Narcissism in Freud
WILLY BARANGER
In psychoanalytic theorizing, the concept of narcissism occupies a position similar to that of identification: both led to a profound restructuring of psychoanalytic theory. Identification gave rise to a radically different view of psychic structure when it was discovered that psychic structure stemmed largely from the vicissitudes of the object relation by way of the structuring role of identification. Once introduced, narcissism completely overturned the theory of instincts; the ultimate root of psychological conflict now became situated in the struggle between libido and destructiveness, Eros and Thanatos.
The concept of narcissism has another aspect, however, which is highly relevant to our present subject. The theory of narcissism directly affects the concept of an object and that of psychic agencies (the ego and even the superego). It also raises some extremely complex problems. Only the most scrupulous examination of Freud’s numerous references to narcissism can yield an idea of the multiplicity of meanings he assigned to the term, as well as the labyrinth of inherent theoretical problems. Even the exact sense in which he used the term is not simple. Freud introduced the concept into psychoanalytic theory in 1909 or 1910, but its use gradually increased until it eventually embraced phenomena apparently inconsistent with its original meaning. At the same time, as frequently occurs in the evolution of Freud’s thought, the new concepts coexisted with formulations that they should logically have supplanted. Two questions must therefore be answered prior to any theoretical study: What made it necessary for Freud to introduce the concept of narcissism? What are the major stages in the evolution of this concept?
The first question can be answered quite easily. The need to introduce the concept stemmed first of all from the study of homosexuality in Freud’s essay on Leonardo da Vinci. Another important factor was Freud’s growing interest and that of some of his disciples in what we would nowadays call “psychotic states”—an interest that culminated in Freud’s admirable analysis of Schreber’s memoirs. The relevance of these two sources that necessitated the introduction of the concept persists.
The historical development of the concept involved a drastic change of direction, as was noted by Laplanche and Pontalis1 and by Strachey.2 In spite of these authors’ efforts to resolve the contradictions introduced by Freud in The Ego and the Id (1923), which represented a far-reaching modification of the concept, they end with a question that leaves the problem unsolved. Again, what are we to make of the contradiction in Abraham’s thinking,3 with his unqualified espousal of the “simple” solution that there is an auto-erotic phase in the development of the libido followed by a narcissistic phase characterized by oral incorporation of the object?
The concept of narcissism turns out to be one of the most problematic and obscure in all psychoanalytic theory. We must therefore begin with a purely semantic approach, examining individually the many meanings of the term in Freud’s work and then considering the problems they raise.
THE NINE SENSES OF THE TERM “NARCISSISM”
Nine meanings of “narcissism” can be identified and classified in three groups of three. The first group relates essentially to narcissism as one of the forms or vicissitudes of libido. In the second group, the emphasis falls on the object in the narcissistic states, and the problems of narcissism come together with those of identification in its introjective form. The final group consists of extensions of the term to refer to attitudes, feelings, and character traits indicative of the valuation, devaluation, or overvaluation of some aspect of the person.
The following meanings are subsumed in the term “narcissism”:
1. A developmental stage of the libido characterized by the concentration of all libido within or toward the ego. In this sense, it denotes an intermediate stage between a “phase” regarded as autoerotic and a phase of object choice (S.E. 14: 69). Thus all the stages in which the libidinal interest of the ego is detached from the external world—in particular, sleeping (S.E. 14:225), psychosis (S.E. 12:72), fetal life as it is assumed to be (S.E. 14:222)—will be described as narcissistic.
2. The processes that make this state possible. For instance, we speak of “primary narcissism,” the concentration of libido within the ego (or, after Freud introduced the differentiation between the ego and the id, the concentration of the id’s libido in the ego), or the withdrawal toward the ego of libido formerly directed toward external objects: “secondary narcissism.”
3. The point of fixation corresponding to this stage of development, which is involved both in the predisposition to homosexuality and in the etiological equation of the “narcissistic neuroses” (psychoses in present-day terminology).
4. The term is used in a different sense in the phrase “narcissistic object choice,” which means that the subject chooses his object on the basis of his own characteristics, in accordance with some actual feature of his own being (for example, his sex) or with what he has been or would have liked to be, and so on. (S.E. 11:100).
5. Similarly, the situation of narcissistic choice may be introjected (“narcissistic identification”; S.E. 14:250). Here, the “narcissism” refers not directly to the ego but to its ego-ideal (S.E. 14:94) or the idealized object, which it imitates and whose commands it obeys. The center of narcissism in this case consists not of the ego but of the superego or ego-ideal, to which it tries to adjust and which alone is truly admirable.
6. By extension, the term “narcissism” is used for a set of attitudes, states, and even character traits extending from simple self-regard to megalomaniacal omnipotence on the part of the subject (S.E. 12:72), by way of all degrees of self-evaluation or overvaluation of some characteristic of the subject or of himself as a whole. Narcissism thus refers here to the subject’s pride in his or her beauty {S.E. 12:138), to the overvaluation by children and “primitives” of the power of their own thoughts (S.E. 13:89-90), to a characteristic of the psychology of women: the wish to be admired and loved (S.E. 14:253), the overvaluation of the penis by men, and so on. Freud even describes various forms of a character type whose dominant feature is the “narcissistic” nature of its libido.
