Abstract
Freud is often negatively contrasted with object relations and relational theorists for holding to a metapsychology in which drives are understood as innate and predetermined in their development, are thought to follow the pleasure principle in a “hydraulic” manner, and are not seen as influenced by real objects. While that theory is certainly one dimension of Freud’s thinking, it is paralleled by another, quite different model positing a “mirror” relation in which affectively charged reversible self-object dyads, as opposed to purely internal dynamics, are the constituent elements of mental life. This view, more compatible with theories placing greater emphasis on the constitutive role of self-object relations, may further the reconciliation of Freudian and object-relational theories.
My intention in this paper is to bridge the abyss often seen to exist between Freud’s drive theory and later object-relational, interpersonal, and relational conceptions of a mind structured not by drives but by relations with significant others. Greenberg and Mitchell’s highly influential Object Relations in Psychoanalysis (1983) may be cited as a detailed and comprehensive statement of this dichotomization of psychoanalytic theories. Critics of such views have countered that the characterization of Freud’s theory as monadic overlooks the complexity of his writings, in which dyadic formulations are pervasive. Thus, Bachant and Richards (1988) point out that “Freud’s ideas about human development and about infantile sexuality include a theory of object relations” (p. 433), while Lothane (2003) asserts that throughout Freud’s writing “there are interpersonal, or dyadic, interactions and there are monadic, private responses to them” (p. 612). These authors concede that Freud’s theorizing is indeed generally monadic, but argue that this is offset by other strands in his writing, such as his clinical works, his case studies, and his developmental ideas. And of course there is Freud’s own remark (1921) that “in the individual’s mental life someone else is invariably involved, as a model, as an object, as a helper, as an opponent” (p. 69). That a drive, in Freud’s sense, requires an object to satisfy it seems to me beyond dispute.
My intent here, then, is not to beat this dead horse, but to demonstrate, through a careful reading of several of Freud’s papers written between 1910 and 1917, that while it is quite true that in what came to be taken as Freud’s “official” or “classical” theory of the mind, which he referred to as his “metapsychology,” the drives are indeed portrayed as operating as monadic intrapsychic forces regulated automatically by the internal dynamic of the pleasure principle, as Greenberg and Mitchell maintain, there can also be found in some of these writings a very different, “unofficial” theoretical construct that, while never explicitly recognized or stated as such, is based on the conception of a dyadic relationship between oneself and another person in a way that anticipates the formulations of object relations theorists.
I will refer to this formulation, based on my reading of Freud’s covert schema, as a reversible self-object dyad, a tripartite schema consisting of oneself as a subject, an other as an object, and a transitive, affectively charged action between them that can go in either direction. Because of this reflexive, reciprocal aspect of the dyad, and also because the image of mirrors appears in the texts I will be explicating, I also refer to this as a mirror relationship. When one looks at oneself in a mirror, one can imagine either that one is oneself looking at another person who is an exact copy of oneself, or that one is the image in the mirror, and that one is looking at him/her as if one were another person seeing oneself as an other. This is the “reversible” aspect of the self-object relation that crops up in Freud’s writing during the period when he was trying to elaborate a complete theory of the mind implied by his clinical findings, while also trying, not entirely successfully, to find a place in that theory for the novel concept of narcissism. This is relevant to the appearance of the self-object reversible dyad: as Ovid tells us, Narcissus himself, the original narcissist, saw An image in the pool, and fell in love With that unbodied hope, and found a substance In what was only shadow. . . . . Foolish boy, He wants himself, the loved becomes the lover [c. a.d. 8, p. 70; emphasis added].
When the loved becomes the lover, or vice versa, this in a phrase is the essence of the dynamic aspect of the reversible self-object dyad.
Before proceeding, I want to stress that my method here is simply to read some Freudian texts and reveal what is implicit in them (I will use simplifying diagrams to make these hidden concepts more plainly visible). I am in no way trying either to criticize or to endorse Freud’s ideas; I am simply finding something previously unnoticed in his thought that directly links his own subterranean, unofficial theory of the drives with the later constructs of various writers in the tradition of object relations. I am certainly well aware of the many ways in which Freud’s theories fall short from today’s perspective, as well as the ways they have been superseded or replaced by newer ideas. I do not think the mirror relationship is a comprehensive theory of the mind, of object relations, or of anything else. Its importance for me lies in its allowing us to see unity in psychoanalytic theory in Freud, and after him where others have seen an irreconcilable polarity.
The Reversible Self-Object Dyad
In a clinical vignette they characterize as “commonplace” and “very ordinary,” Greenburg and Mitchell open their book by describing a male patient who anticipates an upcoming visit from his niece and her boyfriend by cleaning up his apartment as if “preparing for the onslaught of critical, parental intruders.” In the session after the visit, however, the patient complains that his visitors were “bums” and “slobs” who only wanted to lie in bed all day. The authors observe that “the only consistency between the two sessions lies in the relationship described, that of an angry and self-righteous parent and a misbehaving, shameful child”; they continue: In the first session the guests are assigned the role of the parental figures; in the second the role is assumed by the patient. . . . [This] man’s account of an experience with other people is decisively shaped by a pattern of relationship that includes a template of the other and, in ordinary language, is “carried around in his head” [1983, pp. 10–11].
The aspect of this vignette I want to highlight is that it is the relationship that remains constant, while the question of who is playing which role in the relationship in any given moment is subject to a 180-degree reversal: one day the patient is the misbehaving shameful child, while the next day his guests are assigned that role while the patient becomes the judgmental angry parent.
