Jacques Lacan
Jacques Lacan; drawing by David Levine


It is commonly said, by way of depreciating Freud, that he left us not a new science of man but a new picture of man. He opened our eyes.

This is to give Freud less than his due, but also more. For there is no complete picture of man that emerges autographically from Freud’s own hand. There are several reasons for this. The chief one is that Freud never got the two sets of concerns, theoretical and clinical, between which he divided his working life, fully to cohere. And there are several ways in which this shows, one of which is the absence of any account of cognitive development—of how functions like reasoning, perception, and memory mature in the individual. Another (and related) way is the absence of any account of symbolism, of how the individual acquires and uses the system of internal representations with which he encodes reality. What Freud has left us is a sketch toward a picture of man, but he never worked this up into the finished thing.

Awareness of the need to say something about cognitive development and about symbolism is now common in the two principal schools that can make a good claim to be within the Freudian tradition: the New York school of ego psychology, and the so-called “English school” which derives from Karl Abraham and Melanie Klein. There is also an awareness that, since the two topics are connected, something needs to be said about how they connect. Does cognitive development presuppose symbolism (as philosophers tend to think), or does symbolism emerge in response to the needs of cognition (as psychologists tend to think)? And on all these topics both schools have made contributions of insight and interest.

But the thinker who would appear to have taken the challenge of making good these deficiencies most seriously is the legendary. Jacques Lacan. For many years now Lacan’s name has been widely known as that of someone who not only is a practicing analyst whose technique has been the topic of much controversy, but who has, largely through a series of seminars, magisterially conducted and faithfully recorded, 1 brought about an extensive revival of interest in Freud’s thought among French intellectuals and littérateurs. In the Anglo-Saxon world, he has been professionally taken up by some non-psychoanalysts, and he has been professionally ignored by nearly all analysts; but his name remains the best-known thing about him. Now, with the long-awaited translation first of a selection of his Ecrits, which includes his most important lectures and addresses, and then of the transcript of a seminar conducted through the first half of 1964 and put together under the title The Four Fundamental Concepts of Psycho-Analysis, something of an opportunity has been given to the English-speaking reader to assess the phenomenon.2

“Something of an opportunity.” Two things make the qualification necessary. One is that the translated work is still only a small fraction of the total output. The other is that the translator, set no easy task, can claim only partial success: he has got Lacan’s prose out of French but barely into English, with the result that the reader who can manage it would be best advised to have both text and translation in front of him and to use each to decipher the other. He may also want to consult Anika Lemaire’s study, which is agreeably modest, straightforward, and workmanlike.


If we start by thinking of Lacan’s work as an attempt to elaborate the sketch Freud left us, there are three observations to be made about the way he goes about it.

In the first place, Lacan assigns clear priority to symbolism over cognitive development. Advances in cognition depend upon the entry into symbolism. Secondly, symbolism is entered into in two stages. In the earlier stage the infant makes do with a form of pre-symbolic representation, which Lacan calls “the Imaginary,” and only in the later stage does it acquire symbolism proper, or language. Before these two stages, for which there is direct evidence, we have to guess at an inaugural phase. The newly born infant, victim of the prematurity of birth peculiar to man, is at the mercy of unbounded and unmediated instinct: it is (Lacan tells us) a broken egg, “une hommelette.” Thirdly, Lacan treats the whole process as best understood through its outcome, so that to ask at any point what stage of development the infant has reached is to ask how close it is to, or how far from, being a language-user. The crucial question now is what is language, and Lacan’s answer is that he follows Saussure, who has been the major influence upon that whole body of European thought loosely called “structuralist.”

Ferdinand de Saussure, professor of philology at Geneva from 1891 to 1913, was preoccupied all his working life with the question of the fundamental subject of linguistics: how it should be defined. The book toward which all his intellectual efforts were directed was never written and the posthumous Cours de linguistique générale on which we have to rely for his ideas is a compilation of students’ notes taken from his lectures. We remain ignorant how far the difficulties in the book stem from Saussure himself or whether they are not partially due to misrepresentation.


The central idea in the Cours, which must be authentically Saussurean, is that theoretical linguistics, as opposed, say, to various historical inquiries, fundamentally treats of the sign, and the sign is best represented in the formula

S by s

where “S” stands for “signifier” or signifiant, and “s” for “signified” or signifié.3

Saussure understood his formula to convey two essential facts about the sign. The first, which it does convey, is that the sign is a complex: it is made up of two constituents which may be distinguished though they may not be separated. The second fact, which it doesn’t seem to convey, is that the sign is inherently arbitrary. But by arbitrariness Saussure had in mind two different things which he took to be linked. He had in mind that—with the rare exception of onomatopoeia—there is no natural accord between the signifier and the signified that make up a given sign. Generally it is a convention how any language pairs off Ss and ss. Saussure also had in mind that any given signifier and any given signified have the value that they have solely because of the system to which they belong and the relative position that they occupy within it. Each signifier, each signified, is what it is because of the other signifiers used in, or the other signifieds articulated by, the same natural language. The value of the signifier and signified is differential or “diacritical.”