7. Everything that reduces the self-regard of the ego or its feeling of being loved by valued objects is called a “narcissistic wound.”
8. There is a reference to the “narcissism of small differences” and even to the “small differences” between man and woman (S.E. 11:199).
9. Finally, mention must be made of perverse narcissism, which gave its name to the other phenomena and consists in taking one’s own body as an object of contemplation and love.
NARCISSISM AND AUTOEROTISM
It is interesting to note the chronological distance between the introduction of the concepts of autoerotism and narcissism. The former appears for the first time we know of in a letter from Freud to Fliess dating from 1899; the latter does not occur until 1909—five years before its official “introduction” into analytic theory. The first text suggests that the concepts of autoerotism and narcissism were originally undifferentiated and that it was only in 1909 that their differentiation became necessary in Freud’s thought. As we shall see, he did not achieve this differentiation without confronting a number of difficulties:
The lowest sexual stratum is auto-erotism, which does without any psychosexual aim and demands only local feelings of satisfaction. It is succeeded by allo-erotism (homo- and hetero-erotism); but it certainly also continues to exist as a separate current… . Paranoia dissolves the identification once more; it re-establishes all the figures loved in childhood which have been abandoned … and it dissolves the ego itself into extraneous figures. Thus I have come to regard paranoia as a forward surge of the auto-erotic current. (S.E. 1:280)
This passage is truly prophetic of a number of subjects developed fifteen years later in “On Narcissism: An Introduction”; it clearly implies, on the one hand, the concept of autoerotism as a phase of libidinal development and, on the other, the inclusion in the situations constituting this phase of certain relations with objects. This was to induce Freud years later to differentiate the concept of narcissism. From this point of view, narcissism is clearly separated from autoerotism when objects enter into the theory, in regard to the homosexuality of Leonardo da Vinci and Schreber’s psychosis —that is, when the demand for clinical understanding arises at the point where the differentiation of the libidinal quality can no longer dispense with the relation to the object and its structure.
It might appear that the introduction of the concept of narcissism into analytic theory does not greatly disturb the theory and might on the contrary help clarify it. This is not the case. As the following quotations show, Freud oscillates among a number of conceptions of autoerotism, narcissism, and the relations between them without any clear sign of a chronological development (except perhaps a tendency to abandon autoerotism as a phase after 1929; as we shall see, however, this oscillation is replaced by another that is just as important). In some passages Freud considers autoerotism to be a phase of development of the libido prior to that state of narcissism. In others he regards it as a mode of satisfaction characteristic of the narcissistic phase. In some of his works he defines autoerotism as the absence of objects; in others he recognizes that it coexists with object relations, or even that it appears after these relations have been established.
Autoerotism as a Phase
This was Freud’s first idea (see the letter to Fliess dated 1899 quoted above). He returned to it in 1910: “Since at this first phase of infantile sexual life satisfaction is obtained from the subject’s own body and extraneous objects are disregarded, we term this phase (from a word coined by Havelock Ellis) that of auto-erotism” (S.E. 11:44).
He returned more systematically to this idea of development in his study of Schreber:
Recent investigations have directed our attention to a stage in the development of the libido which it passes through on the way from autoerotism to object-love. This stage has been given the name of narcissism… . There comes a time in the development of the individual at which he unifies his sexual instincts (which have hitherto been engaged in auto-erotic activities) in order to obtain a love-object; and he begins by taking himself, his own body, as his love-object, and only subsequently proceeds from this to the choice of some person other than himself as his object. (S.E. 12:60-61)
Logically enough, he chooses an external object very similar to himself (of the same sex) and can only then move on to a heterosexual choice. Here Freud clearly delineates a development that breaks down into four phases: autoerotism, narcissism, homosexual object choice, and heterosexual object choice. The same picture is presented in Totem and Taboo (S.E. 13:88-90). In “On Narcissism,” Freud gives a clear statement of the need to distinguish autoerotism and narcissism as developmental stages: “A unity comparable to the ego cannot exist in the individual from the start; the ego has to be developed. The auto-erotic instincts, however, are there from the very first; so there must be something added to auto-erotism—a new psychical action—in order to bring about narcissism” (S.E. 14:77).
Autoerotism as a Type of Behavior Characteristic of Narcissism
Freud writes: “Originally, at the very beginning of mental life, the ego is cathected with instincts and is to some extent capable of satisfying them on itself. We call this condition ‘narcissism’ and this way of obtaining satisfaction ‘auto-erotic’ At this time the external world is not cathected with interest (in a general sense) and is indifferent for purposes of satisfaction.” (S.E. 14:134). The contradiction between this passage and those quoted above (dating from the same period) is flagrant. And the following passage makes matters even more complicated:
We have become accustomed to call the early phase of the development of the ego, during which its sexual instincts find auto-erotic satisfaction, “narcissism,” without at once entering on any discussion of the relation between auto-erotism and narcissism. It follows that the preliminary stage of the scopophilic instinct, in which the subject’s own body is the object of the scopophilia, must be classed under narcissism, and that we must describe it as a narcissistic formation. (S.E. 14:132)
Here appears what gives rise to most of the difficulties in the elucidation of the relations between autoerotism and narcissism: the lack of definition of the concepts of the ego and of the subject’s own body. Scopophilia here constitutes a link between two contradictory concepts: objectively, a neonate “has” a body and “does not have” (for Freud) an “ego” as an organized agency. The body is at first the seat of all satisfactions and all pains. The ego is structured only afterward. Scopophilia, the pleasure of looking at one’s own body, represents—as Freud suggests and Jacques Lacan emphasizes (the “mirror stage”)—a crucial moment in the structuring of the ego as an organized agency possessing (or accommodated in) a complete body of its own.
