Literature, Culture, Theory

General editors

RICHARD MAC KSE Y, The Johns Hopkins University

and MICHAEL SPRINKER, State University of New York
at Stony Brook

Samuel Weber's Return to Freud is the opening title in a new Cambridge

series. Literature, Culture, Theory is dedicated to theoretical studies in the
human sciences that have literature and culture as their object of enquiry.
Acknowledging the contemporary expansion of cultural studies and the
redefinitions of literature that this has entailed, the series includes not only
original works of literary theory but also monographs and essay collections
on topics and seminal figures from the long history of theoretical
on the arts and human communication generally. The concept
ol theory embraced in the series is broad, including not only the classical
disciplines of poetics and rhetoric, but also those of aesthetics, linguistics,
psychoanalysis, semiotics, and other cognate sciences that have inflected
the systematic study of literature during the past half century.

Title published
Return to Freud: Jacques Lacan's dislocation of psychoanalysis
SAMUEL WEBER
(translated from the German by Michael Levine)

Other titles in preparation

Wordsworth, dialogics, and the practice of criticism
DON H. BIALOSTOSKY
The subject of modernity
ANTHONY). CASCARDI
Paratexts
GERARD GENETTE
(translated from the French by Jane Lewzn)
Onomatopoetics: theory of language and literature
JOSE PH GRAHAM
Twentieth-century Russian literary theory
(an anthology in two volumes)
CHRISTOPHER PIKE(ed.)
Marxist aesthetics: a history
MICHAEL SPRINKER
Kenneth Burke: a rhetoric of the subject
ROBERT WESS
 Return to Freud
Jacques Lacan's dislocation of psychoanalysis

SAMUEL WEBER
translated by
MICHAEL LEVINE

CAMBRIDGE UNIVERSITY PRESS

Cambridge
New York Port Melbourne Sydney
 Published by the Press Syndicate of the University of Cambridge
The Pitt Building, Trumpington Street, Cambridge c,iz
40 West zoth Street, New York, NY 10011, USA
to Stamford Road, Oakleigh, Melbourne 3 t66, Australia

Parts of this work were published in an earlier form as Rückkehr zu Freud: Jacques
Lacans Entstellung du (Verlag Ulistein GmbH, Berlin, 1978). An
expanded second edition of the German work has appeared under the same title
(Passagen Verlag, Vienna, I 990)

All German editions © Samuel Weber

First published in English by Cambridge University Press i991 as Return to

Freud: Jacques Lacan's dislocation of psychoanalysis

English translation © Cambridge University Press, 1991

Printed in Great Britain at the University Press, Cambridge

British Library cataloguing in publication data

Weber, Samuel
Return to Freud: Jacques Lacan's dislocation of
psychoanalysis.
i. Psychoanalysis, Lacan, Jacques, 1901 — 198 I
Title ii. (RUckkehr zu Freud) English
150. 19 5091

Library cataloguing in publication data

Weber, Samuel M.
[Rückkehr zu Freud. Englishj
Return to Freud: Jacques Lacan's dislocation of psychoanalysis /
Samuel Weber; translated by Michael Levine.
p. cm. — (Literature, Culture, Theory: i)
Translation of: Rückkehr zu Freud.
Includes bibliographical references and index.
ISBN 0-511-37410-3 (hard). — iSBN o-511-3777o-6(pbk.)
,. Lacan,Jacques, 1901—u. z. Psychoanalysis. i. Title.
ii. Series.
1W 173.W412.13 1991
150. 19'5'o91—dclo 90-44979 CII'

ISIINO 511 374103 hardback

ISBN 0 511 377706 paperback
\ I:
•1

In memory of

Eugenio Donato
This page intentionally left blank.
 Contents

Preface xi
translator's introduction xv

; Introduction
Mistaken identity: Lacan's theory of the "mirror stage" 7
The unconscious chess player 20
4 The rise and fall of the signifier 38
s Significant fallout: metonymy and metaphor 59
6 Spades and hearts: the subject as stylus 76
7 The subject as "fader": the imaginary and the symbolic 99
H "When someone speaks, it gets light": demand 120
9 The signification of the phallus 139
Appendix A Beyond anxiety: the witch's letter 152
Appendix B Transferring the heritage: psychoanalysis
and criticism i68

Index 183

Ix

The University of Iowa

LIBRARIES
This page intentionally left blank.
 ++ ++ + + ++ ++++++ ++++ +
Preface
+ + + ++ ++ + +++ + +++++++ + +++++ ++ + ++++++++ +

This book has had a rather curious history. It was initially

conceived not as a book at all, but as notes for lectures held at
the Free University of West Berlin in the early seventies. At that
time, there was very little material of or on Lacan available in
German, and consequently the lectures I held were designed to
serve as a general introduction to the Ecrits, which had been
published in French some years earlier. The most salient
characteristic of these scripts was, and doubtless remains, their
pedagogical aspect: they constitute an attempt to delineate the
contours of what at the time, at least, was — and for many, still is
— a tantalizingly enigmatic rereading of Freudian psychoanaly-
sis. As is often the case when a work emerging from one
intellectual field — in this instance, that marked by French
Structuralism — is transposed to a different linguistic and
cultural area, the contextual underpinnings tend to disappear.
This explains why so much of the book is devoted to
reconstituting the Saussurean conception of language, which at
the time was little known in German, outside of specialized
linguistic circles.
In this respect, the situation has changed since then, in
Germany no less than in the English-speaking world. And yet,
such changes are not as simple or as straightforward as one
might suppose. Freud's remarks about repression hold for these
changes as well: they do not take place once and for all, but
rather nmst be renewed constantly in order to remain effective.
What often happens, by contrast, is that what we call "proper"
names begin to circulate widely, suggesting a sort of perma-
nence or at least durability. But these names, far from rendering
what they name accessible, function as screens, isolating rather
than simply repressing, by seemingly arresting the movement
of signification. This can be a powerful means of justifying the

XI
 Preface

familiar. Indeed, if culture, as Emerson remarks, consists in the

number of things that can be taken for granted, the currency ol
such proper names is both an index of and a factor in the
formation and stabilization of culture. In this sense, "Lacan" has
become an integral part of our culture.
The notion of language that underlies the writings attached
to that name, however, moves in a very different way, and
this difference may well explain both the fascination and the
resistance it has elicited. To open the questions of naming and
predication as problematic processes of signification is to call
attention to the irreducible alterity that inhabits language and
that renders it so uncanny: all too familiar in its very strange-
ness. This sort of move can only undercut the desire for clear-cut
relations and decisions that is a particularly conspicuous feature
of North American society.
In the face of insistent efforts to foreclose such questions,
it may still be useful, even today, to retain the notion of
"Poststructuralism." If Structuralism can be described as the
rediscovery of language as a semiotic system of binary relations,
Poststructuralism marks the delimitation of that semiotics in
the wake of the system. What emerges is a movement of marks
and traits, of differences and deferrals, that can no longer be
contained or comprehended in the oppositional categories that
constitute the system, as such and in general.
The writings of Lacan, together with those of Derrida, remain,
today perhaps more than ever, two of the most powerful forces
working to keep the alterity of language from being isolated and
foreclosed. The efficacy of these texts, however, is never
indifferent to the singular idiom — and not just to the particular
language — in which they are articulated. Whence the risk, but
also the challenge of a book like this one. The difficulty in whose
shadow it is inscribed is not just that of translating a difficult text
from one language to another, but even more, of rendering a
sense of signifying movement that is irreducible to conceptual
discourse. And yet, reduced it must be by a style informed,
grosso modo, by the conventional rules and norms of academic
discourse. The ineluctability of this reduction undoubtedly
constitutes the most serious internal limitation of this kind of
undertaking.
It is not, to be sure, its only limitation. To avoid confusion,
xl'
 Preface

however, it should be kept in mind that the subject-matter of

this book, to which the name "Lacan" refers, consists primarily
ol the writings published in 1966 and entitled, simply, Ecrits. In
•.hort, the name "Lacan" in this book could be described as a
were there a "whole" of which the Ecrits could be
tonsidered a part. Although it is a matter of record that much
material was written and published under the name
"Lacan" since this text was written, the question of whether it
can be considered to form a whole remains entirely open. For
my part, I doubt that it can be assembled into anything like a
system without the most incisive and pathbreaking aspects of
I .acan's "return to Freud" being lost in the process.
Although the texts discussed in the following pages comprise
a relatively small portion of Lacan's published work, their dis-
cussion can perhaps still help readers to make their ways in(to)
the Ecrits, which remain the most densely charged laboratory
of the Lacanian experiment. Contrary to the laboratories of
experimental science, however, the walls of Lacan's laboratory
are as unstable as the margins of a text: not entirely inchoate,
to be sure, but also never completely under control. It is this
instability that distinguishes what Lacan calls "l'expérience
psychanalytique" — a phrase which also means psychoanalytical
experiment — from its scientific homonym. To make one's way
under conditions that can never be entirely controlled is part
of what constitutes psychoanalytic truth. To learn how to read
Lacan involves making one's way in this sense. The pace of
such reading is laborious, if also often exhilarating; it has little
in common with the rush to judgment that often goes by the
same name.
If this book itself is now able to contribute to such reading, it
is as a result of the efforts and dedication of Michael Levine,
who initiated and undertook the arduous and thankless task of
translation. Bruce Fink provided invaluable assistance in revis-
ing the manuscript, as did Don Eric Levine. In going over the
English text, I have made certain modifications, not so much to
the translation, as to the initial German version. Since the two
final chapters were written after the book's German publication,
they are placed in an Appendix. Both appear here for the first
time in English. The result is a considerably revised and
expanded English text. A new German edition, published by
xli'
 Preface

Passagen Verlag (Vienna, 1990), contains certain essays that

have not been translated into English.
Finally, I would like to dedicate this book to the memory of
the person who introduced me, as he did many others, to the
work of Lacan, with an unforgettable vivacity and enthusiasm:
to Eugenio Donato. His absence is keenly felt by those fortunate
enough to have studied and worked with him.

xlv
 Translator's introduction
++ ++ ++++ + ++ ++ +++++ +++++++ ++++ ++++ +

Accentuating Ent-stellung
•tmuel Weber's Return to Freud, which appears here in English
ir,inslation for the first time, has for some time had the
•.tatus in German intellectual circles of what they call a
(eheirntip, a hot piece of information circulating semi-privately
through unofficial channels. Between its short-lived distribution
by Ullstein (1978) and its re-publication by Passagen Verlag
(1990) copies of this insider's guide to Lacan have passed from
hand to hand and have been much xeroxed throughout the
lederal Republic's university towns and "alternative scenes."
No less popular than the book itself in these circles is the story
often accompanying it about its author, a young American
writing with ease and rigor in German about a notoriously
impenetrable Frenchman. While it is certainly no exaggeration
to say that Samuel Weber is someone who is very much at home
in these languages, a more accurate description might compare
him to the convalescent of Nietzsche's Zarathustra — that is, to
someone perpetually on his way home.1 Indeed, in many ways
the text which follows traces the itinerary of this "return."
As the reader will have noticed, Weber's Return is itself
already a quotation, already a repetition of another "return,"
and of another's "return to Freud." Moreover, if one reads the
subtitle of the text as a gloss on this quotation, it becomes clear
that this itinerary is not simply a return home, a return to a
proper name, or a return to the authenticity of an original, but is
instead a movement of dislocation. In order to follow this
movement we should recall how Freud instructed his followers

See Martin Heidegger, "Who is Nietzsche's Zarathustra?" in The New Nietzsche,

Cambridge: The MIT Press, 1985, pp. 64—79.

xv
 Translator's introduction

to listen a little differently, to displace accents, and to hear as

though with the ears of a foreigner, while Lacan's advice to
future analysts was to "do crossword puzzles." If Weber's text
may be said to heed the call of a "return to Freud," it also seems
to transfer that call in its own peculiar manner, for it is itself a
practical lesson in the cultivation of foreign accents and unusual
ways of listening. In short, his repetition of Lacan's return to
Freud is an act of transference and translation that only
familiarizes foreign audiences with the language of psycho-
analysis in general and with Lacan's French in particular to the
extent that it alienates the reader from his or her "own"
language.2
While the signs of this kind of defamiiarization are present
throughout Weber's text, they are most immediately and para-
digmatically evident in the subtitle of the German edition,
which describes what has been translated here as "Jacques
Lacan's dislocation of psychoanalysis" as a process of Ent-
stellung. On the simplest level Weber's strategic hyphenation of
this term has the effect of drawing attention to the signifying
matter, which, instead of making itself transparent as it conveys
a particular meaning, becomes somewhat opaque like a piece of
stained or faceted glass. Thus, in the most basic way the reader
is invited to look at rather than through the linguistic surface.
It is also significant that this invitation to attend to the
signifying surface is only silently extended; that is, a shift in
focus is effected exclusively via the strategic deployment of a
written punctuation mark. A mere change in spacing draws
attention to the spatial, graphic, and literal dimension of the
text, a gesture which also prefigures Weber's approach to the
vexed question of the relationship of speech and writing, spirit
and letter in Lacan.3 The hyphenation of Ent-stellung prompts
the reader not only to view this signifier in terms of its concrete,
visual, and graphic aspects — what might be called the
materiality of the letter — but also to take it literally, a la lettre, (or
in German buchstäblich). —

While such hyphenations have in recent years become an all

too familiar tic of critical writing, this particular "literalization"
2
See Walter Benjamin, "The task of the translator," in Illuminations, New York:
Schocken, 1969, pp. 69—8z.
See chapter 4 on "The rise and fall of the signifier."

XVI
 Translator's introduction

of a term seems to reflect the spirit of Freud's own very literal

understanding of Entstellung. Whereas this term is usually trans-
lated as "distortion," Freud comments in Moses and Monotheism
how
one could wish to give the word. . . the double meaning to which it has
a right, although it is no longer used in this sense. It should mean not
only "to change the appearance of," but also "to wrench apart," "to
put in another place." That is why in so many textual distortions we
may count on finding the suppressed and abnegated material hidden
away somewhere, though in an altered shape and torn out of its
original connection. Only it is not always easy to recognize it.4
It should come as no surprise then that such a potentially
ambiguous term should be linked to questions of repressed
conflict in Freud's writing. As Weber points out in chapter one,
Entstellung is used in The Interpretation of Dreams to designate the
general distortion of dreams. These distortions, which also
involve the distorted articulation of repressed wishes, are
brought about through the interaction of the mechanisms of the
dreamwork — condensation, displacement, considerations of
representabiity, and secondary revision — with the dream
censorship. While much remains to be said about the precise
nature of this interaction, it is sufficient at this point to notice
how from the outset the Freudian notion of Entstellung is
involved in the power struggles of (self)censorship in particular
and in psycho-linguistic conflict in general.5
What distinguishes this particular form of conflict is that it is
never simply waged by opposing forces meeting on a common
field of battle. The implications of this are twofold: first, that the
seemingly opposed forces of censorship and the mechanisms of
the dreamwork are more intimately connected (or are more
intensely at odds with themselves) than it at first appears;
secondly, that the site of such conflict is difficult to locate.
This brings us to the other more "outmoded" sense of
Lut-stellung as dis-placement or dis-location referred to by Freud
which Weber's hyphenation of the term helps to sound out. By
Sigmund Freud, Moses and Monotheism, trans. Katherine Jones, New York:
Vintage Books, 1939, p. 52.
These interactions are discussed at length in my essay "Censorship's
self-administration" in Psychoanalysis and Contemporary Thought, 9, no. 4
(1986), pp. 605—640.

XVII
 Translator's introduction

re-introducing the issue of displacement through the disarticula-

lion of a key Freudian term, Weber also seems to suggest that in
order to understand psycho-linguistic conflict and its disloca-
tions, immunity cannot be granted to the terms used to describe
it. Thus, rather than privileging the level of the signified of
Freud and Lacan's texts, which would involve reading them
"qualitatively" as positive descriptions of an objective reality,
Weber pays particular attention to the "quantitatively" defined
level of the signifier; that is to say, he treats the language of
psychoanalysis as a network of negative, differential terms (or
chains of signifiers), whose specific value can only be estab-
lished in relation to their "surroundings" and in the context of
other relevant texts.
There is a certain affinity between Weber's approach to
Lacanian terminology and Freud's way of dealing with images
in dreams, which, he says, should not be read according to their
"pictorial value" [Bilderwert: their value as images of something
elsej, but instead according to their "semiotic relationship"
[Zeichenbeziehung] — that is, like a rebus according to the relations
obtaining among the signifying elements themselves. Thus,
rather than dealing straightforwardly with Lacanian "concepts"
in terms of their positive content, Weber reads Lacan through
Saussure for whom concepts are purely differential and not
defined positively by their content but negatively by their
relations with other terms of the system. Their most precise
characteristic is in being what the others are not.
Yet in contrast to Saussure, Weber's "differential" approach
to Lacan's texts involves reading his conceptual apparatus not
only negatively, but moreover as particular avoidances, displace-
ments, and perversions of a received tradition. As he says, Lacan's
"aim is to disorient and if possible to transform the psychoan-
alytic orthodoxy."6 Thus, it is never simply a question of Lacan's
adoption or application of a particular model (even — and
especially — that of Saussure's linguistic theory), but instead of
a specific displacement, "critical appropriation," and "strategic
deployment" of such models.
As Weber suggests, Lacanian discourse can only be situated
with regard to what and whom it is polemically setting itself off
6 See chapter 7 Ofl "The subject as 'fader': the imaginary and the symbolic".

XVIII
 Translator's introduction

from at a particular point in time. Thus, the place of that

discourse, which is differentially and conflictually determined, is
always only a negative dis-place, an "Ent-stellung."
In order to delimit these "dis-placements" of Lacanian
psychoanalysis Weber is compelled by the logic of his own
argument to stage a series of encounters — which are often not so
much "face-offs" as aversions and strategically skewed encounters
— with Freud, Hegel, Saussure, Jakobson, Benveniste, Husserl,

Laplanche, and others.

Yet while the particular "dis-placements," which are (or
rather will have been) Lacanian discourse, may perhaps only be
differentially located in terms of what they are not — that is,
through a process of determinate negation — these negative
determinations are rarely as fixed and defined as they at first
seem.7 In order to stress the necessary and inevitable indeter-
?nznacy of these "dis-placements" and in order to set them off
however provisionally from the related Hegelian notion of
determinate negation [bestimmte Negation J, Weber introduces the
term "differential articulation" in his reading of Saussure and
uses it throughout the text as a means of avoiding the traditional
language of representation [Darstellung]. His plays on Ent-
stellung not only serve to accentuate this tricky veering away
from more traditional notions of Dar-stellung, but also to
link the process of "differential articulation" to other psycho-
analytic notions such as dream distortion, displacement,
and condensation. Indeed, the signifier Ent-stellung functions
as a veritable nodal point in Weber's text, the specific value
of which can only be appreciated by examining the various
signifying chains in which it is enmeshed. Some of the more
crucial links in the chain run as follows: Ent-stellung — as dis-
tortion — as displacement in the sense of differential articulation
— displacement as a particular mechanism of the dreamwork —
displaceabiity in the sense of Clbertragbarkeit: the volatility
of psychical energy in the primary process — displace as a
dis-location or negative site — in German literally an Ab-ort, a
term whose more conventional sense of toilet is linked to the
See Weber's discussion of the importance of the future perfect tense for Lacan,
its difference from the present perfect of Hegelian discourse, and its
connection to issues of belatedness in Freud in chapter z on "Mistaken
identity: Lacan's theory of the "mirror stage"

XIX
 Translator's introduction

closed restroom doors examined in Weber's reading of Lacan's

"Instance of the Letter in the Unconscious."
While Weber's text thus acknowledges and accedes to the
particular indeterminacy of "Lacan's dis-location of psychoanaly-
sis" through its pointed distprtions of standard Freudian
terminology and its strategic exploitation of linguistic ambi-
guities, this indeterminacy also seems to manifest itself as an
oscillating movement of the argument itself. A decisive first step
of this movement, which is often accompanied by a rhetoric of
"radicalization," traces a movement "beyond" or "outside" a
particular tradition (be it metaphysics or the "idées recues of
psychoanalytic orthodoxy"), while a more tentative second
step examines how what has presumably been supplanted,
displaced, or "precipitated out of the signifying chain" still
functions as a determinate absence, as a spectral signifier that
continues to haunt what supplants it.
While one might indeed link this rhythm of dis-placement to
the Freudian mechanism of repression, a term coined by Lacan
to describe the way that "Saussure genuit Jakobson" is worth
examining in this context since it adds a twilight cast to the crisis
of succession (beyond-not beyond) repeatedly enacted by
Weber's text. As though no one term could quite describe this
hesitant movement, which is neither progessive nor regressive,
neither generative nor degenerative, Lacan forms the oxymor-
onic verb genuire by extracting the "genes" from terms like
genese and generation and combining them with the noxiousness
of the verb nuire thereby begetting not so much a monstrous
product as a disturbing [gênantj mode of "production."
Rather than simply describing this crisis of succession
involving the linguists Saussure and Jakobson, Lacan's lan-
guage awkwardly "deproduces" it by enacting its own gener-
ational crisis; that is, without introducing a new, competing,
alternative mode of (linguistic) production Lacan displaces the
anthropomorphic terms in which we usually discuss "gener-
ational" issues (in the largest sense of the term). The excessive
word play and coinage of an oxymoronic neologism are at the
very least an indication of the disruptive energy which seems to
be required to shake up traditional thinking about such matters.
What erupts in this word play is not so much an ersatz linguistic
or "genetically engineered" mode of production, but rather the
xx
 Translator's introduction

conflicts which are necessarily repressed and misconstrued in

the production of an identity, in the fetishistic identity of a
production process
To understand how these deproductive or generational crises
manifest themselves (or rather silently "insist" since this
movement is precisely not one of pro-duction) as a rhythm or
cadence of Weber's text, let us briefly turn to his introduction of
the notion of differential articulation in his chapter on "The
unconscious chess player: the linguistic theory of Ferdinand
de Saussure." According to Weber,
'l'he radical difference between Saussure's theory and this [metaphysi-
call tradition is to be sought. . . in the radicalization of the idea of
difference as the principle of the linguistic sign [...] What Saussure
describes as the "two amorphous masses" of acoustical and ideational
material can only crystallize and precipitate out as signifier and
signified by virtue of their differentiation; a sound can only operate as a
signifier insofar as it is distinguishable from other sounds; a thought
can only be signified insofar as it is distinguishable from other
thoughts. Thought of in this way, signification is no longer conceived
a process of representation, but as one of articulation. (p.
This first movement thus has all the markings of a "radical"
break, a decisive step beyond, and a supplanting ("signification
is no longer conceived as a process of representation, but as one
of articulation"). Yet in a second step Weber goes on to describe
"this radicalization of differentiality" as that which "defines
Saussure's ambivalent position with respect to traditional
metaphysical approaches to language." (p. z8) This second
more hesitant pas seems to suggest that it is less a question of
supplanting or deposing a powerful tradition than of bringing
out certain contradictions and ambivalences within it.
The point is not merely that Saussure does not go far enough in
breaking decisively with tradition (although this is also the case
particularly in light of the long chapter of the Cours devoted to
the traditional subordination of writing to speech); rather, what
is most important for Weber is that the radicalization of difference
•is "articulation" will never become a new principle or a position
decisively beyond a superceded tradition of metaphysical
linguistic theory precisely because it will have always already
that is, "misconstrued" in the Lacanian sense of mEconnaissance which is
discussed in chapter z on "Mistaken identity."

xx'
 Translator's introduction

been linked — as it is here — to sources of contradiction and

"ambivalent positions." As Weber states at the opening of
chapter four
To properly understand the significance of structural linguistics for
Lacan, one cannot overlook its internal contradictions. Saussure's
writings are of interest to him less as the site where a certain strain of
modern linguistics sought to pose its foundations, than as the theater in
which the structure of language and its relation to the subject are staged
as questions. (p. 38)
That Weber has recourse to dramatic language here (and
indeed throughout the text) is no accident, for as he says in an
important footnote, "the theatrical aspect of articulation tends
to emerge whenever phonocentric conceptions of language
are no longer taken for granted." (p. 73 note) Obviously, the
theatricality of Weber's own writing goes beyond his use of
certain metaphors to include the cadences and hesitations
which scan the text. These rhythms subtly displace accents in
a way that effectively re-introduces repressed conflicts, logical
aporia, and linguistic ambiguities. They open an equivocal space
— perhaps only as wide as a hyphen (or a missing letter) in the

term Ent-stellung — in which foreign accents may and

other unfamiliar ways of listening may be cultivated. Or to
put it a little differently, the graphic dis-articulation of Ent-
stellung silently re-enacts the uncertain shift in signification
from a notion of re-presentation to a process of differential
articulation. For the reader this means that instead of being
able to narcissistically view the word as a discrete body and
unit of meaning, one is compelled to focus on the letter — that
is, on the dislocated joints, hinges, and differential spacing of
articulation.9 In short, accentuating Ent-stellung is but a way of
displacing and dislocating an accent, a way to begin to beat out
the complex rhythms of a "return to Freud."
I would like to thank Bruce Fink for his careful revisions of
the translation, the Mellon Foundation and the Center for
Advanced Study at the University of Virginia for their generous
support of this project, and my grandfather, Harry Levine, for
the endless delights of his immigrant's accent.
See chapter z on "Mistaken identity" and chapter 6 on "Spades and hearts: the
subject as stylus."

xx"
 1

Introduction

Ihe attempt to translate Lacan's French and to discuss his work

in another language might well be met with a certain amount
of skepticism. We usually expect from a theoretical discourse
transparencyof language and a clarity of concept that in
principle seems to guarantee the possibility of translation.
Particularly in the realm of theory, the verbal representation is
supposed to efface itself before that which it represents. We
assume that an articulation in one language can be translated
into another so long as the "represented" may be said to be
truthful, since truth is supposed to be the same in all languages.
if, as in recent times, we conceive of truth as being
inseparable from its verbal expression, we have little doubt that
its self-identity makes possible the kind of repetition in different
languages that we commonly call "translation."
Why then should the attempt to translate Lacan arouse
skepticism? His discourse is undoubtedly theoretical and claims
siot only to speak about truth but indeed to enact it. Neverthe-
Ii'ss, what his texts give voice to and in a certain sense "stage" is
not simply something represented, an object that would be
•.i'Ii-identical, but is itself representation, translation, staging.
I he "object" of Lacan's entire theoretical discourse, its "sense,"
a mode of articulation which might best be described as a
slip-up" or derailing of sense: the language of the unconscious,
the unconscious as language.
What distinguishes this particular linguistic form is that it /

,iever simply speaks directly (in the first or any other person)
rather misspeaks itself, concealing, denying, disavowing. Its
always involves distortion and dislocation. In this
way the unconscious forms a language of representation that is
itot constituted by what it designates, but that instead always
 Return to Freud

deconstructs the "represented," a translation without an orig-

inal or, as Freud would say, another scene.
In order to demonstrate that this understanding of the un-
conscious as a translation without an original or as a represen-
tation without a "represented" is not simply an invention of
Lacan's, but is already prescribed in the texts of Freud, it suf-
fices to recall the description of the primary process and dream-
work in The Interpretation of Dreams. The "other scene" of the
unconscious is governed by the primary process, which is
characterized by an instability of energy cathexes (under the
sway of the pleasure principle); this instability manifests itself
by interacting with the censorship in dream distortions, mak-
ing use of the mechanisms of displacement, condensation, and
considerations of representability. Freud, though using linguis-
tic concepts such as translation, transmission, and coding in
order to describe the unconscious' necessarily distorted forms
of articulation, can still be read as having never called into
question the existence of an original text. For instance, his
famous distinction between "latent" and "manifest content"
seems to presuppose just such an original:
The dream-content seems like a transcript of the dream-thoughts into
another mode of expression, whose characters and syntactic laws it is
our business to discover by comparing the original and the translation.'
A closer reading of The Interpretation of Dreams, however,
shows that the so-called original constituted by the latent dream
thoughts always takes the form of grammatically correct
sentences that as such owe their structure to the secondary
process of preconscious-consciousness. The primary process of
the unconscious appears only in the distorting mechanisms of
displacement, condensation and considerations of represent-
ability. In a long note added in 1925 to The Interpretation of
Dreams, Freud stresses the importance of the dreamwork — in
the sense of a distorting translation:
I used at one time to find it extraordinarily difficult to accustom readers
to the distinction between the manifest content of dreams and the
S. Freud. The Interpretation of Dreams, ch. 6, The Standard Edition of the Works of
Sigmund Freud, translated and edited by James Strachey, [henceforth: SE],
London: Hogarth, 5953—74, vol. iv, p. 177. Here and throughout, I have
modified the translation when necessary.

2
 Introduction

latent dream-thoughts. Again and again arguments and objections

would be brought up based upon some uninterpreted dream in the
form in which it had been retained in the memory, and the need to
interpret it would be ignored. But now that analysts at least have
become reconciled to replacing the manifest dream by the meaning
revealed by its interpretation, many of them have become guilty of
falling into another confusion which they cling to with equal obstinacy.
They seek to find the essence of dreams in their latent content and in so
doing they overlook the distinction between the latent dream-thoughts
and the dream-work. At bottom, dreams are nothing other than a
particularform of thinking, made possible by the conditions of the state
of sleep. It is the dream-work which creates that form, and it alone is the
essence of dreaming — the explanation of its peculiar nature.2
A dream is thus a form of thought and this form is constituted
by the dreamwork, not by the dream-thoughts, the latent con-
tent. For any particular dream there is a manifest and a latent
content, and yet this content is not the essence of the dream
qua content, but only its material. Furthermore, as a result of the
characteristic overdetermination of dreams, the wish realized in
it is itself enmeshed in a network of other wishes, among which
the infantile ones are the most decisive. This gives the ostensibly
original text the character of a palimpsest, superimposed upon
other texts that are both referred to and effaced by it. This
double and antithetical movement, which Freud calls, signifi-
cantly, "transference" (Llbertragung), entails a process of trans-
lation which is potentially interminable and which assumes
a relatively stable form only when the structures of early child-
hood sexuality have been stabilized through the "decline" of
the Oedipus complex.
Indeed, although at times in The Interpretation of Dreams, Freud
seems to suggest that representations have their origin in some
prior presence, upon closer scrutiny such presence reveals itself
to be a representation that in turn refers to other representa-
tions. Dream images thus should not be read simply according
to their "pictorial value" [Bildenvert], but instead according to
their "sign-relationship" [Zeichenbeziehungj; like a rebus, this
relationship treats concrete images as material that serves to
signify "a syllable or word."3
SE, V, pp. 506—507.
'Interpretation of Dreams, SE, iv, p. 278. Significantly, Freud describes the dream
here as "a pictographic script," a Bilderschnfl.

3
 Return to Freud

We thus gradually arrive at a notion of the unconscious as a

movement of translation without an original, as a process of
representation without a "represented," something that, "logi-
cally" speaking, is unthinkable — whether as a substratum or as
a substance. This logical scandal requires linguistic markers like
the curious German prefix, ver-, which is found in so many
of Freud's decisive terms ( Verdrdngung: repression, Verschiebung:
displacement, Verleugnung: disavowal, Verneinung: denial), and
which prefigures the way in which those terms slip away and
out of the grasp of traditional conceptual discourse.4 The
unconscious has no identity, "is" radically other, without being
the Other as such. Thus, when Lacan writes that "the uncon-
scious is what one says,"5 we should not forget to add: insofar
as one says something other than what one means, i.e. intends to
say. For Lacan, no less than for Freud, it is never a mere accident
when language and intention diverge: such divergence derives
from the signifying structure of language. As a signifying
medium, language is the articulation of non-identity and this is
The German prefix, ver-, stems from the Gothic, where it signifies "out,"
"before," "past," and "away from." The Latin per-, por-, pro-, as well as the
Greek pen-, par-, pro- and para- belong to the same word family. In modern
usage, the meaning of yen- points in two distinct, if interrelated directions: first,
a movement away from, a dislodging or dislocation (as in Verschiebung:
displacement); second, the execution or intensification of an operation (as in
Verdichtung: literally, "thickening," "condensation"). A word such as Verdrän-
gung combines both meanings: it signifies a certain dislodging expulsion, and
an intense impulsion (Drang, drangen). This etymological—lexical survey,
however summary, turns out to read like an abbreviated description of the
Unconscious, a word that Lacan glosses as follows: "Freud didn't find a better
one, and there's no going back on it. The disadvantage of the word is that
it is negative, which allows one to assume anything at all in the world about it,
plus everything else as well. Why not?; To that which goes unnoticed, the
word everywhere applies Just as well as nowhere. It is nonetheless a very precise
thing." ("Freud n'en a pas trouvé de meilleur, et il n'y a pas ày revenir. Ce mot
a l'inconvenient d'être negatif, cc qui permet d'y supposer n'importe quoi au
monde, sans compter Ic reste. Pourquoi pas? A chose inapercue, le nom de
'partout' convient aussi bien que de 'nulle part'. C'est pourtant chose fort
precise.") J. Lacan, Television, English translation by Denis Hollier, Rosalind
Krauss and Annette Michelson, in: October, 40 (Spring, 1987), p. 9. French:
Editions du Seuil: Paris, 1973, p. 's.
Jacques Lacan, Ecrits, Editions du Seuil: Paris, 1966, p. 830. To avoid
confusion with the French edition of the Ecrits, references to the English
translation will be to the name of the translator, Alan Sheridan. Throughout this
book, references to Lacan's writings will be first to the English translation,
wherever available, and then to the French edition, in order to
facilitate consultation of the original by the reader. As with Freud, published
translations will be modified wherever it is deemed necessary.

4
 Introduction

what allows the unconscious to be described as the discourse of

the Other.
In the pages that follow, this relation of the unconscious to
language will be elaborated; it is of decisive importance for
l,acan's entire doctrine and for the very distinctive character of
its "return to Freud." Insofar as this return entails more than
a mere paraphrase or translation of an "original," it calls for
a reconsideration of the notion of "repetition," and of its
particular relation to psychoanalysis. The processes studied by
psychoanalysis almost always involve repetition, not however
as a return of the same, in any simple sense, but rather as
he recurrence of a difference separating that which is repeated
from its repetition. An instance of this difference is provided
by Freud's description of the genealogy of the wish, in The
Interpretation of Dreams, as deriving from the hallucinatory
memory of an "experience of satisfaction": the hallucination
•.eeks to repeat the remembered experience in the "identity of a
perception," while at the same time confirming its absence, qua
hallucination. It is this distance, between repeated and repeti-
tion, that opens the space of the wish and therefore allows the
dream as such to take place.
Lacan's "return to Freud" also follows the ambivalent law of
m'petition. It draws attention to itself, not so much as a faithful
tendering of a self-identical original, but as a turn of phrase or a
trope. It is precisely the tropicality of Lacan's use of language
that bars the way to any simple presentation. Lacan's "retour" is
thus also a detour which describes the Unconscious not as an
object, but as a movement whose trajectory it retraces. This is
why Lacan's theoretical discourse is at the same time very
practical, a "signifying practice" in which the laws of significa-
tion, and above all those of the signifier, are not merely
.tbjectified or named as much as staged.
Such staging, to be sure, should be identified neither with the
language of the unconscious nor with so-called "free associa-
lion." It remains a theoretical discourse that must be held
•it countable for the consequences of its conceptuality. Such
onceptuality, however, challenges the conventional criteria of
.ncepted academic discourse. In a preface to a dissertation on
his work, Lacan asserts that "my Writings (Ecrits) are unsuitable
br theses, especially academic theses: antithetical in nature,
S
 Return to Freud

since one either takes what they formulate or one leaves them."
(puisqu'à ce qu'ils formulent, ii n'y a qu'à se prendre ou bien a les
laisser).6 The alternative is drastic, and yet anything but simple.
For to "take" to such texts is inevitably to be taken by them: to be
moved elsewhere by a practice of language in which sense is
often overtaken — surprised — by sound, just as se prendre might
easily be taken for surprendre. To take to these texts is perhaps
above all to follow the lead of such surprises, even if this means
taking on more than can be reasonably reckoned with.
For if language is a condition of reason, the games it plays are
not always reasonable. We can learn about them, therefore, only
by playing along, at least for a while. It is only then that their
sense — i.e. their direction — begins to emerge.

6
Cf. Anika Lemaire, Jacques Lacan, translated by D. Macey, London: Routledge
& Kegan Paul, 1977, p. VII; French edition: Jacques Lacan, Brussels: Pierre
Mardaga, 1970, p. 10.

6
 2
.4'

Mistaken identity: Lacan's theory

of the "mirror stage"

In "Of our antecedents," a short note written for the publication

of his Ecrits, Lacan comments in retrospect on his early writings:
We thus find ourselves replacing these texts in a future anterior: they
will have anticipated our insertion of the unconscious in language.'

Lacan uses the future anterior tense to describe not only his
own development, but also the historicity of the subject in
general, insofar as the unconscious plays a part in its con-
In his programmatic text, "Function and field of
•peech and language in psychoanalysis," Lacan writes:
What is realized in my history [i.e. in that of the individual subject] is
not the past definite of what was, since it is no more, or even the
Iresent perfect of what has been in what I am, but the future anterior of
what I shall have been for what I am in the process of becoming.2
The peculiarity of this future anterior tense, matrix for the
historicity of the subject, can perhaps be explained best by
means of a short comparison to jiegel. The (present) perfect
is undoubtedly the temporal medium of Hegelian dis-
tourse, a discourse that presents itself as a self-realization of
[Geist]. Present in this tense is a spirit or mind that
(virtually at least) has always already been perfect. Without this
present tense, absolute knowledge and philosophical certainty
"Nous nous trouvons donc replacer ces textes dans un futur antérieur: us
,iuront devancé notre insertion de l'inconscient dans le langage." "De nos
,intecédents," Ecrits, p. 7!.
'J. Lacan, The Language of the Self The Function of Language in Psychoanalysis,
trans. Anthony Wilden, Baltimore: Johns Hopkins University Press, 1968, p.
63. ("Ce qui se realise dans mon histoire, n'est pas le passé défini de ce qui füt
puisqu'il n'est plus, flu même le parfait de ce qui a ete dans ce que je suis, mais
Ic (utur antérieur de ce que j'aurai ete pour ce que je suis en train de devenir."
tents, p. 300.) Sheridan, p. 86.

7
 Return to Freud

— i.e. scientific knowledge as such — would never have been

representable in Hegel's writings. In his "Introduction" to The
Phenomenology of Spirit, Hegel criticizes a notion of thought
construed as a mere instrument, serving only to recognize
something exterior to it. He concludes:
If the Absolute were only to be brought on the whole nearer to us by
this agency [Werkzeug: instrumenti [of knowledge or (re)cognition:
Erkenntnisj, without any change being wrought in it, like a bird caught
by a limestick, it would certainly scorn a trick of that sort, if it were not
in its very nature, and did not wish to be, beside us from the start.3
The Absolute, the mind, must then according to Hegel
already be — and want to be — in and for itself with us. Without
this prior presence both philosophy as rigorous science and the
identity of the subject would be impossible.4 For how could a
subject come to know itself, dialectically realize its identity, if
this matrix of presence were not (and did not want to be) at least
virtually already there as a form to be filled or as an interiority to
be unfolded and explicated over time. The thinking subject (or
thought as subject) can only come to know itself to the extent to
which the form of this Self has always already been there as a
potential presence. The course of the Hegelian dialectic may be
infinite, but this infinity, according to the claims of the dialectic
itself, must not be a "bad" one. Rather, it should be the
self-realization of an identity that has always already been
virtually present to itself. For the self-conscious mind as it is
expressed and represented in Hegel's Logic, this in-and-for-
itself-having-been-with-us [An-und-fur-sich-bei-uns-gewesen-Sein]
of the Absolute is articulated in the (present) perfect tense. The
process of negation and determination, the movement of the
dialectic, is defined in the Logic from the very start, not simply as
the transitional movement of "passing by," but as the perfected
past presence of a passage, of an Clbergegangensein, that will have
always already taken place as the determinate negation of being
and nothing:

G.W.F. Hegel, Phenomenology of Mind, trans. Baillie, New York: Macmil]an

Co., 1961, p. 131.
See: M. Heidegger, Hegel's Concept of Experience, trans. K.R. Dove, New York:
Harper & Row, 1979.

8
 Lacan's theory of the "mirror stage"

The truth is neither being nor nothing, nor that being passes over into
nothing, nor nothing into being, but rather that each has passed over into
the of her.5

The history of the subject of metaphysics attaining its most

powerful and complete articulation in Hegel, is conceived
according to the most self-contained form of presence, that of
the present (made) perfect. The temporal structure of the subject
that Lacan's reading of Freud strives to articulate, stands in
marked contrast to this perfected present. The perfect tense is
by the future anterior, thus calling into question the
very foundations of subjective identity conceived in terms of an
interiori.zing memory. In invoking the future anterior tense,
I acan troubles the perfected closure of the always-already-
having-been [des Immer-schon-gewesen-Seinsj by inscribing it in
he inconclusive futurity of what will-always-already-have-been
I Immer-schon-gewesen-sein-wird], a "time" which can never be
intirely remembered, since it will never have fully taken place.
It is an irreducible remainder or remnant that will continually
Irevent the subject from ever becoming entirely self-identical.
In the psychoanalytical perspective, then, (men ry becomes
•.omething very different from what it was for metaphysics — not
tit'cause ofa future that the subject will never be able to catch
with fully, but because every attempt by the subject of the
unconscious to grasp its history inevitably divides that history
into a past that, far from having taken place once and for all, is
•ilways yet to come. Consequently, the living present [lebendige
.igenwart] (Husserl) of the subject emerges as a focal point
whose actuality can reside in an anticipated belatedness.
The consequences of such belatedness can be developed
through the examination of a second aspect of the future
interior, in which it is considered less as a tense, designating a
liiture past, than as a mode, entailing a conjecture. "You will
have understood Freud" (Vous aurez compris Freud. . .). Used
ulus way, modally, the "future anterior" designates a surmise, a
prediction and hence, a proposition bearing upon an
's.iertain state of affairs. This uncertainty, which cannot be
S Science of Logic, trans. A.V. Miller, New York: Humanities Press, 1969,
H; [translation altered; my emphasisl.

9
 Return to Freud

identified simply with a future or a past, typifies the language of

a subject whose self-consciousness is structured I ieiiilipf
anticipated belatedness. In accordance with the split temporality
of the future anterior, this will have been subject to and of the
unconscious.
This split, quasi-temporality affects not only the unconscious
as an object of theoretical discourse, but also, and perhaps more
significantly, the discourse itself. Thus, whereas Lacan stresses
the conceptual character of the unconscious — "The unconscious
is a concept" — the structure of this concept distinguishes it
fundamentally from that of the philosophical tradition culminat-
mg in Hegel. For Lacan, the concept is construed not as the
presentation of a representation in thought through the deter-
minate negation of its properties, but rather as the vehicle of a
search. As a concept, the unconscious is thus "forged on the
trace of what works to constitute the subject."6
The tendency of Lacan's earlier texts to use the conjectural
quasi-time of the future anterior, thereby deferring the closure
of comprehension indefinitely, suggests that the singular
movement of the unconscious requires a theoretical discourse
capable not merely of describing its trajectory, but rather of
staging its movement. Lacan's use of the future anterior does not
imply an absolute knowledge that has, is and will always have
been present to itself. Rather, it returns to the theatricality that is
so powerfully at work in Freud's writings and through which
they are clearly distinguished from those of his followers. Lacan
renews this Freudian tradition through which psychoanalytic
writing gestures towards a stage whose borders are only
provisionally determined by what is called the "reader" (or
"auditor": the difference in this case is not decisive). This
"reader" or "audience," as the provisional "representative of
the other," as Freud called serves to delimit the borders of
a stage that will always have been at a remove from the place we
occupy as self-conscious subjects. Hence, the frequent apos-
trophes that give Lacan's writings a tone that is even more
theatrical than it is pedagogical. For if discourse can be said to
set the stage, the unconscious marks the eccentricity of its
6
"L'inconscient est un concept forge sur La trace de ce qui opere pour constituer
Le sujet." "Position de I'inconscient/' Ecrits, p. 830.

10
 Lacan's theory of the "mirror stage"

snabling limits. There — is the unconscious, not as an object of

or of intuition, not as a clinical object, but as a
theatrical scene that in turn is inscribed in an ongoing scenario.
I he future anterior announces the disjunctive immediacy of this
•'thc'r scenario, to which we shall have occasion to return.
One may draw a further conclusion from the above: if the
• quasi-temporality of the future anterior characterizes
'tot only the historicity of the subject, but also its theoretical
.irticulation in Lacan's texts,7 this in turn determines Lacan's
hoice of a synchronic perspective and his rejection of the
developmental point of view that has dominated
urthodox psychoanalysis ever since Freud, and even more, ever
'.tnce Karl Abraham schematized and codified the theory of
Ithidinal stages.8 Thus if Lacan's entire theoretical project is
itncerned primarily with replacing the predominant genetic
Iloint of view with a synchronic one — his "technique of
ti'ading," Lacan writes, "is concerned simply with replacing
i.tch of [Freud's] terms in their synchrony"9 — then what counts
br him, as in case of the future anterior tense, is to replace
the temporality of the conscious subject, whose basis is the
iresent) perfect tense, with that of another, split, disjointed
It ow that would be more adequate to the movement of the
unconscious. Against a conception of development based upon
uilentity and presence, Lacan advocates a synchronic perspec-
tive that, however, has little to do with mere simultaneity.
I lowever apodictic and systematic Lacan's language some-
limes sounds, it is accessible only to a reading that does not stop
.,l the seeming stability of individual propositions, but instead
•i'u'ks out the process of enunciation, and the contradictions that
luit'I it. Like Freud's unconscious, Lacan's language seems "to
now no contradiction," which means, of course, not that there
.s'•' no contradictions but that the "law of non-contradiction"
•toes not hinder them from proliferating and determining
thought in its "truth."

I,ican characterizes psychoanalysis as a "conjectural science": cf. Ecrits,

l'P 47Z, 863.
I Karl Abraham, On Character and Libido Development, trans. Douglas Bryan
md Alix Strachey, New York: Norton, 1966.
• Il M'agit simplement de remplacerchacun de ses termes dans leur synchronie"
.rits, p. 856.

i:i
 Return to Freud

With these preliminary remarks in mind, let us now turn to

the essay where Lacan's discourse finds its initial articulation
his essay on "the mirror stage." At this point it is important to
bear in mind that any interpretation of a discourse whos
temporal medium is the future anterior must itself be caught ug
in processes of repetition difficult to master or to situate
temporally. In short, such interpretations will themselves have
been inscribed precisely in the temporal non-identity of the texi
read. They are inscribed literally as inter-pretation — as attempts
to ascertain the price (pretium) of textual gaps — of the inter.
At the same time, interpretation itself strives to reach that
future anterior, when it will have done its work, repeating
and displacing the gaps, replacing them with other, supple.
mentary gaps that will in turn elicit yet other retrospective
anticipations...
Let us then begin with Lacan's first psychoanalytic text,
"The mirror stage as former of the function of the I." The mirror
stage described by Lacan can be briefly summarized as follows:
between the ages of six and eighteen months a child displays a
reaction to its mirror-image that strikingly distinguishes it from
other creatures such as chimpanzees. A chimpanzee loses
interest in its mirror-image as soon as it recognizes it to be an
image; a child, on the contrary, displays a jubilant reaction
when it recognizes its own reflection. From this jubilant
acknowledgment of one's mirror-image, Lacan does nothing
less than to derive the constitution — and above all: the destiny -
of the ego. At this point in time, the child is not yet in control of
its body and finds itself in a state of total helplessness and
dependency. This situation is an effect of the "premature" birth
peculiar to human beings, a consequence of which is that visual
perception is much more highly developed than the motor func-
tion. A human being is thus able at a much earlier stage to perceive
the unity of an image than it is to produce this unity in its own
body. The sight of another human being, be it the mother,
caretaker, or even one's own mirror-image, becomes the matrix
of a sense of unity, identity and continuity which the child's
bodily existence is incapable of providing. Furthermore, the
identification of a similar Gestalt constitutes the exact opposite of
the turbulence and inadequate coordination that the child
experiences most immediately. The jubilant reaction of a child

Tz
 Lacan's theory of the "mirror stage"

who has recognized its mirror-image is a sign not of the

of the subject's identity but of its constitution. Later,
I .ican will formulate the role of the other as image in the
I.sllowing way: "The mediation of the Other," he asserts,
i.rnnot be considered as deriving from a second instance or as
I.i'ing of secondary importance, if the first or the one does not
i't exist."1° The ego is thus initially constituted through the
• hild's identification with an image whose otherness is precisely
•.verlooked in the observation of similarity. Despite the effort to
it, however, such alterity can never be entirely effaced,
it is what permits the identification to take place. Thus, it is
• ,nly the anticipated motor and mental unity, visually repre-
in the perception of the image and still wholly lacking in
ike child, which allows the mirror-image, so crucial for the
institution of the ego, to have its effect. The perceived image
•'Ilers a semblance of wholeness that contrasts sharply with
what the child has experienced of its own body: lack of
iiiotonc control, deficiency, and dependency. In order to
liresent this semblance of unity and exercise its power, the
image need not be a mirror-image; but the recognized relation of
'he reflecting image to the body reflected, heightens its
power, while the body's symmetrical inversion in
the mirror-image endows the latter with a supplementary
moment of alterity, thus heightening its ambivalent attraction.
According to Lacan, the mirror stage hereby locates the
,)nstitutlon of the ego in a dimension of fictionality and of
•.i•If-deception, which will have an alienating effect on the
existence and development of the subject. Or
•,ither, the "self" — insofar as it is determined by the ego — is, as
a result of alienation and deception (which also makes it
i'roblematic to retain the notion of "alienation". . .). The sense
• 'I self-identity — of unaltered and enduring homogeneity and
isnity — derives from the internalization of a relationship that is
,'Ifective only by virtue of its heterogeneity: i.e. by virtue of the
between the semblance of unity of the image and the
•Itsunity of the subject's motoric, bodily functions. The sense of
iikntity and even of reality that the subject obtains from its ego

"Discours de Jacques Lacan" (September z6, 1953), Actes du Congrès de Rome,

in: La Psychanalyse i p. 103.
 Return to Freud

thus harbors in it the irreality, deception, and non-identity that

will take on a variety of (familiar) guises, among which Lacan
mentions: phantasies of dismembered bodies (studied by
Melanie Klein), hallucinations of doubles, Hieronymus Bosch's
paintings, and Hans Beilmer's puppets. The stade du miroir is
thereby defined not primarily as a genetic moment, but rather as
a phase and as a turning-point or trope, destined to be repeated
incessantly, in accordance with a schema whose moments are
inadequacy, anticipation, and defensive armoring, and whose
result is an identity that is not so much alienated as alienating,
caught up in the "inexhaustible squaring of its own vicious circle
of ego-confirmations" ("Ia quadrature inépuisable des récole-
ments du moi").1' This vicious circle produces an
that in Lacan's view is initially the effect neither of social
conditions nor of subjective interaction; rather, its roots are
intrasubjective, deriving from a relationship of mis-recognition,
through which the ego comes to be by taking the place of the
imaginary other.
The subject is thus caught up in a future anterior — it will have
been the image whose place it takes. But in order to take place, it
must also repudiate or foreclose the alterity of that futural past.
In order to say, I am, it must deny the irreducible alterity of
the image upon which the ego depends, and instead interiorize
that relationship. A hetero-reflective relationship is thus turned
into an auto-reflective one marked by the transparency of self-
consciousness. The ego forgets, in the words of Rimbaud,
whom Lacan quotes in his essay on "Aggressivity in psycho-
analysis," that "1 is another" (Je est un autre).'2
The ego is thus determined "by a primary identification that
structures the subject as competing with itself,"13 long before
the Oedipus complex manifests itself. There are therefore good
reasons to suspect that, in view of this kind of imaginary origin,
the ego will be perenially alienated, and above all, from itself.
For that Self consists essentially in an alienation: not from itself,
but from the other.
Already this very early essay, apparently dealing with a
developmental problem — the formation of the ego — bears the

Ecrits, p. 97; Sheridan, p. 4. '2Ecrits, p. 1i8; Sheridan, p.