Saussure illustrates his formula with the example Lacan, perceiving that this example conveys the complexity of the sign and its conventional character, but barely its diacritical nature, substitutes his own example which he thinks makes the latter point more perspicuously. We may wonder if it does, just as we may wonder why Lacan thinks it an advantage of Saussure’s example that “arbre” (= “tree”) and “barre” (= “line in the formula”) are anagrams. But he does.

Set out so skeletally, Saussure’s conception of the sign presents certain fundamental problems of interpretation which further reading in the Cours doesn’t conclusively resolve, and which any account that makes use of it is therefore bound to inherit.

The most fundamental question is the most persistent, and it is just how labor is divided between signifier and signified. Over the centuries most of those who have reflected hardest upon language have had forced on them, in some form or another, a distinction which ordinary consideration of the “meaning” of a word easily overlooks. Roughly, the distinction is between that about a word which allows it to pick out things in the world and those very things (if there are any) which it thereby picks out. The distinction has been variously pinned down by the contrast between intension and extension, connotation and denotation, sense and reference. Not all these contrasts are equivalent: some theorists of language have ultimately dispensed with the distinction altogether. But the trouble with Saussure is that he gives one no clear indication how his formula stands to this tradition. Does connotation (to use one dichotomy) go on the side of signifier or signified? Or (to use another) is signified equivalent to sense, or is it just reference? In Saussure’s diagram does the drawing of the tree represent a tree or does it represent some internal representation we have of a tree?

And there are other problems. If every signifier and every signified is to be understood entirely (the crucial word) in terms of all the other signifiers, all the other signifieds, how does meaning ever get started? Will Saussure’s formula do for all signs—or has it been worked out with only one part of speech particularly in mind, i.e., that part which can occupy a subject-place in a judgment? Finally, is it Saussure’s hope that the whole of syntax can be covered by the way signifiers may be permissibly combined—and then the whole of semantics by the way signifieds get linked up by the permissible combinations of signifiers—or do we need from the start something that provides more structure, like the sentence or the fact?

It will be surprising, I have suggested, if Lacan’s account of the infant’s entry into symbolism avoids all these problems. Let us look at this account.

Lacan’s account opens, like Freud’s account of the origins of human culture in Totem and Taboo, on a single catastrophic event. In each case, if it seems mythical to assign such weight to a single event, some sense of reality may be restored by thinking of it as summing up a series of interrelated happenings.4 In Freud’s account, it is the slaying of the primal father. In Lacan’s account, it is the infant’s first sight of its own reflection, which cuts short the inaugural phase of its life and precipitates it toward language. This hypothesis of the stade du miroir was formulated as long ago as 1936 and first presented to the International Psychoanalytical Congress at Marienbad. The original paper was heavily reworked for the 1949 congress, and it is this version that appears in Ecrits.


Characteristically Lacan adduces no evidence for the significance of the stade du miroir, and it seems that the idea was first suggested to him by studies of animal behavior. Nor is the precise significance of the event all that clear. The crucial thing is that the infant is presented with an image, for it is typical of the ensuing stage, which, as we have seen, Lacan calls “the Imaginary,” that the infant lives with images or its mind is inhabited by them. Lacan gives several different descriptions of how these images function, some positive, or saying what the image does for the infant, some negative, or saying what the image doesn’t do for it. In keeping with what I have said about the nature of Lacan’s account I take the negative descriptions, which in effect say how images fall short of language, as the more fundamental or informative.

Briefly, according to Lacan, the image lacks generality. The infant’s confrontation with it is a confrontation with brute fact. What the infant cannot do is to put it to use. It can take various attitudes toward it or experience various emotions in front of it. Struggling to overcome the gulf between itself and the image, it tries to assume it or get inside it: Lacan calls this “primary identification.” Primary identification with the mirror image is going to be of major importance in the infant’s development, and it is crucial for Lacan that the pre-history of the individual—for that is where we are still at—originates in an “alienating” experience.

What the image denies the infant is just what language, once acquired, grants it. The infant gains a way of articulating reality, outer and inner, and it can now have thoughts, form desires, and enter into relations with others. And this is so because language isn’t brute; it possesses generality, it bears meaning.

A favored and ultimately highly significant way in which Lacan distinguishes between the two stages of symbol-acquisition is to say that, whereas in the Imaginary stage the infant is involved in a dyadic or two-term relation, in the Symbolic stage it is involved in a triadic or three-term relation. The two terms to the Imaginary relation are, of course, infant and image, but what is crucial is how Lacan characterizes the third term of the Symbolic relation. The three terms are infant, sign—and the Other. The Other—properly spelled, though not always by Lacan, with a capital O translating l’Autre with a capital A—is a highly powerful Lacanian concept and is notable for the voracity with which it swallows up ever new connotations.

