This is the first of two articles looking at the theory of the mirror stage in Lacan’s work. This first part looks at the presentation of the mirror stage as we find it in the Ecrits, specifically in the 1949 paper, ‘The Mirror Stage as Formative of the Function of the I as Revealed in Psychoanalytic Experience’. This is usually seen as the main text on the mirror stage theory, but there are a number of other texts roughly contemporary with it, through which Lacan develops related ideas. By looking at these as well we can learn something about the development of the mirror stage theory and its place in Lacan’s thought at that time.

The second article on the mirror stage will look at how it is presented and developed over the course of Lacan’s Seminar. Rather than leaving the subject in the late forties, Lacan continued to develop the mirror stage theory from Seminar I in the early fifties, right up until Seminar XXII in 1975.

Context of the mirror stage theory in Lacan’s work

The paper in the Ecrits published in 1966, ‘The Mirror Stage as Formative of the I Function as Revealed in Psychoanalytic Experience’, was first delivered in 1949 at the International Psychoanalytical Association congress in Zurich. It was presented less than a year after the paper on ‘Aggresssivity in Psychoanalysis’ was delivered in Brussels in May 1948. Although the original paper on the mirror stage was presented in Marienbad in 1936, the fact that both the 1949 and 1948 papers were delivered less than a year apart suggests that it helps to read both together. Indeed, these two papers contain very similar ideas, and although the aggressivity paper is chronologically the earlier, it is much longer, and in the published Ecrits follows the paper on the mirror stage. Lacan clearly approved of their publication in reverse chronological order, which we can perhaps take as an indication that the aggressivity paper is an expansion or development of ideas put forward in the mirror stage papers of 1936 and 1949. Lacanian psychoanalyst Philippe Julien suggests just this, seeing the mirror stage theory as a compression of two phases: narcissism and aggressivity:

“In the mirror stage, Lacan compressed the two phases into one. At the very moment when the ego is formed by the image of the other, narcissism and aggressivity are correlatives. Narcissism, in which the image of one’s own body is sustained by the image of the other, in fact introduces a tension: the other in his image both attracts and rejects me” (Julien, Jacques Lacan’s Return to Freud, p.34).

If we turn to the index at the back of the Ecrits that Jacques-Alain Miller has carefully compiled, we find references to the mirror stage stretched throughout the papers that comprise the Ecrits. It is listed, perhaps curiously, under the heading ‘The Defiles of the Signifier’, alongside which are related concepts such as narcissism, aggressivity, the superego and the ideal ego.

But the theory of the mirror stage is not an entirely original contribution. Lacan stands on the shoulders of many theorists (some might say appropriating their work) in order to formulate, what is probably still in the Anglo-American world, his best known contribution to psychoanalytic theory.

Lacan’s antecedents in the mirror stage theory

Lacan has many antecedents whose ideas he draws on for the mirror stage theory. Between them, Leader in his excellent Freud’s Footnotes, and Borch-Jacobsen in his equally good Lacan: The Absolute Master give a fairly comprehensive resumé of these (Leader, Freud’s Footnotes, p.197; Borch-Jacobsen, Lacan: The Absolute Master, p.46). Whilst Lacan may or may not have been aware of all of these works, and whilst some of their work may not be directly related to the development of the mirror stage theory, it might be useful at this point to list them in order to provide some context pre-Lacan for the theory’s lineage:

  • Charles Darwin, in an 1877 paper entitled ‘A Biographical Sketch of an Infant’ had already noticed how at eight month old his son was able to associate his reflection in the mirror with his name. Darwin writes, “I may add that when a few days under nine months old he associated his own name with his image in the looking-glass, and when called by name would turn towards the glass even when at some distance from it” (text available here). Darwin also reports the different reactions of humans and apes when confronted with their reflection in a mirror. In contrast to the child, who remains fascinated with its image despite understanding it is only an image (as opposed to another child), “The higher apes which I tried with a small looking-glass behaved differently; they placed their hands behind the glass, and in doing so showed their sense, but far from taking pleasure in looking at themselves they got angry and would look no more” (text available here).
  •  In 1894 James Mark Baldwin published his paper ‘Imitation: A Chapter in the Natural History of Consciousness’ in the journal Mind. In this exhaustive study he argues, long before Freud or Lacan, that the infant is at first unable to properly recognise objects (the care-giver, for example) outside himself as not being part of his own body, treating the other’s body as his own. Nevertheless, he is dependence on the imitation of the other for his own maturation (paper available here, see p.42-43 in particular).
  • The American sociologist Charles Horton Cooley, whose intriguingly-named concept of the ‘looking glass self’ was developed in his 1902 work Human Nature and the Social Order. Here is just a small sample from this book that is likely to tickle the ears of anyone familiar with Lacan’s theory of the mirror stage: “In a very large and interesting class of cases the social reference takes the form of a somewhat definite imagination of how one’s self – that is any idea he appropriates – appears in a particular mind, and the kind of self-feeling one has is determined by the attitude towards this attributed to that other mind. A social self of this sort might be called the reflected or looking-glass self…. As we see our face, figure, and dress in the glass, and are interested in them because they are ours, and pleased or otherwise with them according as they do or do not answer to what we should like them to be; so in imagination we perceive in another’s mind some thought of our appearance, manners, aims, deeds, character, friends, and so on, and are variously affected by it” (Human Nature and the Social Order, 2009 edition, BiblioBazaar, p.183-184).
  • It is claimed by many writers that Lacan borrows a concept developed by Roger Caillois, ‘legendary psychasthenia’ (the theory that an animal will alter its physical appearance in accordance with its environment) to support the theory of mimicry the mirror stage theory entails. In the mirror stage paper in the Ecrits Lacan uses Caillois’ theory to dismiss as “ridiculous” the idea that animals might do this because of some kind of adaptation to their environment, a criticism which even ardent Lacanians must admit is a bit too sweeping (Ecrits, 96). Whilst Caillois gets a mention in the 1949 paper (ibid), it is only in passing, and whilst passing references in Lacan ‘s work are never a sign of how important a theorist is for him, we should probably be wary of overstating the importance of Caillois for the mirror stage theory.
  • According to Roudinesco, it is from Louis Bolk that Lacan gets the idea of “the anatomical incompleteness of the pyramidal system in infants and their imperfect powers of physical co-ordination during the early months of life” (see Roudinesco in The Cambridge Companion to Lacan, p.30). However, Lacan does not acknowledge Bolk in the 1949 article. If we look at this passage in the Ecrits that Roudinesco cites we find that in its entirety it reads: “The objective notions of the anatomical incompleteness of the pyramidal tracts and of certain humoral residues of the maternal organism in the newborn confirm my view that we find in man a veritable specific prematurity of birth” (Ecrits, 96). What is interesting to note is that Lacan attributes this view to himself – it is “my view'”. In the following paragraph he simply says “Let us note in passing that this fact is recognised as such by emryologists, under the heading ‘fetalisation'” (Ecrits, 97), but he does not credit the theory of fetalisation to Bolk by name.
  • William Preyer, a pioneer of developmental psychology, chronicles in his 1882 work, The Mind of the Child, the observations he makes, at each stage in the child’s development, the interest it shows in its reflection in a mirror: “Eleventh week, child does not see himself in mirror…. Seventeenth week, joy in seeing image in mirror….Thirty-fifth week, his image in mirror is grasped at gayly…. Sixty-sixth week, child strikes at his image in mirror. Sixty-seventh week, makes grimaces before mirror” (text available here).
  • Lacan credits the idea of transitivism, to which he makes reference in the paper in the Ecrits, to Charlotte Buhler‘s work on infant behaviour (Ecrits, 98).
  • Lacan’s interest in and reliance on gestalt theory for the mirror stage might stem from the work of Paul Guillaume, a French psychologist working at the same time as Lacan and one of the principal representatives of gestalt theory in France. His work of 1925, L’Imitation chez l’enfant, concentrates on phenomena of imitation in the infant and is a text that Lacan was likely aware of.
  • Another gestalt theorist worthy of mention is Wolfgang Köhler, a key figure in the development of gestalt psychology. Lacan’s strange reference to the Aha-Erlebnis or epiphany (literally: ‘ah-ha moment’) at the start of the mirror stage article in the Ecrits is borrowed from Köhler (Ecrits, 93).