'3Ecrits, j1. 117; Sheridan, p. ii.
14
 Lacan's theory of the "mirror stage"

unprint of a structuralist approach that only much later will be

illed by that name. The narcissistic structure of the ego and the
tensions that derive from it — which Lacan will later
to the Freudian notions of primary masochism and the
k'ath drive — are as difficult to overcome as the ego itself.
Insofar as the subject conceives of its history as that of its ego —
tisofar that is, as it identifies itself with the ego — there can be no
from the aforementioned vicious circle of self-affirmations.
I or Lacan this vicious circle attains its fullest expression in
..irtre's existential philosophy. Given his postulate of subjective
iii tonomy, Sartre's philosophy is for Lacan not only a "jeu de
I t'sprit" but a "je de l'espnt", a spiritual ego that traditional
pihilosophy has never seriously called into question insofar as it
tended to equate subjectivity with self-consciousness.
onsidered in this way, this ego, Lacan argues, is identified
with that part of the subject that renders a true
historicity of the individual impossible. This point should be
I'orne in mind when considering not only Lacan's theory, but
iPso what has often — all too indiscriminately — been called the
• 'structuralist attack on history." It is not history as such that is
,iticized by structuralism, or by Lacan, but the metaphysical
inception that construes history in terms of a self-identical
'ubject of self-consciousness.
In these early writings, then, Lacan thus challenges the
onception of history as the self-production and self-realization
ot the subject. Instead, he points to the ego's inertia, which
irises from a constitutive identification that leaves it unable to
I'wak out of the orbit and spell of its aggressive, narcissistic
origins. Not yet able to define structurally an alternative
IMjectory for the ego's inherent altenty, Lacan can only insist on
uIie non-coincidence of ego and subject. And yet, even at this
p'oint in time, it is already clear that this non-coincidence
originates in a splitting of the ego that is primary, and not
iswrely external or accidental. In apparent opposition to Freud,
who identifies the ego at first with the system perception-
onsciousness, and later with reality testing within the psyche,
I .ican emphasizes the ego's defensive functions, such as
misapprehension [Verkennung] and denial. As a response to a
immndamental predicament of the human being, what is most
• of the ego is the function not of reality testing, but

'5
 Return to Freud

rather of what Lacan calls "derealization," in accordance with

the fictive nature of the ego. Lacan elucidates this fictive
dimension by reference to a work of fiction: the myth of
Narcissus.
In his essay on the "mirror-phase," Lacan calls attention to
the perspicuity of psychoanalytic theory for having been
sensitive to the "semantic latencies" that resonate in the term
"primary narcissism." The recourse to the myth of Narcissus
entails two moments which, Lacan argues, are all too often
forgotten in psychoanalysis. First, the fact that Narcissus, in the
myth, is not merely in love with himself, but rather with an
image of himself. And second, the fact that Narcissus' fascina-
tion with this image leads him to take his own life. Lacan links
the suicidal aspect of the Narcissus myth to the process of
identification and to its aggressive consequences, as already
mentioned. More important, however, is the function of the
image, which defines the imaginary realm as such. In philo-
sophy, as well as in everyday usage, an image is generally
construed as a representation that reproduces, with diminished
intensity, a predetermined, structurally and temporally prior
sensation, impression or idea: a subject places an object before
its mind's eye, it imagines something; a picture depicts. Lacan
slightly but decisively displaces this notion of the image: it
de-picts not so much by reproducing or representing an object,
as by taking it apart, dismantling it. Nothing can be said to stand
bejbre the image — its model, for instance — that does not in effect
come after it, just as the ego comes after the mirror-image and
depends upon it. This belated arrival of the "model" inscribes it
in a chain of doubles, as the "splitting" image of its ostensible
original.
Something similar to this is already intimated in Freud's
discussion of narcissistic identification, insofar as the latter does
not merely express a subject, but also re-acts upon it and
transforms it. For Lacan, however, the effect of the image
becomes, as we have seen, constitutive of the ego itself. To be
consistent with this reinterpretation of the image's constitutive
power, its origins can no longer be located in a reality held to be
temporally and structurally prior to it, since this reality, at least
insofar as the ego is concerned, is itself only an after-effect of the
image. That this shift in the conception of the image and of
x6
 Lacan's theory of the "mirror stage"

Ilowers should raise a number of logical difficulties should

•mc as no surprise, considering the fact that it is the task of
logic to determine an order in which first things
me first: the model before the copy, the repeated before the
1'etition, identity before difference. This order of precedence is
h.tllenged by Lacan's notion of the semblable, meaning the like
the similar. When Lacan claims that the condition of
identification is the reconnaissance du semblable:
ognition, reacknowledgement of the like, one might ask
'w the imago can be re-acknowledged as being similar,
ithout identity — like Hegel's Absolute — being necessarily
Distinct from this Hegelian position, which places
Isfltity first, ego-identity for Lacan is produced first by the
by the representation.
We thereby find ourselves before the very problem already
in connection with the future anterior: "is" (the
.eMence of the present tense) should be understood only as an
past," which has yet to arrive. The paradox can
.t'dly be resolved according to the familiar rules of logic, and
is perhaps due to the fact that the subject implied by the
I ,i'udian unconscious cannot be conceived according to the
logic of metaphysics. Whether this must lead to a
• s,unciation or revision of Freudian doctrine, or instead to a
thinking of the scope of traditional logic, must be regarded as
ti open question. Nevertheless, by virtue of what is illogical in
view we can understand why ego-recognition — whether as
objective or subjective genitive, or as both — must for Lacan
and of necessity remain a mis-recognition, a miscon-
uing implied in the reiterative structure of (re)cognition as
'u h, a misconstruction and ultimately: a case of mistaken
Ii'ntity. For Lacan, the "coming into its own" of the thinking
'prit, of the subject as and as self-consciousness, always
"coming to another": an other, and never its other, in the
sense of Hegelian determinate negation. In the
• .Inge light of Lacan's mirror-stage, there is the mirror, and
image; first the image, and then that which it depicts. And
we seek to be rigorous, we cannot even assert that the mirror
that is, is present, as an entity, a being, but rather only that
h,,t there is the mirror, and that there, on that other stage, the
superimposes itself upon its mirror-image, forever to be
17
 Return to Freud

haunted by that reflection, beyond the wildest stretch of the

imagination.'4
The imaginary can be defined as that fictive, illusory realm of
the mirror-image, of the optical illusion, of the image which can
no longer hope to be an accurate and faithful representation of
its model, since the latter is fashioned in its image: I am the
image of that image, a mirror without end.
Beyond the looking glass, there is nothing. Does that mean
that there is nothing else? The mirror-stage, Lacan remarks, is
"the moment that decisively tilts all of human knowledge into
mediatisation through desire of the other."5 With this remark,
desire enters the (narcissistic) picture. Before, the discussion of
the mirror stage dealt only with the physiological prematuration
of the human organism and with the formatory effects of images
in the maturation of pigeons and migratory locust. Although in
this essay desire is only mentioned in passing, as it were, it
indicates the way that Lacan's future thought will take, a way
that will lead away from physiology and which wifi allow the
mirror stage to appear in a very different light: no longer as
the expression of an organic deficiency, but as the response
to the linguistic, "symbolic" nature of human being. Looking
back years later on the mirror stage, in "Of Our Antecedents,"
(1966), Lacan remarks that it defines the rule by which the
imaginary differs from the symbolic "at this moment of capture
by an historical inertia. To be sure, the rules of this
differentiation, in which the symbolic dimension of desire is
articulated, will only begin to be worked out from 1953 on,
some fifteen years after the first formulation of "The Mirror
Stage," which was initially presented in 1936 and then
published for the first time in 1949. In the intervening years,
much has happened, but for Lacan nothing will have been of
greater importance than his encounter with the linguistic
14 on the problem of the "there is," cf Martin Heidegger, "On time and being,"
trans. Joan Stambaugh, New York: Harper & Row, pp. 5ff. The relation
of Lacan to Heidegger should be explored less in the manifest adoption of
philosophemes, than in the ways the linguistic and rhetorical practices of both
authors tend to jar established theoretical discourse into motion.
15 Ecrits, p. 98; Sheridan, p. 5.
16 "Le stade du miroir donne Ia regle de partage entre l'imaginaire et le
symbolique a ce moment de capture par une inertie historique Ecrits,
p.

i8
 Lacan's theory of the "mirror stage"

of Ferdinand de Saussure, the founder of structural

I

iiguistics. It is this encounter that turns the stade de miroir from

•tadium (to which Lacan compares it), to a staging area where
's and figures of a very different kind are preparing to make
Ilsir appearance.

'9
 3

The unconscious chess player

The name, Ferdinand de Saussure, is generally associated with

two insights: first, a semiotics based on the principle of the
"arbitrary" or "unmotivated" nature of the sign, and a division
of language into language-system, langue, and speech, parole.
Were this all that Saussure had to say, however, it would be little
enough. The conception of the arbitrary character of the sign is
at least as old as Plato's Cratylus, and the separation of language
into system and speech is hardly any less ancient. If Saussure
was able to exercise a certain influence upon French Structural-
ism, and through it, upon its "post-structuralist" transforma-
tions, it was only insofar as his rethinking of language diverged,
in decisive ways, from the more familiar conceptions that his
terminology, it is true, often suggests. Let us therefore, as a first
step in defining this divergence, briefly recapitulate certain main
traits of what might be called the dominant metaphysical theory
of language, as explicated by the thinker who first laid the
groundwork of that theory: Aristotle.
According to Pierre Aubenque, Aristotle's entire philosophi-
cal project, far from being simply a critique of Platonism, should
be seen as an effort to repair a fatal flaw in Plato's thought, one
which rendered the truth-claims of philosophy vulnerable to the
skeptical challenge of the Sophists.1 At the end of the Cratylus,
Socrates reminds his interlocutor that "knowledge of things
cannot be derived from their names, but only from the things
themselves which must be studied and examined independent-
ly." (439d) Yet, Plato never explains how a thought process that
is only accessible through the medium of language, would be
able to examine things "independently" of their "names," nor
what the role of language would thereby be.
Pierre Aubenque, Le problèine de l'êfre chez Aristote, Paris, 1964; see especially
the chapter entitled "Etre et langage" (Being and language), pp.

20
 The unconscious chess player

The Sophists, by contrast, had made language the focus of

their thinking and had elaborated two seemingly contradictory
accounts of its relation to reality, both of which worked
effectively to undermine philosophy's claim to truth. The first,
lormulated by Gorgias, held that it is impossible for language to
correspond adequately to extra-linguistic things, since language
itself is a thing among things — namely a phonetic entity that, in
its specificity, is unable to render entities different from itself.
Language, in short, is one thing among others, which, by virtue
of their otherness, cannot be rendered by it. While language,
according to this view, can serve as a means to intersubjective
ends, its peculiarity excludes it from being a medium of
objective knowledge.
The second, contrasting position, associated with Antis-
thenes, a disciple of Gorgias, has the following formulation:
MI discourse has an element of truth, for whosoever speaks, says
and to say something means to say some entity, and
whosoever says an entity, says something true.2
I hese two apparently contradictory views of language share a
ommon basis: both conceive of language as something self-
be it as one natural thing among others, be it as an
entity that partakes of Being. Most importantly, both radically
.1ticstion the idea of objective truth, since according to Gorgias,
is separated from other things by virtue of its
.Wference from them, whereas for Antisthenes, language, by
'Iways referring to something beyond its own immediate being,
possesses an element of truth (but only an element). The
of these arguments is that truth itself becomes in-
difficult to distinguish from non-truth.
Aristotle's linguistic theory seeks to counter such Sophistic
by arguing that language is inherently neither true
.i false, but rather a process of signification. In his treatise "On
Aristotle states that
sounds are symbols of affections in the soul, and written marks
of spoken sounds. And just as written marks are not the same
ill men, neither are spoken sounds. But what these are in the first
signs of — affections of the soul — are the same for all; and what

led by Aubenque, p. zoo.

2.1
 Return to Freud

these are likenesses of — actual things — are also the same. (Dc
Interpretatione, i 6a)
Aristotle thus sets up two relationships: that of resemblance, a
natural correspondence of things and mental experiences, and
that of signification or symbolization, which is not related in any
natural way to what is signified; this relation is first produced by
convention. One can already see here the division into referent
(the things themselves), the ideational content or the "desig-
nated" (the mental experiences), and the designating function
(language). Linguistic symbols are arbitrary, since they bear no
natural resemblance to that which they designate; they neces-
sarily indicate an ideational content, which in turn is a faithful
reproduction of a state of Being of an entity. Error is then
possible insofar as the designated things — here the mental
experiences — are infinite, whereas the number of available
linguistic media — the linguistic signs — is limited. A situation
thus arises which we would call ambiguity and which Aristotle
calls "homonymy."
It is not possible here to follow Aubenque in his fascinating
account of Aristotle's struggle to construct a theory of language
and of argumentation which effectively answers the Sophists'
challenge to the truth-claims of philosophy. The interested
reader can consult the study of Pierre Aubenque. Instead, I
want to cite a passage in which Aristotle polemicizes against the
Sophists, since it relates to our topic in a number of important
ways. Towards the end of the fourth book of the Metaphysics,
Aristotle attempts to demonstrate the untenable nature of the
Sophistic position by means of the following reductio ad
absurdum:
But if all are alike both right and wrong, one who believes this can
neither speak nor say anything intelligible; for he says at the same
time both "yes" and "no". And if he makes no judgment but thinks
and does not think, indifferently, what difference will there be between
him and the plants? — Thus, then, it is in the highest degree evident that
neither any one of those who maintain this view nor any one else is
really in this position. For why does a man walk to Megara and not stay
at home, thinking he ought to walk? Why does he not walk early some

The Complete Works of Aristotle, edited by Jonathan Barnes, Princeton

University Press: Princeton, NJ, 5984, vol. I, p. 2.5.

ii
 The unconscious chess player

morning into a well or over a precipice, if one happens to be in his way?

Why do we observe him guarding against this, evidently because not
thinking that falling in is alike good and not good? Evidently, he judges
.'ne thing to be better and another worse. And if this is so, he must
tudge one thing to be man and another to be not-man, one thing to be
.weet and another to be not-sweet. For he does not aim at and judge all
things alike, when, thinking it desirable to drink water or to see a man,
ht proceeds to aim at these things; yet he ought, if the same thing were
man and not-man. But, as was said, there is no one who does not
.hviously avoid some things and not others. Therefore, as it seems, all
men make unqualified judgments, if not about all things, still about
what is better and worse. And if this is not knowledge but opinion,
they should be all the more anxious about the truth, as a sick man
hould be more anxious about his health than one who is healthy; for
in who has opinions is, in comparison with the man who knows, not in
healthy state as far as the truth is concerned. (ioo8b)4
lo begin with, it should be noted that, in order to demonstrate
'lie legitimacy of truth-claims in language, the philosopher finds
himself forced to resort to an ad hominem argument, a pragmatic-
ii move that might rather be expected of a Sophist. One is
teininded of the kind, and above all of the tone, of the arguments
.iuployed by Callicles, in Plato's Gorgias, in which Socrates'
itming death is made the occasion to deprecate those phio-
ophers who are unable or unwilling to avert approaching
l.i nger. Aristotle's pragmatics in this polemic nevertheless
here to be a necessary result of his theory of language,
it is precisely the priority of the pragmata — of the things
— over the language which symbolizes them, that

seeks to defend at all costs against the Sophists.

\. cording to Aristotle, the referents of language (the things
'lnrnselves) and their signifieds (the mental experiences) are
i.1h present; they exist prior to all language, although human
can only have cognitive access to them through lan-
nige. Aristotle can only demonstrate the priority of the
by referring to factual, pragmatic human behavior.
P human beings thus differ from plants insofar as they choose to
1i onething rather than another; for example, if someone
inuoses first thing in the morning not to throw himself or
iself right into a well or a ravine, this shows that things are

Book iv, chapter 4, Complete Works of Aristotle, vol. z, p. 1591.

23
 Return to Freud

not indifferent and of equal importance and that such differ-

ences are also commonly presupposed as known, and as such
form the basis of decisions, preferences and of action: one
drinks water when one is thirsty and above all protects oneself
from certain things rather than from others. Even anxiety,
insofar as it determines human behavior, indicates, according to
this argument, that a human being is not indifferent to
everything, and perhaps above all, that one cannot be at once
human and non-human. The reality principle and the necessity
of self-preservation thus become arguments in favor of the real
and knowable existence of objective essences.
It is precisely the existence of such beings that determines
language in its function as symbol — that is, as a substitute for
things, or rather, for their Being. Language is conceived of by
Aristotle — and by much of the metaphysical tradition that will
follow — as representation, and more precisely, representation
in the sense of a substitute, proxy, deputy, or stand-in. Its
immediate reality resides in the absence of the thing it signifies,
which in turn is construed as already self-identical, prior to all
symbolization. Such self-identical things can thus be called to
mind as such, without mediation, whereas in language they can
be present only in mediate form, qua representation. Language
accordingly has no identity of its own, if it is not that of a
medium of representation: it is not a full-fledged entity, in the
Gorgian sense, for it only is as signification, as the absence of
that which it signifies or represents. Language construed as
representation becomes a place of irreducible difference, de-
fined in terms of the identity and presence of the things it
signifies. Language exists only in view of its meaning or sense,
and this sense is in turn conceived as being prelinguistic: it
proceeds from the speaker's intention and abides in the mind.
Thus Aristotle's rebuttal of Sophistic argumentation must
ultimately have recourse to the speaker's intention, for though
one can expose contradictions and paralogisms in an argument,
and one can trace their necessity to the poverty and dependence
of the linguistic medium, precisely because of this poverty one
cannot convincingly demonstrate the principle of contradiction
in purely intralinguistic terms. The presence of the things
themselves can at best only be inferred from the day to day
behavior of men, as we have just seen.
24
 The unconscious chess player

Language, determined as representation, is thus considered

by Aristotle from the vantage-point of its meaning and refe-
rents. What thereby remains to be seen is just how something
capable of functioning as a symbol: how language is able to
'signify. It is precisely this question which defines the point of
departure of Saussure's reflections upon language. Like Aristo-
tle, he is concerned with the conditions of possibility of scientific
knowledge, but unlike most of his predecessors, his ultimate
is to lay the groundwork for a rigorous science of language.
Metaphysically speaking, this project is not without its contra-
dictions: on the one hand it is informed by a notion of episteme,
ssl scientific knowledge; on the other hand, the object of this
knowledge is not a domain constituted by meaningful objects or
but a medium characterized by non-identity and
.ibsence. It is this contradiction, and the manner that Saussure
it, that constitute the originality of his conception of
lrnguage, and it is to this contradiction that I now turn.
The first requirement of an autonomous science of language
the construction or delineation of a self-contained and
object of study. Guided by this principle, Saus-
.iire divides the field of language into three spheres: langage,
nigue, and parole. Le langage, language in the largest sense,
the ability to constitute and to exercise language in
Yet precisely because of this generality, language in this
%..'idest sense can never serve as the object of study of a single
tence: "Straddling various domains. . . it belongs to none of
it cannot be classified within any particular category of
human relations for one does not know how to determine its
At the other end of the spectrum is speech (parole),
liich, like langage, does not possess a self-contained structure,
entails rather the empirical realization and actualization of
I
unguage by individual speakers. Thus, in order to investigate
j'i'ech, it is necessary to study the conditions of individual
I 5k Saussure, Cours de linguist ique gEnérale, Paris: 1964, p. zs. Course in
neral Linguistics, trans. by Wade Baskin, London: Fontana/Collins, 1974,
Future references to this work will be given in the body of the text, first to
French edition and then to the Baskin translation. References to this
.un.lation are given for purposes of comparison: the English translations used
are largely my own. The reader may also wish to consult the recent
I translation of Saussure by Roy Harris, Lasalle: Open Court Press,

25
 Return to Freud

existence from different points of view, including those of

psychology, physiology and physics. An independent science of
speech is therefore no less difficult to conceive than a science
of language as langage. Only language considered as a self-
contained system of signs, which Saussure designates as la
langue — can provide the basis upon which an autonomous
science of linguistics can be constructed. Although language
considered as a sign-system forms part of langage, it is only by
virtue of the closure of that semiotic system that linguistic
phenomena can be said to have any degree of consistency. For
Saussure, it is only "the language-system (la langue) that
constitutes the unity of language (du langage)." (27/I i) Thus,
although la langue is actualized in speech and has no existence
independently of it, it still retains a structural priority, relating
to speech as does the collective to the individual. As a sign-
system, Saussure argues, language forms part of a field that is
destined to become the privileged domain of a future science of
semiology; although linguistic signs form only a relatively small
part of this domain, they provide a privileged access to
understanding the nature of the semiological problem.
Considered as a closed system, language is determined by the
structure and function of the sign. And it is precisely in his
theory of the sign that Saussure's paradoxical and original
position in relation to metaphysics emerges most forcefully. His
discussion of the "Nature of the linguistic sign" begins by
the notion of namin&as a relevant category in analyzing
the operation of language. Compared to many of his predeces-
sors, this move of Saussure's does not at first seem to be very
radical: "The linguistic sign does not unite a thing and a name,
but rather a concept and an acoustic image." Aristotle,
we recall, had in a similar way already differentiated between
referent, signified (i.e. psychic state) and verbal sign (spoken or
written). Moreover, as has frequently been observed, Saussure's
introduction of the terminological distinction between signifier
and signified — "We propose to retain the word sign to designate
the whole, and to replace concept and acoustical image
respectively by 'signified' and 'signifier" (99/67) — continues a
venerable tradition going back to the Stoical distinction between
signans and signatum. The fact that Saussure thus distinguishes

z6
 The unconscious chess player

sign from name, that he construes the former as constituted by

the opposition of signifier and signified, and finally, that he
conceives this constitutive relationship to be "arbitrary" — all of
this can be inscribed easily in the metaphysical concept of the
sign. The radical difference between Saussure's theory and this
tradition is to be sought elsewhere.
It resides not in the theory of the arbitrary structure of the
sign, but rather, as Derrida has argued in Of Grammatology, in
the radicalization of the idea of difference as the principle of the
linguistic sign. For, as we have seen in Aristotle, the notion of
semiotic arbitrariness is entirely compatible with a conception of
language as the representation of entities existing independent-
ly of and prior to all such representation. In Saussure, however,
the primary distinction is neither that of representation and
referent, nor that of signifier and signified. Rather, it is that of
difference as the principle upon which the function of the signi-
fier as well as that of the signified is "founded." What Saussure
describes as the "two amorphous masses" of acoustical and
ideational material, can only crystallize and precipitate out as
signifier and signified by virtue of their differentiation: a sound
only operate as a signifier insofar as it is distinguishable
from other sounds; a thought can only be signified insofar as it is
distinguishable from other thoughts. Thought of in this way,
signification is no longer conceived of as a process of representa-
tion, but as one of articulation. Instead of language being
tonsidered from the vantage-point of a hierarchically and
temporally prior presence as its point of departure, it is
tonstrued as an articulation, determined and defined by a
difference that produces identities only belatedly and retroac-
tively: as concrete and individual signifiers and signifieds. Thus,
br Saussure, the "concrete entities" of the language-system are
iiot the perceptible, substantial unities of traditional grammar;
they are "values," whose structure is relational or
.yntactical, not grammatical:
In all these cases [i.e. in the analyses of grammatical and lexical
• .itegories] we are thus surprised that instead of pre-given ideas [the
grammatical categories] we encounter values emanating from the
'ystem. When they are said to correspond to concepts, it is implicitly
understood that the concepts are purely differential, defined not
2.7
 Return to Freud

positively, by their content, but negatively, by their relations with other

terms of the system. Their most precise characteristic resides in being
what the others are not. (162/117)
This radicalization of differentiality as the structuring princi-
ple of linguistic signs defines Saussure's ambivalent position
with respect to traditional metaphysical approaches to lan-
guage. The ambivalence is perhaps most evident in the manner
that Saussure conceives the relationship of the signifier to the
signified. The relationship can be described as "arbitrary" or
"unmotivated" only insofar as there is no necessary, "natural"
or perceptible connection determining the link between signifier
and signified in any particular sign. Yet Saussure also emphasizes
that the sign as such forms a unity, and that in general, a signifier
without a signified is unthinkable. Indeed, the function of the
language-system, he states, is "to maintain the parallelism
between the two orders of differences." (x 67/I ix) This amounts
to asserting that the function of language consists precisely in
stabilizing those differences. Such stability is required if linguis-
tics is to constitute itself as a rigorous science, at least as science
has traditionally been understood. This leads Saussure to
modify and moderate the principle that "in the language system
there are only differences" so that it becomes compatible with
the stability required of language qua scientific object:
But the statement that everything in language is negative is true only if
the signified and the signifier are considered separately: as soon as
we consider the sign in its totality, we are confronted by something
positive in the realm of language [dans son ordrej. A linguistic system is a
series of differences of sounds combined with a series of differences of
ideas; but the pairing [mise en regard] of a certain number of acoustical
signs with as many segments [decoupuresl made in the mass of thought
engenders a system of values; and it is this system that constitutes the
effective link between the phonic and psychic elements within each
sign. Although both the signified and signifier are purely differential
and negative when considered separately, their combination is a
positive fact; it is even the sole type of fact that obtains in language,
/
since the property peculiar [le propre] to the linguistic institution is
to maintain the parallelism between the two orders of differences.
(166—167/20—121)

Language is, as it were, called to order, invested with positivity,

its unruly differential principle subordinated to the category of

z8
 The unconscious chess player

binary opposition, which, as it were, contains and stabilizes the

but necessary connection between signifier and sig-
nified. Language is thereby rendered an appropriate object of
linguistic science: well-defined and self-contained. Since the
determining linguistic relationship is now the binary opposi-
tion, the significance of syntax is rejected: according to Saussure
syntactic structures play no role in language apart from "the
concrete material unities" constituted through the a priori
principle of binary opposition.
It would be a mistake to believe that there is an incorporeal syntax
outside of these material unities distributed in space [. . .1 in reality,
the material unities, arranged in a certain order, alone create [syntactic]
value. Outside of a sum of concrete terms, a case of syntax would be
incomprehensible. (191/139)
Language is now — and in terms of the metaphysical tradition,
once again — considered as an established institution, maintaining
itself, i.e. its "peculiar properties." These reside in the dual,
parallel "order" of signifier and signified, combining in stable
oppositions to form total, positive signs. Syntax thus is seen as
nothing more than as a function of these parallel, synchron-
ized, linear dimensions, segmented into series of individual
signs. The linear arrangement of the linguistic sign is largely
taken for granted, and is indeed justified in terms of an ideal of
understanding and communication, which Saussure presup-
poses to be a simple but decisive fact of language:
Besides, the very fact that we understand a linguistic complex
• . . shows that this sequence of terms is the adequate expression of the
thought. (191/139)
insofar as language is held to be essentially a means of
Thus,
communication and of understanding, linguistic difference is
ultimately brought under the sway of a theory of representation:
the function of the sign is to serve as the adequate expression of
thought.
Nevertheless, this line of reasoning, informed by the neces-
sities of determining language as a stable system of signs,
contradicts the thrust of Saussure's insight that it is difference,
not binary opposition, that keeps the process
moving. That the former is in no way reducible to the latter in
the functioning of language is evident throughout Saussure's

29
 Return to Freud

lectures, as for instance, in the description of linguistic value as

determined solely "by what is outside and surrounding it."
(i6i/x i6) The difference that constitutes value is identified not
with a fully determinable relationship of opposition, but in-
stead with an externality or a surrounding. The examples that
Saussure provides in order to illustrate the operation of
difference, for example, the difference between the letters of
the alphabet required in order that handwriting be legible, or
the morphological, etymological, acoustical relations between
words, clearly indicate that signification depends upon a play of
differences that are in no way reducible to binary oppositions.
Perhaps more important still is the fact that difference only
seems to function symmetrically in the realm of signifieds and
signifiers. However, since it is only through difference that
these two realms are determined as such, the signifier leaves its
imprint, "as signifier," upon the realm of the signified. In order
for the "undifferentiated, amorphous mass of thought" (i 55/
iii) to be articulated in and as particular thoughts — that is, in
order for it to become thought at all — its constituent elements
must first set themselves apart from each other, demarcate and
distinguish themselves; in short, they must relate to one another
gs in order to be determinable as The
Saussurian bar, stroke, or slash that is supposed to separate
signifier from signified, cuts through the signified, dividing
each and every signified from itself, and allowing its identity
qua signified to appear only as an effect of this separation.
Meaning thus emerges as a function of signification. Thus,
when Saussure observes that "there are no pre-established ideas
and [that] nothing is distinct before the apparition of language"6
he is not merely referring to the connection of two
separately constituted, structurally symmetrical spheres, but
rather to the indispensable operation of difference working
through signifiers, without which identities could not be
determined, whether as "signifieds" or as "signifiers."
Two fundamental and contradictory consequences may be
drawn from the above: first, the "concrete," "positive" entity of
language — the sign as a binary opposition of signifier and
signified — appears as the effect of one of its parts: of the
6
See Samuel Weber, "The apparition of language," MLN, vol. 9 i/no. 5 (1976).

30
 The unconscious chess player

signifier, insofar as it materializes and realizes the operation of

difference. Secondly, it is precisely the materiality and realiza-
tion of the signifier that constantly elude rigorous definition
or determination. This is reflected in Saussure's contradictory
attitude toward sound as the medium and material of language.
On the one hand, Saussure is obliged by the principle of
difference as articulation to consider every materialization of
language — thus also sound — as a mere vehicle or support of
difference. "Thus," Saussure remarks, in attempting to define
language as the object of linguistics, "language cannot be
reduced to sound" (24/8), and "the essence of language [. . .] is
unrelated to the phonic character of the linguistic sign." (21/7)
At the same time, however, Saussure writes a long chapter in
which the sign-system of writing is subordinated to phonetic
language, repeatedly asserting that the signifier is "of a phonetic
nature," and that the "natural connection," between signifier
and signified, "the only true one, is that of sound." (46/25) Why
then this striking contradiction in Saussure?
Derrida has shown in Of Grammatology7 how closely Saus-
sure's privileging of phonetic language and his reduction and
subordination of writing as phonetic writing continues the logo-
and phono-centric tradition of Western metaphysics. But, what
motivates this repetition and the contradictions implied by it?
Whence does it derive its force? The response to these questions
— which also play a determining role in Lacan's use of structural
linguistics — is related first of all to Saussure's attempt to
establish linguistics as a rigorous science; and secondly, to the
wish that this attempt strives to fulfill.
As we have already seen, in order for language to be
considered as an object of scientific study, it must be auton-
omous, homogenous, concrete, and self-contained. Neverthe-
less, it is precisely difference that in a curious way is not
autonomous: neither a transcendental a priori nor a founding
principle, it exists only in and through the play of articulation;
still, it is never fully present in this play. It is impossible to
reduce or determine, in terms of a binary opposition, since
Jacques Derrida, Of Grammatology, trans. Gayatri Chakravorty Spivak,
Baltimore: The Johns Hopkins University Press, 196?; see especially the
chapter, "Linguistics and grammatology." The following discussion of
Saussure is deeply indebted to Derrida's reading.

31
 Return to Freud

differential articulation is always both more and less than any

given "concrete" pair of opposites. Or to put it another way:
though language as difference and articulation produces repre-
sentation as an effect, this effect is inscribed in a network of
differences that only retroactively produces the possibility of
presence — of meaning, of an object or of a subject. In Derrida's
apt formulation, "the signified must always already have been
in the position of the signifier,"8 always already differing and
deferred, always in relation to something else, in order to be
itself. To be able to construe language as a closed system,
however, Saussure has to disavow just this implication of
difference; he must depict it as a means of representation and
"concretize" it. Thus Saussure finds himself forced — in his
chapter on "The concrete unity of language" — to repeat art
Aristotelian gesture and declare meaning to be the criterion of
language: "A sequence of sounds is linguistic only if it functions
as the vehicle of an idea; taken by itself it is only the material of
physiological study." (144/103) Here one can see how Saus-
sure's argument restricts difference in language by treating the
latter as a medium of expression: since a sound as such cannot
be linguistic, it must be determined through meaning; yet
meaning taken by itself is just as indeterminate as sound, it only
becomes meaning — as signified — through the play of difference.
This play, however, produces identities only retroactively and
in a certain context that must of necessity resist closure. To serve
as a means of communication or of understanding is un-
doubtedly one of the functions of language, but not necessarily
its most essential one. In any event, the conception of language
as communication entails certain presuppositions concerning
the subject of language, or rather the subject of linguistics, and
these demand attention.
Saussure's contradictory attitude toward the linguistic func-
tion of sound reappears in his description of the role of the
speaking subject. On the one hand, he claims that the
"language-system [la langue] is not an entity and [. ..] exists
only in the speaking subject" (19/5 note); on the other hand, he
also emphasizes that "the language-system is not a function
of the speaking subject." (30/14) This contradiction is only

Ibid., p. 73.

32
 The unconscious chess player

apparently resolved by Saussure's distinction between the

individual and the social or collective subject, the former being
said to belong to the realm of speech, the latter to Ia langue.
Speech (parole) implies an historical, "diachronic" dimension,
langue a static, "synchronic" one. Speech, moreover, as a
function of the individual speaker, is essentially random and
unpredictable, while the language-system is by nature accessible
to conscious consideration:
Synchrony knows but one perspective, that of speaking subjects, and
its entire method consists in collecting their testimony; in order to know
to what extent something is real, it is necessary and sufficient to
t'xamine to what degree it exists for the subjects' consciousness.
(12.8/90)

The closure of the synchronic dimension, its self-presence,

dnd its priority over the diachronic dimension of language, are
determined by its capacity to be present, that is, conscious for a
"collective" subject. This priority of synchrony over diachrony
depends primarily on the regulated closure of the language-
system, as compared with its actualization and realization in
speech:

Ihe synchronic law is general, but it is not imperative. Doubtless it

imposes itself on individuals through the constraint of collective usage,
I'ut this relates only to speaking subjects. What we mean (by its
non-imperative character] is that, within the language-system, whatever
force may be that happens to prevail at any given moment, it can
ili'ver guarantee the maintenance of this regularity. The synchronic law
ii therefore merely the expression of an existing order at a given
iiioment. [... J Diachrony, on the contrary, supposes active forces
Ilirough which an effect is produced, a thing executed. But this
imperative character is not sufficient to warrant applying the term
I,mw". . . [since] diachronic events always have an accidental and
.mngular character. (131/92.—93)

11w synchronic system of language does not completely

.li'tcrmine the actual behavior of individual speakers, but it does
their behavior, subordinate it to a system of laws, and
uli'reby render it comprehensible as a scientific object of study.
I 'i,ichronic "events," on the other hand, while they can produce
can never generate a language, insofar as language must
J'mm.sess the quality of a system.

33
 Return to Freud

Saussure attempts to illustrate this difficult relationship

between synchrony and diachrony by means of two compari-
sons. First he compares synchrony to
the projection of a body on a flat surface. Any projection depends
directly on the nature of the body projected and yet differs from it — the
projection being a thing apart. Otherwise there would be no science of
projections; it would suffice to consider the bodies themselves. In
linguistics the same relationship obtains between the historical facts
and a language-state; the latter is like the projection of the facts at a
particular moment. It is not in studying bodies, that is, diachronic
events, that one learns about synchronic states, any more than one
acquires an idea of geometric projections by a detailed study of
different types of bodies. (1z4—1z5/87)
The diachronic dimension of language, that of speech, and
the individuals that correspond to it, thus appear in this
comparison to constitute the body of language, its living
presence. This presence, however, cannot as such be formu-
lated in laws or codified; it is essentially unconscious and can at
best be described. Only by virtue of a projection of the body on a
surface, does it become visible, consciously perceptible, and
susceptible of regulation. Saussure here resorts to the same
image as does Freud, when he compares the formation of the
ego to a projection upon a screen; in both cases, an ostensibly
self-contained, identifiable and systematic structure is produced
by projection. For Freud, however, the body that is projected is
itself already a "The ego is first and foremost a bodily
one, it is not merely a surface entity but is itself the projection of
a surface" [my italics].9 Thus, for Freud, the body is not defined
by depth, in opposition to surface: it functions itself as surface.
And if we recall Lacan's theory of the imaginary-fictive
constitution of the ego, we can see how his description of
this function in terms of misapprehension [verkennen] is directly
derived from Freud's insight into the "superficial" character of
the ego. For Saussure, by contrast, the bodily nature of language
as speech is regarded from the very beginning as a living
Sigmund Freud, The Ego and the Id, SE, XIX p. z6. In a footnote, added in 192.7,
Freud explained his remark as follows: "The ego is ultimately derived from
bodily sensations, chiefly from those springing from the surface of the body.
It may thus be regarded as a mental projection of the surface of the body,
besides, as we have seen above, representing the superficies of the mental
apparatus."

34
 The unconscious chess player

substance and as self-identical, although according to Saus-

sure's own comparison, it only comes to be and acquires its
systematic identity as language-system through a mechanism of
projection. As in the case of the signifier and the signified,
Saussure is thus impelled to posit an identity prior to all
articulation and difference: that of the speaking subject in la
parole.
And yet, strangely and significantly enough, this subject, as
we have seen, lacks what in modern thought, at least since
Descartes, has been considered to be the constitutive attribute of
all identity: self-consciousness. As it turns out, for Saussure,
everything in of speech is unconscious not
merely in the contingent, descriptive sense, but in the Freudian
sense of being necessarily inaccessible to consciousness. This
becomes clear through the second comparison used by Saussure
to clarify the relationship between diachrony and synchrony in
language:
Rut of all comparisons that might be imagined, the most fruitful is the
one that might be drawn between the play of language [langue] and
chess. In both instances we are confronted with a system of values and
with their modifications. A game of chess is like an artificial realization
of what language offers us in a natural form. (x
Saussure goes on to elaborate the similarities between language
chess games: the value of particular elements depends upon
their given position within the system; the system is only
momentary and changes from one moment to the next; the
values, however, depend upon an immutable agreement,
through which the rules of the game are set; these exist before
the game starts and continue to exist after each move. And
whereas changing the position of the elements (making a move)
changes the state of the system in accordance with the rules of
the game, it does so in a way that is never entirely predictable.
linally, the manner in which this change comes about has
.ibsolutely no bearing on the course of the game or on the
significance of a given state of the system. Saussure neverthe-
less adds that his comparison is unsatisfactory concerning one
essential point:
While the chess player has the intention of operating a shift and
the system, the language-system premeditates nothing. The

35
 Return to Freud

pieces of language are displaced, or rather modified, spontaneously

and fortuitously. . . In order for chess to resemble the play of language
in all points we would have to imagine an unconscious or unintelligent
player. (127189)
In order, then, to do justice to the game of language, it does
not suffice to trace its system back to a collective consciousness;
at most this consciousness can belatedly conceive the system in
its static condition, and in so doing, it coincides with the
consciousness of the linguist. Yet in contrast to the chess game,
Saussure argues, the rules by which language is transformed
remain scientifically inconceivable.10
In thus acknowledging that his comparison is deficient on an
essential point, which has to do with the decisive relationship of
the game to the rules of transformation, Saussure comes very
close to reopening the entire issue his example is intended to
close: the relation of synchrony and diachrony. For the
comparison with chess, if pursued, turns out to be incompatible
with the opposition between the two perspectives Saussure is at
pains to establish. What Saussure does not mention — although
he, like Freud's patients, "knew it all the time" — is that there are
two kinds of rules in chess. There are those that regulate the
formal conditions of the game: the squares on the board, the
ways in which the different pieces move, etc. But there are also
those which are less fixed, that regulate the different ways
games have been played. These are rules of strategy, and
although they never allow
predictable, they are also far from being merely "individual,"
"contingent" and "empirical." They structure the transforma-
and hence, the game of che!s And if they can
is is because the "synchronic
structure" that Saussure saw exemplified in the state of the
chessboard, at any given moment, is never present as such in
any chess game. Rather, the synchronic state of the board is, in
principle, and not just in practice, split, divided from itself, by
the move. It is this that distinguishes the "position" of the board
from that of the synchronic state of language: the position of

This is one of the reasons why it is perhaps even more apt to speak of the
apparition of language in English than in French. See the article of that name
mentioned above.

36
 The unconscious chess player

ihe board is, as it were, structurally dislocated, disarticulated,

through the fact that the game requires that there be two players,
•ind that their relation to eachóthèr is rntrinsically dissymmetric-
,iI, since one of them has the next move. In this sense, the example
chess suggests that the temporal, diachronic dimension of
liscontinuous successivity is inscribed in each "present state" of
the system; and that the subject of this system, far from being
unitary — be it as individual or collective consciousness — is split,
Through his example, then, Saussure anticipates
the notion of language-game, developed years later by Wittgen-
•btein; but he anticipates it with a Freudian flair, for the essence
that game is an unresolvable, constitutive conflict. And its
tules, therefore, can only be strategic. The rules of this strategy
•ind their unconscious calculation will constitute one of the
major concerns of Lacan's return to Freud. We begin to see how
•md why that return should pass by way of Saussure, although
it will be a Saussure read in the same manner that Freud reads
I unconscious: in terms of what is not said, but what implicitly
he
has been "known all the time." In terms, that is, of a knowledge
that eludes consciousness through a certain inscription.

37
 4

The rise and fall of the signifier

To properly understand the significance of structural linguistics

for Lacan, one cannot overlook its internal contradictions.
Saussure's writings are of interest to him, less as the site where a
certain strain of modem linguistics sought to pose its founda-
tions, than as the theater in which the structure of language and
its relation to the subject are staged as questions. As we have
seen with the example of the chess game, this staging is only
indirect and implicit, rather than thematic and declared. This is
due not to individual inadequacy, but to the incapacity of a
discipline such as linguistics to address the problem of the
subject as such. Though Saussure's determination of the
language-system in terms of differential articulation subverts
the traditional concepts of object (as referent) and of subject (as
self-consciousness), this disturbance is minimized or effaced so
as not to disturb the project of establishing linguistics as a
rigorous science. As we have seen, the principle of linguistic
difference is partially, at least, revoked when Saussure situates
the reality of language not in "value," but in the "totality of the
sign," itself a result of binary opposition. It is precisely this
ambivalence in the thought of Saussure that serves as Lacan's
point of departure. "The instance of the letter in the uncon-
scious or reason since Freud" was written in 1955, three years
after the Rome discourse in which 'The function of speech and
language in psychoanalysis" is set forth with the help of
Saussurean concepts for the first time. This address, which has
the character of a manifesto, resembles in this respect the lecture
delivered in Vienna in 1955, Ofl "The Freudian thing" (La Chose
freudienne ou sens du retour a Freud en psychanalyse), which also
displays a certain programmatic character, in demarcating, often
polemically, the Lacanian position from the prevalent psychoan-
alytic tendencies of the period, including, above all, the
38
 The rise and fall of the signifier

powerful New York School of Ego Psychology (Hartmann, Kris,

I but also with respect to those efforts to assimilate
Freudian thought to neurophysiology. It is a demarcation that
insists upon the importance of Saussurean linguistic theory for
psychoanalysis, without, however, giving a detailed account of
just what in Saussure is of decisive significance. Only with the
lecture on "The Instance of the Letter," does Lacan begin to put
his cards on the table, while at the same time making clear that
his brand of "teaching" — "enseignement" is a favorite term of
— in no way implies putting all one's cards on

the table. We will have occasion to examine the structure and

certain of the consequences of the argumentation deriving from
this strategy. But it is advisable to begin with the essay that is
surely most explicit about the linguistic moment in Lacan's
thought, and in particular its relation to Saussure, "The instance
of the letter," which, as Lacan remarks, is situated halfway
between the written and the spoken word. Only after having
worked out some of the more general aspects of this rela-
tionship, will we return to some of the issues raised in the
earlier, more programmatic lectures of Rome and Vienna.
As is well-known, Lacan takes as his point of departure the
Saussurean concept of the signifier. The hidden problematic of
this concept, as we have attempted to elaborate it, is silently
hut unmistakably foregrounded by Lacan, in one bold graphic
stroke, as it were, through which the Saussurean diagram of
the sign is stood squarely on its head. The S of the Signifier
is no longer on the bottom, but on the top. The reverse, which
is how Saussure represented the sign, is itself a symptom of
the ambivalence of his semiotics, torn, as it were, between
the notion of radical difference as the principle mechanism in
•tllowing signs to signify, and the fear that this would render a
systematic classification of signs ultimately impossible. The
consequent reduction of the differential notion of "value" to
"the totality of the sign," now determined to be the basic
"reality" of language, could only result in the privileging of
the signified over the signifier, of meaning over articulation in
Saussure's approach to linguistic processes. Considered from
he perspective of the totality of the sign, it is hardly fortuitous
that Saussure places the signified above the signifier in his
diagrammatic representation of the sign. And it is no less

39
 Return to Freud

significant that this representation would ultimately take the

form of an image, in which linguistic noun and extralinguistic
referent would be properly matched to one another. Saussure's
own graphic representation thus winds up contradicting his
attempt to conceive of language in a way that would not render
it structurally dependent upon extralinguistic reference. And
yet, as already indicated, the project of construing language as
an object capable of scientific study leaves Saussure little choice.
If language is to be conceived as a closed system, it is not
difference, but rather opposition, not the signifier but only the
signified that is capable of closing it and of containing the
centrifugal and exorbitant tendencies of "value," "difference"
and of the "signifier." Thus, although Saussure distinguishes
between meaning (signification) that re-presents an idea, and
value, which is an intralinguistic relation, it is meaning that has
the last word. How could it be otherwise insofar as language is
conceived from the point of view of the word, since what is
commonly used to define the word is precisely the meaning it
is held to signify.'
In contrast, and without wasting words, Lacan inverts the
Saussurean sign. The relationship of signifier to signified is,
however, not merely inverted; it is also twisted, dislocated,
turned around. The Saussurean symmetry, ultimately based
on the priority of the signified, is displaced: the line used by
Saussure to separate the two dimensions of the sign from each
other, did so only in order then to unite them all the more
definitively. This line is now turned into a "bar" — an obstacle
or barrier — which, while making meaning (signification) pos-
sible, at the same time resists it.2 What is brought into relief
here is not merely the separation between signifier and sig-
nified, but instead what is implicitly at work in Saussure but
Saussure's description of the linguistic function of the word, however, betrays
once again a certain ambivalence. He begins — in the published notes of
his lectures, at least — by declaring the word to be far too abstract and
indeterminate, to be considered the "concrete entity" of language, that
a scientifically rigorous linguistics requires as its foundation. Elsewhere, in
determimng the sign as an oppositional but total structure, he refers to the
word as though it were precisely the fundamental linguistic phenomenon. On
some of the Implications of this oscillation, cf. "The apparition of language."
2 Sigrufier and signified, Lacan observes, designate "distinct orders initially
separated by a barrier resistant to signification." "The instance of the letter in
the unconscious," Sheridan, p. 149; Ecrits, p. 497.