For the moment, however, it is adequate to think of the Other as something like this: it is what mediates between the mind and the world, or it is meaning. Or, if there is a further connotation that already needs to be taken note of, it is that meaning is essentially something public. The Other is that preexistent “world of rules” into which we are born. Though the most evident influence here is Lévi-Strauss, Lacan is also in the mainstream of mid-twentieth-century philosophy which, irrespective of tradition, is intent on denying the possibility of an inherently private language. “The notion of egocentric discourse,” Lacan writes against Piaget, “is a misunderstanding.”

If, however, it is through language that the infant becomes constituted as an individual, this transformation is, according to Lacan, achieved at a price. Language makes distinctions and thus causes divisions. It splits the inner world from the outer. Within the inner world it divides the “I,” the spoken pronoun, from the self which it vainly claims to denote. Then, rustling about in the mind among materials left over from the Imaginary stage, language translates the self’s new double into the illusory ego—which, Lacan charges, American ego-psychology solemnly proceeds to study. This alienating effect is only intensified by the way in which, as the child realizes himself in language, he is thereby bound into an external and collective entity—the community—whose values and, above all, whose prohibitions he absorbs from its speech. He becomes the mouthpiece (literally) of an external agency. And, finally, if language-learning makes knowledge possible, it makes the failure to recognize or a tendency to misconstrue inevitable. Connaissance brings in its train méconnaissance.


So long as we continue to think of Lacan’s work as primarily an attempt to supplement Freud, to work up the sketch into a picture, it cannot be thought of as very original either in the materials it uses or in the ends to which it puts them.

Apart from his general indebtedness to Saussure, Lacan produces an account of the good consequences of language, or of language in its constitutive role, from which only the crassest empiricist would dissent. His account of the bad consequences of language, or its alienating effect, is borrowed, exaggerations and all, from standard Hegelianism. If there is something to the idea that language distorts even while it describes, and that, more particularly, introspective language falsifies internal reality, Lacan does not seem the man to make it clear. The need to distinguish within symbolism, taken broadly, between a primitive or more concrete and a developed or more discursive kind of representation has been felt by psycho-analysts at least since Ernest Jones; and, in the work of both Jones and Melanie Klein, the distinction has been developed not only with greater elegance but also with a regard for the clinical material that makes it necessary.

Again the two places within Freudian theory where Lacan stresses the importance of symbolism would be worth emphasizing only if one thought that they had been overlooked.

In the first place, Lacan points out that nearly all the mental phenomena that psychoanalysis deals with—desires, beliefs, anxieties, thoughts—are invariably of something. They are directed on to an object, or are what philosophers call (technically) intentional, and Lacan argues that it is hard to see how they could get their objects without the aid of symbolism. The point is sound, but has not been neglected by psycho-analysts. Indeed the dispute that has long raged within psychoanalysis between those who ascribe a very early psychosexual development to the infant and maintain that at the age of four or five months it entertains Oedipal phantasies and those who hold to a much later psychosexual development has in large part been a dispute about the age the infant can have the kind of symbolism that these phantasies require for their representation. (Incidentally Lacan identifies the “intentionalist” thesis with the denial that the mind has any biological basis. So he thinks it an obvious philosophical error to believe that “drive,” which is mental, is grounded in “instinct,” which is physical. In his glossary of terms the translator concurs, but neither of them gives an argument.)

Secondly, Lacan stresses that speech is the medium of the psychoanalytic process and that, in a session, at least as important as what is said is how or when it is said. The point is correct, but if there are readers likely to be ignorant of this, surely the best thing would have been for him to include some case histories, which Lacan never gives us. It is also true that Lacan exaggerates the point. For him speech is everything, and his neglect of the nonverbal aspects of the analytic process may be partly responsible for some of his more extreme innovations in technique: for instance, the ten-minute session.

But to criticize Lacan on the assumption that his aim is to supplement Freud’s work is beside the point. We can look at it in this way, but to do so is to leave out what is most distinctive, most original about it. Lacan’s aim is not to add on to psychoanalytic theory, it is to provide a base or ground, and this he claims to find in the theory of symbolism. His recipe is this: Take the best available theory of symbolism. (This for Lacan is the Saussurean theory.) Construct around it the most plausible account of how such symbolism is acquired. (This we have seen Lacan doing.) And from this psychoanalytic theory follows. “Everything,” he writes in The Four Fundamental Concepts of Psycho-Analysis, “emerges from the structure of the signifier.”

