2

Primary Narcissism: Structure or

State? (1966-1967)
In memory o f J.M.

No concept in the theoretical apparatus of psychoanalysis has under­

gone so many modern reappraisals as the ego. Its complexity, not to
mention the contradictions which seem unavoidable when one
attempts to formulate a theory of it, has been such that a good many
post-Freudian authors, emphasising one particular aspect of the func­
tions it is supposed to fulfil, have given very different versions of it.
Furthermore, many other authors have claimed that it was necessary
to complete the Freudian theory of the ego and to add to it the
concept of the Self (the Anglo-Saxon concept) as the agency repre­
senting narcissistic cathexes. Of the post-Freudian authors, Hartmann
was probably the one who most vigorously defended the need for a
complement to ego metapsychology. Hartmann was followed by
Kohut, who became the most eminent herald of a line of thought to
which he himself made an important contribution. However, in
France, Grunberger had already preceded him in this field, arousing a
certain am ount of surprise, and a good deal of controversy, when he
suggested that narcissism should be considered as an agency, in the
same way as the ego, the id and the super-ego. Many analysts,
following in Hartmann's steps, or sometimes adopting a quite
different perspective, included the Self in their conceptual framework.
Thus, authors as different from one another in their outlook as Spitz,
W innicott, Lebovici, and even the Kleinians, prefer to refer to the Self
rather than the ego. Edith Jacobson introduced the notion of a psycho-
physiological primary Self. Related concepts, such as identity, which one
finds in the writings of Erikson, Lichtenstein and Spiegel, /or person­
ation (Racamier), are also closer to the Self than the ego.
It is true that, after giving up his earlier hypotheses on the
antithesis between ego-libido and object-libido in favour of the
fundamental conflict between Eros and the destructive drives, or
between the life drives and the death drives, Freud did not devote
much attention to the study of narcissism, and particularly to its
future in the theory. 'O n Narcissism: An Introduction' (1914)
nonetheless remains one of Freud's most important texts. Whatever
the reasons for Freud's eventual loss of interest in narcissism - for
instance, the polemic with Jung - it is nevertheless astonishing that
the inventor of the concept did not even consider it useful to explain
why what he had so convincingly described at an earlier juncture
should now be re-evaluated and incorporated within another theor­
etical ensemble. He had not failed to do this, for instance, with the
unconscious, when he replaced the first topography with the second.
This is all the more surprising in that the role of the ego was to be
given more importance with the elaboration of the second topog­
raphy. Freud's readers, psychoanalysts, first and foremost, had more
than one reason therefore to expect a re-evaluation of narcissism,
which in fact never occurred.
It is not surprising that this half-forgotten concept returned in
force to haunt the works of psychoanalysts; for the clinical reality of
narcissism is a fact, even if the way it is interpreted may vary from one
author to another.
Of all the questions relating to narcissism, primary narcissism is
the most muddled and controversial. Furthermore, there is no other
issue which calls the ego's status into question more. How can one
adhere to a line of development which goes from non-differentiation
or primitive fragmentation to a unified image of the ego, whereas the
epistemological revolution, based on the concept unconscious, postu­
lates an unsurmountable split, as the title of one of Freud's last
articles, The Splitting of the Ego in the Process of Defence' (1940b
[1938]) indicates? This is all the more true since, after 1923, the ego
was said to be largely unconscious, especially of its defence mecha­
nisms. Linking narcissism to the accomplishment of Eros alone, an
essential characteristic of which is to carry out increasingly extensive
syntheses - which implies in particular a synthesis of the ego drives -
leads us to wonder what the effect of the destructive drives on narcis­
sistic cathexes and primary narcissism might be. This will be the main
focus of the discussion that is to follow and it will often lead us far
away from this central issue. The point of view that I shall adopt chal­
lenges a certain idea of primary narcissism which regards it as a mere
stage or state of psychical development. I shall endeavour to go
beyond the level of mythical description - as is the case in any recon­
struction based on genetic assumptions - in order to understand a
structure of the psychical apparatus based on a theoretical model. In
Beyond the Pleasure Principle (1920) Freud wrote:
 I do not think a large part is played by what is called 'intuition' in
work of this kind. From what I have seen of intuition, it seems to
me to be the product of a kind of intellectual impartiality.1

Trying to make one synthetic interpretation of all the figures or states

described by Freud under the denom ination of narcissism is not
necessarily a feasible task. The contradictions one comes across in so
doing leave narcissism in a state of open inquiry.

Absolute Primary Narcissism: Narcissism of Dreams or

Narcissism of Sleep?
The condition dominating all the other aspects of narcissism, and
which seems to govern the configuration that has been given to these
forms as a whole, is primary narcissism. W hen Freud used this term
for the last time, he added a qualifier which gives the impression that
he was trying to radicalise the notion. He spoke of absolute, primary
narcissism.2 But let there be no mistake; narcissism is cited here not
in the sense of an experience but in the sense of a concept, or perhaps
as part of a concept. In any case there is nothing which resembles a
positive quality of an experiential order. Sleep might stand such a
comparison, but not dreams. Sleep requires the subject to lay aside his
wrappings, which Freud light-heartedly compares with leaving
behind in the dream's antechamber the accessories (for instance,
spectacles and other prostheses) used to compensate for organ defi­
ciencies. Although Freud compared this with returning to the
beginnings of life, the period spent in the maternal womb does not
take place in an atmosphere of victory or blossoming of any sort. The
conditions fulfilled there, just as in intrauterine existence, are 'repose,
warmth and exclusion of stimulus'.3 Entering the condition of sleep
can only occur if the ego leaves behind its attachments, acquisitions
and possessions, drawing in its cathexes or investments.
Consequently, if primary narcissism is indeed an absolute state, it
is because it represents the lim it of what we can imagine about a
state totally free of excitation. But this notion of lim it is itself open
to confusion. By resorting to it, one immediately introduces a
quality, an affective tonality, the presence of which is explained by
arguing that the experience that is specific to it is encountered on
the way towards primary narcissism, before there is any possibility of
it being accomplished. Unless one is prepared to give up the idea that
the lifting of tensions is the m ain purpose of narcissism, these states,
which are described by using terms denoting a state of bliss, cannot
be amalgamated without destroying the principle of quietude postu­
lated by absolute primary narcissism. Freud does not consider dreams
as a manifestation on the path towards sleep as such but, on the
contrary, as the expression of that which refuses to be silenced, and
which sleep has to integrate if it is not to be interrupted (a breach of
narcissism: this is what Freud has to say about the unconscious
thoughts which are the origin of dreams, showing the sleeping ego
that its capacity to impose its will is limited). In the same way, for
Freud, narcissistic elation or expansion, connoting narcissistic regres­
sion are, so to speak, extraneous, and a sign of the subject's
reluctance to let himself slip away into silence. For when the
analysand has the feeling that the analyst is no longer present in the
session, one has to explain why he does not remain silent and why
he does not stop talking. Is it not precisely when there is a danger
that his own discourse results in his no longer being seen or heard
by the analyst that he swallows him like an egg, incorporating him,
so that the discourse is not interrupted but continues after the threat
of an absence - which could very well be his own - has been warded
off? Even when this feeling is experienced during a pause, the aware­
ness and expression of it are the signs that such a moment has been
ruptured.
Freud seemed to want to distinguish clearly between the narcissism
of dreams and the narcissism of sleep. If we read the text4 carefully,
we realise that two very similar formulations should be understood
more as reflecting two different modalities than as the orientations of
a single process for which Freud does not supply a theory. In fact the
narcissism of dreams is the same as the narcissism of the dreamer. The
latter is unmistakably the principal dream character who, as it were,
always glorifies the dreamer - a point of view which is by no means
undermined by dreams of self-punishment or nightmares. Whereas
the narcissism of sleep surpasses, so to speak, the dreamer's wishes; it
supports the movement of the dream and withdraws from it into a
region that is out of reach, in which the dreamer himself vanishes.
When, in a dream, there is an unrecognisable character or an
unknown face or, one whose features are not even recognisable, it
represents the dreamer or his mother. I shall come back to this later.
This blank face which is only present in outline, or which is only
marked by its location in space, is perhaps the leitmotif that will
guide us in constructing the theory left in abeyance by Freud.

The Principle of Constancy or the Principle of Inertia?

The separation I have just referred to between the narcissism aimed at
removing tension, of which sleep is not an illustration (how can one
speak of sleep without dreams?) but an abstract model, and the
narcissism of dreams, or of the dreamer who experiences states of bliss
or states in which his bodily limits are overwhelmed during waking
life, was never completely explained by Freud. It is customary to link
Freud's loss of interest in narcissism to the theoretical reformulation
which led to the final drive theory and, above all, to the introduction
of the death drive; a point of view that is no doubt justified. But redis­
tributing the connotations of the drive according to a different
configuration, and in terms of the drive orientation, was not the only
innovation as far as narcissism was concerned.
The aspiration for a state of zero excitation - the insusceptibility to
excitation of uncathected systems already alluded to in the 'Project' -
is a constant element in Freud's thinking. This is how he described the
organism's tendency, which thereby ensures its mastery over stimuli,
in his first formulations, inspired by psychobiology. Focusing there­
after on the vicissitudes of desire, he likened pleasure to the cessation
of sexual tension, to the removal of the pressure of desire by its satis­
faction causing a pleasant state of relaxation. But experience probably
taught him that the longing for a lowering of tension was, so to
speak, independent of it. It was no longer to be seen merely as a m ani­
festation of the mastery of the psychical apparatus but perhaps, or
even probably, as a state, even if there was no way of telling whether
it was a consequence of its functioning, one of its aims, or if it itself
had to obey it as though it were an exigency. In An Outline of Psycho­
Analysis (1940a [1938]) he says:

The consideration that the pleasure principle demands a reduc­

tion, at bottom the extinction perhaps,5 of the tensions of instinctual
needs (that is, Nirvana) leads to the still unassessed relations
between the pleasure principle and the two primal forces, Eros and
the death instinct.6

