\

University of California Press

Berkeley and Los Angelel. California
Fil"!it California Paperback Printing 1989
Publislled by Agreement with Farrar. Straus and Girou.lnc_

Tnnllat.ion oopyriJht C 1986 by Farrar. SIrIUI and GifOU.l.lnc.

QriainaIIy Plblilhed in French ufA br1 de fa kllll '"
Copyripl C 1984by Edition. du Seuil
All ri&btl raeted
Prinaed in the United StIleS of AmericIo
Drsie'"Ed by Jack HmUon

45618910
-ro WriU!, An lntnnlitive Verbr fmt Plbliahed lin Enal-ishl in Tit. Ulll_8Uc!
C,iliNm tuod lIN ;,,., cfAI",,: Tit. SfrVd_Ii.r, C(}flm_I1. edited by Rkh.,d
M ...
.. ey. C 1970 by the JoM. H",*im Univcl1iay Pral. "The Rustle <:L l.anl"l,e
fIM publiabed in Vcn _ UJUliqwlWU 'II/raVU. C 1975b y U.G.E.
AMl)'lis fitlt publUhed in LiJ,INfIlT, " SDciill. C 1961 by L'InJliluI de JOcioIo&;e de
i'Univcnitt Iib!e de Bnuek' "The Division <:L lMI&UI'es fU'lt publi$hed in U.u
civiJi.roliole c? HAl;:"" d: G,or8U FrUd.:wNC, C 1m by Oallimard.
and EruUle" fU'lt published in La Corpsl1nJli8cn by Jean c.yroI, C 1964 by U.O,E.
BIoy" fimpubfuhcd in TQW... de klliJlir/JI.."frtu!f4ise, C 1974 by Gallinmd.. "What
Br ;Ollel<:L1he Sipif:er" fml published in &kif., Ed".. de". by Picm= Guywl, C 1970
by Oillimard. OutcoulCl of !he THot" fin! publilhed in &lIIiJl. in the 1011& collection.
C 1973 by U.G.E. "Readin, BriIlal.sav.rin fint published inPIojDl..,;" dll.8"" by
BrilW-SIovlrin, C 1975 by C . Hem::ann. "Preface 10 Rcnaud Camus', Trb: C 1979 by
Penona. "The Im..e fU'lt published inP,Iuu,: RoI4nd Bal1i:u in !he 10118
eoI1eclm.C 1978 by U.G.B.

ubnuy <:L Conglal c.ta1ogin, inPublicat.ion

TRII IIation of: Le hnlislEl"U,1 de IIlanJIIC'

I. PhiJoloay. 2. DUcoul1e analysis, 3. SuniQlte.,
L Title.
N9,B3513 1989 401'.41 88-29579
ISBN 0-5 (Ilk. paper)

The paper u.sed in this public.tion m1i the minimum requirements of American
Nlliooal Standard for Information Scienc,s-Permanence of Paper for Printed
Librlry M.terials, ANSI Z39.48-1984. e

B
 Contents

1 I FROMSCIENCETOJ.JTERATlJRE
From Science to Jjterature 3
ToWrite:AnIntransitiveVerb? )
)

ReflectionsonaManual 22

Writing Reading 29
On Reading 33
Frttdom to Write 44

2 I FROMWORKTOTEXT
The Death of the Author 49
From Work to Text 56
Mythology Today 65
Research: The Young 6g
The Rustle of Language ,6

3 I LANGlJAGES AND STYLE

Rhetorical Analysis 8g
Style and Its Image go
PoxCuhuralis 100

The War of Languages ,06

The Division of Languages I II

..
vu

B
VIII Contents

4 I FROM HISTORY TO REALITY

The Discourse of History 127
The Reality Effect 141
Writing the Event 149

5 I THE LOVER OF SIGNS

Revelation 157
A Magnificent Gift 159
Why I Love Benveniste 162
Kristcva's5mujptiJre 168
The Return of the Poetician 172
To uarn and to Teach 176

6 I READINGS
ONE

Carrol and Erasure 181

Blay 191
Michelet, Today 195
Michelet's Modernity 208
Brecht and Discourse: A Contribution to the
Study of DiSCUT5ivity 212
TWO

F.B. 223
The Baroque Side 233
What Becomes of the Signifier 236
Outcomes of the Text 238
Reading Brillat-Savarin 250
An Idea of Research 271