However probably the most important amongst these antecedents is Henri Wallon. Let’s look briefly at Lacan’s debt to Wallon.

Lacan’s debt to Wallon

It might be surprising to learn that the term stade du miroir appears in Henri Wallon’s 1931 article, Comment se développe chez l’enfant la notion de corps propre, (in Journal de Psychologie, November-December 1931, pp.705-48, cited in Roudinesco’s ‘The Mirror Stage, an obliterated archive’, Cambridge Companion to Lacan, CUP, 2003, p.26). But Roudinesco notes that Lacan neglects to cite Wallon as his source – quite astonishing given Wallon is his main intellectual reference here. Indeed, Roudinesco points out that “Wallon’s name is not mentioned either in Lacan’s 1949 paper or in the bibliography of the Encyclopedie francaise” article he wrote in 1938 (ibid, p.27). In fact, there is only one reference to Wallon in the text of the Ecrits, and that comes in the 1948 paper ‘Aggressivity in Psychoanalysis’, where Lacan name-checks Wallon, but not in the context of the mirror stage (Ecrits, 112). Roudinesco writes,

“In 1931, Henri Wallon gave the name épreuve du miroir (mirror test) to an experiment in which a child, put in front of a mirror, gradually comes to distinguish his own body from its reflected image. According to Wallon, this dialectical operation takes place because of the subject’s symbolic comprehension of the imaginary space in which his unity is created…. On 16th June 1936, Lacan revised Wallon’s terminology and changed the épreuve du miroir into the stade du miroir – that is, mixing two concepts, ‘position’ in the Kleinian sense and ‘phase’ in the Freudian sense” (Roudinesco in The Cambridge Companion to Lacan, p.29).

Mikkel Borch-Jacobsen, who says that “Lacan’s description of the mirror stage is far from being truly original” (Borch-Jacobsen, Lacan: The Absolute Master, p.47) also credits Wallon as being its true originator:

“[The mirror stage] had already been given a detailed presentation in 1931-32 by the psychologist Henri Wallon, who relied on previous work by Darwin, Guillaume, Preyer, and Charlotte Buhler. Moreover, Wallon had already drawn much the same conclusions as Lacan.” (Borch-Jacobsen, Lacan: The Absolute Master, p.46).

The non-acknowledgement of Wallon as such an important source for the mirror stage theory is made all the worse by the slightly too self-indulgent bragging Lacan engages in when crediting himself for the idea. Take, for example, this short passage from ‘On My Antecedents’:

“… If I presented the ‘mirror stage’ in 1936, when I had yet to be granted the customary title of analyst, at the first International Congress at which I had my first taste of an association that was to give me plenty of others, I was not lacking in merit for doing so” (Ecrits, 67).

Borch-Jacobsen comments that,

“One cannot help being struck by Lacan’s stubborn silence concerning this important debt [to Wallon]. The article on the family complexes… makes no mention of Wallon. Likewise, in the text tentitled ‘The Mirror Stage’, Lacan says not a word about Wallon… and he attributes to Baldwin an approximate periodisation that in fact belongs to Darwin. Wallon extensively quotes Darwin; Lacan does not. Via the false reference to Baldwin, he skips over Wallon” (Mikkel Borch-Jacobsen, Lacan: The Absolute Master, p.248-249).

What we know about the 1936 article

Speaking about the gestation of the mirror stage theory between 1936 and 1949 in ‘Presentation on Psychical Causality’, Lacan says,

“I duly presented it at the Marienbad Congress in 1936, at least up to the point, coinciding exactly with the fourth stroke of the ten-minute mark, at which I was interrupted by Ernest Jones who was presiding over the congress…. I did not submit my paper for inclusion in the proceedings of the congress; you can find the gist of it in a few lines in my article about the family published in 1938 in the Encyclopédie Francaise, [the ‘Family Complexes’ article, available here] in the volume on ‘The Life of the Mind’ (Ecrits, 184-185).

