Seminar II – The Ego in Freud’s Theory and in the Technique of Psychoanalysis

1954 – 1955

Chapter XIV – The Dream of Irma’s Injection (conclusion)

(All quotations refer to The Seminar of Jacques Lacan: Book II: The Ego in Freud’s Theory and in the Technique of Psychoanalysis, 1954 -1955, Edited by Jacques-Alain Miller, translated by Sylvana Tomaselli, with notes by John Forrester, WW Norton: 1991)

Lacan begins the concluding session on Freud’s dream of Irma’s injection by restating what he sees as its importance in enabling Freud to take the decisive theoretical step towards the unconscious, at a time, 1895, when he is still two years away from abandoning the seduction theory. Lacan believes that the very meaning of the dream is entwined with this discovery – “The dream Freud had is, as a dream, integrated in the progress of his discovery” (p.162) – but he also suggests that despite its status as the paradigmatic dream in the history of psychoanalysis, Freud was not able to provide us with an adequate enough analysis of the dream itself. Although placing it at the start of his major work – a work that was to inaugurate psychoanalysis with a theory of dreams as its keystone – Freud could not have given us the reason for the importance of this dream at a time when his theoretical orientation was towards the effects of seduction in childhood as determinative for the different forms of neurosis in adulthood. “The value Freud accords it”, Lacan writes,

“… went far beyond what Freud himself is at this point in time capable of analysing for us in what he writes. What he weighs up, the balance-sheet he draws up of the significance of the dream is far surpassed by the de facto historical value he grants it by placing it in this position in his Traumdeutung. That is essential to the understanding of this dream.” (p.162).

It is not therefore the interpretation of the dream, its meaning, that Lacan feels accords it importance; rather it is because the dream is illustrative of the point which Freud had reached in his thought, and in particular for Lacan, the way that it throws light on the question of regression.

As Lacan points out, we have to be careful to distinguish a topographical regression from a temporal regression. Whilst the latter is the stuff of psycho-sexual development, libidinal phases and so on, this is not the regression Freud is referring to here. In its topographical sense, we are once again instead referring to Freud’s schema of the psychical apparatus found in chapter VII of The Interpretation of Dreams (SE V, p.537, 538 and 541).


On this model, excitation usually follows in a progredient direction from perception to motor discharge, but because in sleep the body is paralysed and so the excitation cannot proceed to motility, it regresses instead towards perception (Pcpt.), presented as the semi-hallucinatory images we experience when we dream. As Lacan summarises it,

“On the level of topographical regression, the hallucinatory nature of the dream led Freud, in accordance with his schema, to articulate it with a regredient process, to the extent that it would bring back certain psychic requirements to their most primitive mode of expression, which would be situated at the level of perception” (p.163).

However, Lacan adds that the images we see in dreams are not simply perceptions, as if we were going backwards in the schema above. The images themselves have too much symbolic or associative value, as Freud’s analysis of the dream of Irma’s injection shows clearly. Lacan argues, “These images are further and further away from the qualitative level on which perception occurs, more and more denuded, they take on a more and more associative character” (p.163). What Lacan finds interesting is that the associations that are made to the mnemic dream elements – Freud’s Mnem., Mnem.’, Mnem.”- lead to a point of perception. “Do we have to consider”, he asks, “that what happens at the associative levels… brings us back more closely to the primitive point of entry of perception” (p.163). When we become conscious of processes that were unconscious, consciousness would surely be at the end of the model, between preconsciousness and motor discharge; but by placing it at the opposite end of his model Freud is clearly showing that consciousness and perception cannot be equated.

Lacan returns to what he sees as the two crucial moments in the phenomenology of the dream, the first being the point at which Freud looks into Irma’s mouth, and the second being the emergence of the formula for trimethylamine. His comments about the first scene might strike us as a bit melodramatic. Looking into Irma’s mouth is what he calls,

“An anxiety-provoking apparition of an image which summarises what we can call the revelation of that which is least penetrable in the real, of the real lacking any possible mediation, of the ultimate real… something faced with which all words cease and all categories fail, the object of anxiety par excellence” (p.164)

However, as we noted last time, nowhere in Freud’s account of the dream in The Interpretation of Dreams (SE IV, p.107, p.111) does he give us the impression that there is something horrifying about the image. Moreover, the multitude of associations connected to it, and the fact that Freud goes on dreaming and producing signifiers – the doctors’ diagnoses, trimethylamine – seem out of step with the impression that this is of the order of the Real, if we understand it according to Lacan’s assertion that the in the Real “all words cease and all categories fail” (p.164).