40
 The rise and fall of the signifier

.ilso explicitly disavowed: a certain structural primacy of the

signifier over the signified, the latter considered as dependent
upon the former. This primacy of the signifier implies in turn
that language is no longer understood as re-presentation, but
instead as differential articulation. If Lacan thus writes that
"there is no signification that sustains itself other than by re-
to another signification," he concludes from this that
'the function of the signifier does not have to answer for its
existence to any signification whatsoever."3 On the contrary,
it is by referring to other signifieds, that is by means of the
that the signified first becomes self-identical, that is,
•i signified. Its identity thus must be conceived as an effect of
the signifier, insofar as the signifier embodies the process of
in terms of the play of differential relations. The
•4ignifier, conceived as a movement of difference, thereby be-
omes co-extensive with language itself, insofar as the latter is
no longer understood as a function of representation, but in-
as articulation. "The structure of the signifier is, as is
tommonly said of language in general, that it is articulated."4
What is criticized here is not only the metaphysical view of
Lrnguage as re-presentation, but also the ontological premises
upon which this conception of language is based, above all,
the priority of the referent, held to be self-identical above and
beyond the differential relations of language.
With the status of the referent, that of the object is also
tadically problematized. Lacan indicates this by describing how
"the signifier actually enters into the signified; that is, in a form
that, not being immaterial, poses the question of its place in
reality."5 To illustrate the process, Lacan adduces an example

'"Dans cette voie les choses ne peuvent ailer plus loin que de démontrer qu'il
n'est aucune signification qui se soutienne sinon du renvoi a une autre
signification [...]" However, Lacan continues, we shall never be in a
position to take on the question of language, which it imposes upon us, "as
long as we have not rid ourselves of the illusion that the signifier answers to
the function of representing the signified, or, better put: that the signifier has
to answer for its existence to any signification whatsoever." Sheridan, p. 150;
Ecrits, p. 498.
'"Or Ia structure du signifiant est, comme on le dit communément du langage,
qu'il soit articulé." Sheridan, p. Ecrits, p. so'.
"Ceci [est] pour montrer comment le signifiant entre en fait dans le signifle; a
savoir sous une forme qui, pour n'étre pas immatérielle, pose Ia question de
sa place dans Ia réalité." Sheridan, p. ip; Ecrits, p. 500.

4'
 Return to Freud

that takes the place of the tree (arbre) used by Saussure to

represent the sign, and which at once places the latter in a
distinctly different light. Lacan supplements Saussure's diagram:

with his own:

LADIES GENTLEMEN

The tree is replaced by two words: Hommes and Dames, Gentle-

men and Ladies. The doubling of the signifier by Lacan — or is it
the splitting of the signifier? — into Gentlemen and Ladies seems
to do greater justice to the Saussurean difference than the single
word "tree," with all of its organic and holistic associations. Yet,
it is the signifieds chosen by Lacan which make the difference
strikm . the two words are inscribed at the top of the figure, as a
kind of title, underscored by a line beneath them; below,
separate from the legend, as it were, two doors, with rectangles
where the words should go. The two doors point to the verbal
signifiers above, which however are separated from them. In
addition, then, to choosing an example designed to remind one
of "the urinary segregation typical both of western culture and
of 'primitive' societies,"6 the separation of signified (the words)

6
On voit que, sans beaucoup étendre Ia portee du signifiant intéressé dans
I'expérience, soit en redoublant seulement I'espece nominale par Ia seule
juxtaposition de deux termes dont le sens complementaire parait devoir s'en
consolider, Ia surprise Se produit d'une precipitation du sens inattendue; dans
I'image de deux portes jumelles qui symbolisent avec l'isoloir offert a l'homme
occidental pour satisfaire a ses besoins naturels hors de sa maison, l'impératif
qu'il semble partager avec La grande majorite des communautés primitives et

42
 The rise and fall of the

signifier (the doors) illustrates the dependency of meaning

upon relations of signification, which cannot simply be taken for
granted. Without the signifier, no signified, and furthermore no
identifiable, definable referent. Yet how do things stand with
he subject, with the gentlemen and ladies, who take the lead of
the signifier in satisfying their "needs"? To answer this question
I acan tells a short, "true" story:
A train arrives at a station. A little boy and a little girl, brother and
are seated in a compartment face to face next to the window
through which the buildings along the station platform can be seen
p.Issing as the train pulls to a stop. "Look," says the brother, "we're
•st Ladies!"; "Idiot!" replies his sister, "Can't you see that we're at
;cntlemen."7
Are we confronted here with a simple example of perspectivism
nr relativism? Hardly, since the first thing that has to be taken
into account is that "the rails" (les rails: and the story is also a
tuillerie; a raillery or jest) materialize the bar of the Saussurean
but "in a form designed to suggest that its resistance
might be other than dialectical."8 The subject here is literally
borne by the bar separating the signifier from the signified: its
is on the rails, and yet on rails that simultaneously derail.
I he quarrel between the little girl, who claims to have arrived in
'Gentlemen," and the little boy, for whom the place can only be
,mlled "Ladies," is impossible to arbitrate, at least as long as the
at issue — the relationship of the sexes — remains
• nuched in the binary structure of an alternative: either "gentle-
isis.'n," or "ladies," each ostensibly in its proper place. It is
therefore no accident that the place described in Lacan's illustra-
live figure is a toilet, a place that is never entirely "proper," and
Iii which access is generally more or less regulated. In Lacan's
ligure, the doors seem to be closed. We can, therefore, never be
•ntirely certain about what stands behind it all, and for Lacan,
'his is probably the most important aspect of the story:
We should add that only someone who has a cockeyed view of things
Iliterally: "who doesn't have holes in front of his eyes" — n'avoir pas les

soumet sa vie publique aux lois de Ia segregation urinaire." Sheridan,

mci; Ecrifs, p. 500.
.lwridan, p. 152.
'Iieridan, p. s 52..

43
 Return to Freud

yeux en face des trous), (it's the appropriate image here) could possibly
confuse the place of the signifier and that of the signified in this story.
or not see from what radiating (rayonnant) centre the signifier sends
forth its light into the shadow of incomplete significations.9

The traditional pathos of a metaphorics of light and darkness?

What radiates behind the bar of the signifier is the light of a hole.
Not the light that shines through holes, but that of the hole
as such. For with what Lacan calls the "precipitation of the
signifier," with which it leaves its mark upon the signified, the
light of the object goes out, or at best becomes a half-light,
chiaroscuro of difference. We can see "objects" only because
we can see holes: that is, the interstices, through which
relate to one another and delineate themselves. And such
interstices are not just intervals, not just spaces between objects
and words, but also cracks and fractures within them. The railing
of the rails opens to allow the raillery of ridiculous word-plays to
interrupt semantic seriousness. Insofar as the object is consti-
tuted in and through the play of signifiers, it falls through the
rails of its railing rays, as is the case of the word, "thing":

The thing [Ia chose], when taken literally, breaks up into the double,
divergent ray of the "cause" (causa), where it has found shelter in
French, and the nothing [rienJ to which it has abandoned its Latin garb
(rem).1°

This etymological word play — chose/cause, rien/rem reflects

the ambivalence of an object as soon as it is viewed in its
dependence upon the signifier: while the object may function as
a cause, insofar as every signifier as such refers — and by virtue of
its form must refer — to a signified, the belated priority of the
signifier as difference and articulation marks the signified and
inscribes it in a structure of radical heterogeneity that cannol
be fully contained or comprehended in terms of identity. Thus,

"II faudrait, c'est bien I'image qui convient, n'avoir pas les yeux en face dc.
trous pour s'y embrouiller sur Ia place respective du signifiant et du signific.
et ne pas suivre de quel centre rayonnant le premier vient a refleter sa Iumièrt
dans Ia ténèbre des significations inachevées." Sheridan, p. 15z;
p. soo.
10 "et que Ia chose, a se réduire bien évidemment au nom, se brise en le doubli
rayon divergent de Ia cause oü elle a pris abri en notre langue et du den a qui
elle a fait abandon de sa robe latine (rem)." Sheridan, p. '5°; Ecrits, p. 498

44
 The rise and fall of the signifier

I(lative to the achieved identity of an object, this heterogeneity

s1 signification can be determined as a hole or blank, a trou,
ince what allows one to see — differential relations — is not itself
usible and in terms of visibility can therefore be compared to a
'hole." Seeing is only possible "through" difference, or de-
marcation. It is, however, quite a different question whether
.me is ready to see this hole — or, to stay with the terms of
I acan's story — whether one feels obliged to open the restroom
door. Here we touch on a problem which Freud called that of
the "lost object," a problem to which we shall return later, in
onnection with Lacan's theories of the phallus and of desire.
Nevertheless, it is already apparent that this problem cannot be
from the differential operation of language, or from
the influence of the signifier on the signified and through it,
upon reality as well.
Lacan's use of structural linguistics is thus no simple ap-
of an otherwise unproblematized model, although
this is the impression often left by the polemical pathos that
his recourse to Saussure and which is directed
.igainst the state of the psychoanalytic theory that dominated
the International Association at the time. Lacan's inversion of
the Saussurean formula, the emphasis thereby placed on the
..ignifier, develops the internal contradictions of structural
linguistics beyond the limits of linguistics as such. Nevertheless,
I .acan's adaptation of the Saussurean theory of the signifier by
no means entirely frees itself from the aporia of that theory. And
the latter are destined to play a determining role in
I acanian thought, they merit careful examination.
The problem can be localized as, precisely, one of localization.
I acan localizes the signifier in what he calls the letter. His point
.1 departure is the articulated structure of the signifier: "The
•.tructure of the signifier. . . is that it is articulated." In theori-
iing this articulation, Lacan at first follows the determination of
the signifier as phoneme, elaborated by the Prague School —
identified above all, with the names of Trubeckoj and Jakobson.
According to this approach, linguistic articulation entails a
double mechanism: signifying units "are subjected to a dual
• ondition: that of being reducible to basic, differential elements,
.ind that of combining in accordance with the laws of a closed
'.vstem." "These elements," Lacan continues, which constitute

45
 Return to Freud

one of the decisive discoveries of linguistics, are phonemes; we must not

expect, however, to find any phonetic constancy in the modulatory
variability to which this term applies, but rather a synchronic system of
differential couplings necessary for the discernment of words in a given
language)'
At first, then, Lacan follows Trubeckoj and Jakobson in
determining the signifier as essentially phonemic, and hence, as
binary opposition; he thereby also assumes their view, which
was shared by Saussure, that language consists of a closed,
synchronic system, in turn composed of a limited number of
binary oppositions. The heterological heritage of Saussure is
thus assumed, as well as its ambivalent revocation. Jakobson's
description of the essence of the phoneme as "mere other-
ness,"2 is thus cited, while at the same time this "otherness" is
identified with relations of binary opposition. Since the notion
of binary opposition presupposes a principle of identification for
which it, as relation, cannot account, it inevitably if
inconsistently recurs to the privilege of the semiotic medium of
sound.
In thus assuming the heritage of the Prague School, Lacan
comes close to what Derrida has called "phonologism."3 This
phonologism, which, as we have seen, is already at work in
Saussure, regards spoken discourse as the exemplary model of
language in general. In so doing, it ignores the implications of
its own insights into the differential structure of the linguistic
sign, which is as little to be identified with a particular medium,
as with determinate referents. The most consistent linguistic
critique of this "phonological deviation," is to be found in the
work of the Danish linguist, Louis Hjelmslev, founder of the
so-called school of "glossematics." To the Saussurean distinc-
" "Ces éléments, découverte decisive de Ia linguistique, sont les phonemes oü ii
ne faut chercher aucune constance phonetique dans Ia variabilité modulatoire
oü s'applique ce terme, mais le système synchronique des couplages
différentiels, nécessaires au discemement des vocables dans une tangue
donnée." Sheridan, p. 253; Ecrifs, p. 502.
12 Roman Jakobson & Moms Halle, Fundamentals of Language, The Hague, 1956,
p. i6.
II Derrida, On Grammatologij. In "The purveyor of truth," (La facteur de Ia
véritE), Derrida has analyzed what he calls the "phallogocentrism" of Lacan,
with particular reference to the latter's reading of Poe's 'Purloined letter."
English translation by Alan Bass. In The Post Card, Chicago: University of
Chicago Press, 1987.

46
 The rise and fall of the signifier

lion between signifier and signified, Hjelmslev adds that of

'lorm" and "substance." According to this view the tonal or
.I(oustlc aspect of sound would belong to the substance of
l.inguage, while its differentiality would belong to its form. Thus
instead of equating the "signifier" with sound or even with the
error that goes back to Saussure, despite the latter's
word — an
,.sdical critique of the linguistic pertinence of the concept of
• word" — Hjelmslev's distinction serves to emphasize that
.ignification is a function of differential relations as such, and
siot of a particular, privileged material. In this respect, Hjelm-
faithful to the most original aspect of Saussurean
is
linguistics, whose decisive insight is that of the differential
•.tructure of language and not, as phonologism — and at this
Lacan as well — holds, that of binary oppositions, above
•sll in the medium of sound. Indeed, the predominance given by
I acan to the opposition of the "full" and "empty" word: "parole
and "vide," would seem to place him unequivocally in
Ihe tradition of Western Phonocentrism.
On the other hand, insofar as Phonocentrism serves to
.ubordinate language as re-presentation or as expression to an
s'xtra-linguistic referent, whether this is construed as things
themselves or as the subject of self-consciousness — it is clear
that Lacan's endorsement of this tradition, and in particular of
its privileging of the parole, cannot remain unequivocal. And
indeed, after having given the (already cited) classical deter-
mination of the signifier as phoneme, Lacan goes on to
undermine this determination in his own distinctive fashion:
I hrough this, one sees that an essential element of the spoken word
itself was predestined to flow into the mobile characters which,
whether Didots or Garamonds, crowding together in lower-case render
validly present what we call the letter, namely, the essentially localized
'.tructure of the signifier.14

'Through this one sees" — par quoi l'on voit —this recurrent figure
of speech typifies Lacan's discourse, which — like that of the
'
'Par quoi I'on voit qu'un element essentiel dans Ia parole elle-même était
predestine a se couler dans les caracteres mobiles qui, Didots ou Garamonds
se pressant dans les bas-de-casse, presentifient valablement ce que nous
appelons Ia lettre, a savoir Ia structure essentiellement Iocalisée du sig-
nifiant." Sheridan, p. '53; Ecrits, p. 501.

47
 Return to Freud

unconscious — often seems to "know no contradictions," or

rather, to allow itself to be bound by the "law" of non-
contradiction. In this particular case, it is difficult to see just
what, in the determination of the signifier as phoneme, would
predestine it to flow into typography, much less into its "lower
cases." We need only recall the writings of Jakobson, to
recognize just how much this figure of the phoneme disfigures
the tenets of linguistic phonologism. For Jakobson, as for
Saussure, writing is unequivocally subordinated to speech:
Only after having mastered speech does one graduate to reading and
writing. There is a cardinal difference between phonemes and graphic
units. Each letter carries a specific denotation — in a phonemic
orthography, it usually denotes one of the phonemes or a certain
limited series of phonemes, whereas phonemes denote nothing but
mere otherness. Graphic signs that serve to interpret phonemes or
other linguistic units stand for these units, as the logician would say.
This difference has far-reaching consequences for the cardinally
dissimilar patterning of letters and phonemes. Letters never or only
partially reproduce the different distinctive features on which the
phonemic pattern is based and unfailingly disregard the structural
relationship of these features. There is no such thing in human society
as the supplantation of the speech code by its visual replicas, but only a
supplementation of this code by parasitic auxiliaries, while the speech
code constantly and unalterably remains in effect.

This gesture — the reduction of writing to its phonetic form in

order then to assign it a subordinate position as a "parasitic
auxiliary" — is not merely an almost verbatim replay of
Saussure's attack on the "usurpation" of writing in linguistics; it
also repeats a movement as old as metaphysics itself. It is
significant, therefore, that such a movement is nowhere to be
found in the writings of Lacan — even in their most "structural-
ist" phase. On the contrary, already in 1956, when this text was
written — and thus seven years before the appearance of
Derrida's first writings explicitly problematizing the power of
phonocentrism — a very clear anticipation of this problematic is
evident in Lacan. In drawing attention to Freud's characteriza-
tion of the dream as a rebus and as a system of inscription,
Lacan emphasizes that the dream's plasticity [Bildhaftigkeit] —

Jakobson/Halle, Fundamentals of Language, pp. 16—17.

48
 The rise and fall of the signifier

called by Freud "considerations of representabiity" — must be

understood as part of a hieroglyphics (faithfully
rendering Freud) and adds:
This fact could perhaps shed light on the problems involved in certain
modes of pictography which, simply because they have been aban-
doned in writing as imperfect, are not therefore to be regarded as mere
evolutionary stages.'6
Lacan thus clearly distances himself from the teleological
tendencies of phonocentrism, in which the latter reveals its
complicity with Western ethnocentrism. And this is most clearly
the case when dealing with the relationship between letter and
spirit:
Of course, as it is said, the letter killeth while the spirit giveth life. We
can't help but agree. . , but we should also like to know how the spirit
.

could live without the letter. Even so, the pretensions of the spirit
would remain unassailable if the letter had not shown us that it
Produces all the effects of truth in man without involving the spirit at
all. It is none other than Freud who had this revelation, and he called
his discovery the unconscious.'7
According to Lacan, "considerations of representability," constitute "a
limitation operating within the system of writing, rather than dissolving
it into a figurative semiology which would approach the phenomena
of natural expression ("une limitation qui s'exerce a l'intérieur du système de
l'écriture, loin qu'elle le dissolve en une sémiologie figurative oü ii rejoindrait
les phenomenes de l'expression naturelle. On pourrait probablement eclairer
par là les problèmes de certains modes de pictographie, qu'on n'est pas
autorisé, du seul fait qu'ils alent ete abandonnés comme imparfaits dans
l'écriture, a considérer comme des stades évolutifs.") Sheridan, p. i6i; Ecrits,
p. sir. Lacan's prudence illustrates at its best what could be called the
"structuralist vigilance" — which is in fact a Nietzschean vigilance — with
respect to teleological conceptions of history that legitimate prevailing values
as the "intrinsic ends" of all "development." Derrida's deconstruction of
"phonocentrism" proceeds from a similar suspicion of the normative
hypostasis of phonetic writing at the expense of non-phonetic forms of
Inscription. The term "post-structuralist" can serve, even today, to designate
the continuing vigilance with regard to teleological thinking, at least in its
more naive forms. The emphasis placed by Lacan on parole and on the letter
should not be isolated from his no less insistent foregrounding of the
scriptural — and above all, non-phonetic — nature of unconscious articulation.
'Certes la lettre tue, dit-on, quand 1'esprit vivifie. Nous n'en disconvenons
pas (. . .] mais nous demandons aussi comment sans Ia lettre I'esprit vivrait.
I es prétentions de I'esprit pourtant demeureraient irreductibles, si la lettre
navait fait Ia preuve qu'elle produit tous ses effets de vérité dans l'homme,
'sans que l'esprit ait le moms du monde a s'en mêler. Cette révélation, c'est
a Freud qu'elle s'est faite, et sa découverte, il la appelée I'inconsdent."
p. xg8; Ecrits, p.

49
 Return to Freud

Lacan's views that differential articulation — the signifier — is not

a transcendental principle and that the signifier must h&'
localized — namely in the letter — do not co-exist easily with
tendency to conceive of language as a closed system actualized
in speech. If it is no accident that in the train example already
discussed, the signifiers are clearly inscribed in reality — th
signs on the restroom doors — it is no less fortuitous that Lacaii
makes no explicit mention of this; for on the one hand, he is
concerned with the principle by which the signifier is radically
separated from its materialization: the letter is determined from
the very beginning as an effet — as a signifying effect

or as an effect of the signifier — a distinction to which Lacaii

subsequently returned in a discussion of the work of Serg
Leclaire:

That which I have written about the formations of the unconscious,

with the aid of letters that themselves are effects of the signifier, does
not warrant making the letter into a signifier, and even less, according
the letter priority over the signifier.18

The letter is thus an "effect" of the signifier, its "material

vehicle" and its necessary localization. Yet, on the other hand,
the medium of this vehicle is concrete discourse: "By letter I

designate that material support that concrete discourse borrows

from language."9
How concrete, however, is this discourse? What is its rela-
tionship to the different forms and functions of language: in
Saussurean terms, to language in general, including its non-
verbal forms? to the language-system? to spoken discourse? to
writing? These questions draw us further into Lacan's text.
First, however, a word may be in order on the notion of
"discourse," so prevalent in Lacan's writing as in much of
structural linguistics, and yet as elusive and hard to pin down as
its etymology would suggest: from dis-currere, the word "runs
here and there," like a broken-field runner. And indeed,
perhaps that is precisely its place: a broken field. In Saussure,
18
j, Lacan, "Lituraterre," in: Littérature 3, Paris, 1971, p. 5.
19
"Nous designons par Iettre ce support materiel que le discours
emprunte au langage." Sheridan, p. Ecrits, p. 495.

50
 The rise and fall of the signifier

iht notion is never used systematically, but seems to be

tresupposed, when Saussure distinguishes the two ways that
•'k'ments combine to form a linguistic structure: the "syntagma-
tic" and the "paradigmatic." It is in the former that something
like "discourse" has its place:

I )n the one hand, in discourse words enter into relations with each
tither through concatenation, based on the linear nature of language,
which excludes the possibility of pronouncing two elements at the
ime time. The elements are arranged one after the other in the chain
iii speech. These combinations, which have [spatial] extension as their
.upport, may be called syntagms. The syntagm is thus always composed
tif two or more consecutive units (. ) In the syntagm a term acquires
.

ih value only because it stands in opposition to what precedes or

lollows it, or to both.2°

In thus construing discourse as syntagmatic, Saussure, without

any further questions, simply transposes the linearity of
to the language system itself. Thus, although discourse
may be called upon to mediate between language as system
(langue) and language as speech (parole), it itself is structured by
speaking. And if the syntagm as such is attributed by Saussure
to the language system, not to speaking, this is because the
according to Saussure, is not a product of the free-will
cif the speaker. To use a post-Saussurean term, the syntagm
o coded. The fact remains, however, that the linear structure of
the syntagm, in its one-dimensionality, as an irreversible
sequence, is conceived on the model of speech.
By contrast, the second, "associative" aspect or axis of
linguistic relations cannot be conceived as a mode of discourse:
On the other hand, outside discourse, words having something in
common are associated [s'associentj in memory and groups are thus
lormed among which very diverse relations prevail [.. I They are not
.

linearly; their seat is in the brain; they are part of that inner
treasure that constitutes the language of each individual. We will call
them associative relations.21

While associative relations may constitute the "treasure of the

language system," and may bear the imprint of the language-
Cours, p. 170—171/12.3.
Ibid., p. 171/113.

51
 Return to Freud

system, they still seem to lack one of the essential properties of

that system: the closure consisting of a limited number of binary
oppositions. In contrast to the determinate and limited sequen-
tial order of the syntagm, "terms in an associative family occur
neither in fixed numbers nor in a definite order. [...] A
particular term is like the center of a constellation; it is the
point of convergence of an indefinite number of co-ordinated
Associative relations thus are located outside of
discourse, are virtual, in contrast to the actuality of the syntagm,
and yet are as much a part of la langue as the syntagm.
This division of the linguistic function was taken up and
developed by Roman Jakobson in his essay on the "Two types
of language and two types of aphasic Using
Saussure's distinction as a point of departure, Jakobson distin-
guishes between two "modes of arrangement" at work in every
speech act: (r) combination or contexture, Saussure's syntagms,
and (z) selection or substitution, involving the equivalence or
similarity of linguistic elements. Based on his investigation of
different disturbances displayed in aphasia, Jakobson tries to
show how disorders involving the linguistic function can be
grouped along two axes: either as the inability to recognize
similarity or equivalence and consequently the inability to select
or substitute; or disturbances involving relations of contiguity
limiting one's ability to combine and contextualize. Though
these two different operations are carried out by the individual
speaker, listener or reader, their roots should be sought, accord-
ing to Jakobson, in the structure of the linguistic sign itself, the
operation of which is no longer conceived of as being linear, but
instead as being multidimensional. Saussure's "associative"
relation is thereby de-psychologized and his linear conception
of the sign broadened.
Perhaps even more important, however, is that concrete
discourse no longer is construed in a one-dimensional manner.
Instead, it has to be considered in terms of the multi-
dimensionality of the linguistic sign. Jakobson designates these
two functions or operations as the metaphoric (based on
similarity) and the metonymic (based on contiguity) poles o

174/12.6.
Reprinted as Part 11 of The Fundamentals of Language.

52
 The rise and fall of the signifier

I inguage. In the realm of literature, Jakobson interprets

a predominance of
'1w metaphoric function, based on similarity or equivalence,
while explaining literary Realism as the use of metonymy, that
'.. of contiguity and contextuality. In addition, he interprets
I rt'ud's displacement and condensation as procedures
.inploying what he considers to be a form of metonymy:
.vnecdoche, while claiming that identification and symbolism
in' based on metaphoric similarity.
From this brief review, it should be clear that Jakobson's
sl,thoration of Saussurean theory provides Lacan with a decisive
,"ecedent in relating linguistic operations to Freud's description
"I unconscious mechanisms in the Interpretation of Dreams.
I.ikobson's precedent, however, by no means serves as a model.
I does not merely apply it unchanged: it is subjected to
interpretation and transformation, and indeed, the
litter are not without implications for the theories of Jakobson
himself. The distinction between metaphor and metonymy, as
likobson develops it, tends to relapse into a pre-Saussurean
• of language, insofar as linguistic functions are
• instrued in terms of notions — similarity and contiguity — that
ire determined semantically, rather than syntactically (that is,
in terms of differential relations). It is therefore not without
.ignificance that Jakobson introduces his discussion of meta-
ithor and metonymy by distinguishing between the "general
iiwaning" of a sign, governed by the function of substitution,
md its "contextual meaning" based on combination. This
•listinction, however, presupposes that a general meaning can
Itt' generated independently of context, and that it is only the
l.;tter which is differentially constituted through the differential
of signifiers. In short, the Jakobsonian distinction
l't'tween metaphor and metonymy subordinates the notion of
linguistic difference — once again — to a logic of binary opposition
l'.ised upon the priority of identity over difference.
It is evident, therefore, that Jakobson can offer a precedent,
limit in no way a model for the Lacanian approach to language,
which, in however complex a manner, is based upon the
of the signifier, and hence, of differential, syntactical
im'lations over semantic functions. Lacan's use of Jakobson,
of Saussure, is commanded by the strategical imperative of
53
 Return to Freud

distinguishing the authority of a precedent from that of a model.2'

Thus, although at first Lacan appears to accept Jakobson's
version of Saussure's linear view of the sign, there is an implicit
critique of the linearization of context in his notion of the
"signifying chain":
There is, in effect, no signifying chain that does not have, as if attached
to the punctuation of each of its units, a whole articulation of relevant
contexts suspended "vertically," as it were, from that
The fact that the "vertical" dimension of language is, in this
account, "suspended" to "the punctuation of each of its units,'
indicates that the "paradigmatic" or "metaphorical" axis is
construed in terms of syntax, rather than of semantics. The
"associative" aspect of language can no longer be considered in
terms of the multiplication of meaning; rather, the category of
meaning itself is inscribed in the movement of contextual
relations out of which the signifier emerges. The sturdy
Saussurean image of the tree decomposes into the different
meanings of the signifier "tree" (arbre), which may also be read
as an anagram of the Saussurean barre; such decomposition
problematizes the unity of the word as the basic building block
of language. The linguistic significance of a word can just as
easily be a function of its graphic, phonic or typographic fea-
tures as of its different meanings. Thus, it is not only in the
verse of Valery, which Lacan inscribes in his text, that the "tree
says No!":
Non! dit l'Arbre, ii dit: Non! dans l'étincellement
De sa tête superbe
Que Ia tern pete traite universellernent
Comme elle fait une herbe.26

24 In their astute reading of "The instance of the letter," Philippe Lacoue-

Labarthe and Jean-Luc Nancy analyze Lacan's use of predecessors in terms of
a défournement, the more or less violent, more or less perverse appropriation of
a movement by giving it a different direction. In current French usagt',
the word also designates "hijacking." See: Ph. Lacoue-Labarthe, Jean-Lu
Nancy, I.e fit re de Ia left re, Paris: Galilee, 1973. Second edition: Paris x99o.
25 "Nulle chaine signifiante en effet qui ne soutienne comme appendu a Ii
ponctuation de chacune de ses unites tout ce qui s'articule de
attestés, a Ia verticale, si l'on peut dire, de ce point." Sheridan, p. 154;
P. 503.
26 Sheridan, p. Ecrits, p. 504.

54
 The rise and fall of the signifier

Ike tree says "no!" to the storm that in this text and context
hi'comes the signification, which only negatively, through its no
md its difference lets the tree have its say, in a verse that is
not by the signified, but, as Lacan remarks, by the
'law of the parallelism of the signifier." This parallelism, Lacan
notes, is marked by a "comme" — "as" — which operates
independently of the meaning of what it "equates": the tree is
treated like a blade of grass by the tempest, which "reduces" its
'majestic head" by "treating [itj universally." This movement of
and "contradiction," Lacan continues, culminates in
the "condensation" of tête and tempête, effectuated through
.issonance. To this, we might also add that of traite. What Lacan
*.tlls "the law of the parallelism of the signifier," thereby
tonsists in a repetition of sound, which may produce meaning,
hut which is not dependent upon it.
Summing up his argument, Lacan writes: "What this struc
ture of the signifying chain discloses," is the subject's possibility
"to signify something entirely different from what it says."27
Insofar as this movement of signification is constitutionally
.rnchored in language itself, and does not depend upon the
onscious intention of the subject, the function of discourse is
no longer that of disguising — or, one might add: of expressing
thought. Rather, it is to "indicate the place of this subject in
the search for truth."28
"Truth," is therefore no longer determined as the adequation
a thought to its object, or that of an expression to its thought,
hut rather has to do with a relation between signifiers, which
here is in turn associated with "metonymy." In the "Instance
the Letter. . . ", metonymy is redefined as "the properly
'.ignifying function,"29 which supplements its traditional defini-
tion as a relation of "word to word" (mot a mot). The traditional
s'xample that Lacan cites: "thirty sails," used to designate a
Ilotilla of thirty ships, can serve to introduce this redefinition.
the figure of thirty sails, which, in the strict sense, is more of
m synecdoche than a metonymy, provides us with little reliable
information about the fleet of ships it is said to designate, since a

Sheridan, p. 155; Ecrits, p. 505.

'Ibid., p.

55
 Return to Freud

ship can as easily possess eight sails as one. Although one might
be tempted to reply that a sail remains part of a ship, even if the
ship has many of them, Lacan's concern here seems to
problematize a notion that is presupposed by all synecdoche,
and which itself necessarily depends upon the identity of the
signified: that of totality. And indeed, the figural movement 01
this signified — by means of which the ship seems to split or
double itself in the process of conferring its identity upon the
sails — is far more pertinent to the effect of metonymy Lacati
is discussing than is the familiar relationship of part and whole.
The term, metonymy, is designed to call attention to the
fact that the chain of signification is constituted of signifiers,
linked to one another by their differential function. What,
however, does this imply concerning the metaphor? Is it
less concrete?
Before addressing this question, it may be helpful to recall
how Lacan defines metaphor in the first place. Lacan adopts
Jakobson's notion of substitution, but with an essential differ-
ence: insofar as the substitution takes place between signifiers,
it cannot be based upon a semantic or substantial equivalence
or similarity, as Jakobson often presumes. Lacan's clearest
statement on this question is to be found not in the "Instance oI
the Letter," but instead in a short note entitled "The Metaphor
of the Subject."3° Here Lacan criticizes the theory of metaphor
elaborated by Charles Perelman, according to which metaphoric
substitution is based upon analogy, and hence upon similarity.
In response, Lacan asserts that:
Metaphor is — radically seen — the effect of substituting one signifier lo,
another in a chain, without anything natural predestining it to thi'.
function of p/iota [vehicle], except for the fact that it deals with tw,
signifiers, which as such are reducible to a phonematic opposition.3'
Metaphor is thus not based either on a substantial similarity
or on equivalence — which would once again imply the primacy
of the signified — nor is it, as the surrealists (following a long
tradition) claimed, the product of the simultaneity of two
actualized signifiers; rather:
30Published in an appendix to the second French edition of the Ecr,!
pp. 889—89z.
" Ibid., p. 890.

56
 The rise and fall of the signifier

It breaks out between two signifiers, one of which has taken the place
the other in the signifying chain, the occulted signifier remaining
present through its (metonymic) connexion with the rest of the chain.32
A clearly determined relationship of presence and absence of
the signifier is thus described here: the replaced, absent signifier
is driven under the bar, as it were, into the realm of the signi-
lied — one could say it is "repressed" — yet, as an excluded and
absent signifier it still remains present through its syntagmatic
relationship to the rest of the chain. This suppression, in the
most literal sense, is, however, in no way a removal, lifting
or surpassing of the bar itself: though it may be crossed over,
it still stays in place, for the repressed signifier remains a
signifier even in the position of the signified.
Thus, if metonymy marks the proper function of the signifier
that is, the formation of the signifying chain — the function of
metaphor is no less indispensable, insofar as no signifying chain
an exist without simultaneously depending upon the signified.
Metaphor confers its name on that movement of dependence,
already noted in Saussure: the self-precipitation of the signifier
as signified, which in virtue of the differential structure of sig-
nification must have always already been a signifier, in order
to become a signified. "One sees," writes Lacan, "that meta-
phor occurs at the precise point where sense takes place in
non-sense."33 This taking-place, Lacan argues, is described in
theory of jokes, as well as in his account of condensation
I
Verdichtung).
The "precipitation" by which the signifier drops down, as it
were, to become a signified, discloses here a further aspect,
which we could at most only suppose in Saussure: the process is
a deadly one. The verse of Hugo, from "Booz endormi," which
I .acan cites to exemplify the function of metaphor, leaves no
doubt about this. If the proper name of Booz is replaced, and
indeed repressed, in the verse Lacan quotes — and repression,
we should remember, always involves the supplanting or

"Elle jaihit entre deux signifiants dont l'un s'est substitué a l'autre en prenant
sa place dans Ia chalne signifiante, he signifiant occulté restant present de sa
connexion (metonymique) au reste de Ia chaine." Sheridan, p. 157; Ecrits,
p. 507.
"On voit que ha metaphore se place au point précis ou he sens se produit dans
le non-sens." Sheridan, p. Ecrits, p. 508.

57
 Return to Freud

translation of one term by another — this verse: "Sa gerbe n'etai!

pas avare ni haineuse" (His sheaf was neither miserly nor
spiteful), still retains the place of Booz by means of the
syntagmatic relationship to the predicates denied by the sheaf.
The exclusion of Booz becomes definitive, however, when the
possessive pronoun "his" (sa gerbe) arrives on the scene. Thus —
and we are forced to simplify Lacan's commentary here
considerably — on the one hand, the surprising fertility, the
unexpected paternity of the old man is alluded to, yet at the
same time, his ability to produce life is reduced to "less than
nothing," when contrasted with the fertility of the sheaf.
One sees here how, by virtue of its significance, Hugo's verse
emerges as an exemplary representation of the process 01
signification itself. For there is no movement of the signifier that
would not be dependent upon the signified: the question is only
how the signified relates to the signifier, as cause or effect? In
this particular case, the structure and meaning of paternity are
performed, as it were, through the repression of the name
of the father as a proper name. Lacan emphasizes that this
repression must be seen in relation to the myth of the parricide,
as expounded in Totem and Taboo. Without being able to pursue
this point further we have perhaps caught a glimpse 01
how the structure of language, like that of the object, comes to
be determined by the movement and operations of the signifier.

Cf. "On a question prior to all possible treatment of psychosis," in: Sheridan,
p. 179—z24, Ecrits, pp. 531—583. have discussed Lacan's reading of th
1

Schreber case in a preface to the American edition of Schreber's Memoirs of My

Nervous Illness, Cambridge Mass.: Harvard University Press, 1989.
 5
I. + + ++++ ++ ++++++ + + ++ +++ + ++++++ + ++

Significant fallout: metonymy

and metaphor

Ilefore proceeding to the next section of "The instance of the

letter in the unconscious," it may be useful to attempt to
summarize Lacan's remarks in the first part of that article. He
takes as his point of departure the insight that what he calls in
lrench l'expérience psychanalytique, and which can be translated
as "psychoanalytic experience," but also as "psychoanalytic
experiment," discovers in Freud's account of the unconscious,
not merely speech (Ia parole), but moreover "the entire structure
of language."1 This structure is determined through the distinc-
tive properties of the signifier, which is constituted on the one
hand through opposition, on the other, through a capacity to
tombine with other signifiers to form a chain, "in accordance
with the laws of a closed order."2 This order consists of two
radically heterogeneous but interdependent dimensions: that of
the signifier and that of the signified, which are separated from
each other by a bar — derived from the Saussurean division of
the sign, but above all, from its graphic representation. But this
l'ar is not merely a mark of separation: it is also a limit that must
necessarily be transgressed, for the signifier is also precipitated
into the realm of the signified, leaving its mark upon the latter,
and allowing it to be designated as an "effect." Lacan there-
lore emphasizes a certain "primacy of the signifier over the
'.ignified."
The signifier is associated by Lacan with the phoneme, which
Jikobson, following Trubetskoj, defines in terms of its "distinc-
tive features," and as "mere otherness." However, from the
• Notre litre (l'instance de la lettre] fait entendre qu'au-dela de cette parole, c'est
toute Ia structure du langage que Iexperience psychanalytique découvre dans
linconscient." Sheridan, p. 147; Ecrits, p. 495.
4heridan, p. Ecrifs, pp. soi—soz.

59
 Return to Freud

translation of one term by another — this verse: "Sa gerbe n'etait

pas avare ni haineuse" (His sheaf was neither miserly nor
spiteful), still retains the place of Booz by means of the
syntagmatic relationship to the predicates denied by the sheaf.
The exclusion of Booz becomes definitive, however, when the
possessive pronoun "his" (sa gerbe) arrives on the scene. Thus —
and we are forced to simplify Lacan's commentary here
considerably — on the one hand, the surprising fertility, the
unexpected paternity of the old man is alluded to, yet at the
same time, his ability to produce life is reduced to "less than
nothing," when contrasted with the fertility of the sheaf.
One sees here how, by virtue of its significance, Hugo's verse
emerges as an exemplary representation of the process of
signification itself. For there is no movement of the signifier that
would not be dependent upon the signified: the question is only
how the signified relates to the signifier, as cause or effect? In
this particular case, the structure and meaning of paternity are
performed, as it were, through the repression of the name
of the father as a proper name. Lacan emphasizes that this
repression must be seen in relation to the myth of the parricide,
as expounded in Totem and Taboo. Without being able to pursue
this point further we have perhaps caught a glimpse of
how the structure of language, like that of the object, comes to
be determined by the movement and operations of the signifier.

Cf. "On a question prior to all possible treatment of psychosis," in: Sheridan,
p. 279—224, Ecrits, pp. 531—583. 1 have discussed Lacan's reading of the
Schreber case in a preface to the American edition of Schreber's Memoirs of My
Nervous Illness, Cambridge Mass.: Harvard University Press, 1989.

58
 5

'4,

Significant fallout: metonymy

and metaphor

Before proceeding to the next section of "The instance of the

letter in the unconscious," it may be useful to attempt to
summarize Lacan's remarks in the first part of that article. He
takes as his point of departure the insight that what he calls in
French l'expérience psychanalytique, and which can be translated
as "psychoanalytic experience," but also as "psychoanalytic
experiment," discovers in Freud's account of the unconscious,
not merely speech (Ia parole), but moreover "the entire structure
of language."1 This structure is determined through the distinc-
tive properties of the signifier, which is constituted on the one
hand through opposition, on the other, through a capacity to
combine with other signifiers to form a chain, "in accordance
with the laws of a closed order."2 This order consists of two
radically heterogeneous but interdependent dimensions: that of
the signifier and that of the signified, which are separated from
each other by a bar — derived from the Saussurean division of
the sign, but above all, from its graphic representation. But this
bar is not merely a mark of separation: it is also a limit that must
necessarily be transgressed, for the signifier is also precipitated
into the realm of the signified, leaving its mark upon the latter,
and allowing it to be designated as an "effect." Lacan there-
fore emphasizes a certain "primacy of the signifier over the
signified."
The signifier is associated by Lacan with the phoneme, which
Jakobson, following Trubetskoj, defines in terms of its "distinc-
tive features," and as "mere otherness." However, from the
"Notre titre (l'instance de la left reJ fait entendre qu'au-dela de cette parole, c'est
toute Ia structure du langage que l'experience psychanalytique découvre dans
l'inconscient." Sheridan, p. 147; Ecrits, p. 49g.
2 Sheridan, p. Ecrits, pp. 501—501.

59
 Return to Freud

exemplary status of the phoneme, Jakobson still tends to derivt

the linguistic priority of the spoken word over all forms ot
inscription, whereas Lacan stresses the letter — a graphic sign
albeit a phonetic one — as the necessary materialization and
localization of the signifier. As in Freud, Lacan thus attributes to
inscription the function of articulation, a fact that considerably
relativizes the great emphasis he places on the parole.
This foregrounding of the importance of writing and its
intimate relation to the signifier leads Lacan to problematiz&'
certain other linguistic concepts he adopts from the Saussurean
heritage of structural linguistics: above all those of "speech" and
of "discourse," although such problems tend to be implied,
rather than explicitly stated. These concepts are no longer, as
in traditional, metaphysical theories of language, conceived as
forms of verbal exchange, as expressions of an identical subject,
or as designations of things. It is not the linearity of speak-
ing that is actualized in discourse, but instead the multi-
dimensionality of the signifying chain formed by the twin
operations of metonymy and metaphor. These two concepts are
originally defined by Jakobson as designating the two functions
of language: the first, the combination of signs, in the dimension
of contiguity or of contextuality; the second, the selection and
substitution of signs, based on relations of similarity. Lacan
does not, however, merely apply these concepts, he plies them,
giving them a new twist in the process. By redefining them as
a movement of signifiers, the contextuality of metonymy and
the similarity of metaphor become a function of differential
opposition and cease to depend upon the signified. The only
contiguity that metonymy can therefore count on, is that of the
signifying chain itself. And the sole "similarity" presupposed
by metaphorical substitution is the purely formal similarity
among signifiers.
This rapprochement of Saussure and Jakobson in Lacan's text
does not so much reconcile the two points of view as mutually
dislocate them: Saussure's linear view of the syntagma and the
sign gives way to the ambiguity of metaphorical—metonymical
operations, while Jakobson's naive realism — according to which
metonymy functions "within a predetermined context," instead
of producing contextuality as such — is corrected by recourse
to Saussure's implicit if highly ambivalent privileging of the
6o
 Metonymy and metaphor

signifier. Insofar as the signifier remains the primary concept of

language, Lacan can state, in a later radio interview, published
under the title, "Radiophonie," that metaphor and metonymy
are "places where Saussure genuit Jakobson:"3 that is, where
Saussure begets Jakobson but also besets (gene) and troubles
(twit) him. Literally, then, the statement, "Saussure genuit
Jakobson," resounds, radiophonically, as: "Saussure gene (et)
nuit a Jakobson." How, then, do things stand with the trouble-
maker himself?
It can equally be said that Saussure genuit Lacan. The
radicalization of the signifier as the constitutive element of
language, throws into question the distinction between meta-
phor and metonymy — at least in the form in which they
have heretofore been considered. Lacan's example taken from
traditional rhetoric, thirty sails for the flotilla, is not only a
"combination," but a substitution as well: not simply a "word
for word," but rather "one word for another" (Un mot pour un
autre), which is the definition Lacan gives of "metaphor." Every
operation of the signifier consists of both substitution and
concatenation, each depending upon the other, since the
signifier is determined only through its relationship to its
surroundings, and more precisely, through exchanges that
define linguistic "value" as a function of substitution. In his
seminar, "The formations of the unconscious," Lacan's remark
upon this is rendered by Pontalis as follows:
What characterizes the signifier is not that it can replace an object
but rather that it can become its own substitute, which presup-
poses a concatenation and a law that governs the signifiers.4
The search for this law and the attempt to define it in a way
that would not make it contingent upon substantial factors
(and thereby exclude it from being the law of the signifier), in-
creasingly comes to determine the direction of Lacan's think-
ing. In "The instance of the letter," Lacan's initial use of an
algorithmic model determines the formalization of metaphor
"Et pourquoi Saussure se serait-il rendu compte, pour emprunter les termes
de votre citation, mieux que Freud lui-même de ce que Freud anticipait,
notamment Ia metaphore et Ia metonymie lacaniennes, Ileux oü Saussure
genuit Jakobson." Scilicet 2/3, '97°, p. 58.
"Les formations de l'inconscient", in the Bulletin de Psychologie XII/4, 1958—59,
p. 251.

6x
 Return to Freud

and metonymy in the direction of a mathematical topology that

plays an ever-increasing role in his later writings.
In order, then, to be able to read Lacan in that future anterior,
which he designates as the temporality not only of the discourse
of the unconscious, but also of his own discourse, we must
strive to read his own texts no less metonymically than they
read the unconscious.
To do this, however, the concepts and intellectual habits
associated with an Euclidean notion of space, as a homogeneous
and empty continuum — in which two bodies cannot occupy the
same place at the same time — are only of very limited value. For
the laws of the signifier, like those of the unconscious,
presuppose a very different kind of space. The Saussurean bar,
on the other hand, is Euclidean through and through: it
separates two internally coherent spheres and thus allows us to
conceive of a movement between and within them, without,
however, putting their respective identities into question. In
contrast, the Lacanian bar traces the impossible trajectory of the
signifier, precipitated into the realm of the signified, which only
takes shape as a result.
Lacan's radicalization of the principle of difference thereby
deconstructs not only the realm of the signified, but also — at
least implicitly — that of the signifier, since every realm as a
"closed" order must first of all be articulated as such, and
this articulation can be understood only as an "effect of the
signifier". That Lacan should retain a spatial notion here, in
separating signifier and signified, is not surprising, for his
primary concern at this stage in his thinking is to emphasize the
radical disparity of the two orders. Nevertheless, the ultimate
consequence of this kind of spatial separation would be nothing
other than the transformation of the signifier into its own
signified and its consequent hypostatizing as a self-identical
transcendental principle.
Indeed, just such an interpretation of the signifier was ad-
vanced in an article on "The Unconscious" written by two well-
known psychoanalysts who studied with Lacan, Jean Laplanche
and Serge Leclaire.5 There, they claim that a signifier can sig-

"L'inconscient, une étude psychanalytique," Actes du Colloque de Bonneval, D.

de Brouwer, 1966.

6z
 Metonymy and metaphor

order subsequently to ontologize the unconscious

nify itself — in
as the precondition of language. Lacan sharply criticized this
interpretation6 and we will have an opportunity later, in the
context of our discussion of Lacan's concept of the unconscious,
to examine the question more closely. What needs to be stressed
here, is that only those readings of Lacan's texts guided by the
"metonymic movement of signification" — i.e. its contextuality —
will do them justice. This means not only taking Lacan "at his
word," but moreover taking him literally, which is not necessarily
the same, any more than a "letter" is the same as a "word."
Indeed, the difference between these two kinds of readings may
well turn out to be the decisive question of hermeneutics today.
To take a text at its word is to assume that signifier and signified
are united, as it is precisely this unity that defines the word as
such. Thus, to construe the significance of Lacan's writings
primarily in terms of the meanings his words convey, however
sensitive one may be to their connotations, is to practice reading
in a manner that inevitably presupposes the priority of the
signified over the signifier, even (and often especially) where
one explicitly proclaims the contrary thesis. The form of the
"thesis" as such, i.e. as a proposition, assertion or statement,
inevitably subordinates the differential relation of signification
to its significant "payoff," as the Pragmatists used to call it.
It is just this payoff, however, that Saussure, Freud and Lacan
seek to problematize.
Concretely, this suggests that the term "signifier" — formally
considered, a word — has neither a simple nor a clearly
determinate meaning, since what it designates and points
toward — a configuration of differences — engenders meaning
only retroactively, as the result of the "pointing," as it were. It is
for this reason that Lacan insists on the distinction between
signifier and letter; the significance of this distinction, however,
lies not so much in neatly separating otherwise clear concepts,
as in designating the ineluctability of difference, insofar as it
coincides with the movement of articulation. The signifier is
thus materialized and localized in the letter, which should not
be identified with the signifier, since the latter, strictly speaking,
has no identity, but instead designates the process through
6
In his Preface to: A. Rifflet-Lemaire, Jacques Lacan, Brussels, 1970, p. i8, 404.