Later on, in the Introductory Lectures on Psycho-Analysis, Freud again adopts this second approach:
This narcissism is the universal and original state of things, from which object-love is only later developed, without the narcissism necessarily disappearing on that account… . Many sexual instincts begin by finding satisfaction in the subject’s own body—auto-erotically, as we say—and … this capacity for auto-erotism is the basis of the lagging-behind of sexuality in the process of education in the reality principle. Auto-erotism would thus be the sexual activity of the narcissistic stage of allocation of the libido. (S.E. 16:416)
Quite probably, some of these contradictions on Freud’s part are due to an implicit change in his conceptual model. Autoerotism has a meaning within a purely “instinctual” framework; we might say that it is confined to a closed system: the body-ego with its instincts, conceived in an energetic form—that is, as processes of discharge taking place within the subject’s own body. Freud writes: “In the auto-erotic instincts, the part played by the organic source is so decisive that, according to a plausible suggestion of Fedem … and Jekels, … the form and function of the organ determine the activity or passivity of the instinctual aim” (S.E. 14:132-33). Narcissism now involves the object (it should not be forgotten that the concept itself stems from the study of object relations: homosexual choice/replacement of external objects by the ego as an object). For this reason, we should not be surprised by the oscillations in Freud’s thought concerning the relations among autoerotism, narcissism, and the object.
On the one hand, autoerotism is defined as libido without an object. On the other, Freud allows the existence of libidinal sexual objects from the beginnings of postnatal life:
At a time at which the first beginnings of sexual satisfaction are still linked with the taking of nourishment, the sexual instinct has a sexual object outside the infant’s own body in the shape of his mother’s breast. It is only later that the instinct loses that object, just at the time, perhaps, when the child is able to form a total idea of the person to whom the organ that is giving him satisfaction belongs. As a rule the sexual instinct then becomes auto-erotic. (S.E. 7:222)
In terms of genetic stages, we thus arrive at the paradoxical solution that an object stage precedes autoerotism.
Our problems do not end here, however. Logically, there is a third solution, and this too is to be found in Freud: autoerotism coexisting with object love: “Alongside these and other auto-erotic activities, we find in children at a very early age manifestations of those instinctual components of sexual pleasure (or, as we like to say, of libido) which presupposes the taking of an extraneous person as an object” (S.E. 11:44). Freud’s formulations as to timing (“the very beginning of mental life,” “the first beginnings of sexual satisfaction,” “a very early age,” and so on) are so vague that they afford no basis for an attempt at a coherent reordering that would be truly faithful to his thought. Freud is trying out solutions and formulating each of them by switching alternately from one scheme of reference to another.
But our tribulations continue. Freud’s response to these thorny problems is to introduce new concepts and new discriminations: primary and secondary narcissism, narcissistic-libido and object-libido—all these complicate the concept of narcissism still further. The complication is made worse with the discovery of the death instincts. It was already difficult to formulate sadism in relation to the discovery of the narcissistic state; it is even more difficult to give a coherent formulation of the compatibility of these last concepts.
PRIMARY AND SECONDARY NARCISSISM
The distinction between primary and secondary narcissism, introduced by Freud in his first great work, actually followed from earlier writings. Schreber’s megalomania had already induced Freud to think about the recovery of infantile omnipotence; it was also in connection with the latter (and possibly thinking of Schreber) that he coined the term “primary narcissism.”
[In megalomania] the libido that has been withdrawn from the external world has been directed to the ego and thus gives rise to an attitude which may be called narcissism. But the megalomania itself is no new creation; on the contrary, it is, as we know, a magnification and plainer manifestation of a condition which had already existed previously. This leads us to look upon the narcissism which arises through the drawing in of object-cathexes as a secondary one, superimposed upon a primary narcissism that is obscured by a number of different influences. (S.E. 14:75)
Freud was to return to this argument in the Introductory Lectures. In “On Narcissism: An Introduction,” he investigates the vicissitudes of this primary narcissism more thoroughly:
The development of the ego consists in a departure from primary narcissism and gives rise to a vigorous attempt to recover that state. This departure is brought about by means of the displacement of libido on to an ego ideal [Ichideal] imposed from without; and satisfaction is brought about from fulfilling this ideal.
At the same time the ego has sent out the libidinal object-cathexes. It becomes impoverished in favour of these cathexes, just as it does in favour of the ego ideal, and it enriches itself once more from its satisfactions in respect of the object, just as it does by fulfilling its ideal.