,
 ox

Conunts

Longtnnpl,jt mt lUis couch; dt bonne

hmre. . .277
Preface to Renaud Camus's Triclu 291
One Always Fails in Speaking of What One
Loves 296

7 I ENVIRONS OF THE IMAGE

Writers, Intellectuals. Teachers 309
To the Seminar 332
The Indictment Periodically Lodged. . .
343
Leaving the Movie Theater 345
The Image 350
Deliberation 359

C;.pvnghted malaria
 t
FROM SCIENCE TO
LITERATURE

Malena :>m d reitl al


 From Science to Uterature
"Man cannot speak his tlwught witlund thinAing his speech."
-BONALD

French university faculties possess an official list of the social

and human sciences which constitute the object of a Tttognized
instruction, thereby necessarily limiting the speciahy of the
diplomas they confer: you can a doctor of aesthetiC5, of
psychology, of sociology-not of heraldry. of semantics, of
victimology. Thereby the institution direct1y determines the
nature of human knowledge, imposing its modes of division
and of dassification,jusl as a language, by its "obligatory rubrics"
(and nO[ only by its exclusions), compels us to think in a certain
way. In other words, what defines scintce (the word will hence
forth be used, in this text, to refer to all the social and human
sciences) is neither its content (which is often ill defined and
labile) nor its. method (which varies from one science to the
next: what do the science of history and that of experimental
psychology have in common?). nor its morality (neither serious
ness nor rigor is the property of science), nor its mode of
communication (science is printed in books, like everything else).
but only its status, i.e
. its social determination: the object of
science is any material society deems worthy of being transmit
ted. In a word. science is what is taught.
Literature has all the secondary characteristics of science, i.e.,
all the attributes which do not define it. Its contents are precisely
those of science: there is certainly not a single scientific matter
which has not at some moment been treated by universal
literature: the world of the work is a total world, in which all
(social, psychological, historical) knowledge takes place. so that
for us literature has that grand cosmogonic unity which so

Malena :>m d reitl al

4 From Science to Literature
delighted the ancient Gruks but which the compartmentalized
state of our sciences denies us today. Further, like science.
literature is methodical: it has its programs of research. which
vary according to schools and periods (like those of science,
moreover), its Tules of investigation. sometimes even its exper
imental pretensions. Like science. literature has its morality, a
certain way of extracting its rules of procedure from the image
it assumes of its being. and consequently of submitting its
enterpris to a certain absolute spirit.
One last feature unites science and literature, but this feature
is also the one which divides them more certainly than any other
difference: both are discourses (which was well expressed by the
idea of the ancient logos), but science and literature do nOt
assume do not profes5-t.he language which constitutes them
in the same way. For science, language is merely an instrument,
which it chooses to make as transparent, as neutral as possible,
subjugated to scientific matters (operations, hypotheses, resul(5),
which are said to exist outside it and to precede it: on one side
and first of all, the contents of the scientific message, which are
everything; and on the other and aftnwards, the verbal form
entrusted with expressing these contents. which is nothing. It is
no coincidence if, since the sixteenth century, the combined rise
of empiricism, of rationa1ism, and of religious evidence (with
the Reformation), i.e., of the scientific spirit (in the very broad
sense of the t.erm), has been accompanied by a regression of
the automy of language, hencefonh relegated to the status of
"instrument" or of "fine style," whereas in the Middle Ages
human culture, as interpreted by the Stptenium. shared almost
equally the secrets of language and those of nature.
For literature, on the contrary-at least for that literature
which has issued from classicism and from humanism-language
can no longer be the convenient instrument or the sumptuous
decor of a social, emotional, or poetic "reality" which preexists
it and which it is responsible, in a subsidiary way, for expressing,
provided it abides by a few rules of style: no, language is the
being of literature, its very world: all literature is contained in