What is remarkable about Lacan’s confidence in taking credit for the mirror stage theory  is that the encyclopaedia article to which he refers here was penned on the suggestion of Wallon himself!

Whilst there is scant record of the contents of the lost 1936 lecture, Roudinesco discovers that Fransoise Dolto’s notes from the Marianbad conference show that SPP members did not exactly follow what Lacan was getting at and asked him to give a clearer definition of his terms:

“They asked him to define his attitudes more clearly, in particular his view of the relation between weaning and the death impulse, and his conception of the link between the I (je), the body and fantasy. Is the I (je) one’s body? Is fantasy the specular image? Another question asked is: what is the relationship between the I (je) and the ego (moi), and between the I and the personality?” (Roudinesco in The Cambridge Companion to Lacan, p.26-27).

Whilst this does not necessarily show that his audience were unreceptive to the talk – their questions may be an indication of the extent to which they were intrigued and engaged by it – Roudinesco, rather kindly, sees these difficulties in understanding as a consequence of Lacan’s attempt to introduce the philosophical concept of the subject into psychoanalysis.

The use of gestalt theory

To what extent is the theory of the mirror stage a gestaltist theory? Lacan makes multiple references to gestalt theory in the 1949 paper, and it is fair to say that he sees the mirrored image itself as a gestalt:

“For the total form of his body, by which the subject anticipates the maturation of his power in a mirage, is given to him only as a gestalt, that is, in an exteriority in which, to be sure, this form is more constitutive than constituted…. this gestalt, whose power [prégnance] should be considered linked to the species… symbolises the I‘s mental permanence, at the same time as it prefigures its alienating destination. This gestalt is also replete with the correspondences that unite the I with the statue onto which man projects himself, the phantoms that dominate him, and the automaton with which the world of his own making tends to achieve fruition in an ambiguous relation” (Ecrits, 95).

According to gestalt theory the image is unified by the actions of the brain – it makes a recognisable, unitary form or image out of purely geometrical shapes, curves and lines. It is precisely this that Lacan is getting at when he refers to the image as being “more constitutive than constituted” in the passage quoted above.

It would seem that we could use gestalt theory to help us solve one of the seeming contradictions thrown up by the mirror stage theory: what recognises the image as our image if the ego itself (which is supposed to do this recognising) is formed by the mirror stage (that is, formed at the point at which this recognition supposedly takes place)?

However if, according to Lacan’s use of gestalt theory, the image is “more constitutive than constituted”, perhaps there need be no agency that does this recognising. We might imagine a kind of cognitive processing involved, but we do not need an ego or I function for this to happen. Instead, what gestalt theory implies is to all intents and purposes a kind of automatic cognitive recognition of our reflected image as our own.

Mikkel Borch-Jacobsen also comments on this apparent dilemma. He proposes a slightly different solution, which complements the one lent by gestalt theory, that he finds in Freud’s work on identification:

“”Identification”, he [Freud] says, “is the original form of emotional tie with an object”. This is certainly a difficult proposition, but one that nevertheless means that identification is not initially a question of optical representation (not an ‘ideal’, ‘objectivising’, ‘spatial’ identificaftion). Not only is identification “possible before any sexual object-choice has been made”, as Freud also writes, but it must also, by all rights, precede every ob-jection, and every view in general, because it gives birth to the ‘ego’. How, then, could this ego see anything at all – and particularly any ‘model’ or ‘image’ – since it is nothing before the identification?” (Borch-Jacobsen, Lacan: The Absolute Master, p.66).

Rather than appealing explicitly to gestalt theory to get round the problem of their being no ‘agency’ as such to recognise the image as one’s own, Borch-Jacobsen needs to look no further than Freud’s model of identification, clearly envisaged by the latter as prior to ego-processing.

Is there a difference, then, between the infant’s automatic assumption of the gestalt and identification? Are both concepts solving the same problem which the mirror stage theory implies from different angles? Indeed, when we look at how Freud defines identification in 1922 we find it is very similar to the way Lacan sees the gestalt operating in the mirror stage. For Freud, “Identification endeavours to mould a person’s own ego after the fashion of the one that has been taken as a model” (SE XVIII, 106). So perhaps what Borch-Jacobsen is trying to make clear is that Freud’s view of identification does not from the first involve an ego-to-ego relationship, one identifying with the other. Rather his point, with Freud, is to show that the ego is formed via this process of identification: “the character of the ego is a precipitate of abandoned object-cathexes and… it contains the history of those object-choices”, Freud writes (SE XIX, 29).

In any case, through Lacan’s multiple appeals to gestalt theory it is clear that he invokes this theory to do a lot of the work in explaining what goes on in the mirror stage. If, for example, “The total form of his body, by which the subject anticipates the maturation of his power in a mirage, is given to him only as a gestalt” (Ecrits, 95), this most certainly implies that the child’s assumption of the image in the mirror as his own image is only possible through the gestalt. Indeed, he states quite clearly his view that “a gestalt may have formative effects on an organism” (Ecrits, 95) and to illustrate this refers to the biological experiment that purports to show the maturation of a pigeon’s gonads as dependent on seeing the image of another of its species (Ecrits, 95. Incidentally, Lacan makes reference to a similar phenomenon in the male stickleback in Seminar I, 18.01.56. A more recent study on this can be found here). We can say that the entire mirror stage theory depends on an acceptance of the formative, capturing function of the image as gestalt. This is clear from the definition of the mirror stage that Lacan provides in the 1949 paper:

“The function of the mirror stage thus turns out, in my view, to be a particular case of the function of imagos, which is to establish a relationship between an organism and its reality – or, as they say, between the Innenwelt and the Umwelt [the ‘inner world’ and the ‘outer world’] (Ecrits, 96).

Is the mirror stage theory too vision-based?

