at having “killed off” another therapist.

The combination of severe self-destructive tendencies and antisocial tendencies may be expressed in provocative and litigious behavior, blaming the therapist for not having been attentive to the risk of a severe suicidal attempt by the patient, indirectly exerting an omnipotent control over the therapist by means of induction of paranoid fears and guilt feelings. Here, the aggressively infiltrated pathological grandiose self provides the patient with an illusionary power, not only over the therapist, but over life and death over pain and suffering, and opens an escape into a “liberating” death from a world that cannot be controlled.

In a randomized controlled trial comparing TFP, dialectical behavior therapy (DBT), and supportive psychotherapy based on psychoanalytic principles (SP), we found that TFP was as effective as DBT in reducing suicidal and parasuicidal behavior, in contrast to SP, which was less effective. Only TFP increased mentalization (Kernberg et al. 2008).

I believe that it is important that the therapist accept honestly, in his or relationship with the patient and the patient’s family, and in the elaboration of his or her countertransference, the possibility that the patient may commit suicide and that the treatment may not be able to prevent it. The therapist may have to acknowledge openly, in diagnostic sessions with the patient and the family, that outpatient psychotherapeutic treatment carries a serious and unavoidable risk of suicide, given the patient’s severe condition, and yet may be still preferable to the long-term, indefinite duration of hospitalization. The fact that here suicide is neither predictable by the presence of severe depression nor preventable because of its deep characterological basis may have to be made explicit verbally and, at times, in documentation that protects the therapist legally. At the same time, the patient’s knowledge that he or she will not be able to blackmail the therapist with suicidal threats may reduce the secondary gain of this symptom and limit the patient’s sense of omnipotence. The therapist needs to be assured of his or her own physical, psychological, and legal safety to be able to be dedicated to helping these extremely difficult patients.

In all these cases, one must interpret openly and calmly that the patient, while willing to undergo a psychotherapeutic treatment that he or she hopes to bring about improvement of his or her condition, is nonetheless under the sway of powerful and destructive internal forces. An acknowledgment of objective concern (the excluded third) for the patient’s self-destructiveness conveys the possibility of discussing destructive impulses openly to assure the patient that one respects the severity of his or her condition and the power of this part of him or her. The need to protectively dissociate the destructive part of himself or herself from the apparently “cheerful” participation at other moments in the treatment can be clarified and resolved, and the patient’s open acknowledgment of his or her dominant self-destructive motivation becomes the main issue to be explored in the sessions.

An important related technical requirement is the exploration of the patient’s apparent absence of concern over this terrible control that the part of him or her wishing to die has over whatever part wishes to remain alive and the reasons for the devaluation and implicit hatred of that part that wishes to be alive. Here, the gradual discovery of the dangers of remaining alive, the terrible suffering related to the implied loss of control and superiority that is implied in identification with death, having to experience oneself alone and abandoned, having to face envy toward people not condemned to such self-induced destruction—all may emerge in the context of more specific aspects of the patient’s infantile and childhood history. Sometimes, the patient unconsciously hopes for an omnipotent good object that will rescue him or her from that desperate state and projects his or her own omnipotence to an illusional godlike rescuer. This fantasy, in turn, needs to be explored for its potentially self-destructive implications.

The “Dead Mother” Syndrome

A related group of patients, who on the surface seem much less severely ill but at a deeper level evince a relentless determination to destroy all relationships, the efforts of the therapist, and even their own sense of being alive, is reflected in the syndrome of the “dead mother” described by André Green (1993a, 1993b). Under these conditions, there is a rejection of any significant relationship out of identification with an internalized imago of a dead mother, frequently derived from the early experiences with a severely depressed, unavailable mother. The patient unconsciously attempts to maintain the relationship with this absent, nonresponsive mother, unconsciously enacting the fantasy that in his or her own emotional death and loss of self, he or she will be reunited with an idealized mother and protected from any further suffering. To give up one’s own existence as an autonomous self, one’s own need to depend on anybody else, to devalue completely all representations of significant others would, in the fantasy of these patients, provide definite restfulness, security, and equanimity. These are patients who may, on the surface, attentively listen to the therapist’s interpretations and then react, saying, “All this is very interesting, but doesn’t touch me at all.” Their attitude of “So what?” in response to interpretations is persistent and unmovable. By the unconscious dismantling of all relationships, what Green (1993b) calls “de-objectalization,” they obtain the same effect as patients whose repeated severe attempts at self-destruction reflects their dominant motivation in life.