Malena :>m d reitl al

From Science to Literature 5

the act of writing, and no longer in that of "thinking," of

"painting." of "recounting," of "feeling." Technically. according
to Roman Jakobson's definition, the "poetic" (i.e., the literary)
designates that type of message which takes for object iu own
fonn. and not its contents. Ethically, it is solely by its passage
through language that literature pursues the disturbance of the
essential concepts of our culture. "reality" chief among them.
Politically, it is by professing (and illustrating) that no language
is innocent. it is by employing what might be called an "integral
language" that literature is revolutionary. Literature thus is
alone today in bearing the entire responsibility for language;
for though science needs language, it is not, like literature,
within language; science is taught. i.e., it makes itself known;
literature fulfills more than it transmits itself (only its history is
taught). Science speaks itself; literature writes itself; science is
led by the voice, literature follows the hand; it is not the same
body, and hence the same desire, which is behind the one and
the other.
Bearing essentially on a certain way of taking language-in
the former case dodged and in the latter assumed-the oppo
sition between science and literature is of panicular imponance
to structuralism. Of course this word. generally imposed from
outside, actually overlaps very diverse, sometimes divergent.
sometimes even hostile enterprises, and no one can claim the
privilege of speaking in its name; the author of these lines
makes no such claim; he merely retains the most particular and
consequently the most peninent version of contemporary struc
turalism, meaning by that name a certain mode of analysis of
cultural works, insofar as this mode is inspired by the methods
of contemporary linguistics. Thus, itself resulting from a lin
guistic model, structuralism finds in literature, the work of
language. an object much more than affinitary: homogeneous
to itself. This coincidence does not exclude a certain embar
rassment. even a cenain laceration. depending on whether
structuralism means to keep the distance of a science in relation
to its object. or whether, on the contrary. it is willing to

Matena dlreitl
6 From Science to Litl'TatUTt
compromi and to spoil the analysis it wields in that infinitude
of language of which literature is today the conduit-in a word.
depending on whether it seeks to science or writing.
As science, structuralism "finds itself," one might say. on every
level of the literary work. First of all, on the level of contents,
or more exactly, on the level of the form of contents, since
structuralism ek5 to establish the "language" of the stories
told. their aniculations, their units. the logic which links some
to otheTS-in short. the general mythology in which each literary
work participates. Next. on the level of the forms of discourse:
structuralism. by virtue of its method, pays special attention to
classifications. orders, arrangements; its essential object is tax
onomy, or the distributive model inevitably established by any
human work, institution. or book, for there is no culture without
classification; now discourse, or ensemble of words superior to
the sentence, has its forms of organization; it too is a dassifica
tion, and a signifying one; on this point, literary structuralism
has a glamorous ancestor, one whose historical role is in general
underestimated or discredited for ideological reasons: Rhetoric.
grandiose effort of an entire culture to analyze and classify the
forms of speech. to render the world of language intelligible.
Finally. on the level of words: the sentence has not only a literal
or denoted meaning; it is crammed with supplementary signi
fications: since it is at once a cultural reference. a rhetorical
model. a deliberate ambiguity of the speech-act, and a simple
unit of denotation, the "literary" word has the depth of a space,
and this space is the field of structural analysis itself. whose
project is much greater than that of the old stylistics, entirely
based as it was on an erroneous idea of "expressivily." On all
its levels-that of the argument, that of discourse, that of the
words-the literary work thereby offers structuralism the image
of a structure perfectly homological (present-day investigations
tend to prove this) to the structure of language itself; derived
from linguistics, structuralism encounters in literature an object
which is itself derived from language. Henceforth. it will be
understood that structuralism may attempt to found a science
From Science to Lilnalure 7
of literature. or more exactly a linguistics of discourse. whose
object is the "language" of literary forms. apprehended on many
levels: a new project. for hitherto literature has been approached
"scientifically" only in a very marginal fashion-by the history
o f works. or of authors. or of schools, or of texU (philology).
New as i t may be. this project is nonetheless not satisfactory
or at least not sufficient. It leaves untouched the dilemma I
mentioned at the beginning. one that is a1legoricalJy suggested
by the opposition between science and literature. insofar as
literature assumes iu own language-under the name of writ
ing-and science avoids it, feigning to regard it as purely
instrumental. In a word, structuralism will never be anything
but one more "science" (several of these are born every century.
some quite ephemeral), if it cannot make its central enterprise
the very subversion of scientific language. i.e., cannot "write
itself": how can it fail to call into question the very language by
which it knows language? Structuralism's logical extension can
only be to join literature no longer as "object" of analysis but as
activity of writing, to abolish the distinction, born of logic. which
makes the work into a language.-object and science into a meta
language, and thereby to risk the illusory privilege attached by
science to the ownership of a slave language.
It remains therefore for the structuralist to transform himself
into a "writer." not in order to profess or to practice "style." but
in order to recognize the crucial problems of any speech-act.
once it is no longer swathed in the kindly cloud of strictJy realist
illusions which make language the simple medium of thought.
This transformation-still rather theoretical, it must be admit
ted-requires a certain number of ciarifications---o r acknowl
edgments. First of all. the relations of subjectivity and objectiv
ity-or, to put it another way, the subject'S place in his work
can no longer be conceived as in the palmy days of positivist
science. Objectivity and rigor. attributes of the scholar which we
still hear so much about, are essentially preparatory virtues.
necessary to the work's moment, and as such there is no reason
to mistrust them or to abandon them; but these virtues cannot