Suppose we were to try and relieve the mirror stage theory of its dependence on gestalt theory. This would also have the effect of making the mirror stage not so dependent on the fact of vision – of actually seeing a reflection in the mirror. Nobus believes that Lacan’s successive reconceptualisations of the mirror stage in the fifties and sixties have this effect by virtue of Lacan’s privileging the symbolic over the imaginary. In fact in those later works, he argues, Lacan had “stripped the mirror stage of its psychological dimensions” (Nobus, ‘Life and Death in the Glass – A New Look at the Mirror Stage’, in Key Concepts of Lacanian Psychoanalysis, p.119):

“Whereas Lacan’s 1949 paper was still largely devoted to what actually happens when animals and human beings are confronted with their mirror images, in his subsequent analyses of the experience the physical presence of a mirror was subordinated to the intervention of the symbolic order (the Other). By 1960, Lacan even identified the mirror with the Other, a construction which had already been foreshadowed in Seminar I, where he had claimed ‘that the inclination of the plane mirror is governed by the voice of the other” (Nobus, ‘Life and Death in the Glass – A New Look at the Mirror Stage’, in Key Concepts of Lacanian Psychoanalysis, p.120).

The refocusing of the mirror stage away from the imaginary realm and towards the symbolic would seem to get us round the dependence of the theory, as Lacan presents it in the Ecrits paper, on vision. The obvious problem is that of the blind child: if the mirror stage as Lacan conceives it in the late forties is so heavily reliant on the image does this imply that a blind child would be unable to pass through the mirror stage, given that it would be unable  to actually see its own image? And if this was the case, what would be the consequences in terms of the development of a self-image, or perhaps even an ego? This is an objection voiced by philosopher Raymond Tallis:

“If epistemological maturation and the formation of a world picture were dependent upon catching sight of oneself in a mirror, then the theory would predict that congenitally blind indiv­iduals would lack selfhood and be unable to enter language, society or the world at large. There is no evidence whatsoever that this implausible consequence of the theory is borne out in practice” (Raymond Tallis, Not Saussure: A Critique of Post-Saussurean Literary Theory, Macmillan, 1988, p.153; cited in Webster’s ‘The Cult of Lacan’, available here).

Nobus’ proposed reading of Lacan’s later development of the mirror stage addresses this issue directly:

“The symbolic control of the imaginary implies that the assumption of a ‘self-image’ can occur outside the field of vision. As such, Lacan’s purification of the mirror stage entails a reduction of its basis in the physiology of perception. Insofar as the symbolic governs the imaginary, a blind child can still assume a self-image, as long as the symbolic is there to replace and control its eyes, for it will then see itself through the words of the Other” (Nobus, ‘Life and Death in the Glass – A New Look at the Mirror Stage’, in Key Concepts of Lacanian Psychoanalysis, p.120).

Whilst this might appear to be an apt solution to the problem, we have to accept that if Nobus is right Lacan is making quite a substantial shift from the conceptual basis he gives the mirror stage theory in the 1949 paper. We have to also accept that Lacan evidently did not consider any major modifications to that paper to be necessary prior to its inclusion in the published version of the Ecrits in 1966. Lacan is quite clear there that the mirror stage is prior to any socio-symbolic insertion: “These reflections”, he says, “lead me to recognise in the spatial capture manifested by the mirror stage, the effect in man, even prior to this social dialectic, of an organic inadequacy of  his natural reality” (Ecrits, 96, italics mine). We remember as well that Lacan also says in that paper that the “total form of his [the infant’s] body, by which the subject anticipates the maturation of his power in a mirage, is given to him only as a gestalt” (Ecrits, 94-95, italics mine). The fundamental point is the same as we concluded above: Lacan does not seem to accept a gestalt of an image that occurs outside the field of vision (if, indeed, there can be said to be such a thing); it has to be either a mirror or the image of another child – in any case, some kind of image.

However, one thing we can note in contrast to Tallis’ rather superficial reading, is that Lacan gives enough examples in the Ecrits paper for it to be clear to his readers that he does not want us to take the ‘mirror’ too seriously – it does not have to be an actual mirror, but can be instead the mirroring behaviour that the child sees in other children of the same age. The female pigeon’s gonads mature on the sight of another pigeon, regardless of the sex of that other (Ecrits, 95), and the behaviour of migratory locust can be altered in a similar way (Ecrits, 95). Tallis’ argument against Lacan testify, at the least, to a very selective reading of Lacan’s paper as he chooses to disregard the ethological material Lacan uses to support his argument.

Nevertheless, the key point is that for Lacan, at the time of the 1949 paper, the image is purely specular – it is a gestalt-imago and does not need to have any sanction or investment from the Other for it to function in producing the mirror stage. This is a point privileged by many Lacanian writers. For Philippe Julien the relation to the image is essential:

“The image has a morphogenetic power. It is not a pure, passive reflection but the engendering of the infant’s ego. What we call the feeling of one’s own body, or intero-ceptive sensation of the body, comes from this matrix that is the image of the other. The child does not exteriorise itself. It does not project itself in an image. Rather, the reverse occurs. The child is constituted in conformity to and by means of the image” (Julien, Jacques Lacan’s Return to Freud, p.32).

The focus on the specular image in the mirror stage paper of 1949 is in fact a hangover from Wallon’s theory which, as we have seen, Lacan is heavily indebted to. To quote Wallon’s French paper, Les origines du caractere chez l’enfant, as cited by Borch-Jacobsen:

“What other exteroceptive image could he have of himself but the one given him by the body’s eyes, that is, one that is necessarily exterior to him who perceives it?” (Wallon, cited by Borch-Jacobsen, Lacan: The Absolute Master, p.47).