To this bold claim Lacan adds a coda: Freud thought so too. For years now Lacan has been saying that he is different from other psychoanalysts. But the difference, according to Lacan, lies not in his greater intelligence, or in his more powerful imagination, or in his familiarity with philosophy and mathematics and classical learning: it is not his high culture or his high spirits or the high priestliness of his personality that sets him apart—all claims for which we might be prepared. What sets him apart is his Freudian fundamentalism. It is total. Eroticizing, like other French intellectuals, le texte, Lacan exhibits himself as the Slave of the Freudian text. If Freud too thought that psychoanalysis came out of psycholinguistics, it is Lacan’s pride that it was left to him to discover this and the pride of discovery notably inflates one of the best-known écrits, the “Discours de Rome.”5

In the New Introductory Lectures Freud contrasts the different ways we would react to someone who speculates (against all good evidence) that the interior of the earth is filled with water saturated with carbonic acid and to someone who tells us that it is filled with marmalade. Lacan’s hypothesis that Freud anticipated Lacan seems to me to fall into the second category. Evidence passes it by, and our curiosity soon shifts from the hypothesis to the kind of person who put it forward. So rather than get involved in the labyrinthine reinterpretations of the text that Lacan encourages—excellent examples are to be found in “On a Question Preliminary to Any Possible Treatment of Psychosis,” which is Lacan’s reading of the Schreber case—I suggest that we let the question of fidelity to Freud take second place and ask whether Lacan’s interpretation of psychoanalytic theory is of intrinsic interest. One feature of his presentation we must be prepared for. And that is that the constant appeal to case histories, the making sense of everyday actions, the richness of psychological detail, all of which make the reading of Freud such a remarkable and vivid experience, give way, in Lacan’s prose, to a far more abstract mode of exposition. All is either argument or rhetoric.


To assess Lacan’s project it is necessary to get the resultant theory into some kind of shape, and it is a good idea to think of it as layered like a cake. At the bottom of the pan Lacan places his account of the infant’s entry into symbolism. Then he builds it up layer by layer, each more specific than that below, and the secret of the dish is that any gap or split referred to in the base gets itself reflected all the way up. When the theory is complete, the test is, does it have, in richness, in subtlety, the quality of that confected in Vienna?

The first layer that Lacan lays down concerns the affective side of the subject’s life or what could be called the “instincts” if it weren’t that for Lacan this implied an undesirable biologism. Anyhow, he is referring to aggression and sexuality, and he calls them “drives.”

In The Four Fundamental Concepts Lacan produces a general argument to show that “drives” are impossible without some progress toward symbolism. The argument runs: No drives without subjective experience, no subjective experience without some approximation to meaning. As he says elsewhere, drive “implies in itself the advent of a signifier.” But more interesting are the specific histories ascribed to aggression and to sexuality.

The origins of aggression, which are treated in the écrit “Aggressivity in Psychoanalysis,” are placed in the Imaginary stage. Aggression is the infant’s reaction to early mirror-derived images of its body. The simpler account is that the infant reacts aggressively to certain particular images it entertains which have a brutal and mutilated character. Surely such images are themselves expressions or projections of the infant’s aggression. So a more sophisticated account is that aggression is the infant’s reaction to its general relationship to images. Aggression breaks out for, having internalized its image, the infant now finds within its inner world a rival to itself—the rival in the mirror. Aggression is the infant’s response to the tensions, threats, and, above all, confusions attendant upon primary identification.

Lacan’s account of the origins of sexuality—for which a good source is “The Transference and the Drive” in The Four Fundamental Concepts—is more complex both in the materials it surveys and in the time span it allows them to work themselves out. For not only does this account spread itself forward into the Symbolic stage, it reaches back into the inaugural phase. Indeed the early start that Lacan allows sexuality ought to make it difficult for him to insist on the full dependence of sexuality on symbolism and to deny its biological base. Sometimes Lacan bows to this difficulty and settles for only a partial dependence of sexuality on symbolism. But at other times Lacan circumvents the difficulty. How does he manage this?

Lacan, as we have seen, thinks of the inaugural phase as originating in the anatomical incompleteness of the newly born infant. This incompleteness is experienced as what Lacan calls “déhiscence and what his translator (on mature reflection presumably) translates as “dehiscence”: that is, the opening-up of a gap to be filled. This sense of a gap precipitates the infant into symbolism—Lacan’s idea being, I think, that language through its capacity to represent absence offers to make good this gap. Now Lacan finds a very significant parallel between the trajectory described by the process of symbol-acquisition and the trajectory described by the sexual drive. The parallel is on a highly abstract level, but it is typical of Lacan to think that the nature of something very material like sexuality comes out clearest when thought of most abstractly. Both articulate a loop. The sexual drive sets out from some part of the subject’s body, moves outward, encircles some thing or object in the environment, controls it, and then returns with it to find satisfaction in that very part of the subject’s own body from which it set out. This is the erotogenic zone.

But what of the erotogenic zone itself? The sexual drive may be given some kind of symbol-linked function, but surely the erotogenic zone is not to be explained as a psycholinguistic phenomenon?

Freud, as we know, identified four such zones—the mouth, the anus, the phallus, the genitals. His treatment of them as functionally equivalent throughout the body’s maturation, in that the libido organizes itself around each in turn, together with his account of how one libidinal organization gives way to another underlie his spectacular extension of the concept of sexuality. Now Lacan accepts the zones that Freud identified. But he denies any biological account of how they get singled out and how they succeed one another. For him the central feature of an erotogenic zone and that which he thinks gives it its significance is—that it is a zone. The significance of the mouth, the anus, the phallus, the genitals, for the developing infant, is, in each case, that it is an area of the body marked off from those other areas which it is not. Answering an interlocutor after one of his seminars Lacan is recorded as saying:

It is precisely to the extent that adjoining, connected zones are excluded that others take on their erogenous function and become specific sources for the drive. You follow me? [Four Fundamental Concepts, p. 172]

This assimilation of the erotogenic zone to the “diacritical” sign as Saussure conceived of it is for Lacan confirmed by the way that each zone is demarcated by a rim and that sexual pleasure is always experienced at the rim. Pleasure at the rim, Lacan implies, is pleasure in the rim.