C :lpynghled malenal
8 From Science to Literature
be transfe to discourst:, except by a kind of hocus-pocus, a
purely metonymic procedure which identifies precaution with its
discursive effect. Every sechact supposes its own subject.
whether this subject expresses himself in an apparently direct
fashion, by saying /, or indirect. by designating himself as hi.
or in no fashion at all, by resoning to impersonal lurns of
speh; what is in question here are purely grammatical strat
agems, simply varying how the subject constitutes himself in
discourse, i.e., gives himself, theatrically or fantasmatically, to
others; hence they all designate forms of the image-repertoire.
Of these forms, the most specious is the privative form, precisely
the one usually employed in scientific discourse. from which the
scholar excludes himself in a concern for objtttivity; yet what
is excluded is never anything but the "person" (psychological,
emotional, biographical), not the subject; moreover, this subject
is filled, so to speak, with the very exclusion it so spectacularly
imposes upon its person, so that objectivity, on the levd of
discourse an inevitable levd, we must not forget-is an image
repertoire like any other. In truth. only an integral formalization
of scientific discourse (that of the human sciences. of course.
for in the case of the other sciences this has already been largely
achieved) could spare science the risks of the image-repertoire
unleS5. of course, it consents to employ this image-repertoire
with Julllr.1Wwltdgt, a knowledge which can be achieved only in
writing: only writing has occasion to dispel the bad faith attached
to every language unaware of its own existence.
Again. only writing-and this is a first approach to its defi
nition-effectuates language in its totality. To resort to scientific
discourse as to an instrument of thought is to postulate that a
neutral state of language exists, from which would branch off.
like so many gaps and ornaments, a certain number of special
languages, such as the literary language or the poetic language;
this neutral state would be, it is assumed, the code of reference
for all the "eccentric" languages which would be only so many
sub-codes; by identifying itself with this referential code, basis
of all normality, scientific discourse arrogates to itself the very
authority which writing must contest; the notion of "writing"

,
From Scimct to Literaturt 9
implies in effect the idea that language is a vast system of which
no single code is privileged-()r, one may say, central and of
which the departments are in a relation of "fluctuating hier
archy." Scientific discourse believes it is a superior code; writing
seeks to be a total code, including its own forces of destruction.
It follows that only writing can break the theological image
imposed by science, can reject the paternal terror spread by the
abusive "truth" of contents and reasonings, can open to research
the complete space of language, with its logical subversions, the
mixing of its codes. with its slippages, its dialogues, its parodies;
only writing can set in opposition to the savant's assurance
insofar as he "expresses" his science-what Lautreamont called
the writer's "modesty."
Last, between science and writing, there is a third margin,
which science must reconquer: that of pftasurt. In a civilization
inured by monotheism to the idea of Transgression, where
every value is the product of a punishment. this word has an
unfortunate resonance: there is something light, trivial, partial
about it. Coleridge said: "A poem is that spies of composition
which is opposed to works by science, by purposing, for its
immediate object, pleasure, not truth"-an ambiguous declara
tion, for if it assumes the "erotic" nature of the poem (of
literature), it continues to assign it a special and guarded canton,
distinct from the major territory oflruth. "Pleasure," however
we admit this more readily nowadays-implies an experience
much wider, more significant than the simple satisfaction of
"taste." Now. the pleasure oflanguage has never been seriously
considered; the old Rhetoric had, in its fashion, some idea of it
when it set up a spial genre of discourse dedicated to sptacle
and to admiration, the tpidictic; but classical art wrapped the
pleasing which it claimed as its law (Racine: "The first rule is to
please ... ") in all the constraints of the "natural"; only the
baroque, a literary experiment which has never been more than
tolerated by our societies, at least by French society, dared some
exploration of what might be called the Eros of language.
Scientific discourse is remote from this; for if it accepted the
notion, it would have to renounce all the privileges with which