For his part, Borch-Jacobsen believes that Lacan abandons a theory of identification based on an affective bond to the other in favour of one based on the specular relation somewhere between the publication of the Family Complexes article in 1938 and the Mirror Stage article in 1949. For Borch-Jacobsen this is a shame because Lacan misses the opportunity to assert that the first imago the child identifies with not its own in the mirror, or that of the fraternal counterpart, but the maternal imago:

“The maternal imago is not an image or an object or a form or a representation…. As Freud said, “the original form of… tie with an object” is an “emotional” one, and its nature is profoundly one of identification, since it precedes the very distinction between “subject” and “object”, between “ego” and “other”. The important thing here is not merely that Lacan explicitly recognises the existence, previous to the mirror stage, of a pre-specular relation to the image. More important, he plants – something he will never do again – the subsequent dialectic of specular identifications in that initial affective ambivalence, seeing in this ambivalence the “undialecticisable matrix” of that dialectic” (Borch-Jacobsen, Lacan: The Absolute Master, p.68).

The importance of the affective relationship to the (m)other, over and above the relationship to the mirrored image, is one that Lacan’s harshest critics reproach him for. Richard Webster, for instance, in his attack on Lacan that forms the unpublished chapter to his book attacking Freud, cannot understand why the child’s relationship to the mirror image should be of greater importance than the affective relation to its caregivers:

“The possible objections to Lacan’s theory [of the mirror stage] are so numerous that an entire book would be needed to anthologise them. One of the simplest would point to the inherent implausibility of a theory of human development in which a child’s relationship to a mirror is held to be more significant than its relationship to its parents” (Webster, ‘The Cult of Lacan’, available here).

Of course, like Tallis whom he quotes Webster is taking the ‘mirror’ of the mirror stage far too literally, believing it to have to always be an actual mirror (something that, as we have seen, Lacan advises against). Nevertheless, his concern has the same roots as Borch-Jacobsen’s: that Lacan is over-reliant on a vision-based conception of the mirror stage.

Despite this, we can think of the experiment known as the rouge test – which attempts to study infant self-recognition – as providing an experimental basis for some of the observations of infant behaviour in relation to their image in front of the mirror on which Lacan posits the theory of the mirror stage. It is interesting to note that the results claimed by the rouge test align very closely to the age range at which Lacan says the mirror stage takes place (in both cases at 18 months most children are believed to be able to recognise their reflection in the mirror as that of their own image).

Phantom limbs

The strange phenomena known as ‘phantom limb’ syndrome might also be thought about in the context of the mirror stage, specifically Lacan’s description of an essentially “fragmented body”, as he calls it in the Ecrits paper (97). Phantom limb syndrome is the sensation that a limb persists despite its having been amputated. Whilst extremely common – 60-80% of amputees report the experience – it is often felt as extremely painful. One of the interesting things to note when we think about this syndrome using the theory of the mirror stage is that one of the most effective treatments for this pain involves the therapeutic use of mirrors. The so-called mirror box treatment involves a patient inserting both his good and ‘phantom’ limb into a box divided into two, the sections separated by a mirror. When the patient moves his good limb, he sees in the mirror next to it the reflected image of a symmetrical limb in the place where he feels his phantom limb. By this method the patient is able to alleviate the feelings reported by many sufferers – that their phantom limb is clenched or contorted in a painful position. From a Lacanian perspective we might wonder whether this kind of visual-kinaesthetic feedback is not replicating exactly the kind of motor mastery over the “fragmented” body, the corps morcelé, that Lacan describes as the experience of the infant prior to the mirror stage?

Indeed, Lacan makes a reference to phantom limb syndrome in ‘Some Reflections on the Ego’ in 1951:

“The meaning of the phenomenon called’ phantom limb’ is still far from being exhausted. The aspect which seems to me especially worthy of notice is that such experiences are essentially related to the continuation of a pain which can no longer be explained by local irritation; it is as if one caught a glimpse here of the existential relation of a man with his body-image in this relationship with such a narcissistic object as the lack of a limb…[‘phantom limb’ syndrome] leads us to suspect that the cerebral cortex functions like a mirror, and that it is the site where the images are integrated in the libidinal relationship which is hinted at in the theory of narcissism” (Lacan, Some Reflections of the Ego, reproduced in Influential Papers from the 1950s, Karnac, 2003, p.298).

From primary narcissism to aggressivity

Primary narcissism is a pre-Lacanian idea – we find it first in Freud – but despite its elevation to the status of a ‘concept’ by many post-Freudians both Lacan and Freud give it little attention. Nevertheless, it is worth discussing in the context of the mirror stage and the related issue of aggressivity.

Roudinesco defines primary narcissism as follows:

“Primary narcissism is a first state, prior to the constitution of the ego and therefore auto-erotic, through which the infant sees his own person as the object of exclusive love – a state that precedes his ability to turn towards external objects. From this ensues the constitution of the ideal ego. Secondary narcissism results from the transfer to the ego of investments in objects in the external world. Both primary and secondary narcissism seem to be a defence against aggressive drives” (Roudinesco in The Cambridge Companion to Lacan, p.29).

What is odd about this concept, or at least Roudinesco’s definition of it, is the fact that if primary narcissism is auto-erotic it is surely not really narcissism. We can wonder then whether there is really any such thing as primary narcissism.

Laplanche and Pontalis’ description of primary narcissism is similar: “‘Primary narcissism’ denotes an early state in which the child cathects its own self with the whole of its libido” (Laplanche and Pontalis, The Language of Psycho-Analysis, London: Karnac, 2004, p.337). Perhaps we are to understand that the difference between narcissism and auto-erotism is that in the former the whole of the libido is employed towards the self?

What does Freud say about primary narcissism? Remarkably, we find only a few references to it in the index of the Standard Edition. Freud’s inaugural pronouncement on the subject comes in the paper On Narcissism: An Introduction from 1914:

“We say that a human being has originally two sexual objects – himself and the woman who nurses him – and in doing so we are postulating a primary narcissism in everyone, which may in some cases manifest itself in a dominating fashion in his object-choice” (SE XIV, 88).

Later in that paper, Freud separates auto-erotism from narcissism (interestingly, he does not mention primary narcissism when making this separation), but accepts that there is a significant gap between the two and that he himself cannot provide an answer of how to fill it:

“I may point out that we are bound to suppose that a unity comparable to the ego cannot exist in the individual from the start; the ego has to be developed. The auto-erotic instincts, however, are there from the very first; so there must be something added to auto-erotism – a new psychical action – in order to bring about narcissism” (SE XIV, 77).