The modern versions of primary narcissism which have been put

forward provide us with many partial images of these relations,
particularly where the links between the state of Nirvana and Eros are
concerned; but they tell us nothing about the relationship between
Nirvana and the death drive. Either it is completely ignored, or the
states described - which can only be interpreted as resulting from the
fusion of Nirvana and Eros - are only thought of as being stages on
the way to complete Nirvana when the death drive would take over
from Eros but would not be its antagonist.
As was often the case, Freud forgot that he had already begun to
examine and even to resolve these questions which had escaped eval­
uation. The idea of a state free of excitation had haunted Freud from
the time of his neurological formulations on neuronic inertia to his
search for backing in Fechner's psychology (.Beyond the Pleasure
Principle, 1920). He paid for his allegiance to the banner of his illus­
trious elder by renouncing an original point of view which he was
only to rediscover many years later; for in the 'Project' there was
already tendency to make an absolute of that which, later, would be
known as absolute primary narcissism. The principle of inertia - and
not the principle of constancy - was the first to be stated by Freud.
The 'original trend' of the nervous system to inertia is to 'bring the
level of tension to zero'. For Freud, this original trend was the primary
function, whose aim was to keep the system in a state free of excita­
tion. Constancy obeys secondary processes which are governed by the
necessity of maintaining a m in im um level of cathexis.7 It is impor­
tant to realise that Freud only used the word principle in the case of
the principle of inertia; maintaining excitation at a constant level was
not raised to the same rank. The principle of constancy was, however,
frequently mentioned by Freud in the Letters to Fliess (Manuscript D,
May 1894, letters dated 29 November 1895 and 8 December 1895;
Manuscript K, the letter of 1 January 1895) at the time when he was
elaborating the 'Project'. Moreover, the first mention of it comes even
earlier, in the Studies on Hysteria (1895c [1893-5]).8 But although the
principle of constancy was attributed to Fechner, the principle of
inertia was purely Freudian. W hich means that, in the cursive allu­
sions or correspondence, the only reference is to maintaining
excitation at a constant level, as low as possible; whereas, in the
'Project', where Freud was trying to systematise his ideas, his wish to
make a theory of it led him to pursue his assumptions to their logical
limit and to give preference to the principle whose purpose is to
attain the level zero, and not merely the 'lowest level possible'. Here
we can see the origin of a duality of principles whose order of prece­
dence was to fluctuate in Freud's writings thereafter. But if we are to
understand their permutations or their later fusion properly, it is first
necessary to notice in what ways they are different. For Freud, the
principle of inertia was fundamental,9 belonging to the order of
primary aims (characteristic of the primary neuronal system). It owes
its existence to the neuronal system's attribute of totally suppressing
excitation by means of flight which, on the other hand, is totally
impossible in the case of internal stimuli. In the light of this impossi­
bility the only solution is to keep tension at its lowest possible level.
Freud describes this function as secondary.10
This is a good opportunity for noting the liberties that Freud took
with the genetic point of view, while he endeavoured to make a
division between primary and secondary functions. It is perfectly clear
that the possibilities of suppressing excitation by means of flight in a
young organism are strictly limited; and, that the most intense and
frequent stimuli unquestionably come from the major needs which,
logically, should be in a position of primarity. But Freud did not stop
there. W hat was important for him was to focus on the efficiency or
success of the operation of taking flight from disturbing stimuli and
to establish the configuration unexcitability - tension - flight - abol­
ishing tension - unexcitability as a model; that is, as a fundamental
aspiration, within a psychological perspective, even if it was unrealis-
able in practice. This is why keeping tension as low as possible and
guarding against any increase of it was, at this stage in his thinking, a
second choice, as the English say; that is, a secondary function. It is
this difference that Freud apparently gave up in Beyond the Pleasure
Principle when he merged the two principles into one. Taking refuge
behind the authority of Fechner, just when he was at his most auda­
cious, was rather characteristic of him. By making Fechner's principle
of constancy the organising principle of which lowering tension to
the level zero was only one special case, he increased his under­
standing of primary-secondary relations. Primarity was granted to the
principle of constancy from which he inferred the pleasure prin­
ciple,11 and secondarity to the reality principle.
It is understandable, therefore, that this change has been a source
of confusion. There is a temptation to regard the lifting of tension,
accompanied by a return to calm produced by the satisfaction of a
drive which had hitherto generated unpleasure, as equivalent to the
state of absolute elimination of tension in the initial model, where
non-excitation, a state in which the system is out of action, was the
absolute criterion. At first sight, the difference between inertia and
calm is clear; just as it is between night and obscurity. This transposi­
tion is all the more significant in that Freud recast the relations
between the principle of constancy and the pleasure principle in
terms of the relations between an abstract theoretical model and its
concrete illustration.12 Apparently, however, he was overlooking the
fact that he had already applied the notion of the relativity of
pleasure to that of m aintaining excitation at a constant level when
faced with the total extinction of stimuli towards which the inertia
principle tends. Let us recall, however, that this apparent retreat coin­
cided with the introduction of the death drive.13 And yet there is an
indication of uncertainty surrounding this relegation of the inertia
principle to a position of secondary importance. In the penultimate
chapter of Beyond the Pleasure Principle - and one might have thought
that since Freud no longer felt the need for backing, and had now
expressed his opinion on the death drive, he would now be able to
return to it - he wrote: '

The dom inating tendency of mental life, and perhaps of nervous

life in general, is the effort to reduce, to keep constant or to remove
 internal tension due to stimuli (the Nirvana principle, to borrow a
term from Barbara Low [1920, 73]) - a tendency which finds
expression in the pleasure principle; and our recognition of that
fact is one of our strongest reasons for believing in the existence of
death instincts.14

A Theory of States and a Theory of Structures

So the order of things had now been restored: the ultimate tendency
of the Nirvana principle was to eliminate excitation and the pleasure
principle was simply derived from it. The first theory of the 'Project'
was thus legitimised and was to be reinforced even more indisputably
a few years later in the first paragraphs of 'The Economic Problem of
Masochism' (1924), where Freud went a considerable way towards
clarifying his ideas. He proclaimed the divorce between the Nirvana
principle and the pleasure principle and warned against any further
confusion between them .15 Their respective characteristics were
described as follows:

The Nirvana principle expresses the trend of the death instinct; the
pleasure principle represents the demands of the libido; and the
modification of the latter principle, the reality principle, represents
the influence of the external world. (SE, XIX, p. 160)

The task of reducing tension no longer fell to the pleasure principle -

the notion of constancy disappeared from this theoretical reorganisa­
tion - and remained the exclusive task of the Nirvana principle,
whereas the function of the pleasure principle was very closely
connected with the 'qualitative characteristics of stimuli'. We are
therefore justified in postulating that all states comprising an affec­
tive characteristic, or pleasure and its derivative forms (elation,
expansion, or any other manifestation of the same order), are foreign
to absolute primary narcissism.
Let me point out right away that proposing this trinity does not
infringe upon Freud's epistemological rule, which keeps all the
antitheses within a context of duality. The reality principle is merely
a modified pleasure principle. In fact, there is no other solution than
to consider that the problem has two aspects to it: first, there is a
primary antithesis between Nirvana and the pleasure principle; and
second, between the pleasure principle and the reality principle, the
latter of the two being more common. For although Freud uses the
same terms to describe the transformation of the Nirvana principle
into the pleasure principle and the relation between the pleasure
principle and the reality principle,16 he does not connect the two
operations. We have no other choice but to assume that Freud could
not connect these two modifications because they belong to funda­
mentally different registers or spheres and do not tolerate being
mixed or amalgamated.
In his first text on narcissism in 1914 Freud gave an indication of
how the two aspects of this problem might be explained:

The individual does actually carry on a twofold existence: one to

serve his own purposes and the other as a link in a chain, which
he serves against his will, or at least involuntarily. The individual
himself regards sexuality as one of his own ends, whereas from
another point of view he is an appendage to his germ-plasm, at
whose disposal he puts his energies in return for a bonus of
pleasure. He is the mortal vehicle of a (possibly) immortal
substance - like the inheritor of an entailed property, who is only
the temporary holder of an estate which survives h im .17

Is it fair to suppose that the role Freud attributes to the legacy of the
species can be sacrificed with impunity, on the basis that this elabo­
ration stems from a metabiological romanticism which a healthy
scientific reflex requires us to shun? It may be felt that Freud's state­
ment is outdated and that his assumptions with regard to this part of
the theory are clumsy and debatable. But there is m uch less justifica­
tion for refusing to examine the fundamental problem, which is
neither the role of the species, nor the heritage of acquired character­
istics. It is a two-sided problem: the current general trend in
psychoanalysis is resolutely ontogenetic; if anything, its error is that
it is not enough so. Freud was more ontogenetic in that he did not
allow himself to be paralysed by a linear conception of time. But he
was constantly having to navigate between a theory of states, which
continued to retain the descriptive aspect of clinical forms, and a
theory o f structures which created models, if not as pure conventions,
at least as the development of these states to the point at which their
function and their meaning was revealed in the most abstract terms.
The antithesis between the pleasure principle and the Nirvana
principle is, I think, a good example. If Freud took a wrong turning
with the principle of constancy, surely it was because this notion was
halfway between a theory of states - in this case the state of pleasure
- and a theory of structures, the constant level of excitation holding
the middle ground between the extinction of excitation and the
heightening of internal tension. If one gives the matter careful
thought, it will be noticed that the theory of states which- gave rise
to the hybrid monster of psychoanalytic phenomenology is, in the
last resort, a theory of the subject's manifestations, but not a theory
of the subject. And while the conflict has not gone away, it has, as
one says today, been 'personalised'. In the end the subject is always
a volitional being who wants something or cannot do something,
who allows himself, or does not allow himself something, who longs
for, or is frightened by something. This being the case, it is difficult
to understand why an analysis conducted along these lines would
not remove the obstacles when the invisible hindrances have been
brought to light and identified. It is easy to see that, even when the
analyst's good will is accompanied by lucidity and vigilance, it has
little mutative effect. Although there is some consistency in the
concept of the subject's Entzweiung, it is not to be understood as the
antithesis and reconciliation of two wills, but as a conflict between
two systems, inspired by two antithetical and obstinate ways of
thinking, which can be noticed even in the effects produced by the
way the discourse is constituted, or in the utterance itself (in the
suture and cutting of the elements of a section of the enunciation
and in what follows them), in which the signs of the work of this
division are reflected. The theory of structures seeks to establish the
conditions under which discourse is possible; the organisation of the
latter being such that the subject can only be apprehended in terms
of his life trajectory whose functioning is the mark of its reality. The
subject is therefore not in a position of m odality18 in which the
index at the origin of the utterance designates the operation of
thinking as being distinct from the representation it is aiming at;
neither is he at the end of the sentence at which point, once the
utterance is over, one may be able to throw light on everything that
has preceded it by looking backwards. He is the vehicle by means of
which there is an utterance.
It should not be thought that we repudiate wholesale everything
in psychoanalysis pertaining to the theory of states. It represents a
first level of psychoanalytic epistemology; and, in their silent
com m unication with their analysands, or with other analysts,
psychoanalysts cannot avoid expressing themselves as follows: in fact
he wants this or that; at bottom he is saying this or that; he is again
reliving, and so on. But this inevitable stage cannot be considered as
the degree of organisation which accounts for the analytic process.
W hat guarantees the functioning of this process is the analyst's
silence which, in the last resort, has no other justification. The great
merit of the impetus Lacan gave to this kind of research was to show
how the results of our psychoanalytic investigations, notw ith­
standing the fact that their structural intent is respected, refer to
organisations that are already structured.
The Psychical Apparatus and the Drives
Let us turn our attention for a m om ent to the psychical apparatus.
There can be no doubt that this construction is linked in Freud's
thinking with a theoretical model situated off the path between the
brain and conscious thought, creating an essential discontinuity
between them. But Freud gives this model a space19 and a time (since
he speaks of the relations of seniority between the agencies). He does
not take the trouble to say which space and which time are involved
but, since it is space and time that are at stake, the psychical appa­
ratus has been reintegrated w ithin a pre-Freudian universe of
representation by treating it as one of the multiple organisms defined
by our conscious space and time. This indicates the beginnings of an
attempt to look for a structure within an ontogenetic framework. The
psychical apparatus becomes a kind of self-codification in which the
subject constructs himself.
As one might well suspect, this shift tends to shrink and, ulti­
mately, to superimpose the dimensions of the psychical apparatus on
the ego, and flouts Freud's remark that individual experience, as it is
registered by the ego, only determines 'accidental and contemporary
events'.20 It is logical to assume that the effect of structuring has to
come from elsewhere if the ego is thus involved in the instantaneity
of the present.21
In order to preserve the metaphorical value of this apparatus, the
question needs to be turned round the other way; rather than trying
to find out what kind of apparatus psychic life may relate to, the
question we should ask ourselves is: what is an apparatus from the
point of view of the psyche which is assumed to be its function? Can
the principles which we have examined at length be considered as
original first causes or as regulators of functioning? In the latter case,
all 'governing' power would be taken away from them and there
would be no further reason to use the word principle. Regarding them
as first^causes or, at least, as that which conceptualises such causes, is
to see in them the ultimate foundation of any kind of psychical
organisation. However, a careful examination of the last systematic -
dogmatic, even, Freud says - theoretical work, that is, the Outline,
shows that it accords equal status to the theory of the drives and the
principles of mental functioning, that is, the same conceptual worth.
Even the values of the first topography (conscious, preconscious,
unconscious) are confined to psychical qualities whose status can only
be explained by the structure of the psychical apparatus, in the same
way that the development of the sexual function - albeit the origin of
everything we know about Eros - is subordinated to the theory of the
drives.22 Freud shows in the seventh chapter of Beyond the Pleasure
Principle that he was aware of the difficulties involved in these rela­
tions, when he turned to the question - m uch too briefly,
unfortunately - of the differences between function and tendency.
Notably he says there that the pleasure principle is a tendency oper­
ating in the service of a function