De-objectalization refers to an extreme manifestation of the death drive, whereby an attack is leveled against the very structures of the mind that sustain introjective processes and permit the establishment of representation of self and object within an internal space. As an ultimate defense against intolerable dominance of aggression over libidinal investments, the patient’s very capability to experience an internal world is attacked: the sense of time is frozen, the aggressive and libidinal investments of object and self are dismantled, and a total void occupies the mind (Green 1993b, 2010). This self-destructiveness, in Green’s view, goes far beyond masochism. In our experience, these patients are remarkably free of conscious awareness of both aggressive and libidinal impulses. Often effective in their work and superficial social relationships, they are basically loners and convey a sense of total isolation and lack of interest to live. They are not depressed, and the initial depressive countertransference reaction they evoke in the analyst shifts, over time, into a sense of exhaustion and emptiness. If it becomes possible during lengthy treatment to unmask their unconscious efforts to deaden the relation with the analyst as a defense against severe frustrations in their need for love and dependency, and to explore the aggressive resentment of the analyst’s alleged indifference, self-centeredness, and unconcerned enjoyment of his or her own life, there may be hope. Under the most favorable circumstances, intense primitive hatred and envy may become reactivated and a “revival” of internalized object relations possible.

Severe Sadomasochistic Transferences

Another group of patients, who are somewhat less severe in the manifestations of their self-destructiveness, unconsciously attempt to transform all relationships into hostile interactions, or severely sadomasochistic involvements. It is as if the only way in which they can trust that someone cares about them is to provoke an attack from that person. Within the frame of a psychotherapeutic treatment, this need to provoke attachment through hostility may lead to disastrous stalemates and breakdowns. These patients, at least, still attempt to maintain some sort of relationship with the therapist, in contrast to patients with the “dead mother” syndrome (discussed below), in which the destruction of all relationships seems to be the overriding goal. Here, the therapist’s interpretive clarification of this consistent attempt to provoke the therapist to attack the patient may open the possibility of modifying this pattern. These cases, in fact, represent the most severe form of negative therapeutic reaction.

Regarding these most severe cases described, we have found that technical requirements include, first, the establishment of realistic conditions under which the treatment can be carried out. The structure of the treatment, as determined through contract setting, must create conditions in which the therapist is protected physically, psychologically, socially, and legally. The involvement of the family in the process of setting up the conditions for the treatment is essential. During the treatment itself, it may be necessary to maintain ongoing family contacts to maintain and reinforce the treatment frame and realistic expectations. In a litigious culture such as the United States, it is particularly important that the therapist be solidly protected from any risk of being involved in threatening lawsuits related to a patient’s attempted or completed suicide. If the safety of the therapist cannot be established and maintained, the treatment is not possible.

Amid all these struggles, the patient’s attempt to destroy the therapist, spoil his or her reputation, expose his or her impotence, and blacken his or her image to his or her family members needs to be explored as an expression of that part of the patient that, at bottom, is attempting to destroy himself or herself. The exposure of a sadistic, murderous internal object and the patient’s fascination with, submission to, and identification with that object, and the savage suppression of his infantile self and aspirations for love, are frightening aspects of the transference. They also may emerge as very disturbing aspects of countertransference developments. This, again, requires a combination of objective security of the therapist and “space” for working through one’s countertransference reactions. The therapist’s survival despite a patient’s consistent unconscious efforts to transform him or her into a sadistic and devalued object and the tolerance of the feelings of frustration, delusion, envy, triumph, and loneliness dominant in the intersubjective field, are the heavy price for carrying out these treatments, as well as the material on the basis of which understanding may be gained, and both success and failure evolve.


Although narcissistic pathology with antisocial features reflects a higher level of engagement with the external world than the most severe group of narcissistic constellations examined previously, the practical absence of protective superego functions facilitates uncontrollable acting out of aggression that may easily destroy the therapeutic relationship and represents a negative prognostic feature for any therapeutic intervention. A high degree of secondary gain and/or the presence of severe antisocial features represent negative prognostic indicators in the treatment of the narcissistic pathology we are exploring (Kernberg 2007). We have found that it is essential for the therapist to carry out an adequate diagnosis at the beginning of the treatment, to differentiate antisocial personality proper from narcissistic personalities with significant antisocial behavior and from the syndrome of malignant narcissism. The latter conditions may be psychotherapeutically treated, in contrast with antisocial personality proper (see the criteria outlined by Hare [Hare et al. 1991], Stone [1993], and Kernberg [2004]).