Lacan’s reference in the 1949 paper to the concept of primary narcissism  – which he puts, perhaps as a sign of disdain, in inverted commas – comes on page 98 of the Ecrits. He equates the move from primary narcissism to secondary narcissism with the moment at which “the specular I turns into the social I” (Ecrits, 98) that is, the moment at which the child stops looking at its own reflection in the mirror and starts seeing other people as its imaginary terms of reference, its ‘semblables’, its imaginary rivals.

For Lacan, a fundamental part of this shift from primary to secondary narcissism is the appearance in the infant of aggressivity and jealousy. This assertion is based on what he detects as happening to the Freudian libido theory after Freud publishes the narcissism paper. Lacan says that his distinction between the two narcissisms in the terms presented above “sheds light on the dynamic opposition between this libido and sexual libido, an opposition they tried to define when they invoked destructive and even death instincts in order to explain the obvious relationship between narcissistic libido and the alienating I function, and the aggressiveness deriving therefrom in all relations with others” (Ecrits, 98).

But it is unclear who the ‘they’ he is referring to here actually is, as this problem is Freud’s own. At the start of the paper on narcissism, Freud is arguing (in stated opposition to Jung) for why he has chosen to stick to his guns about the dualism of the ego and sexual drives (SE XIV, 80-81). Freud has a tough time accounting for why, if there is, as he puts it, a primary libidinal cathexis of the ego, there is then any need to make the distinction between the sexual drives and the ego or self-preservative drives. As we know, in later years this dualism of the libido theory morphs from one between the ego or self-preservative drives and the sexual drives, to one between the life and death drives, Eros and Thanatos.

In the passage from the mirror stage article quoted, Lacan is proposing an answer to the problem Jung presents, but perhaps stealthily we can detect he is siding with Jung instead of Freud. His solution is not to introduce a new distinction between the life and death drives, as Freud eventually does, but to propose the idea that what follows the creation of an Urbild – a kind of prototype ego – from the gestalt or imago of the mirror image is a relation to the images of other people. The jubilation that Lacan describes the infant as exhibiting when its counterpart enables it to achieve mastery over its motor functions is a libidinal cathexis of the imago preceding that of the ego. The “new psychical action” (SE XIV, 77) that Freud describes in the introduction to On Narcissism as the hinge or bracket between auto-erotism and the formation of the ego is for Lacan a libidinal cathexis of the image rather than the ego. The twist in Lacan’s idea that might return us to the problem Freud confronts is that the relationship to the image of the other is so ambivalent that it might even account for what Freud saw as the manifestations of a death drive. Where Freud saw primitive destructiveness, Lacan sees all the phenomena of aggressivity – namely, jealousy, envy, dual imaginary rivalry, transitivism, and so forth.

One last implication of Freud’s assumption of a primary libidinal cathexis of the ego should not be missed: the ego itself is already an object. It is precisely this fact that entails the ambivalence of the phenomenon of transitivism. As Lacan describes it in Seminar III:

“… In the beginning the subject is closer to the form of the other than to the emergence of his own tendency…. The desiring human subject is constructed around a centre which is the other insofar as he gives the subject his unity, and the first encounter with the object is with the object as object of the other’s desire” (Seminar III, p.39).

Does the mirror stage create the ego?

Lacan’s 1949 paper in the Ecrits carries the title ‘The Mirror Stage as formative of the function of the I as revealed in psychoanalytic experience’. Fink has justifiably decided to translate Lacan’s literal ‘the I function’ in the French title of Lacan’s paper as ‘the function of the I‘, the original French being La fonction du Je (Ecrits, 773). But this raises an interesting question: why does Lacan not just refer to the moi, the French term for ego? Why in the title does he refer instead the I function, Je?

It is almost taken for granted in many commentaries that the mirror stage leads to the development of the ego, as if once we have passed through the mirror stage we emerge at the other end with an ego. But when we look in Lacan’s work for a statement to back up this assumption we have a hard time finding it. What we find instead again and again is Lacan being very careful with his pronouncements on the subject. Let’s survey what Lacan says.

To begin with, in the article on Family Complexes from 1938 we find the following:

The unity it [the image] introduces into the tendencies will nevertheless contribute to the formation of the ego. However, before the ego affirms its own identity it confuses itself with this image which forms it, but also subjects it to a primordial alienation” (‘Family Complexes in the Formation of the Individual’, p.22 of the Gallagher copy published by Karnac; p.32 of the online copy available here, italics mine).

We find a similar statement in the paper on Aggressivity in Psychoanalysis from 1948. Notice how careful Lacan is to avoid saying that the mirror stage creates the ego:

“What I have called the “mirror stage” is of interest because it manifests the affective dynamism by which the subject primordially identifies with the visual gestalt of his own body. In comparison with the still very profound lack of co-ordination in his own motor functioning, that gestalt is an ideal unity, a salutary imago. Its value is heightened by all the early distress resulting from the child’s intra-organic and relational discordance during the first six months of life, when he bears the neurological and humoral signs of a physiological prematurity of birth” (Ecrits, 113).

The definition of the mirror stage Lacan provides here is of a process that unifies the body image, as a response to the prematurity of birth, not the need for an ego. Instead, what introduces the ego is “the notion of an aggressiveness linked to the narcissistic relationship and to the structures of systematic misrecognition and objectification that characterise ego formation” (Ecrits, 115-116). Aggressiveness has a significant role to play in ego formation, more so that the mirror stage, which only results in a kind of proto-ego. Indeed, Lacan elaborates later in this paper a view that implies narcissism and aggressivity go together, that you cannot have one without the other:

“The notion of aggressiveness as a tension correlated with narcissistic structure in the subject’s becoming allows us to encompass in a very simply formulated function all sorts of accidents and atypicalities in that becoming” (Ecrits, 116).