I suggest that we diagram this basic building block of the patient’s internal world in schematic form:
This relationship is constant; what changes is whether the patient takes one or the other role and places the other(s) in the complementary role demanded by the schema, as in figures 2 and 3:
One can easily imagine, too, how this template might play out in the transference: in one session the patient might, for example, apologize to the analyst for putting his dirty shoes on the couch, while on another day he might point out to the analyst that the magazines in the waiting room are all out of order, or something along those lines. Again, what remains constant is the form of the relationship, which the analyst can discern at the root of these surface transformations, and which structures how this patient structures his experience of others.
This is all, as Greenburg and Mitchell say, ordinary, everyday analytic experience. Self-object dyads of this sort and their reversals form key elements in the thought of theorists including Melanie Klein, Ronald Fairbairn, and, perhaps most particularly, Otto Kernberg (2019). For Kernberg, the “drives,” as he formulates them, are accretions, formed by developmental affective experiences, of elementary units that consist of three parts: the self or subject, the object, and an affective relationship between them. In the vignette above, the relationship is not just affectively charged; it also implies an action, a transitive verb, with an actor and the one acted upon: the angry parent shames and judges the misbehaving child. Thus, in my schematic figures the third element is an arrow representing the direction of the action, a transitive verb, from the actor to the one acted upon.
We can think of these dyads as mirror relations because it is by means of them that individuals, interacting with others in their environment, overcome a purely first-person perspective and, taking the role of the other, are able to see themselves through the eyes of the other, and learn to shape themselves from and become able to communicate with others. The idea that the self or ego is not something pre-given, but is constructed in interaction with others, is at the root of the schools of analytic thought that descend from Sullivan and Fairbairn, and is also found in recent academic psychology in the work of such authors as Prinz (2012) and Rochat (2009). These authors draw inspiration from the foundational ideas of George Herbert Mead (1934) and Lev Vygotsky (1962), theorists who stress the role of mirroring by significant others in the development of mature self-consciousness and self-regulation. Indeed, the American social psychologist Charles Cooley (1902) referred explicitly to what he called the “looking-glass self” over a hundred years ago.
“Mirroring” has also featured in the discourse of psychoanalysis in the very different formulations of Winnicott (1968), Kohut (1977), and Lacan (1966). Freud too, of course, suggested that the analyst should act in a manner that, like a mirror, reflects back to the analysand his or her inner mental world. However, that is not the aspect of Freud’s thought on which I will concentrate. What I want to show is that there is a strand in the rich, variegated, and—not to put too fine a point on it—inconsistent tapestry of Freud’s writings that exemplifies and prefigures the later versions of the idea of the dyadic self-object relation. The aim is to recuperate that aspect of his thinking for psychoanalytic thought. It is often argued (e.g., by Greenburg and Mitchell themselves) that object relations theorists from Fairbairn and Sullivan onward made a radical break with Freud by discarding Freud’s “one-person psychology,” which emphasizes the internal regulation of drives understood as finite forces of energy, and by replacing the very idea of drives with a model of the internal world as composed of self- and object representations, not of drives understood as innate physical forces regulated by pre-given structures. In the interest of showing that there is more continuity of psychoanalytic thought than is sometimes implied from the relational perspective, I want to bring to the fore an aspect of Freud’s thinking that fits better with more recent understandings.
Freud’s Two Theories Of The Drives
Anyone wishing to argue that Freud’s description of the drives is mechanistic, biologistic, and impervious to the effects of external experience need only point to what is generally taken as his most important statement on the drives: in “Instincts and Their Vicissitudes” (1915), he states clearly that an instinct [trieb] . . . never operates as a force giving a momentary impact but a constant one. Moreover, since it impinges not from without but from within the organism, no flight can avail against it. . . . What does away with a need is ‘satisfaction.’ This can be attained only by appropriate (‘adequate’) alteration of the internal source of stimulation [pp. 118–119]. When . . . we find that the activity of even the most highly developed mental apparatus is subject to the pleasure principle, i.e. is automatically regulated by feelings belonging to the pleasure-unpleasure series, we can hardly reject the further hypothesis that these feelings reflect the manner in which the process of mastering stimuli takes place—certainly in the sense that unpleasurable feelings are connected with an increase and pleasurable feelings with a decrease of stimulus [pp. 120–121].
In this formulation, the object of the drive is characterized as that which is “most variable about an instinct and not originally connected with it, but becomes assigned to it only in consequence of being peculiarly fitted to make satisfaction possible” (p. 122). It is passages like these, with no room for other people or their representations and with the object seemingly relegated to a secondary and contingent status, that lead critics to reject Freud’s metapsychology in favor of a theory that places relations with other people, beginning with the parents, at the center of human mental life, reflecting humankind’s essential sociality.
And yet only a few pages further on in this same 1915 essay, Freud’s discussion morphs from a depiction of the mental apparatus on the model of a network of neurons automatically regulated by the rise and fall of instinctual quantities of energy to one in which dyadic self-object representations come into the foreground, as he turns to a discussion of the “vicissitudes” a drive may undergo. The change from activity to passivity, for example, as one of the forms of reversal of a drive, necessarily involves two people. This is clearly illustrated by two subtypes of libidinal drive, sadomasochism and scopophilia, the latter encompassing both exhibitionism and voyeurism. The sadist requires a masochist in order fulfill his desire, and vice versa; in the case of the voyeur and the exhibitionist, the former needs to see someone and the latter needs to be seen by someone.