63
 Return to Freud

which identity is produced in the first place. If this process

designated by the signifier forms the condition of possibility of
the word, qua meaningful unit, which in turn is an indispens-
able constituent of the concept, the signifier cannot be grasped
in terms of a particular content, but instead can be represented
only formally, by what Lacan calls an "algorithm"; this
unspeakable formula must be written:

If, on the other hand, one attempts to put what is designated

here into words, one could say, in French, "l'incidence du
signifiant sur le signifie."7 With regard to its content this
formulation is still not much clearer or more complete than the
formal algorithm. The decisive word, incidence, decomposes in
an ambiguity that again seems to defy any conceptual subsump-
tion or synthesis: taken literally "incidence" means "falling in,"
"on," or occurring unexpectedly. I would, therefore, propose,
as a tentative translation, or rendition, of the Lacanian phrase,
"the falling out of the signifier with the signified." This has at
least the virtue of suggesting that the signified is a kind of
"fallout" of the signifier, the effect of its decomposition, and
at the same time hinting at the conflictual aspect of their
relationship. Through its falling out, the signifier bears upon the
signified, to be borne in turn by it, in what must be considered
to be an intrinsically uncanny "incident."
Lacan's use of algebraic representations as well as his witty
and playful discursive language games constitute serious
attempts at keeping the "signifying chain" from congealing into
the hypostatized meanings of that essentially denominative,
conceptual, constative discourse we call the language of theory.
By calling attention to the formal processes through which
meanings are generated and articulated, the errant and unpre-
dictable "back and forth" of "dis-cursivity" is associated with a
surplus of significance that no proposition or concept can fully
comprehend. As we have seen, Lacan calls this movement a
result of "metonymy" and "metaphor." The actual function of

This formula, as well as the two following, designating metonymy and

metaphor, are to be found in Sheridan, p. 564, Ecrits, p. 515.

64
 Metonymy and metaphor

the signifier, we recall, is embodied in metonymy, insofar as the

signifier can only be determined as such by being related
differentially to other signifiers, that is by means of the contiguity
of a discontinuous concatenation. The signifier only "is" as an
element of a signifying chain, which in turn is part of a net-
work of such chains. The constitution and reproduction of this
network, as a concatenation of intrinsically meaningless ele-
ments, constitutes the operation of metonymy. Metonymy thus
could be said to actualize the differential articulation of the
signifier. The distinctive particularity of such an occurrence — in
the sense of the falling-out mentioned above — is that its
particularization is always at issue, implicitly at least, since the
metonymic movement depends upon something else, upon
another missing signifier: indeed, upon another scene and
stage, what Freud, in German called "the other scene" of the
unconscious, in order for it to "take place." The place of the
signifying change is always elsewhere, but that elsewhere is also
never simply accessible as such.8 In German, the literal designa-
tion for this would be an Abort, the restroom of Lacan's famous
example. In English, it could be called a dis-location. Put another
way, one could say that metonymy names the Saussurean
"bar," insofar as it dislocates the sign. This formulation
provides a fresh look at the dis-located place of the signifier,
which, however, remains barred. For its place is none other
than that marked by the bar, which divides the sign right down
the middle while at the same time turning it inside out, forcing it
to point elsewhere. Lacan's formula for metonymy:
f(S..
represents this movement, in which the signifier at hand (S)
fulfills its metonymic function only by
(S')
referring to another latent signifier: which determines the end of
one chain, only by pointing toward the beginning of another.
And yet at the same time, that further beginning must be

All this is an effort to find a suitable circumlocution for what in the

original German text was called an Abort: literally, a place that is off-limits, but
in everyday language, a toilet.

65
 Return to Freud

resisted, excluded, suspended or deferred, if anything is to take

place at all. The signified is consequently always excluded or
barred by the bar (—).
We can now attempt to trace how Lacan's formalization 0
metonymy contrasts with that of metaphor. In the latter the
signifier that produces an effect of "significance" is not some-
where else, in a dis-location, but instead appears directly
in the chain itself, albeit as a substitute that takes the place of
another signifier, thereby driving it from the chain, repressing
and supplanting it. As a result of this substitution, the repressed
signifier is banned to the realm of the signified, or more
precisely, to a "place" where it functions as a signified while,
however, remaining a signifier. Only here does significance, or
more precisely, meaning emerge. The bar is cleared (cf. the + in
the formula for metaphor):

S(+)s

Is the dis-location then finally accessible? Hardly, since the

signifier that clears (or limbos under) the bar remains a signifier
and acts as a signified only in order to operate as a signifier both
"in the realm of the signified" and in the signifying chain to
which it is still linked as a determinate absence. We see that this
"metaphorical" operation, as Lacan calls it, is nothing other
than the precipitation of the signifier that produces the signified
and at the same time bars it as long it does not cease to be a
signifier. Metaphor thus produces an effect of meaning, but in
the double meaning of the French sens: sense and direction. If
metonymy can be said to describe the movement, metaphor
provides the sense of direction. But to have a sense of direction is
not necessarily to arrive at one's destination.
We are thus led to the conclusion that both metonymy and
metaphor are "functions of a uniform movement of the
signifier," which, on the one hand, can only function in and
through its concatenation, and on the other, is always depen-
dent upon what is not part of the chain, the signifier to which it
refers. Nevertheless, if the two aspects are necessary and
interdependent, they do not have the same status: the signifier
only becomes a signifier by means of a concatenation, and this

66
 Metonymy and metaphor

would seem to suggest a priority of metonymy over metaphor.9

The effect of meaning (that is, of a determinate signified), pre-
supposes the functioning of the signifier in a chain. We will be
able to examine more closely this priority of metonymy when
we discuss Lacan's theory of desire as essentially metonymic
and his definition of the symptom as metaphoric. But in order to
follow the difficult relationship between metaphor and metony-
my, we must first examine how Lacan uses them to reinterpret
Freud's analysis of the "dreamwork" in the Interpretation of
Dreams.
We will leave aside, for the moment, Lacan's discussion
of the dream as a system of writing, at the very beginning of
this section, in order to proceed to those mechanisms of the
dreamwork in which he rediscovers metaphor and metonymy.
What concerns us here are those processes that Freud called
condensation and displacement. Lacan writes:
Verdichtung, or "condensation," is the structure of the superimposition
of the signifiers, which metaphor takes as its field, and whose name,
condensing in itself the word Dichtung, shows how the mechanism is
con-natural with poetry to the point that it envelops the traditional
function proper to poetry.
In the case of Verschiebung, "displacement," the German term is
closer to the idea of that veering off of signification that we see in
metonymy, and which from its first appearance in Freud is represented
as the most appropriate means used by the unconscious to foil
censorship.'°
If we examine Freud's concept of condensation, we find at
first not so much the idea of substitution as that of accumulation
or "compression,"11 and it is significant that in his definition
Lacan speaks not of substitution but rather of surim position,

"It should be noted that Lacoue-Labarthe and Nancy arrive at an opposite

conclusion; according to them, Lacan's thought is marked by "the bias of a
preference accorded (against difference, in short) to metaphor, including the
choice of the paradigmatic (vertical) axis of language against syntagmatic
linearity" (Le titre de Ia lettre, p. '4N, note). The question around which such
an evaluation turns, may be formulated as follows: To what extent does
Lacan's writing imply a metalinguistic position for its own propositions? To
what extent, in short, should Lacan be read as a theoretical discourse? The
analyses in this book constitute an attempt to address this question (which,
however, remains largely implicit throughout).
"Sheridan, p. i6o; Ecrifs, p. 511. "SE, iv, p.

67
 Return to Freud

superimposition. A single idea or representation serves as the

nodal point of different associative chains; considered from an
economic perspective, this idea unites in itself the energetic
cathexis of the chains with which it is in contact. This process
takes place as much in the latent dream-thoughts and repressed
wishes as in the manifest dream-content. Not only dreams, but
also jokes, symptoms, and slips, are structured by this mechan-
ism of condensation. As one of the forms of unconscious
articulation, condensation can be employed in all expressions of
the unconscious, insofar as these expressions are necessarily
"overdetermmed." Nevertheless, although Freud discusses
condensation in terms of compression and concentration, it is
clear that this process can take place only by virtue of an
exchangeability of representations: the condensed representa-
tion — be it a nodal point, a composite person, or a montage —
replaces other ideas, which it excludes from the manifest chain
and yet which it at the same time is bound to allude to.
Overdetermination, which is a distinctive aspect of condensa-
tion, is conceivable only as a relationship of substitution, even
if substitution alone does not seem to describe the compactness
of condensation adequately. This difficulty should no longer
surprise us, however, since we have already seen that substitu-
tion alone does not suffice to distinguish Lacan's conception of
metaphor, since it applies equally to metonymy.
One of the questions raised by this problem, therefore, is
whether the priority of metonymy that is implicit in Lacan's
theory of the signifier, is also to be found in the writings of
Freud, in particular with respect to his notion of "displace-
ment." To respond, we must first differentiate various mean-
ings in the Freudian use of the term. In the Interpretation of
Dreams, displacement primarily has to do with the way elements
of the dream-thoughts relate to those of the dream-content.
Freud writes:
What is clearly the essential content of the dream-thoughts need not be
represented in the dream at all. The dream is, as it were,
centered from the dream-thoughts — its content has different elements
as its central point. 12

12 SE, iv, p. 305.

68
 Metonymy and metaphor

In thus speaking of the de-centering of the dream-content,

Freud seeks to account for the way the demands of censorship
are met while still allowing the unconscious wish to be fulfilled.
The narrowness of this interpretation makes it possible for
Freud to speak of dreams in which no displacement occurs, as
in the dream of Irma's injection, where "the different elements
are able to retain.. . the approximate place which they occupy
in the dream-thoughts." Yet as soon as we turn to Freud's
metapsychological discussion of the dream-mechanisms, we see
that there is a second, more general concept of displacement at
work in his writing, one that is presupposed by the notion of a
"primary process" of thought, and hence, of the unconscious.
What is decisive here is the displaceabiity — or as Freud also
writes, the Ubertragbarkeit: the transferability and translatability,
infectiousness, and contagiousness — of psychic energy, of
ideational cathexes, whose instability constitutes the difference
between primary (unconscious) and secondary (preconscious-
conscious) processes. The reason for this instability of cathexes
is to be sought in the structure of the wish and of the pleasure
principle (or as Freud still calls it in The Interpretation of Dreams,
the "unpleasure principle"). We will have occasion later to
return to this structure and to discuss it in greater detail, but
here it will be sufficient to indicate the universality of displace-
ment as an essential mechanism of the primary process in order
to show that, considered in this sense, displacement is at the
basis of all dream-distortion, that is, at the basis of the entire
dreamwork. Both condensation and displacement in the
narrow sense are only made possible by the displaceability of
cathexes.
The structure of this displaceability is nothing other than that
of the metonymic movement of the signifier. With this in mind,
itis possible to put in perspective a criticism often made of
Lacan: namely, that the structuralist interpretation of psycho-
analysis neglects or even ignores the economic aspect of the
drives. The drive, which according to Freud represents "a con-
cept straddling the psychological and the somatic,"3 can be
determined only in terms of a structure of representation: to
proceed otherwise would be to abandon the liminal position of

""Drives and their vicissitudes," SE, xlv, pp.

69
 Return to Freud

what Freud calls Trieb, situated between the physiological and

consciousness, in favor of biologism. However, in determining
this structure of representation, Freud himself wavers between
two versions, or rather, two emphases: on the one hand, the
drive itself is described as being a "psychic representative
[Reprasentantj of a stimulus emanating from within the organ-
ism," albeit one that defines itself precisely by "making its way
into the psyche,"4 and thereby distancing itself from its physio-
logical origins; on the other hand, the drive is described as itsel
a Vorstellung, a "representation," or more precisely, by a Vorstel-
lungsrepräsentanz, a term forged by Freud himself, extremeR
difficult to render in English, and generally translated as:
"ideational representative":
A drive can never become an object of consciousness, only th
Vorstellung that represents it can be consciously perceived. In tht
unconscious as well, it is only by means of the Vorstellung that lth&
drive] can be represented.'5
Freud's difficulty in deciding whether the drive is a repre-
sentative, or that which is represented, is rooted in the
ambivalent nature of the concept of drive, which entails both
aspects. Our discussion of the process of signification can be
helpful in determining their configuration: the drive is deter-
mined by Freud as the movement of representation qua
signification; that is, insofar as it is not grounded in a predeter-
mined signified (the traditional notion of representation), but
instead in a differential relation of signifiers. That Freud's
concept of the drive can best be construed in these terms, may
be demonstrated by recalling his approach to the categories of
quality and quantity in this context. Freud's insistence on the
quantitative aspect of the drive is often summarily dismissed as
evidence of an outmoded scientific bias. Such a view, however,
misses the significance of Freud's quantitative notion of psychic
economy: far from trying to make the drive into something
quantifiable, Freud's insistence on the quantitative aspect of
psychic energy defines its distinctive articulation as a repre-
sentation that is not constituted by its qualitative aspect, that is
Ibid.
15
"The Unconscious," SE, xiv, p. 577.

70
 Metonymy and metaphor

by its "ideational content," but rather by its incommensurability

to any qualitative object. This is what distinguishes what Freud
i',iIls the "primary process," from the more familiar "secondary"
thought processes. Whereas the latter is "qualitative" in nature,
the former is determined by relations that are essentially
differential: by the so-called "pleasure principle," consisting of
"reductions in tension." The latter are only "quantitative" in
the sense of being irreducible to qualities, i.e. to identities:
they are not quantitative in the sense of being quantifiable. Such
differential relations function according to the laws of the
signifier, as Lacan describes them. In Totem and Taboo, Freud
brings these laws together under what he calls, in German,
Iteruhrung, that is, touch, contact or contiguity:
It is further to be noticed that the two principles of association —
'.imilarity and contiguity —are both included in the more comprehen-
'4ve unity of "contact." Association by contiguity is contact in the literal
•.ense; association by similarity is contact in the metaphorical sense.'6
If "association by contiguity" might be understood as corres-
ponding to a limited concept of metonymy, the more compre-
hensive concept of contact suggests a broader interpretation of
Inetonymy, entailing nothing less than "the actual function of
the signifier," its discontinuous concatenation. The manner in
which contact contributes to the formation of ideational contents
described by Freud with the help of what he calls "perceptual
identities." This notion is developed as an attempt to account
br the distinctive plasticity of the dream. Freud conjectures that
perceptions first become identities by means of their essentially
bortuitous convergence — that is contact — with an "experience of
satisfaction"; the association of a particular perception with such
an experience then recurs at the next sign of distress. Insofar as
.i perception is experienced together with a satisfaction, it will
bend to be repeated — that is re-presented — whenever satisfac-
lion is lacking. Through such iterability, the perception acquires
a certain degree of identity, which in turn, however,
is linked to a series — or chain — of other perceptions, as Freud
makes clear:

Totem and Taboo, SE, x,uu, p. 85.

7J
 Return to Freud

Our perceptions prove also to be connected with one another in

memory, and this is especially so if they originally occurred simul-
taneously. We call this the fact of association.'7
Perceptual identity, the condition of all ideational content,
shows itself to be in turn an effect of a process of repetition based
on a fortuitous encounter: on contact and contiguity. The place,
however, that makes contact and contiguity possible — the place,
where such an encounter takes place — is itself determined not
by the "quality" or identity of the elements encountering each
other, but rather by the "quantitative" variations in tension, that
is by the relations that constitute the pleasure (or
more precisely: unpleasure) principle. What Freud attempts to
retrace in his account of the formation of "perceptual identity,"
is thus nothing other than the determination of the signifier.
We are perhaps better prepared at this point to turn our
attention to the third mechanism of the dreamwork, which,
Lacan asserts, constitutes the distinctive trait of dream lan-
guage: its "consideration of representability" [Rucksicht au/
Darstellbarkeit]. Lacan takes some liberties in translating it as
"consideration of the means of staging" (egards aux moyens de lu
mise en scene).18 The aptness of this translation, however, lies in
the emphasis it puts on the theatrical: for what Freud describes as
the Darstellbarkeit of the dream involves less a representation,
a Vorstellung, in the sense of a mental image, than what in
German is condensed into the word, Entstellung. This word
normally signifies "distortion,"9 but it literally, and etymologi
cally, also suggests a shift in place, a dis-placement, such as that
we have already encountered in the movement of the signifier
Above all, however, the term points us toward the precis
difference between a mental representation and Freud's notion
of Darstellbarkeit: whereas the former seems to imply only
binary relation between an object represented and a represent
ing subject, the latter, replacing the "Vor-" of Vorstellung
Dar-, "there", opens what might be called a more public
Interpretation of Dreams, SE, v, p. 539.
18
Sheridan, p. ,6r; Ecrits, p. six.
Chapter of the Interpretation of Dreams is entitled, "The distortion of dream.
(Die Traumentstellung), and must be understood as designating not simply
effect of the dream, but the dream itself. To be sure, the structure of the drv.lII
cannot be entirely separated from its effects, a fact that underscores ii
pragmatic aspect of the Freudian unconscious.

7,-
 Metonymy and metaphor

involving relations more complex than mere binary opposition.

If the dream involves distortion, Entstellung, it is because of the
intervention of a third party — which Freud calls "censorship" —
in the figuration of the dream, a party that plays the role both of
spectator and of judge in the dream-representation. And it is
precisely this that distinguishes theatrical representation from
other forms: the explicit reference to a third party, to an
addressee, and hence, a structure that is at least triadic. This is
why the plasticity of the signifier of the dream, as described by
Freud, does not have the re-presentative structure of a simple,
mimetic image, but rather that of a scene or of a scenario; and it is
also why that scene also inevitably resorts to forms of inscription.
In this context Freud employs the term pictographic
or ideogrammatic writing:
The dream-content is, as it were, given in a pictographic script
whose characters [Zeichen] must be transposed individual-
ly into the language of the dream-thought. If we attempted to read
these characters according to their pictorial value instead of according
to their semiotic relations, we should clearly be led into error.2°
While Lacan's translation of Darsteilbarkeit as mise en scene
emphasizes the scenic quality of dream-distortion, his explana-
tion of this translation underscores the necessity of that scene
becoming a scenography, that is, a writing system. If Lacan
refuses to render Freud's term as the French figuration, it is
because Darsteilbarkeit "takes place within a writing system" and
thus clearly distinguishes itself from figuration in general, be it
even that of "a figurative semiology."2' The written structure of
the dream language sets itself off not only from the representa-
tive image, but also from the spoken word. Speech and word
only provide the dreamwork with "material" with which to
•%rticulate its distortions. Freud comments:
A word, as the nodal point of a number of ideas, possesses, as it were,
.i predestined polyvalence, and the neuroses (obsessions, phobias)

SE, Iv, p. 277. Since the pictography of the dream is also distorted, Freud goes
on in the next sentence to compare it with a "picture puzzle," a "rebus."
It should be noted that Lacan's insistence upon the non-phonetic, scriptural
quality of the dream converges here with an emphasis upon the latter's
theatricality. The theatrical aspect of articulation tends to emerge wherever
1'honocentric conceptions of language are no longer taken for granted, as
I )errida has shown in regard to Artaud.

73
 Return to Freud

take advantage of the opportunities for condensation and disguise

afforded by words quite as eagerly as do dreams.u
According to the law of the signifier, all elements of the
dream — whether images, words, utterances, or their syntactical
arrangement"— only become significant as nodal points or as
"determinatives," that is, through their contextual relations:
Perhaps the dreamer will say: "My mother was there too" (Stekel).
Such an element of the dream-content is then comparable to a
determinative in hieroglyphic script, which is not meant to be
expressed, but is intended only to explain another sign.u
Although Freud seems to present the "determinative" here as
comprising a special case of the dream langauge, it should be
considered to be its exemplary element. It does not signify
by means of its expressive value, but instead solely through
its relation to other signs which it serves to "explain." The
constitutive function of a shifting context also rules out the
possibility of a dream interpreter ever being able to depend on a
symbol or code book that might, as it were, serve as a universal
key to deciphering dream symbolism. "These (symbols) often
possess many and varied meanings, so that, as in Chinese
script, only the context can furnish the correct meaning."24 This
also implies, however, that the shifting context of the dream can
never be completely determined or exhaustively interpreted,
since it itself is only part of a more comprehensive context that
must in principle be impossible to close. If the impossibility of
closure suggests — as an image — something like an infinite
expansion, the figure that Freud uses to name this aspect of the
dream shows that such openness leads not only to the periphery
of the dream, but also to its most intimate center:
Even in the best interpreted dreams, there is often a place that must be
left in the dark, because in the process of interpreting, one notices a
tangle of dream-thoughts arising, which resists unravelling, but has
also made no further contributions to the dream-content. This, then, is
the navel of the dream, the place where it straddles the
22 SE, v, pp. 340—34 i. In this context Freud refers to the use of verbal wit in
dreams.
23 SE. iv, p. 311.
SE, v, 353.
25 SE, p. 525. 1 have discussed this passage at greater length in The Legend of
V.
Freud, University of Minneapolis Press: Minneapolis, 1982, pp. 75ff.

74
 Metonymy and metaphor

If the meaning of the dream requires a certain interpretation

in order to be articulated, it is apparent that the latter cannot
be conceived as entirely detached from what it is interpreting;
interpretation itself cannot avoid being contaminated by the
sense-distorting articulation of the dream. This fact leads Freud
to introduce another figure in describing the peculiar imbrica-
tion of interpretation and text in the dream, one which may well
exemplify the peculiar structure of that relationship: the palimp-
sest. Commenting on the use of this image by the English
psychologist, Sully, to describe dreams, Freud asserts that "no
statement found in the literature [on dreams] so closely
my own account."26
If the dream "itself" already consists of a system of superim-
posed inscription, its interpretation adds a new layer to the
dream-text. The latter, as Freud insists, is not merely the
representation of a desire, but its "fulfillment," however
distorted. It is, therefore, as "wish" that the palimpsest receives
its contours and frame, which is why the discussion of the
dream as signifying text leads us to the question of the subject
its desire.
"
If Freud identifies Sully as his closest predecessor in the exploration of
dreams, it is because "he was more firmly convinced, perhaps, than any other
psychologist that dreams have a disguised meaning." (SE, iv, p. 6o). The
following passage from Sully's article, "Dreams as a revelation," cited by
Freud, emphasizes precisely the refusal to regard dreams as absurd: "It would
seem, then, after all, that dreams are not the utter nonsense they have been
said to be by such authorities as Chaucer, Shakespeare and Milton. The
chaotic aggregations of our nightfancy have a significance and communicate
new knowledge. Like some letter in cipher, the dream-inscription when scrutinized
closely loses its first look of balderdash and takes on the aspect of a serious intelligible
message. Or, to vary the figure slightly, we may say that, like some palimpsest, the
dream discloses beneath its worthless surface-characters traces of an old and precious
communication." (SE, iv, p. 135 note; emphasis Freud's)

75
 6

Spades and hearts: the subject

as stylus

Until now we have considered the laws governing the move-

ment of the signifier in terms of a metonymic string that in
principle is never closed, since it produces determinate mean-
ing, and hence closure, as a retroactive effect, thereby rendering
meaning a function of the string. Can one therefore ever hope,
from this point of view, to ascertain regularities by which
metonymic contexts themselves are determined? Or to put it
even more radically: are the "contiguity" or "contact" presup-
posed by metonymy even thinkable without some recourse to a
grounding signified functioning as a principle of closure?
Both contiguity and contact necessarily require some sort of
site, delimited in some way, in order to be able to "take place."
Yet, in order for such a place to be taken, it must first be
delimited, marked out and cordoned off. How is this to be
accomplished, if not by some sort of signified? If, until now, we
have investigated how the signified is produced by the signifier
in unconscious articulation, we now must ask why the move-
ment of the signifier, so resistant to closure, nevertheless
succeeds in delimiting itself or in being delimited. How in short
are particular signifiers and individual signifying chains formed?
Without such delimitation, no single signifier could ever take
place, could ever encounter another, even to demarcate itself
from it; without such delimitation, there would be no place for
either contact, contiguity or difference. It is imperative, then,
that the site of signification be located, and it was such a
necessity that led Freud himself, once he had described the laws
of the dreamwork, to take a further, and fateful step into
darkness:
For it must be clearly understood that the easy and agreeable portion
of our journey lies behind us. Hitherto, unless I am greatly mistaken,

76
 The subject as stylus

all the paths along which we have travelled have led us towards the
light — towards elucidation and fuller understanding. But as soon as
we endeavour to penetrate more deeply into the mental process
involved in dreaming, every path will end in darkness.1
This path into darkness which, in the steps of Freud and
Lacan, we must follow, has nevertheless been on the agenda
for quite some time. It was already inscribed in the ostensibly
simple notion of the dream as wish-fulfillment. In analyzing
the dream not simply as wish-fulfillment, but also as an almost
inevitably distorted one, which therefore encrypts itself in a
hieroglyphics and hides the key, Freud raised, at least implic-
itly, the question: a wish-fulfillment for whom? Who or what
is the subject whose wishes are fulfilled by the dream in this
way? And, perhaps even more significantly: where is its place?
The seventh chapter of The Interpretation of Dreams may be
read as an attempt to outline this place and in so doing, to track
down the subject. In following its trail, Freud stumbles upon
the question of the unconscious. He approaches it cautiously,
indeed, with a certain trepidation. How is one to conceptualize
that which, in its essence, is inaccessible to consciousness? Or,
put somewhat differently: does the unconscious lend itself to
thought? In seeking to respond to such questions, Freud follows
two, not necessarily compatible paths. First, he portrays the
unconscious as the most comprehensive of psychic domains,
implying, as it were, that its very comprehensiveness allows
comprehension by consciousness:
The unconscious is the larger sphere, which includes within it the
smaller sphere of the conscious. Everything conscious has an uncon-
scious preliminary stage; whereas what is unconscious may remain at
that stage and nevertheless claim to be regarded as having the full value
of a psychical process. The unconscious is the true psychical reality
(...)
As "the truly psychical," the unconscious appears, in such
formulations, to be essentially intrasubjective in nature. At
the same time, however, Freud tends to portray the conflicts
and differences between the unconscious and preconscious-
consciousness as intersubjective phenomena, as if they involved
Interpretation of Dreams, v, p. 51 i.
2 SE, V. pp. 61z—613.

77
 Return to Freud

entirely distinct persons, something like a master and a slave.

And when he writes of an unconscious "intention," this might
seem to imply that the unconscious is a kind of second ego,
insofar as "intentionality" is generally closely associated with
self-consciousness. His initial division of the psychic "apparatus"
into dynamic, conflictual instances or systems, leaves the
question of the structure of subjectivity to which both belong
unanswered. Nor is this lacuna an accident: before Freud, the
subject had been largely identified with self-consciousness,
the instance that comprises the "smaller circle" of the psyche;
hence, for its "larger circle" no theoretical account was readily
forthcoming. The theoretical discourse that would be adequate
to the subject of the unconscious had to be invented by Freud,
although the task was hardly facilitated by his highly ambivalent
relationship to those philosophical predecessors who might
have aided him in this task, in particular, Nietzsche.
Thus, when Freud claims that the dream is not only a wish-
fulfillment, but also utterly "egoistic,"3 it cannot be taken for
granted that we know just which ego he is referring to. Certainly
not to the ego that coincides more or less with the system of
perception-consciousness — that is, with the secondary process.
In this connection, a passage from The Introductory Lectures on
Psychoanalysis, which Freud later inserted as a footnote into The
Interpretation of Dreams, is helpful:
A second factor, which is much more important and far-reaching, but
which is equally overlooked by laymen is the following. No doubt
wish-fulfillment must bring pleasure; but the question then arises, "Tu
whom?" To the person who has the wish, of course. But, as we know,
dreamer's relation to his wishes is a quite peculiar one. He repudiates
them and censors them — in short: he doesn't like them. Their fulfill
ment can therefore give him no pleasure, but just the opposite; and
experience shows that this opposite appears in the form of anxiety.
something still to be explained. Thus, in his relation to his dream
wishes, a dreamer can only be compared to an amalgamation of two
separate people linked by some important common element. Instead
of enlarging on this, I will remind you of a familiar fairy tale in which
you will find the same solution repeated. A good fairy promises a pow
married couple to fulfill their first three wishes. They are delighted, and
make up their minds to choose their three wishes carefully. But th

SE, V, p. 311.

78
 The subject as stylus

woman lets herself be tempted by the odor of sausages being grilled in

the cottage next door and wishes for a pair. In a flash they are there;
this is the first wish-fulfillment. But the man is furious, and in his
rage wishes that the sausages would hang from his wife's nose. This
happens too; and the sausages are not to be dislodged from their new
position. This is the second wish-fulfillment; but the wish is the man's,
and its fulfillment is most disagreeable for his wife. You know the rest
of the story. Since after all they are in fact one — man and wife — the
third wish can only be that the sausages should be removed from the
woman's nose.4
We should not be distracted by the seemingly intersubjective
character of the scene Freud recounts, since the two "are in
fact one, husband and wife;" and yet this unity is split and
conflictual. Nor is it accidental that the story recounts a split
between husband and wife, in German, between Mann and
Frau, Man and Woman, since it thereby links the conflictual
aspect of difference to sexual difference, where "bisexuality,"
far from naming the reconciliation of this difference, indicates
its ineradicable character. By being "one," "Men" and "Women"
do not become a harmonious unity, but on the contrary, are
inseparably linked in their differences. It is the fairytale version
of the irreconcilable difference touched upon in the quarrel over
the two signs, "Gentlemen" and "Ladies." Nor is the example
itself chosen at random; the woman wishes she had a pair of
"little wieners," the man spitefully complies by having a few
hung on her nose, and both then finally decide to wish them
away, thereby almost returning to their point of departure —
almost, but not quite. Freud's story confronts us with nothing
less than the constitutive relationship between desire and the
phallus — or more exactly, between desire and castration; we
shall have occasion to return to this later. Here, what should be
remarked is that wish-fulfillment, governed as it is by the un-
conscious, comes paradoxically close to converging with "wish-
renunciation" [Wunschversagen]. Husband and wife, in Freud's
story, represent not two separate persons, but a situation that is
"one" in its disunity. Fulfillment and renunciation of a wish turn
out to amount to the same thing because the wish, drawing its
energy from unconscious desire, must always be abhorrent to

• SE, V. pp. 580—58!, note.

79
 Return to Freud

consciousness. A wish only becomes unconscious, insofar

it must be repressed. Such repression thus constitutes the stil
ject, insofar as it is a subject of the unconscious. Whence, th
"special relationship" of the dreamer "to his wishes": he 'ii
pudiates them, censors them, in short, he doesn't like them
For such a subject, pleasure and unpleasure coincide.
Thus if what comprises the subject is not simply liii
separation of the unconscious from consciousness (precot
scious), but rather their active exclusion of one other, the sitt
occupied by the subject is riven by divergent drives, since
wish is unconscious without at the same time being fractunti
and split by a striving incompatible with it. If in the
of the early Freud, "pleasure" and "unpleasure" are
exchangeable terms, it is because they designate not scit
identical feelings or states, but a relation of "tension," which i
irreducible to any one of its components.
When, therefore, Freud designates all dreams as
"egoistic," the ego to which he refers is not that of an
unifiable self-consciousness, but rather of a certain dispersiot
and repetition:
It is my experience, and one to which I have found no exception, thu
every dream deals with the dreamer himself. Dreams are
egoistic. Whenever my own ego does not appear in the content of th'
dream, but only some extraneous person, I may safely assume that ni
own ego lies concealed, by identification, behind this other person;
may fill in my ego EIch darf mein Ich erganzen]. On other occasions, whit
my own ego does appear in the dream, the situation in which it occu
may teach me that some other person lies concealed, by identificatiout
behind my ego. [...] Thus, my ego may be represented in a dream
several times over, now directly and now through identification with
extraneous persons.5

Thus, according to Freud all dreams are egoistic, regardless of

whether the ego of the dreamer is directly represented or not
Since, however, according to the law of distortion, the most
important elements of the dream-thoughts seldom appear a
such, there is every reason to think that the ego that appears ii
the dream will only rarely signify the ego of the wish-distorting

SE, IV, pp. 322—323.

8o
 The subject as stylus

Ireamer. Behind this manifest ego, "another person is usually

isncealed by means of identification." When "in the dream-
[. .1 it is not my ego but a strange person who comes
.

I may easily assume that my ego is hidden by means of

"1 may fill in my ego," [Ich darf mein Ich erganzen]
Ireud remarks. The German phrase is notably ambiguous: "to
usfl in my ego" means both to complete my ego, and to complete

'he scene by adding my ego to it, in the place of the non-ego that
But it would be more accurate to say, "I may conceal
md distort my ego," for this is precisely what the subject does
in the dream. And yet, precisely that is what "filling in one's
'go" amounts to: creating the illusion of fullness, of complete-
ness, of the ego as a self-identical subjective instance. If this is
the inevitable illusion of all dream-content, as such, and
independently of its specific signification, then this might
the rather curious "may" in Freud's phrase, "Ich darf
mnein Ich erganzen." The dream indeed allows, permits, but even
snore, it almost obligates the dreamer to "fill in" the ego: that is
the "law" of dream-distortion, of the dream as distortion. And
vet, is this filled-in ego the same as the subject of the dream? In
the light, or shadow, of the dream, are we certain that we know
what an ego is?
Ever since Descartes, modem philosophy has constituted
in part at least, through the attempt to provide an answer
lit this question. The ego is said to be a thinking being which,
'yen if it knows nothing else, knows itself to be thinking. The
is supposed to be this being thinking itself: cogito me cogitate,
reflexive identity of thought itself. Yet at
kast since Freud's great philosophical contemporary, Husserl, a
new specification has been added: the ego is not only an
instance of thought, but an instance of language as well. In his
logical Investigations, (which appeared in 1900, the same year as
I:reud's Interpretation of Dreams), Husserl includes the ego, as
first person pronoun, in that group of expressions which he
describes as "essentially occasional"; while these expressions
may have a "conceptually fixed" [begrzfflich-einheitliche] mean-
ing, they are nevertheless oriented "by the occasion, the
speaker and the situation." Hence, "only by looking to the
•tctual circumstances of utterance can one definite meaning be

8i
 Return to Freud

constituted out of all this mutually connected class for ii

hearer.
The meaning of the word "I" thus "can only be drawn I
living speech and from the perceptual circumstances relevant I
it."7 Accordingly, "I" is intelligible only as an indication of ii
discursive context in which it occurs. Yet, as Husserl stressr.
this indexing function does not coincide with the concnI
meaning of the word, since "otherwise we could
substitute for it the phrase 'whatever speaker is
himself," which is clearly too general to explain the specili
meaning of the word as it is sometimes used. Husserl there It
distinguishes "the universal semantic function [Bedeutungsfu iiA
tion] of the word 'I' to designate whoever is speaking," from it
concrete meaning, which is "essentially realized in the immed i
ate idea [Vorstellung] of one's own personality" as it i

experienced in our own silent thoughts: in the mind's i'

soliloquy" with itself.8 Husserl thereby attempts to take tlii
"essentially occasional" indicative function of the word "I
which refers primarily to the context of linguistic enunciation Ii
the sense of Benveniste's énonciationj, and raise it to ideality tit
thought. The context of the enunciation — the linguistic context
is thereby negated and surpassed, aufgehoben, in that of thought
and of transcendental self-consciousness.
Opposing this effort by Husserl to intellectualize the
tic context, Roman Jakobson defines the status of the word
"I" in purely linguistic terms, in his essay on "Shifters, verbal
categories and the Russian verb."9 Whereas Husserl attempt.
to separate the concrete meaning of the "I" from the linguistit
situation by identifying it with "the immediate representation itt
one's own personality," and thereby with pure thought.
Jakobson emphasizes the word's "general meaning," which, h
asserts, remains the same regardless of changes in context, b'
referring to the process of enunciation — to the "message."
6
E.Husserl, Logical Investigations, vol. i, trans. by J.N. Findlay, New Yorl.
Humanities Press, 1970, P. 315. Cf. Logische Untersuchungen, vol. ii, part
Tubingen: Niemeyer, 1968, p. 8,.
Ibid.
'Ibid. pp. 315—316.
"Shifters, verbal categories and the Russian verb," Roman Jakobson, Select 1

Writings, vol. ii: Word and Language, The Hague: Mouton, '97', pp. 130—14 -
'°Ibid., p. 179.

8z
 The subject as stylus

I lusserl would not have disagreed with this; yet Jakobson then
on to distinguish the specific forms this indexing function
may take. First of all, he divides the linguistic process into
.tatement and enunciation, on the one hand, and into process
md protagonist, on the other. Shifters are distinguished by the
they refer not only to language as that which is
l.ict that
inunciated, but also to the process and the protagonists of
language qua enunciation.
The distinction Jakobson draws between the enunciated and
inunciation will be important for Lacan, not only in his
definition of the subject, but also in his choice of terminology. In
mirder to appreciate fully the importance of this distinction, it will
he helpful to consult briefly two short essays by the French
linguist, Emile Benveniste, which elaborate upon Jakobson's
distinctions. The first essay, "The nature of pronouns," was
published in 1956 in the collection, For Roman Jakobson." In
tontrast to Husserl, and in agreement with Jakobson, Benve-
niste argues that the meaning of the word "I" "can only be
identified by the instance of discourse that contains it and by
that alone."2 Yet this discourse is itself comprehensible only
with respect to the linguistic form "I," which in turn exists only
part of a speech act. Drawing on Jakobson's distinction
between enunciated and enunciation, Benveniste concludes that
"there is a combined double instance in this process: the
instance of 1 as that which refers, and the instance of discourse
containing I as that which is referred to."3 The word "1" thus
entails a double reference: on the one hand, it refers to the
speaker designating himself as part of the content of a particular
statement [enoncel; on the other hand, and at the same time, it
refers to the speaker designating himself as the subject of a more
general process of enunciation that is irreducible to any
determinate statement.
This distinction, although implicit in Benveniste's analyses, is
(lualified by the fact that the linguistic constitution of the subject

"In the meantime, this has been republished in: Emile Benveniste, Problems in
General Linguistics, trans. Mary Elizabeth Meek, Coral Gables: University of
Miami Press, 1971, pp. 217—liz. Cf. in the same volume, "Subjectivity in
Language," pp. 213—230.
° Ibid., p. zr8.
lbid; translation modified.

83
 Return to Freud

described above, relates to the operations of language in the

same way as it does, for instance, in Husserl; that is, language
continues to be construed from the teleological point of view of
appropriation — that of the subject becoming conscious of itself.
A speaker referring to herself as "1," positions herself vis-à-vis
another subject, a "you." Language thereby is conceived as a
medium of intersubjective communication, whereby "language
turns into instances of discourse," thus enabling "the process of
appropriation by the speaker" to take place. 14
Is this "1" — whether determined philosophically by Husserl,
or linguistically by Jakobson and Benveniste, the same as the "I"
which, as Freud writes, is "represented in a multiple manner"
in dreams? Is it really, or merely, an "I" that is "filled in" in
dreams? And does Lacan's adoption or adaptation of linguistic
terms, such as shifter, enunciated/enunciation, involve an
endorsement of the linguistic conception of the "I"?
Let us return to the passage quoted from The Interpretation of
Dreams. After considering the multiple representation of the ego
in the dream, Freud tries to reduce the apparent strangeness of
this process by comparing it to phenomena from waking
thought:
The fact that the dreamer's own ego appears several times, or in several
forms, in a dream is at bottom no more remarkable than the fact that the
ego should occur at various times, in different places and in other
contexts in a conscious thought, for instance, in the sentence, "When I
think of what a healthy child I was."15
The example, however, hardly demonstrates what it is sup-
posed to, since in fact there would be nothing "remarkable"
about sentences an ego could consciously remember — or more
exactly, could pronounce at will. The unconscious "I" of the
dreamer, were it ever to stage such a sentence, would certainly
have something quite different in mind. The example that
would enable us to broach the question of the dreaming "I"
cannot be one of a speaker reflexively positing its own identity,
since this "I" almost always replaces itself with something else —
particularly when it is most immediately concerned. Lacan's
formulation of the question takes this into account and thus
Ibid. p. ho.
Interpretation of Dreams, SE, IV, p. 313.

84
 The subject as stylus

comes much closer to what Freud, in The Interpretation of Dreams,

is trying to articulate:
It is not a question of knowing whether I speak of myself in a way that
conforms to what I am, but rather of knowing whether, when I speak of
this, I am the same as the one of whom I am speaking.16
Whereas the metaphysical tradition — be it in the philosophy
of Husserl, or in the linguistics of Jakobson and Benveniste — has
generally considered the "I" to be the sign of a reflexive and
self-identical subject, it is precisely this self-identity that Lacan,
following Freud, calls into question. Making explicit a question
that remains implicit in Freud, Lacan asks:
Is the place that I occupy as the subject of a signifier concentric or
eccentric with respect to the place I occupy as subject of the signified? —
that is the question. 17

The answer for a philosopher such as Husserl is clear: the "I"

spoken about acquires its veritable meaning only as an expres-
sion of the very same I/ego that is now speaking. Where the
continuity of such a relationship can no longer be taken for
granted, as is the case in written texts, the meaning of the word
"I," is, as Husserl puts it, "alienated": "If we read the word
without knowing who wrote it, we have a word that is, if not
meaningless, at least alienated from one of its normal
meanings."18
As Derrida demonstrates in his pathbreaking reading of this
text,19 it is no accident that Husserl should choose, as his
example of "alienation," a written text, since it is predsely in
II ne s'agit pas de savoir si je pane de moi de facon conforme a ce que je suis,
mais si, quand j'en park, je suis le méme que celui dont je parle." Sheridan,
p. 165; Ecrits, 517.
La place que j'occupe comme sujet du signifiant est-elle, par rapport a celle
7

que j'occupe comme sujet du signifle, concentnique ou excentrique? Voila Ia

question." Sheridan, p.
1-lusserl, Logical Investigations, p. 8z.
Jacques Dernida, Speech and Phenomena trans. David B. Allison, Evanston:
Northwestern University Press, 5973. Derrida demonstrates how Husserl is
compelled here to contradict his own position concerning the ideality of
meaning. For Husserl's "ideality" demands that a meaning be independent of
any actual or possible intuition of an object. The example of a written I, cut off
from its authorial referent, would therefore have to be considered not
"abnormal," but the normative case of ideality. Derrida's conclusion: "My
death is structurally necessary to the utterance, 'I." p. 96. (Translation
slightly modified.)

85
 Return to Freud

writing that the presence of the author is least assured. And yet
this is no less the case in dreams: for if the dreamer articulates
something, he does so not as a speaker, but instead as a scribe.
Speech, we recall, provides the dreamwork only with its raw
materials, as it were. And if the term, "dreamer," suggests a
certain activity, that of scribe — as distinct from "writer," much
less "author" — denotes an operation that cannot be easily
accommodated by the alternative, active/passive. The subject of
the dream "receives" the dream the way a scribe receives the
text to be inscribed. To "have" a dream is to "open" oneself to
impulses which cannot be controlled consciously or voluntarily.
It entails an "active passivity," the readiness to receive and to
retain, but also to follow.
What Husserl therefore describes as an anomaly in the
functioning of the "I" is recognized by Freud as that which
constitutes the norm. This "norm," then — a word, to be sure,
that Freud rarely employs, — entails a certain relation, not to self,
but to the other, an alterity that Lacan will variously describe as
that of the signifier, of enunciation, or of the unconscious. The
instance that sums up the effects of such alterity is what Lacan
calls the "subject": it is defined, quite literally, by being subject
to the other.
While structural linguistics defines the "I" as a shifter, Lacan
demonstrates how this shifting extends far beyond the limits
imposed on it by a linguistics still under the sway of metaphy-
sics. If, as Benveniste writes, the "I" has "no reference
other.. . than the actuality of discourse,"2° then such "actual-
ity," we must conclude, extends beyond the present moment of
discourse — and thereby also beyond the "I": at the same time, in
so extending, it also reaches back to what comes "before." Let us
attempt to retrace the contours of this divided movement. To do
so we will have to make a short detour, by way of Freud's
definition of the "ego"; more precisely, we shall reconsider a
particular aspect of this definition.
This aspect could be called "the linguistic condition of thu
ego." For it is nothing but language, or more exactly, a
particular function of language, that permits the ego to
constitute itself, at least insofar as the ego harbors what Freud
20
Problèmes, p. z6z.

86
 The subject as stylus

describes as preconscious and conscious thought. To be sure,

the ego cannot be simply identified with such thought; but it is
its psychical "home," as it were. The role of language in the
constitution of the ego becomes clear when Freud explains how
it is possible for something to be admitted to consciousness, or
more specifically, how it enters the preconscious. The problem
is dealt with at length in his study of "The Unconscious." In that
text, Freud seems to have a sudden insight, which leads him to
modify his earlier, topological view of the "double inscription"
held to be characteristic of the unconscious:
It strikes us all at once that we now know what the difference is
between a conscious and an unconscious representation. The two are
not, as we supposed, different inscriptions of the same content in
different psychical localities, nor yet different functional states of
cathexis in the same locality; but the conscious representation com-
prises the thing-representation plus the corresponding
word-representation, the unconscious one consists in the thing-
representation alone. The system Ucs contains the thing-cathexes of
the objects, the first and authentic object-cathexes; the system Pcs
originates in a hypercathexis of this thing-representation, linking it to
the word-representations that correspond to it. [...]We can now also
formulate precisely what it is that repression denies to the rejected
representation in transference neurosis: translation into words capable
of remaining attached to the object. The non-verbalized representation,
or the non-cathected act, then remains repressed in the Ucs.21
It might seem at first that for Freud, as in the Gospel of St.
John, "In the beginning was the Word," and that consequently,
language begins with the word and belongs exclusively to the
system, preconscious-consciousness, while the unconscious, by
contrast, is concerned only with images of objects. And yet, as
the discussion of "considerations of representability" already
has suggested, what the unconscious involves is not just
different from such objects and images, but in fact radically
distinguished from them. Perceptual identities, which may be
equated with what Freud refers to here as "thing-repre-
and which provide the unconscious, primary
process with its privileged material, become signifiers through
the processes of condensation and displacement; word-

S. Freud, "The unconscious," SE, xiv, pp.

87
 Return to Freud

representations too are subjected to the same transformation,

insofar as they undergo the elaborations of the primary process.
An example, not from dreams this time, is furnished by a case
of schizophrenia:
Now it is in this respect that the essential difference between the
dreamwork and schizophrenia becomes clear. In the latter, what
becomes the subject of modification by the primary process are the
words themselves in which the preconscious thought was expressed; in
dreams, what are subject to this modification are not the words, but the
thing-presentations to which the words have been
Whether or not the difference singled out by Freud here suffices
to distinguish dream and psychosis is dubious, since, as he
himself acknowledges in The Interpretation of Dreams, it is
entirely characteristic of the dream to submit words to distortion
and transformation through the primary process: one need only
recall the dream in which "the vividly remembered word
Autodidasker" plays such an important More significant,
however, in Freud's description is the implication that what
determines whether or not a representation is admissible to
consciousness or not is its translatability. In the sense that Freud
gives to the term here, this means the ability to translate a
signifier into the signified that belongs to it. It is this notion of
"belonging," of Zugehorigkeit or of Entsprechung, that defines
admissibility to consciousness, and hence, consciousness itself.
And this in turn is tantamount to privileging the "word," at
least as it is traditionally understood — namely, as a unity ol
signified and signifier — as the basic unit of language. The
language of consciousness is thus defined as that of verbal
discourse, in this sense. In the word, so understood, the
movement of signification proceeds from the signified, as its
beginning, traverses the signifier, as its middle — and manifesta-
tion — and comes full circle, ending again in the signified.
In Lacanian terms, the movement of the signifier through
displacement and condensation — through metonymy and
metaphor — is thereby brought to a standstill; unstable,
unconscious cathexes are arrested, psychic energy is bound up or

22
S. Freud, "Metapsychological supplement to the theory of dreams," SE,
p. Z29.
23
Interpretation, SE, iv, Z99.

88
 The subject as stylus

"cathected" to a signified; and the principle of identity asserts

itself — asserting as well the identity of the ego. That identity is
based on the signified, for it is the signified that marks the
arresting of signification — the movement of signifiers — and
makes possible a repetition or fixation of the same. This
repetition of the same authorizes the formation of the apparent
identity of the two "I's" in Freud's phrase, "When I think of
what a healthy child I was." The form of this identity is based,
as we have seen, on the unity of the "I" as a word, having a
fixed, albeit formal, meaning. In the history of Western thought,
this form appears most clearly in the Cartesian cogito ergo sum.
"The philosophical cogito," Lacan writes, "is at the center of the
mirage that renders modern man so sure of being himself even
in his uncertainties about himself."24 Why a mirage? Because
the cogito construes the subject in terms of the signified, rather
than as an effect of the signifier. In one of his most important
essays, "The subversion of the subject and the dialectic of
desire," Lacan makes this clear with unusual directness:
The promotion of consciousness, in the historical aftermath of the
Cartesian cogito, to the essence of the subject, signifies for us [both]
the deceptive accentuation of the transparency of the I in action, at
the expense of the opacity of the signifier that determines it, and
the slippage through which the Bewu/itsein serves to cover up the
confusion of the Selbst (. .