He does also say, however, that this moment of looking into Irma’s mouth heralds what he labels “the fundamental destructuation” (p.164) of Freud’s ego in the dream. When the doctors enter the scene to confirm Freud’s diagnosis or offer their own, “there’s no Freud any longer, there is no longer anyone who can say I” (p.164), as Freud appeals to his colleagues’ professional opinions. Freud’s ego dissolves into the imagos, or semblables, out of which it has been constructed. Can we call this the ‘regression’ of the ego, asks Lacan. If there is a regression of the ego it has to regress to something, and that would seem to imply a development of the ego, possibly through developmental or libidinal stages, a trajectory which could easy be taken as normative.

Lacan is adamant that this would be a false step. Referring to Anna Freud’s The Ego and the Mechanisms of Defence (1936), he argues that “we can in no way introduce the notion of a typical, stylised development of the ego” (p.165). Lacan does not believe that a defence mechanism corresponding to a particular developmental stage of the ego can be identified, to which in turn would correspond a symptom, such that you could trace an entire neurosis back from its symptom to the corresponding point at which the development of the ego stalled. Neither can this be used to fashion a developmental theory of the instincts. On the contrary, what the dream of Irma’s injection shows us about the ego is that in the dream “We’re not dealing with an antecedent state of the ego, but, literally, with a spectral decomposition of the function of the ego” (p165). The ego is not something that flowers from a tiny bud; it is nothing more than the sum total of identifications formed at any particular point in a subject’s life. The ego therefore does not regress; it dissolves because it is constructed from images radically alien to the subject, but as Lacan says in the last chapter, “borrowed from what I would call the bric-a-brac of its props department” (p.155). As the ego falls apart, the images on which it was founded come into sharper relief, which is why Lacan says that “This spectral decomposition is evidently an imaginary decomposition” (p.165).

This is a refrain familiar to us from Lacan’s work on the mirror stage and Lacan utilises that theory again here to make the point that what happens in perception itself is modeled on the construction of the ego. The way the ego is constructed is a prototype for all relations to the external world. And man’s desire, insofar as it is never truly his own, never simply something he knows he wants, is only accessed via the mediation of the image. Although “The image of his body is the principle of every unity he perceives in objects…. he only perceives the unity of this specific image from the outside” (p.166). This seems a bit paradoxical, perhaps even contradictory. If everything I perceive has its unity thanks to being based on the image of my own unity, but this unity itself is anticipated from identifications made with objects from the outside, how do I have any sense of the unity of my own image in the first place? This is one of the difficulties in the theory of identification, and it is probably no accident that when Lacan devotes a year long seminar to the topic of identification in 1961 and 1962, he makes extensive use of topological models such as the Möbius band, the defining feature of which is the impossibility to distinguish one side from another.

Lacan seems to cling to this paradox, making it constitutive of subjectivity. The subject oscillates between between recognising the identity of his perceptions, attributing an identity to objects, and on the other hand grasping his own unity. This is a zero-sum game: he wins his own unity at the cost of that of his objects, and vice versa, despite the fact that for desire to be experienced as such it needs to be hinged to objects:

“If the object perceived from without has its own identity, the latter places the man who sees it in a state of tension, because he perceives himself as desire, and as unsatisfied desire. Inversely, when he grasps his unity, on the contrary it is the world which for him becomes decomposed, loses its meaning, and takes on an alienated and discordant aspect. It is this imaginary oscillation which gives to all human perception the dramatic subjacency experienced by a subject, in so far as his interest is truly aroused” (p.164).

Where the image is broken, where the imaginary fixing comes apart, both the ego and the object face obliteration, as it is impossible to separate them from each other, to find them a place on one side or the other of the Möbius band:

It is in the nature of desire to be radically torn. The very image of man brings in here a mediation which is always imaginary, always problematic, and which is therefore never completely fulfilled. It is maintained by a succession of momentary experiences and this experience either alienates man from himself, or else ends in a destruction, a negation of the object” (p.166).

Apropos of the question of regression, which was the subject of the chapter of this seminar which preceded Lacan’s analysis of the dream of Irma’s injection, Lacan’s central argument therefore is that if we want to know why dreams take an imaginary form, with quasi-hallucinatory representations stacked upon each other, we do not need to reach for the concept of regression, as Freud had done when he constructed the model of the psychical apparatus that he presents in seventh chapter of The Interpretation of Dreams. The scene in which Freud looks down Irma’s mouth precipitates the appearance in which the three fellow doctors appear, which demonstrates only an “imaginary decomposition which is only the revelation of the normal component parts of perception” (p.166). Perception has this imaginary character which is exactly the same as the construction of the ego, and is fragile and liable to break down. “So we do not have to look to regression for the reason why it is imaginary apparitions [surgissements] which are characteristic of the dream” (p.166). We only have an identity of perception because our relationship to the world is fundamentally narcissistic. This anchors us in the external world; our own image helps reconcile us to it. As Lacan puts it,

“If the picture of the relation to the world is not made unreal by the subject, it is because it contains elements representing the diversified images of his ego, and these are so many points of anchorage, of stabilisation, or inertia. That is exactly how I teach you to interpret dreams in supervisions – the main thing is to recognise where the ego of the subject is” (p.167).