One part of self-regard [Selbstgefuhl] is primary—the residue of infantile narcissism; another part arises out of the omnipotence [Allmacht] which is corroborated by experience (the fulfillment of the ego ideal), whilst a third part proceeds from the satisfaction of object-libido. (S.E. 14:100)
This passage is very important indeed, not only because it shows very clearly the three vicissitudes of narcissistic libido but also because it accepts the narcissistic satisfaction that the subject may receive from the external world —which, as we shall see, was to lead to the distinction of two classes of libido and of object—and also because it introduces as a powerful (or all-powerful) source of narcissistic satisfactions another agency, the ego-ideal, which was to change its name and content in 1923 and become the superego. In introducing narcissism, however, Freud came upon a theoretical difficulty, because at the time he assumed, on the one hand, the existence of two groups of opposing instincts (sexual instincts and ego-instincts)—an opposition he could not abandon without relinquishing the cornerstone of his entire edifice, psychic conflict—and, on the other hand, the existence of a libido whose seat was the ego and which for this very reason proved hard to distinguish from the ego-instincts. This could not but disturb the theory profoundly.
The following passage shows an attempt to escape from the impasse:
Some of the sexual instincts are, as we know, capable of this auto-erotic satisfaction, and so are adapted to being the vehicle for the development under the dominance of the pleasure principle… . Those sexual instincts which from the outset require an object, and the needs of the ego-instincts, which are never capable of auto-erotic satisfaction, naturally disturb this state [of primal narcissism] and so pave the way for an advance from it. Indeed, the primal narcissistic state would not be able to follow the development… if it were not for the fact that every individual passes through a period during which he is helpless and has to be looked after. (S.£. 14:134)
This is a surprising conclusion: first, not all the libido can be narcissistic, but only a part of it requires an object from the outset; and, second, the ego-instincts are never capable of autoerotic satisfaction, so that they need an external object (maternal care in the state of helplessness of early infancy, as described by Freud). It is legitimate to wonder whether, in speaking of libido, which “from the outset requires an object,” and ego-instincts, which require objects for the survival of the subject, Freud is not using the term “object” in two radically different senses. He is more specific on this point in other works. In any case, the idea of a total primary narcissistic state is discarded (because obviously no one could ever get out of it).
Abraham’s model (“A Short Study of the Development of the Libido”) is also rejected from the outset because it presupposes the existence of some autoerotic phase without objects and of a narcissistic phase characterized by a “total incorporation of the object.” Now, in “On Narcissism: An Introduction,” there are objects everywhere.
A Failed Revolution
Freud’s difficulty in integrating narcissism within the opposition “ego-instincts versus libido,” as well as his immersion in the study of the psychoses, of the role of guilt in normal development and in neuroses, of mourning, and so on, induced him to undertake a radical modification of the theory of instincts and at the same time brought about a revolution in which disorder was not lacking. The revolution began in 1920 with Beyond the Pleasure Principle, when Freud put together the ego-instincts and libido as two alternative forms of Eros, two life instincts: “We are venturing upon the further step of recognizing the sexual instinct as Eros, the preserver of all things, and of deriving the narcissistic libido of the ego from the stores of libido by means of which the cells of the soma are attached to one another” (S.E. 18:52). Apart from the confusion of levels (the biological and the psychological), which is much more frequent in this work of Freud’s than in others, narcissistic libido is used in a deeper sense, arising directly from the origins of the soma, and in any case at a lower level than the ego. The revolution had begun and soon its principles were to be formulated in all their clarity, in The Ego and the Id (1923).
But in the same year, 1923, Freud published “Two Encyclopaedia Articles,” in which he described the history of his ideas, and here conceptual confusion reigns:
Narcissism. The most important theoretical advance has certainly been the application of the libido theory to the repressing ego. The ego itself came to be regarded as a reservoir of what was described as narcissistic libido, from which the libidinal cathexes of objects flowed out and into which they could be once more withdrawn. By the help of this conception it became possible to embark upon the analysis of the ego and to make a clinical distinction of the psychoneuroses into transference neuroses and narcissistic disorders. (S.E. 18:249)
In the second article, Freud refers implicitly to Beyond the Pleasure Principle, noting: “The libido of the self-preservative instincts was now described as narcissistic libido and it was recognized that a high degree of this self-love constituted the primary and normal state of things” (S.E. 18:257). Times and models are mixed up: the ego, as at the beginning of “On Narcissism: An Introduction,” is the “reservoir” of this libido (and not something at a lower level than the ego, as in the previous quotation). The self-preservative instincts become “narcissistic libido” again, as Freud momentarily forgets the essential differences he had established between the ego-instincts, whose satisfaction can never be autoerotic and requires objects from the beginning, and narcissistic libido, which by definition can be satisfied autoerotically. Yet Freud’s forgetfulness is not total, as he acknowledges only a “high degree” of self-love as the primitive stage of things, suggesting that a certain amount of the ego-instincts remains outside this situation or that a part of the initial libido is not narcissistic. We are in a tangle of contradictions.
It was legitimate to hope that they were temporary. Indeed, in the same year, 1923, in The Ego and the Id, Freud appears to have decided to take the discovery of 1920 to its ultimate conclusion—that is, to locate the “great reservoir” of narcissistic libido in the id, now instituted as an agency:
This would seem to imply an important amplification of the theory of narcissism. At the very beginning, all the libido is accumulated in the id, while the ego is still in process of formation or is still feeble. The id sends part of this libido out into erotic object-cathexes, whereupon the ego, now grown stronger, tries to get hold of this object-libido and to force itself on the id as a love-object. The narcissism of the ego is thus a secondary one, which has been withdrawn from objects. (S.E. 19:46)
We thus arrive at a revolutionary but coherent conception: the ego is no longer the “great reservoir” of libido; this is now the id. It is the id that sends out the first and most important object cathexes.