To take the case of masochism as an example, Freud analyses the action of this drive into three moments: the first is sadism proper, the exercise of violent power upon another person, which (at this point in his own development) he takes as the primary form of the sadomasochistic drive. The second step involves giving up the object and replacing it with the self. “With the turning round upon the self the change from an active to a passive instinctual aim is also effected.” Finally, in a third step, “An extraneous person is once more sought as an object; this person, in consequence of the alteration which has taken place in the instinctual aim, has to take over the role of the subject” (p. 127).
Here, then, is a turning round of a self-object dyad of exactly the sort seen in the vignette I cited from Greenburg and Mitchell. It may be diagrammed as in the next set of figures, first in its basic, abstract form, then with actors inserted:
Figure 4 represents the basic pattern in the abstract without specifying particular persons: Person A inflicts pain on (or dominates) Person B. When the self and an other are introduced into the schema, it takes the form of figure 5.
The reversal entailed in Freud’s second step of this vicissitude can be represented in two ways, as in Figures 6 and 7:
In Figure 6, the active verb has been reversed into the passive voice. In Figure 7, the actual roles of subject and object of the active verb have been reversed.
But this does not complete the reversal. As Figure 8 illustrates, Freud’s vicissitude by which primary sadism turns into masochism can further reverse the object: that is, instead of hurting you, I hurt myself.
Here the masochist has reversed the object of the verb from the other to himself. To state it in plain language: it is as if the masochist says to his object: “You be me and I’ll be you, and you act out my sadism for me while I undergo it as you.” Then the whole schema can be internalized and enacted in unconscious fantasy with oneself in both the active and the passive role: “I hurt myself.” (This arrangement, as I assume Freud would argue, enacts the forbidden sadistic fantasy, as well as the punishment—undergoing pain—for the transgressive wish.)
It is important to note that in this discussion of Freud’s use of the reversible self-object dyad, as in subsequent discussions in this paper, I am not making any judgment regarding the validity of Freud’s understanding of the dynamics at work, in this instance the dynamics of sadomasochism. In fact, in this particular instance we know that in “The Economic Problem of Masochism,” Freud (1924) would actually reverse himself and argue that it is masochism, not sadism, that is the primary form of the drive. (We might see this reversal as itself a case of parallel process.) My aim here is rather to demonstrate that Freud sees the sadomasochistic drive as a mirror relationship, that is, a reversible self-object dyad, not to judge whether this is a correct understanding of sadomasochism.
The operative schema of the drive, then, in this model as analyzed by Freud, has three parts: self, object, and the affectively charged verb. We have seen that each of these can be reversed to produce a different surface manifestation. This is a far cry from the theory of the drive pictured as a force with a certain quantity of energy automatically regulated by the pleasure-unpleasure principle.
Freud summarizes his analysis of the drive by pointing out the three different ways in which the affect “love” can be reversed: it can be turned into hate; it can be turned from active to passive; and it can be reversed as “indifference.” It will be noted that both “love” and “hate” can as words represent either affective states or transitive actions, while “indifference” can be only an affective (here affectless) state. In my reformulation of Freud’s schema the third terms of the schema connecting the two persons are active verbs, not merely affective states. We might then redescribe “indifference” as “is indifferent to” or “ignores.” So the transformations that “I love you” undergoes would take the forms represented in the following set of figures:
OR
Summing up, Freud (1915) concludes that we shall come to better understanding of the several opposites of loving if we reflect that our mental life as a whole is governed by three polarities, the antitheses Subject (ego)–Object (external world), Pleasure–Unpleasure, and Active–Passive [p. 133].
To translate this triad of dyads into the much simpler reversible self-object dyad, one can combine the dyads subject-object and active-passive into one, since, as Figures 11a and 11b show, they are simply restatements of each other. The dyad pleasure-unpleasure refers to the affective charge of the relation joining the self and the object, that is, whether it is felt as good or bad. Thus Freud has constructed the drive to love—libido—as the interaction of three affectively charged dyads or, in my simplification, as a single self-object reversible dyad, rather than to a primary energy regulated by the mental apparatus. Whether or not this also is a “drive” is not the point, and would depend on one’s presuppositions about what a drive ought to be. What is clinically true or useful is that it acquires a compulsory status in the unconscious that determines thought and behavior.
I have demonstrated, then, that even in the course of Freud’s most “classical” formulation of the drives, a paper explicitly about the drives (or “instincts”), he has actually resorted to two entirely different theories of how a “drive” is to be conceptualized: in one case it is seen as an energy regulated by the laws of the primacy of the aim of reducing mental stimulation either to constancy or to zero; in the other it is a psychic dyadic template or “deep structure” that, through transformations of reversal wrought on its different parts, leads to the fantasies one actually encounters and analyzes to its unconscious roots in clinical practice, such as sadomasochism or scopophilia. This second theory of the “drives” could thus be said to be much more experience-near than the first, which is a hypothetical construction based on premises we now know to be inadequate: the nervous system is not an apparatus that operates only to reduce stimulation or “unpleasure”; as Freud himself ultimately recognized in “The Economic Problem of Masochism” (1924), “Pleasure and Unpleasure . . . cannot be referred to as an increase or decrease in a quantity (which we describe as ‘tension due to stimulus’)” (p. 160).