Lacan says he finds Freud’s ego represented right throughout The Interpretation of Dreams, but in the dream of Irma’s injection in particular he says it crystallises around the rupture caused by the intrusion of the real, the moment at which “something at its most unfathomable” (p.167) appears. Nevertheless, this does not seem to be a prerequisite, as Lacan’s examples earlier in this seminar show. He tells us about two dreams: in the first, the subject dreams of a helpless infant lying on its back, flapping its arms and legs about; in the second, the subject is swimming in a sea of numbers which relate to his date of birth and age (p.41-42). Both, he says, represent the subject, and whilst he is not talking here simply about the ego, but about the subject’s “symbolic assumption of his destiny” (p.42), it is clear that Lacan is alert to the plasticity with which representations of the subject and his ego are presented in dreams. Indeed, the subject is non-reducible to the ego of the individual’s psychology. What Lacan sees in the dream is “the imaginary plurality of the subject, of the fanning out, the blossoming of the different identifications of the ego” into the persons of the three doctors (p.167). Lacan says that this subject-without-an ego, a headless or acephalic subject, is an apt representation of the Freudian unconscious – a subject that speaks without hearing the meaning of his own words, a subject supposed of a knowledge of which he himself is unaware.

If the scene in which Freud looks into Irma’s mouth is, as Lacan insists, an experience of the real, and if the appearance of the three doctors represents the dissolution of Freud’s own ego into the identifications by which it is constituted, then the mysterious appearance of the formula for trimethylamine can be seen as “The coming into operation of the symbolic function in its most radical, absolute, usage”, such that it “ends up abolishing the action of the individual so completely that by the same token it eliminates his tragic relation to the world” (p.168). Taking into account Freud’s suggestion that the dream fulfills the wish of removing culpability for the failure of Irma’s treatment which he believes his friend Otto tacitly blames him for, this can be read as the most extreme solution to Freud’s predicament – a total abolition of the individual, an “ataraxia in which any individual is justified” (p.168), as Lacan calls it. But we can still ask why the formula for trimethylamine appears and not something else? If the symbolic function manifests itself so starkly, surely this does not take an arbitrary form?

Lacan however turns to explore the junction between the imaginary and symbolic. Recalling his slightly absurd analogy from an earlier chapter (p.51), in which he imagines an automaton that depends for its continued movement on the perception of another machine having reached that stage, he claims we come across the same thing in humans. The imaginary relation is hinges on a single opposition – same or different, you or me – which is a zero-sum game. This entails that “On the imaginary level, the objects only ever appear to man within relations which fade. He recognises his unity in them, but uniquely from without. And in as much as he recognises his unity in an object, he feels himself to be in disarray in relation to the latter” (p.169). The unity of the image of oneself therefore comes at the expense of a fundamental discordance, a “lack of adaptation… characteristic of the instinctual life of man” (p.169).

Freud had noted in his 1919 essay on ‘The Uncanny’ the fundamental ambivalence to the image. In discussing Otto Rank’s article on ‘Der Doppelgänger’ published in the journal Imago in 1914, he says that the ‘double’ was “originally an insurance against the destruction of the ego” (SE XVII, p.235), and is perhaps the reason why the Ancient Egyptians made images of the deceased from such durable materials. But he also notes that “when this stage has been surmounted, the ‘double’ reverses its aspect. From having been an assurance of immortality, it becomes the uncanny harbinger of death” (ibid). We still find this idea in folklore – that the encounter with one’s double is a sign of impending death.

So perhaps we can assert that the terrifying “point of entering the order of anxiety… [the] drawing nigh of the ultimate real” (p.166) is not quite so separate from the imaginary order as we might expect, given that it is necessary to distinguish these three orders from each other in Lacanian jargon. What causes anxiety is not an experience of the real, understood as that which is unnameable or unsymbolisable; rather, it is more precisely an encounter with your own image, but divorced from your perception of it as your image, an exact likeness, but unrecognisable all the same. This is an idea represented in the 1950s horror classic Invasion of the Body Snatchers. The heroine begins to realise that whilst her loved ones and those around her are physically the same, something that she cannot put her finger on is different about them. They become uncanny, ‘unhomely’ or unheimlich, to use the German Freud reaches for in the title of this article. Clinical psychiatry even grants this the status of a disorder – the Capgras delusion – in which the subject believes their loved one has been replaced by an identical-looking impostor.