The narcissism of the first kind, the narcissism of the ego, is no longer a primary narcissism but always a secondary narcissism. It stems from objects “cathected” by the id with which the ego identifies (which it introjects). The terminology is reversed: what Freud previously called “primary narcissism” is now necessarily “secondary narcissism.”
Let us not rejoice too soon at this recovery of coherence, however. From 1923 to the end of his work, Freud returned to his first concept of narcissism, accommodating it basically in the ego and not in the id. This is confirmed by quotation after quotation: “the ego itself is cathected with libido. … the ego, indeed, is the libido’s original home, and remains to some extent its headquarters. This narcissistic libido turns toward objects, and thus becomes object-libido; and it can change back into narcissistic libido once more” (S.E. 21:118). Again: “the ego is always the main reservoir of libido, from which libidinal cathexes of objects go out and into which they return again, while the major part of this libido remains permanently in the ego” (S.E. 22:103). Even in the Outline of 1938—admittedly an important work, but one that remained unfinished and was not revised conceptually as a whole by Freud—the most significant passage is the following:
It is hard to say anything of the behaviour of the libido in the id and in the super-ego. All that we know about it relates to the ego, in which at first the whole available quota of libido is stored up. We call this state absolute, primary narcissism. It lasts till the ego begins to cathect the ideas of objects with libido, to transform narcissistic libido into object-libido. Throughout the whole of life the ego remains the great reservoir from which libidinal cathexes are sent out to objects and into which they are also once more withdrawn, just as an amoeba behaves with its pseudopodia. (S.E. 23:150)
Despite the clarification of 1923, the confusion remains unresolved. The revolution has left behind, side by side, mutually incompatible institutions (in this case, concepts). It may be that the great difficulty that prevented Freud from remaining true to his 1923 concepts was the idea that the id (because it lacked organization) could “cathect” objects. In order to love (or hate) an object, it is necessary to perceive it, to recognize it and to distinguish it from others. An entity that can do this is a subject endowed with an organized ego. Although the 1923 solution was more coherent, it came up against this problem of the subject, which we encounter upon each difficult change of direction in analytic theory.
The “Great Reservoir” and the “Amoeba”
The problem just raised—of Freud’s contradictions in the theory of narcissism—did not escape the expert and perspicacious eye of James Strachey. who tackled it particularly in Appendix B to The Ego and the Id, where he referred to “apparently conflicting views.” He ingeniously maintains that Freud, in his many descriptions of narcissism, resorts either concurrently or consecutively to two metaphors: that of the amoeba which sends out and draws in pseudopodia, and that of a reservoir which may empty itself of a content to fill something else and may at the same time take back the amount it has released, refilling itself with it. We have one biological model and one hydraulic model and this gives rise to contradictions. For this reason, it is difficult to accept Strachey’s attempt to reconcile these contradictions. Let us see.
Strachey says that the analogy of the reservoir is ambiguous: it may mean either a storage tank for a liquid (or something similar) or a source of supply of this liquid or other substances. The id might be this source and the ego the tank in which its product accumulates. In the beginning, however, the ego and the id, the tank and the source, are not differentiated; this, according to Strachey, diminishes “the apparent contradiction in [Freud’s] expression.” I in turn should like to interpret Strachey. When he speaks of Freud’s “apparently conflicting views,” he may be referring to an obvious contradiction (and this is the case) or to an apparent contradiction, which, however, is not fundamental (which is what Strachey means). I do not in general believe in the utility of biological metaphors, and still less of hydraulic ones, in psychoanalysis. In this case, the metaphors become incoherent when it is a matter of a source of energy that at the same time sends out pseudopodia.
But as Strachey says (and I agree with him), this is a minor point. What is more important is that Freud sometimes maintains that the object cathexes emanate from the id, reaching the ego only secondarily and indirectly by way of “centripetal” (to avoid the term “introjective”) identification; on other occasions he holds that the totality of the libido is conceived as going from the id toward the ego, reaching objects only indirectly.
Strachey’s argument does not seem to be illuminating here. I cease to understand when immediately afterward he states that perhaps the two processes are not incompatible and that “it is possible that both may occur.”
LIBIDO AND OBJECT IN NARCISSISM AND OBJECT RELATIONS
Of course, though Freud defines narcissism principally as a specific vicissitude of libido, it cannot be isolated from a concomitant vicissitude of objects and psychic agencies or structures. I have thus far concentrated on the former aspect, introducing the latter when necessary. I now propose to examine narcissism from the second point of view; here again, isolation is not feasible and repetition cannot be completely avoided.
The Two Types of Object Choice
As we have seen in connection with the opposition of ego-instincts and libido, Freud considers that the former require an object “from the beginning” and that the latter must also require one from the beginning in respect of a part of itself, while the other part is focused on the ego. The ego, thus converted into a repository of libido, sends out object “cathexes” that may subsequently be withdrawn toward the ego itself. There is, then, an opposition between the libido that is directed toward objects and that which is retained within the ego. There is a kind of balance between these libidinal forms, but never a strict equivalence, as a certain “amount” of narcissistic libido necessarily remains within the ego, and the libido directed toward objects may likewise qualitatively retain narcissistic features: “We see also, broadly speaking, an antithesis between ego-libido and object-libido. The more of the one is employed, the more the other becomes depleted” (S.E. 14:76).