But doesn’t the reference to pleasure invalidate this whole “semiotic” account of the erotogenic zone? Surely it is the fact that pleasure can be got out of them that explains why we esteem certain parts of our body—and why we esteem different parts at different stages of development? Lacan finds it in him to deny this too, and suggests that it reverses the order of explanation. The primary item in a libidinal organization is an organ, and that we use it, for instance to gain pleasure, comes second. (In The Four Fundamental Concepts he traces the way in which the use of the eye is an antidote to the dominance of the eye in our thinking.)

The second layer that we might expect from Lacan concerns the conative side of the infant’s life; that is, the striving, effortful side of life through which the drives get realized. In the inaugural phase the infant is confined to the single conative state of Need. Need is an “intransitive” state, in that, when the newborn child has needs, there is nothing of which it can be said that this is just what it needs. This is because the infant at this stage cannot represent to itself an object. So, when the infant acquires a system of representation, we should expect it to move into a “transitive” state, or a conative state with an object, which is what Desire is. To express the dependence of Desire on symbolism, Lacan reuses his notion of the Other and says, “Desire is the desire of the Other.”

But if symbol acquisition is a prerequisite to Desire, it also puts obstacles in its way. From Need the infant may graduate to Desire, but it may be shunted into what is mysteriously called Demand. So, to be faithful to Lacan, let us consider the smooth transition from Need to Desire as a preliminary idealization, look at the obstacles across its path, and then return.

The third layer of Lacan’s theory concerns the formation of the unconscious or repression. In the écrit “The Agency of the Letter in the Unconscious” Lacan denies any original or instinctual unconscious. Everything that is in the unconscious has to find its way there. And there is only one way to get there: it must first get symbolized, and only then is it ripe for repression. In the preface provided for Lemaire’s book Lacan defines his theoretical position by contrasting it with one he assigns to the French analyst Jean Laplanche. For Laplanche the unconscious is the precondition of language; for Lacan “language is the condition of the unconscious.”

But for Lacan language is not just the precondition, it is also the content, of the unconscious. Lacan constantly says that the unconscious is like, or is structured like, a language. What he appears to think is that the unconscious is a language. It is a language having three distinctive features. In the first place, it is made up not of signs but just of signifiers. Secondly, the signifiers that make it up are those which have undergone repression and also those signifiers related to them by principles of association.

Freud too thought that unrepressed material gets dragged into the unconscious through association with the repressed. Freud specified what he thought the principles of association were. He called them “condensation” and “displacement.” Following Roman Jakobson Lacan calls his principles “metonymy” and “metaphor” and, of course, claims they are identical with Freud’s. “Metonymy” and “metaphor” seem to me to have all the disadvantages and none of the advantages of technical terms, and here I only want to point to one significant difference between Freud’s principles of association and Lacan’s. It is we, speakers of the language, who condense and who displace: we forge associations within the system of language. But for Lacan metonymy and metaphor are intrinsic features of language itself. So in holding that the unconscious is formed in accordance with such principles, he edges himself a little further toward where he wants to be: that is to say, to a view of human psychology as constituted by the impersonal reality of language.

Thirdly, the chains of signifiers that form the unconscious are inaccessible to the subject. Access to them is gained through the dialectic of the analysis which restores to the patient “true” or “full” speech. If psychoanalysis is psycholinguistics in its theory, in its technique it is speech therapy.

A slogan which expresses the Lacanian view of the unconscious and also exemplifies the Lacanian form of the slogan is “The unconscious is the discourse of the Other.” Whereupon “the Other” acquires two further connotations. It means first “the unconscious,” and then he who restores the discourse of the unconscious to its owner or “the analyst.”

Lacan sets out his notion of the unconscious in the “Discours de Rome,” “The Agency of the Letter in the Unconscious,” “The Direction of the Treatment,” and the untranslated Position de l’Inconscient: and it might be thought feasible to work back from this to his notion of repression and to reconstruct how the unconscious presupposes symbol acquisition.

One account immediately suggests itself: In repression what happens essentially is that the link between signifier and signified gets broken, and the signified slips out of the picture. The signifier now rides free, and the associative links with other signifiers, effected through metonymy and metaphor, become all-important. One signifier gets freely exchanged for another signifier in accordance with these links so that the subject loses all grip upon what his signifiers mean. He fails to understand them either when he asserts them (in speech) or when they assert themselves (in symptoms). Understanding returns only with the reconstitution of the sign in the analytic process.