... whose business it is to free the mental apparatus entirely from

excitation or to keep the amount of excitation in it constant or to
keep it as low as possible. We cannot yet decide with certainty in
favour of any of these ways of putting it; but it is clear that the
function thus described would be concerned with the most
universal endeavour of all living substance - namely to return to
the quiescence of the inorganic world.23

In addition to this assertion, he announced an entirely fresh series of

investigations which, for lack of time, he was never able to carry out
fully, in which the relations between principle and drive can be
inferred; and, he maintained that there was a contradiction, if not
between the particular and the universal, at least between the
personal and impersonal. At this point, we can say that the principles
are at the crossroads of the relations between the psychical apparatus
and the theory of the drives.24
At the core of a drive there is a principle which enables us to gain
an understanding of the equipment that governs its functioning in a
way that on no account should be understood as the impact of a force
that is external to it, but rather as a process within the constituents of
the drive. As a result, the latter is deployed, distributed, and ampli­
fied, the structure of the apparatus allowing its elements to be linked
up and primitively condensed, quasi-tautologically, within a system
of relations. This could not, for example, be an act of repression, an
operation that is itself subject to the pleasure-unpleasure principle.
The 'universal' function of the drive is particularised in an individual,
but on the condition that the individual acquiesces to it, which can
only happen if there is a 'tendency'. Nonetheless, this word should
not lead us astray; it is not a synonym for an 'attempt' (tentative) but
of a 'tension towards' (tension vers). And if the aim here is the inac­
cessible absolute, the absolute is displaced on to the effort of tension.
Now, the relations between the psychical apparatus and the drives
have rarely been examined in detail. It is quite usual to talk about the
situation of the drives in the id25 (the id as the seat or reservoir of the
instinctual drives). W hat is less frequently discussed is the articula­
tion between the theory of the drives and the psychical apparatus.
It is generally acknowledged that the theory of the psychical appa­
ratus represents the last stage of psychoanalytic theorisation and, in a
certain sense, this is true. It is true at this first level of the theory, that
which Freud calls the part of the individual whose type of organisa­
tion is reflected in this construction. But for Freud, drive theory
brings into play what is already-structured, a notion I referred to earlier,
the structure of which organises the possible conditions of func­
tioning in which a subject reveals himself. While, like Freud, one is
reluctant to see in this a manifestation of the species, one needs at
least to recognise this dimension of depuis-toujours-deja-la; this
montage which is never immediately accessible but to which all
organisation refers. It is not possible to say whether the drives are
always for the psychical apparatus or if the psychical apparatus is for
the drives. 'Already structured' does not mean that the structural
mode is identical in all cases. Indeed, the interest of the system resides
in its heterogeneity.
The psychical apparatus represents the construction that drive
activity would be capable of if its mode of functioning were other than
agonistic and antagonistic. But conversely, one would have no idea
what the fundamental nature of such agonism and antagonism might
be if there was no psychical apparatus to represent it for us. A better
idea of these relations can perhaps be obtained by recalling Freud's
view that the drives act essentially within dynamic and economic
dimensions. They do not have any localisation, even within the frame­
work of a conventional abstract model.26 Whereas, the psychical
apparatus has the characteristic of being extended in space, that is, of
converting modes of transformation originating from the dynamic
economic system - and we will see in due course which ones - into an
interdependent system of surfaces and spaces (or locations) capable of
receiving qualitative and quantitative modes of varied inscriptions,
filtering and retaining them in forms that are appropriate for them.
Between the undifferentiated drive, which some authors present
variously as a current of force, a tide or as a tachist painting, and
Freud's elegant and precise montage, which nonetheless seems to
many td'be too restrictive today, the final theory of the drives may
provide a point of mediation. Here the functions of Eros and the
destructive drives converge with the major categories of the tendency
towards union and the tendency towards division, of fusion and
defusion. Using more modern vocabulary one might speak of conjunc­
tion and disjunction, of suture and cutting.27 But Freud was not
content with bringing together, in the manner of classical antitheses,
two terms of equal value in order, through the repetition and setting
up of new relations, to end up with an organising, ordering power.
Eros and the destructive drive do not form a pair of equal terms.
An indication of this is that Freud always refrained from nam ing the
death drive in any other way than this (except by the related formula
of the destructive drives). For if the compulsion to repeat is the mode
of activity characteristic of any drive - which, as F. Pasche puts it
aptly, would be like the instinct of the instinct - it can be said that
something of the essence of the death drive has passed over to Eros;
or, that Eros has tapped into it for its own benefit, which disqualifies
the death drive and means that one can no longer speak of it merely
as the invisible and silent term of a couple whose difference can only
be determined by a shadow cast over the sparkling light of Eros.
Recasting the antithesis would enable Freud to say - a first duplication
- that the two drives could operate against each other or combine
with each other. W hile drive defusion - in the case of discordant
work, examples of which can be found in pathological states such as
melancholia or paranoia - may give an idea of it in love-hate rela­
tions, the collaboration of the two drives is of course puzzling if one
does not accept the idea of hate being neutralised by love, and, if one
is not satisfied with employing arguments of a quantitative nature to
suppress the issue.
The internalisation of this contradiction led to the rediscovery of a
duality in Eros which was to be a second duplication; namely the
division of Eros between ego-love and object-love and the division
between self-preservation and the preservation of the species.
Although, at first sight, it is tempting to put ego-love and self-preser­
vation together on one side, and object-love and the preservation of
the species on the other, one soon realises that, by so doing, one is
doing away with the antithesis between personal erotism, of which
object-love is a part, and impersonal erotism whose heuristic value is
of such importance. This perhaps offers a way of exploring further the
richness of the Lacanian theory of the subject as a structure. W hen
Lacan writes: 'Only the signifier can tolerate coexistence; only the
disorder consisting (in synchrony) of elements deployed (in
diachrony) in the most indestructible order. This associative rigour of
which it is capable, in the second dimension, even merges with the
commutability which it shows is interchangeable in the first',28 one
may wonder whether such commutability does not have a bearing on
the two double registers that I have just mentioned. We should there­
fore not overlook the troubling expression used by Freud in the
'Project' - so skilfully interpreted by Jacques Derrida29 - according to
which, the processes uncovered in the study of neuroses, which only
differ from the normal in their intensity, are quantities in motion. The
question of primary narcissism appears to have been overshadowed
by the problems involved in the theory of the drives. We shall see that
this is not in fact the case when we come back to the question from
another angle, by examining the following problem: is narcissism
merely a consequence of the orientation of cathexes?
The Origin and Vicissitudes of Primary Cathexes
Our thinking on the first forms of exchange is dominated by the
paradigm of the amoeba. However, although Freud only made use of
this analogy to compare the movements of cathexes being sent out
and withdrawn, the peripheral phenomena, which were the main
reason for resorting to this image, have themselves become peripheral
in our thinking, thus making room for the idea that the general form
of amoebas should be taken as a model for the first forms of psychical
organisation, and particularly of the ego.
Nevertheless, while this analogy may, at a pinch, be congruent
with the ego Freud described before the last topography, contradic­
tions inevitably arise from continuing to make use of the comparison
after the final conception of the ego.
The vesicle, a small self-enclosed sphere, suggests the existence of
a mode of functioning which does not fit in easily with Freud's ambi­
guities or imprecise observations on the earliest relations between the
ego and the id ... Another paradigm, that of the reservoir, may be
regarded as consubstantial with it, and Freud even condenses the two
of them in some texts. It required all of Strachey's penetrating vigi­
lance to analyse this image.30 Yet is it not enough to distinguish
between the storage function and the source of supply; or to note that
the contradictory versions Freud gives of the origin of the first
cathexes - alternately in the ego (before its differentiation in the last
topography), then in the id and finally, paradoxically, in the ego
again - are resolved in the conception of the undifferentiated ego and
id? This provides a useful clarification but needs examining more
closely. The primitive, undifferentiated ego-id 'originally' serves two
functions at once: as a source of supply and a storage tank. As a source
of supply, it sends its cathexes out in two directions: towards objects
(centrifugal orientation) and towards the future ego (centripetal
orientation), thus contributing to the second function. As it develops,
the undifferentiated ego, is basically a storage tank. And while the ego
unquestionably plays a role as a source of energy supply for object-
cathexes, it also watches over the reserves of narcissistic libido. In
other words, ego-id differentiation introduces a functional separa­
tion. But the ego recuperates part of the function it gave up in the id's
favour in order to secure in priority a store of narcissistic cathexis. It
therefore acts upon the object-cathexes originating in the id in such
a way that they do not compromise too much the narcissistic cathexis
under its control. However, it is the detail of this differentiation that
needs clarifying. W hat cannot be doubted is that Freud, as I have
already shown, linked the state of absolute primary narcissism with
the lifting of tension and an ego-relationship. Although he fully
insisted on the possibility of a conversion in the exchanges between
narcissistic-libido and object-libido, he equally stressed the durability
of a narcissistic organisation which never disappears. The libido
cathects the ego and in this way provides itself with a love-object, a
process that can be observed throughout life. But Freud never associ­
ates the state of absolute primary narcissism with the id. Freud quite
frequently employs the term ego to designate either the ego stricto
sensu or the primitive undifferentiated ego-id. But the opposite is not
true. Freud never associates the id with functions or processes
belonging exclusively to the ego.
Now, to define narcissism in terms of the qualities of elation or
expansion or any other affect of the same order, even when one is
referring to the undifferentiated ego-id, is to refer to properties which
are only meaningful within the system of the id.31 This means that,
in order to define their relation to narcissism, they are being intro­
duced in a context which is not appropriate to ego-cathexes. It is not
enough to liken them to omnipotence; for elation and expansion are
consequences of omnipotence and not the process by which om nipo­
tence sets in. This consists in suppressing the object's power of
resistance or the power of reality by denying dependence on them
and not by merging with them. Such fusion, if it occurred, would
only be possible once the ego had assured itself that it had the upper
hand over the object's powers which it appropriates to this end.
The Nirvana principle - which, as I have pointed out, has its place
in a theory of structures, but is in fact absent from a theory of states
in which only signs of a lessening of tensions are perceptible - has
undergone a modification in living organisms. It is true that it is often
necessary to turn to the pleasure principle (which is fundamentally
different, however, as it is tied up with qualities of pleasure) in order
to find a trace of it. In the Freudian system, where modifications
never completely erase the state they modify, we perhaps need to see
whether a displacement of values makes it possible to rediscover what
appears to have disappeared. And since we feel condemned by the
death drive to seeing nothing but the invisible, and to questioning
only that which is silent, we have to explore that aspect of Eros which
resembles it.
It is clear, I think, that the love the ego has for itself (ensuring its
independence from the external world and saving it from expending
cathexes on the object), the flowing back of object-libido towards the
ego, and the absence of conflict - provided that the quality of this
ego-love compensates for the libidinal quality intended for the object
and that it protects against the disappointments that the latter can
inflict - succeeds in creating an enclosed system, and comes very close
to the condition which the ego strives for in dreamless sleep. A border
situation is created here in which the 'clamour of life7of Eros, and the
struggle against Eros, manage against all odds to introduce that which
is at the origin of death into the very heart of love, settling their debts
at the object's expense. But how does this come about? We shall have
to take a roundabout route before being in a position to answer this
question.