This “simple” equivalence becomes complicated when Freud points out that the objects of the ego-instincts in turn become objects of the libido. In this case the libido becomes added to the ego-instincts, giving rise to a certain type of object choice: the “anaclitic,” or “attachment” type (Anlehnungstypus); to this he opposes a narcissistic choice—that is, the choice of an object similar to the subject (or similar to what the subject was or would like to be, or to the form in which the subject has been loved or of the same sex as the subject, and so on). Note that the word Anlehnung refers not to the relation of the libido to an object to which it is “attached” but to the attachment of the libido to an ego-instinct which leads it toward a particular object:
The sexual instincts are at the outset attached to the satisfaction of the ego-instincts; only later do they become independent of these, and even then we have an indication of that original attachment in the fact that the persons who are concerned with a child’s feeding, care, and protection become his earliest sexual objects: that is to say, in the first instance his mother or a substitute for her. Side by side, however, with this type and source of object choice, which may be called the “anaclitic” or “attachment” type, psychoanalytic research has revealed a second type, which we were not prepared for finding. We have discovered, especially clearly in people whose libidinal development has suffered some disturbance, such as perverts and homosexuals, that in their later choice of love-objects they have taken as a model not their mother but their own selves. They are plainly seeking themselves as a love-object, and are exhibiting a type of object-choice which must be termed “narcissistic.” In this observation we have the strongest of reasons which have led us to adopt the hypothesis of narcissism. (S.E. 14:87-88)
In these cases, “the path back to narcissism is made particularly easy” (S.E. 16:426).
Love obviously stems from the Anlehnungstypus and not from narcissistic choice. Freud shows the first type of choice is more characteristic of man than of woman, but this topic deserves to be treated in a different context.
In the case of the narcissistic choice (that is, the type of external object chosen) the narcissistic withdrawal is necessarily more pathogenic (and doubtless more complete); the amoeba puts out and draws in its pseudopodia, and so, metaphorically, does a human being normally, so that he can make new object choices when appropriate. In pathological mourning, in melancholia, in the “narcissistic” diseases, this ability for moving is lost and the withdrawal becomes irreversible. Freud opens here to psychological investigation an extremely fertile field. The structure of the narcissistic object still has many surprises in store for us.
Love and Narcissism
Freud’s description of the state of being in love is of very great theoretical importance. This state implies a considerable involvement of narcissistic libido, which has been deposited in the object as a condition of its gratification. Love by the object for the subject becomes indispensable to the narcissism of the subject: “A person who loves has, so to speak, forfeited a part of his narcissism, and it can only be replaced by his being loved. In all these respects self-regard [Selbstgefuhl] seems to remain related to the narcissistic element in love” (S.E. 14:98).
There is a further step, in which something different occurs, similar to a change of center of the narcissistic libido; we might call this the depositing of valued aspects of the person in the object: “When we are in love a considerable amount of narcissistic libido overflows on to the object. It is even obvious, in many forms of love-choice, that the object serves as a substitute for some unattained ego ideal of our own. We love it on account of the perfections which we have striven to reach for our own ego, and which we should now like to procure in this roundabout way as a means of satisfying our narcissism” (S.E. 18:112-13).
Another case of being in love described by Freud takes matters even further, presupposing a much more complete structural transfer, in which the ego has become radically impoverished and the ego-ideal has taken up a position within the object, investing it with all perfections. “A worm in love with a star,” as one of Victor Hugo’s characters says. This implies complex structural processes that we can understand only on the basis of work subsequent to Freud’s, such as that of Melanie Klein.
Freud shows his awareness of this complexity and of the impossibility of formulating it in exclusively instinctual terms in the following passage, to which I believe the greatest importance must be attached: “We might at a pinch say of an instinct that it “loves” the objects towards which it strives for purposes of satisfaction; but to say that an instinct “hates” an object strikes us as odd. Thus we become aware that the attitudes [Beziehungen] of love and hate cannot be made use of for the relations of instincts to their objects, but are reserved for the relations of the total ego to objects” (S.E. 14:137). Here is the solution to many of the difficulties we have been considering since the problem of the “great reservoir.” The great reservoir neither loves nor hates. The entity that loves and hates is “an” ego, a person, a subject. This essential distinction in analytic technique does not escape Freud when he writes: “Wo Es war, soil Ich werden” (Where id was, there I shall be): “Ich” (I) and not “das Ich” (the ego). The distinction is pointed out by Lacan, contrary to the usual translations of the German phrase to date.
Being in love naturally leads us to the process of idealization mentioned by Freud in many passages. It is in a way comparable to the aggrandizement of the ego observed in the megalomaniacal aspect that commonly accompanies paranoia. It is also closely bound up with the formation of the ego-ideal: “The sexual ideal may enter into an interesting auxiliary relation to the ego ideal. It may be used for substitutive satisfaction where narcissistic satisfaction encounters real hindrances. In that case a person will love in conformity with the narcissistic type of object choice, will love what he once was and no longer is, or else what possesses the excellences which he never had at all” (S.E. 14:101).
The Structuring Function of Narcissism
Freud has no doubts about the close relationship between infantile narcissism and the formation of the ego-ideal. The ideal ego (that is, the idealized ego) of the narcissistic state is the prototype of the ego-ideal, resulting from a reincorporation of the former, previously projected on to an external object.