Insofar as language is defined as a system of articulation

governed by the play of signifiers, a subject constituted through
this play can never be reduced to the reflexive identity and
transparency generally associated with the ego. Instead, it must
go the way of the "I" construed as a shifter, or, we could add, as
a drifter, whose only "home" is the rails of the signifying train
(or the metonymic chain). What deranges in this re-inscription
of the subject is that its destiny is no longer simply to follow the

"II n'en reste pas moms que le cogito philosophique est au foyer de ce mirage
qui rend l'homme moderne si sCir d'être soi dans ses incertitudes sur lui-
même Sheridan, p. 165; Ecrits, p. 517.
' "La promotion de Ia conscience comme essentielle au sujet dans La sequelle
historique du cogito cartésien, est pour nous I'accentuation trompeuse de Ia
transparence du Je en acte aux depens de I'opacite du signifiant qui le
determine, et le glissement par quoi le sert a couvrir Ia confusion
du Selbst [.1" Sheridan, p. 307; Ecrits, p. 809.

89
 Return to Freud

rails of the signifier, but rather to be derailed. The differential

nature of the signifier thereby affects the metonymic movement
itself: it does not merely carry the subject somewhere else, but
rather locates the subject in a place where it can never arrive.
The signifying structure of the subject assumes value only
through its position relative to others, within the chain, but also
without. The place of the subject thereby becomes impossible
to demarcate fully, since it is always "there where I am not,
because I cannot situate myself there," as Lacan puts it.26 One
can even go so far as to say that the subject, in the Lacanian
perspective, determines itself in and as this impossibility. The
latter, translated into an image, is called: fader. The subject only
appears, insofar as it fades. The emergence of the subject is its
fading.27 Lacan describes this movement in the following way: "I
am not, wherever I am the plaything of my thought; I think of
what I am, wherever I do not think [that I amJ thinking."28
Not to think that one is thinking, is to think of what one is.
Or rather, of what the "I" is: a mode of thought, that, far from
attaining self-reflexivity, is determined precisely by its destiny
of not thinking that it thinks. A form of thought that is not
self-identical, that only thinks of itself when it is not thinking
that it is thinking, this way of thinking is incompatible with
self-consciousness. Rather, it is the thought-process of the
unconscious, the unconscious as a mode of thought, albeit one
whose essence consists in "the radical eccentricity of the self to
itself." Put in general terms, "the truth discovered by Freud"
can be said to be the truth of a "radical heteronomy."29
Perhaps it is possible to indicate the location, or — for it
amounts to the same — the dislocation of the subject by
examining a particularly telling anecdote recounted and ana-
26
Sheridan, p. 166; Ecrits, p. 5
27
Lacan's notion of "truth," here, seems close to that of Heidegger, as, for
instance, it is discussed in the essay on "The origin of the work of art,"
namely as "the dispute of clearing and concealing" (als der Urstreit von
Lichtung und Verbergung). M. Heidegger, Der Ursprung des Kunstwerks,
Stuttgart: Reclam, 1967, p. 6o. The Heideggerian motif of the (Rifl), could
be compared to the Lacanian notion of Ia béance, the gap, gash or gaping hole
that opens and traverses the "symbolic."
28 "Je ne suis pas, là oft je suis le jouet de ma pensee; je pense a ce que je suis, Ia
oft je ne pense pas penser." Sheridan, p. 166; Ecrits, p. 517—518.
29 "L'excentricité radicale de soi a lui-même a quoi l'homme est affronte,
autrement dit Ia vérité decouverte par Freud (. .j l'hétéronomie radicale
.

Sheridan, pp. 171—172; Ecrits, p. 524.

90
 The subject as stylus

lyzed by Freud. The example I have in mind involves "the

forgetting of proper names," the case of "Signorelli." It was first
discussed in print by Freud in 1898, in "On the psychical mech-
anism of forgetfulness," and was later included, in modified
form, in The Psychopathology of Everyday published in 1904.
Because of the extremely complex nature of the story, I prefer
not to try to summarize it, but rather to quote Freud's account of
it at length:
During my summer holidays, I once took a carriage-drive from the
lovely city of Ragusa to a town nearby in Herzegovina. Conversation
with my companion centered, as was natural, upon the condition of the
two countries (Bosnia and Herzegovina) and upon the character of their
inhabitants. I talked about the various peculiarities of the Turks living
there, as I had heard them described years before by a friend and
colleague who had lived among them for many years as a doctor. A
little later, our conversation turned to the subject of Italy and to
painting, and I had occasion to recommend strongly that my compan-
ion visit Orvieto some time, in order to see the frescoes of the end of
the world and of the Last Judgment, with which a great artist had
decorated one of the chapels in the cathedral there. But the artist's
name escaped me and was not to be recalled. I exerted my powers of
recollection, let all the details of the day I had spent in Orvieto pass
through my memory, and convinced myself that not the smallest part
of it had been obliterated or become indistinct. On the contrary, I was
able to conjure up the paintings more vividly than is usual for me. I saw
the artist's self-portrait before my eyes, with unusual clarity: a serious
face and folded hands — which he has put in a corner of one of the
pictures, next to the portrait of his predecessor in the work, Fra
Angelico da Fiesole; but the artist's name, ordinarily so familiar to me,
remained obstinately in hiding, nor could my travelling companion
help me out. My continued efforts met with no success, beyond
bringing up the names of two other artists, which I knew could not be
the right ones: Botticelli and, in the second place, Bolt raffio.3°
It is only much later that Freud is able to recall the forgotten
name, and only then with the help of a
cultivated Italian who freed me from it by telling me the name:
Signorelli. I was myself able to add the artist's first name, Luca. Soon
my extra sharp memory of the master's features, as depicted in his
portrait, faded away.3'

"On the psychical mechanism of forgetfulness," SE, iii, pp. 290—29!.

31 Ibid., p. 29!.

9!
 Return to Freud

Let us review the situation: the name that Freud is trying to

recall has been forgotten, repressed; in its place, paintings by
the repressed artist appear with a visual intensity that is highly
unusual for Freud's memory. Even the self-portrait of the pain-
ter appears with extraordinary clarity. And finally, the names
of two other painters of the Italian Renaissance, Botticelli, and
the lesser-known Boltrafflo, appear in place of the forgotten
name. Yet as soon as Freud is told the name, Signorelli, "the
extra sharp memory of the master's features, as depicted in his
portrait, faded away." We discover here both the result of a
regression from word-representations to thing-representations,
occasioned by the repression of a proper name, "Signorelli,"
and the effect of the primary process — namely displacement
and condensation; these operations increase the intensity of
the remembered picture far beyond what might otherwise be
expected. As soon as the thing-representation is reunited with
the word-representation, the intensity diminishes. And yet
perhaps most importantly, the mental image does not function
as a mere reproduction of the original picture, but rather as
a significant nodal point at which various metonymic chains
converge. Its role as a signifier is apparent in Freud's explana-
tion. Viewed as a process of repression, forgetting cannot be
explained in terms of any kind of "immanent" meaning that the
forgotten information might contain, since it is not the picture
that Freud forgets — far from it — but rather the name of the
painter. Forgetting, here, can be explained only by considering
the particular significance, the contextual value [Stellenwert] of
the name within the context of the conversation leading up to it:
Shortly before I asked my travelling companion if he had been in
Orvieto, we had been discussing the customs of the Turks living in
Bosnia and Herzegovina. I had related what I heard from a colleague
who was practicing medicine among them, namely that they show full
confidence in the physician and complete submission to fate. When one
is compelled to inform them that there is no help for the patient, they
answer: "Sir [Herr], what is there to say? I know that if he could be
saved, you would save him." It is only in these sentences that we find
the words and names Bosnia, Herzegovina and Herr which can be
inserted in the series of associations between Signorelli and Botticelli -
Boltraffio.32

32
S. Freud, The Psychopathology of Everyday Life, SE, vi, p. 3.

92.
 The subject as stylus

Freud thereby uncovers the linguistic material that the

process of repression works on, but we still do not know the
motivation for the repression. Why "Signorelli"?
I recall that I wished to relate a second anecdote which was close to the
first in my memory. These Turks value sexual pleasure above all else,
and upon encountering sexual disturbances, fall into utter despair, one
which contrasts strangely with their resignation when faced with the
danger of losing their lives. One of my colleague's patients once told
him: "For you know, sir [Herr], if that doesn't work any more, life is no
longer worth living."33
Freud suppresses this anecdote not only because he does not
dare speak about such delicate matters with a stranger, but also
because he himself is particularly preoccupied with the problem
of sex and death at this time.
I was at that time under the after-effects of a message I had received a
few weeks before, during a brief sojourn in Trafoi. A patient on whom I
had spent much effort, had ended his life on account of an incurable
sexual disturbance. I know positively that this melancholy event, and
everything connected with it, did not come to my conscious recollection
on that trip in Herzegovina. However, the similarity between Trafol
and Boltraffio forces me to assume that this reminiscence was at that
time activated, despite the deliberate distraction of my
Freud thus draws the conclusion that his forgetting was not only
motivated, but that the forgotten material itself was significant
only in view of its symbolic function:
I wanted to forget something, I had repressed something. To be sure, I
wished to forget something other than the name of the master of
Orvieto; but this other thought produced an associative connection
between itself and this name, so that my act of volition missed its goal,
and I forgot the one against my will, whereas what I intentionally wished
to do was to forget the other. The disinclination to recall directed it-
self against the one content, the inability to remember appeared in
another.35
This story contains two moments which are of particular
interest here: first, it shows how the mechanism of repression
works as a metonymic—metaphoric movement; and second, in

"Ibid.
Ibid.,
Ibid.,

93
 Return to Freud

so doing, it articulates the structure and motivation of repres-

sion. In order to explicate these processes further, let us refer to
the diagram used by Freud, on which he comments in the
following manner:
The name Signorelli has undergone a division into two pieces. One 01
the pairs of syllables (elli), recurs without alteration in one of the
substitute names, while the other, by means of the translation of Signor
into Herr, has acquired numerous and diverse relations to the names
contained in the repressed theme, but for this very reason is lost to
[conscious] reproduction. Its substitute has been arrived at in a way
that suggests that a displacement along the connected names ol
"Herzegovina and Bosnia" had taken place, without considering the
sense or acoustic demarcation of the syllables. Thus, the names have
been treated in this process like pictograms of a sentence that is to be
transformed into a picture-puzzle (or rebus). Of the whole course 01
events that have in ways like these produced the substitute names
instead of the name Signorelli, no information has been given to
consciousness. Any relation between the theme in which the name
Signorelli occurred, and the theme, the repression of which must have
chronologically preceded it — any such relation, other than that
consisting in the mere recurrence of syllables (or rather of sequences ol
letters), seems at first to be impossible to
Freud's diagram of the incident is shown in Figure

Signor elli tticelli I traffio

Her zegovina and Bo

Trafoi

what is there to be said? etc.

Death and Sexuality

(Repressed Thoughts)
Figure I

4—5 37Ibid., 5

94
 The subject as stylus

Before turning our attention to the apparent absence of a

thematic connection between Signorelli and the motifs of sex
and death that led to the repression, let us first consider the
movement and structure of signification operating here. As
Lacan remarks in his seminar on the "Formations of the
unconscious," the properly metaphorical dimension here is the
absence of the signifier, "Signorelli," in the manifest and
conscious chain, and the substitution of "Botticelli" and "Boltra-
ffio" for it. An effect of sense is produced here; yet the sig-
nifier that falls out as a signified, still functions as a signifier,
although in relation to the substitute formations that replace
it, it has now become a signified. As a signifier, "Signorelli" is
over-determined in at least three ways: first, as a translation of
"Signor" into "Herr," it refers to Her-zegovina and to "Herr"
("what is there to say," and "Herr, if that doesn't work
anymore. . ."); second, as the name of the painter of the "Three
Last Things," it refers to the Day of Judgment; third, it signifies
the instance to which the statement is addressed, the addressee.
Other significant moments, such as the painter's narcissistic
relation both to his image and to his predecessors, might also be
mentioned. Even though these relationships do not appear in
the manifest chain, they may still be considered as being
metonymic in nature, since they are less dependent upon a fixed
meaning than upon a movement of signification. Botticelli and
Boltrafflo refer to Bosnia, and Boltraffio's connection to Trafoi, like
Botticelli's to Signorelli, may also be described as metonymic.
The names acquire signification as vehicles of repetition and
recurrence, and this function is effective even if there is no
connection between the signifiers and the "proper" meaning of
their vehicles.
Let us now turn from the metonymic—metaphoric movement
of the signifier to the problem of repression and of its moti-
vation. This in turn again returns us to our former question,
namely: what is the place of the subject in this process? When
Freud writes, "I thus wanted to forget something, something I
had repressed," who is this "I" and where is it located with
respect to the movement of signification? It would be beyond
the scope of our investigation to attempt an exhaustive response
to this question — if one can be envisaged at all; I will therefore
refer the reader to an article by Anthony Wilden, the English
translator of Lacan's Rome discourse, that examines the matter
95
 Return to Freud

in great detail.38 Nevertheless, what should already be clear is

that the strength of the so-called "death and sex" motifs, and
the effect of the news Freud receives in Trafoi and his patient's
suicide, are intelligible only in a context that is organized "in
terms of the structures of infantile sexuality" (this holds not only
for the case at hand, but also for dreams and in fact for all
expressions of the unconscious). Neither any conscious concern
about the issues of "death" and "sexuality," nor even the shock
of the patient's suicide, can suffice to account for the strength of
the repression; only when these factors are reinforced by the
phantasies of infantile sexuality, reported by Freud in the
context of his "self-analysis," can they elicit such a vigorous
response. The signifier, "Herr," refers both to the role of the
father, and to the problem of the Oedipus complex (sexuality
and death). If one now asks where the subject Freud is in all of
this, the answer, as in the dream, is quite simply, everywhere: the
Herr, both as the physician and as the Creator who depicts
himself in his work; the follower, who portrays (ambivalently)
his predecessors — i.e. his own "Herren" — in his work; but also
in the heart [Herz], since Freud is concerned at this time about a
heart ailment. Everywhere, and yet also: nowhere, i.e. in no single
place, for everywhere here is an alibi, since it elsewhere:

On one occasion (. .) when I was meaning to describe the same small

.

incident to a colleague of mine, the name of my authority for the stories

about Bosnia suddenly escaped me. The reason for this was as follows.
Just before, I had been playing cards. My [Bosnian] authority was called
Pik. Now "Pik" and "Herz" ["Spades" and "Hearts") are two of the
four suits in the deck. Moreover the two words were connected by an
anecdote in which this same person pointed to himself and said: "I'm
not called 'Herz,' but 'Pik.' "Herz" appears in the name "Herzego-
vina," and the heart itself, as a sick bodily organ, played a part in the
thoughts I have described as having been repressed.39
"I'm not called Herz, but Pik": what better description could
there be of the place of the "I" in the metonymic chain of sig-
nifiers; that is, the movement of signification does not come to
an end with hearts and spades, but instead is effectively con-
tinued by them. At the same time, the relation of the "I" — as
Anthony Wilden, "The repression of the signifier," American Imago, 13/t\
pp. 312—366.
"The psychic mechanism of forgetfulness," SE, in, p. 196 note.

96
 The subject as stylus

instance of identity in the subject — to the calling of a proper

name is also manifest: the Ego is the signifier that would call
itself by its proper name, if such were possible.4° But such
"property" is always caught up in the web of language, in
which a name can only signify — that is, name — by demarcating
itself from other, improper names. Not Herz, but Pik; not the
innermost organ of body, symbol of the soul, but an instrument
that punctures: this is a true, and truly improper name, for the
ego.
Through this reading of Freud, one can gain a sense of what
Lacan calls "the radical eccentricity of the self to itself," of the
"radical heteronymy" that betrays its heritage. We also see that
if Freud describes the unconscious as the "core of our Being," at
its heart what we find is a pick and a pike, which do not just
puncture, but also punctuate. At the heart of the subject is a
certain punctuation. As such, it can never in and of itself
constitute "an object of cognition,"41 since it consists in the
interruption and suspension of beings, rather than in their
codification. Such were the aberrations and slips that drew
Freud's attention to dreams, symptoms, jokes, and parapraxes,
none of which is consciously controlled or intended. Which
is why the subject of the unconscious is not the author or
proprietor of "its" language, but rather, the bearer of a name
that is never entirely proper, never quite a proper fit. A remark
of Lacan's about neurosis could be extended to the unconscious
in general:
It does not pose (the question) before the subject (. . .) but in place of it
(a la place du sujet), i.e. it poses it in this place with the subject, as one
poses a problem with a pen. .
The question of the subject thus becomes a question in (the)
stead of the subject, which is neither the author nor the object
40
The game of cards, here, becomes a game of naming, which in Lacanian terms
involves, ultimately, the "name of the father." Since the name of the father
must be able to function in the radical absence of the named — "death and
sexuality," in Freud's words — it is hardly accidental that the only suit suitable
to name the "1" is: spades.
Sheridan, p. 174; Ecrits, p. 516.
42
ne Ia pose pas devant le sujet puisque le sujet ne peut venir a Ia place oü il Ia
pose, mais ills pose a la place du sujet, c'est-à-dire qu'a cette place il pose Ia
question avec le sujet, comme on pose un probleme avec une plume et comme
l'homme antique pensait avec son âme." Sheridan, p. i68, Ecrifs, p. 520.

97
 Return to Freud

of the question, but its vehicle, or more precisely: its stylti

The apparently generic name, "subject," and the ostensihi
"proper" pronoun, "I," thereby emerge as punctuation mark
with all the depth of a pack of cards. Saussure's chess gaiii
has shifted to a game of cards. Between hearts and spad"
however, the stakes have become slightly clearer.

98
 7

The subject as the

imaginary and the symbolic

The polemical moment, the attack and parry style of distin-

guishing one's own position that is so characteristic of Freud,
plays an even more decisive role in the evolution of Lacan's
thinking. Almost from its inception, Lacan's "return" to Freud
is polemically aimed at the psychoanalytic establishment.1
Many issues are at stake in Lacan's controversy with main-
stream psychoanalysis, especially as institutionalized in the
International Psychoanalytical Association: the length of the
psychoanalytic session and the training of analysts being those
explicitly mentioned by the IPA in excluding him from its ranks.
But from a theoretical point of view, the decisive difference that
separates Lacan from the conceptions predominant in the IPA
involves the concept of the subject implied in the Freudian
notion of the unconscious. Looking back upon the development
of his thought, Lacan demonstrates his awareness of the
importance of "strategic" factors, and in particular, as they
relate to this question:
I shall not return here to the function of my "mirror stage," that first
strategic point that I developed in opposition to the favor accorded
in psychoanalytic theory to the supposedly autonomous ego. The
academic restoration of this "autonomous ego" justified my view that a
misunderstanding was involved in any attempt to strengthen the ego in
a type of analysis that took as its criterion of "success" a successful
adaptation to society — a phenomenon of intellectual abdication that
was bound up with the aging of the psychoanalytic group in the
'On the history of Lacan's relationship to the International Psychoanalytical
Association, see the Dossier on the Institutional debate, trans. Jeffrey Mehlman,
October 40 (Spring, 1987), 51—81, and L.a scission de 1953, supplement to
Ornicar?, no. 7, edited by Jacques-Alain Miller, with a note by Lacan, Paris:
1976. Cf. also Elisabeth Roudinesco, L'histoire de Ia psychanalyse en France, vol.
ii, Seuil: Paris, 1986.

99
 Return to Freud

of the question, but its vehicle, or more precisely: its stylus.

The apparently generic name, "subject," and the ostensibly
"proper" pronoun, "I," thereby emerge as punctuation marks,
with all the depth of a pack of cards. Saussure's chess game
has shifted to a game of cards. Between hearts and spades,
however, the stakes have become slightly clearer.

98
 7

The subject as "fader": the

imaginary and the symbolic
+++ + + + +++++ ++++ +

The polemical moment, the attack and parry style of distin-

guishing one's own position that is so characteristic of Freud,
plays an even more decisive role in the evolution of Lacan's
thinking. Almost from its inception, Lacan's "return" to Freud
is polemically aimed at the psychoanalytic establishment.1
Many issues are at stake in Lacan's controversy with main-
stream psychoanalysis, especially as institutionalized in the
International Psychoanalytical Association: the length of the
psychoanalytic session and the training of analysts being those
explicitly mentioned by the IPA in excluding him from its ranks.
But from a theoretical point of view, the decisive difference that
separates Lacan from the conceptions predominant in the IPA
involves the concept of the subject implied in the Freudian
notion of the unconscious. Looking back upon the development
of his thought, Lacan demonstrates his awareness of the
importance of "strategic" factors, and in particular, as they
relate to this question:
I shall not return here to the function of my "mirror stage," that first
strategic point that I developed in opposition to the favor accorded
in psychoanalytic theory to the supposedly autonomous ego. The
academic restoration of this "autonomous ego" justified my view that a
misunderstanding was involved in any attempt to strengthen the ego in
a type of analysis that took as its criterion of "success" a successful
adaptation to society — a phenomenon of intellectual abdication that
was bound up with the aging of the psychoanalytic group in the
'On the history of Lacan's relationship to the International Psychoanalytical
Association, see the Dossier on the Institutional debate, trans. Jeffrey Mehiman,
October 40 (Spring, 1987), 51—81, and La scission de 1953, supplement to
Ornicar?, no. 7, edited by Jacques-Alain Miller, with a note by Lacan, Paris:
5976. Cf. also Elisabeth Roudinesco, L'histoire de la psychanalyse en France, vol.
ii, Seuil: Paris, 1986.

99
 Return to Freud

diaspora of the war, and the reduction of a distinguished practice to a

label suitable for exploiting the "American way of life."2
Such lines make it clear that polemics are indispensable to
Lacan not only as a strategy directed against opponents, but also
as a constitutent of his thought. In its incipient stages, that
thought is characterized by a series of oppositions, designed to
invert prevailing conceptual and axiological hierarchies: subject
of the signifier, and not of the signified; subject of enunciation
and not of the énoncé; subject of the unconscious and not the ego
of self-consciousness. Only in this light does Lacan's insistence
upon the philosophical category of the subject — despite its
dismissal in the thought of Heidegger, with which Lacan was
well acquainted3 — become understandable; his aim is to
disorient and transform the received ideas of psychoanalytic
orthodoxy, and the notion of "subject" provides an alternative
to that of the "self" so highly accredited by that orthodoxy,
although the word, not accidentally, is rarely to be found in the
writings of Freud.
If after the previous chapters we are better prepared to follow
the Lacanian conception of the subject through a new concep-
tual opposition, we should not forget that Lacan's language
must itself be understood in the context of the theory it
elaborates, that of the signifier. Lacan's discourse is, not
surprisingly, itself a signifying practice, which means that the
terms inscribed in it function less as traditional concepts, than as
ciphers in a dream. We recall how the dream's pictorial writing
acquires meaning through the position the dream-elements
occupy in relation to one another, and not through the
particular content they appear to depict.
Lacan's own language must be read in this light: "Our
statements [enoncés] are designed primarily for a function that

2
"Nous ne reprendrons pas ici Ia fonction de notre 'stade du miroir,' point
strategique premier dressé par nous en objection a Ia faveur accordée dans Ia
théorie au pretendu moi autonome, dont Ia restauration academique justiflait le
contresens propose de son renforcement dans une cure désormais déviée vers
un succès adaptatif; phénomene d'abdication mentale, lie au vieillissement du
groupe dans Ia diaspora de Ia guerre, et reduction d'une pratique éminente
a un label propre a I'exploitation de I'American way of Ecrits, pp. 808—809;
Sheridan, p. 306—207.
Lacan translated Heidegger's essay, "Logos," in the first number of the
journal he edited, La Psychanalyse, I, 1956.

100
 The imaginary and the symbolic

they perform only in their place."4 Inasmuch as it is subject to

the signifier, however, this place can be determined only by
means of an alibi, since it can only take place by taking the place
of another, for which it then stands and fills in. Such a place can
never be definitively located, which is why the place of a
signifying element is always the dissimulated place of another
and hence its alibi.
In returning to the mirror-stage, then, it should come as no

"Nos énoncés sont faits premierement pour Ia fonction qu'ils ne rem plissent
qu'a leur place." Ecrits, p. 834. Lacan's theory of enunciation, which is pro-
posed in his discourse explicitly, that is, as statement, as énoncf, can hardly
fail to have consequences for its own mode of articulation. Where, one might
ask, is the proper place of a "statement"? That Lacan himself is quite aware of
this aspect of his thought, is indicated a few lines before the passage just cited:
"Psychoanalysts themselves comprise part of the concept of the unconscious,
since they constitute its addressee. We cannot therefore avoid including our
own discourse on the unconscious in the thesis it enunciates, namely that,
insofar as it is situated in the place of the other, the unconscious must be
sought in all discourse, in its enunciation." ("les psychanalystes font partie
du concept de l'inconscient, puisqu'ils en constituent I'adresse. Nous ne
pouvons des lors ne pas inclure notre discours sur l'inconscient dans La these
méme qu'il énonce, que Ia presence de l'inconscient, pour se situer au lieu de
l'Autre, est a chercher en tout discours, en son énonciation.") For the reading
of Lacan's writings — whether in the Ecrits or elsewhere — certain consequ-
ences result from this, including the following: first, such writings call for a
reading, in the strong sense of the word. Notwithstanding the calculated play
with oral rhetoric in the Ecrits, Lacan's utterances are eminently textual in
character. Not, to be sure, in the sense of an opposition to oral delivery, but
rather in distinction from that language of statement and of proposition, of
énoncé and of the signified, that it is incessantly at work to deflect and derail.
Thus, if one can speak of a priority of enunciation over enunciated in Lacan's
writing, it is because it does not merely utter statements, but also de-scribes
the movement of what it says, i.e. inscribes its statements in its movement.
This description determines and localizes itself as writing: in Freud's analysis
of the pictography of the dream no less than in Lacan's determination of the
letter as localization of the signifier. Whatever divergences separate Lacan's
emphasis on parole, truth and what Lacoue-Labarthe and Nancy have called
his "negative ontology," (Le titre de Ia left re, p. 131) from Derrida's
tug of writing, the motifs of utterance and of the signifier retain considerable
deconstructive force. The latter does not stop short of the notion of text itself.
What in Lacan's writings takes the place of textuality, is theatricality, and in this
respect, it anticipates Derrida's own "pragrammatological turn": each
utterance localized in the text, "in its place," is determined, post facto as it
were — and in this, very much like the dream — by addressees that it did not
necessarily intend. The most explicit model for this is undoubtedly to be
found in Freud's analysis of the function of the "third person," upon which
the joke depends, and which endows it with its social character. It is perhaps
this that Lacan, in the wake of Freud, but also of Heidegger, anticipates most
dearly: a form of address not governed by (conscious) intentionality. This is the
indelible, if inaccessible place of the other in the theater of the unconscious.

101
 Return to Freud

shock that we do not go back to the same place: to the immutable

to pos of an unaltered text, but rather to another place and to
other texts, even if they appear under the same title. "The
mirror stage," Lacan observes in retrospect, "situates the line
dividing the imaginary and the symbolic at that moment in
which [the subject] is seized by an historic inertia (. As we.
.

will see, this claim only holds, if we read the "original" text in
terms of what "it will have been": that is to say, in relation to
two later essays which supplement the earlier one in important
ways. These two texts are, first, an essay written in 1958 in
response to a paper by the French analyst, Daniel Lagache, and
entitled, "Psychoanalysis and the structure of personality"; and
a note, giving a brief sketch of Lacan's intellectual
genealogy: "On our antecedents," written in 1966. Revisited in
this perspective, the text of the mirror-stage provides us — at
least implicitly — with a preliminary account of what Lacan will
subsequently call the "imaginary."
The elements that make up this account, such as the mirror-
image, cannot be considered in isolation, as though they were
intrinsically meaningful. Rather, they can be understood only
within a process of representation that necessarily misrepresents
itself, and in so doing produces a semblance of autonomy. Yet,
in what sense can the mirror stage be said to describe a process
of representation? The child, between the age of six and
eighteen months, as a subject-in-the-making, recognizes itself in
the Gestalt of its mirror image. At a time when motoric control
of its body is still inadequate, the child's powers of perception
are already able to grasp Gestalten, coherent images that
compose a whole. The child's discovery of its reflection provides
it with a model for all future feelings of identity. Such an image
exemplifies an instance that strives to stay the same no matter
how much it may change; it appears as enduring, substantial
and solid. This instance in which the subject seems to be
unified, transparent and identical with itself, develops into the
ego; that is, it develops into the subject of self-consciousness — a
being that strives to be present in, for and to itself. This
"Le stade du miroir donne Ia regle de partage entre I'imaginaire et le
symbolique a ce moment de capture par une inertie historique dont tout ce qui
s'autorise d'être psychologie porte Ia charge, füt-ce par des voies a prétendre
s'en degager." Ecrits, p. 69.

xoz
 The imaginary and the symbolic

in-for-and-to-itself, this self-identical, unified and self-present

subject, is thus the effect of a particular mode of representation:
that of its own (or of a similar) body, perceived and introjected
as a Gestalt that will provide the matrix for all future presence
and identity.
Yet — and this is what makes all the difference in Lacan's
theory — the process of projective identification with an image,
that is: with something other, something external, with some-
thing that appears to be similar or subtly different (the
mirror-image symmetrically inverts whatever it reproduces) —
this process remains a constitutive force and factor in the ego
that develops from it. Thus, although the ego may claim to be
self-identical, although it may strive to subordinate and to
appropriate heterogeneity as its other, it is in fact in constant
rivalry with itself and with everything else; aggressive tension is
fundamental to it. The autonomy of the ego, conceived as an
instance of consciousness, of reality, of perception, or of all of
these at once, is constituted only through a misconstruing
[Verkennung] and denial [Verleugnungj of its structural depen-
dence on others.6
The misconstruing we are dealing with here can be neither
avoided nor reduced to an instance of false consciousness. The
structure of consciousness is not merely misrepresented by the
ego, in order then to be all the better comprehended by it, in
accordance with a dialectical model of absolute knowledge. This
movement would be possible only if the ego actually were what
it claims to be: self-identical. Only then could it hope to know
and recognize itself. Yet if the ego always has and will have
been another, all such attempts to "come into its own," must
inevitably entail misconception and disavowal.
Having thus touched on the structural necessity of this
misconception, let us now turn to its real basis. Lacan's
description of the latter is reminiscent of Aristotle's polemic
6
"Au principe des véritables resistances a quoi on a a faire dans les dedales de ce
qui fleurit de theoretique sur le Moi dans Ia psychanalyse, ii y a le simple ref us
d'admettre que le Moi y sojt en droit qu'il s'avère être dans I'experience:
une fonction de méconnaissance." Ecrits, p. 668. ("At the origin of the
veritable resistances with which one is confronted in the labyrinthian excesses
of psychoanalytical ego-theory, stands the simple refusal to acknowledge that
the I, considered theoretically, is nothing other than what it turns out to be in
experience: a function of misapprehension.")

103
 Return to Freud

ridiculing the inconsistency of the Sophists,7 who, when all is

said and done, still prefer not to jump into an abyss first thing in
the morning. In his "Remarks on the paper given by Daniel
Lagache," in which Lacan discusses at length the significance of
the ego in Freud's second topology, Lacan describes the
resistances that any critical understanding and theory of the ego
— namely as a function of misrepresentation — would inevitably

evoke:
)
Such resistance is nourished by the fact that it is certainly indispensable
to know something of reality in order to survive in it, and that practical
evidence shows us that experience accumulated in the ego, and
particularly in the preconscious, provides us with the surest bearings
for getting around in this reality. Yet one thereby forgets — and it is all
the more surprising that it is psychoanalysts who forget — that this
argument breaks down when it is a question.. . of the effects of the
unconscious. Moreover, the sphere of influence of these effects extends
to the ego itself: it was precisely to affirm this that Freud introduced his
theory of the relations of the ego to the id. Its purpose was thus to
extend the field of our ignorance, not of our knowledge.. 8
The pragmatic demands of self-preservation thus require a
perceptible reality, that is, one that may be identified with itself,
assumed to be present and coherent, in order, in its turn, to
permit the self-preserving subject to identify with itself, and to
identify that Self as a unified and identical ego. What is most
important, however, about this subject of self-preservation is
that it can be constituted only on the basis of a particular
structure, or more precisely, of a particular representation.
Subject and object, ego and reality, self and other, all presup-
pose a form and a matrix of presence in which representation is
conceived as the copy of an original [Abbild eines Lirbildes], as the
sign of referent, the signifier of a signified. It is this very
See above, chapter 3.
"Cette résistance s'appuie sur le fait qu'il faut bien que nous connaissions
quelque chose a Ia réalité pour y subsister, et qu'il est pratique quc
l'experience accumulée dans le Moi, specialement dans le Préconscient, now.
foumit les reperes qui s'y avèrent les plus sürs. On y oublie seulement, et nt
faut-il pas s'étonner que ce soit des psychanalystes qui l'oublient, que cel
argument echoue quand il s'agit. . . des effets de I'lnconscient. Or ces effets
étendent leur empire stir le Moi Iui-même: c'est méme pour l'affirme'r
expressement que Freud a introduit sa théorie des rapports du Moi au
c'est donc pour étendre le champ de notre ignorance, non de notre savoli
[. . Ecrits, p. 668.

104
 The imaginary and the symbolic

interpretation of representation and the notion of the subject

that results from it that Lacan addresses in his theory of the
mirror stage. In so doing, he attempts to point out both the
necessity and the limits of such an interpretation.
According to Lacan, the child's ego is formed:
i. through the child's perception of its own mirror-image, or of
that of someone similar.9
z. through the recognition that this image represents its own or
a similar body.
3. through the projective, "heteropathic" identification with this
other as other in order thereby to anticipate that corporeal-
motoric identity which the child still lacks.
This account has significant implications for a theory of
representation. First and foremost, the logical priority of the
"represented" over the "representative," a priority that defines
traditional conceptions of representation, is called into question.
In the mirror-stage, the representative, the reflected image, in a
certain way produces the "represented": the body "proper,"
conceived as a totality and as the matrix of the ego's identity.
However, if a certain image of the body thus produces identity,
it at the same time also threatens it. For the perception of one's
own body as a whole perpetuates what it dissimulates: the
disunity of the subject, which "returns," as it were, in the
phantasies of physical dismemberment described so vividly by
Melanie Klein.
Insofar, then, as it describes the immutable structure of the
ego, the theory of the mirror-stage suggests that the meta-
physical interpretation of representation entails a conception of
the subject that Lacan calls narcissistic. The subject of the signi-
fied is, ultimately, the narcissistic ego. As such, it is determined
by a form of articulation which dissimulates the play of difference
— that is, that of the signifier — in order to prioritize the signified.

Lacan will later designate this difference-dissimulating form of

articulation as the "imaginary." The imaginary is defined as an
order of representation which misrepresents difference as the
It is therefore in no way indispensable that the child see its own image in the
mirror, in order to traverse the mirror stage. All it needs is to perceive, or
recognize, an image as being similar, as, for instance, the image of the
mother.

105
 Return to Freud

image of identity. One might say that it seeks to impose a ground

upon the groundless "reality" of differential articulation. It is a ii
effort that can never fully succeed, but that in missing the mark
engenders powerful effects.
To be sure, the notion of reality implied in the imaginary
should in no way be confused with Lacan's concept of the "real"
(le reel), which in turn is derived from Freud's notion of "psychit
reality." In Lacan, as also in Peirce, the "real" is defined by
resistance, which includes resistance to representation, includ
ing cognition. It is, therefore, in a certain sense at the furthest
remove from the imaginary. At the same time, one could with
equal justification describe it as residing at the innermost core ut
the imaginary insofar as the latter is constituted by an ambiva I
ence and a conflict that, precisely, resists imaginary
tion, and in so doing goads it on. For the imaginary aims not at
the "real," but at a reality, which would be its self-contained and
autonomous legitimation. Its constitutive incapacity to produce,
or rather: to reproduce such a state results in a highly unstabk'
struggle for power. Lacan's analysis of the imaginary corres
ponds — mutatis mutandis — to Heidegger's account of the "will
to will": both are seen as the necessary but ominous culmination
of the Western tradition of autonomous and constitutive
subjectivity. Like Heidegger, Lacan stresses that the sell
dissimulation of the imaginary, however "unreal" it may be, can
have very powerful and dangerous consequences.
What distinguishes the imaginary, and constitutes its peculia I

"reality," is not dissimulation or estrangement as such, since

as the Heideggerian comparison would suggest — they are
inevitable effects of the signifier. What distinguishes the
imaginary is that it is estranged from this inevitable estrange
ment. For the identity it seductively and alluringly promises is
not simply unattainable, or structurally aporetic: it denies and
dissimulates the repetition to which it is indebted. Which is to
say, it denies its debt to the other.
The result of this denial is an antinomical relation to the other.
Repressed, rejected and subordinated to identity, the debt 01
difference returns, as it were, in a dangerous spiral of destruc-
tive rivalry. If the ego is formed in the image of a mirror, this still
presupposes a minimal but decisive difference between image
and "imaged." It is precisely this difference, after all, that
I 06
 The imaginary and the symbolic

endows the image with its fascinating power. To regard it as

an image of wholeness is to overlook the frame that delimits
its contours, that separates the "figure" not simply from the
"ground," as Gestalt-theory would have it, but from the rest.
But can this rest: everything that remains "outside" the Gestalt,
be reduced to a mere "background"? Can one ever be certain
where the background stops, where the back hits ground.
Lacan's theory of the mirror-stage, which is initially articulated
with the aid of Gestalt psychology, thus tends to put the latter
very much into question.
It is only by means of identifying with another, even if that
other is its "own" mirror-image, that the ego is constituted. Yet
to seek to appropriate that other is tantamount to denying the
difference that makes it a suitable object of identification in the
first place. The imaginary thus becomes a "trap," or a double-
bind: the ego can only emerge by binding itself to the other; but
for it to fulfill its image of wholeness, as a Gestalt, it is bound
to deny the bond that constitutes it. No wonder the ego, as
Lacan sees it, is the subject tied up in knots. Its efforts at self-
affirmation and preservation tend to undermine the relationship
upon which it depends. Far from appropriating the other, qua
mirror-image, the ego is thus permanently suspended in denials
that are ultimately as threatening to itself as to the other denied.
The phenomenon of scapegoating suggests itself as an instance
of such a dual threat.
Such narcissistic ambivalence can be discerned in psycho-
analysis itself. To define the goal of psychoanalysis as the thera-
peutic strengthening of the ego is to place both the practice and
the theory of psychoanalysis increasingly under the self-
destructive sway of the imaginary. This, Lacan concludes, is the
price psychoanalysis has paid for acceptance into the American
establishment. Official psychoanalysis has thereby lost touch
with the symbolic processes that were at the heart of Freud's
discoveries.
Although the "imaginary" order evidently derives from the
"image," it would be a mistake to conceive it as constituting a
separate realm. For this would be to overlook that whatever
coherence and structure may be attributed to the imaginary
must in turn be seen in relation to the signifier as the condition
of all articulation whatsoever. As Lacan notes in his remarks
107
 Return to Freud

on Lagache, "structure is defined by, and in turn defines, the

signifying articulation as such, [...] that is, the effects deter-
mined by the pure and simple combinatorics of the signifier in
the reality where it manifests itself."1°
This reference to Saussure can help to emphasize just how
different the "reality" in which the signifier "manifests itself"
is from its imaginary counterpart. The reality of the imagination
situates the imaginary as "mere" representation. "Reality" is
thereby construed as a domain that both precedes and outlasts
imagination, its origin and end. The reality in which the
symbolic shows itself, by contrast, is that of the showing itself,
not of the show. In its movement of "manifestation," the
signifier is never fully identical with its necessary localization
and materialization — that is, with its realization — but is always
both "more and less," more and less than itself. In coming forth,
the signifier, as we have argued, falls out. The signifier "is" the
falling-out, the fall-out of "being" with "itself." It thus both
exceeds the reality of its manifestation, and at the same time
falls short of it. The signifier falls out with itself as it falls into the
signified. No signifier without a signified. But the signified is
ultimately only the signifier dissimulating its signification. It is
through such dissimulation that meaning takes place. The place
of meaning, or rather: the taking-place of meaning, is (the)
imaginary. The imaginary halts the fall-out of the signifier. But
the halt remains imaginary, for it does not put an end to the
falling-out. The imaginary arrests the fall-out of the signifier, but
only at the cost of falling-out with itself, or rather: with the Self.
If the arresting images of the imaginary order are effectively
inscribed in the force-field of the signifier, the latter would have
neither field nor force without the dissimulation by the
imaginary. Left to its own devices, the symbolic, like the
primary process, would tend to dissolve and to displace the very
determinations upon which it "itself' depends. In short:
without the imaginary, the symbolic would self-destruct. It is
therefore no less dependent on the imaginary than the imagin-
ary is on it, although it is this latter aspect that Lacan chooses to
emphasize, almost to the exclusion of the former. The reason for
"La structure definie par l'articulation signiuiante comme telle [. . a savoir
tes effets que Ia combinatoire pure et simple du signifiant determine dans Ia
réalité oü elle se produit." Ecrits, p. 649.

io8
 The imaginary and the symbolic

this one-sided emphasis is above all strategic: it is the signifying

function of the symbolic that had to be introduced into
psychoanalysis at the time Lacan was writing, whereas the
dissimulation of the imaginary was simply identified with
reality per Se. Nevertheless, to misconstrue their reciprocal
relationship would be, ultimately, only to supplant one ethical
norm, that of the autonomous, if imaginary ego, with another:
the ontological, but no less hypostatized priority of the signifier.
By contrast, the imaginary and the symbolic constitute neither
an ethical opposition nor an ontological hierarchy, but a
differential relationship that disorders each of these "orders," as
Lacan often calls them; each order sets itself apartfrom the other,
but in so doing reveals its dependency upon the other and
thereby sets itself
Such mutual interdependence and undoing of the symbolic
and imaginary recall the relationship between metonymy and
metaphor. If the metonymic movement may be said to consti-
tute the symbolic function "proper," its "slippage" must in turn
be held in check, given direction and hence, distorted by the
metaphorical function if it is not to dissolve into sheer
indeterminacy. In short, metonymic displacement must itself
be dislocated and disfigured — entstellt, to use Freud's term — by
metaphoric condensation, in order to function at all.
To be sure, for Lacan the metaphorical movement belongs to
the symbolic order, and hence, in it the narcissism of the
mirror-stage is subordinated to the movement of the signifier, as
the following passage, from Lacan's essay on "The situation of
psychoanalysis in 1956," asserts:
Without doubt, the imaginary is not simply the illusory, and it supplies
material for the Idea. But what enabled Freud to plumb the depths that
were to enrich his followers was the symbolic determination to which
the imaginary function is subordinated; we are constantly reminded of
this in Freud, whether we are dealing with the mechanism by which
words are forgotten, or with the structure of fetishism.12
I have discussed how such setting apart works in the writings of Freud in The
Legend of Freud, part i.
12 "Sans doute l'imaginaire n'est-il pas l'illusoire et donne-t-il matière a l'idée.
Mais ce qui permit a Freud d'y faire Ia descente au trésor dont ses suivants
furent ennchis, c'est Ia determination symbolique oü Ia fonction imaginaire se
subordonne, et qui chez Freud est toujours rappelee puissamment, qu'il
s'agisse du mécanisme de l'oubli verbal ou de Ia structure du fetichisme."
Ecrits, p. 464.

109
 Return to Freud

We have already discussed the forgetting of proper

and we will come to the phenomenon of fetishism shortly. iti
the moment, however, we are concerned only with
the role of the imaginary within the structure of the symboli
As we have seen, this structure is also that of the unconscious
primary process; its metonymic movement produces
transference and displacements, concatenations but also frag
mentation. Nevertheless, just as the allegedly "primary" pm
cess of the unconscious requires a no less "primary" censorship.
or more precisely, repression, in order to constitute itself as
process, the symbolic needs the imaginary-metaphorical mu
ment of fixation, in order to assure the minimum of
tion necessary for any articulated structure.
Here we touch on the problem of what Freud described as
Urverdrangung, "primal repression." Without recourse to sonit'
such irreducible, initiating but also internally incoherent notion
it is difficult to conceive of the possibility of a drive's "attaching
itself to — that is defining itself in terms of — an ideational
representative; and since "attachment" — also known as cat hexis
— is an essential aspect of the drive as such, the theoretical fiction

of a "primal repression" must be regarded not as an external

"vicissitude" of drives, but as a necessary part of then
destination. Considered in a purely subjective register, this
reciprocal dependency can only appear to be paradoxical and
circular: on the one hand, the unconscious presupposes (primal)
repression as its enabling other (be it in the form of the ego or ul
the super-ego); on the other, primal repression likewise presup
poses the differentiation of the psyche into unconscious
preconscious-consciousness, in order to have a space within
which it can occur in the first place.
Such paradoxes, however, lose something of their logical
absurdity when it is recognized that they are structural effects ot
differential articulation. The apparently vicious circle describes
the necessary form of signification. Meaning — the signified -
is an effect of the signifying chain; yet without some such
"meaning-effect," the chain itself would be unthinkable.
Thus, viewed from the structural perspective of differential
signification, the imaginary and the symbolic must be consi-
dered primarily as forms of articulation. The subject of thy
imaginary, as described in the mirror-stage, is le moi, tht'
I 10
 The imaginary and the symbolic

reflexive ego of self-consciousness. Yet while this subject may

appear to be present and identical, the very process of
identification disrupts and divides it. Insofar as its identity is
determined as self-consciousness, any and all divisions must be
comprehended within the unity of a self; by contrast, the
conception of a subject as a configuration of signifying chains is
a notion that is not compatible with the conception of the
autonomous ego as self-consciousness, however dialectically
conceived. For the symbolic network of signification can never
be totalized, even by dialectical negation. This essential incom-
patibility should be kept in mind when encountering Lacan's
use of dialectical terminology: as is the case whenever he
"borrows" a particular philosophical vocabulary, his use of it is
always commanded by strategic imperatives: that of revealing
an element of irreducible heterogeneity and difference there,
where identity had previously been assumed.13
Independently of its volition or designs the narcissistic ego of
the imaginary signified — of the énoncé, the statement — is thus
reinscribed in the symbolic as the eccentric subject that signs
itself over to the signifier with a slip of the pen. It thereby
subjects itself not merely to the utterance, but in the process of
uttering (l'énonciation), in the sense of the shifter discussed
earlier. This subject of the utterance is not simply diametrically
or symmetrically opposed to the subject of the statement, the
moi. Rather, and in sharp contrast to the purported unity and
identity of this moi, the subject of the uttterance is necessarily
split and suspended between the statement as meaningful
utterance, and the excessive overdeterminations of the signify-

13
Mikkel Borch-Jacobsen has argued, in Lacan, le maître absolu, Flammarion:
Paris 1990, that Lacan's theoretical armature is decisively determined by
Alexandre Kojeve, whose lectures on Hegel (1933—1939) were enormously
influential in France. Although such influence should obviously not be under-
estimated, it cannot, I believe, sufficiently account for the Heideggerian
elements in Lacan's elaboration of the heterological nature of language. In
Lacanian/Sassurian terms, the Hegelian dialectic ultimately subordinates the
signifier to the signified, difference to identity. This explains why Kojeve
himself refuses to acknowledge that the Heideggerian notion of history
constitutes a viable alternative to Hegel. Cf. A. Kojève, introduction a Ia lecture
de Hegel, Gallimard: Paris, p. 575, note: "Heidegger has taken up the
Hegelian themes of death; but he completely neglects those of Struggle and
Work; moreover, his philosophy does not succeed in accounting for History."