One has to wonder however why the relation to the image is marked by this aggressiveness. Why should the ‘libidinal dynamism’, that is the correlate in the human of the ovulating pigeon in the example Lacan chooses from ethology, be marked by such an ambivalent attitude to the gestalt, especially given this gestalt has delivered the unity and motor co-ordination the infant hitherto lacked? Perhaps the infant is aware of and frustrated by the distance that separates its experience of the body from the image that unifies it? However, when defining the mirror stage in the Aggressivity paper Lacan appears to put the accent on the achievement this imago offers to the child:

“What I have called the “mirror stage” [Lacan actually borrows the term from Wallon, as we saw earlier] is of interest because it manifests the affective dynamism by which the subject primordially identifies with the visual gestalt of his own body. In comparison with the still very profound lack of coordination in his own motor functioning, that gestalt is an ideal unity, a salutary imago” (Ecrits, 113).

In the 1951 address to the British Psycho-Analytical Society, ‘Some Reflections on the Ego’ we find Lacan summing up his mirror stage theory that he had by then elaborated twice before, in 1936 and 1949:

“The theory I there advanced, which I submitted long ago to French psychologists for discussion, deals with a phenomenon to which I assign a twofold value. In the first place, it has historical value as it marks a decisive turning-point in the mental development of the child. In the second place, it typifies an essential libidinal relationship with the body-image. [Note the hyphenated term here, indicating they are not separate: your body is not a given that you have an image of]. For these two reasons the phenomenon demonstrates clearly the passing of the individual to a stage where the earliest formation of the ego can be observed” (Lacan, Some Reflections of the Ego, reproduced in Influential Papers from the 1950s, Karnac, 2003, p.300).

Notice how in this passage all he says is that “the earliest formation of the ego” can be observed in the mirror stage. This is not the same as saying that the ego is formed through the mirror stage; what we have instead as a result of passing through the mirror stage is some sort of prototype ego, a kind of sketch or outline of the ego but not yet fully ‘fleshed-out’.

Some remarks Lacan makes in Seminar I a few years later finesse this question of whether the ego is formed through the mirror stage. Note that in what follows, as in the instances above, he is careful to avoid saying that the mirror stage results in the formation of the ego; instead, he says that the mirror stage reveals the relation of the child to his image, and that the image is the Urbild or prototype of the ego:

“As I have often underlined, the mirror-stage is not simply a moment in development. It also has an exemplary function, because it reveals some of the subject’s relations to his image, in so far as it is the Urbild of the ego” (Seminar I, p.74).

What we have in the mirror stage is a unification of the image that produces something akin to a rudimentary ego, rather than the ego as such. So it is perhaps more correct to say that the mirror stage is an identification with an image, and that this image is the precipitate or Urbild of the ego:

“It suffices to understand the mirror stage in this context as an identification, in the full sense analysis gives to the term: namely, the transformation that takes place in the subject when he assumes [assume] an image – an image that is seemingly predestined to have an effect at this phase, as witnessed by the use in analytic theory of antiquity’s term, ‘imago'” (Ecrits, 94, Lacan’s italics).

Returning to the 1949 paper in the Ecrits, Lacan’s definition of the mirror stage there is as follows:

“The function of the mirror stage thus turns out, in my view, to be a particular case of the function of imagos, which is to establish a relationship between an organism and its reality – or, as they say, between the Innenwelt and the Umwelt” (Ecrits, 96).

Again, Lacan bypasses the chance to state that the mirror stage creates the ego, and as such we need to refine the simplistic idea that “the image in the mirror, far from being a simple reflection, forms the ego”, as Mikkel Borch-Jacobsen perhaps too hastily summarises it (Lacan: The Absolute Master, p.63). The mirror stage is a response to the problem of prematurity – not the need for an ego as such. It is only if we were to extend what Lacan calls the ‘drama’ of the mirror stage far enough, beyond the creation of the Urbild, that we can say it results in the creation of the ego. This appears to be what Lacan is getting at in the following passage from the 1949 paper, and which probably gives some of his readers the idea that passing through the mirror stage rewards us with a fully-formed ego:

“The mirror stage is a drama whose internal pressure pushes precipitously from insufficiency to anticipation – and, for the subject caught up in the lure of spatial identification, turns out fantasies that proceed from a fragmented image of the body to what I will call an ‘orthopaedic’ form of its totality – and to the finally donned armour of an alienating identity that will mark his entire mental development with its rigid structure. Thus, the shattering of the Innenwelt to Umwelt circle gives rise to an inexhaustible squaring of the ego’s audits” (Ecrits, 97).

But this nonetheless leaves a rather large question unanswered: how do we go from the gestalt, the imago or ideal-ego, to the Urbild of the ego to the final stage, the ego itself? Lacan seems to account for the creation of the Urbild or prototype of the ego, but then stops. As such,  we might wonder after all whether we can really think of the mirror stage as being about the formation of the ego as such?

Lacan has a go at putting this final step in place when he talks about the mirror stage coming to an end at “the time at which the specular I turns into the social I” (Ecrits, 98). We do not get our ego from the mirror, from our own image, or from the mirror stage as such; we get it socially. The move from the specular I to the social I is a process which “inaugurates, through identification with the imago of one’s semblable and the drama of primordial jealousy… the dialectic that will henceforth link the I to socially elaborated situations” (Ecrits, 98). In other words, the mirror stage terminates when you stop looking at the reflection in the mirror and start looking at other people, your fellow beings. This is the kind of thing that Lacan had spent some time talking about in the Family Complexes article of 1938 (p.13 onwards), but here he simply says,

“It is this moment that decisively tips the whole of human knowledge [savoir] into being mediated by the other’s desire, constitutes its objects in an abstract equivalence due to competition from other people, and turns the I into an apparatus to which every instinctual pressure constitutes a danger” (Ecrits, 98).