When, in the final transformation of masochism, “I” enact the fantasy by hurting “myself,” or causing myself suffering, this is a mirror relation in the sense that I have split myself into a subject and an object. The place where I can actually perceive and experience this division is in a mirror, where “I” see “myself” as an apparently external representation of me. Standing before the mirror, I observe and am myself observed. I will now show how this mirror imagery and the reversals entailed in it appear in Freud’s thinking.
Leonardo
In his essay on Leonardo da Vinci, Freud (1910) clearly recognizes and uses the mirroring self-object dyad. This is especially appropriate in the case of Leonardo, who exhibited a good deal of interest in mirrors and mirroring. For example, as Freud notes, Leonardo, who was left-handed, kept his diary in a form that reads “backwards,” from right to left, that is, in a form that allows it to be read only if held up to a mirror—indeed, this is known as “mirror-writing.” The artist also addressed himself in his diary in the second person, as if he were addressing the image of himself in a mirror, or, perhaps, as if he himself were the image in the mirror talking to his embodied self. For example, Freud quotes Leonardo as writing such diary entries as these: “Get Master d’Abaco to show you how to square the circle”; or “You have to show in your treatise that the earth is a star, like the moon or something like it, and thus prove the nobility of our world” (p. 102). Freud also notes that Leonardo instructed his pupils to hold their paintings up to a mirror in order to see how well they captured three-dimensional space on a two-dimensional plane. So Leonardo, as Freud depicts him, is someone for whom the mirror experience is important. How did this come to be?
Freud’s essay tries to construct Leonardo’s distinctive psychology on the basis of the very few known facts about his early life. (Again, before proceeding, I want to be clear that I am making no claims regarding the validity of either Freud’s depiction of Leonardo or his interpretation of Leonardo’s sexuality; my aim is solely to demonstrate Freud’s use of the reversible self-object dyad. Gulati and Pauley [2020] provide a rich contemporary perspective on Leonardo and on Freud’s paper, in which many of Freud’s premises and conclusions are countered.)
According to existing documents, Leonardo was the illegitimate son of Ser Piero da Vinci, a notary of Florence, by a single peasant woman named Caterina. At about the same time that Leonardo was born, Ser Piero married a well-born woman named Donna Alfiera, but this marriage proved childless. By the age of five, according to town records, Leonardo was living as a member of Ser Piero’s household. Freud deduces from these facts that Leonardo spent his first three or four years as the sole child of a single mother, and that he was adopted into his father’s family only when it became clear that there would be no offspring forthcoming to carry on the family name. Alone with an adoring mother (as Freud imagines the situation) and no father in his earliest youth, Leonardo was set on a path that would determine the future course of his love life and his art.
As for the former, Freud follows the authorities of his day in assuming that Leonardo never had a sexual relationship with either a woman or a man; instead in his mature years he lavished asexual (aim-inhibited) care and affection on young male pupils who were chosen for this role more on the basis of their charm and beauty than for their (modest) talent. As for the rest of Leonardo’s sexual libido, Freud assumes it was sublimated into his art and into his scientific curiosity and passion for research. Leonardo’s first works of art were sculptures of laughing young women and little boys’ heads; Freud reads this as an indication of Leonardo’s memory of a blissful early childhood with an adoring happy mother, a memory that in his mature work is re-created in the Mona Lisa smile that is to be seen not only on the face of La Gioconda but also in other paintings in which it is a beautiful young man who displays the enigmatic smile, as in his depictions of St. John and Bacchus. The sexual ambiguity implied suggests to Freud that as a small child Leonardo identified with his mother, whom he would have imagined as having the attributes of both sexes, and therefore saw both himself and his mother as androgynous.
Leonardo’s sole recorded childhood memory is of a bird forcing its tail into his mouth as he lay in his cradle. (Too much has been made of the fact that the bird Leonardo writes of was a kite, not a vulture, as Freud assumed; correcting this detail has only a minimal effect on the rest of the argument except for the part about the mythology surrounding the vulture, which is quite dispensable, and can be ignored without subtracting much from the paper.) Freud reads this scene as a memory, or fantasy, of both nursing at his mother’s breast and of fellatio, which are related for Freud, the former being the anlage for the latter. Freud assumes that there was an intense erotic attachment to the mother encouraged by the absence of a father for him and of a husband for her, and that this led to the foundational unconscious fantasy template determining the future course of Leonardo’s erotic life. In cases such as these, Freud argues, The boy represses his love for his mother: he puts himself in her place, identifies himself with her, and takes his own person as a model in whose likeness he chooses the new objects of his love. . . . What he has in fact done is to slip back into auto-erotism: for the boys whom he now loves as he grows up are after all only substitutive figures and revivals of himself in childhood—boys whom he loves in the way in which his mother loved him when he was a child. He finds the objects of his love along the path of narcissism, as we say; for Narcissus, according to the Greek legend, was a youth who preferred his own reflection to everything else . . . [p. 100].
In invoking Narcissus, Freud is directly conjuring up a mirror relation, since, as I have noted, what Narcissus fell in love with, in the myth, was his own image reflected up to him from a pool of water as if in a mirror. Through an identification of himself with his beloved mother, Freud argues, one puts another, whether real or, as in the case of Narcissus, illusory, in his own place and loves him as a combined self- and maternal representation. The lover becomes the loved, as Ovid said, and the loved becomes the lover.