B
10 From Scitnct to LiUralure
the social institution surrounds it and agrtt to return to that
"literary life" Baudelaire calls, apropos of p, "the sale element
in which certain diclassiJ can breathe."
Mutation of consciousness, of structure, and of the purposes
of scientific discourse that is what must be demanded today,
precisely where the Hourishing, constituted human sciences seem
to leave less and less room for a literature commonly accused
of unreality and inhumanity. But precisely: the Tole of literature
is to refwt$tnt actively to the scientific institution just what it
rejects. i.e., the sovereignty of language. And structuralism
should be in a good position to provoke this scandal; for.
intensely conscious of the linguistic nature of human works.
only structuralism today can reopen the problem of the linguistic
status of science; having language-all languages-for object, it
has very quickly come to define itself as our culture's meta
language. This stage, however, must b e transcended, for the
opposition of language-objects and their meta-language remains
ultimately subject to the paternal model of a science without
language. The task facing structural discourse is to make itself
entirely homogeneous to its object; this task can be accomplished
by only two methods, each as radical as the other: either by a n
exhaustive formalization, or el.se by an integral writing. In this
second hypothesis (which we are defending here), science will
become literature, insofar as literature-subject, moreover, to a
growing collapse of traditional genres (poem, narrative, criti
cism, essay)--is already, has always been, science; for what the
human sciences are discovering today, in whatever realm: soci
ological, psychological, psychiatric, linguistic, etc., literature has
always known; the only difference is that literature has not said
what it knows, it has written it. Confronting this integral truth
of writing, the "human sciences," belatedly constituted in the
wake of bourgeois positivism, appear as the technical alibis our
society uses to maintain the fiction of a theological truth,
superbly-abusively-disengaged from language.

TM Tinw Li/n'ary Supplnnrnt, 1967

,
 To Write: An Intransitive Verb?

1. Uterature and linguistics

For centuries, Western culture conceived of literature not as we

do today. through a study of works, authors, and schools. but
through a veritable theory of language. Thi.s theory had a name,
Rlutoric. and it triumphed in the West from Gorgias to the
Renaissance, i.e., for over two thousand years. Threatened since
the sixteenth century by the advent of modern rationalism,
rhetoric was ahogether ruined when rationalism was trans
formed into positivism, at the end of the nineteenth. By then,
there was no longer any common zone of reSection between
literature and language: literature no longer regarded itself as
language, except in the work of a few precursor writers, such
as Mallarme. and linguistics claimed only very limited rights
over literature, these being enclosed within a secondary philo
logical discipline of uncertain status: stylistics.
As we k.now, this situation is changing, and it seems to me
that it is in pan to take cognizance of this change that our
colloquium has been assembled: literature and language are in
the process of recognizing each other. The factors of this
rapprochement are various and complex; I shall cite the most
obvious: on the one hand, the action of certain writers who
since Mallarme have undertaken a radical exploration of writing
and who have made their work a search for the total Book,
such as Proust and Joyce; on the other, the development of
linguistics itself, which henceforth includes within its scope
pottics. or the order of effects linked to the message and not to
its referent. Hence. there exists today a new perspective of
reflectioo-<ommon. I insist. to literature and to linguistics, to

"

B
 From Scinu:e to Literature
the creator and the critic, whose tasks, hitheno absolutely self
contained, are l>f=ginning to communicate, perhaps even to
converge, at least on the level of the writer, whose action can
increasingly be defined as a critique of language_ It is in this
perspective that I want to indicate by a few brief observations,
of a prospective and not conclusive nature, how the activity of
writing can today l>f= expressed [rnoncee] with the help of certain
linguistic categories_