This ideal ego is now the target of the self-love which was enjoyed in childhood by the actual ego. The subject’s narcissism makes its appearance displaced on to this new ideal ego, which, like the infantile ego, finds itself possessed of every perfection that is of value. As always where the libido is concerned, man has here again shown himself incapable of giving up a satisfaction he had once enjoyed. He is not willing to forgo the narcissistic perfection of his childhood; and when, as he grows up, he is disturbed by the admonitions of others and by the awakening of his own critical judgement, so that he can no longer retain that perfection, he seeks to recover it in the new form of an ego ideal. What he projects before him as his ideal is the substitute for the lost narcissism of his childhood in which he was his own ideal. (S.E. 14:94)
This is the process of “narcissistic identification,” which is “the older of the two” (S.E. 14:250). Freud is here hinting at a very important conclusion. Identifications (“introjective” identifications, in today’s parlance) are more easily made at first with a narcissistic basis. We might say that the object that is introjected has first been modeled in accordance with infantile necessities and is reintrojected with only slight modifications. Although the terminology differs somewhat from Freud’s, the idea is quite similar.
This ego-ideal in turn becomes the source of many satisfactions of a narcissistic character: the satisfaction of “being able to think oneself better than others” (S.E. 21:143), or “the consciousness of a difficulty overcome” (S.E. 23:118), even where this runs counter to the demands of the libido.
At this point the narcissistic state with its fantasized perfections converges with identification for the structuring of very important aspects of the personality.
NARCISSISM AND SADISM
Freud has convinced us of the validity and necessity of the concept of narcissism in order to cover a number of phenomena commonly observed in our practice. But considerable difficulty attaches to any attempt to reconcile the infantile narcissistic “perfection” and its convenient denial of a quite unpleasant real world with the internal or intrinsic existence—still more after 1920—of destructive instincts that are fundamentally incompatible with narcissistic felicity.
Freud was, of course, aware of the problem and solved it in different ways during the course of his development. All these solutions are interesting and surprise even the most assiduous reader of his work.
The subject is mentioned in “Instincts and Their Vicissitudes” (1915), where Freud compares the fates of scopophilia and sadism: “Similarly the transformation of sadism into masochism implies a return to the narcissistic object. And in both these cases [that is, in passive scopophilia and masochism] the narcissistic subject is, through identification, replaced by another, extraneous ego” (S.E. 14:132). Another quotation to the same effect is found in the same work:
If we take into account our constructed preliminary narcissistic stage of sadism, we shall be approaching a more general realization—namely, that the instinctual vicissitudes which consist in the instinct’s being turned round upon the subject’s own ego and undergoing reversal from activity to passivity are dependent on the narcissistic organization of the ego and bear the stamp of that phase. They perhaps correspond to the attempts at defence which at high stages of the development of the ego are effected by other means. (S.E. 14:132)
This would suggest a narcissistic organization of sadism. How can these quotations be understood other than as an expression of primary masochism, which is, however, explicitly denied in the same work? “A primary masochism, not derived from sadism in the manner I have described, seems not to be met with” (S.E. 14:128). In the same work, Freud puts forward an easier solution, which has been used ad nauseum by certain present-day analytical schools: “Indeed, it may be asserted that the true prototypes of the relation of hate are derived not from sexual life, but from the ego’s struggle to preserve and maintain itself (S.E. 14:138). And again: “Hate, as a relation to objects, is older than love. It derives from the narcissistic ego’s primordial repudiation of the external world” (S.E. 14:139).
Up to this point we have two alternative solutions: either there is a narcissistic organization of sadism, which implies a sadism directed toward the ego (that is, a primary masochism—a conception that is now rejected but that Freud was to arrive at later), or sadism has nothing to do with narcissism (which is essentially libidinal) and stems from the undeniable frustrations imposed by reality.
Fortunately, Freud does not content himself with the second (and easier) solution. It is only in Civilization and Its Discontents (1930) that we find another view of the matter:
It is in sadism, where the death instinct twists the erotic aim in its own sense and yet at the same time fully satisfies the erotic urge, that we succeed in obtaining the clearest insight into its nature and its relation to Eros. But even where it emerges without any sexual purpose, in the blindest fury of destructiveness, we cannot fail to recognize that the satisfaction of the instinct is accompanied by an extraordinarily high degree of narcissistic enjoyment, owing to its presenting the ego with a fulfillment of the latter’s old wishes for omnipotence. The instinct of destruction, moderated and tamed, and, as it were, inhibited in its aim, must, when it is directed toward objects, provide the ego with the satisfaction of its vital needs and with control over nature. (S.E. 21:121)
Not even this last quotation answers our original question, although it does approach an answer.
An entire line of Freud’s thought tends toward allowing that there is “a sadistic organization” in narcissism. The id, with its “death instincts” (after 1920), is a source not only of destructiveness but also of self-destructiveness. This does not result, as Freud might at some point have thought, from the ego’s need to defend itself from a more or less hostile world but from purely instinctual factors.