The diagram of the relationship just described for Leonardo has as its abstract root form the following:
Putting this into the context of the case of Leonardo, this basic schema becomes:
At this point we can consider a new element in the template, that is, the gender of the respective characters. At first thought we might obviously describe Caterina as female and Leonardo as male. However, that would not do justice to Leonardo’s own fantasy (as Freud constructs it), according to which, as we have already surmised, both Caterina and Leonardo, in identification with her, may be thought of as possessing the physical characteristics of both sexes.
What happens next is that the original schema undergoes a reversal whereby the adult Leonardo places himself in the position of Person A, who bestows doting love on the young men he takes as his protégés, while the young men enact the role of Leonardo.
It is as if Leonardo were to say to his pupil: “We are going to enact my unconscious fantasy schema of a mother-son dyad composed of two people of ambiguous or androgynous gender; I will play the role of my doting mother, and you will play the role of me as the happy recipient of maternal love.” This would of course be quite consistent with a person who writes his diary in mirror writing, addresses himself in the second person as “you,” and employs mirrors as a pedagogical technique. The schema would be this:
If we picture Freud’s Leonardo looking into a mirror, he sees a reflection of himself that he transposes onto a young man who resembles him as a youth, and also resembles his youthful smiling mother. We might take from Winnicott (1968) the idea that the original (good-enough) mother does actually “mirror” the young child by conveying to him love and care, visible to him in her eyes and perhaps even in the reflection of himself in those eyes; and from Lacan (1966) the idea that the mother can be the original mirror of the “mirror phase,” upon whose perceived image the child models his own ego. Kohut (1977), meanwhile, showed us that the narcissist can create a “mirror transference,” by which he means an admiring attitude, in the analytic dyad. This combined fantasy self/mother is then enacted in the relational schema as a splitting of Leonardo himself into two aspects: as he experiences himself, and as he perceives himself in the mirror or in the visage of the youthful mother with whom he identifies. Like Narcissus, Leonardo falls in love with a mirror image of himself. As an adult, acting and experiencing himself as mother in the original dyad, he loves an idealized mirror image, embodied in his succession of favorites, each of whom reflects back to him an image of himself as identical with this idealized mother/self.
One can see in Leonardo’s paintings of beautiful people—the Mona Lisa, St. John, Bacchus—images both of a beloved other, and a mirror image of himself. These portraits combine the images of himself, his mother, and his young protégés, all imagined as having the androgynous characteristic that lends them much of their air of mystery. A drawing recently attributed to Leonardo, and that was therefore not available to Freud as evidence, provides a vivid visual representation of the androgyne of Leonardo’s fantasy, which could be seen to lend support to Freud’s analysis: the figure in this drawing, Angelo Incarnato (angel incarnate) has an erect penis, and appears to have drawn back an upper body garment to reveal what can be taken as either a small female breast or a man’s breast and prominent nipple. Either way, this image would seem to be the prototype of the smiling androgynous fantasy figure that reappears in so many of his paintings. (Gulati and Pauley reproduce this striking image in their paper.)
On Narcissism
The passage quoted above in which Freud describes Leonardo’s libidinal object choice as “narcissistic” marks the first appearance of that term in Freud’s published work. It is worth underlining that it appears as an illustration of the mirror relationship in the self-object dyad. As such, it is quite at odds with the picture of narcissism that is often understood as Freud’s theory of narcissism, This “classical” theory attributed to him is based on the idea of the drives operating like a flow of energy, imagined as a finite substance, through a network of channels regulated by the tendency to minimize neuronal stimulation. This theory is put forward in “On Narcissism: An Introduction” (1914), which Freud frames as an attempt to understand schizophrenia (for which he substitutes his preferred term, “paraphrenia”). According to this view, the person with this disorder “seems really to have withdrawn his libido from people and things in the external world, without replacing them in others by phantasy. When he does so replace them, the process seems to be a secondary one and to be part of an attempt at recovery . . .” (p. 74).
As for the question of what happens to the libido that is withdrawn from external objects, or more accurately from the representations of these objects, megalomania . . . points the way. This megalomania has no doubt come into being at the expense of object-libido. 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 [pp. 74–75]. Thus we form the idea of there being an original libidinal cathexis of the ego, from which some is later given off to objects, but which fundamentally persists and is related to the ego much as the body of an amoeba is related to the pseudopodia which it puts out [p. 75].
The pseudopodia of an amoeba can be retracted, and it is this retraction, leading to the engorging of the ego with all the libido that had previously been invested in representations of real-world objects, that results in narcissism or, in the more extreme case, paraphrenia or delusional megalomania. “We see . . . , broadly speaking, an antithesis between ego-libido and object libido. The more of the one is employed, the more of the other becomes depleted” (p. 76). This is clearly a hydraulic model of a materialized form of energy—a “U-tube” model, as Kohut astutely pointed out.
But just as we saw was the case in “Instincts and Their Vicissitudes,” Freud shifts gear midway through “On Narcissism,” if only briefly, to frame narcissism in terms of the self-object dyadic relation I have been describing. After arguing that a child’s first love object is the one who cares for it, feeds, it, and protects it—in the usual case, the mother—he goes on to posit another form of object choice: Side by side, however, with this type and source of object-choice, which may be called the ‘anaclitic’ or ‘attachment’ type, psycho-analytic 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, . . . that in their later choice of love objects they have taken as their 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’ [1914, pp. 87–88].