For Lacan, it is in the face of this terrifying ambivalence to the proximity of the image that the symbolic relation asserts itself. “This is where the symbolic relation comes in”, he tells us. “The power of naming objects structures the perception itself. The percipi of man can only be sustained within a zone of nomination. It is through nomination that man makes objects subsist with a certain consistence” (p.169). The symbolic can provide a designation to the image which offers it a meaning, just as the way that the function of the ego ideal – I(A) in Lacan’s algebra – is to give an image a symbolic sanction, investment or guarantee that enables us to assume it as our own image. Why does the symbolic have this special ability that distinguishes it from the imaginary, and provides a supplement to the imaginary that reconciles us to our image? Lacan says that “The word doesn’t answer to the spacial distinctiveness of the object, which is always ready to be dissolved in an identification with the subject, but to its temporal dimension” (p.169). The symbolic gives a space between the images, it differentiates them from one another and prevents them from collapsing into one another, or being super-imposed on one another in the way that we find in Freud’s dream, in which a single person in the dream is a composite of a number of representations of other people – Otto has Freud’s brother’s limp, Irma is reticent to open her mouth like the other female patient, etc. So this is the point at which we see the joint, “the emergence of the dimension of the symbolic in relation to the imaginary” (p.170).

Lacan says that in the dream of Irma’s injection, “it is just when the world of the dreamer is plunged into the greatest imaginary chaos that discourse enters into play” (p.170) but he only says that this is “discourse as such, independently of its meaning” (p.170, my italics). So does Lacan mean by this that the formula for trimethylamine has no associative connections, that there is no significance in the choice of signifier? Surely we still need to ask why it is that this element appears to Freud in the dream and not another? Lacan appears quite adamant, however: “This word means nothing except that it is a word” (p.170). Here he seems to be presenting a treatment of the trimethylamine formula as a pure, reified symbolic element, a symbolic of the symbolic, perhaps. It does not matter for Lacan so much what it is as what it does. At its conclusion the subject of the dream is simply overtaken by the signifier. When the unity of the image fails, the signifier emerges as a pure signifier, such that the dream element presented – the formula – escapes a psychology.

We need to challenge Lacan on this however, but from a point which surpasses his own knowledge of the case. In the first instance, it is not a simple word that Freud sees but a formula, and whilst it functions as a signifier in the sense that it is unhinged from any referent, Lacan nevertheless picks up on the associative connections which must have been familiar to Freud at the time – that trimethylamine is a decomposition product of sperm, for instance. In summing up his own analysis of the dream Lacan does not deviate far from that of Freud’s (though perhaps he was unaware of the alternative interpretation Freud provided for his dream in the 1908 letter to Abraham, cited in our comments on the first of these two chapters). Lacan’s reading is that Freud displaces his desire as a psychoanalyst onto the figures of the three doctors, who offer what becomes their interpretations of the cause of Irma’s suffering from within the dream itself.

But returning to the enigmatic formula for trimethylamine, we can connect it not only to the sexual associations Freud himself notes – the ideas that Fliess “confided… to me on the subject of the chemistry of the sexual processes… [that]… one of the products of sexual metabolism was trimethylamine” (SE IV, p.116) – but also to the recurring sets of three that emerge in the dream (the three doctors, the three women, and the three sons of Jakob Freud). In 1968, the Freud researcher Joseph Sajner made what might be considered a very significant discovery whilst searching through the records in Freud’s birthplace of Frieberg. There he found evidence that Freud’s father, Jakob, had not two wives as previously thought, but three (Sajner, Josef (1968), ‘Sigmund Freud’s Beziehungen zu seinem Geburtsort Freiberg (Pribor), und zu Mähren’, Clio Medica, 3, 167 – 180. (‘Sigmund Freud’s relationship with his birthplace, Freiberg (Privor), and Moravia’)). Whilst the evidence for this is only indicative, it is nonetheless extremely tantilising in respect to this dream. Could this be what the series of threes in the dream refer to? Was Freud himself even aware of his Father’s first wife, or is this yet another example of transgenerational phenomena asserting themselves unconsciously through the displacement of the signfier? And, if so, did he not wish to acknowledge this, or did he just prefer an alternative explanation? And if so, which explanation – the one provided in The Interpretation of Dreams, or the one given to Abraham in the 1908 letter? Lacan’s interventions on this dream provide us with some some very provocative insights, but the dream’s apogee, its enigmatic centerpiece, surely deserves further investigation.

By Owen Hewitson,

Creative Commons Licence
All content on is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported License.