Once the decision has been made that, although the prognosis may be guarded, TFP would still be indicated, it is important to set up a treatment arrangement by which not only family are involved, but the therapist has the assurance of adequate outside information that may protect the patient, the therapist, and the treatment from severely destructive behavior that is kept split off from the sessions and reflects a long-standing unresolved dishonesty of the patient. Antisocial behavior with deceptive communications by the patient in the treatment hours may still be treatable if the patient does not present an antisocial personality disorder in a strict sense. The systematic analysis of dishonest behavior becomes a central precondition for structural change in these patients. Often a team structure may be necessary to assess the patient, as well as to protect him or her from severe acting out of antisocial behavior, and the therapist must be prepared to analyze all symptoms of deceptiveness as the highest priority of his or her interventions. The fact that patients cannot be trusted to convey honest information about what is occurring in their lives, what they think and feel, must be accepted as a given reality whose conscious and unconscious motivation needs to be explored in the sessions. Obviously, deceptiveness and dishonesty protect the patient from imagined dangers that would occur if, in his or her mind, he or she were not to deploy such dishonest behavior.

This means that severely paranoid features are constantly present as conscious and unconscious motivation protecting antisocial behavior in the transference, and the transformation of “psychopathic transferences” into “paranoid transferences” is a major tactical aim in these treatments. Often the atmosphere during sessions, at least initially, is apparently friendly and distant, only to become extremely paranoid and hostile when this transformation is achieved. The patient seems to be worse, acting out may intensify, the treatment may appear more at risk, and yet a more realistic relationship may be taking place. The combination of antisocial defenses and the pathological grandiose self makes this transition from a psychopathic into a predominantly paranoid transference much more difficult and laborious.

During this process, patients may attempt to seduce the therapist into colluding with some of their deceptive behavior under the guise of a more honest collaboration with the therapist. It is important, of course, to resist this temptation, and it is under such circumstances when the therapist’s responsibility to the patient, to his or her own moral values, and to society may enter into conflict. In all cases, any collusion with antisocial behavior carries a heavy price and, eventually, the risk of treatment failure. It may be preferable to interrupt a treatment when therapeutic conditions are no longer compatible with the therapist’s commitment to an honest and moral relationship. Once again, the therapist’s safety represents the first condition that determines whether a treatment is viable, and psychoanalytic psychotherapy is no place for inordinate heroics.

Some patients’ antisocial behavior is expressed under dissociated conditions. For example, one patient would physically abuse his wife in the middle of rage attacks, a behavior rationalized as part of a patriarchal ideology that required her to accept orders from him without raising questions. This behavior, infused with sadistic pleasure, gradually became ego-dystonic outside such crises, but the behavior persisted despite professed guilt feelings at other times. It was as if the expression of guilt and concern, instead of influencing that behavior pattern, contributed to stabilize it by serving to replace any reparative action based on these guilt feelings. In another example, a research scientist falsified the data derived from his experiments and then replicated these experiments to annul the impact of the false results previously communicated. Sometimes the antisocial behavior is rationalized as part of the entitlement professed by the pathological grandiose self: one patient refused to pay taxes, disobeyed parking laws, and ignored his incoming mail to the extent of seriously damaging consequences for him.

The transference implications of these behavior patterns vary widely: from generalized projection of superego precursors to enactment of identification with specific dishonest or poisonous representations of infantile objects, to provocative testing of the therapist’s commitment to the patient. The countertransference developments in such cases, in turn, challenge the analyst’s commitment to the patient and sometimes tend to induce a paralyzing denial of the gravity of the situation. The analyst’s sense that the unremitting nature of the antisocial behavior of his or her patient threatens to make the analyst an accomplice of a sort may contribute to a realistic consideration as to whether the treatment is still viable.