This formulation, which Freud describes as the strongest reason to adopt the hypothesis of narcissism, seems to clearly follow from and reflect his analysis of Leonardo. As we have seen in that instance, the form taken by narcissism there is the choice of an object who is a real person, but who stands in for, or embodies via projection, Leonardo’s own self-representation. This is very different from the image of an amoeba pulling in its tentacles. The latter picture evokes a reservoir of a substance charged with energy that can flow out and then be retracted. The substance is then depicted as filling to excess its original unitary container. This is not a depiction of an object relation or an object choice.
Further reflection shows that the “ego” that overinflates itself with libido is not the same thing as the “ego” that can become the object of libidinal desire. The former is the presumed subjective agency of a person who experiences and desires, and has at its disposal a certain amount of libidinal energy. The latter is a mental representation of a person who can be perceived, in this case, the ego’s “self.” That self is not subjectively experienced directly but is seen, either as projected onto a suitable other person, or in an illusory form, as in a mirror. Freud has made use of the fact that the word “ego” (and this is true also of the now widely used term “self”) has these two distinct meanings, which he can then equate on the basis of their shared name. But they are actually two quite different sorts of things.
When he returns to a consideration of narcissism in The Ego and the Id nine years after the publication of “On Narcissism,” he once again employs both the amoeba model and the object-relations model. He argues that the ego tries to draw libido to itself by disguising itself as an object by means of identification; he continues: This would seem to imply an important modification of the theory of narcissism. At the very beginning, all the libido was 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 . . . 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 [1923, p. 45; emphasis added].
Here, instead of a self that sends out pseudopods of libido into representations of an object world, and then draws them back again, the self (or ego) attracts libido to itself by passing itself off as an object. An ego or self can both experience itself directly and perceive itself—for instance in a mirror, or in the form of a real other person regarded as somehow similar to the self—as one among the various objects of perception. A perception is very different from a subjective experience of being a desiring self.
While the “libido as energy-charged substance in a state of regulated flow” model of narcissism is, as I have shown, a rather confusing concept in Freud’s exposition, it does provide the basis for a useful metaphor, as when we say of a narcissistic person that he is “full of himself” or “puffed up.” But we need to clearly distinguish a good metaphor from a good theory. The amoeba theory is the former; the mirror theory is the latter.
Schreber
Freud’s idea that narcissism and psychosis are different points along the same continuum, representing lesser or greater degrees to which the pseudopods of libido have been withdrawn into the ego, is rooted in his analysis of the case of Daniel Paul Schreber, the judge who had published a detailed account of his own mental illness. It was in a paper providing an interpretation of Schreber’s psychotic ideas that Freud (1911) tried to explain delusional psychosis in terms of the libido theory, and it was there too that his ideas about narcissism, which found full expression in “On Narcissism,” began to develop. (Once again, my discussion addresses Freud’s mode of reasoning in his paper, not the validity of his views about Judge Schreber’s sexuality and illness; these have been subjected to criticism and brought up to date, notably in Lothane’s detailed and authoritative book [2019].)
Schreber, as Freud portrays him, suffered from a delusion that began with the paranoid belief that he was being tortured and sexually abused, but evolved into the megalomaniacal conviction that he was being gradually transformed into a woman who, impregnated by God, would regenerate the entire human race, which in his fantasy had been annihilated in an apocalypse. Freud interprets the experience of the end of the world as representing the collapse of Schreber’s cathexis of external reality and his withdrawal of all libido from the objects into which it had been invested. This is the point of origin of Freud’s “amoeba theory.” Anyone reading only the first two parts of Freud’s three-part essay on Schreber would quite naturally come away with the impression that Freud’s model of the mind is that of a mental apparatus dominated by a hydraulic view of the distribution of libidinal drive-energy regulated by the pleasure principle.
In the third section, however, Freud once again changes his theoretical model over to that of the reversible self-object dyad, just as we have seen him do in “Instincts and Their Vicissitudes” and “On Narcissism.” And here Freud presents his most impressive tour de force exploiting the generative explanatory power of the mirror relationship. He begins by proposing that the various manifestations of delusional psychosis, which appear as very different kinds of fantasy formation on the surface level, can all be derived from systematic transformations of a single disavowed proposition: “I (a man) love him (a man).” That is, he supposes that the various forms of delusional psychosis known to psychiatry are all different defenses against a powerful but repressed homoerotic wish.
(While it would not be surprising for a man like Schreber in his strongly heteronormative sociocultural context to vehemently disavow any erotic attraction to a man, one can also, if one wishes, quite easily understand Freud’s interpretation against the backdrop of his own erotic infatuation with and subsequent break with Fliess, whom Freud understood as having fled from their mutual attraction into a paranoid alienation from him; the same dynamic was just then about to repeat itself in the case of Freud’s relationship with Jung. In any event, we would today be more likely to see Schreber as transsexual rather than homosexual.)
The root schema that Freud proposes to explain the manifold delusions of delusional psychosis once again has three components: a male self, a male other, and the relationship “love”:
The different delusional fantasies, then, all arise from different reversals of one or the other of these three components.
The first move of a person experiencing the classical persecutory ideation of paranoia, Freud argues, is to negate the forbidden relationship by reversing “love” into “hate,” as in figure 18:
But, Freud says, the mechanism of symptom formation in paranoia requires that internal perceptions be replaced by external perceptions, that is, by projection. Therefore the new proposition “I hate him” is further transformed by changing active to passive, that is, by reversing the direction of the transitivity of the verb: “I am hated by him,” or, more simply, “He hates me.”