CONCLUSIONS
History of the Concept in Freud’s Work
The concept of narcissism has an exceedingly complicated history in Freud’s work; the line of development it follows is not straight but broken and full of fluctuations and even changes of meaning. Originally, what was subsequently to settle out as narcissism was mingled with the concept of autoerotism (1899). The process of settling out took place gradually between 1900 and 1914, by virtue of the need to take account of a number of phenomena including homosexual object choice and megalomania.
Autoerotism and narcissism then tended to become distinguished from each other. The former signified an objectless state prior to the formation of an ego and a mode of satisfaction of the libido with the subject’s own body. The latter at first connoted a relation of the libido to the external choice in which it (the libido) gave up this object and turned back upon the ego itself, which recovered a former state in which it was the prototype of all future objects.
We thus have five terms, autoerotism as a stage of the libido, autoerotism as a mode of libidinal satisfaction, secondary narcissism, primary narcissism, and the ego-instincts, which are not capable of autoerotic satisfaction or distinguishable into phases or stages as the libido is. Freud oscillated among these five terms, sometimes partially reconciling them in one way and sometimes in another (for example, by maintaining that there were two phases, one autoerotic and the other narcissistic, contemporary with the ego instincts).
In 1923, with The Ego and the Id he appeared to arrive at a systematic conception. He now posited a primary narcissism, in which all the libido was concentrated in the id, while the ego was in the process of formation. The id cathected objects, and the ego’s subsequent identification with these external objects (coupled with the corresponding orientation of the cathexis from the id toward the ego) constituted secondary narcissism; what had previously been called secondary narcissism was now referred to as primary narcissism. Autoerotism was nothing more than the mode of satisfaction of a structurally defined state, which was narcissism. At the same time, the concept of narcissism was progressively enriched, as narcissistic satisfaction could derive (1) from the id which loved the ego as it had loved external objects, (2) from the ego’s feeling of being loved by the superego, or (3) from the approval of the superego congratulating the ego for obeying its commands. The id loved the ego; the superego loved the ego; the ego loved the id and the superego. Narcissism would then be the paradisiacal harmony rediscovered between the agencies—a kind of blessing by God the Father on the lovemaking of Adam and Eve in a paradise full of food, devoid of hate, and in which all fruits were permitted.
Freud was unable to sustain the formulation of The Ego and the Id on primary and secondary narcissism or this definition of autoerotism. For reasons connected with the theoretical frame of reference and the development of the theory of instincts, the entire problem of narcissism was raised once again: all the instincts—those of the ego—were no longer fundamentally different from the libido, and, on the other hand, both were radically opposed to another group of instincts, the death instincts, which now shook the existing order from top to bottom.
The problem then became insoluble, and Freud returned to earlier definitions of autoerotism and narcissism, even including a primitive narcissistic organization of sadism (which was absolutely logical but inconsistent with many other concepts). The historical examination shows that (1) the concept of narcissism is indispensable; (2) Freud never managed to harmonize it completely with the rest of the analytic theory (which was, moreover, continuously changing); and (3) we shall never know whether the great reservoir of libido was for Freud the ego or the id.
Results of Freud’s Explorations into Narcissism
1. Paradoxically, the study of narcissism gives a fundamental boost to the study of object relations and the structure of the object.
2. A new chapter in Freudian “objectology” is opened, in particular as regards the perversions, states of being in love, groups, the psychoses, and normal development.
3. We begin to understand the relations between the structure of the object and the (specular or fanciful) characteristics of the subject himself and his agencies. Narcissism is structuring.
4. Any simplistic model of the developmental stages of the libido (Abraham) is eliminated from the beginning. Neither autoerotism nor narcissism can now be considered as other than relatively simple stages of a lineal evolution, (a) because Freud did not solve the problem of whether autoerotism should be defined as a developmental stage or as a mode of satisfaction; and (b) because a powerful antagonist of libido, Thanatos, appears after 1920, together with the inescapable idea of a “sadistic-narcissistic organization.” Had due consideration been given to this point, subsequent analytic thought would have been saved from a number of errors and impasses (I say this in full recognition of Abraham’s attempt at synthesis, which had its value, although it later became an obstacle to investigation owing to the obstinate decision of many analysts to stop thinking about it).
5. It is evident that Freud, in his last works, gradually abandoned the concept of autoerotism (except in the sense of a masturbatory mode of satisfaction).
Problems Raised by Freud
1. Freud’s oscillations and contradictions give rise to a need to redefine narcissism, not now in terms of a libidinal or “thanatic” phase but in terms of object relations.
2. The metaphors of the reservoir and the amoeba do not facilitate this redefinition. If they were eliminated, one of the bases of the contradictions in Freud’s thought would disappear.
3. Freud’s discovery of the two types of object relations (the attachment type and the narcissistic type) carries within itself the potential for extremely fruitful development provided that Anlehnung is regarded not exclusively as the attachment of one instinct to another but also as a type of relation to the object.
4. The concept of narcissism appears essentially to include a scopic element, as in the myth of Narcissus. The object of narcissism fluctuates between the body, the image of the body, the ego as an agency, and the person in some of his actual or imaginary characteristics or as a whole.
Notes
1. The Language of Psycho-Analysis (London: Hogarth Press and Institute of Psycho-Analysis, 1973), 255.
2. S.E. 19:63-66.
3. Karl Abraham, “A Short Study of the Development of the Libido, Viewed in the Light of Mental Disorders,” in Selected Papers (London, 1927), 496.