Such an account has a certain amount to recommend it, including its comparative clarity, but whether we are right to attribute it to Lacan depends on how we think he interprets the Saussurean distinction between signifier and signified when he says that the unconscious is populated entirely with signifiers. Does he mean, as I take him to mean, signs which have lost their sense? Or does he interpret Saussure differently? Or is it possible, as some commentators suggest,6 that Lacan is much more casual with Saussurean terminology than his professions of discipleship prepare us for?7

Both repression and the unconscious have a dual aspect in Lacan’s theory. Once a thought is repressed, the person who has the thought does not recognize it. Additionally, he cannot recognize it as his thought. He misunderstands what his speech says, and where it comes from.

This symbiosis of alienation and repression becomes significant when we turn to the fourth layer that Lacan lays down, which concerns what he calls Need, Demand, and Desire. This is the realistic version of that simple progression from Need, which life does not offer. The crucial difference between Demand and Desire seems to be that Demand has its roots in the Imaginary, whereas Desire is structured within the symbolic order. Each has the defects of its origins, and each brings with it its own attendant dissatisfactions.

Being represented within the Imaginary order, the object of Demand is always brute. The infant demands it for its immediate allure, not because of any meaning it has for him. Accordingly, when one demand is met, a new demand is presented. Being represented within the symbolic order, the object of Desire is never brute, it is always sought after as if for its meaning. But the “as if” here is crucial. Rooted in symbolism, and therefore prone to repression, desire is essentially a substitutive phenomenon, so that one desired object always does duty for another, with a third lying in wait to take over. “Man’s desire is a metonymy,” is how Lacan puts it, adding, “however funny people may find the idea.” And this process of substitution goes back historically to the very beginnings of desire in the individual’s life, or to the earliest attempt to formulate that lack or gap, which is the original psychic representation of need. Accordingly, desire too is insatiable, but not because when one desire is satisfied a new desire arises, as with demand, but because, more radically, desires do not thus split themselves up: there is one desire, which is continuous.

There is for Lacan another dimension to the difference between Demand and Desire. So far we have contrasted them as they relate to their objects: but they also relate differently to the individual. In breaking out of the miasmic condition of Need, the infant claims recognition and love, and every demand and every desire is also a vehicle of this general claim. They express a kind of primitive assertiveness.

But they carry the claim in different ways. Demand invariably makes the claim from the outside, peremptorily, and therefore, when it gets what it asks for, this invariably seems extorted and therefore unacceptable. By contrast Desire makes the same claim from the inside, insinuatingly, in that it tries to take over the desire of the person upon whom the claim is made. (A young girl, the daughter of a paranoiac impotent father and a frightened authoritarian mother who is terrified of change, goes mad. She is retarded, her speech is incoherent, she has phobic attacks. Her madness is the assumption of her mother’s desire that nothing should be different, that mother and daughter should never be separated.)8 The fact that desire is from the beginning an encroachment upon another extends the meaning of “Man’s desire is the desire of the Other.” For, since the earliest encroachment is upon the mother’s desire, the Other is, in appropriate contexts, the mother.9

Buried in this rather confusing material—and I regard it as no accident that I have failed to find a coherent account of the distinction between Demand and Desire in any of Lacan’s commentators 10—are several reasons why Desire has ultimately more to offer the individual than Demand.

In the first place, Desire, being registered in language, can be understood. And understanding may be the best we can achieve. Secondly, the registration of Desire in language, being a social phenomenon, automatically gives the individual part of what he claims: it gives him recognition—if not love. Thirdly, however fugitive or elusive the object of desire may be, at the causal end Desire is firmly fixed. It is rooted in the original lack or manque-à-être, or (better perhaps) in that primitive phantasy in which the filling of this lack was hallucinated. This original moment of bliss, which Lacan calls “l’objet petit a” (“a” for “l’autre,” “the other” as opposed to “A” for “l’Autre,” “the Other”), and which a less abstract psychology might think of as the mother’s breast, lies at the back of all the intersubstituted objects of desire, and at one point Lacan suggests that this, the cause of the desire, may also be its (true) object. If we can only recognize this in ourselves, or that what we desire stands in for a lost object, at least we may get beyond the stop-go of “demand.”

But with four layers laid down, how far on are we toward psychoanalytic theory? Not very far, it might be said. Out of rather unpromising elements Lacan has elucidated the chief categories of the mind and the general principles of its functioning. But this is about the same point that Hegel reached by the end of The Phenomenology of Mind, starting from roughly similar material. Lacan, it is true, has updated Hegel by adding certain twentieth-century ideas about symbolism, but the enterprise he set himself was, after all, that of backdating Freud, or showing that he can be derived from the general principles of symbol acquisition. Where Freud differs from Hegel is that he described not only the structure of the mind but its content. He talked not just about the general possibilities of human development but about how these are actualized: he talked about the Oedipus complex, and castration anxiety, and penis envy, about the origins of homosexuality and about paranoia. Does Lacan think that these too can be derived from psycholinguistic material?