But what is perhaps interesting in the last quotation but one is the “drama of primordial jealousy” that needs to come about to accomplish this passage from the identification to the gestalt or imago to the creation of the ego or ‘I’ function. What place does this jealousy or aggressivity have? To try to answer this we need to take a look at the phenomenon of transitivism that interests Lacan so much at the time at which he is working on the theory of the mirror stage.

Transitivism and the formation of the ego

In the aggressivity paper delivered in the year prior to the mirror stage paper, Lacan discusses the creation of the ego not in terms of the mirror stage but in connection with a phenomenon concurrent with the mirror stage – transitivism. This phenomenon has two important features: it shows a certain ambivalence to the other’s image and it provides the crucial jump to the formation of the ego:

“There is a sort of structural crossroads here [at the period at which transitivism manifests itself] to which we must accommodate our thinking if we are to understand the nature of aggressiveness in man and its relation to the formalism of his ego and objects. It is in this erotic relationship, in which the human individual fixates on an image that alienates him from himself, that we find the energy and the form from which the organisation of the passions that he will call his ego originates” (Ecrits, 113).

So the ego is formed not at the point of looking into the mirror, but at the point at which the image of others, almost super-imposed onto our own, come into play. We can describe the genesis of the ego as less a process internal to us as it is external; perhaps as a kind of ‘internalised externality’. This latter description might remind us of Freud’s definition of the ego as “a precipitate of abandoned object-cathexes” (SE XIX, 29). The libidinal dynamics involved in the formation of the ego do not go unnoticed by Lacan as we see from the quotation above in which he refers to an “organisation of the passions”. The ambivalence of the erotic relation to the image of the other however is the price that is paid for this formation.

But why this aggressivity towards the (image of the) other that we see in transitivism? Perhaps because the infant exhibits a certain confusion at the level of the image – he is simply not sure whether it is its own. His mastery over his own body is a result of its identification with the image outside him, but tragically this image is always going to be somehow ‘other’ to him, and a fundamental aggressivity to his mirrored reflection, or the image of his counterpart, is a consequence of this tragedy. The resulting ego therefore is always going to be founded on an image that is experienced as alienating. Lacan had already developed a similar line of thought in his doctoral thesis on paranoia in 1932. As Lacanian psychoanalyst Darian Leader describes it the infant is “… trapped in an image fundamentally alien to me, outside me. Mastery of one’s motor functions and an entry into the human world of space and movement is thus at the price of a fundamental alienation” (Leader, Introducing Lacan, p.22).

What is implied here is a resolutely ‘social’ aspect to the mirror stage. As we noted above, the mirror stage does not just include the child’s relation to his own image in a mirror, but to other children, his semblables or fellow beings, who function in the same way as a mirror: to present to him an image of physical mastery. Lacan stipulates that these other children should be roughly the same age as the infant himself, but this is not to say that they have achieved a greater degree of mastery over their own bodies. Rather, what is important is that they manifest a unified image from the infant’s own perspective.

It is perhaps more helpful therefore to talk about mirroring behaviour rather than focusing on the child’s relation to his image in the mirror. Mirroring behaviour is the mimicry of other children – we can see transitivism as one manifestation of this – and does not just simply refer to the relection in a mirror. The effects of the mirror stage are therefore observable in “phenomena of transitivism in which one finds the infant taking as equivalent his own action and that of the other. He says – François hit me, whereas it was him who hit François” (Seminar I, p.169). The drama involved here is that the child does not simply accede to mastery over his own body, but to an understanding of how the other sees him. This in turn opens up all kinds of questions about what the other thinks or wants of him. We can say it is a vista onto the Other’s desire.

What is the difference between humans and animals when confronted with the mirror?

Transitivism appears to be a uniquely human phenomena. In his 1951 paper on the ego Lacan contrasts the human and animal experience when confronted with their mirror images:

“The child’s attitude is just the reverse of that of animals. The chimpanzee, in particular, is certainly quite capable at the same age of detecting the illusion, for one finds him testing its reality by devious methods which shows an intelligence on the performance level at least equal to, if not better than, that of the child at the same age. But when he has been disappointed several times in trying to get hold of something that is not there, the animal loses all interest in it. It would, of course, be paradoxical to draw the conclusion that the animal is the better adjusted to reality of the two” (Lacan, Some Reflections of the Ego, reproduced in Influential Papers from the 1950s, Karnac, 2003, p.301).

Chimpanzees recognise the image in the mirror as one of their own species and try to reach behind the mirror to grab the other chimp, but Lacan observes that they soon get bored of this (Ecrits, 93). Humans on the other hand are spellbound by the image. For Lacan they are duped by it because, unlike the chimpanzee, the human infant does not see the reflection as just an image to be investigated so as to ascertain its authenticity; rather the infant quickly identifies with the image as their own:

“It suffices to understand the mirror stage in this context as an identification, in the full sense analysis gives to the term: namely, the transformation that takes place in the subject when he assumes [assume] an image” (Ecrits, 94).

The tragic consequence of this, as Lacan points out in the 1949 paper, is that as humans what we identify as our ego – our I function, as Lacan calls it here – will be born from a process that begins with this identification of our image with our being:

“This form situates the agency known as the ego, prior to its social determination, in a fictional direction that will forever remain irreducible for any single individual or, rather, that will only asymptotically approach the subject’s becoming, no matter how successful the dialectical syntheses by which he must resolve, as I, his discordance with his own reality” (Ecrits, 94).

As Lacan points out here, in our identification with the image we will forever remain ignorant of the artificiality of the image, precisely because of the fact that without that image we would not have the ‘sense of self’ that the ego grants us. This is the mistake the chimpanzee does not fall prey to perhaps, we might say, because an I function resembling the ego is not present in that species. What is important to remember is that it is not that we are just ignorant of the gap that separates the unreality of the mirrored image with our ‘true’ selves; rather, this gap is constitutive of our sense of self in the first place – without it, we are completely stuck in a state of primordial dependency. This identification of the image is thus necessary for the creation of the ego, but not sufficient.


The second part of this investigation of the mirror stage will look at the way Lacan develops the theory through his Seminars from the early 1950s to the mid-1970s.

By Owen Hewitson,

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