III
 Return to Freud

ing chain. Left to its "own" devices — that is, to the metonymic
movement of the signifier — the subject would become a
hopeless drifter; it is as moi that the drift is stopped, but only
through imaginary (dis-)simulation. As shifter, the subject
remains suspiciously shifty.
The subject of the utterance thus emerges not only as a drifter,
always on the run, on the "rails," at home nowhere, but
moreover — to cite the English word used by Lacan — as a
•14 Lacan illustrates this "fading" by referring to certain
so-called "expletives," like the French ne, used primarily in
subjunctive phrases to stress the counterfactual moment of
desire or of affect. An approximate equivalent, in American
English, would be the word, "really?!", meaning everything
and nothing, and confirming, by antiphrasis, the tenuous
"reality" of the communicative process. A counterpart in
conversational German would be the expression "genau!,"
"exactly". Such "expletives"fihl out the fall out of the signifier.15
But as its name indicates, such filling is even shiftier than
the shifter. For the "I" as shifter still would seem to occupy one
place at a time, and thus to be entirely compatible with the
subject of the signified, one whose identity and presence — as
author of a message and as creator of meaning — is never
radically called into question. By contrast, a "filler," such as ne,
is closer to what Freud, in his discussion of dreams, refers to as
"determinatives": markers that have no semantic meaning of
their own, but which function purely syntactically. In this
particular case, ne also alludes to a process of denial or
distancing, which is particularly appropriate for the "fading" of
the subject of the unconscious:
The unconscious, beginning with Freud, is a chain of signifiers which
repeat themselves insistently somewhere (on another stage, he writes),
thereby intervening in the fissures offered it by actual discourse and by
the thinking that it informs.16

Sincethe subject constitutes itself through the movement of the signifier, "it
disappears as subject in the signifiers, for it only becomes" and "is" in and
through the "fading that constitutes its identification." Ecrits, p. 835.
Such "expletives" can also play a decisive role in jokes, as I have sought to
demonstrate; cf. Samuel Weber, "Laughing in the meanwhile," MLN (Fall,
1987), Johns Hopkins University Press, Baltimore 1988, pp. 704—705.
16 "Subversion du sujet", Ecrits,
p. 799; Sheridan, p.

112
 The imaginary and the symbolic

The "fissures" mentioned by Lacan are, as it were, the mater-

ialization of difference in "actual discourse." Difference mater-
ializes as a gap, a fissure, a hole or shadow that interferes with
the semantic progress of discourse, puncturing and punctuating
it. Such interruptions, as Freud emphasized in his discussion
of dreams, constitute the favored markers of the unconscious.
Despite its name, then, an "expletive" like ne does not so much
fill up a hole as mark its borders.17 As with the (in)famous
Watergate tapes, expletives are always deleted from official
declarations, since they are obscene, not only in what they say,
but in their interruption of all meaningful statement. If what is
obscene belongs off-stage — a dubious etymology but neverthe-
less a suggestive one — expletives are obscenity itself: self-
effacing, they are often barely heard, interrupting the flow of
speech, while at the same time engendering the illusion of a
flow by covering up its incoherence. In spoken, communicative
discourse, they are the perfect cover-up, Watergate avant la
lettre. Which is why they are virtually inaudible in ordinary,
spoken language, whose semblance of continuity they assure; it
is only in the merciless rendition of transcription that the full
force of the expletive is revealed (if it is not censored out before
publication).
Perhaps the most prevalent "expletive" of this sort, in spoken
American English, is hardly a word or part of discourse at all, in
the grammatical sense, at least. It is the ubiquitous "uh. . ." that
scans the speech of most Americans, in varying degrees. In so
doing, it functions as an exemplary trace of what constitutes the
subject. The notion of the unconscious is Freud's attempt to
describe this lack. In American spoken English, its most
characteristic expression, perhaps, is the uh.
The structure of the symbolic, the symbolic as structure, thus
stands out in sharp contrast to the imaginary. Symbolic is the
play of differences that manifests itself in and as the concatena-
tion of signifiers. While the imaginary reduces differential

In Zerman, what most dosely corresponds to the French, ne, is, perhaps
surprisingly, the word, ja, as, for instance, in the phrase, "Das habe ich ja
nicht besteilt." The "ja" functions merely to intensify the statement ("I didn't
order thati"), it says "yes" to the statement, which is often a denial, and in so
doing, says "no" to the real or suspected "no" of the other. See: "Laughing in
the Meanwhile," pp. 691—706.

''3
 Return to Freud

articulation to the ostensibly dualistic relationship of representa-

tive and "represented," in which the latter term is presumed to
guarantee preconstituted identity and presence, the symbolic is
representation primarily in the sense of a Darstellung, of a
performance destined for another stage, or more precisely, for a
theater in which the representation simultaneously produces and
dislocates the represented. What is thereby "represented,"
however, is not, as one might expect, the signified, but what
Lacan insists upon calling: the subject. In one of his most
famous formulations, he describes the place of this subject by
demarcating the sign from the signifier. If, as in the celebrated
definition of Peirce, quoted by Lacan, the sign can be said to
represent "something to someone," the "signifying order," by
contrast,
constitutes itself through the fact that a signifier represents a subject
for another signifier. This is the structure of all the formations of the
unconscious, be it the dream, the lapsus, or the joke. It is also the
structure that explains the originaly division of the subject. 18
representation of the subject by the signifier is thus no
The
simple reproduction of something that — virtually or actually —
was already present. Rather, it is a movement of division that
constitutes the subject. Represented by this movement of
signifier to signifier, the subject cannot be suitably articulated in
the present indicative or in any of its modalities, such as the
present perfect. By contrast, the present participal, through
which the subject is determined through its participation in an
on-going (verbal) process, marks the subject's subjection to a
movement of signification. Another aspect of this subjection is
articulated, as already discussed, by the future past. And yet,
there is another tense in French — if indeed it is really a tense at
all — capable of indicating the elusive status of the subject of
signification: the imperfect. In French, the imperfect designates
not merely a "before," which once was, but which no longer is;
by virtue of its incompleteness and imperfection, it can also
articulate a "not yet." In this sense, it functions as what
grammarians call a "past future" (frtur du passé); like the future
18
"Le registre du signifiant s'institue de ce qu'un signifiant represente un sujet
pour un autre signifiant. C'est Ia structure, réve, lapsus et mot d'esprit, de
toutes les formations de l'inconscient. Et c'est aussi celle qui explique Ia
division originaire du sujet." Ecrits, p. 840.

114
 The imaginary and the symbolic

past of the "will have been," this "past future" also implies a
conjectured "reality." Take, for example, the French sentence:
"Un pas de plus, ii était dans la rue," which in English can be
rendered idiomatically only with the far less ambiguous con-
ditional phrase: "One step more and he would have been in the
street," but which literally says: "One step more and he was in
the street." What is decisive here is precisely the undecidability
of this temporal—conjectural distinction, and the fact that it
results from the differential structure of the signifier. As part of
a process of signification, the signifier always anticipates a
signified. At the same time, however, it also refers back; and
inasmuch as it is always more or less determinate, each signifier
derives from, or has its point of departure in, a previously
determined signified.
Unlike the sign in Peirce's definition, then, Lacan's signifier is
based on a distinctive function; rather than being construed in
relation to a central presence or present tense (as is the case of
the sign), the before and after of the signifier are "centered"
around a difference. As a distinctive element, the signifier
always differs from itself and "is" only in this difference: not
merely to other, similarly differential terms, but to itself. To be
identifiable, each signifying mark must be repeated, must be
iterable, as Derrida argues.'9 Its identity is thus imperfrct in the
most "literal" sense: it can never fully take place because its
place is a function of divergent metonymic chains. Which is
why, as Lacan often writes, the signifier takes place "in [the]
place of the other," ("au lieu de l'Autre"): both "in the place of
the Other," and "instead of another." As signifier, language is
"intrinsically" substitutive, and hence, figurative. But since
each figure always gestures toward another figure, the process of
configuration is addressed at a destination it can never attain.
The subject of both the signifier and the symbolic is
necessarily incomplete, and it is there, in the force-field of such
imperfectabiity, that the ostensible identity of the imaginary
ego will have been inscribed, albeit in invisible, self-effacing
characters. The "I" takes place, takes its place in the shadow of
an "uh To retrace the genealogy of this shadow, it may be

'9J. Derrida, Limited Inc., trans. S. Weber, Northwestern University Press:

Evanston,

''5
 Return to Freud

useful to return, once again — assuming, that is, that we ever left
it — to the mirror stage.
The mirror stage appears to be clearly pre- and extra-linguistic
when considered from a genetic perspective, as a develop-
mental "stage." The process of narcissistic identification sets in
before the child has learned to speak, and it appears to occur
independently of language, in the ostensible silence of infancy.
It seems to involve a purely dualistic relationship of the child to
its reflection. Furthermore, the relation Lacan establishes be-
tween the precocity of birth and the belated development of
motor and sensory powers in humans, as compared with other
mammals, would seem itself to betray what Lacan might call an
imaginary mode of argumentation, structured upon the opposi-
tion of the organic and the perceptual. Moreover, the conse-
quences of the mirror stage appear to be conceived in terms of a
dualistic structure: the ego competes with itself as with an alter-
ego. However, the text of the "mirror stage," upon which this
reading is based, turns out itself to be incomplete and imperfect;
what is lacking, or rather, effaced, is precisely that aspect of the
figure which distinguishes it, as signifier, from a Gestalt: its
gesture. It is this that will become — or rather, will have been — its
most decisive moment: the gesture of desire. This aspect
appears in print only in 1958, in an article already mentioned,
Lacan's "Remarks on the paper by Daniel Lagache." Moreover,
it is introduced in such a matter of fact manner, that its omission
from the published paper is almost instantaneously forgotten.
The context is defined by the question of the Other (capital 0):
that is, by the function of alterity or heterogeneity in discourse.
In contrast to Lagache's "personalistic" interpretation of Freud-
ian doctrine, Lacan stresses the impersonal structure "of this
Other, where discourse is situated"; such alterity, he continues,
reaches to "the purest moment of the mirror relation." What is
this "purest moment"? Lacan locates it
in the gesture by which the child at the mirror, turning around to the
person carrying it appeals with a look to the witness who decants, by
verifying it, the recognition of the image from the jubilant assumption,
in which, to be sure, it [such recognition] already was.2°

20
"Car l'Autre oü le discours se place, toujours latent a Ia triangulation qui
consacre cette distance, ne l'est pas tant qu'il ne s'étale jusque dans Ia relation

ix6
 The imaginary and the symbolic

The present participle indicates an ongoing movement: that of

the child, turning around, seeking to "decant": to purify,
separate, decontaminate "the recognition of the image" from its
"jubilant assumption," in which such recognition already was:
was already at work. What is not said here, but what is implicit,
is that in that recognition something else was at work as well,
something capable of adulterating it, and which thereby
requires a confirmation that Lacan compares here with "decant-
ing." A liquid is decanted for at least two reasons: to separate it
from its sediment or precipitate, and, correlatively, to preserve
its homogeneity, or purity. We recall that one of Lacan's favorite
figures for describing the manner in which the signifier falls out
into the signified is precisely that of a certain "precipitation."
The signifier precipitates out as a sediment that attaches to every
signified. This is why the "recognition of the image" that
"already was" at wqrk in its "jubilant assumption," must
necessarily be as imperfect as the tense through which it is
articulated: such recognition "already was" in the "jubilant
assumption," and yet it never was fully there, since it is precisely
the jubilant assumption that makes "verification" necessary.
It is this imperfection of the child's jubilant recognition — its
contamination by something else, something unsettling, that
impels the child to turn around, for such verification can only
come from somewhere else, from another place. It does not take
place in the ostensibly perceptual relationship of the child to the
mirror, of subject to object, or even, despite appearances, of
subject to subject. It takes place in the encounter of look and
look. The look is no longer determined by the object it seems to
perceive: the mirror image, but rather by its encounter with
another look. This turning, from object or image, to look, marks
not merely the movement of the child, but the Kehre of Lacan as
well. In it, the ethological perspective that initially marks the
mirror stage, turns away from the descriptive discourse of
Gestalt psychology and moves toward the theater of the
symbolic. What is now added to the mirror image is the glance of
the other and the gesture of turning around that returns the child
to the signifier.
speculaire en son plus pur moment: dans le geste par quoi l'enfant au miroir,
se retournant vers celui qui Je porte, en appelle du regard au témoin qui
décante, de Ia reconnaissance de l'image, de I'assomption
jubilante, oü certes elle était déjà." Ecrits, p. 678.

"7
 Return to Freud

With the appearance of this gesture, a problem noted in on I

previous discussion of this text begins to be clarified, if fl( It
decanted. In the original version of the text of the mirror stagt'
the effect of the reflection upon the child seemed to result froiii
its recognition of the image as its own likeness. Yet what
mained unclear was just how this recognition could produ t
such jubilation. In the later essay, recognition as such is no
longer enough; instead, in its stead, there appears the
which causes the child to twist back, turning around, and in
gesture, to seek the confirming look of another. Recognition
of the "same," in the guise of a mirror image, is no longci
described simply as a means by which the child seeks to corn
pensate for physiological helplessness. Instead, the jubilant
reaction does not relieve the child from having to seek
something like an acknowledgment of the other. In this sense
recognition is no longer a process organized around two poles
child and mirror image, subject and object. Instead, it emerges
as a triadic relation in which acknowledgment emanates not
from the self-identical ego, but from the "person who carries it,"
that is, from the place of the Other.
At the same time, the twisting and turning of the child marks
a decisive articulation in Lacan's thinking itself. The linear
temporality of a before and after is no longer adequate to
measure the gesture at stake here. The acknowledgment "was" -
and Lacan stresses the imperfect here — "already" at work in the
jubilant reaction (including, perhaps, Lacan's own jubilation at
his discovery of the signification of the mirror stage. . .). That
the tense of this turn should, as I have indicated, be the im-
perfect, suggests that the discovery involves more than simply
meets the eye:
But this "already" should not deceive us about the structure of that
tertiary presence: it owes nothing to the anecdotal figure that incarnates
it.2'

In the later text, "Of our antecedents," dealing with the same
problem, Lacan stresses the fact that the personal identity of the
other is a matter of indifference; what is important in the
21 "Mais ce déjà ne dolt pas nous tromper sur Ia structure de Ia presence qui
est ici évoquée en tiers: elle ne dolt nen a I'anecdote du personnage qui
l'incarne." Ibid.

i i8
 The imaginary and the symbolic

production oithis reaction is the sheer fact of "being there."22

The other, whose role as witness is, in Western societies, at
least, (but not always) assigned to the mother, can in
essence be ddermined neither as an individual, nor as a social
function, noras a subject in general. Indeed, it is nothing more
or less than differentiality upon which discourse depends,
but which itself is not so much discursive as gestural. Like the
"uh. . .", this other takes the place of the signifier, whose
imperfection appears as a fading in and out, or as an
interruption.The realm of the signifier is not confined to the
verbal discourse it makes possible, but includes the appeal and
the exchange of looks, "that most evanescent of objects, for it
appears only on the margins." Such gestures and appeals may
antedate verbal discourse, but their structure is homologous to
its condition, that of the signifier.
The on the mirror stage thus sets
the stage of a very different theater, albeit one that was already
"on the scene." The mirror stage is not negated or invalidated,
far from it; it is given a new twist by a gesture that turns away
from the image in a silent appeal to the look as such. The place of
the subject is thus no longer framed by the mirror, but relegated
to the enablartg margins of the visible. On this margin, the
manifestation of a certain "prematuration" is replaced by
another lack, a "much more critical one, the concealment of
which is the secret of the subject's jubilation."23 The new twist,
the look that is lacking, sets the stage for a very different kind of
drama: that of desire conceived as the desire of the other.

"Ce qui se manipule dans le triomphe de l'assomption de l'image du corps au

miroir, c'est cet objet le plus évanouissant a n'y apparaitre qu'en marge:
l'echange des regards, manifeste ace que l'enfant se retourne vers celui qui de
quelque facos l'assiste, füt-ce seulement de ce qu'il assiste a son jeu." ("What
ismanipulated in the triumph of the reception of the bodily image in the
mirror is thai most evanescent object, which only appears on the fringe: the
exchange of glances, manifest in the turning-around of the child toward the
person who aids it, be it merely through the fact that the person is there
witnessing the game.") Ecrits, f5. 70. In this mention of the "most evanescent"
of objects that appears "only on the margin," Lacan alludes to that most
elusive theory of the "objet petit a" — the object small a — which thereby makes a
fleeting appearance here, in the margins of this book. See also below,
Appendix A, where the relation of the object a to anxiety is explored.
"Cette fonction est d'un manque plus critique, a ce que sa couverture soit le
secret de Ia jubilation du sujet." Ibid.

''9
 8

"When someone speaks, it gets

light": demand

Nowhere is the peculiarity of Lacan's "return to Freud" more

apparent than in his insistence upon a term Freud rarely
employs, and certainly never as a central theoretical concept:
desire. The German word generally held to correspond to the
French, désir, Begehren, is not to be found in the voluminous
Index to the German edition of Freud's collected works; instead
one finds the term Begierde, without any page reference, but
with cross-references to Gier, and in parentheses to Erregung,
Geluste, Wunsch.1 One may thus infer that the problem is at best
of peripheral importance in Freud's writings; and insofar as one
takes these texts at their word, this conclusion is difficult to
refute. Indeed, it is difficult to situate desire in respect to the
words that take its place in Freud, words such as wish,
excitation, drive, libido, eros, etc. In thus ignoring the word,
Freud would appear to continue a tradition of thought which
has increasingly excluded "desire" from rigorous scientific
discourse and relegated it to the languages of poetry, theology
or at best, morals.2 One should thus recognize the degree of
provocation involved in Lacan's breaking with this tradition by
determining desire to be the "the manifestly constitutive vector

'S. Freud, Gesammelte Werke, vol. xviii, S. Fischer Verlag: Frankfurt am Main,
1968.
2 An exception, perhaps, is still to be found in the philosophy of Hegel, and
in particular in The Phenomenologij of Spirit, where the term Begierde signifies
a relation of sensibility that must be overcome in the development of self-
consciousness, If Hegel, possibly through Kojeve (see below, p. 128), alerted
Lacan to the significance of this notion, the Lacanian "dialectic of desire" is
ultimately incompatible with the Hegelian notion of their negation and trans-
cendence, their Aufhebung, in and through self-consciousness. The "dialectics
of desire" will turn Out to be precisely what excludes all such Aufliebung, the
transparency of self-consciousness and the reflexive totalization of the subject
through conceptual discourse.

120
 Demand

of the Freudian field of experience [or experiment: experience],"

and by defining the subject as a "subject of desire."3
If Lacan's provocative emphasis upon desire should not be
overlooked, to accept it as self-evident would also be to blunt its
force. If the interpretation of a text seeks to be more than mere
paraphrase, the introduction of terms absent or even alien to
the text is hardly to be avoided. However, in order to impose
themselves, such terms must call attention to an absence that
structures the text. Let us attempt to trace the contours of this
absence in Freud's work by investigating the concept which
most closely approaches that of desire, without however
completely coinciding with it: that of the wish.
The theory of the wish is developed by Freud primarily in The
Interpretation of Dreams. In the section on wish-fulfillment, he
reconstructs the genesis of the wish as follows: i. The psychic
apparatus strives to keep itself as free as possible from stimuli, a
process described by Freud as that of a reflex apparatus; z. The
"exigencies of life", however, interfere with this striving and
force the apparatus to move in a different direction. These
exigencies confront the psyche first in the form of major somatic
needs: "The excitations produced by internal needs seek
discharge in movement. . . but the situation remains unaltered,
for the excitation arising from an internal need is due not to a
force producing a momentary impact, but to one which is in
continuous operation." Freud adds:
A change can only come about if in some way or other (in the case of
the baby, through outside help) an "experience of satisfaction" can be
achieved which puts an end to the internal stimulus. An essential
component of this experience of satisfaction is a particular perception
(that of nourishment, in our example) the mnemic image of which
remains associated henceforward with the memory trace of the
excitation produced by the need. As a result of the link that has thus
been established, the next time this need arises, a psychical impulse
The entire sentence reads: "Pour nous, le sujet a a surgir de Ia donnée des
signifiants qui le recouvrent dans un Autre qui est leur lieu transcendental:
par quoi il se constitue dans une existence oü est possible le vecteur
manifestement constituant du champ freudien de l'experience: a savoir ce qui
s'appelle le désir." ("For us, the subject must emerge from the given of the
signifiers that cover it over in an Other that is their transcendental site:
whereby it constitutes itself in an existence that makes possible the vector that
is manifestly constitutive of the Freudian field of experience: which is to say,
desire.") Ecrits, pp. 655—656.

I2I
 Return to Freud

will at once emerge which will seek to re-cathect the mnemic image ol
the perception and to re-evoke the perception itself, that is to say, tu
re-establish the situation of the original satisfaction. An impulse of this
kind is what we call a wish; the re-appearance of the perception is tht
fulfillment of the wish; and the shortest path to the fulfillment of th
wish is a path leading directly from the excitation produced by the need
to a complete cathexis of the perception. Nothing prevents us from
assuming that there was a primitive state of the psychical apparatus iii
which this path was actually traversed, that is, in which wishing ended
in hallucinating. Thus, the aim of this first psychical activity was to
produce a "perceptual identity" — a repetition of the perception which
was linked with the satisfaction of the need.4
Let us interrupt Freud's account here, which goes on to
describe the genesis of reality-testing as a necessary "detour
on the way to wish-fulfillment," in order to underscore the
essential moments of his wish theory. We should begin by
examining the relationship and the difference between need
and wish. A condition of the wish is the so-called "experience
of satisfaction," i.e. the satisfaction of a somatic need. The
satisfaction of this need involves three moments: first the need
can be filled only by particular objects determined by the need;
hunger, for example, can, insofar as it is a physical need, be
relieved only by some kind of nourishment and not by water.
Secondly, the elimination of tension brought on by need
[Bedurfnisspannungj is linked to a perceptual image; here we
should again stress the fact that this image depends not upon
the perceived object of satisfaction, but rather upon its spatio-
temporal coincidence with the experience of satisfaction. Thus,
from the very beginning, the image [BildJ functions less as a
copy [Abbild] than as a signal. Yet it is also clear that such signals
cannot be completely arbitrary. Certain objects, persons, and
scenes, including perspectives, such as those studied by Spitz,
from which the nursing child sees the mother's face, are
necessarily linked to experiences of satisfaction and to the
perceptions associated with them. The fact that the mother
plays an important role in this initial experience of satisfaction,
sets the stage for her later psychical function.
Thus, according to this scenario, wish differs from need. Yet,
at the same time, it appears as a kind of second-order
S. Freud, Interpretation of Dreams, SE, v, pp. 565-566.

I ZZ
 Demand

satisfaction of a need. It is described in terms of a tendency to

reproduce the initial perceptual image — which meanwhile has
become a "memory-image" — as something that is independent
of external reality, namely as an hallucination. Hallucination,
accordingly, would constitute the essence of wish-fulfillment.
Nevertheless, even in this schematic presentation, it should
already be clear that the fulfillment of a wish is not entirely
analogous to the satisfaction of a need. If hallucination is the
essence of wish-fulfillment, its way of "fulfilling" a wish is
certainly different from the manner in which a real object
satisfies a need. In contrast to the satisfaction of needs,
wish-fulfillment is characterized by a certain negativity: its scene
is not that of physical objects, but the psychic realm of repetition
and representation. What constitutes this realm is precisely the
absence of that which is represented and wished. For something
like a wish to arise in the first place, objects of need must be
lacking. But this lack is not that by which an object, for instance,
can be said to be absent from or negated by an image of it. What
counts here is not the constitutive absence of the original in the
copy, but a more general absence of object as such. It is this
condition that makes it possible for perceptual images, or more
precisely: for memory traces to function more or less arbitrarily
as signals or signs. Finally, insofar as wishing relies primarily
on outside help, the "helper" helps to constitute the wish, by
means of a certain power. This power helps to bring about an
experience of satisfaction, but at the same time transcends it.
In light of the ensuing description of the psyche, this initial
model of the wish, which treats it still primarily as the
epiphenomenon of unsatisfied need, may seem inadequate. For
as they develop, wishes acquire a relative autonomy; they
can no longer be derived from unsatisfied need, and operate
according to their own laws: those of the primary process. Yet
insofar as the unconscious can "do nothing but wish," as Freud
remarks in The Interpretation of Dreams, its laws must themselves
be grounded in the structure of the wish. First sketched out in
that book, this structure is more fully elaborated in later texts,
in which Freud foregrounds the peculiar negativity already
mentioned. Principal revisions to his initial account of the wish
stem from the related discoveries of infantile sexuality and the
Oedipus complex. Concerning the latter, we shall limit the
I 2.3
 Return to Freud

discussion, at this point, to those aspects of it that are of

particular relevance to the structure of the wish.
Of primary importance is the prohibition of the original love
object, the mother, and the consequences it has for desire. If this
word now supplants the term "wish," it is because the Oedipus
complex involves a degree of structuration difficult to reconcile
with the concept of the wish. With respect to the prohibition of
incest, Freud describes, in Totem and Taboo, the movement of
what he calls Trieblust, translated by Strachey as "instinctual
desire":
The prohibition owes its strength and its obsessive character precisely
to its unconscious counterpart, the concealed and undiminished desire
— that is to say, to an internal necessity inaccessible to conscious
inspection. The ease with which the prohibition can be transferred and
extended, reflects a process that falls in with the unconscious desire
and is greatly facilitated by the psychological conditions that prevail in
the unconscious. The instinctual desire is constantly shifting in order to
escape from the barrier [Absperrungl against which it finds itself, and
endeavors to find substitutes — substitute objects and substitute acts —
in place of the prohibited ones. In consequence of this, the prohibition
itself shifts about as well, and extends to any new aims which the
forbidden impulse may adopt. [My italics]5
Both here and in the case of infantile sexuality, prohibition
seems to be the result of some initial event, and to apply to an
object that at first really was there. Yet, we cannot read the
above lines too literally: the drive "finds itself" only through
the resistance of the barrier, which is nothing but the bar of the
signifier. Such a reading seems to say more than Freud wants to
say. And yet, it is perhaps the only account consistent with his
notion of a wish, which, as we have seen, emerges from the
absence of the object as such; a wish can come about only if its
object (as the object of a need) is missing. Thus, what occurs
diachronically in the experience of the individual subject, must
already be structurally prescribed by the way in which the
absence is articulated. The absence of the object, condition of
possibility of the wish, develops into the incest prohibition
forbidding that the mother (or father) be taken as the primary

S. Freud, Totem and Taboo, SE, xiii, p. 30.

124
 Demand

love object. This very prohibition, however, opens the possibil-

ity of displacement and condensation, the two moments
constitutive of the primary process. We may thus conclude that
although, as a matter of empirical fact, the prohibition applies to
the parents, the latter function as personifications of a structural
category: the signified. The prohibition of incest thus articulates
the movement of articulation as such, insofar as it depends
upon the barring of the signifier, which splits it and causes it to
fall out as a signified. It is this movement that Freud retraces
when describing the trajectory of Trieblust, which "is constantly
shifting in order to escape from the barrier [Absperrung] against
which it finds itself, and endeavors to discover substitutes —
substitute objects and substitute acts — for those that are
forbidden. In consequence of this, the prohibition itself shifts
about as well, and extends to any new aims which the forbidden
impulse may adopt."
It might be objected that Freud is concerned here only with an
abnormal form of desire — namely the ambivalent desire of
obsessional neurosis — which he then links to the rites and
ceremonies involved in taboos. This is doubtless correct, and
yet, if neuroses are singularly illuminating objects of study, it is
because their very singularity brings general tendencies and
structures of psychic life to light. While the specificity of the
different kinds of neuroses should not be overlooked, Freud
never permits any doubt concerning the ubiquity of ambiva-
lence — above all in respect to the intimate and constitutive
complicity between wishes and prohibitions. In "The most prevalent
form of degradation in erotic life," (191 z) Freud writes:
However strange it may sound, I think the possibility must be
considered that something in the nature of the sexual drive itself is
unfavorable to the achievement of absolute gratification. When we
think of the long and difficult evolution the drive goes through, two
factors to which this difficulty might be ascribed at once emerge. First,
in consequence of the double inception [des zweimaligen Ansatzl of
object-choice, with the incest barrier arriving in between, the ultimate
object of the sexual drive is never again the original one, but only a
surrogate for it. Psychoanalysis has shown us, however, that when the
original object of a wishful excitation becomes lost in consequence of
repression, it is often replaced by an endless series of substitute objects,
none of which ever give full satisfaction. This may explain the lack of

125
 Return to Freud

stability in object-choice, the "craving for excitement," which is so

often a feature of the erotic life of adults.6
The second "inception," involving the reduction of polymor-
phous sexuality to genitality, is inscribed in the structure of the
first. Sexuality is afflicted and undermined by the substitution
that constitutes it, and which in turn is the effect of prohibitions
and repression. These define the unconscious character of
sexuality, for whether we are concerned with wishes, drives or
desires, we are always dealing with unconscious structures.
Thus, while the goal of sexuality may be understood as
wish-fulfillment, it begins to emerge that the latter is very
different from an "experience of satisfaction": there is "some-
thing in the nature of the sexual drive that is unfavorable to the
achievement of total satisfaction." The wish, we recall, was
described by Freud as the reproduction not of experienced
satisfaction, but of a representation or perception associated
with such an experience. We now begin to see just how tenuous
such an association can be: both precarious, and yet by virtue of
its very precariousness, all the more tenacious.
In Beyond the Principle, Freud describes this structure
of desire in terms of the dynamics of the drives. What we
previously have referred to as "negativity" now emerges as
"difference":
The repressed drive never ceases to strive for complete satisfaction,
which consists in the repetition of a primary experience of satisfaction.
No substitutive or reactive formations and no sublimations will suffice
to remove the repressed drive's persisting tension; and it is the difference
in amount between the pleasure of satisfaction which is demanded and
that which is actually achieved that provides the driving factor which
will permit of no halting at any position attained. [My italicsj7
Given the resistance of such a movement to closure, we can
see why the process described here cannot be covered by the
concept of "wish" — no matter how closely related to it it may be.
The wish that is "fulfilled" in dreams has the form of a
predication constructed from latent dream-thoughts. These,
however, are "thought" according to the laws, and above all,
6
SE, xi, pp. 188—189.
S. Freud, Beyond the Pleasure Principle, SE, xviii, p. 41. See the discussion of
this passage by Serge Leclaire, in: Psychoanalyser, Paris, 1968.

iz6
 Demand

according to the subject-object grammar of the secondary process,

of the system preconscious-consciousness, while the movement
of displacement, involving what Freud also calls "facilitations"
or "paths," operates according to the laws, and above all,
according to the differential graphics of the unconscious, the
primary process, which "writes" in the wake of the signifier. In
his later writings, Freud increasingly uses the term "drive"
[Trieb]; yet even this notion is too narrow to describe adequately
the particular way in which psychic energy is structured by
"representatives"; and we should be mindful of the consider-
able ambiguity of this term in Freud; it can refer either to the
energy psychically represented by ideas and affects, or to the
representative itself. Indeed, it seems as though a third concept
is needed to account for the movement of desire. Let us attempt
to retrace its structure as described by Lacan.
Desire for Lacan — and undoubtedly for Freud as well — is
essentially unconscious in structure. The latter is therefore
determined by the unconscious, which in turn, as we have
seen, is an effect of the signifying structure of language. As
Lacan writes:
The unconscious exists, not because there is unconscious desire, in the
sense of something impenetrable, heavy, a Caliban or even animal-like,
which emerges from the depths in all of its primitiveness, in order then
to raise itself to the higher level of consciousness. Quite to the contrary,
if there is desire, it is only because there is the unconscious, i.e. a
language, whose structure and effects escape the subject: because at the
level of language, there is always something that is beyond conscious-
ness, which allows the function of desire to be situated.8
Desire is thus structured differentially and as a metonymic
movement; it is oriented less by objects than by signifiers. In
"The instance of the letter," Lacan writes that desire is always
a desire for something else: "le désir d'autre chose — de la
métonymie."9
8
Jacques Lacan, "Psychanalyse et medicine," Left res de I'Ecolefreudienne, no. z.
(1967), p. 45; cited in: M. Safouan, "De Ia structure en psychanalyse,"
Qu'esf-ce que le strucfuralisme?, Paris: Editions du Seuil, 1968, pp. 251—153.
Sheridan, p. 167; Ecrits, p. In "Subversion of the subject" Lacan
comments upon his formula for desire: "Human desire is the desire of the
Other" — in the sense of a subjective genitive: i.e. man desires not the other,
but as another, "en tant que I'Autre." Desire of the Other means accordingly,
desire instead of the other, in a place, however, that is continually displacing

1Z7
 Return to Freud

Yet insofar as desire is directed towards something else which

"itself" can never simply be a self-identical object, it is not only
desirous of another, but is "itself" another's desire. It is "the
desire for the other's desire," the desire of a signifier, defined as
the signifier of another desire. Let us elaborate upon this
formula, which comprises the core of Lacan's theory of desire.
This is perhaps the point in Lacan's thought where he is
most indebted to Kojève's reading of Hegel. Lacan never
concealed this debt, acknowledging Kojève as his "master,"
even if he went on to place increasing emphasis upon the
distance that separated his thinking from the philosophy of
Hegel.'° Kojève's reading of The Phenomenology of Spirit, which
places considerable emphasis on the master-slave dialectic and
on the liberating process of work, traces how the struggle for
recognition necessarily develops out of Hegel's problematic of
desire [Begierde]. So long as the desired end entails the
consumption of natural objects, the subject is defined as a
natural being caught up in the struggle for self-preservation.
Only a desire seeking not a natural object, but rather acknowl-
edgment through another desire raises the subject above its
natural and material existence to the level of pure self-
consciousness. What ensues from this account is the struggle for
mastery. In this interpretation, desire seeks recognition from
another desire. Recognition, however, is a moment in the
autonomous constitution of self-consciousness. The master-
slave dialectic thus represents only a stage in the development
of consciousness, played out between two of its forms. As a
reader of Freud, however, Lacan maintains that this dialectical
moment of self-consciousness cannot set the stage for the scene
of desire, since the latter is essentially unconscious. Yet how can
Lacan at the same time adopt the notion of desire — as desire for
the other's desire, and as desire for recognition through the
other — without situating this process in a dialectic of self-

itself. The subjective genitive of desire is contrasted by Lacan with the

unconscious as the "discourse of the Other"; in the latter case the genitive is
objective, in the sense of de A!io in orafione, as Lacan remarks. Cf. Ecrifs,
p. 8r4ff; Sheridan, p. 312.
10 Cf. above all "Subversion of the subject , where Lacan discusses in some
detail his relation to Hegel and declares it to be "entirely didactic." Ecrits,
p. 794. Sheridan, P. 293. Needless to say, such dedarations hardly can be
taken at their word.

x z8
 Demand

consciousness? To do so, he must develop a radically non-

dialectical notion of the Other.
In a paper on "The direction of treatment and the principles of
its power," Lacan writes:
If desire functions in the subject by virtue of the conditions imposed
upon him by the existence of discourse, namely, that his need must
go by way of the processions [défilésj of the signifier; — if, on the other
hand [. ] the concept of the Other is to be determined as the locus of
. .

the deployment of speech (as that "other scene," of which Freud

speaks); — then it must be supposed. . that human desire is the desire
of the Other.
As Lacan remarks, we are not dealing with a process of
identification here — for, as we will see, the metonymic
movement of desire excludes any kind of identification — but
with "the condition that obliges the subject to find the
constitutive structure of its desire in the very rift [beance] opened
by the effect of the signifiers in those who come to represent the
Other for it, insofar as its demand is subjected to them."12
We encounter here the three terms that will play a determin-
ing role in Lacan's theory of desire: need, demand, and desire.
Need is the least difficult to grasp; it refers to the dependency of
the human organism which requires empirically determined
objects to satisfy certain natural wants. Although Freud makes
no strict terminological distinction between need and desire, our
examination of the Freudian notions of "wish" and "drive" has
suggested how different the two are from each other. The
difficulty begins with Lacan's notion of demand. Let us start by
citing a passage from "The signification of the phallus," in
which demand is distinguished from need:

"Si le désir est en effet dans le sujet de cette condition qui lui est imposée par
l'existence du discours de faire passer son besoin par les défilés du signifiant;
— Si d'autre part, comme nous l'avons donné a entendre plus haut, en ouvrant

Ia dialectique du transfert, il faut fonder Ia notion de l'Autre avec tin grand A,

comme étant le lieu de deploiement de Ia parole (l'autre scene, em anderer
Schauplatz, dont pane Freud dans Ia Traumdeutung); — il faut poser que, fait
d'un animal en proie au langage, le désir de l'homme eSt le désir de l'Autre."
Ecrits, p. 6z8. Sheridan, p. 164.
12 "Cette condition que Ic sujet a a trouver Ia structure constituante de son désir
dans Ia même béance ouverte par I'effet des signiflants chez ceux qui viennent
pour lui a representer I'Autre, en tant que sa demande leur est assujettie."
Ecrits, p. 6z8, Sheridan, p. 164.

I 2.9
 Return to Freud

Demand in itself bears on something other than the satisfactions it calls

for. It is demand of a presence or absence. Which is what is manifest in
the primordial relation to the mother, pregnant as it is with that Other
to be situated this side of the needs it can satisfy. Demand constitutes
the Other as already possessing the "privilege" of satisfying needs, that
is to say, the power of depriving them of that alone by which they are
satisfied. This privilege of the Other thus outlines the radical form of
the gift of that which the Other does not have, namely, that which is
called its love. In this way, demand annuls (aufhebt) the particularity of
everything that can be granted by transforming it into a proof of love,
and the very satisfactions that it obtains for need are degraded Isicl,
erniedrigt] to the level of being no more than the demolishing of the
demand for love (. )13 .
.

Thus demand, which as Lacan writes elsewhere, is always a

demand for love, does not seek concrete objects of satisfaction.
Instead, it is demand for presence or absence, for power. And
while this power may indeed make possible the satisfaction of
needs, as that "outside help" mentioned by Freud in his
description of the child's dependency, as such it transcends any
of its concrete actions. The "privilege" of this power is
expressed first, in the withholding or deprivation of objects of
satisfaction, and second, in its peculiar way of giving, for it
bestows what it does not have: love. Thus, the demand for love
not only differs from the satisfactions it obtains, it also tends to
"depreciate and degrade" them, in order to defend its trans-
cendence against obliteration through the particularity of such
satisfaction.
How then are we to understand this notion of demand? At
first sight, it seems linked to the child's condition of depend-
ence, of Bedurftigkeit, as Freud called it. To satisfy its needs, the
child is of necessity dependent upon outside help. Yet, at the
same time, demand stands in sharp contrast to the real sphere of
13
"La demande en sot porte sur autre chose que sur les satisfactions qu'elle
appelle. Elle est demande d'une presence ou d'une absence. Ce que Ia relation
primordiale a Ia mere manifeste, d'être grosse de cet Autre a situer en deça des
besoins qu'il peut combler. Elle le constitue déjà comme ayant le 'privilege' de
satisfaire les besoins, c'est-a-dire le pouvoir de les priver de cela seul par quoi
ils sont satisfaits. Ce privilege de l'Autre dessine ainsi La forme radicale du
don de ce qu'il n'a pas, soit ce qu'on appelle son amour. C'est par là que Ia
demande annule (aujhebt) Ia particularité de tout ce qui peut être accordé en La
transmuant en preuve d'amour, et les satisfactions même qu'elIe obtient pour
le besoin se ravalent (sich erniedrigen) a n'être plus que l'ecrasement de Ia
demande d'amour Ecrits, pp. 690—691. Sheridan, p. z86.

130
 Demand

need. As the transcendental condition of possibility of satisfac-

tion, demand cannot coincide with any concrete need. Lacan's
pairing of the words "privilege" and "deprivation" suggests
that demand might be defined as a kind of general negation of
dependency, which could lead one to infer that the subjective
basis of demand is not need, but anxiety. Although this line of
thought is not explicitly pursued by Lacan,'4 a footnote in Three
Essays on the Theory of Sexuality gestures in this direction:
For this explanation of the origin of infantile sexuality I have to thank a
three year old boy, whom I once heard calling out of a dark room:
"Auntie, speak to me! I'm frightened because it's so dark." His aunt
called back to him: "What good would that do? You can't see me."
"That doesn't matter," replied the child, "when someone speaks, it
gets light." Thus, what he was afraid of was not the dark, but the
absence of a loved one, and he could promise to calm down once he had
obtained proof of that person's presence.
Anxiety is thus relieved by "proof" of the "presence" of "a
loved one," and it is in this manner that "demand" is answered.
This answer in turn involves a linguistic operation: it is voiced
and elicits a promise (the child "promised to calm down once he
had obtained evidence of the beloved person's presence"). It
marks a new stage in the assumption by the subject of its
relation to language; instead of merely crying, the child
increasingly uses articulated speech to draw attention to its
needs. As Lacan remarks, "with demand we find ourselves in
the domain of what is actually articulated."16 It is demand, not
need, that "opens the way to unconscious desire"17; it is "an
approach" to desire and not desire itself. Why?
A scene from Beyond the Pleasure Principle provides elements of
a response. The passage has received widespread attention ever
since Lacan interpreted it as exemplifying the relation of
demand to desire. Freud describes a "game invented and played
by" his grandson, "a little boy of one and a half":
The child was not at all precocious in its intellectual development. At
the age of one and a half he could say only a few comprehensible

14
See below, chapter 10, "The witch's letter."
S. Freud, Three Essays on the Theory of Sexuality, SE, vii, p. 224, note.
16
Lacan, "Les formations de l'inconscient," Bulletin de Psychologie xi, p. 2.55.
Ibid.

131
 Return to Freud

words; he could also make use of a number of sounds which expressed

a meaning intelligible to those around it. He was, however, on good
terms with his parents and their one servant-girl, and tributes were
paid to his being a "good boy." He did not "disturb" his parents at
night, he conscientiously obeyed orders not to touch certain things or
go into certain rooms, and above all he never cried when his mother left
him for a few hours. At the same time, he was greatly attached to his
mother, who had not only fed him herself, but had also looked after
him without any outside help. This good little boy, however, had an
occasional disturbing habit of taking any small objects he could get hold
of and throwing them away from him into a corner, under the bed,
and so on, so that hunting for his toys and picking them up was often
quite a business. As he did this he gave vent to a loud, long-drawn-out
"o-o-o-o," accompanied by an expression of interest and satisfaction.
His mother and the writer of the present account were agreed in
thinking that this was not a mere interjection but represented the
German wordfort ["gone"]. I eventually realized that it was a game and
that the only use he made of any of his toys was to play "gone" with
them. One day I made an observation which confirmed my view. The
child had a wooden reel with a piece of string tied round it. It never
occurred to him to pull it along the floor behind him, for instance, and
play at its being a carriage. What he did was to hold the reel by the
string and very skillfully throw it over the edge of his curtained cot,
so that it disappeared into it, at the same time uttering his expressive
"o-o-o-o." He then pulled the reel out of the cot again by the string and
hailed its reappearance with a joyful "da" [there]. This, then, was the
complete game — disappearance and return. As a rule one only
witnessed its first act, which was repeated untiringly as a game in itself,
though there is no doubt that the greater pleasure was attached to the
second act.18
Freud interprets the game in two ways and, as we shall
see, the two need not be mutually exclusive. First, the child
"compensates himself" for his "great cultural achievement — the
renunciation of instinctual satisfaction — which he had made in
allowing his mother to go away without protesting." The way
he compensates for this is "by himself staging the same disap-
pearance and return with objects within his reach," thereby
repeating and symbolizing the mother's departure and return
with the wooden reel — that is, with a language of things.
Second, Freud suspects that behind this game is some act of
vengeance; for the child not only passively accepts or symboli-
18
S. Freud, Beyond the Pleasure Principle, SE, xviii, pp. 14—15.

132
 Demand

cally recuperates the mother's absence, but instead acts it out

again in its game. It is as if the child were "defiantly saying, 'All
right then, go away! I don't need you. I'm sending you away
myself." Freud's interpretation was later confirmed — this time
in relation to the father who had been called away to the front.
In this example we see how the presence and absence of the
mother is at first repeated and symbolized by means of a
thing-language. Yet beyond this, we observe the beginnings of a
linguistic form — that of the signifier, in Lacan's terms. The
presence-absence alternative is signified by the difference
between the two sounds, "o-o-o-o" and "a-a-a-a," functioning
at first as phonemes. What interests Lacan are not so much the
objects used to symbolize the mother's departure and return, as
this phonematic opposition, which forms a bridge between
sign-language and the language of the signifier. What role then
do presence and absence play in the game?
First of all, they signify the possibility of narcissistic identifica-
tion as in the mirror stage. This is evidenced in a footnote Freud
adds to his description:
A subsequent observation fully confirmed this interpretation [i.e. that
the game symbolized the mother's disappearance and return]. One day
the child's mother had been away for several hours and on her return
was met with the words "Baby 0-0-0-0!", which were at first incom-
prehensible. It soon turned out, however, that during this long period
of solitude the child had found a method of making himself disappear.
He had discovered his reflection in a full-length mirror which did
not quite reach to the ground, so that by crouching down he could
make his mirror-image "gone."19
The presence-absence of the mother forms the matrix of
narcissistic identification; according to Lacan, it is out of this that
the ego first emerges as the figure of irremediable alienation.
Yet, as we have already heard, the demand for presence or
absence must itself be situated in the realm of actual articulation.
Unlocalizable in the imaginary register alone, demand opens
the way to unconscious desire, i.e. to the symbolic. As Lacan
remarks in reference to the story of Freud's grandson, "de-
mand is first linked to the premises of language." Yet, what
are these "premises"?2° There are at least two: first, that
19 SE, xviii, p. note.
20 "Les formations de l'inconscient," p. 251.

'33
 Return to Freud

language presupposes the absence of what it speaks about — its

referents. This is similarly the presupposition of language con-
ceived as a signifying system; the referent is not only pre-
sumed to be absent; it is the absent other. And this is precisely
the first aspect of the child's game: in throwing away and
retrieving objects, it can at will set in motion and control the
Fort-Da (gone-there) alternation. In so doing, the child is an
active subject mastering both the game and its symbolism. The
child acknowledges the mother's absence, yet only insofar as it
can make use of it — that is, only insofar as absence can be used
(and appropriated) to further its own identification and ego
constitution. Representation and articulation operate only in a
limited sense here — only insofar as they can be enlisted and
put to use in the making-present of an identity. It is in this
sense that we understand Lacan's assertion that "demand is
linked to an identifying and idealizing function."21 So long
as language qua representation is viewed essentially as
absence, it virtually effaces itself in the voice of a speaker, to
whose presence it seems dialectically to bear witness.
Yet a second presupposition of language arises when it is no
longer considered merely as a signifying system, but instead
as a distinctive one. Here language is seen as a movement of
difference in which the moment of absence is given priority; this
absence is always related to the negation of a presence and of a
signified. This presupposition is implicit in the doubling of the
object-language by the phonemes o - A: in this apparently
superfluous doubling of symbolization, the sign is replaced by
the signifier and the ego is supplanted by the subject of the
unconscious. With the entry of the subject into the language of
the signifier — into verbal language in this — it enters into a

21
Ibid.
22
As the previous discussion of Saussure has suggested, it would be a mistake
to equate "verbal" discourse with "language" qua signifying function. To be
sure, "verbal" is not simply a descriptive category: to speak of "words"
presupposes an interpretation of what constitutes a word. If the word is
"intuitively" defined by its morphemic content: i.e. as having a minimal
meaning, in linguistic practice words are ambiguous, and such ambiguity, far
from simply adding to the minimum semantic condition, tends to foreground
the dependence of its signification, like all signification, upon "contextual"
and syntactic factors. In the particular case under discussion, two remarks are
in order: first, that it is differential signification, presented in the form of the
phonemic (or semantic) opposition, which defines the couple, o-o-o/a-a-a.

'34
 Demand

structure of articulation in which direct identification no longer

functions; the subject can never again hope to find itself in the
signifier, because the latter only receives its identity by virtue of
its place in the signifying chain. Signifiers "are," only by virtue
of their difference. The subject is thereby split between the
"said" and the "saying," between the enunciated and the
enunciation; it is inscribed in a structure of representation that
cannot be traced back to an original presence, but is instead
constituted by an irreducible movement of repetition.
This, however, also modifies the status of the Other: as the
addressee of the demand, the Other offers a guarantee of its
claims. In a linguistic perspective, this Other is the reservoir and
resource of meaning, "the treasure trove of signifiers," as Lacan
puts it ("trésor des Moreover, it is a treasure in the
sense of a code; the signifier here functions primarily as a sign,
as the representative of a signified and as the sender of a
message whose content is the presence of the sender himself.
And it is certainly no accident that in the case described by
Freud, what assures the child of the presence of the person it
loves is not seeing her, but hearing her voice. "When someone
speaks, it gets light," says the child, thereby rediscovering what
Jacques Demda describes as the metaphysics of presence, that
is, phonocentrism.24
In this sense, the Other of demand guarantees the meaning
and presence of articulation. Lacan can thus write that demand
is "at home in what is actually articulated," since only demand —
and the Other it calls upon — can claim any such "actuality" of
articulation. But demand, as articulation, is not merely imagin-
ary, though it constitutes the basis and medium of the
Imaginary. In demand, need goes the way of the signifier, that
is of symbolic articulation, even if demand itself seeks an
imaginary realm, "beyond articulation": an absolute presence or
absence.
The opposition of presence and absence of the Other in
demand, however, initiates a movement which cannot be

And second, that it is this relationship of binary opposition, presence or

absence, either/or, that prevails in the constitution of the ego.
23
Ecrits, p. 8o6; Sheridan, p. 304.
24
Of Grammatology, passim.