As mentioned previously regarding several cases, the systematic working through of the transference related to the pathological grandiose self gradually unmasks the component self and object representations of that pathological self-structure. A female patient presented a syndrome of malignant narcissism, with chronically deceptive social behavior, exploitive relationships, paranoid reactions, exhibitionistic behavior, and sexual promiscuity. She had a history of childhood sexual abuse by close relatives, an intensely ambivalent and symbiotic relation with her mother, and a seductive but distant father. In analysis, the transference shifted gradually from a grandiose, aloof demandingness and devaluation of the analyst to an intense negative therapeutic reaction whenever she felt helped by him. In her four-sessions-per-week analysis, these episodes of feeling worse as a result of her treatment, lasting for weeks, seemed at first to reflect unconscious envy over the analyst’s capacity to help her. Gradually, however, they reflected increasing paranoid fears over the analyst’s presumed seductive intentions involved in helping her, representing the projected identification with the sexual abusers of her childhood. It also reflected the projection of the patient’s own seductive and promiscuous behavior. Later on, severe aggressive conflicts over the dependency on an unavailable, teasing, and withholding mother emerged. Her irresponsible handling of money increased, and she fell behind in her payments for the treatment, linked to the developments of resentful depressions over any frustrations in the therapeutic setting, enacting the relationship between a hungry infant and a withholding mother, with corresponding role reversals in the transference. Now the withholding, sadistic mother representation reflected another of the patterns of the pathological grandiose self, initially expressed in her teasing and withholding behavior toward men with whom she would get involved. At this stage of her treatment, for the first time, guilty feelings and depressive reactions evolved as a consequence of her becoming aware of the deceptive behavior in the transference of lying and withholding important information to maintain an illusory superiority over the analyst. All these transference developments could be explored gradually in her chaotic relations with men, leading to more stable and deepening love relations. She was able, eventually, to obtain a very profound change in her characterological structure, with normalization of her capacity to achieve a gratifying, stable love relationship and an effective improvement in her professional function and social life.

While the chronic dishonesty in this patient’s social interactions and irresponsibility regarding financial matters were subtle and relatively minor and did not constitute a major objective danger to others or to herself, this is not always the case. Many narcissistic patients with severe antisocial behavior or passive exploitive or aggressive behavior may, indeed, threaten other persons, society at large, and/or themselves. Whenever the patient’s behavior seems to pose a threat to his or her physical, psychological, or legal survival, clear limits may be required as a condition for the treatment. In some cases, it may be necessary to have access to corroborating information from external sources, with the patient’s permission, of course. For example, patients who are chronically deceptive and who carry out violent behaviors against sexual partners or other members of their family that they avoid communicating to the therapist, or patients involved in stealing, drug dealing, or other severe conflicts with the law that reflect serious risks to themselves or others but that they are withholding from the therapist, may require such interventions. Such arrangements may be limited to early periods of the treatment. The technical approach in TFP for such severe cases includes the combination of setting appropriate limits, availability of external sources of information, open communication with the patient regarding all involvement with third parties, and ongoing interpretation of the transference implications of the therapist’s temporary abandonment of a position of technical neutrality, should circumstances warrant the therapist’s having to take a more active intervention.

In cases of severe secondary complications such as alcoholism, drug addiction, or severe eating disorders, we have found that psychoanalysis or TFP is not effective without first getting these obstacles to treatment under control. There are alternative ways to achieve this, depending on the nature of the complication, the degree of pathology of the patient, and the actual danger that such a complication presents.

As a general rule, alcohol or drug abuse, but not dependency, may be tolerated as a complication that can be dealt with in ordinary therapeutic settings. In contrast, definite dependency requires either that treatment by detoxification and rehabilitation before a psychoanalytic psychotherapy or psychoanalysis be started or that the therapy be timed to coincide with the start of a rehabilitation program that allows open communication between the treatment team and the therapist. Ancillary arrangements such as regular attendance at 12-step programs or ongoing work with an eating disorder specialist are part of the general strategy of setting up a therapeutic contract and free up the therapist to focus on the patient’s object relational world. Again, if the major problem is deceptiveness, but without chronic behavior outside the sessions threatening the patient and/or his or her psychosocial environment, ordinary treatment arrangements and the therapist’s preparedness to deal with the dominance of psychopathic transferences in the early stages of the treatment usually are sufficient.


I have attempted to provide an overview of different constellations of narcissistic personality disorders as they present clinically along a spectrum, from the relatively best functioning and organized forms of pathological narcissism to the most regressive and potentially most threatening to the patient’s psychosocial and physical survival, that may yet be helped within our present clinical understanding and therapeutic approaches.