2. Language
This new conjunction of literature and linguistics, which I have
just mentioned, might provisionally be called, for lack of a better
name, snnio-criticism, since it implies that writing is a system of
signs. Now, semio-criticism cannot be identified with stylistics.
even in a new form, or in any case, stylistics is far from exbausting
it. It involves a perspective of an altogether different scope.
whose object cannot be constituted by simple accidents of form,
but by the very relations between the scriptm- and language. This
perspective does not imply a lack of interest in language. but.
on the contrary, a continual return to the "truths," however
provisional, of linguistic anthropology. Certain of these truths
still have a power of provocation. in respect to a cenain current
idea of literature and of language. and for this reason. we must
not fail to consider them.
I. One of the teachings of contemporary linguistics is that
there is no archaic language, or that, at least, there is no relation
between a language's simplicity and its age: ancient languages
can be as complete and as complex as the recent ones; there is
no "progressive" history of languages. Hence, when we try to
recognize in modern writing certain fundamental categories of
language. we make no claim to reveal a cenain archaism of the
"psyche"; we are not saying that the writer harks back to the
origin of language, but that language is for him the origin.
2. A second principle. especially important with regard to
literature. is that language cannot be considered as a simple

B
To Write: An Intransitive VtrM '3

instrument-utilitarian o r decorative of thought. Man does

not exist prior to language. either as a species or as an individual.
We never encounter a state where man is separated from
language, which he then elaborates in order to "express" what
is happening within him: it is language which teaches the
definition of man, not the contrary.
3 . Moreover, from a methodological view, linguistics accus
toms us to a new type of objectivity. The objectivity hitherto
required in the human sciences is an objectivity of the given,
which must be accepted totally. Linguistics suggests, on the one
hand, that we distinguish levels of analysis and describe the
distinctive elements of each of these levels, in short. that we
establish the distinctness of the fact and not the fact itself; and
on the other, it asks us to recognize that, unlike physical and
biological facts. cultural facts are twofold, that they refer to
something else: as Benveniste has observed, it is the discovery
of language's "duplicity" which gives Saussure's reHection all its
value.
4. These few preliminaries are contained in a final proposition
which justifies all semio-critical research. Cuhure increasingly
appears to us as a general system of symbols. governed by the
same operations: there is a unity of the symbolic field. and
culture. in all its aspects, is a language. Hence, it is possible
today t o foresee the constitution of a unique science of culture,
which will certainly be based on various disciplines, but all
devoted to analyzing. at different levels of description, culture
as language. Semio-criticism will obviously be only a part of this
science, which will always remain a discourse on culture. This
unity of the human symbolic field authorizes us to elaborate a
postulate which I shall call a postulate of homology: the structure
of the sentence, object of linguistics, can be recognized homo
logically in the structure of works: discourse is not only a sum
of sc:ntences, it is, itself, one great sentence. It is in terms of
this working hypothesis that I would like to confront cenain
categories of language with the writer's situation in relation to
his writing. J am not concealing the fact that this confrontation

C JPYnghied mater1
14 From Science to Literature
does not have a demonstrative force and that for the moment
its value remains essentially metaphorical: but perhaps, too, in
the order of objects which concerns us, metaphor has-more
than we suppoa methodological ex.istence and a heuristic
force.

3. Temporality
As we know, there is a linguistic temporality, equaJly different
from physical time and from what Benveniste calls "chronicle"
time, or the time of calendars and computations. This linguistic
time receives extremely various comours and expressions in
various languages-for example. certain languages like Chinook
employ several pasts, including a mythic one but one thing
seems certain: the generating center of linguistic time is always
the present of the spttch-act (inonciationJ. This leads us to ask
whether there is, homologous to linguistic time, a time specific
to discourse. On this point. Benveniste offers an initial clarifi
cation: in many languages, specifically lndo-Euroan ones, the
system is twofold: (I) a first system, or system of discourse
pror, adapted to the temporality of the saker, whose sech
act is always the point of origin; (2) a second system, or system
of history, of narrative, appropriate to the recounting of past
events, without the saker's intervention and consequently
deprived of present and future (except riphrastically), its
specific tense the aorist (or its equivalents, like the French passe
simple), precisely the one tense missing from the system of
discourse. The existence of this a-rsonal system does not
contradict the essentially logocentric nature of linguistic time
we have just assened: the second system merely lacks the
characteristics of the first: one is linked to the other by the
opposition mar/c.td I unmar4td: consequently, they participate in
the same field of rtinence.
The distinction between the two systems is not at all the same
as the one traditionally made between objective discourse and
subjective discourse, for we cannot identify the relation of the