'35
 Return to Freud

restricted to these two poles. Lacan indicates the trajectory ol

this movement in "The signification of the phallus":
It is necessary then that the particularity thus abolished [aufgehoben]
demand reappear beyond demand. It does in fact reappear there,
preserving, however, the structure contained in the demand-for-love's
unconditionality. By a reversal that is not simply a negation of thu
negation, the power of pure loss emerges out of the residue of an
obliteration. In place of the unconditionality of demand, desire
substitutes the "absolute" condition: this condition unties the knot ol
that which in the proof of love rebels against satisfaction. Thus desire is
neither the appetite of satisfaction, nor the demand for love, but the
difference that results from the subtraction of the first from the second,
indeed, the very phenomenon of their splitting
Desire thus entails not only the difference between the
satisfaction of particular needs, and an unconditional demand
for love, but difference itself, i.e. "the phenomenon of their
splitting." Desire is the absolute condition insofar as it desig-
nates a movement of differential articulation based on the other
— on difference. Yet at the same time it preserves the structure

"contained" in the demand-for-love's "unconditionality," for

desire's own movement is interminable; as such, desire must
also function "unconditionally." The "object" of desire, signifier
of another desire and of the Other's desire, always points to
another signifier. In so doing, it refers not only to its own
condition but beyond it as well — to other conditions. Just as
hallucinatory wish-fulfillment presupposes the absence of the
original object of satisfaction; just as the displacement of desire
is occasioned by the loss of the original object (the mother) and
the assumption of the incest prohibition, so desire presupposes
the denial of need while pnvileging demand. Desire thus
emerges on the fringes of denial.

"II y a donc une necessité a ce que Ia particularite ainsi abolie reparaisse

au-delà de Ia demande. Elle y reparait en effet, mais conservant Ia structure
que recèle l'inconditionné de Ia demande d'amour. Par un renversement qui
n'est pas simple negation de Ia negation, Ia puissance de Ia pure perte surgit
du résidu d'une obliteration. A l'inconditionne de Ia demande, le désir
substitue la condition 'absolue': cette condition dénoue en effet ce que Ia
preuve d'amour a de rebelle a Ia satisfaction d'un besoin. C'est ainsi que le
désir n'est ni l'appétit de Ia satisfaction, ni Ia demande d'amour, mais Ia
difference qui resulte de Ia soustraction du premier a Ia seconde, le
phénomene méme de leur refente (Spaltung)." Ecrits, p. 691; Sheridan,
pp. 286—z87.

136
 Demand

This Other of desire can thus no longer be located in some

kind of code, since a code implies a collection of signs based on a
particular system of signifieds. This Other of desire is instead the
locus of the discourse of the unconscious; it can only be placed as the
difference between the "said" and the "saying," between
signified and signifying, or more exactly as the movement of
signifiers which itself always takes place upon "another stage."
This Other locale thus traces the contours of that dislocation,
that "transcendental" locus, where any possible combination or
configuration of signifiers must in turn always be another
signifier referring to something beyond itself. This Other, like
the other of demand, befalls the subject in a variety of ways: for
example, in the form of the third "person." As Freud stresses in
his book on jokes, the third person is an essential element of the
joke (though not of humor), for a joke only is one insofar as the
listener laughs at it. Yet just as the exemplary embodiment of
the Other of demand is the mother, so the Other of desire is
personified in the father, for it is he who introduces the law of
desire through the incest prohibition and the threat of castra-
tion. If we consider the Other as the dislocation of the signifier,
it cannot be incarnated in the living identity of a person; here as
well, the locale remains closed, barred. What is important is not
the person of the father, but his role as guardian of the law.26
Lacan often stresses the fact that this Other "does not exist",27 it
is barred, always elsewhere, inaccessible. Lacan names the
mode of this inaccessibility the name-of-the-father (le nom du père).

26 The "father" is guardian of the Law, but not Legislator. The law is not given
by anyone, "there is the Law," and the "there" marks the place of the Other.
This place can be represented by the father, whom Lacan describes as
"representant onginel" [original representative] of the Law. But if the father
can exercise this function, it is less as a person than through his name, which
in turn, as signifier, presupposes the radical absence of the signified. This is
why it is above all m the place of the dead father that this name assumes its
structuring power. This is the sense of the parricide, allegorically recounted in
Totem and Taboo. It is only in inter-dicting himself, that the "father," qua
"function," can become the effective site of the Name of the Father. It is
therefore of more than anecdotal interest that in November, 1963, Lacan
interrupted the seminar he had planned to devote to the "Names-of-the-
Father," after its first session, as a response to the exclusion of his name from
the list of training analysts. The text of that session has now been translated
into English by effrey Mehlman, and published in the Dossier on the
Institutional Debate, October 40, pp. 81—95.
27
Ecrits, pp. 8zo, 8z6; Sheridan, pp. iii.
'37
 Return to Freud

Yet while the name-of-the-father is the signifier of that place

from which desire receives its law via prohibition, this law is
enforced only by virtue of another signifier, which structures
desire and which perhaps can only be named improperly, as the
"phallus."

138
 9

++ + ++ + + ++ + ++ + ++++++ + +++++ + ++ +
The signification of the phallus
+ + ++++ +++ + + ++ + ++ + ++++++ ++ + ++++++ +

Let us begin with a scene from a film: a little girl stands naked
before a mirror and observes herself. Suddenly, she places her
hand on her body, covering her sex. End of scene. Or perhaps
the beginning of another? (To be re-viewed in "Of our
antecedents."1)
What the mirror image represents for a subject that cannot yet
control its body is an image of totality. The unity of the ego will
follow from an identification with this reflection of the body as a
full Gestalt. What counts here is a sense of wholeness. Yet what
happens when the image is no longer whole? And what if it is
lacking in a way that can never be remedied? This is precisely
what the little cinematic scene displays: observing herself in the
mirror, the girl suddenly notices that something is missing; she
responds to this "perception" by trying to conceal (or in some
way, make up for) the absence. The mirror image is no longer
whole. Would it have been different if, instead of a little girl,
there had been a little boy?
Let us hear what Lacan has to say:

Certainly there is in all this what we call a hitch. . . a margin that all
thought has avoided, skipped over, circumvented, or blocked whenev-
er it apparently succeeds in sustaining itself through a circle, be it
dialectical or mathematical.2

What then is this margin that ever since Freud appears as the
shadow of the circularity with which thought seeks to sustain
itself, haunting it with the ghost of the squaring of the circle?

Ecrits, p. 70.
2 "Assurément ii y a là ce qu'on appelle un os. [. .
I Cette marge que toute
.

pensee a évitée, sautée, contournée ou bouchee chaque fois qu'elle réussit

apparemment a se soutenir d'un cercie: qu'eIIe soit dialectique ou mathémati-
que." Ecrits, p. 8zo; Sheridan, p. 318.

'39
 Return to Freud

Yet while the name-of-the-father is the signifier of that placc

from which desire receives its law via prohibition, this law is
enforced only by virtue of another signifier, which structurcs
desire and which perhaps can only be named improperly, as th4.
"phallus."

138
 9

The signification of the phallus

Let us begin with a scene from a film: a little girl stands naked
before a mirror and observes herself. Suddenly, she places her
hand on her body, covering her sex. End of scene. Or perhaps
the beginning of another? (To be re-viewed in "Of our
antecedents."1)
What the mirror image represents for a subject that cannot yet
control its body is an image of totality. The unity of the ego will
follow from an identification with this reflection of the body as a
full Gestalt. What counts here is a sense of wholeness. Yet what
happens when the image is no longer whole? And what if it is
lacking in a way that can never be remedied? This is precisely
what the little cinematic scene displays: observing herself in the
mirror, the girl suddenly notices that something is missing; she
responds to this "perception" by trying to conceal (or in some
way, make up for) the absence. The mirror image is no longer
whole. Would it have been different if, instead of a little girl,
there had been a little boy?
Let us hear what Lacan has to say:

Certainly there is in all this what we call a hitch. . . a margin that all
thought has avoided, skipped over, circumvented, or blocked whenev-
er it apparently succeeds in sustaining itself through a circle, be it
dialectical or mathematical.2

What then is this margin that ever since Freud appears as the
shadow of the circularity with which thought seeks to sustain
itself, haunting it with the ghost of the squaring of the circle?

Ecrits, 70.
ii y a là ce qu'on appelle un Os. I. I Cette marge que toute
2
. .

pensee a évitée, sautée, contournée ou bouchee chaque fois qu'elle réussit

apparemment a se soutenir cercle: qu'elle soit dialectique ou mathémati-
que." Ecrits. p. 82.0; Sheridan, p. 318.

'39
 Return to Freud

This stumbling block, or as Freud calls it, this

Fels,"3 has a name: castration. The question, what that
names — whether a fear, a threat or a "complex" — can for h
moment be left unanswered. Only two things seem cert,
first, whatever castration may be, it's not good. Second, it I

nothing less than the hinge upon which Freudian psychoana

turns. How so and why, are the questions to which the Lacaiii,ii
theory of the phallus seeks to respond.
To answer these questions we must attempt to retrace th
path that led Freud to the theory of castration. Freud's thinkiiiy,
on the subject crystallizes slowly. The matter first comes up
the context of his "Analysis of the phobia of a five-year-old
boy,"4 "little Hans," and it is discussed at length in his
"On the sexual theories of children"5 published in 1908. Yet,
is not until 1923 that Freud will elaborate fully the significa 111
of his discovery in his essay on "The infantile genital organiii
tion of the libido,"6 where he sets forth the theory of the phalli
phase. Let us first summarize the main features of this
as itself a theory of meaning and of signification befon
attempting to describe its significance for psychoanalysis.
The essay, "On the sexual theories of children," describes thl
castration complex as one of those "false theories of
that nevertheless contain "a grain of truth." (2.25) "The first ol
these theories, which builds upon a neglect of sexual difti.',
ence," according to Freud, "consists in attributing to
including women, a penis just like the one the boy knows 'i
from his own body." (2.25)
As Freud clearly states, the reasons for this attribution an
essentially narcissistic:

The penis is already in childhood the key erogenous zone and the mo'.l
important auto-erotic sexual object, and the child's appreciation of it'.
value is logically reflected in his inability to imagine a person similar I
himself without this essential part. 1215—2.16, my emphasis)

S. Freud, "Analysis Terminable and Interminable," SE, x, p. zsz; GW XVI, F'

99.
"SE, x.
S. Freud, "On infantile sexual theories," SE, ix. Page references to this worL
will be given in parentheses in the body of the text.
SE, xix, pp. 141—145. References to this essay will be given in parentheses in
the body of the text.

140
 The signification oft/ic phallus

Freud continues:
If a little boy obtains a sight of his little sister's genitals, what he says
shows that his prejudice is already strong enough to warp the
perception; he does not remark on the lack of the penis but invariably
says, as if to console and reconcile: the. . is still small, but when she
.

gets bigger, it, too, will grow. (z16)

The narcissistic moment is already apparent in this first
formulation, which actually has more to do with the point of
departure of the castration complex than with the complex itself.
The ego — the boy's in this case — persistently endeavors to find
itself, its mirror image, in others. Hence, the boy's "inability to
imagine a person similar to himself without this essential part,"
viz, the penis. Hence, too, his disavowal of the perception that
does not confirm this expectation; the boy does not remark on
the lack of the penis, but only that it is still small. Castration is
discussed here only as one of many infantile theories of
sexuality. Nevertheless, Freud already claims that it applies to
young girls as well as to small boys; he does not, however,
elaborate on this. As already stated, it is in 1923 jO "The
infantile genital organization of the libido," that Freud first
grounds the universality of castration in a theory of the
primacy of the phallus, or the so-called "phallic phase." He
begins by correcting his earlier thesis that "the primacy of the
genitals is not completed in the early period of childhood, or
only very imperfectly."7 He now claims instead that "the
approximation of childhood sexuality to that of adults goes
much farther" than he had at first believed; yet at the same time,
Freud points out a significant difference between the "infantile
genital organisation" and "the final genital organisation of the
adult." The difference lies in the fact that
for both sexes in childhood only one kind of genital organ comes into
account — the male organ. The primacy reached is, therefore, not a
primacy of the genitals, but of the phallus.8
Thus begins one of the liveliest and most significant debates
in the history of psychoanalytic theory; although Freud admits
he does not have enough observational material to support fully

7SE, XIX, 141.

l SE, XIX, 142.

141
 Return to Freud

his presumption of the primacy of the phallus in both sexes, h'

nevertheless insists that it is invested with the same psychi&
significance for girls as well as for boys. And despite increasing
skepticism and criticism from within the psychoanalytic movc
ment itself — on the part of Ernest Jones, Melanie Klein, and
Karen Homey, among others — Freud sticks to his thesis, even
if the empirical evidence in support of his conviction that "th'
real female genitals never seem to be discovered" by the young
girl appears ever more dubious. It is difficult, therefore, to
avoid suspecting that Freud's position in this decisive debate is
determined by "phallocentrism," as Jones calls it. Is one to
conclude, then, that Freud, and certain central psychoanalytical
tenets, are as badly tainted by "male chauvinism" as has often
been charged?9 Why privilege the phallus in such an obviously
one-sided manner?
To even address this question adequately, much less respond
to it, the precise significance of this privilege must be discussed,
which means, first of all, to determine the place it occupies
within the general context of Freudian theory. The first,
obvious, but nevertheless hardly trivial aspect that should bi'
noted about that place is its proximity to what Freud refers to as
castration. At first glance, it would seem as if Freud construes
that proximity in terms of a genetic derivation:
It seems to me [. . ] that the significance of the castration complex can
be rightly appreciated only when its origin in the phase of phalik
primacy is also taken into account.
Although Freud concedes that there are experiences 01
separation in childhood development that prepare the child for
the experience of castration — e.g., the loss of the nipple after
nursing and the daily production of its feces — "the castration
complex," he insists, "should be a term reserved for th'
occasion when the idea of such deprivations comes to bt'

For a counter-argument, cf. Juliet Mitchell, Psychoanalysis and Fe,ninisn,.

London, 1974, p. xlii: "The greater part of the feminist movement has
identified Freud as the enemy. It is held that psychoanalysis claims women
are inferior and that they can achieve true femininity only as wives and
mothers. I. .11 would agree that popularized Freudianism must answer tn
.

this description; but the argument of this book is that a rejection iii
psychoanalysis and of Freud's works is fatal for feminism." Mitchell's
understanding of Freud is strongly influenced by her reading of Lacan.

141
 The signification of the phallus

associated with the loss of the male organ." note) Is the

priority Freud here attributes to the phallus sheer evidence of
the naturalistic bent of his thought, the fetishization of an
anatomical condition, rather than a theory that is itself supposed
to explain fetishism? Or is it an adequate description of a state of
affairs which has its basis not in anatomy, but in a specific
cultural and social tradition, which, however vast, is still far
from being universal? Can the theory of the phallic phase and
the universality of castration be applied to women as well as to
men? Why should the feminine psyche be conditioned by the
loss of something which in reality cannot have been lost, since it
was never possessed?
As Lacan comments, "it might be a good idea to re-examine
the question by asking what could have imposed upon Freud
the evident paradox of his position."° His response is that
Freud's phallocentrism is the result of "the intrusion of the
signifier into the human psyche, which is strictly impossible to
deduce from any pre-established harmony of this psyche with
the nature it expresses." Accordingly, the phallus must be
understood as that which marks the passage from the imaginary to
the symbolic, from demand to desire, as a discontinuous and
conflictual one. The favored name for this discontinuity and
conflict is "castration." Having established just what the
Lacanian hypothesis is, let us try to determine the degree to
which it is inscribed or prefigured in the writings of Freud.
We have already pointed out that castration presupposes the
conviction that all human beings have a penis. This belief has its
origins in a narcissism said to be common to both sexes. Freud
maintains this position by asserting that what counts, for the
psyche, is the perception not of one's own body, but of others'.
The belief in the ubiquity of the penis is always held in relation
to the other. The totality of one's body-image, the narcissistic
condition for the development of the ego, is, however, called in
question by the perception of the absence of the penis. Called
into question, but not simply shattered. The invisibility of the
penis can in fact help to confirm its existence; according to some
early childhood phantasies interpreted by Melanie Klein, the
'° Lacan, "La signification du phallus," Ecrits, p. 688; Sheridan, p. 2.84.
J.
Ecrits, p. Sheridan, p. 198.

'43
 Return to Freud

(father's) penis is contained inside the mother's body. Here we

have a concrete example of the theory of perceptual identities
discussed earlier at work, for a child's perception is certainly not
disinterested or objective. It is motivated and conditioned by a
curiosity which itself serves the interests of the narcissisni
expressed in the demand for love.
Nevertheless, self-expression here is at the same time self-
alienation, since the demand is dependent upon the hetero-
geneity of the other in the movement of signification. Since the
other addressed by the infant is (in modern Western middle-
class culture) generally the mother, it is here that the drama ol
castration unfolds. Demand, however, is articulated in the
medium of the imaginary and this in turn has particular
implications for castration. First, the perception of the female
genitals is misconstrued as the perception of an absence, of a
"lack" or "loss"; that is, as castration in a narrow or even literal
sense. The alternative logic that structures both demand and its
medium, the imaginary, thus remains in force, since the subject
is presented with the choice: absence or presence. The Other -
in this case, the mother — is the seat of a code formed as a system
of signs; its message thus presents itself as one consisting of
signs that refer to an (absent) referent. The exemplary organ of
this is therefore the voice, insofar as it can claim to assure the
presence of the (beloved) speaker: "When someone speaks, it
gets light." In its pretension to presence, however, the voice
sets itself apart from articulation understood as a movement 01
signifiers or as a play of difference. On the contrary, it tends to
reduce difference by determining it as the interval between itseli
and the being of the speaker — between sign and referent.
Hence, we see how the imaginary structure of demand — with
the mother in the position of the utterly Other — is actually
supported by the invisibility of the maternal phallus. The
mother does not have a penis, and yet precisely by virtue of this
not-having, she can present herself as being the phallus. For
what is decisive, Lacan insists, to all understanding in this
question, is the awareness that the phallus functions psychically
not as an organ, but as a particular kind of representation, a
simulacrum:
In Freudian doctrine the phallus is not a phantasy, if by that we mean
an imaginary effect. Nor is it as such an object (partial, internal, good,
144
 The signification of the phallus

bad, etc.), in the sense that this term tends to accentuate the reality
obtaining in a relation. It is even less the organ, penis or clitoris, that it
symbolizes. And it is not without reason that Freud used the reference
to the simulacrum that it represented for the Ancients. For the phallus
is a signifier, a signifier whose function in the intrasubjective economy
of analysis is perhaps to raise the veil of the function it had in the
mysteries. For it is the signifier destined to designate the effects of the
signified in their entirety, i.e. insofar as these effects are conditioned by
the presence of the signifier. 12
The phallus is thus situated, decisively and incisively, on the
border that separates the imaginary from the symbolic. It
emerges out of the gaps of a perception that apprehends only
presence or absence. Within the phantasmatic economy of
human desire the phallus is, therefore, a simulacrurn: it presents
similarity instead of the dissimilar, symmetry in place of the
dissymmetrical. It is a perception that strives to be identical, but
it is not a perceptual identity (except in the Freudian sense of
that term). What it represents is not the absence of a presence,
but a difference impossible to apprehend in terms of presence or
absence. What it represents, but only by effacing it, is the
differential relation of the sexes. The trace of this effaced
difference it then names: "castration." This name also desig-
nates the "falling out" of the signifier with — and as — the
signified. This is what Lacan is driving at when he writes that
the phallus is the signifier that designates "the effects of the
signified in their entirety, insofar as they are conditioned by the
presence of the signifier." What, however, are these effects of
the signified and how does the "presence of the signifier"
condition them?
These effects belong to the imaginary mode of articulation,
whose structural basis, the transcendental signified, is always

12 "Le phallus dans Ia doctrine freudienne n'est pas un fantasme, s'il faut
entendre par là un effet imaginaire. II n'est pas non plus comme tel un objet
(partiel, interne, bon, mauvais etc. .) pour autant que ce terme tend a
apprecier Ia réalité intéressée dans une relation. II est encore bien moms
l'organe, penis ou clitoris, qu'il symbolise. Et ce n'est pas sans raison que
Freud en a pris Ia reference au simulacre qu'iI était pour les Anciens. Car le
phallus est un signifiant, un signifiant dont Ia fonction, dans l'économie
intrasubjective de l'analyse, soulève peut-etre le voile de celle qu'il tenait dans
les mysteres. Car le signifiant destine a designer dans leur ensemble les
effets de signifie, en tant que le signifiant les conditionne par sa presence de
signifiant." Ecrits, p. 690; Sheridan, p.

145
 Return to Freud

already conditioned by the signifier. In this respect, it must he

remembered that, whatever else it may be, the phallus is first of
all the idea of something that in reality has never existed: the
maternal penis. Once the child establishes that this idea has no
positive or objective referent, the significance of the phallus is
split for it: on the one hand, the phallus becomes a kind of
negative idea of a particular non-being; it signifies the total
presence of the mother — a presence that can claim to be
non-objective and transcendental, since it withdraws itself, or is
withdrawn, precisely in the granting of the satisfaction of need.
It is as the negation of the particular satisfactions and objects 01
satisfaction it grants; it is precisely as their degradation and
belittlement. The phallus thus operates as the pure representa-
tion of absence, a representation that is pure because it
represents nothing, and hence, coincides, qua representation,
with what it represents, without leaving the slightest trace or
residue. In this sense, the phallus is the perfect simulacrum: one
that can claim to be utterly self-identical in the pure ideality of
representation. The phallus then would be pure sign, were it
not for the fact that it does not completely disappear or dissolve
into diaphany; instead, it reappears inscribed as the distinctive
trait of the male body, and above all, of the paternal body. Or
more precisely: it is not the phallus that reappears, but instead
that which it initially seems to represent, before it reveals itself
in the claim to be the non-being of the maternal penis, before.
that is, it takes off upon the path of pure representation.
The phallus now appears in a second phase Lacan calls
"privation." Rather than appearing as the mark of the transcen-
dental being of the mother, it now appears as that which she
neither is nor has: as the exclusive property of the father and
simultaneously as that which the mother, in her negativity, caii
neither be nor have, but only desire. In the first phase,
coinciding with the mirror stage, the child tries to be the phallus
in order to satisfy the desire of the mother: to confirm her
completeness, on the one hand, and to partake in it, on the
other. Now, however, the child sees itself excluded by the father
from identifying narcissistically with the phallus as absolute
being and with the mother as presence. Moreover, the child
only feels the force of this exclusion when it learns that even the
father cannot possess the phallus, but only speak in its name.
146
 The signification of the phallus

Thus, while the father may be the locus of a prohibition, this

prohibition also applies to the "law-giver" himself, turning him
into a symbolic father or into what Lacan calls the name-of-the-
father, and which might also be rendered as the naming of the
father (subjective and objective, genitive and genitor). Only with
the naming of this name, does the phallus become for the
subject the signifier of its desire, that is, of its relationship to the
symbolic, and the symptom of its split articulation in and
through the signifier.
Thus, the phallus, in the course of its movement, otherwise
known as castration, describes the operation that transforms the
signified demanded (and the signified of demand) into the
signifier of desire; that is, it transforms it into what it always wi/l
have been, insofar as the subject is structured in and through the
signifier. In the problematic of the phallus and of castration,
we see how the structure of language conditioning the subject
converges with the destination of the individual. As Lacan
remarks, "the phallus is the privileged signifier of the mark in
which the role of the Logos and the arrival of desire converge."3
Two moments thus come together in the phallus: first, the
structure of the signifier as the differential element of articula-
tion. The phallus is the mark of this structure insofar as it
functions as a negative representation, which however does not
negate (and surpass) itself in the process; that is, it does not
function according to the totalizing teleology of dialectical
sublation. The signification of the phallus is constituted through
castration, even and especially if "nothing" is castrated. In this
sense the phallus is the signifier of signification in general, the
signifier of difference. To put it more precisely, the phallus
marks the bar and the barrier, which itself is meaningless, but
which renders signification possible. Lacan alludes to this
relationship and function in his description of the moment in
the famous painting of the Pompei villa when the phallus,
which otherwise is always depicted as veiled, is revealed.
It then becomes the bar which, at the hands of this demon (shame)
strikes the signified, marking it as the bastard offspring of this
signifying concatenation.'4

"La formation de I'inconscient."

Ecrits, p. 69z; Sheridan, p. z88.

147
 Return to Freud

It is through castration that the phallus is constituted as a

signifier. Castration is thus nothing other than an effect upon
the subject caused by the falling out or striking down of the
signifier. This function of the phallus thus derives from the
structure of language conceived as a movement of signifiers, a
movement to which the subject is sub-jected. This subjection
not only splits the subject, but also supports and suspends it in
the incessant reiteration of an irremediable division: not be-
tween subject and object, but between enunciated and enuncia-
tion, signified and signifier. It is in this sense that the individual
subject is dependent upon the symbolic, or as Freud writes in a
letter to Fliess, upon that "prehistoric, unforgettable other, who
is never equaled by anyone later."5 This Other, however, is not
a person or even a subject, but instead that "somewhere else,"
that localized difference which emerges from and moves
towards the signifying chain. This negative place, this "dis-
location" comes to be occupied by various instances: first by the
mother as that utterly Other in the sense of the demand for love;
then by the father as the forbidding, castrating instance which
also introduces the law; and finally by the phallus "itself," as the
selfless, self-effacing mark that bars the place and splits the
subject. While the phallus thus marks the decisive moment of
bifurcation in the trajectory of the subject, it at the same time
remarks the structural condition of the latter's subjection — to
language as medium of articulation and of difference. After this
mark, there is no going back; and yet before it, there is no going
forward. As the phallic object of the mother's desire, a child is
marked by castration even before it is born.'6
In this respect, it may be said that structure, in a synchronic
sense, enjoys a certain priority over the diachrony of individual
development. But it is a structure of dislocation, not of deter-

5
S. Freud, Letter of December 6, 1896, in: The Complete Letters of Sigmund Freud
to Wilhelm Fliess, 1887—1904, translated and edited by Jeffrey Moussak'It
Masson, Cambridge & London: The Belknap Press of Harvard University
Press, 1985, 213.
lb In support of the priority of the symbolic over the empirical, genetic
development of the individual, Lacan often cites the passage from Freud's
discussion of "little Hans," in which Freud tells the boy that "long before h'
was in the world. . I had known that a little Hans would come, who would
.

be so fond of his mother that he would be bound to feel afraid of his

father SE, X, p. 41.

148
 The signification oft/ic phallus

mination, much less of predetermination. This is a priority that

is implicitly at work in the writings of Freud, for instance, in his
metapsychological reflections on "primal repression," a primacy
that cannot be construed coherently in chronological terms. This
form of repression is supposed to differentiate the psychic
apparatus and yet, as it is also a purely intrapsychic instance, it
must presuppose this differentiation. Yet as soon as one ceases
to consider primal repression as an intrasubjective product and
begins to see it as an effect of the signifier in the structuring of
the subject, the paradox becomes intelligible. It can be under-
stood as an allegory of the difference that renders discursive logic
and its linearity possible while simultaneously restricting and
displacing them. It is in this sense that Lacan can claim that
language is the condition of the unconscious, for the uncon-
scious "is" only (before) the bar that strikes and sub-jects the
subject to the signifier.'7
On the one hand, then, the phallus performs the function of
the signifier in general, as it falls out with the signified; on the
other hand, however — and we now come to its second moment
— it is a specific, determinate signifier, not a transcendental one. It

is from this perspective that one can begin to understand the

particular importance Freud attributed to castration vis-à-vis
other experiences of separation, such as the loss of the nipple or
the production of feces. What sets the phallus apart is a
particular fixation on a signified, which, to be sure, is the case
for every particular or determinate signifier. This means that the
imaginary function of the phallus — or of anything else, for that
matter — can never be wholly surpassed by its symbolic function.
For the imaginary bars the way, not just in the sense of blocking
the way, but also in that of staking it out, demarcating it and
fixing its trajectory. Without such an imaginary bar, the phallus
would be nothing but a "pure signifier," and we would be
unable to speak of it or name it in any other way. It would
thereby dissolve into a diaphany so pure that the symbolic
function itself would disappear. For the symbolic to function,
the signifier can never be pure or self-identical, but instead must
17
"Le langage est Ia condition de l'inconscient," Lacan asserts in Radiophonie
not the reverse. If one can speak of an
(Scilicet 213, Paris, 1970, p. ç8) — and
ontology, therefore, in respect to Lacan, it is not one of the Unconscious, but
of language qua signifying function.

'49
 Return to Freud

always be slightly out of sync, slightly deranged, slightly:

imaginary. This is the imaginary moment of the phallus: its
localization and materialization as the representative of the male
member; and since it is imaginary, it can never entirely fulfill its
symbolic function. Some other imaginary determination is
always possible, indeed, inevitable. That is, the possibility of a
substitution is inevitable. It is this necessary possibility that
marks the symbolic moment of the imaginary, contaminating it
no less than it, in turn, contaminates the symbolic that it
nevertheless renders possible.
Which is why, as the (in part imaginary) signifier of the
signifier, the phallus displays the limits of formalization, and at
the same time the necessity of naming; for it must be
determined as a particular mark, linked to a particular content
and context, in short: to a signified. The peculiarity of this
signified, however, resides in the fact that, within the economy
and development of the subject, it indicates that the signified
can appear only as an effect of the signifier. As the particular
signifier (of a specific signified), the phallus can thus take its
place in numerous "symbolic equations": it may be equated
with the child, as "the little one"8; with excrement, "as some-
thing that can be detached from the body" (i.e. the phallus
menaced by the threat of castration); with the body of the
mother, as that which is lacking, and with money, as that which
circulates. Its representation by the Ancients as an erect penis
points just as much to the movement and tension of desire, as it
does, negatively and implicitly, to a detumescence marking the
frailty and evanescence of pleasure. And even if it might be
difficult to verify the etymological assumption that "phallos"
derives from "phanein," this association reminds us of its
indubitably intimate and ambivalent relationship to the imagin-
ary realm of phenomenality, as well as to that of phantasy and
phantasm. In this respect, the phallus is the phenomenon that is
conspicuously absent: it is the conspicuousness of absence.
Perhaps, then, instead of trying of find out "what the phallus
is", we would do better to ask, "What about the phallus?"
"How do things stand with it?" To this the response will he,
18
S. Freud, "On transformations of instincts as exemplified in anal eroticism.
SE, xvii, p. iz8.

150
 The ion of the phallus

about the phallus stand circum-stances, which cause it to fall

due constantly. Because its stance depends upon circumstance,
it stands to reason that it must fall due. For it marks an out-
standing debt, a debt out of all proportion that can only be
settled — i.e. deferred — according to the laws of the symbolic.
Thus the phallus "is" the "Fall" itself, fortunate or unfortunate:
the case of the signifier falling out with(in) the signified. The
aura of the phallus is this fallout.
In the light of this aura, we see why Lacan's response to the
charges directed against the theory of the phallus is unequivoc-
al:the phallocentrism of Freud aims at deconstructing all
"centrism," insofar as the phallus stages the drama of the
signifier. As the phallus, the signifier is struck down as the
signified. The signified is the signifier struck down. This act of
determination is violent and illicit, and yet as such, it is the
origin of all law. The striking down of the signifier is performed
in the name of the father, but the "father" here names the name
of a name that has no present or presentable reference. What it
names is the interminable circulation that it terminates, precise-
ly through being struck down. This is why the phallus does not
debut as the product of a particular culture. Rather, it is part of
the theater of language, the condition of "culture" as well as of
"nature," as of everything that bears a name. Yet no one knows
better than Lacan, perhaps, that the phallus nevertheless
remains a particular signifier, tied to the margin of a text. Of a
text, and not of the text. Which is why no formalization can ever
entirely replace reading. And why reading never takes place in
general.
Reading in this sense — and it is the sense of the Return to
Freud announced by Lacan — means relearning how to be struck
by the signifier. And by its stage. In the theater of the
unconscious, one never gets over being stage-struck.

'5'
 Appendix A
+++ ++ + + +++ +++ ++ + + + + ++++ +++++ ++++ ++++ +

Beyond anxiety: the witch's letter

The incompleteness in the story of the Fall, the fact that anybody could
have the idea of telling Adam something he could not possibly
understand, disappears once we consider that the speaker is language,
and that therefore it is Adam himself who speaks. That leaves the
serpent. I. ] And here I must confess that I am incapable of attaching
. .

any definite thought whatsoever to it.

That the man of science ought to forget himself is entirely true;

nevertheless it is therefore also very fortunate that sin is no scientific
problem, and thus no man of science has an obligation F. .]to forget
.

how sin came into the world. If this is what he wants to do, if he
magnanimously wants to forget himself in the zeal to explain all of
humanity, he will become as comical as [the Copenhagen booksellerj
Soldin, who, in a fit of self-oblivious enthusiasm, and carried away by
his chatter, has to ask his wife: "Rebecca, am I the one who is
speaking?"
S. Kierkegaard, The Concept of Anxieti,'

I
In the "Descriptive index of major concepts" that Jacques-Alain
Miller assembled to guide readers of the Ecrits, a term is missing
which, indeed, occurs infrequently in the volume. The term is
angoisse, anxiety, and its infrequency in the 1966 collection of
texts deserves more consideration that it has received. For it is
anything but self-evident. Three years previously, in 1962—63,
Lacan had lectured for an entire year on the subject. And to
judge from his opening remarks — as they are recorded in the
unauthorized, but nevertheless widely circulated text to which I
shall refer, while awaiting its publication as volume x of Le
Séminaire — the importance of anxiety as a problem was no more
self-evident at the time than it has proved to be since. The choice
of it as the subject of the year's lectures having apparently come

152
 The witch's letter

as a surprise to many of his listeners, Lacan begins by

emphasizing its necessity within the context of his thought:
You shall see, I believe, that anxiety is precisely the point of rendez-
vous where all of my previous discourse awaits you, including a certain
number of terms which up to now may have seemed to you to be
insufficiently related. You will see how anxiety is the terrain where
these terms tie in with each other and thereby assume more clearly
their place.'
If anxiety was thus to supply the "terrain" upon which the
Lacanian discourse was to demonstrate its coherence, it is not
entirely suprising if the "setting" has been eclipsed by the
scenario, It is as if the destiny of the problem reflected the
particular theoretical role it was called upon to play: that of
serving as an introduction to another problem, one more
specifically and (in the meanwhile) more recognizably Lacanian,
that of the lack:
In other words, anxiety introduces us, with the greatest possible accent
of communicability, to the function of the lack or fault (In fonction du
man que), inasmuch as it is radical for our field.
The topic of anxiety, then, which allows the hitherto somewhat
disparate membra disjecta of Lacan's terminology to come
together and take their proper places, at the same time and in
the same process has its own place clearly defined: that not
merely of a "terrain," but of a passage-way. This passage must be
negotiated, but it also must in a certain sense be left behind. For
the work to be done, and in particular the work of the analyst,
takes place elsewhere, as Lacan recalls at the very end of these
lectures:
I have often asked you what the analyst's desire must be so that work
might be possible there, where we are trying to bring things, beyond the
limit of anxiety. (7.3.63)
The lectures on anxiety thus have the task of establishing this
"limit," and thereby of preparing its overcoming: "Anxiety can

"Vous le verrez, je pense, l'angoisse est très precisement le point de

rendez-vous oü vous attend tout Ce qu'il était de mon discours antérieur et oü
s'attendent entre eux un certain nombre de termes qul ont Pu jusqu'a present
ne pas vous apparaitre suffisamment conjoints. Vous verrez sur Ce terrain de
l'angoisse, comment, a se nouer plus étroitement. chacun prendra encore
mieux sa place." (rI.14.6z)

'53
 Appendix A

only be surmounted when the Other (capital 0) has named

itself." This self-naming of the Other, Lacan announced, would
comprise the topic of his lectures of the following year, which
were to deal with the Names of the Father. But as is well known,
this project had to be deferred for a decade, following the
interruption of his teaching at Sainte Anne. When finally the
time had arrived for the Other to name itself, les noms du père had
now been doubled by "les non-dupes errent." And anxiety,
presumably, had long since been left behind.

Lacan was not the first who had sought to put anxiety in its
(proper) place, to be sure. Freud himself had already made a
similar attempt. With dubious success. For Freud, too, anxiety
presented itself as an unavoidable station along the way to the
new discourse he sought to establish. In one of his earliest
essays, he endeavored to demarcate "anxiety neurosis" from
the then more familiar and more general category of "neurasthe-
nia." Libido that could not be discharged through sexual
activity, he argued, was transformed into anxiety. Subsequen-
tly, Freud modified this thesis in the context of his later theory,
asserting that anxiety was the reaction of the ego to danger, and
above all, to that of an object-loss (of which "castration" was
considered to be the exemplary instance). In the later perspec-
tive, anxiety is described less as a disturbance of the psyche,
than as its attempt to protect itself from such disturbances.
Freud's discussion of anxiety thus turns upon a question that
it will never entirely resolve: Is anxiety a constitutive process by
which the psyche maintains its coherence and identity, or does
it ultimately entail their dissolution? To be sure, as Freud
pursues the notion of "danger," to which anxiety is said to be a
reaction, the opposition — between identity and nonidentity,
internal and external — that such a question presupposes, has to
be refined. A danger, by definition, entails a certain exteriority
with respect to that which it endangers, an aspect that Freud
retains in his notion of "real" or objective danger. The morc
realistically danger is considered, however, the less it can bc
used to define anxiety, which, as Freud laconically remarks, is
rarely the most realistic, in the sense of effective, response to au

154
 The witch's letter

objective menace. The danger to which anxiety may be

considered a response, therefore, cannot be defined purely from
its realistic, external side; it must in some way be related to that
which it endangers, the functioning of the I. If the latter consists
above all in the production and maintenance of stable cathexes,
in particular of a visual nature, this explains why Freud defines
the danger faced by anxiety in terms not merely of the "loss of
objects," — which would still suggest a relation to objective
reality — but rather, more rigorously, as a loss of perception, a
Wahrnehmungsverlust, since it is in such a loss that a true danger
to the ego is articulated.
The notion of "perceptual loss" can be used to describe the
peculiar indeterminacy that, according to Freud — who here
follows a venerable tradition — distinguishes anxiety from fear.
The danger of a loss of perception is determinate, inasmuch as it
is necessarily attached to familiar images, situations and
representations. The threat of "losing" these, by contrast,
entails much more than the possibility of their simple disappear-
ance. The danger of perceptual loss cannot be understood
strictly in terms of the negation or absence of a particular
perception, although this can often be the triggering factor. The
danger itself, however, is not perceptual; rather, it involves
what Freud calls "trauma": a shock that prevents the I from
retaining and maintaining fixed cathexes (that is, energy more
or less stably invested in representations).
In this context, it should not be forgotten that such a
"trauma" is only the reverse side of the process by which not
only "perceptions," but also "desires" themselves are consti-
tuted. If the essence of danger, qua "perceptual loss", can be
described by Freud as that of the trauma, it is because the
formation of what The Interpretation of Dreams calls "perceptual
identities" (Wahrnehmungsidentitaten), and hence of "wishes" (if
not of desire itself), entails the very relationship out of which the
trauma emerges: the radical divergence of representation and
cation. A perceptual identity, it will be remembered, arises
when a memory trace or image is reproduced in a quasi-
hallucinatory manner, in order to repeat the experience of
satisfaction with which it has remained associated. The radical
discrepancy between the representation, and that which it
signifies, but cannot represent — for the simple reason that a

155
 Appendix A

shift in tension is not as such representable — is the structural

essence both of the heterogeneity that separates all desire from
its "object," and of the "trauma" discerned by Freud as con-
stituting the ultimate danger to which anxiety responds.
Considered from this perspective, it is precisely Freud's
inability to resolve the problem in terms of the alternative
already described — Is anxiety functional or dysfunctional? Is the
danger to which it responds essentially external or internal? —
that constitutes the interest of his discussion. The reality of
anxiety and of the danger to which it reacts, emerges as neither
simply external nor internal, neither straightforwardly func-
tional, nor dysfunctional; the functioning of the psyche is,
intrinsically, as it were, bound up with an irreducible exterior-
ity, with an alterity that it simultaneously denies and affirms. It
denies it by giving it the form of an object that can be
represented, and it affirms it by treating that representation as
the mere material of a "cathexis," of a libidinal "investment,"
the signification of which must be sought elsewhere.
But, the reader will doubtless want to protest, just who or
what is this "it" being referred to? Does not such an account
confuse, or at least conflate, Freud's theoretical discourse with
its object, the psyche? To this charge, I would reply as does
Freud at the beginning of Chapter 7 of the Interpretation of
Dreams, concerning the criticism that all accounts of dreams
only falsify the latter. Freud affirms the statement as a descrip-
tion (yes, the recounting of the dream does disfigure it), but
denies it any critical value (no, such disfiguration does not
disqualify itself as an access to the dream). This is the distinctive
involvement of what I have elsewhere called "psychoanalytical
thinldng."2 The "it" of the discourse of psychoanalytical theory,
like the Id that theory seeks to describe, entails a region of
indeterminacy in which object and subject, signifier and
signified, far from being clearly distinct, tend to redouble one
another in a play of mirrors and shadows that one should not
too hastily disqualify as "imaginary." It is this area of in-
determinacy that Freud's discourse highlights, precisely by

2
See Samuel Weber, The Legend of Freud, University of Minneapolis Press:
Minneapolis, 1982.

156
 The witch's letter

demonstrating, more implicitly than explicitly, how and why

the oppositionally structured alternative of "real" and "neuro-
tic," of "external and internal," of functional and dysfunctional,
is incapable of articulating the dynamics of anxiety. At the same
time, the synthetic aim of comprehension makes it inevitable
that the same discourse seeks to define and delimit that unruly
"area" and thus to reinstate precisely the distinctions that it also
tends to neutralize. The result is a text like Inhibition, Symptom,
Anxiety, in which the argument follows a seesaw movement of
self-revocation not unlike that described by Derrida in his
reading of Beyond the Pleasure Principle in The Post Card.3 The
Freudian theory of anxiety is perhaps most revealing in the
manner in which it comes to resemble what it is ostensibly
intent on describing: the tension between the traumatic tendency
of object-cathexes to dissolve, and the counter-effort of the
Ego to arrest and channel that tendency. If, despite such
resemblance, Freud's discourse on anxiety avoids becoming a
discourse of anxiety, it is precisely because it allows its own
movement to go on in function of that tension, rather than
seeking to protect it from it.

This unresolved ambiguity and tension in the Freudian theory

of anxiety reflects the manner in which the latter remains bound
to the I: bound up, that is, with the effort of the subject to bind
itself to and with representations. Or, to put it in more tradi-
tional terms: for Freud, anxiety is inseparable from the effort of
the psyche to organize itself and to maintain its identity, how-
ever problematic, divided, and conflictual such identity will
necessarily have to be. In Lacan's lectures, by contrast, the accent
rests upon the need to go beyond the notion of the I — not neces-
sarily in the sense that term has in the writings of Freud, but
rather in those of the Ego-Psychology that dominated the Inter-
national Psychoanalytical Association in the early sixties. To
read Freud properly for Lacan is to uncover the necessity of

Jacques Derrida, The Post Card: From Socrates to Freud and Beyond, trans. Alan
Bass, University of Chicago Press: Chicago, 1987.

'57
 Appendix A

looking beyond the Ego in order to understand the origins and

structure of anxiety:
How does it happen that the movement of reflection [.. I brought .

[...] first Rank and then Freud, who follows him on this point, k
situate the origin of anxiety at a pre-specular, pre-autoerotic level, at
that level of birth where no one, in the entire analytical chorus, would
ever dream of speaking of the constitution of an Ego. There is
something here which proves that even if it is possible, once the Ego is
constituted, to define anxiety as a signal [.. .1 in the Ego, this can
hardly be exhaustive.4 (1.13.63)
Lacan returns to this point in a later lecture:
The Ego is the site of the signal. But it is not for the Ego that the signal is
given. [...] It is so that the subject — it cannot be called otherwise -
may be alerted to [soit averti] something. (1.17.63)

That "something," to be sure, is not merely an "original lack or

fault," but rather the object o, the "cause" of desire and also the
elusive embodiment of the Other. The latter can only embody
itself — that is, its heterogeneous unself — by falling out with
itself, and this leaves behind an "object" that is other than
the object as we usually construe it: that is, as an object that can
be represented, or at least symbolized. The object o is, by
contrast, precisely not accessible to symbolization, it is what
falls out of the signifying chain, or what it leaves behind. Nor
can this strange object be reflected in a mirror, since, like the
Moebius strip with which Lacan at times compares it, it has no
definable, fixed border; indeed, its structure seems to be that of
such a twisted border. More precisely, it seems to be situated at
the intersection of the Imaginary and the Symbolic, and to serve
the function of guaranteeing "the alterity of the Other":
With respect to this Other, dependent upon this Other, the subject
inscribes itself as a quotient; it is marked by the unique trait (le trait
unaire) of the signifier in the field of the Other. Well, it is not for nothing
if I can put it this way: that it slices up the Other (met l'Autre en rondelles).
This is a remainder, in the sense of a division, a residue. This
remainder, this ultimate other, this irrational [number], this proof and
sole guarantee, finally, of the altenty of the Other, is the o. (11.21.61)

Lacan finds support for this assertion in "the phenomena of depersonaliza-

tion" known to accompany severe anxiety, phenomena that are "the most
contrary to the structure of the Ego as such."
 The witch's letter

[It is just this residue, produced by the precipitation of the

signifier, that the child encounters when it averts its gaze from
the mirror and turns around in search of a verification that
might "decant" the moment of recognition from the jubilation
that accompanies it; what it finds is the look of the other, which
in turn, as Lacan emphasizes, is one of the paradigmatic
manifestations of the object o. [See above, p. i i 6ff}
Since this o, object of desire par excellence, safeguards the
alterity of the Other by localizing the lack/fault (le manque) in its
place, the Freudian notion of object-loss is, for Lacan, mislead-
ing. The object o is constituted rather by its very ability to be lost,
or more precisely, to be "ceded." Anxiety, therefore, arises not
from the loss of an object, but rather from the loss of this loss,
or, as Lacan puts it, "when the lack comes to be lacking" (quand
le man que vient a man quer). At the same time, such a process is
necessarily associated with the o, since it has the paradoxical
function of embodying what precisely cannot have a body
without being distorted and defaced.5 The "loss" that triggers
anxiety, then, is not that of an object that once was possessed,
but rather the effacement, through objectification, of what
Lacan refers to as the "lack."
Anxiety results, then, from the process by which the Other

Lacan describes this self-effacement of the man que in terms obviously indebted
to Heidegger's description of the "disclosure" of Being, which conceals itself
all the more thoroughly in and through revealing itself. The "lack of a
lack" similarly echoes Heidegger's account of the (self-)effacement of the
ontological difference between Being and beings. It should be added that
Lacan's relation to Heidegger itself appears to follow a similar pattern.
Heidegger is mentioned at the beginning of the lectures on anxiety, but only
in order to remark that the latter's questioning of the Being of beings, and in
particular, his thematization of Being-toward-Death, "does not really take the
way of anxiety," but is oriented rather toward the question of care (Sorge,
souci). (is. 54.62) That Lacan should approach the problem of anxiety in terms
of lack and of fault; that the o is determined as a mode of falling; that the
uncanny is described by him as the "key" to anxiety, which in turn is related
essentially to the effect of isolation: all this will be quite familiar to readers of
Being and Time, § 40. But Lacan's posture toward Heidegger here merely
repeats that of Heidegger toward his own forerunners, in particular toward
Kierkegaard. Whether or not such an effort to blur the traces of a debt is itself
not part and parcel of the theory (and practice) of anxiety — as the anxiety of
theory itself, beyond any particular set of "influences" — is a question that
deserves more attention than can be given it here. I have discussed this
question in a different context in the essay "The debts of deconstruction" in:
Samuel Weber, Interpretation and Institution, University of Minnesota Press:
Minneapolis, 5986.