I would place the general approach to the treatment of narcissistic pathology represented in this chapter as very close to Kleinian theory and, particularly, to the contributions of Herbert Rosenfeld (1987), John Steiner (1993), Ronald Britton (2004), and Wilfred Bion (1967) to the study of narcissistic personality, and the clinical and theoretical contributions that André Green (1993a, 1993b) has provided throughout the years developing the concept of narcissism of death, the psychology of the negative, and de-objectalization as a manifestation of severe narcissistic pathology. My own contributions to the technical approach of narcissistic personalities, particularly the most severe cases, in which functioning is at a borderline level, represent a significant aspect of this therapeutic approach, as well as the technical implications of the development of TFP as a general treatment approach to severe personality disorders (Kernberg 1984, 2004, 2007). Within this general technical approach, I believe that the treatment of patients with severe narcissistic pathology, as well as patients presenting borderline personality organization in general, should be approached with an interpretive psychoanalytic stance from the very beginning of treatment. The early analysis of the alternation of primitive unconscious dyadic relations in the transference in the most severe cases facilitates their psychoanalytic treatment. This particular technical approach facilitates mentalization and constitutes an important aspect of the technical approach at severe levels of transference regression (Kernberg 2012). In short, I have presented an approach that reflects the application of psychoanalytic theory of pathology and technique to a broad spectrum of patients and expands the realm of classical psychoanalysis with its extension of a specific type of psychoanalytic psychotherapy.


Bion WR: On arrogance, in Second Thoughts: Selected Papers on Psychoanalysis. New York, Basic Books, 1967, pp 86–92

Britton R: Subjectivity, objectivity, and triangular space. Psychoanal Q 73(1):47–61, 2004 14750465

Clarkin JF, Yeomans FE, Kernberg OF: Psychotherapy for Borderline Personality: Focusing on Object Relations. Washington, DC, American Psychiatric Publishing, 2006

Green A: On Private Madness. Madison, CT, International Universities Press, 1993a

Green A: Le Travail du Négatif. Paris, Les Éditions de Minuit, 1993b

Green A: Pourquoi les Pulsions de Destruction ou de Mort? Paris, Éditions du Panama, 2007

Green A: Illusions et Désillusions du Travail Psychoanalytique. Paris, Odile Jacob, 2010

Grinberg L: Countertransference and projective counteridentification. Contemp Psychoanal 15:226–247, 1979

Hare RD, Hart SD, Harpur TJ: Psychopathy and the DSM-IV criteria for antisocial personality disorder. J Abnorm Psychol 100(3):391–398, 1991 1918618

Kernberg OF: Technical aspects in the psychoanalytic treatment of narcissistic personalities, in Severe Personality Disorders: Psychotherapeutic Strategies. New Haven, CT, Yale University Press, 1984, pp 197–209

Kernberg OF: Aggressivity, Narcissism, and Self-Destructiveness in the Psychotherapeutic Relationship. New Haven, CT, Yale University Press, 2004

Kernberg OF: The almost untreatable narcissistic patient. J Am Psychoanal Assoc 55(2):503–539, 2007 17601104

Kernberg O: The concept of the death drive: a clinical perspective. Int J Psychoanal 90(5):1009–1023, 2009 19821849

Kernberg OF: Mentalization, mindfulness, insight, empathy, and interpretation, in The Inseparable Nature of Love and Aggression: Clinical and Theoretical Perspectives. Washington, DC, American Psychiatric Publishing, 2012, pp 57–79

Kernberg OF, Yeomans FE, Clarkin JF, et al: Transference focused psychotherapy: overview and update. Int J Psychoanal 89(3):601–620, 2008 18558958

Rosenfeld H: Impasse and Interpretation: Therapeutic and Anti-Therapeutic Factors in the Psychoanalytic Treatment of Psychotic, Borderline, and Neurotic Patients. London, Tavistock, 1987

Steiner J: Psychic Retreats: Pathological Organizations in Psychotic, Neurotic, and Borderline Patients. London, Routledge, 1993

Steiner J: A personal view of Rosenfeld’s contribution to clinical psychoanalysis, in Rosenfeld in Retrospect: Essays on His Clinical Influence. London, Routledge, 2008, pp 58–84

Stone MH: Abnormalities of Personality. New York, WW Norton, 1993


Only gold members can continue reading. Log In or Register to continue

Mar 29, 2020 | Posted by in PSYCHIATRY | Comments Off on An Overview of the Treatment of Severe Narcissistic Pathology
Premium Wordpress Themes by UFO Themes