Heinz Kohut (1913-1981) started his professional life as a traditional analyst, but gradually moved away from a focus on drives and conflict related to the oedipal period to earlier developmental stages, and the establishment of the self. Kohut characterized narcissism by a:
lack of genuine enthusiasm and joy
sense of deadness/boredom
frequency of perverse activities
He believed there is a developmental need for the infant to endow caretakers, particularly the mother, with idealized capacities for power and omniscience which the infant can then identify with and borrow from. There is also the primitive need to be noticed/admired/approved in ones grandiose aspirations.
These developmental aspects of the self precede the development of drive. In pathological narcissism, there is a deficiency, an arrested development of adequate psychic structure, due to the failure of the caretaker to meet these needs, so that a crucial developmental task is left uncompleted . Traditional psychoanalysis (PSA), according to Kohut, prevents the emergence of this deficit, with its focus on conflict and the oedipal period. Further, in more traditional PSA the analyst’s muted responses repeat experiences of early deprivation.
Kohut's approach to treatment was to allow the idealizing and mirror (i.e. need to be noticed/admired) transferences to emerge in the early phases of treatment. For example, when a patient would start to say something along the lines of, "Dr. Kohut, you are the best doctor in the world," instead of questioning the need for the patient to think of him that way, or pointing out the denial of aggression and envy in such a statement, he would just let it ride. In fact, for Kohut, the emergence of such a transference early in the treatment was diagnostic of narcissistic pathology.
He also employed reconstruction, in which the inevitable failures of empathy by the analyst could be used to reexamine the original failures of empathy in the patient's life.
In this way, he believed the developmental task that had been uncompleted in childhood, the establishment of a sense of self, could now be completed in adulthood, and the patient could then go on to address less narcissistic issues.
A contrasting view of narcissism is that of Otto Kernberg (1928-).
Kernberg characterizes pathologic narcissism by an incapacity to depend on internalized good objects. These patients look depressed when they're abandoned, but what they actually feel is anger, resentment, and vengeance rather than real sadness over loss. They lack true emotional ties to others, and there is an overall sense of emptiness, and absence of genuine feeling. They lack positive feelings about their own activities. They think of themselves as denigrated, hungry, weak, enraged, fearful, and self-hating. They lack the ability to sustain relationships except as sources of admiration, and they have a tenuous hold on their self-esteem, maintaining it by depreciating others and avoiding dependency. They also experience destructive rage and envy towards those they depend on.
Etiologically, Kernberg views the self as a vital aspect of the early ego, developing originally as a fused self/object internalization. That is, the very young infant views itself as undifferentiated from the primary object, in most cases mother. On the way to thinking of itself as a separate entity, the infant internalizes this idea of itself as fused with the mother, in order retain a sense of omniscience in the face of the helplessness of being a little child. Later, in normal development, the child is able to relinquish the fusion, and can ultimately perceive both itself and the object as separate entities, each with inherent strengths and limitations.
(I'm leaving out a lot of stuff about normal internalization of, as opposed to fusion with, the object and subsequent development of the superego, but suffice it to say that Kernberg sees superego distortions in narcissistic pathology, and feels that antisocial character disorders are a subgroup of narcissistic ones).
In Narcissistic Personality Disorder, stable ego boundaries have been established, (i.e. reality testing is intact, unlike in more primitive pathology), but there is a refusal to accept the differentiation between the idealized object and the self. It's like saying, “That ideal person and my ideal image of that person and my real self are all one and better than the ideal person whom I wanted to love me, so that I do not need anybody else anymore.” These patients are often raised by parents who are cold and aggressive.
Cooper, A. M. Narcissism (1986) in Essential Papers on Narcissism, Andrew P. Morrison Editor, pp. 112-143. New York University Press
For Kernberg, the goal of treatment is for the patient to give up his yearning for perfection, accept the terror of intimacy and the reality of the other person as genuine but flawed. This is where he fundamentally disagrees with Kohut. Where Kohut encourages the idealizing and mirroring transferences, Kernberg sees the goal as undoing pathological idealizations, not encouraging new ones with the analyst. Kernberg views these idealizations as defenses against rage, greed, and emptiness, which need to be interpreted.
It's basically that Kohut and Kernberg have different ideas about the developmental problem that causes narcissistic pathology. Visually, it's like this:
Clinical examples are always helpful in elucidating theoretical concepts, and I use them in my class, but unfortunately, I can't do so here, so my apologies for that. But I hope there's at least some information that may be useful.