'59
 Appendix A

paradox of this is that only such effacement and defacing can

guarantee the place of the Other. For without the intervention
the o, a situation would arise that Lacan describes as being
manic:

In mania what is involved is the non-functioning of the o, and not

merely its misrecognifion. It is through this that the subject is no longer
weighed down (lesté) by any o, and sometimes, without any possibility
of freedom, is delivered over to the infinite metonymy and pure play of
the signifying chain. (7.3.63)

The difficulties involved in the notion of the o are thus the

measure of its importance. It holds the line, as it were, between
the symbolic gone wild and the imaginary gone wrong, between
the infinite metonymizing of signifiers and the unending
mirroring of narcissistic identification. And it is in anxiety that
the subject encounters this peculiar "object," not as something
that is "over against," but rather implied by its own position in
respect to the desire of the Other. It is from this perspective that
Lacan approaches the significance of "danger" in regard to
anxiety. In contrast to Freud, for whom the notion is inseparably
bound up with the traumatic "helplessness" of the I in the face
of excessive quantities of excitation, Lacan construes danger as
related to the desire of the Other insofar as it addresses the
subject as its "cause," thereby placing it radically into question:
[The subject] is alerted (by the anxiety-signal) to something that is a
desire [.. ] which concerns nothing other than my being itself, that is
.

to say, which puts me in question [.. .] which does not address itselt
to me as to someone present but rather as to someone who has been
awaited (attendu), or even more, as to someone who has been lost, and
whose loss is also urged (qui sollicite ma perte), so that the Other may
rediscover itself there — that is anxiety. The desire of the Other does not
recognize me, as Hegel believed [. . .1 It challenges me (me met en
Cause), questioning me at the very root of my own desire as o, as causu
of this desire and not as object. And it is because this entails a relation
of antecedence, a temporal relation, that I can do nothing to break thk
hold other than enter into it. It is this temporal dimension that is
anxiety and that is also the dimension of analysis. (1.17.63)

One begins to envisage here just why Lacan should attribut&

such central significance to anxiety and also why this sign i
ficance should require a perspective that looks "beyond tht
i 6o
 The witch's letter

limits of anxiety." Anxiety assumes a decisive position in

Lacan's "enseignement" — his teaching, but also his doctrine —
because it confronts the subject with "the truth of the lack!
fault," in the form of the object o as "cause" of desire. This
desire, in turn, emanates from an other, whose alterity admits of
no further derivation. The encounter of the subject with this
irreducible heterogeneity, through the intermediary of the o,
must bear the mark of a certain non-knowledge. The latter is
certain, insofar as anxiety, as Lacan puts it, is what "does not
deceive" (ce qui ne trompe pas). But that about which anxiety does
not deceive is the fact that one can know nothing of the
particular, determinate o that the subject represents for the
desire of the Other. This is summarized by Lacan at the end of
his lectures on anxiety:
anxiety manifests itself clearly from the very beginning as relating — in a
complex manner — to the desire of the Other. From the very first I have
indicated that the anxiety-producing function of the desire of the Other
was tied to the fact that I do not know what object o I am for this desire.
(7.3.63)

The certain non-knowledge, the fact that I cannot know just

what I am for the desire of the other, indicates clearly that the
object of anxiety cannot be a cognitive object. In this sense,
Lacan can describe anxiety as that which puts cognition into
question (or, more literally, "into cause": la mise en cause de la
connaissance) (5.8.63). For cognition, at least as traditionally
construed, requires objects constructed according to the model
of narcissistic identification with the mirror-image, with the
"i(o)" (the imaginary object). By contrast, the o, which the
subject encounters in and as anxiety, is precisely that which
cannot be contained in the mirror-image and which Lacan
compares on the one hand to the frame of a window, citing the
drama of the Wolfman, and on the other, to the curtain of a
theater. The o, far from being an object accessible to cognition, is
that which allows a certain place to demarcate itself. At the same
time, the process of demarcation, insofar as it is bound to the o,
is itself in a certain sense incomprehensible.
Where, then, does this leave the discourse "on" or "about"
anxiety? The discourse that seeks to make out the proper place
of its subject, to fix its borders, map out its terrain? How
i6i
 Appendix A

hospitable a terrain can it offer to disparate terms seeking

bind themselves together into a more coherent whole? Is such
scenario even thinkable other than as the acting-out of anxiety
itself? And if it is such, what will its consequences be?
Already in the initial lecture of the series, Lacan indicates that
he is in no way unaware of this problem:
If you know therefore how to come to terms with anxiety (vous arrangci
avec l'angoisse), it may help us to move forwards if we try to see how
that works; as for me, I wouldn't know how to introduce it without
determining it in some way or other — and therein lies perhaps a snag
One should not come to terms too quickly with anxiety. (11 ne faut pas
que je l'arrange trop vite). (i 1.14.61)

That this remark should be followed immediately by a reference

to Kierkegaard and to "existentialist" philosophy, from which
Lacan seeks to distinguish his own approach, does not alter
the fact that the name of the Danish writer recurs throughout
these lectures whenever the theoretical status of anxiety is at
issue. Yet it is only in the final session that this problem is fully
addressed:
I do not know if one has been sufficiently aware of Kierkegaard's
audacity in speaking of the concept of anxiety. What can this possibly
mean if not the affirmation that, either the concept functions in a
Hegelian manner, entailing symbolically a veritable grasp of the real; or,
the sole grasp (prise) that we can have — and it is here that one must
choose — is that afforded us by anxiety, the sole and thus ultimate
apprehension of all reality (seule apprehension et comme telle de toute
réalité). (7.3.63)

If an adequate discussion of the choice embodied in Lacan's

lectures on anxiety must await their publication, one can still, in
a preliminary way, question the manner in which he envisages
this choice: can one enclose anxiety in an alternative of this kind?
If there is a "concept" of anxiety, must its "hold" or "grasp" —
the Griff of the — be defined either in terms of anxiety or

of its Hegelian Aujhebung in thought? The word used by Lacan

instead of the German is prise, which, as the participle of
prendre, to take, suggests not merely "getting a handle on", but
also a pinch, as of salt or sugar. Thus, the word itself is
suspended in the alternative set out by Lacan: either anxiety
can be considered a concept through which we get a true hold

i6z
 The witch's letter

upon the real, or we are beholden to it, we feel its pinch.

Another possible approach can be retraced in Freud's essay,
Inhibition, Symptom, Anxiety, beginning with its remarkable title
that establishes a sequence but not a conceptual or totalizing
synthesis. The serialization of the problem also leads Freud to
question the usefulness of the oppositional logic generally used
in the elaboration of concepts:
It is time to reflect. We are apparently searching for an insight that will
reveal to us the essence of anxiety; we are looking for an either-or that
will distinguish truth from error. But that is hard to come by, anxiety is
not easy to grasp. So far we have reached nothing but contradictions,
between which it was not possible to choose. I now propose that we
proceed differently: we will try to collect, impartially, everything that
we can say about anxiety, and in so doing renounce the expectation of a
new synthesis.6
Since I have discussed this text at length elsewhere,7 I will limit
myself here to noting that Freud's discussion of the "indeter-
minacy" of anxiety does not hesitate to inscribe that indeter-
minacy in the text in ways that conjure up the risk of
compromising the value of its theoretical argumentation, consi-
dered, that is, from the standpoint of traditional scientific
standards. If the core of anxiety relates to what I have described
as the "traumatic" discrepancy between representation and
signification, the same divergence is assumed by Freud as a
principle of his metapsychological discourse: not as the small o,
but as that large "X," which we take over into each new formula
without really knowing just what it represents.8
Yet another approach, however, is suggested by the following
remark of Kierkegaard:
Anxiety is a determination of the dreaming spirit. [.. . ]Awake, the
difference between myself and my other is posited; sleeping, it is
suspended, dreaming it is a signified nothing. The actuality of the spirit
constantly shows itself as a figure that tries out its possibility but which
disappears as soon as the spirit seeks to take hold of it: a nothing, that
can only bring anxiety.9

6 S. Freud, Inhibition, Symptom, Anxiety, W. W. Norton: New York, 1959, p. 58.

The Legend of Freud, pp. 57ft.
8 S. Freud, Beyond the Pleasure Principle, Bantam Books: New York, 1972, p. 58.
S. Kierkegaard, The Concept of Anxiety, Princeton University Press: Princeton,
1980, pp. translation modified.

163
 Appendix A

The discourse of Lacan, in contrast to that of both Kierkegaard

and Freud, is one which seeks and offers "guarantees." The
quasi-concept of the o, "guarantee of the otherness of the
Other," also opens the perspective of that move "beyond the
limits of anxiety," so that another sort of "work can be
possible," that of analysis itself:
To be sure, this requires that the analyst be one who has been able, in
whatever small measure, by whatever way, by whatever edge, to cause
his desire to return to that irreducible o and thereby provide the
question of the concept of anxiety with a real guarantee. (7.3.63)

"Anxiety," Lacan assured his listeners, "is only overcome

where the Other names itself." (7.3.63) In the discourse that
follows the lectures on anxiety, this naming will make ever less
mention of anxiety. The term drops out of Lacan's discourse, in
large measure. Is this a sign that the problem it entails has
indeed been overcome? Or could it be that by thus dropping
out, anxiety assumes the role of the o in Lacanian discourse
itself? In any case, it would not have been the first time that such
an event had occurred. In the lectures on anxiety, Lacan calls
attention to the way in which Freud, by missing the point in his
analysis of Dora, by misconstruing the character of feminine
desire, reveals something that a more insightful procedure
might well have obscured: "La chose freudienne," — the Freudian
thing — "is what Freud let fall, what he overlooked, the matter
he dropped, but it is precisely this that, in the guise of all of us,
still leads the hunt after his death." (1.23.63)
From this, however, to infer that anxiety might be a name for
Ia chose lacanienne, would be to offer a conjecture that little if any
explicit evidence would seem to support, much less authorize.
Except, perhaps, if one recalls the manner in which Lacan
himself construed authorization:
Let us proceed from the conception of the Other as the site of the
signifier. In the context of this site, no authoritative statement can have
any guarantee other than its very enunciation, since it is futile for it to
look for any such in another signifier, which under no circumstances
can appear outside of this site. That is what we formulate in saying that
there is no metalanguage that might be spoken, or, more aphoristically,
that there is no Other of the Other. The legislator, who claims to make
up for this by erecting the Law, does so as an impostor.

164
 The witch's letter

But not so the Law itself, nor whoever acts on its authority.'0
The question raised by Lacan's conclusion is: how can one act on
the authority of a Law that derives, as it were, from a site from
which one is barred? How, unless it is through the bars
themselves, which demarcate an area in which we, as subjects,
first person plural or singular, never quite succeed in taking
(our) place. Let us therefore in conclusion, in lieu of a
conclusion, take a look through, and at, these bars, albeit in an
area that seems quite remote from that we have been discussing.
Peter Greenaway's film, The Draught man's Contract, tells the
enigmatic story of a cocky, successful young draughtsman,
Neville, who accepts an offer to make twelve sketches of the
estate of Lord Herbert, in exchange not just for money, but for
services to be rendered him by Lady Herbert. As his work
progresses, a number of strange, unaccountable objects crop up
in his, and our, field of vision, a field that is determined by an
apparatus that he carries around with him, and that he uses to
frame and construct his drawings. Wholly absorbed by his
work, and by his relation to Lady Herbert, Neville is taken
entirely by surprise when the daughter of Lady Herbert, Miss
Sarah Talman, suggests to him that the objects mentioned may
not be as inconsequential, nor his situation as simple, as he
seems to believe. On the contrary, she continues, these curious
objects could well be indications of a conspiracy to do away with
Lord Herbert, in which Nevile might now be involved,
however unwittingly.
What is of particular interest in the context of our discussion,
however, is the "theory" — Miss Talman's word — with which
she seeks to explain to her increasingly disconcerted listener,
"Partons de Ia conception de I'Autre comme du lieu du signifiant. Tout
énoncé d'autorité n'y a d'autre garantie que son énonciation méme, car II est
vain qu'il le cherche dans un autre signifiant, lequel d'aucune facon ne saurait
apparaItre hors de ce lieu. Ce que nous formulons a dire qu'il n'y a pas de
métalangage qui puisse être parle, plus aphoristiquement: qu'il n'y a pas
d'Autre de l'Autre. C'est en imposteur que se présente pour y suppleer, le
Legislateur (celui qui pretend enger Ia Loi). Mais non pas Ia Loi elle-même,
non plus que celui qui s'en autorise." Ecrita, p. 813. Sheridan, pp. 310—311.
The designation of the Legislator as an "impostor" goes back at least to
Rousseau's Social Contract. For a discussion of this text, see Paul de Man,
"Promises," in Allegories of Reading, Yale University Press: New Haven, 1979,
and Samuel Weber, "In the name of the law," Cardozo Review, vol. ii,
fl05. 5—6, july—August, 1990, esp. 1533—38.

165
 Appendix A

just why and how he might make an easy prey for possible
conspirators:

Mr. Neville, I have come to the conclusion that a really intelligent man
would make a mediocre painter. For painting requires a certain
blindness, the partial refusal to take all the possibilities into account.
An intelligent man will know more about what he draws than what he
sees with his eyes, and in the space between knowing and seeing, he
feels himself more and more.. . confined, incapable of following out
an idea with any consequence, fearing that the judicious will not take
pleasure in his work, if, in addition to what he knows, he does not put
into it what they know. You, Mr. Neville, if you are an intelligent and
therefore a mediocre draughtsman, would have to welcome the fact
that a theory such as I have proposed could indeed be constructed
based upon the clues contained in your paintings. If, however, you are,
as I have heard people say, a talented draughtsman, then I could
imagine that you might assume that the objects which I have brought to
your attention testify to no plan, indicate no deviousness and do not
amount to any sort of accusation.

Sarah Talman's theory, taken by itself, would hardly amount to

more than a clever reworking of Socrates' mockery of the
rhapsode, Ion, who sings about things without knowing what
in the world they really are. But the shot itself accentuates a
dimension that is not simply that of the Platonic mimesis: the
dimension of space itself, or rather, the interval between space
and place, which is not necessarily equivalent to "the space
between knowing and seeing." The shot shows Neville in
profile, silhouetted in the lower left foreground, while Lady
Talman stands above and behind him, addressing not so much
Nevile as the spectators, camera, or both. But what is most
striking in this frame is precisely its framing: almost, but not
quite convergent with the actual frame of the shot is Neville's
frame, which forms a second, heavy black border within which
the artist's head is located, whereas Lady Talman's is outside.
And yet, not entirely either, for the plumed column of her hat
rises first above Neville's frame, and then right out of the picture
as such. The clever and 'talented' artist, Neville, takes the frame
that he carries around with him for granted, as an instrument.
He knows no anxiety, except perhaps when the force of Lady
Talman's words begins to take hold. For the spectator, to be
sure, it is precisely the presence of those curious objects that

i66
 The witch's letter

don't "fit in," that raise the issue of the frame by pointing
elsewhere, to another place, to the "space between knowing
and seeing" within which the "intelligent man" feels increas-
ingly "confined."
Anxiety, it may be recalled — Nietzsche, among others, made
the point — is related etymologically to the idea of "confine-
ment" (Angst, from Enge: narrow): to "lack," if you will, but
above all, to a lack of breath. Anxiety is perhaps what one feels
when the world reveals itself to be caught up in the space
between two frames: a doubled frame, or one that is split, who
can tell? Frames in which we are no less framed than is poor
Nevile.
A final remark. Toward the conclusion of his thesis on The
Concept of Anxiety, Kierkegaard comes up with a striking figure
to describe anxiety, "the pivot upon which everything turns:"11
Anxiety discovers destiny, but just when the individual wants to put
his trust in destiny, anxiety turns around and takes destiny away,
because destiny is like anxiety, and anxiety, like possibility, is a
"magic" picture.'2
The Danish word that is here rendered as "magic picture" is:
Heksebrev, literally: witch's letter. A "witch's letter" is a "set of
picture segments of people and animals that recombine when
unfolded and turned."13 If we ever get "beyond the limits of
anxiety," beyond transference, or to any of the other "beyonds"
one might conceive, we might well find a witch's letter waiting
there to greet us.

" The Concept of Anxiety, p. 43.

12
Ibid., p. 159.
The Concept of Anxiety, translator's note, p. 154.

i 67
 Appendix B

Transferring the heritage:

psychoanalysis and criticism

Shortly before Christmas, 1985, a brief article appeared on the

last page of the Parisian daily, Le Monde, under the headline:
JACQUES LACAN
"BELONGS" TO HIS SON-IN-LAW

The text of the story reads as follows:

On the rith day of December, 1985, the First Chamber of the Paris
Civil Court recognized the rights of Jacques-Alain Miller, as the
testamentary executor and trustee, [. over the work of Jacques
. . J

Lacan. J.-A. Miller, Lacan's son-in-law, and the Editions du Seuil, had
brought several charges against the association, APRES, for publishing
a transcribed version of the seminar of Jacques Lacan, on "Transfer-
ence," in its internal bulletin, Stécriture. [ . . .

The Association, APRES, is found guilty of copyright violation; the

court orders distribution of the bulletin to be stopped, existing copies to
be destroyed, and damages to be paid.
It should be noted, however, that the individual members of the
association are exonerated of any further responsibility. Moreover, the
only part of the sentence to be executed, for the time being, concerns
the ban upon distribution: the court leaves it up to the parties involved
to decide whether or not to publicize the verdict . . le
jugement). This decision, therefore, may perhaps not be the epilogue of
this "murky affair" (cette ténébreuse affaire).
The Balzacian allusion with which the article concludes was
hardly lost upon the readers of Le Monde: the "case" of the
purloined papers of Jacques Lacan raised the very issues around
which the writings of Balzac — but also those of contemporary
literary theory more generally — incessantly swirl: the question
of the rights of an author and, correlatively, that of the status of
a "work," literary or other.1 It is the enduring power of this
I have discussed the relation of author to work in Baizac's writings in

x68
 Psychoanalysis and criticism

question that makes the Paris case more than merely a "fait
divers," more, that is, than a French version of what in English
is called a "human interest story." Which is why this curious
affair merits consideration.
First, however, some necessary background information. The
association APRES, an acronym signifying "Association for the
Research and Establishment of the Seminars," was constituted
in 1983, two years after the death of Lacan, by researchers and
analysts including many former members of the Ecole Freudi-
enne de Paris, the institution founded by Lacan and then dis-
solved by him, amid general confusion and much protest. The
group described its purpose as that of "elaborating a theory of
the transition from the spoken to the written word of Lacan."
The results of its efforts are published in a bulletin which takes
its name from a word play of Lacan: "Stécriture." The groups
sees the "originality" of its "method" reflected in the produc-
tion of a "critical apparatus" and of a text, which, unlike the
authorized edition of Miller, does not efface the multiplicity of
sources that is at its origin. This multiple origin includes: the
stenographic record of Lacan's lectures, the notes of his
listeners, the many tape recordings that were made by them,
and, last but not least, the various interpretive interventions of
the editorial collective itself.
By thus retaining a certain textual plurality, or, if one prefers,
a certain intertextuality, in which not merely the speaker, Lacan,
is inscribed, but also his listeners and even certain of his
readers, Stécriture endeavors to produce "a collective version"
of the Seminar that is "as close as possible to Lacan."
The question, to be sure, is: just how close is close? Or rather,
given the nature of this particular case, just how close is just, or
at least, legal? How close can one get to Lacan without violating
French copyright law — this is the question that the editors, and
lawyer, of Stécriture seek to elaborate, if not to resolve.
"Stécriture does not pretend (pretend: claim) "to publish" Lacan
and thus to compete with the Editions du Seuil."2 But if

Unwrapping Balzac: A Study of"La Peau de Chagrin," University of Toronto Press:

Toronto/Buffalo/London, 1979.
2 The Editions du Seuil have published a number of Lacan's seminars and are
under contract to do yet others. The publishing house was also a co-plaintiff
with Miller in this case.

169
 Appendix B

Stécriture is not publishing "Lacan," what is it publishing under

the title, "Transference in its Subjective Disparity, its Putative
Situation, its Technical Excursions"? How, in short, can one
publish a text that comes as "close as possible to Lacan" without
infringing upon the "droits d'auteur," firmly in the hands of the
literary executor, Jacques-Alain Miller? There is only one
possibility: by contesting that there is any author at all, at least
in this particular case. This is precisely the way taken by
Stécriture.
The text from which I have been quoting, entitled "Who is the
author of 'Lacan's' Seminar?", begins by raising precisely this
question: "The spoken work constituted by the 'Seminar' of
Jacques Lacan poses, in a very particular manner, the question
of the 'right of the author'." Lacan, Stécriture argues, like
Foucault, never thought of himself as an "author", once even
going so far as to assert that, "contrary to my friend,
Levi-Strauss, I will not leave behind a work." Where there is no
work, however, there can be no author, the latter always being
defined as the originator or creator of the former. And if there is
one issue upon which both Miller and Stecriture agree, it is that
"the spoken work constituted by" Lacan's Seminar is not really
a work at all, or at least, not the work of an author. Stécriture
supports this assertion by referring to the peculiar nature of
Lacan's enseignement, his "teaching," which, it argues, did
nothing less than "put into practice the theory he developed."
One of the more conspicuous tenets of this theory is that the
subject receives its message in a more or less distorted — Lacan
says "inverted" — form, on the rebound, as it were, from the
Other that is in part constituted by its interlocutors. Given the
constitutive importance of such interaction, the oral teaching of
Lacan cannot be considered to be the sole or exclusive product
or property of an author. (In view of this argument it is hardly
surprising that the seminar chosen to serve as a test-case of this
approach would be that dealing with "transference." We will
have occasion to return to this matter shortly.
Stécriture might have strengthened this argument, theoreti-
cally if not legally, had it cited a passage from the essay entitled
"Subversion of the subject and dialectic of desire," which
indicates just how complex the issue of authorial rights can
become in a Lacanian perspective. In his essay, Lacan has been

I 70
 Psychoanalysis and criticism

elaborating the significance of what he calls "the paternal

function," in the light — or rather, in the chiaroscuro — of his "con-
ception of the Other as the site of the signifier." He then goes
on to describe the kind of legality that derives from this place:
Let us proceed from the conception of the Other as the site of the
signifier. In the context of this site, no authoritative statement can have
any guarantee other than its very enunciation, since it is futile for it to
look for any such in another signifier, which under no circumstances
can appear outside of this site. That is what we formulate in saying that
there is no metalanguage that might be spoken, or, more aphoristically,
that there is no Other of the Other. The legislator, who claims to make
up for this by erecting the law, does so as an impostor.3
How is it possible, however, to "act on" the "authority" of a
Law that is not so much made, as received, or, as Sheridan
translates it, "assumed"? How can one conceive of "authority"
without an author? How are we to be sure that such a law is
itself legitimate? What if it were only powerful, based on more or
less opaque force? Would it still be legally binding?4
The French expression used by Lacan, s'en autoriser, is
reflexive, but the turn back upon itself that it articulates turns
the subject inside-out, as it were, making it dependent upon a
place that is beyond the compass of subjectivity. It is the turn
itself that appears to constitute the authoritative move, and it
turns out to be a linguistic turn. For the only guarantee of any
statement of authority is, as we have read, "its very enuncia-
tion." Yet, how is this enunciation to be determined, if its only
law is that of difference, or, in Lacanian terminology, that of the
Other? In short, what is the relationship between a determinate
statement, an énoncé, and an énonciation, the process of utter-
ance? How do the rules of the State, or of a certain stasis,
relate to the dynamic play of signification?
The decision of the Paris Court in favor of the plaintiffs and
against Stécriture, offers one response to this question. First of
all: the law of March ii, as well as Article 509 of the French

Ecrits, p. 813. Sheridan, pp. 310—311 (translation modified).

These are questions that Jacques Derrida has addressed in a recent article
dealing primarily with Walter Benjamin's Critique of Force, but which also
discusses Montaigne, from whom Derrida takes the title of the essay: "Force of
Law: The 'mystical Foundations of Authority," Cardozo Law Review,
pp. 919—1046.

'7'
 Appendix B

New Code of Civil Procedure, explicitly include oral lectures as one

of the "works," the reproduction of which is subject to
copyright regulation. Lacan's declaration that, unlike his friend,
Levi-Strauss, he will not leave behind a work, has in this
perspective no legal value whatsoever, however significant in
other ways it might turn out to be. Secondly, there are other
speech acts of Lacan, which, although they may be of lesser
theoretical interest, and indeed, may even seem to contradict
that theory, are legally binding. One such is the duly signed and
notarized will, in which Lacan "names as my testamentary
executor, insofar as the totality of my published and non-
published work (oeuvre) is concerned, M. Jacques-Alain Miller.
He will exercise all the prerogatives attached to the moral right
according to the law of March II, 1957." Between two énoncés,
then, the one stating that Lacan will leave behind no work, the
other, that his entire work, published as well as unpublished,
shall be administered by Miller as prescribed by French
copyright law — how is one to choose? If the "authority" of a
statement derives only from its utterance, how is it to be acted
upon?
In Lacanian terms: If the subject is constituted by its
"submission to the signifier,"5 i.e. to an alterity that can never
be reduced or retraced to the same, how is one to conceive the
process of such an authorization? What is the relation of the Law
or laws of language to the laws of the State? Of its statements
and the process of stating, of enunciating? How in general does
authority as such take place, if this place, as site of the signifier, is
not one that can ever be simply, or definitively, taken?
These are just some of the questions that the dispute over the
Lacan legacy stirs up and calls us to reconsider. But this case
becomes even richer in its enseignement when we consider the
other side, which, in this instance is not that of the Other, not at
least to begin with, but that of the Author: the "co-author" and
"testamentary executor," Jacques-Alain Miller. A year before he
actually brought charges against Stécriture, Miller published an
interview that he had given in which he discusses his work as
the editor of Lacan. In this text, entitled Conversation on the

Sheridan translates Lacan's soumission as "subjection." Cl: Ecrits, p. 8o6;

Selection, p. 304.

172
 Psychoanalysis and criticism

Seminar,6 Miller addresses the criticism later to be formulated by

Stécriture, above all, that of not having satisfactorily fulfilled the
"function of editor and guardian of the work, conferred upon
me by Lacan in the most legal of forms." It has not been easy, he
acknowledges, to do this while "continuing and animating the
truth of his teachings." (66) Coming as it does shortly before the
end of the interview, such an admission suggests that there is
no easy response to the critiques addressed to him. Neverthe-
less, elements of a response are furnished by Miller's editorial
practices, such as his decision not to publish Lacan's texts with a
critical apparatus, as did Stécriture:
the lectures of Lacan could have justified an entire critical apparatus:
references, citations, clarifications of difficulties. However, by common
consent [between Miller and Lacan, presumably] the Seminar is
presented without any critical apparatus. . .

Miller gives no particular reasons for this decision, beyond

alluding to the need for keeping the teachings of Lacan "alive"
(vivant). He also refers to a similar decision taken by the editors
of the Collected Works of Heidegger. Yet the interview as a
whole makes it quite clear just why there is no place in his
edition of Lacan for what he calls "la glose universitaire":
academic notes and commentary. What is to be avoided,
ultimately, is what has happened to psychoanalysis in the
Unites States, or for that matter anywhere where the teaching of
Lacan has not (yet?) taken hold. In France, as in "all other
romance-language countries," the interest in psychoanalysis is
intense, Miller notes. On the contrary, in areas
where this teaching has not been received, for example in the United
States, the fashionable vogue enjoyed by psychoanalysis in the
post-war years has disappeared. If interpretation is predetermined by
norms, which (in turn) always borrow their definition from social
ideals, it loses its distinctive force (vertu), and the subject supposed to
know (le sujet suppose savoir), which is essential to the functioning of the
experience (or the experiment), crumbles (se délite). This has happened
wherever Lacan's teaching has not taken hold.
Miller's allusion to Lacan's notion of the "subject supposed
to know," indicates that in his eyes, the establishment of an
6
J.-A. Miller, Conversation sur le Séminaire, Navarin, Diffusion Seuil: Paris, 1985.

'73
 Appendix B

authoritative text, without distracting notes and references, is

inseparable from a desired reinforcement of "transference" on
the part of the reader:
According to Lacan's definition, knowledge is the structural pivot of
transference. One can, therefore, certainly have a "transference to
Lacan" [.. 1. on the basis of the knowledge laid down and articulated
.

as a work. Moreover, the form of the work is such that this knowledge
conserves a dimension of supposition.

Having thus established the importance, in reading Lacan, of

the supposition of an all-knowing subject, Miller goes on to
point out that this subject should not be confused with an
author:
Lacan undoes the position of the author as someone who knows what
he is saying, such that the dimension of supposition persists, and that
in place of truth — to refer to his discourse — there is precisely this
supposed knowledge, not the author identical to himself. (64—65)

Although the accent here is placed upon the supposition of

knowledge rather than upon knowledge itself, it is clear that
such a supposition cannot impose itself in a void. If knowledge
is to be supposed, and if this supposition is to impose itself upon
readers, it will have to respect the forms of cognition, even if its
contents will prove to be elusive. Miller seeks to describe this by
emphasizing the systematic character of Lacan's thought. The
subject supposed to know articultates itself in Lacan's work not
through a refusal or an absence of systematization, but through
the tireless transformation of each successive system by what
follows in a thought-process that constantly calls itself into
question. "I believe that Lacan continually thought against
Lacan," Miller remarks, while also acknowledging that such
questioning is obscured by the often apodictic, formulaic
character of Lacan's affirmations: "The cutting edge of his
formulas does not alter the fact that their exact theoretical value
is related to the moment of their enunciation." (45—46)
Once again, we find ourselves back at the problem of
"enunciation," the enigma of which is hardly resolved by
introducing, as Miller does, the Hegelian notion of "moment."
For the movement of the Signifier, unlike that of the Concept, is
not circular, its "chain" is no daisy-chain, nor does it spiral

'74
 Psychoanalysis and criticism

toward totality. How "exactly" therefore "theoretical value" can

be determined, insofar as it depends upon a movement of
signification that is not just ongoing, but is dislocated in each of
its moves, is a question that Miller, in this interview at least,
does not address.
In the meanwhile, to be sure, one reply has been furnished,
provisionally at least, by the First Chamber of the Paris Tribunal
de Grande Instance. It should be noted that the name of this court
is not easy to render in English, which has a Small Claims Court,
but no Large Claims Court. More interesting, to be sure, is the
fact that the French title condenses several of the problems we
have been discussing. The term, Instance, recalls both Freud's
use of its German version, Instanz, to designate the different
systems of the psyche, and Lacan's designation of the letter in
psychoanalysis as an instance. The English translation of this
word as agency obscures the conflictual urgency that marks the
word both in Lacan's and Freud's writings, where it designates
not merely the executor or executive, but the precariousness of
all stable institutions given the pull of divergent forces that
prevails. This sense of the term both points to the need for a
judge and judgment in respect to conflicting parties, and
also suggests how fragile such sentences, and indeed, the
authority in whose name they are spoken, inevitably must be.
The etymology of the term, in-stare, designates not just the fact
of being present, but rather the necessity of such presence being
continually renewed and reaffirmed. Hence, the relationship,
often noted by Lacan, between "instance" and "insistence," or
even "persistence." It is this insistent drive that is arrested in
the Instance that the German translators of Lacan sought to
preserve when they translated the word into German as
"Drangen": the "Urging" or "Urgency of the Letter." An
urgency is not entirely the same as an agency.
What gets lost is the notion that the very "instance" that is
called upon to adjudicate, and indeed, to pronounce judg-
ment, is itself part and parcel of the conflicts it seeks to resolve.
In short, the Court has power, but does it have authority over
the Signifier? And if it does, from where does that authority
derive its legitimacy? In democratic societies, one is prone to
point to the people as the source of all authority; but it remains
to be seen in what sense the people, as a collective subject, can

'75
 Appendix B

claim to dictate the law to language, if all subjectivity only comes

to be in and through the movement of signifiers.
The Tribunal de Grand Instance delivers its verdict, recognizes
rights and prohibits all reproduction, distribution, exchange and
circulation by Stécriture of its version of the Seminar on
"Transference." It also "stays" the execution of its verdict,
except for the ban on distribution, pending appeal. "Transfer-
ence" is thus stopped, provisionally, until an authorized version
can be produced. The "law" of language, however, as elabo-
rated by Lacan and before him by Freud, "knows" no such stop-
page, just as the Unconscious is said to "know" no contradic-
tion. The overdetermination of unconscious inscription, as
in dreams, both requires interpretation and at the same time
can never be exhausted or rendered fully by any interpretation.
Interpretation thus is construed and practiced less as a faithful
rendition than as a struggle for power, or rather — in the
Nietzschean sense of the phrase, as a Wille zur Macht, a will
toward power. Such power is not something that can be reached,
attained, in a place that one might hope to occupy (in German:
besetzen, "cathect"), once and for all. Rather, it entails a constant
struggle that Freud in the Interpretation of Dreams describes as one
of "Selbstuberwindung," which is to say, overcoming of Self, and
not, as the Standard Edition would have it, "self-discipline."7
Certainly, inasmuch as all interpretive practice necessarily
attempts to establish its authority, the distinction between self-
discipline and self-overcoming inevitably becomes blurred;
nevertheless, the fact remains that today, psychoanalysis,
where it is informed not only by Lacan, but by the more general
movement of thought of which Lacan is an outstanding
"instance" and yet by no means the only such: that is by what
we call today "poststructuralism" — that psychoanalysis has
become one of the areas in which the a-licit Law of language
struggles to impose the problematic rights of the author.
Another such area is, of course, the study of literature.
This has, of course, been the case ever since the beginnings of
Western thought. It is a case that has been made against

S. Freud, The Interpretation of Dreams, chapter "The forgetting of dreams,"

Avon Books, p. 563.

176
 Psychoanalysis and criticism

literature ever since Socrates — or was it Plato? — excluded the

poets from the just state. They were banished for "forgetting
themselves," for allowing themselves to be carried away by
their mimetic impulses. In so doing, Plato argued — mimetically,
through the mouth of "Socrates" — that the poets forgot or
abdicated their diegetical responsibilities, and thereby forfeited
their place in the polis:
Do you know the first lines of the Iliad, in which the poet says that
Chryses implores Agamemnon to release his daughter, and that the
king was angry and that Chryses, failing of his request, heaped curses
on the Acheans in his prayers to the god? You know then that [. ] the . .

poet himself is the speaker [there] and does not even attempt to suggest
to us that anyone but himself is speaking. But in what follows he
delivers his speech as if he himself were Chryses and tries as far as may
be to make us feel that not Homer is the speaker, but the priest, an old
man. And in this manner he has carried on nearly all the rest of his
narration about affairs in Ilium, all that happened in Ithaca, and the
entire Odyssey. (393a—b)8
As the possibility of such "mimetic" narration, poetry poses a
danger to the statesmen, the "guardians" who if they must
imitate, "should from childhood on imitate what is appropriate
to them." (395c) By "likening himself to another," by speaking
with the voice of another, the poet undermines the authority of
his discourse. The verdict is ironic, but without appeal:
If a man [. ] who was capable by his cunning of assuming every kind
. .

of shape and imitating all things should arrive in our city, bringing with
himself the poems which he wished to exhibit, we should fall down
and worship him as a holy and wondrous and delightful creature, but
should say to him that there is no man of that kind among us in our
city, nor is it lawful for such a man to arise among us, and we should
send him away to another city, after pouring myrrh down over his head
and crowning him with fillets of wool. . . (398a—398b)
Reading this passage today, we are liable to react with a
certain condescension, as though the irresponsibility of lan-
guage was no longer a problem for us, schooled as we are on
Bakhtin and Barthes, and protected by International Copyright
Conventions. Were we to react in this manner, however, we

Plato, The Collected Dialogues, ed. E. Hamilton and H. Carins, Pantheon Books:
New York, 1961, pp. 637—638.

I 77
 Appendix B

would be pulling the wool over our eyes: the issues that
preoccupied Socrates and Plato are still very much with us. To
confirm this, we need only reflect for a moment on the
importance, in our own writing, on the one hand of quotation
marks, and on the other, of proper names, in particular those of
authors and titles. Without the latter, how could we identify
"works"; without the former, their meaning? Imagine what
would become of our jobs, and of our practice as teachers and
students, scholars and critics, were we no longer able to rely
upon quotation marks to distinguish direct from indirect dis-
course, or to demarcate the writing of others from what we
claim as our own?
For almost a century, reflection upon literature has been
occupied, and indeed increasingly so, with the problem of
authoritative discourse: from Henry James and Percy Bullock's
thematization of point of view to Wimsatt and Beardsley's
critique of the intentional fallacy; from Bakhtin's polyphonic-
dialogic theory of the novel, to Barthes' obituary of the author
and to the more cautious, more historical investigation of
Foucault, literary practice and critical theory have grown
increasingly suspicious of authorial positions and discourse.
Until fairly recently, however, criticism almost always stopped
short of reflecting upon the implications of such suspicions for
its own 'position' and project, and with good reason. For if the
"omniscient narrator" is at best unreliable and at worst an
illusion, what of the Critic? To what kind of authority can the
discourse of criticism legitimately lay claim?
The response most recently in vogue, in the US at least, of
certain neo-pragmatist critics such as Stanley Fish, is that it is
the "community of interpreters" alone that authorizes inter-
pretation. But such an answer merely begs the question it
pretends or claims to address. Constructing a collective subject
to serve as the authoritative instance accomplishes little, if that
subject is construed to have the same, self-identical, undivided
structure as the individual critic it is meant to supplant. For the
divisions with which we are confronted, today no less than in
the past, affect communities no less than individuals. The
question therefore to be addressed is not just: how does a
community constitute itself, but also: what does the notion of
community entail? Indeed, if we feel impelled to recur to this
178
 Psychoanalysis and criticism

notion today, it is because our interpretive practice calls into

question the establishment of precisely such communality.
What is involved in interpretation is not so much the analysis of
works, perhaps, as the imposition of meanings always more or
less at the expense of other, competing schemes. Interpretation
would therefore address neither the meaning of works, nor
even the condition under which such meanings take place, but
the very process of "taking place" itself; that is, of taking place
away from other place-holders. The real object of interpretation
would be the place itself, as site of division and conffict, and this
would determine its practice as negotiation and as strategy.
It is in allowing us to explore the nature of such divisions
and conflicts, in their structural and structuring effects, that
psychoanalysis has an important and probably indispensable
contribution to make to current theoretical discussions. Let us
return then to one of the areas opened by psychoanalysis, and
which seems particularly promising in this regard: that of
transference. The history of its use by Freud provides a useful
way of approaching the problem. In The Interpretation of Dreams,
the German word Llbertragung is employed to designate the
distortions of the dreamwork, which shifts from one represen-
tation to another in order to accomplish its goal: that of
producing a distorted, self-dissimulating fulfillment of a con-
flictual wish. Clbertragung, the German word that literally
corresponds to the Greek metaphorein, thus describes both the
particular dream-device of displacement (Verschiebung), and the
more general instability of psychic energy that is characteristic of
the primary process and of the unconscious in general. Freud
describes this primary process in terms of the volatility and
mobility of its cathexes, that is, of the manner in which energy is
attached to representations. This he contrasts with the secon-
dary process, to which he attributes a greater stability. In it,
energy is bound in a more enduring way to representations, to
"intellectual identities" as he calls them (Denkidentitäten), in
contrast to perceptual identities, which are epitomized not in
the perception of stable, self-identical objects, but in the
equivocal imagery of the manifest dream content.
What is striking, however, is that this opposition of primary
and secondary process, of volatile and stable cathexes, does not
suffice to account for the phenomenon of transference, which

'79
 Appendix B

exhibits traits of both processes: as distorted representation,

Ubertragung presupposes the volatile movement, the "carrying-
over" from one place or thought or image to another; but at the
same time, like "metaphor" itself, it also entails an element of
fixity, indeed of fixation. What is perhaps most significant of all,
however, is that these two elements, movement and fixation, do
not simply oppose one another, as one might expect, but rather
converge and overlap: the movement of representation is fixed,
arrested but the fixation/arrestation turns out to constitute
another kind of movement. The movement is fixed and
arrested, inasmuch as the process of symbolization has come to
rest in the manifest dream-content; but it is also in movement,
insofar as the apparently stable content leads us inevitably in
multiple directions: forward, into the future, through the fact
that the dream depends upon its belated narration in order to
function and indeed, to be; it only comes to be the morning
after, as it were, in its distorted reproduction; and backward,
toward the infantile complexes that are always more or less at
the origin of the dream. In this sense, then, the dream does not
merely make use of CThertragung: it has the structure of an
Ubertragung.
This curious conflict of fixation and mobility, which also
entails a form of repetition, is what emerges with increasing
emphasis in Freud's later use of the term to designate the pivotal
mechanism in the analytic situation itself. The analysand,
instead of remembering — that is, instead of representing the
past as past, and hence, as representable — repeats the past as
though it were the present. The past, instead of being
remembered, is re-enacted. Again, we are confronted with a
movement of repetition that is simultaneously submitted to the
constraints of a certain fixation. The differences of the present
are ignored and thereby reduced to sameness.
What, however, gives this latter usage of the term "transfer-
ence" its specific quality is that its fixation is bound up with the
figure of the analyst, who becomes the object of love, hate, or
both at once. Freud stresses that transference becomes in-
creasingly intense as the analysis progresses, that is, as the
analysand begins to approach, to articulate and to assume the
conflicts of desire involved in neurosis. The projective mechan-
ism of transference is a way of both acknowledging and resisting

i 8o
 Psychoanalysis and criticism

that development: the split in the subject is bridged, as it were,

by an amorous (or antagonistic) relation to the Other, whose
role in the analysis is played by the analyst. By being treated as
the object of erotic passion, this Other is thus made into an
object of love, the reality of which is no longer to be questioned.
The Other is no longer analyst, but beloved, no longer agent of
the signifier, but quintessential signified. The conflict is no
longer within the subject, but between subjects construed as
self-identical egos.
Such transference, Freud emphasizes, poses the greatest
dangers to the analytic process, but at the same time it is its only
chance of success. For only by means of such transferential
projection can analyst and analysand hope to "work through"
the resistant and conflictual reality of the signifieds of desire in
order to reach its signifying passion.
There is every reason to believe that something very similar is
at work in our dealings with texts in general, and with literary
texts in particular. If, at least, by "literary" we mean something
akin to what Kant had in mind when, in the Preface to the
Critique of Judgement, he noted that it is "primarily in those forms
of evaluation which are called aesthetic" that we find that
"embarrassment concerning principles" in which judgment has
no universals to fall back on in its confrontation with particular
cases, in which it therefore has no choice but to fall forward, as it
were. Faced with the inexhaustible multiplicity of experience,
with an alterity which cannot be subsumed under existing
knowledge, what the judging subject does is something not so
very different from the analysand, or for that matter, from the
literary critic. In order to judge, the subject considers the
particular thing that confronts it as though it were the product of
an "understanding" like our own, and yet sufficiently different
from ours to comprehend what we cannot. Faced with the
singular unknown, what the subject does is to suppose a subject
that knows, that comprehends what we do not, because it has
produced it according to its knowledge. Through such an
assumption, the judging subject seeks to assure itself that the
unknown is, at least potentially, knowable.
This, for Kant, constitutes the a priori, transcendental princi-
ple of what he calls "reflective judgment": it is "reflective"
because properly understood, it tells us nothing about the
 Appendix B

object, nothing about the other to be judged, but only about a

"law" — if it can be called that — the judging subject gives to itself
in the process of judging the unknown (that for which no
general law or concept can be found). And it is this that Kant
finds at work in that most exemplary case of reflective
judgment: aesthetic judgements of taste. Thus, the entire
Kantian conception of beauty as "form," as purposiveness
without purpose, depends upon this initial, initiating assump-
tion: that of an Author, having produced a — his — work.
However, the fact that this judgment is defined by Kant as being
"reflexive" also renders that assumption of authority fictional: it
applies not to the object but to the judging subject; it is an "as
if." And yet, the status of this as-if, of this assumption, proves
difficult to determine in any univocal way. For what does it
entail to assume such an Author, while at the same time
"knowing" that it is "only" an analogy, a projection of the
knowledge we desire? Can such an assumption, which does not
or should not constitute a statement about reality, be "really"
assumed, as a pure fiction? Were it recognized to be a pure
product of the subjective imagination, would it still operate to
enable investigation, and thus to prepare us to discover the
missing universal law, rule or concept?
What the psychoanalytic theory of transference suggests, is
that such assumptions, or projects, can never be made innocent-
ly or as mere heuristic devices, for the simple, or rather
complex, reason that the subject only comes to be through such
a process. If this is so, there is good reason to suspect that it will
take more than exposure of the various Intentional Fallacies to
rid us of its literary correlative, the Authorial assumption.
Perhaps what we should try to think about are ways not so
much of escaping from it, as of putting it into play; in this case,
however, it might just be criticism itself that turns out to have
a leading part.

i 8z
 Index

a (objet petit a: see o: object small 0): fader (subject as): 90

I 19fl fetishism: 143
Ab-ort: xix Fish, Stanley: 178
Abraham, K.: I i Wilhelm: io, 148
anxiety: 152.—167 fort/da: 131—135
Aristotle: 10—15, 103 future anterior: 7—10
Artaud, Antonin: 73fl
Aubenque, Pierre: 20—2.2 Gorgias: 21, 23,24
Greenaway, Peter (The Draught man's
Bakhtin, Mikhail: 177, 178 Contract): 165—167
bar (of signification): 40—43, 66, 12.4,
'47, 149 Hegel:xix, 8—io, 17—18, 110
belatedness: 9—10 Heidegger, Martin: i i 8n, 9on,
Belimer, Hans: 14 100, loIn, ,o6,
Benjamin, Walter: xvi, xvin Hjelmslev, Louis: 46, 47
Benveniste, E.: xix, 81—83, 85—86 Homey, Karen: 142
Bilderwert: xviii Hugo, Victor ("Booz endormi"): 57
Borch-Jakobson, Mikkel: III n Husserl, E.: xix, 9,46, 52, 54, 6i,
Bullock, Percy: 178 8i—86

castration: 140—150 1: 8i—86, 89—90

cogito (see also Descartes): 89 ideational representative: 70
condensation: XiX, 67—68 identification: 11—16, 133—134
imaginary: i8, 105, 135
demand: 129—130 imperfect: 114—115, 117
Derrida: xii, xiii, 27, 3 1—32, 48, 73n, Inhibitions, Symptom, Anxiety: 157' 163
tøifl, 135 interpretation: i 2.

Descartes: 35, 8i Interpretat ion of Dreams: 2—3, 5,53,

desire: izo, 127—129, 136—137 76—81,111, 12.3, 156
diachrony: 33, 35 IPA (International Psychoanalytic
differential articulation: 6z, Association): 99
dislocation: xvii isolating: xi
displacement: xvii—xx, 2, 67—68 iterability (see also: repetition): 71
dream-distortion: xix
dreams: xviii, 2. Jakobson:xix—xx, 46, 48, 52—54,
drifter (subject as): iii 59—61, 82—84
drive: 124—127 James, Henry: 178
Jones, Ernest: 142.
ego (see also 1): 12—17. 34, 78—81,
157— 158 Kant, Immanuel: ,8,—,8z
ego, fictionality of: 13 Kierkegaard,S.: 152, ,6z, 163, 164
Emerson, R.W.: xii Klein, Melanie: 14, 105, 142
Ent-stellung: xvi—xx, 72—73 Kojéve, Alexandre: 11 in, 120,
experience of satisfaction: 5 128—129

i83
 Index

"Lacan": xii—xiii repetition: 5, 72

Lacoue-Labarthe, Philippe: ioi repression: xi
Lagache, Daniel: io2, 104, Io8 Roudinesco, Elisabeth: 9911
langage: 25
langue: 25 Safouan,M.: 12711
Laplanche, J.: xix, 6z Sartre,j.-P.: 15
law: 137, 176 Saussure: xviii—xxii, 19, 20, 25—40,
Leclaire, Serge: 6z, i 26n 42, 45—48, 51—54, 59—65, ro8,
letter: xxii, 45, 49, 63 I 34fl
Schreber, Paul:
Masson,J.M.: 148n screens: xi
méconnaissance: xxii, 15 signifying chain:
Mehlman, Jeffrey: 99fl, 137fl Signorelli: 91—97
metaphor/metonymy: 51—69 simulacrum (phallus as): 145—146
Miller,J.-A.: 99fl, 168—170, 172—175 Socrates: 177, 178
mirror stage: 7—19 Sophists (theory of language): 21—24,
misrecognition: 14 104
Mitchell, Juliet: 141fl Spitz, René: ,zz
structuralism: xii
Sully,J.: 75
name(s]-of-father: 97, 137—138, symbolic: 18
'47, '5', '54 synchronic-diachronic: 33—37
Nancy, jean-Luc: 101 syntagmatic relations: 51—54
Narcissus, narcissistic: 133,
140—141
taking place: 76, 179
need: izz theatricality: xxii, io, 72—73
Nietzsche, Friedrich: xv, 176 transference: 3, 179—182
translation: i—s, 88
o(object small 0): 119, 158—161, 164 trauma: 155—156
Oedipus complex: i 23—124 Trubeckoj, Nikolai: 46, 59
paradigmatic relations: 5 1—54 uncanny: xi, 64
parole: Urverdrangung (see primal repression):
Peirce,C.S.: ,o6, 114, 115 110
Perelman, Charles (ChaIm):
phallus, phallic: 139—151 Valery, Paul: 54
Plato: lo, 176, 178 Ver-(German prefix): 4fl
poststructuralism: xii Verkennung (misapprehension,
primal repression: 110, 149 méconnaissance): 15
primary identification: 14 Vorstellungsreprasentanz: see ideational
primary narcissism: 16 representative
primary process: 69—71
privation: 146 Wilden, Anthony: 95—6
Wimsatt and Beardsley: 178
real: io6 wish: 121—126
reflective judgment (Kantian): r8,—
182 Zeichenbeziehung: xviii

184
This page intentionally left blank.