The Aim-Inhibited Drive

Even though the current state of psychoanalytic research only sees
the drives of self-preservation at work in the ego (libidinalised since
the introduction of narcissism), the repeated observation32 that other
drives may be operating in it has gone unheeded. As Freud tied the
relations of the ego to reality to safeguard the pleasure principle,
without saying more about the forms of this non-libidinal drive
activity, the conclusion has been drawn that this silence must have
covered one of Freud's mysterious assertions, the secret of which he
took with him to the grave.
Between those drives that were not libidinal, 'which seem to be
operating in the ego', and the elusive work of the death drive, Freud
was to introduce an intermediary series which he regarded as
belonging to the constituents of Eros. Alongside the uninhibited
sexual drive proper and the drive of self-preservation, Freud set the
instinctual impulses of an aim-inhibited or sublimated nature derived from
the sexual instinct.33 He would undoubtedly have resisted any inter­
pretation which gave this contingent autonomy under the umbrella
of the 'social instincts', which were very much in vogue in the
psychology of the period. But after further examination he distin­
guished the aim-inhibited drives. The best description Freud gives of
them can be found in the thirty-second lecture in which he likens
them to sublimation.
s
Besides this, we have grounds for distinguishing instincts which are
'inhibited in their aim' - instinctual impulses from sources well
known to us with an unambiguous aim, but which come to a stop
on their way to satisfaction, so that a durable object-cathexis comes
about and a permanent trend [of feeling]. Such, for instance, is the
relation of tenderness, which undoubtedly originates from the
sources of sexual need and invariably renounces its satisfaction.34

In the end it was the idea of restriction, of holding back, of the non­
development of the cathexis which won through, justifying a special
denomination. By proposing to give a special place to this type of
drive, Freud was pursuing a postulate which he had already had a
glimpse of in 1912.35 W hen he credited the affectionate current of
infantile sexuality with the power of carrying the primitive sexual
cathexes of the component drives along with it, he was raising the
question of where the affectionate current gets so much power from.
And although in his Three Essays on the Theory of Sexuality (1905),
instinctual inhibitions result from the latency period, sexual impulses
being held back by dams impeding the full development of sexual
activity, Freud was then led to distinguish the effect of the action of
the dams - unquestionably repression - and an inhibition internal to
the instinct, as becomes increasingly clear in each of the passages in
which he approaches the question.
It is not repression that is the cause of the aim-inhibited drive,
since it is precisely the manner in which the drive avoids repression
that is the particular feature of this drive vicissitude. And it is thanks
to this status of the drive, which is not dismantled but simply arrested
in its development, that it can assume the power to carry others along
with it which are more attached to component functions.
It should not be thought either that the aim-inhibited drives can
always be placed alongside the pregenital drives with which they are
at variance. The characteristic of the pregenital drives is to aim for
organ pleasure. Later, as a result of the new sexual aim of union with
the object, the genital erotic components carry out transformations
denouncing the orientation of the pregenital drives towards organ
pleasure and subordinating them to ends which confine them to
preliminary pleasure. Some of them will even be excluded. In other
words, the drives which have undergone an inhibition of aim will be
those whose activity has been best preserved. They will be combined
in equal parts with the erotic cathexes proper of the genital phase,
whereas those whose tendency to satisfaction has not been able, like
the ones just mentioned, to make do with an 'approximation', will be
left behind. They will contribute, by exchanging their aims and their
objects, to the complexity of the wish-organisation. Nevertheless,
their time will be limited. As they have not undergone an inhibition
of aim, they will simply facilitate union with the object. One can see
the difference: on the one hand, there is an inhibition of drive
activity which maintains the object by sacrificing the complete satis­
faction of the wish for erotic union with it, yet conserves a form of
attachment which fixes the investment of it; and, on the other, an
unrestrained development of drive activity on the sole condition that
aims and objects enter into the operations of permutation and substi­
tution, the only limitation being the influence of repression and
other drives. The first type of activity, which ultimately dominates,
makes use of the drives of the second type which are compatible with
its purpose and rejects the others. It is evident that the fate of this
contingent of drives with uninhibited aim is necessarily the most
vulnerable and the most suited to supporting the insubordination of
the ego-drives. Paradoxically, the aim-inhibited drives are those
which should, first and foremost, be characterised by their link with
the object. W ithout saying so in so many words, Freud seemed to
consider that what might be called the genital vocation towards the
object, in its quality as the definitive libidinal object, that is, of sexual
union, was present from the outset. It is in order to safeguard this
purpose, thereby avoiding that the scene is not completely aban­
doned to the pregenital drives which give primacy to organ pleasure
before all else, that the inhibition of the aim of the drive occurs.36
The Oedipus complex involves relations of affection and hostility.
Yet there exists a relative independence between the relations of affec­
tion or hostility and the phallic organisation under whose aegis the
Oedipus complex comes. The affectionate relationship towards
parents is tied up with everything involved in the sensual relationship,
censored by the threat of castration. But there is no confusion between
the two. Evidence for this may be found in the fact that maintaining
an affectionate cathexis can be the best way of getting round the fear
of castration, as in the situation described in the article 'On the
Universal Tendency to Debasement in the Sphere of Love' (1912). If
Freud linked the maternal object-cathexes in the Oedipus complex
with those which originally were related to the mother's breast,37 then
it is perhaps at this level that the inhibition of aim needs to be under­
stood; that is, at the moment when the loss of the object-breast goes
hand in hand with perceiving the mother as a whole object.38

The Function of the Ideal: Desexualisation and the Death Drive

W hen one considers the fact that the drive is in contention with
itself, and that this is not due to an evolving process, as well as the
fact that this restriction exists without the intervention of an extra­
neous force, it is difficult not to see at work here the group of drives
that are antagonistic to Eros, that is, the destructive drives. Instead of
the two groups of drives expressing their antagonism in their rela­
tions to the object through defusion, the forces of separation act, on
the contrary, by modifying the erotic drives in an intrinsic manner.
Already in 1912, Freud suspected that a solution of this kind would
ultimately be necessary when, at the end of the second article on the
psychology of love, he maintained that the sexual drive was divided
into components which work against its own satisfaction.39 It is not
the pregenital drives that impede such satisfaction but a factor that
Freud attributes to civilisation and that has become an integral part
of hereditary history.
 Our task would probably be easier if we could assume that an influ­
ence of this order - which, to Freud's way of thinking, is not to be
attributed to any form of transcendence - was a product acquired in
the progressive acculturation of each of us. But such a simplification is
scarcely acceptable here. From The Ego and the Id (1923) onwards, Freud
seems to have attributed psychical life with three centres of develop­
ment. Perception seemed to him to be so closely tied up with the ego's
activity that, on two occasions, he compared it with the relation of the
drive to the id.40 Of course, it is not a crude opposition but a
confrontation between different types of hyper-cathexes, the dialec­
tical outcome of which will be the unconscious representation of the
drive, that is, the ideational representative. A corresponding function
necessarily exists for the super-ego, a role that is fulfilled by the
function of the ideal. Moreover, Freud said that he could not localise
the ego ideal, as he had tried to do for the relations between the ego
and the id. In trying to follow the course of Freud's metapsychological
approach, one might think that the dispersed distribution of the ego
ideal, its quasi-ubiquitous presence in the field of psychical processes,
is a consequence of the topographical relations of the ego and the id.
It is as though the spatial limitation imposed on the id, at least by the
frontier which brings it into relation with the ego, were paid for in turn
by the clear field left to the function of the ideal. For although the ego
has been successful through the binding of psychical processes in
gagging - even if only partially - the id, the latter can only go along
with this by masking its defeat. Consequently, the id now imposes a
new exigency, just as pressing as instinctual satisfaction obeying the
pleasure principle, which is the copy of it or the negative double. This
new exigency will not rest until it has achieved illusory emancipation
from the former. The ego ideal, in terms of which the ego evaluates
itself and tries to achieve perfection, is calibrated according to the
demand made by the body on the mind. The pretensions of the
function of the ideal do not represent any consolation or compensa­
tion. In the very place where instinctual satisfaction occurred, it sets up
its contrary. It accords an even greater value to renunciation. Pride has
become a higher aim than satisfaction; the ideal ego has been replaced
by the ego ideal. There is nothing here that deserves autonomy de jure
or de facto, since this graft can only take root in the drive and reflect
the latter negatively. It is not so much a question of making a virtue of
necessity as of making a necessity of virtue.
Even though this function of the ideal actually originated 'from
the experiences that led to totemism' (from the experiences ... and
not totemism itself), and contains 'the germ from which all religions
have evolved',41 Freud only connects it with the primordial identifi­
cation with the father insofar as the latter is a dead father. This makes
death a necessary condition so that the aggrandisement of the
defunct occurs through signs which do not so much restore his
presence as they ensure his perpetuation in this absence for ever, thus
conferring him with eternal power. It is now necessary to return once
again to the death drive for which death represents the final outcome
of its tendency. The death drive repudiates actual death and restores
cathexis of the father, while endeavouring to eliminate all possible
tension by celebrating renunciation through the function of the
ideal. W hat is the meaning of this reference to the dead father in the
period of ontogenesis? It means that paternity cannot be transmitted
wholly from parent to child, because the father is only one link in the
chain, the succession of ancestors having become the property of
culture of which he is only a representative, the traces of which the
child will have to discover for ‘himself. These traces are written in a
different ink from the one in which experience is recorded. This
process is at the root of the primordial identification with the father.
Let us stop quibbling over the text42 with regard to the mother's
chronological anteriority and recognise once and for all what this
means from a Freudian perspective. Pointing out, as Freud does in an
adjoining note, that it may, perhaps, be safer to speak of the parents
rather than the father, does not mean that this experience is under­
gone twice, first with the mother and then with the father, but that
the motor of this inaugural identification is a principle of kinship, the
condition of being a parent to which the child will be called. Two
requirements will have to be met: the intangible preservation of the
attachment and the no less inevitable emancipation from the object.