The answer is that he does, and accordingly the fifth layer he lays down concerns man’s psychosexual development. One peculiarity about this layer is that, though everything gets explained through language, some phenomena get deep explanations and others shallow. So the incest prohibitions, which are intrinsic trinsic to the Oedipal situation, are connected with language in the most superficial way. They are said by Lacan to be messages explicitly written into the natural languages which we all learn: with presumably the utopian consequence, convenient for Lacan, that they could just as easily be written out of these languages, should society agree.11 By contrast, the two phenomena crucial and also peculiar to the Lacanian account of psychosexuality—the phallus and what he calls the Name-of-the Father—get deep explanations. They are located within the profoundest moments of symbol acquisition. Just how is at best obscure, but also ambiguous.

What are the phallus and the Name-of-the-Father? The best way of looking at them, which Lacan encourages, is as phantasies that the infant entertains. The phallus is the earlier phantasy, originating in the Imaginary stage but persisting. The Name-of-the-Father dates from the Symbolic stage.

If we now ask what the content of these phantasies is, Lacan’s implicit answer is that they are about what their names indicate. The phallus is a phantasy about the erectile sexual organ. The Name-of-the-Father is a phantasy about the male parent: or more specifically the male parent in so far as he issues commands—and, more specifically yet, in so far as he commands in absence, or from beyond the grave.

However, each of these phantasies, bound up as it is with the most elementary movements toward expression, acquires further significance. The phallus dominates the infant’s moments of blissful merging with the mother: it is the phantasized point of union between them.12 And so it comes to stand for totality, or a state in which all is union and nothing is differentiated, and ultimately, when the symbolic stage is entered into, for the completeness of the system of signs. The Name-of-the-Father gets similarly extended. It comes to stand for rule-governed activity, and then for the supreme example of such activity, speech.

So Lancan’s implicit answer shows the importance of symbol acquisition for psychosexuality, but not its priority. Hence Lacan’s explicit answer about the content of the phallus and the Name-of-the-Father. This reverses the whole story, and makes the phallus primarily a phantasy about the totality of a symbol system and the Name-of-the-Father primarily a phantasy about the rules of language.

Plausibility apart, this explicit answer just won’t do because it renders incomprehensible many of the quasi-Freudian things Lacan says about the infant’s life.

For instance, Lacan regards phallic phantasies as, for a variety of reasons, peculiarly precarious. As they crystallize around the symbolic system, they get more precarious. But how does this precariousness evince itself? Phantasy converts itself to anxiety, and the anxiety is experienced as fear of castration. But doesn’t this require that phallic phantasies are indeed about what their name indicates?

Again, the phallus, or phallic fantasy, is at its most precarious when it collides in the infantile mind with the Name-of-the-Father. Why is this? Because the Name-of-the-Father claims the mother from the infant. Because the Name-of-the-Father seeks to subject the mother to its will. And because the Name-of-the-Father instills into the infant’s mind, alongside the warm, primitive hallucination of being the phallus, the more evolved, the more discursive, the more reality-testable thought of having the phallus. But, if we are to make sense of this collision and its baneful aspects, does this not require that we think of the Name-of-the-Father too as being about what its name indicates?

And a final consideration in favor of the implicit over the explicit answer is this: Freud, as we know, thought the appearance of the father in the infant’s awareness sets up a three-cornered conflict in its mind in which the actors are father, mother, and infant, and the stake is the infant’s sexual organ. This is the Oedipus conflict. Lacan also talks of a psychic drama in the infant’s mind. He gives it the same structure as Freud does, he gives it the same dramatis personae as Freud, and he borrows the Freudian title. The Lacanian drama is set into being by the Name-of-the-Father and it is fought over the phallus. Is this a coincidence, or does it show that whatever may be in doubt about the Lacanian scenario, the Name-of-the-Father and the phallus must be given a literal significance primarily, if an extended one derivatively? With this, the attempt to ground psychosexuality in the phenomenon of language collapses.


In expounding Lacan’s theory of the mind I have shown where Lacan takes Freud’s name in vain, which may not be serious, and I have indicated certain confusions and mistakes, which may be eliminable. The big question to which everything leads is whether, charitably read, adequately repaired, Lacan’s theory can be put beside Freud’s—which, after all, has its defects too.

My answer would be a qualified No.

The reason for the No is that Lacan’s theory lacks the explanatory force of Freud’s. Freud’s theory has the following form: It shows man to be endowed with a very complex internal structure. This internal structure changes. It matures, and also it is modified by experience which can be both of outer and of inner reality. But if experience modifies structure, structure mediates experience. It determines how man reacts to experience, and this reaction, like the experience it reacts to, can be either external or internal. Structure, experience, reaction—Freudian theory shows these to be interdependent, and yet capable of being independently studied.

Precious little of this survives in Lacan’s theory. In the first place, Lacan assigns no place to maturation. Indeed, he looks upon any attempt to treat the mental as resting upon the state of the body as an abdication of psychology.