OR
In Schreber’s case, as Freud sees it, the other man whom he loves and hates is the eminent Dr. Emil Flechsig, the head of the sanitarium in which Schreber was housed and successfully treated during the first episode of his illness; in Schreber’s later fantasy, Flechsig became the chief author of Schreber’s torments and tribulations.
But, as Freud goes on to argue, there are, in psychiatric classification, forms of delusion other than the persecutory one that Schreber constructed around the figure of Dr. Flechsig. One of these is erotomania, the imagined conviction that one (a man) is loved by a certain woman, often someone of celebrity status (as in the well-known instance of the actress Jodie Foster). This delusion, according to Freud, is formed by a different transformation of the original proposition “I (a man) love him.” In erotomania, the verb “love” is unchanged, but what is reversed is the gender of the object. Once again the process occurs in two steps. First the gender of the object is changed from male to female; but, once again, because paranoid thinking requires projection, according to Freud, this is further modified by the reversal of the direction of transitivity, or by the turning of active to passive, as in figures 20 and 21:
OR
The third transformation of the original schema occurs in the form of delusional jealousy. Here Freud’s analysis requires that “I (a man) love him” be altered in two respects: the subject is replaced by a different person, and that person’s sex is changed from male to female, thus:
In this case, for example, a man accuses his wife of betraying him with the man who is actually the object of his repudiated desire.
The final transformation of the root schema that Freud treats is megalomania. In this variation, Freud asserts that there is no object; that the idea is “I do not have an object, I love only myself.” But actually, contrary to Freud’s assertion, this version too does have an object, namely, the person’s own self as an object of perception rather than as a locus of desire, as I discussed in the section on narcissism.
As Freud notes, “this kind of contradiction “would give us megalomania, which we may regard as a sexual overvaluation of the ego and may thus [be] set beside the overvaluation of the love object with which we are already familiar” (p. 65).
So, once again, here megalomania as a psychotic form of narcissism is depicted as an object relation derived from a more primary real object relation—the beloved male other—that is being disavowed and defended against, not only as an overaccumulation of a quantity of libido in the self, as in the amoeba theory.
In this way, Freud has elegantly accounted for most of the psychiatric typology of the delusional psychoses as understood in the psychiatry of his day. (The list of the varieties of delusional psychosis in the current DSM is similar but not identical.) But while his analysis explains why Flechsig is turned into a persecutor in the initial phases of Schreber’s illness, the model does not account for the final form that his delusion took, that is, the fantasy that he was being transformed into a woman.
As Freud supposes, Schreber’s fantasy/delusion is constructed as a negation of, and defense against, the original ego-dystonic thought that “I, Schreber, love Dr. Flechsig”; and the reason for the distortion is to avoid acceptance of the proscribed emerging homoerotic desire. So it is quite remarkable that Freud did not seem to see that the most relevant transformation of the original unacceptable thought is the one Schreber in fact ultimately appears to have hit upon, namely:
Or, if we insert the actual players:
(Schreber in his memoir reported having overheard Flechsig referring to himself as “God Flechsig.”) It is as if Freud’s Schreber were to think: “The only way I can tolerate loving and enjoying copulation with a man is if I myself am transformed into a woman, thus preserving my heteronormative self-image.” According to Schreber’s delusional ideation, the Order of Things, ultimately originating from God, requires that he, Schreber, be transformed into a woman so that he can copulate with God and so repopulate the world. This picture is at least minimally more acceptable than the real underlying one, since it is in tune with the heteronormativity of his milieu and of his own presumed conscious self-image.
This fantasy, of being turned into a woman, is in turn directly connected with a very specific and real mirror relationship, as is clear from the following passage from Schreber’s book about himself, as cited by Freud: The only thing which could appear unreasonable in the eyes of other people is the fact . . . that I am sometimes to be found standing before the mirror or elsewhere, with the upper portions of my body bared, and wearing sundry feminine adornments, such as ribbons, false necklaces, and the like [p. 21].
And, as Freud writes of him a few pages later, By means of what he calls ‘drawing’ (that is, by calling up visual images) he is able to give both himself and the rays [of God] an impression that his body is fitted out with female breasts and genitals: ‘it has become so much a habit with me to draw female buttocks on to my body—honi soit qui mal y pense—that I do it almost involuntarily every time I stoop.’ He is ‘bold enough to assert that anyone who should happen to see me before the mirror with the upper portion of my torso bared—especially if the illusion is assisted by my wearing a little female finery—would receive an unmistakable impression of a female bust’ [pp. 32–33].
In other words: because Schreber does not have Leonardo’s gift for actually “drawing” a phallic person with an ambiguously male or female bust, as in Angelo Incarnato, he makes for himself an image of himself as the woman he needs to be to fit into his delusion by creating a visual image—by a mental “drawing”—of himself in the form of a woman, in an actual mirror which he can gaze upon—perhaps, we might speculate, as Leonardo enjoyed gazing at the models for his paintings.
It is illuminating, I think, to look again at Leonardo’s drawing of the incarnate angel with these passages from Schreber in mind.