It may be that this identification is the sole condition under which

the id can give up its objects ... From another point of view it may
be said that this transformation of an erotic object-choice into an
alteration of the ego is also a method by which the ego can obtain
control over the id and deepen its relations with it - at the cost, it
is thie, of acquiescing to a large extent in the id's experiences.43

One cannot help comparing the two types of phenomenon. They are
not equivalent, but reveal two possible destinies in which the condi­
tions are met allowing relations with the object to be maintained at
the cost of a sacrifice. Renunciation becomes the condition for the
survival of the most essential attachment; at the same time, it shows
that this relation takes precedence over all other considerations and
that there can be no question of replacing it solely by permuting the
object or aim. Renunciation or inhibition of aim furnish the best
evidence that nothing can replace the object and that no sequence of
actions is conceivable outside the ongoing relations linking it with
the ego. It is therefore no coincidence that, after these considerations,
Freud introduced the notions of desexualisation and sublimation,
although he had just spoken in the preceding paragraphs of the
earliest cathexes, those of the oral phase, and wondered whether all
forms of sublimation - a question that he was to ask himself three
times in The Ego and the Id44 - originated through the mediation of
the ego, or if it was possible that they resulted from the defusion of
the drives. In fact, we should recognise in this capacity to create
durable and permanent cathexes, the existence of a structural justifica­
tion that is always perceived as such, though never completely
clarified conceptually, which has its roots in the defusion of the
drives, that is, in the operation of the death drive upon the erotic life
drives, which include the drives of self-preservation.45
Linking these processes to operations governed by the pleasure
principle and the Nirvana principle would most likely provide
evidence of the pre-eminence of the latter. In the chapter of The Ego
and the Id devoted to the two classes of drives, Freud pursues his
hypotheses to their logical conclusion: sublimation and identification
are merely forms of transformation of erotic libido into ego-libido
which involve a desexualisation, an abandoning of object-cathexes,
which can even result in an undifferentiated neutral energy, a hybrid
form between the libido of Eros and the libido of the destructive
drives, that is, a 'mortified' libido. A libido, in any case, which is more
vulnerable to the effect of the death drive.
Freud seems to have assigned desexualisation with a very general
function capable of affecting the first object-cathexes:

By thus getting hold of the libido from the object-cathexes, setting

itself up as sole love-object, and desexualizing or sublimating the
libido of the id, the ego is working in opposition to the purposes
of Eros and placing itself at the service of the opposing instinctual
impulses.

The work thus accomplished was attributed by Freud to defusion.46

And if we take into account Freud's next statement describing the
ego's narcissism as secondary narcissism, the course of the investiga­
tion that led Freud to circumscribe, with increasing precision, the
death drive within narcissism, will enable us to trace it back to its
earliest origins.

The Protective Shield and Repression

How does such a stable, durable and permanent cathexis find its way
into the register of dynamic and economic processes? Freud only gave
examples of this by referring to states which analysts will have no diffi­
culty in recognising. The operations which govern the formation of
their structure continue to intrigue us. Now on each occasion that
Freud was pressed to give an explanation of the means by which dura­
bility and even permanence, as against mobility and change, are
acquired, he resorted to the metaphor of the transition from unbound
energy to bound energy. As we do not have an alternative solution to
propose, it is difficult to see how this metaphor can be avoided. It
should be possible to describe everything that has just been said about
the relations between the aim-inhibited drive and the object in the
language Freud used when he endeavoured to describe these processes.
Let it be said without further ado that there is no reason to think
that the inhibition of the drive aim only works in favour of the erotic
drives involving an object-choice - although it is only in those cases
that Freud speaks of it - and it is not clear why it should be excluded
in the case of the erotic drives of self-preservation. As soon as we
assume that the drives of self-preservation also have an antagonist in
the drives which are connected with preserving the species, and
which are accomplished by fusing with the object in the genital rela­
tionship, it can be seen that here, too, the inhibition of aim protects
the object against being completely assimilated into the ego, which
would result in the dissolution of the ego's organisation.
The mechanisms for converting unbound energy into bound
energy, described by Freud, show how the organism protects itself
against excessive amounts of external stimulation by offering a
surface of resistance which has been subject to a neutralisation of
cathexes but is capable of receiving, absorbing and conveying
external stimuli. One can see, then, that this barrier or 'protective
shield' has the twofold function of prohibiting, at its level, all trans­
formation of the reception of stimuli involving alterations in the
register of expression, mutation, combination, and so on. It is simply
a question of absorbing: that is, of transmitting, without deforming,
the Weakened result of what has been recorded. Its functions are
therefore those of blocking - reception and binding - and of trans­
mission by means of circulation. Protection takes precedence over the
capacity for reception. A similar surface receives the impression of
internal stimuli and also seeks to avoid too great an afflux or an exces­
sive am ount of excitation. But it is obvious that, however
homologous they appear to be, the two operations are not equivalent.
For the force resisting the external stimuli eliminates them, whereas
the repudiation of internal stimuli merely results in their returning
towards unconscious processes, that is, in a new charge, leading to a
new thrust towards consciousness, in the face of which the possibili­
ties for repudiation will be limited. A device comparable to the
protective shield can therefore not function here. The link between
the two modes of activity, that which has the function of dealing with
external stimuli and that which protects against internal stimuli, is
thus not conclusive. Once again, Freud made use here of the
metaphor comparing the organism with a living vesicle. The initial
reality ego allows a distinction to be made, it is true, between the
origins of the two sources of excitation, but its action is not without
shortcomings since projection remains a possibility. Furthermore, the
intervention of this projective mechanism occurs on a scale which is
much too great for us to be able to imagine - think, for a moment, of
the case of pain - that a breach of the system could lead to an osmosis
such that what comes from the inside is in fact treated as if it came
from the outside. This operation does not merely involve a rejection;
it has the advantage of creating the possibility of mustering the
means necessary to defend itself - once the externalisation has
occurred - against the source of the projection.
Freud himself expresses some reservations about this way of
picturing things in the Outline.47 The relations between the two layers,
inner and outer, might perhaps offer a better solution. The special
feature of the outer layer of the metaphorical organism is that it has
been 'worked on' to such an extent that it has succeeded in lowering
all the organic processes to a m inim um . It contents itself with
knowing the source and nature of the excitations, which is possible
due to its orientation. In fact, such an achievement is not unrelated to
the types of processes which, under the effect of the Nirvana principle,
make it their aim to abolish all tension. Freud even says that the death
of this outer layer seems to have saved from a similar fate all the
deeper ones which shelter the sense organs, the latter dealing only
with very small and selected quantities of external stimulation.
My conclusion, then, is that such a system cannot be applied to
the internal barrier. But although Freud compares them, he does not
see a similitude between them - which is impossible - but an analogy.
It is as though the model provided by the protective shield provided
a tempting solution for the internal stimuli. The stimuli are thus
treated as quantities of excitation to be reduced, bound, rendered
'inanimate' or mortified. And although some tensions continue to
break through the dams and to produce effects comparable to an
external trauma, this does not occur frequently. The force of binding
will depend on the quantitative level of the cathexes of the system.
As this quiescent force no longer has the function of neutralising or
disqualifying excitations, as the protective shield does, it offers an
equivalent of it: a mirror in which the lure of the removal of tensions
is reflected. The id becomes, to use Freud's felicitous expression, the
ego's 'second external world'.48 In certain circumstances the terminal
organs receiving external excitations also transmit sensations and
feelings such as pain. The work of the internal force of binding is to
render internal stimuli perceptible and to master them (by reducing
tensions). But it has less capacity to discern the source of the excita­
tions so that what it experiences as coming from everywhere - and
Freud, whose statements are never vague, speaks here of 'something'
which corresponds to sensations and enters consciousness - is subject
to all manner of confusion with regard to its localisation. However,
the result of this is that the comparison with the terminal organs
which receive the external excitations makes an analogy possible and
Freud says that 'as regards the terminal organs of sensation and
feeling, the body itself would take the place of the external world'.49
This does not mean that we are justified in speaking of a confusion
between them but simply of a duplication of the latter, which can
also be understood as a division. Yet by interposing a 'second external
world' in the relations between the id and the ego, Freud was re-eval­
uating the relations between the three agencies. The process of
mortifying inertia, established within the envelope which serves as an
intermediary with the outside, corresponds to the system (repression)
which protects against exigencies and pressures. Liberation from the
latter poses more problems than dealing with external stimuli. As we
have seen this initial evaluation is complicated by the action of the
function of the ideal.

Auto-Erotism
A comparable process can be observed here to that which occurs
when the erotic drive is subject to an inhibition of aim, though not
all of its characteristics are the same. Of course, auto-erotism does not
have the durability and imm utability of the affectionate relations to
which Freud refers, but it is quite clear that auto-erotism and narcis­
sism are not only stages. The ego - or, initially, the ego drives - can
present itself as a source of satisfaction by means of mechanisms
which will last throughout life. It is legitimate to want to assign a
beginning, a point of departure for auto-erotism, as Laplanche and
Pontalis do,50 when they insist on the fact that the drive becomes
auto-erotic when it has lost the object. Freud's own statement on this
point is too important for me not to cite it in full:

At a time at which the first beginnings of sexual satisfaction are

still linked with the taking of nourishment, the sexual instinct has
a sexual object outside the infant's own body in the shape of his
mother's breast. It is only later that the instinct loses that object,
just at the time, perhaps, when the child is able to form a total idea
 of the person to whom the organ that is giving him satisfaction
belongs. As a rule the sexual instinct then becomes auto-erotic ...
(SE, VII, p. 222)

W hen Laplanche and Pontalis point out in another passage that the
object does not have to be absent for the auto-erotic condition to
come about, their argument is indisputable. But in that case we should
probably attempt to define auto-erotism more closely.51 One cannot
take Freud's remark out of context, and the interesting point here is
that this process is tied up with introjection. W hat needs accounting
for is the transition from the object of satisfaction 'outside' to the
search for satisfaction, if not 'inside', then at least in the infant's own
body, at its surface of contact, thus concretising in a remarkable way
the proposition that the body takes the place of the external world. I
agree with Laplanche and Pontalis who maintain, following Freud,
that the ideal of auto-erotism is 'lips kissing themselves'. The signifi­
cance of this image needs to be understood in a much wider sense;
that is, as a movement of a more general and fundamental value. It is
not so m uch that the distribution between infant and object has been
abolished but that, before acceding to the status of a subject, that is,
at the moment when the object, which had hitherto been 'outside',
was lost, the 'subject' was still only a centrifugal movement of inves­
tigation. Separation reconstitutes this couple in the subject's own
body, since the image of lips kissing themselves suggests the idea of a
replication followed by a re-gluing which, in this new unity, traces the
line of partition which has enabled the 'subject' to fall back on his
own resources. Auto-erotism occurs during this retreat; it represents a
kind of stopping place on the frontier, and in this respect may be
compared to the inhibition of aim described in relation to the libid-
inal erotic drives.52 We have seen that the inhibition of aim is closely
tied up with retaining the object. Now what is striking in this auto­
erotic situation is the particular status of the drive, with regard to its
aim and object. It is not possible - and, on this point, I agree with
Laplanche and Pontalis - to link auto-erotism with the object's
absence. But on no account can what is happening here be likened to
a substitution of the object, or even to a change of aim, since the latter
remains unchanged; that is, the pleasure connected with sucking, for
which thumb-sucking is not the equivalent but the quintessence. This
is why auto-erotism is, to a certain extent, organ-pleasure; but only to
a certain extent. Saying that the auto-erotic character of the drive is
'an anarchic product of the component drives'53 is perhaps abusing
the theory slightly, since it means situating the said drive on the same
side as the so-called drives with an inhibited aim, characterised by
constant displacement, transformations of energy and repeated
permutations of aims and objects. Primordially, the auto-erotic drive
is a drive capable of satisfying itself, either in the object's absence or
presence, but independently of it We can only understand the issue
clearly if we assume, as Freud did, that there are two categories of
drives: those capable of finding satisfaction in the subject's own body,
and those which cannot do without the object. Consequently, there is
no longer any reason to link auto-erotism to the emergence of
desire,54 as Laplanche and Pontalis do, since the latter is in fact desire
for contact with the object. Moreover, their conception neglects the
role of the drives which require the object's participation. By the same
token, it is not necessary, as Pasche maintains, to postulate an anti­
narcissism,55 since this is implicit in this latter category of drives. This
differentiation in Freud's* work occurs in the context of a remarkable
continuity of thought. For if one does not wish to regard auto-erotism
merely as a stage, then one must exploit fully all the theoretical poten­
tial that the notion contains - not always explicitly - in order to
justify rejecting a genetic position which is over-simplified, incom­
plete and rather unsatisfactory.
Let us pause a m om ent over the following passage taken from
'Instincts and their Vicissitudes':