Secondly, Lacan is extremely hazy about the internal structure that he presumes. He talks of the mechanisms of repression and rejection, and treats them as impairments of the symbolic function. But he says next to nothing about other mechanisms like introjection, projection, projective identification, which later psychoanalysts have carefully and fruitfully distinguished. And we are never told why any of the mechanisms should get employed. As long ago as 1909 Freud thought that internal conflict could not be explained simply by reference to consciousness and the unconscious, but that separate agencies in the mind had to be invoked. Lacan antedates this.

Thirdly, Lacan totally depreciates the contribution of experience to psychoanalytic explanation, and it becomes clear that the absence from his writing of case histories and clinical illustration is not just an eccentricity of presentation. It reflects his theory. For his favored form of explanation is not by reference to the internal structure of the individual plus his experience. He appeals only to how the individual is internally structured.

An artificial example might help. In her account of early development, Melanie Klein laid great emphasis upon the moment when the infant is able to conceive of whole persons, at once good and bad, and not just part-persons, some altogether good, some altogether bad. It can now recognize that it has hated the person whom it also loves, it can feel remorse, and it can desire to set the harm right. This is the “depressive” position, and Klein goes on to make use of it in order to explain how an infant who achieved this position will then respond to subsequent experiences. But if Lacan were a Kleinian, we can imagine him simply appealing to the depressive position to explain an infant’s reactions, and there would be no reference to experience. The Lacanian individual typicaly reacts to himself or to his own being rather than to what happens to or within him.

But the No I would give to Lacanian theory is a qualified No. For judgment must be qualified if it respects the difficulties inherent in any text which disdains examples, which concedes no second thoughts and denies all change of opinion, which modulates from rhetoric to buffoonery to self-pity, which is laden with formulas that respect no formation rules and diagrams that require conflicting principles of interpretation, which uses technical terms like “topology,” “metalanguage,” “Gestalt” decoratively, which is elusive and obscure, and consciously and deliberately so.

And for the reader who is still uncertain what I have in mind in talking about Lacan’s obscurity, the following examples must suffice.

1) On the connection between psychoanalysis and science:

If we can couple psychoanalysis to the train of modern science, despite the essential effect of the analyst’s desire, we have a right to ask the question of the desire that lies behind modern science. There is certainly a disconnection between scientific discourse and the conditions of the discourse of the unconscious. We see this in set theory. At a time when the combinatory is coupled to the capture of sexuality, set theory cannot emerge. How is this disconnection possible? It is at the level of desire that we will be able to find the answer. [Four Fundamental Concepts, p. 160]

2) On the mirror image:

The fact is that the total form of the body by which the subject anticipates in a mirage the maturation of his power is given to him only as Gestalt, that is to say, in an exteriority in which this form is certainly more constituent than constituted, but in which it appears to him above all in a contrasting size that fixes it and in a symmetry that inverts it, in contrast with the turbulent movements that the subject feels are animating him. [Ecrits, p. 2]

3) And:

What one ought to say is: I am not wherever I am the plaything of my thought; I think of what I am where I do not think to think. [Ecrits, p. 166]


Obscurity is not the worst failing, and it is philistinism to pretend that it is. In a series of brilliant essays written over the last fifteen years Stanley Cavell has consistently argued that more important than the question whether obscurity could have been avoided is whether it affects our confidence in the author.

Confidence raises the issue of intention, and I would have thought that the primary commitment of a psychoanalytic writer was to pass on, and (if he can) to refine while passing on, a particular way of exploring the mind. Indeed this is how Lacan himself proposes that his work should be judged. “The aim of my teaching,” he writes, “has been and still is the training of analysts.”

For decades now Lacan has been insisting that the nature of this commitment has been systematically obscured, particularly in North America. Training has become “routinized,” and analysis itself has become distorted into a process of crude social adaptation. There is much here to agree with. Yet two questions must be raised. Has Lacan devised a more effective method of training analysts? And, would one expect this from his writings?

Neither question gets a favorable answer. All reports of his training methods, over which he has now brought three distinct secessions within the French psychoanalytic movement, are horrifying.13 It is now, I am told, possible to become a Lacanian analyst after a very few months of Lacanian analysis. And what pedagogic contribution could we expect from a form of prose that has two salient characteristics: it exhibits the application of theory to particular cases as quite arbitrary, and it forces the adherents it gains into pastiche.14 Lacan’s ideas and Lacan’s style, yoked in an indissoluble union, represent an invasive tyranny. And it is by a hideous irony that this tyranny should find its recruits among groups that have nothing in common except the sense that they lack a theory worthy of their cause or calling: feminists, cinéastes, professors of literature.

Lacan himself offers several justifications for his obscurity, about which he has no false modesty. At times he says that he is the voice, the messenger, the porte-parole, of the unconscious itself. Lacan’s claim stirs in my mind the retort Freud made to a similar assault upon his credulity and by someone who had learned from Lacan. “It is not the unconscious mind I look out for in your paintings,” Freud said to Salvador Dali, “it is the conscious.”

This Issue

January 25, 1979