Mourning And Melancholia
As I discussed in the case of masochism, one fate that a reversible self-object dyad can undergo is to be internalized, so that, as in this example, the self as sadist hurts the self as masochist. This vicissitude of the drive is further developed in “Mourning and Melancholia” (Freud 1917), a paper that, as Ogden (2012) shows, is often recognized as a forerunner of object relations theory. In this paper Freud argues that the reproaches directed against oneself in melancholia (depression) have their origin in reproaches against a formerly loved object that has been internalized into the patient’s own ego. This takes place via identification with the lost object, in which the libido formerly invested in the object is now used for identification: the free libido was not displaced onto another object; it was withdrawn into the ego. There, however, it was not employed in an unspecified way, but served to establish an identification with the abandoned object. Thus the shadow of the object fell upon the ego, and the latter could henceforth be judged by a special agency, as though it were an object, the forsaken object. In this way an object-loss was transformed into an ego-loss and the conflict between the ego and the loved person into a cleavage between the critical activity of the ego and the ego as altered by identification [p. 249].
In other words, in response to the loss of a lost object, the self-object dyad, “I reproach you”
is transformed into “I, as critical agent, reproach you, as me identified with you,” as in Figure 25:
Here the ego is split into two parts, and one part, playing the role of the subject-self, rails against the other part, the object-self, which is set up within the ego as the lost object. And since this critical agency is a precursor of the concept of the superego, which will be formulated six years later in The Ego and the Id, and since it is a parental introject, what we have is an inner representational world in which a parent reproaches a child: exactly the dyad in the example with which I began this paper— the patient and the visiting niece (and her boyfriend)—except that both are now understood to be part of the inner world of the sufferer. The melancholic person, in splitting the ego, acts out the scenario not, as Leonardo did, with an actual other person as a stand-in for himself, or as Schreber did by looking at himself in the mirror in the form of a woman, but wholly within his own psyche.
What is new here is that this process is not just a construction characterizing certain categories of particular individuals, but is the regular basis for the formation of an enduring structure, the superego, which stands or hovers over the ego and judges it, and is a major part of the structural model of the mind. One can draw a direct line of influence from this model to Fairbairn’s model of psychic structure (1944); Fairbairn himself says as much. This latter model posits two “mini-egos” split off from the main ego, each in a dyadic relation with an object. The resulting two dyads are the “libidinal ego and the exciting object,” on the one hand, and the “internal saboteur and the rejecting object” on the other. One could also see these two as the same dyadic mirror relationship but with a different affective valence: love in the first instance, hate in the second. It is far beyond my scope here to launch any further into an exposition of Fairbairn’s thought here, but my point is that it is not too difficult to link Fairbairn up with Freud’s reversible dyad model, rather than seeing them as radically irreconcilable, as would be the case if one recognized only Freud’s tension-reduction model of the drives, as Greenberg and Mitchell did in their ambitious book.
Conclusion
The reversible self-object dyad is an aspect of mental life that underlies and structures the various behaviors, symptoms, and syndromes that can be generated from it through transformations that can be both observed clinically and described clearly in schematic form, as I have done here. All object-relational, interpersonal, and relational approaches to analysis use some form of it, explicitly or implicitly, since they deal with the self’s relations with object representations in what Sandler (1994) described as their representational worlds. If they did not, Greenberg and Mitchell would not have begun a book advocating for object relations theories with a vignette that perfectly illustrates it. That Freud also thought this way clinically is evident, as many authors have pointed out, from his case studies, in which he certainly is focused on the unconscious fantasies of patients about significant others in their lives. I might cite for example the Rat Man’s fear of his own aggression as a defense against an underlying disavowed fantasy scheme taking the reversible forms “Father beats me” and “I beat Father” (Freud 1909).
In his attempt to develop a metapsychology abstracted from real clinical situations, however, Freud relied on a model of the drives that seems to operate independently of relations with other people; and that in any event, in its depiction of the workings of the mind as determined solely by the dynamic interplay of quantities of neuronal energy and the overriding motivation to reduce stimuli according to the pleasure principle, is so unworkable that he himself renounced major parts of it, as I discussed earlier in relation to “The Economic Problem of Masochism.” What I hope to have shown, however, is that even in the centrally important statement of his theory in “Instincts and Their Vicissitudes,” and elsewhere, he also employed a very different model, that of the affectively charged reversible self-object dyad or mirror relation; and that this model is much more compatible with contemporary psychoanalytic thought than what is taken to be the official one.
What has continued to divide the classical and the object relations perspectives, at least as Greenburg and Mitchell see it, is the question of whether the drives are endogenous from the start, or whether what we encounter as “drives” are built up in the course of development, for example through the accretion of affectively colored dyadic self-object relations, as Kernberg proposes. In their critique of Kernberg’s theory that, as they put it, “experience gives rise to motivational systems which are expressed in sexual and aggressive aims,” Greenburg and Mitchell remark, “Good theory can be derived from these premises; good classical psychoanalytic drive theory cannot” (1983, p. 340). Perhaps; but Freud had provided all along an alternative model that in its formal structure is much more compatible with theories like Kernberg’s, as I hope to have shown here.
This paper, then, is a contribution to an effort to see psychoanalysis as a unified and constantly developing body of thought and practice, rather than as being divided by incompatible polarities. Many of Freud’s original ideas, to be sure, have had to be augmented, reinterpreted, or even rejected as the field has evolved since its beginnings well over a century ago; but at this moment in its history I believe it is constructive to see our field as a large and wide-ranging one with many components, old and new, and to appreciate the commonalities that have characterized our field from Freud’s time to the present.
Footnotes
Supervising and Training Analyst, Emory University Psychoanalytic Institute; Adjunct Professor, Department of Psychiatry and Behavioral Sciences, and Charles Howard Candler Professor of Anthropology and Interdisciplinary Studies, Emory University.
Submitted for publication May 31, 2020.