Originally, at the very beginning of mental life, the ego is

cathected with instincts and is to some extent capable of satisfying
them on itself. We call this condition 'narcissism' and this way of
obtaining satisfaction 'auto-erotic'.56

It was when he wrote these lines, which at first sight seem to reinforce
the genetic point of view, that Freud added a note which has caught
the attention of many authors, including W innicott. Freud recognises
here that the group of sexual drives and the drives of self-preservation
are not homogeneous, and that it is still necessary to differentiate
between drives that are capable of finding satisfaction, without neces­
sarily involving the object, and drives which cannot economise on
the link with the object. W hat makes the functioning of auto-erotic
drives possible is the vicariousness of the mother's care. But this does
not am ount to saying, however, that the latter is subordinated to the
drives which require a relationship with the object. It is not because
the mother ensures that the infant's needs are satisfied and makes up
for his immaturity that she has the absolute function of a primordial
object. The latter would deprive the infant's own organisation of all
reality; an organisation which acquires its value not on a biological
level, of course, since without the mother's care the infant would die,
but in the dom ain of desire and of the signifier. The mother shields the
infant's auto-erotism.
 These remarks throw light on the question I was discussing earlier
of the origin of primary cathexes which, depending on the version of
Freud one follows, arise either from the ego or the id. Strachey was
right to define the parameters of the debate by reminding us of the
primitive undifferentiated state of the id and the ego. Would we not
come even closer to the truth if, in an attempt to understand these
relations, we pictured an id which partly included the mother,
cathected primitively and directly, and an ego forming itself around
its own possibilities for satisfaction. The latter are essential for their
founding role but are called into question by the drives which can
only find satisfaction in the object.

Repression and the Ego

Perhaps you can see more clearly now the parallel that I am drawing
between drives with an inhibited aim and auto-erotic drives. Is it a
coincidence that the most common signs of affection, caresses and
kisses, belong to both categories? Auto-erotism is thus one of the
succession of phenomena through which the body takes the place of the
external world.
W hat we need to consider now is how we should conceive of the
protective barrier w ithin the perspective of a structural theory, while
distancing ourselves as much as possible from an approach based on
archaeological reconstitution. Modelled on the protective shield, the
protective barrier makes it possible to register the stimuli coming
from the body, this second external world, on a screen, so to speak.
In some recent metapsychological conceptions this role has been
attributed to repression (Laplanche and Pontalis, Stein), which is
regarded as having the property of founding the conscious and
unconscious registers as well as separating primary and secondary
processes.57 W hile this point of view has the advantage of centring
the distinctions on a founding act, thereby allowing the various kinds
of facts and phenomena to be articulated more easily, it seems to run
the risk of postulating the existence, beyond repression, of an un in ­
telligible chaos that is opposed to the primordial order from which
intelligible structure emerges. The protective shield, which has the
capacity of localising the external source of the stimuli, sees its action
strengthened by the reality principle,58 which makes the distinction
between the ego and the external world quite clear. Repression could
be thought of as its stand-in. From this point of view, some investi­
gators would place primary narcissism in this area beyond repression
- in a world which is without order or limits, in which the ego would
be merged with the cosmos; hence the description ego-cosmic. In my
judgement, this situation pertains more to the id than to narcissism.
Now as L have already pointed out, the characteristic of absolute
primary narcissism is that it strives towards a zero level of excitation.
Doing away with all motion, protecting oneself from all tension, does
not necessarily generate the feeling of expansion, although it may do
so sometimes.
It is important to remember that on numerous occasions Freud
refused to give repression the status of an inaugural function, even
after an interval of nearly twenty years:

Originally, to be sure, everything was id; the ego was developed out
of the id by the continual influence of the external world. In the
course of this slow development certain of the contents of the id
were transformed into the preconscious state and so taken into the
ego; others of its contents remained in the id unchanged, as its
scarcely accessible nucleus. During this development, however, the
young and feeble ego put back into the unconscious state some of
the material it had already taken in, dropped it, and behaved in the
same way to some fresh impressions which it might have taken in,
so that these, having been rejected, could leave a trace only in the
id. In consideration of its origin we speak of this latter portion of
the id as the repressed,59

From this passage it emerges that:

1. the ego is not constituted by repression, but is pre-existent to it;

2. if these traces are only left in an id separated from the ego, the
form in which the primitive contents of the id have been
accepted and admitted nonetheless remains a problem;
3. repression does not carry out a primal separation, but rejects that
which has already been admitted once;
4. the division into unconscious-preconscious is necessary for acti­
vating repression;
5. finally, repression is linked to a mechanism involving the re­
passage, the re-turn, of the re-pressed.

An unavoidable question arises here: why is it that what has been

admitted once is later rejected? Even when one attaches great impor­
tance to counter-cathexis - involving a considerable expenditure of
energy - one should not lose sight of the fact that repression is also 'a
preliminary stage of condemnation'. Linking these two aspects is
undoubtedly interesting from a heuristic perspective. The advantage
is that it makes the processes of judgement consubstantial with those
of energetic energy. But this is perhaps being a bit hasty. Not that
there is any reason to question the connection between the order of
the signifier and the energetic order. But in my opinion, this link
requires further mediation. Freud seemed to have a reason of this sort
in m ind when he wrote:

It has consequently become a condition for repression that the

motive force of unpleasure shall have acquired more strength than
the pleasure obtained from satisfaction.60

Now the only type of pleasure we know of that has claimed to safe­
guard - under the cover of maternal care - such a possibility of
satisfaction, without incurring unpleasure, is auto-erotism.61 The
period of separation from the mother and the period of repression
may come together again afterwards, but they are not merged from
the start, since this conjunction of periods is inferred retrospectively
through the search for the lost object which unites the actual loss of
the object at the time of separation and the loss suffered through
repression. I believe it would be more coherent to account for this
search in a different way. The loss of the breast, occurring when the
mother is perceived as a whole object, which implies that the process
of separation between them has been completed, results in the
creation of a mediation that is necessary to compensate for the effects
of her absence and integration into the psychical apparatus. The latter
occurs independently of the action of repression, whose aim is
different. This mediation represents the constitution in the ego of the
maternal setting as a framing structure.
The continuation of Freud's text helps us to see more clearly:

Psychoanalytic observation of the transference neuroses,

moreover, leads us to conclude that repression is not a defensive
mechanism which is present from the very beginning, and that it
cannot arise until a sharp cleavage has occurred between conscious
and unconscious mental activity - that the essence of repression lies
simply in turning something away, and keeping it at a distance, from the
conscious.62

To say that the essence of repression resides simply in repudiating a

psychical content is not to diminish its importance; it is merely to
specify its function without in any way ignoring its privileged
position.
Certain passages in Inhibitions, Symptoms and Anxiety (1926
[1925])63 go to great lengths to draw a comparison between the
defence offered by the protective shield against external stimuli and
that which resists internal stimuli. We need to be very attentive to the
idea that it is flight, more than repudiation, that is the fundamental
mechanism in the second case. Here linguistic connections are
lacking, for there is the notion of a turning away, of a dismissal in the
Freudian term which, when all is said and done, implies an active
attitude of counter-cathexis, whereas flight is an attitude which, so to
speak, is actively passive.64 The two modes of defence might be
compared - although this can only be expressed partially through
images - to tactics whose very principles are opposed. The first, the
protective shield, m ight be compared to tactics of withdrawal
whereby, depending on the size of one's forces, one periodically turns
and faces the enemy, taking advantage of each engagement to consol­
idate one's defences, so that they are in a position to hold their
ground when the moment comes to thwart the enemy's thrust. The
second, which could be compared to internal stimuli, exercises a
withdrawal using all resources available to carry out a scorched earth
policy until a fortified place is reached where better times are awaited.

The Double Reversal and Primary Decussation

There is no denying that repression partakes of these two forms. In the
same passage, Freud adds that 'repression is an equivalent for this
attempt at flight', but he does not recognise the primary flight itself.65
The correction of the interpretative error which tends to confuse the
two can be found in one of the addenda of Inhibitions, Symptoms and
Anxiety. The concept of defence covers the general category of the ego's
measures of protection against instinctual demands and provides good
grounds 'for subsuming repression under it as a special case'. Freud thus
reneges on the earlier solution he had adopted in which repression
seemed to illustrate the process of defence in general. But he also adds:

In addition we may look forward to the possible discovery of yet

another important correlation.66 It may well be that before its sharp
cleavage into an ego and an id, and before the formation of a
super-ego, the mental apparatus makes use of different methods of
defence from those which it employs after it has reached these
stages of organisation.67

Once again, one could simply put a question mark here against the
text, regretting that the author did not explain himself more fully.
And yet it is practically the same sentence as we find written, eleven
years before, in the text 'Repression' (1915b):

This view of repression would be made more complete by

assuming that, before the mental organisation reaches this stage,
the task of fending off instinctual impulses is dealt with by the
 other vicissitudes which instincts may undergo - e.g. reversal into
the opposite or turning round upon the subject's own self.68

This echoes a similar passage in Tnstincts and their Vicissitudes'.69 In

fact, here Freud described a single process involving two operations.
This concerns, firstly, the orientation - the change of which indicates
that the centrifugal orientation has become centripetal - and,
secondly, the mode of reversal which is not just a reversal of direc­
tion, nor a mere change of sign, but should be thought of as a
decussation. The direct confusion of the two mechanisms would result
in withdrawal which would in no way resolve the problem posed by
the instinctual demand; the only solution for which is a modification
inscribed in the body which leaves a trace of satisfaction. In this
reversal through decussation, it is rather as if the response expected
from the object were carried along by this movement in which the
extreme positions of internal and external in the instinctual current
change places. An overlapping thus occurs between that which, on a
surface, can be localised on the left and right of a hypothetical
frontier. This movement of return permits the drive to reach the
bodily zone which is awaiting satisfaction as if it were the object itself
that had provided the satisfaction there; for, as with the inhibition of
aim, the object has been retained and not exchanged. But this preser­
vation has been paid for by limited satisfaction - something I regard
as the negative of a metonymical operation, since it resists the
suturing of the subject and the object. At the same time such a lim i­
tation preserves it, because this union would exclude any follow-up
to this first and last sequence. W hat is created, then, is a circuit which
does not concern the properties of the object but its response. While
preserving the object in its absence, the circuit delegates the object's
function to the subject in such a way that the subject experiences the
realisation (implied by the response) as if it had been provided by the
object. It is clear that a metaphorical process is at work here.
This may throw light on the mutation that occurs between the
relation to the breast - which can equally well be described as 'the
mother suckling the baby or as her being sucked by it'70 - and the
reversion of lips kissing themselves.
Between the undifferentiated state of ego-id and mother-infant
and the emergence of repression, a mediating process occurs
involving a regulation of the drives, on the basis of which repression
is possible. In other words, between the 'biological' process at work in
the protective shield and the psychological process of repression
which Freud refers to, there is no correspondence as between outside
and inside; but there is an intersection between them, so that what is
inside may be treated in the same way as that which comes from the
outside, provided that the inside is perceived as if it were the outside,
and without there being any fusion between the two. This is exactly what
Freud affirms in Beyond the Pleasure Principle, which links the forma­
tion of an internal barrier to projection. The double reversal allows us
to think about this mediation structurally. If we read the passage on
the double reversal it shows that Freud was describing the work which
prevents a drive from attaining direct satisfaction, not by means of a
force that is alien to it - repression as a psychological process - but by
an internal modification of its own nature.
W hen Freud considered that it was necessary to distinguish two
processes in the reversal of the drive into its opposite: namely, the
change from activity to passivity, and the reversal of its content (love-
hate), nobody seemed to be concerned that, in so doing, he had
introduced a new characteristic of the drive, that is, its content, which
was never mentioned in later descriptions, or, only with respect to the
id. A problem arises here that is closely related to that of auto-erotism
and the relation to the object. Freud denies that the antithesis between
loving and hating can fit into the scheme of the reversal of a drive into
its opposite in the same way as the change from activity to passivity
can, these affects only being applicable to a whole object. Narcissism,
a state in which one loves oneself, appears to represent the form, at this
last level, of the equivalent of it in the transformation from activity to
passivity. There is therefore justification for saying that it is when drive
activity is understood as the ego's relation to sources of pleasure related
to the object, considered as independent of the ego, that the reversal
from activity into passivity assumes the form of the love that the ego
can have for itself. And if we ask ourselves what the preparation of this
structural stage corresponds to, we need to bear in m ind a distinction
Freud considered to be imperative, which led him to separate the oper­
ation of reversing the aims of the drives from the turning-around upon
the subject's own self. He was right to separate them but the situations
which he encountered (sadism-masochism, scopophilia-exhibi-
tionism) soon led him to make the following observation:

We cannot fail to notice, however, that in these examples the

turning round upon the subject's self and the transformation from
activity to passivity converge or coincide.71

Retaining the object, and maintaining certain cathexes in a lasting

and unchanged mode, are closely tied up with the aim-inhibited
drive. Auto-erotism spares the object and does not lose it entirely;
since it is just when the subject is able to apprehend the mother as a
whole that the drive becomes auto-erotic. And if the subject seems to
change object, it is only to turn towards the object o f the object (the
subject's body) in order to create there a second erotogenic zone,
though 'of an inferior kind',72 thereby vouching for its fidelity. On
the other hand, the fact that the loss of the object coincides with the
moment when the organ which was the source of the infant's satis­
faction, that is, the breast, and the mother, to whom it belongs, are
united, as well as the fact that this loss paves the way for the onset of
auto-erotism, suggests that the infant's apprehension of the link
between organ and person may also have been internalised. This
internalisation does not lead to awareness of a bodily form; but,
because the circular mode of cathexis has been closed down, it gives
rise to feelings of autonomy, perfection and of being delivered from
desire through the symmetrical creation, almost simultaneously, of
the global and unifying apprehension that is characteristic of the
infant's ego, as it was described by Lacan in the mirror stage.

The Ego and its Ideal

In this context, inferiority and independence are associated terms.
The sense of inferiority exists because the persistence of a lack with
regard to the object is not abolished by auto-erotism; independence
shows how the tutelage of desire is a most formidable yoke, and,
though it is undoubtedly necessary to psychical organisation, it needs
to be overcome if structure is to be acquired. Once again what we are
faced with here, as was the case with the inhibition of aim, is the
work of the death drive. The mark of the death drive is to be seen not
in its impossibility of attaining a goal but rather in the choice of this
zone 'of an inferior kind', with a special vocation. The lowering of
tension to the level zero and the immediate suppression of all differ­
ence abolishing the object's absence are both hallowed in the temples
of self-sufficiency. The impression this leaves is indelible and it will
persist, if not at each moment of life, at least throughout the whole
of life. 'To be their own ideal once more, in regard to sexual no less
than other trends, as they were in childhood - this is what people
strive to attain as their happiness.'73
But can we be certain that this mediatory stage between the lack of
ego-id differentiation and repression is to be connected with narcis­
sism through auto-erotism? Is there any other way of seeing things
that offers a convincing alternative? Either narcissism is cast out into
the chaos prior to repression or it is specified as a field of illusion; but,
in any case, it lacks its own structure. Freud appears to have had a
solution:

We shall be approaching a more general realisation - namely, that

the instinctual vicissitudes which consist in the instinct being
 turned round upon the subject's own ego and undergoing reversal
from activity to passivity are dependent on the narcissistic organi­
sation of the ego and bear the stamp of that phase. They perhaps
correspond to the attempts at defence which at higher stages of the
development of the ego are effected by other means.74

Narcissism is founded on the ego drives. But it would be wrong to

think that, because I have based my interpretation of auto-erotism on
the drive contingent that is capable of obtaining satisfaction without
the object's assistance, I therefore consider that this mechanism alone
can provide a solution for all the unanswered questions. But I still
want to tackle the problem of the ego's unity which Freud associated
with narcissism. There is a big difference between 'the energy of the
ego drives' and narcissism. For this expression, which Freud took as
his point of departure, needs to be related to the highly indeterminate
character of the primitive drives. The drive finds its vocation during
the course of its actual functioning, which is set in m otion by its aim,
but only discovered in the course of effective action directed at a
specific goal. Inferring the drive's activity does not am ount to intro­
ducing a teleological point of view into Freudian theory, since the
innate spontaneity of the way it is activated is enriched by the
discovery of the aim that motivates the very process of its activation.
By taking the energy of the ego drives as my starting-point, I do not
mean to attribute any sort of biological character to this pre-form; but
I am trying to picture in the most convenient way a current of
cathexes between two limits separated by a difference of potential,
without which it would be impossible to identify specifically any
current whatsoever. In short, I regard it as the prerequisite state for
the formation of a chain. A suitable means of expressing this needs to
be found if we are to understand how Freud could maintain that the
infant is unable to distinguish between its body and the breast, yet is
able to localise the latter when it is absent - although a lack of differ­
entiation persists - 'outside' itself. I share the view of Laplanche and
Pontalis that this whole process is dominated by anaclitic object rela­
tions; but I am tempted to liken this mechanism, in which the
activity of need coincides with the emergence of pleasure at the
points where need is appeased, with the difference between the
'locus' of the satisfaction of pleasure and that which makes its satis­
faction possible. If their association constitutes a demand, I would be
inclined to think that the demand and its circuit are separable. The
circuit is cathected before demand. W hich does not amount to saying -
as Lebovici argues - that the object is cathected before it is perceived,
but rather that cathexis is cathected before the object is. Just as the
repressed does not simply remain banished from consciousness, but is
attracted by the pre-existing repressed and moves towards that which
is ready to take possession of it, similarly, the pathway of cathexis
only exists because it is cathected by the mother. But it is important
to understand that the function of the two currents is placed under
opposite signs. For the mother only unites with her child insofar as
she has consented to separate from him in the future; and to the
extent that the child, in his encounter with her, is forced to recognise
the limitations of self-preservation. By trying to preserve himself, he
endeavours to m aintain the link that has been established, though, in
another sense of this term, he has to appropriate the condition of his
satisfaction along with the source of pleasure.

The First Difference

We cannot go any further without examining the antagonism between
Eros and the death drives. In the initial state of affairs, the id and the
ego, each indistinct, thwart the action of the destructive drives which
were working in the child towards a return to the earlier situation;
whereas, in the mother, the movement of Eros finds an ally in the
desire to reintegrate the product of creation.75 A genuine reversal of drive
values needs to occur in order that decisive change can come about. As far as
the mother is concerned, the forces that are pushing for separation
must succeed in making themselves heard,76 whereas the child needs
to hold together the portion of the mother's id which serves these aims
and everything that has sided with the individual's clamour of life. And
so what, in the initial 'stage', had no other purpose than the suspen­
sion of all disturbance, now has another meaning in this new context
which is to lead the individual to himself, towards submission and the
binding of the ego. Its purpose is not only to muzzle or reduce the
chaotic id to a state of impotence, but also to seal the mark of self­
belonging and belonging to the other. One can see that this reversal of
values necessitates a de-centring of drive polarities from the mother to
the child, and of their common id, so that an ego may be born. The id
has created object-cathexes which the ego takes possession of. This is
the first transgression. But these are not the ego's only origins for it can
also make use of the portion of cathexes that do not necessarily involve
the object. It can be seen that this alliance between the ego and the id
can only come about through a relative synergy; for, although the ego's
action is that of 'binding', this can only be carried out to the extent
that the ego has agreed to support, within itself, the pursuit of the
removal of tension which prevailed in the work of the death drive. 'The
pleasure principle seems to be in the service of the death instinct.'77
It is clearly difficult to maintain a strict opposition between the
two types of antagonistic drives, but each time one of them seems to
have gained the upper hand over the other, the force which got the
better of it in the conflict between them has to be internalised into
this new state of things. Freud's thought does not lend itself to simpli­
fication. In certain places, therefore, the id is conceived of as an
antagonist of the libido. Pleasure, which has become the character­
istic of the libido, is used by the id against the libido. It should not be
thought that the id supplants it, but rather that their aims converge,
provided that pleasure is the beneficiary. Moreover, when I speak of
the forces of union, nothing is further from my m ind than to present
them as the equivalent of physical forces; they should be considered
more as orientations and aims, impersonal and personal. Ulysses' ruse
was to make use of the polysemy of language; the same word desig­
nating both nobody and a particular person. The overlapping of
operations does not always allow us to see what is involved in them.
W hen I allude to the contradiction between preservation and appro­
priation, it will be clear that I do not in any way picture this reversal
in favour of the ego as a monopolisation or a seizure of assets in
favour of the purchaser, so to speak. And although separation from
the mother is what is implied, it would be wrong to imagine that she
has been abandoned or that the cathexes of which she was the object
have been transfused.

Negative Hallucination of the Mother

Let us return now to Freud's assertion that the perception of the
object is linked with its absence. It is against the backcloth of this
absence that signs have to be created which will be inscribed where
there is a lack, as an exchange value and not as a substitute object.
But, as this perception of absence goes hand in hand with an aware­
ness of loss, the two tend to become merged; or, auto-erotism is
considered as the new form that is capable of resolving the problems
raised by these two observations. Although Freud saw the loss of the
breast and the moment when the infant apprehends the mother as a
whole person as contemporaneous, what precedes this apprehension
must potentially include the content of the later appropriation. Not
in the form of a perception because, if this were the case, the object
would be outside, and the representation of this perception would
only be a copy, whose function of replication would not be congruent
with the reversal of polarities which focuses the effort of unification
on the ego itself. O n the contrary, it would take the form of a negative
hallucination of this global apprehension. Auto-erotism at the body's
door is a hallmark of independence from the object; negative halluci­
nation signals, with the total perception of the object, that the latter
has been put 'outside-of-I', which is succeeded by the I-not-I on which
identification is based. While such negative hallucination cannot, by
definition, be represented by an image of any kind, it can be observed
in the constitution of the circuit of double reversal of which auto­
erotism represents only the mark of the function or suture, and is
carried out - and here the operation of reversing activity into
passivity is more fundamental than the reversal against the self -
through the inversion of the polarities between mother and infant.
He treats himself as she treats him once she is no longer merely an
excentric part of himself. The mother is caught in the empty frame of
negative hallucination and becomes a framing structure for the subject
himself. The subject constructs himself in the place where the object's
investiture has been consecrated to the locus o f its investment. Everything
is then in place so that the infant's body can take the place of the
external world.
By resorting to the example of the wooden reel Freud not only
represented the creation of the status of absence. To say that what
mattered to him above all was to stress the aspect of mastery involved
in this activity would be doing violence to his thought. The phonetic
opposition accompanying the game is indeed linked to the signifier.
However, it cannot be separated from the circuit supporting it.78 It
goes without saying that the child is not the creator of this circuit;
otherwise the concepts of the division of the subject and of the
subject of the unconscious would be reduced to nothing. The entire
Freudian system of interpretation is based on the following device:
the wooden reel, the piece of string for pulling it in again, the
curtained cot, the active movement of throwing away and reeling in
again. At this stage, the child makes use of his hands but it is the
mother who plays the active role by returning. The reversal of the
subject's polarity is indicated by the link Freud established between
this game and the appearance-disappearance of the baby's image in
the mirror, as if he had been seen by someone else, even though he
was the one to make the movements allowing an image of himself to
be formed, just where the mother was expected to appear. The child
says: 'Baby o-o-o-o!', providing a further reason for linking the
negative hallucination of the mother with identification. But let there
be no misunderstanding. Not all the cathexes have suffered this fate
but precisely those that are susceptible to binding through auto-satis­
faction. Nothing is detracted from the cathexes of the component
drives which continue, in their mobile, changing, fragmentary form
to enter into contact with the object of those cathexes which the loss
of the object cannot compensate for by means of identification. It is
this contingent of drives that is subject to repression. This throws
light on Freud's idea that what is repressed has already been admitted
into the ego. Such is the lot of the homologous portion of cathexes
capable of auto-satisfaction, which are in no way different from the
others, in as much as they are object-cathexes, before this outcome is
offered to them.
Freud always thought that repression affected the forms of repre­
sentation (the affects which are subjected to this fate are only those
w hich have been linked at one time or another to
Vorstellungsreprasentanz). Are we not justified in inferring that the
negative hallucination of the mother, without in any way repre­
senting anything, has made the conditions for representation possible?
That is, the creation of a memory without content, the transition
from repetition to the suture preceding the presence of the elements
of the suturing which the chain they form will presuppose.

Desire for the One

Narcissism erases the trace of the Other in its Desire for the One. The
difference inaugurated by the separation between mother and child is
compensated for by narcissistic investiture. This is the term which, in
all respects, founded difference on the place that the child occupied
in the mother's desire. W hen difference has not been established,
another difference is created owing to the fact that the mother is
caught in the framing structure. The partial cathexes which were
destined for her, nonetheless enter into the series of exchanges and
transformations which occur between them, of which the forms of
representation will be both the product and the witness. It is at this
point that the barrier of repression, which is the stand-in for this
circuit, forms the wall on which the component drives will be
reflected, facilitating the dissociation that allows representation to
occur. Repression is now able to carry out its task of repudiating the
drive which is considered undesirable. This is a stage which opens the
way towards other forms of exchange in which these intersecting
conversions between object-cathexes and secondary narcissistic-
cathexes 'stolen from objects' occur, the economy of which is
regulated by the structure I have just described. Once this process has
been completed, the ego, making use of the closure for which the
contours of negative hallucination provided the model, will be able
to offer itself as a love-object to the portion of the id which it has
taken possession of by assuming the object's features: 'Look, you can
love me too - I am so like the object.'79
This way of seeing things might account for Freud's remarks on the
indestructible character of early identifications and the ego's narcis­
sism as secondary narcissism. During the first stage, the primitive
stamp left by the object inspires the ego in its endeavours to offer, not
its resemblance with it, but the self-sufficient quality of its own
imprint. The features borrowed from the object can be diversified,
selected and isolated one by one, but they should be able to give the
subject the feeling that they make him independent of desire. It is
possible to imagine here a new form of anaclitic object relations,
between two forms of narcissism.
Once the process is completed, negative hallucination will have
established the boundaries of an empty space, as in a Mobius strip.
This is what Freud repeatedly pointed out when he introduced the
final theory of the drives. The division between ego drives and sexual
drives amounts to replacing a qualitative distinction by a topograph­
ical one,80 which involves a great deal more than simply assigning a
direction to cathexes and lays down the same foundations of a
psychical apparatus as I have postulated in my description. The struc­
ture of the Mobius strip provides the equivalent of this double
reversal and circumscribes the two parts of the empty space I have just
referred to.81 They will be occupied respectively by the object-
cathexes and the ego-cathexes which are denied auto-satisfaction as
the latter depends on the libidinal erotic drives. Circumscribed spaces,
then, with different orientations and contrasting directions; but it is
possible to pass alternately from one to the other by making a detour
along the external and internal surface, the dividing surfaces of each
space allowing the exchange of these two types of cathexes.

Introjection and Projection

It is of course impossible to make links between all these mechanisms
without referring to the fundamental role played by introjection.
When Freud was commenting on the process of introjection during the
phase that bears the stamp of narcissistic organisation, and declared
that the object is consumed, incorporated into the subject but also
destroyed, his remarks are unintelligible if the entire cathexis takes the
path of destructivity; for, how can something be preserved if there has
been total destruction? A satisfactory answer might be that introjection
becomes merged with the inscription of the framing circuit, thereby consti­
tuting the matrix of identifications and coinciding with the object's
disappearance. Introjection is dependent on the closure of the circuit
which, as I have said, results in the abolition of tensions. The emer­
gence of auto-erotism, which proceeds along similar lines to the
satisfaction of drives independently of the object, completes the
process. Later introjections will be dismantled in the same way as the
identifications I have just mentioned, constituting the group of object-
cathexes. It will come as no surprise, then, that projection has a part to
play here, since the whole effect of the reversal of activity into passivity
is to make the subject responsible for what apparently takes place
outside him. The mother's excentric position is invalidated by the
forming of a circuit which re-includes in the individual the polarity he
tends towards, in such a way that it becomes an integral part of
himself. The formation of the Mobius strip no longer allows us to speak
of a wrong side and a right side, of an interior and an exterior, though
they should not become merged in a universe without limits.82
It is a mistake always to situate projection beyond the subject's
limits, as hypochondria provides us with a contrary example.
Hypochondria is often regarded today as the result of an introjection.
Following Tausk, who understood the essence, if not the structure, of
narcissism so clearly, it should be seen as an example of projection
from a distance into the body, the discovery of the lost object. The
hypochondriacal object is 'cut out' of the body by the bodily libido of
the psychical cathexis allotted to the ego. The body has taken the
place of the external world, thus allowing psychical cathexes to be
formed. The hypochondriacal organ represents the negative of auto­
erotism, the point at which the negative hallucination of the mother
is ruptured; a moment when the body (which had taken the place she
occupied primitively), undoing the internalisation of this exteriority,
restores her presence, or rather the presence of the object, whose
absence was a sign that it had been localised outside the infant. But
the hypochondriacal organ is more than this; it is also a source of
scanning, investigation and listening. It is the eye in the body which
feels, senses, guesses and warns.

The Eye of Narcissus

Freud thought certain structures of narcissistic origin had the role of
evaluating the ego, measuring themselves against it, competing with
it and striving for ever greater perfection in relation to it. I link these
structures with secondary narcissism. The struggle in question, which
sustains the ego, occurs between satisfaction and the renunciation of
libidinal satisfactions. The sacrifices it makes seem negligible in
comparison with the sense of pride it derives from doing so. We know
from many examples that the ego ideal can prove to be intransigent
to such a degree that the ego is driven to the very brink of what it can
tolerate.
Myths, artistic creations and personal fantasies have made the
theme of the double familiar to us.83 Romantic and expressionist liter­
ature has drawn deeply on this source of 'uncanniness'. Freud pointed
out that one of the most frequent characteristics of the double is that
of being immortal.84 Here we are bound to recognise a trace of
primary narcissism which leads us to suspect that it plays a part in
these occurrences.
 Strachey noted that Freud wavered between different formulations
of the ego ideal. Sometimes the ego ideal is presented as that which
restores the perfection of the lost narcissism of childhood and, in this
case, another structure guarantees the ego functions of self-observa­
tion, vigilance and evaluation. Sometimes all this merges into a
single unity, the super-ego. Most writers accept there is a link
between narcissism and the ego ideal in order to distinguish it from
the super-ego. But it may be necessary to make a clearer distinction
between the function of censoring which is primarily the task of the
super-ego and that of keeping watch, called self-observation. That
which has the role of looking does not arise from a function analo­
gous to the visual function85 but from the detachment of a part of
the ego from the rest. If one bears in m ind that the double is
immortal, it will be seen that the ego aspires to nothing less than
complete invulnerability. As for primary narcissism it admits of no
division and the veil cast over dreamless sleep remains a mystery.
This division enables us to get a clearer idea of the most extreme
purposes of primary narcissism. There is no contradiction in thinking
of it simultaneously as the state of absolute quiescence from which
all tension is removed; the prior condition for the independence of
satisfaction; the closure of the circuit by means of which the negative
hallucination of the mother is fixed, paving the way for identifica­
tion; and the process of appropriating the ideal so as to be able to
attain the highest degree of perfection in which invulnerability is the
final aim. The stage which would necessarily follow this invulnera­
bility would undoubtedly be that of self-begetting abolishing sexual
difference.

The Phoenix, Narcissus and Death

It will come as no surprise, therefore, that in her analysis of the myths
and rites of bisexuality in ancient times Marie Delcourt86 finds a
synthesis of the raw materials Mind-Body, Sky-Earth and, in the last
analysis, immortality. The legend of the Phoenix is the most striking
example, combining effective androgynous bisexuality and eternal
rejuvenescence which is oblivious to death. In many ways the legend
of Narcissus extends and completes the legend of the Phoenix.
Our reflections on Freud's work help us to understand why, after
his brilliant introduction of narcissism in 1914, he felt obliged to
abandon it for fear of leading us along false paths - just as he felt
impelled to introduce the death drive (1921), which resulted in a
more coherent redistribution of the values of psychoanalytic theory
which he upheld until his death (1939) with ever increasing vigi­
lance. And although he was not explicit about the future of narcissism
after the final drive theory, he said enough for us to be able to develop
his reflections on the matter.
Primary narcissism cannot be understood as a state, but should be
understood as a structure. Most writers on the subject not only treat
it as a state, but also only speak of it as life narcissism, observing a
silence - the very silence that dwells within it - with regard to the
death narcissism that is present in the reduction of tensions to the
level zero. Some themes of Freudian metapsychology demonstrate the
work of the death drive in certain aspects of psychical life: aim-inhib­
ited drives, sublimation, identification and the function of the ideal.
The problem of primary narcissism cannot bypass the question of the
origin and vicissitudes of primary cathexes, of the separation of the
ego and the id, which calls for an examination of the concepts of
repression and defence. On the basis of Freudian theory, I have
defended the existence of defences prior to repression: the reversal
against the self and the reversal into its opposite, which I call the
double reversal. In developing the structure which emerges from this
study we have seen that there is an inversion of drive polarities, an
exchange of aims which culminates in the primary difference
between the mother and child, in which several registers of drives can
be distinguished: the component drives whose object is the breast,
and the aim-inhibited drives whose object is the mother, whose
destiny will be separate until the definitive object-choice has been
made. During the period of primary difference, the loss of the breast
is the equivalent in one register of what the negative hallucination of
the mother is in the other. Ego narcissism is, then, as Freud said,
secondary narcissism stolen from objects - it implies the division of
the subject following upon auto-erotism as a situation of self-suffi­
ciency. From this point of view primary narcissism is Desire for the
One, a longing for a self-sufficient and immortal totality, for which
self-begetting is the condition, death and negation of death at the
same time.