C ;.pvnghted malaria
To Write: An Intransitive VerM
speaker and the referent on one hand .....ith the relation of this
same speaker and the speech-act on the other, and it is only
this second relation which determines the temporal system of
the discourse. These linguistic phenomena were difficult to
perceive so long as literature was regarded as the docile and
"transparent" expression of either so-called objective (or chron
icle) time, or of psychological subjectivity, i.e . 50 long as litera
ture was placed within a totalitarian ideology of the referent.
Today, however. literature discovers in the unfolding of dis
course what I call certain fundamental subtleties: for example,
what is told in the aorist does not appear immersed in the past,
in "what has taken place," but only in the non-personal, which
is neither history nor science nor even the one of so-called
anonymous writing. for what prevails in this one is the indefinite,
not the absence of person: one is marked; he, paradoxically. is
not. At the other extreme of the experience of discourse, the
writer today. it seems to me. can no longer be content to express
his own present according to a lyrical project: he must learn to
distinguish the speaker's present, which remains grounded in
psychological plenitude, from the present of the locution, which
is as flexible as that locution and in which event and writing are
absolutely coincidental. Thus literature, at least in its explora
tions. is taking the same path as linguistics when, with Gustave
Guillaume. it concerns itself with operative time, or the time of
the speech-act itself.

4. Person
This leads to a second grammatical category, quite as important
in linguistics as in literature: that of person. First of all, we are
reminded by the linguists that person (in the grammatical sense
of the term) seems to be universal, linked to the very anthro
pology of language. Every language, as Benveniste has shown,
organizes person into two oppositions: a correlation of person
ality, which sets person (lor you) in opposition to the non
person (IK or it), sign of what is absent. of absence i15elf; and,

C JPYnghied mater1
.6 From Scnce to LittTatuTt
within this first great opposition, a correlation of subjectivity
sets two persons in opposition, the I and the non-J (i.e., you).
For our purposes, we must make three oppositions, following
Benveniste's lead. First of ail, the polarity of persons, a basic
condition of language. is nonetheless very special, for this
polarity involves neither equality nor symmetry: ego always has
a position of transcendence with regard to you, I being interior
LO what is Slaled and you remaining exterior to it; and yet I and
you are reversible. I can always become )'OU, and vice versa; this
is not the case for the non-person (he or it), which can never
reverse itself into person or vice versa. Second, the linguistic I
can and must be defined in an entirely a-psychological fashion:
I is nothing but "the person who utters the present instance of
discourse containing the linguistic instance I" (Benveniste). Last,
the nonperson never reflects the instance of discourse, being
situated outside of it; we must give its full weight to Benveniste's
recommendation that he or it is not to be represented as a more
or less diminished or distanced person: he or it i s absolutely
nonperson, marked by the absence of what specifically (i.e.,
linguistically) constitutes / and yOU.
From this linguistic explanation we shall draw several sugges
tions for an analysis of literary discourse. First of all, we note
that whatever the varied and often cunning forms (marks)
person may take when we proceed from sentence to discourse,
just as in the case of temporality, the work's discourse is subject
to a double system, that of person and that of nonperson. What
produces an illusion, here, is that our classical discourse (in the
broad sense) is a mixed one, which frequently ahernates--at a
rapid rate (for example, within the same sentence}-the personal
speech-act and the apersonal one, by a complex interplay of
pronouns and descriptive verbs. This mixed system of person
and non-person produces an ambiguous consciousness which
manages to keep the personal quality of what i s stated, yet
periodically breaking off the speaker's participation in the state
ment.
Second, if we return to the linguistic definition of the first
person (/ i s the one who says / in the present instance of
To Write: An Intransitive VtrM '7
discourse), we may bener understand the effort of certain writers
today (I am thinking of Sollers's Drame) when they try to
distinguish, on the level of the narrative itself, psychological
person from the author of the writing: contrary to the current
illusion of autobiographies and traditional novels, the subject of
the speech-act can never be the same as the one who acted
yesterday: the I of the discourse can no longer be the site where
a previously stored-up person is innocently restored. Absolute
recourse to the instance of discourse in order to determine
person, which with Damourette and Pichon we might can nyn
egocnurism (consider the exemplary beginning of RobbeGrillet's
novel In the LAbyrinth: "I am alone here now"}-this recourse,
imperfect as its practice may still be, thus seems a weapon against
the general bad faith of a discourse which makes or would make
literary form merely the expression of an interiority constituted
previous to and outside of language.
Last, let u s recall this detail of linguistic analysis: in the process
of communication. the course of the I is not homogenous: when
1 liberate the sign I, 1 refer to myself insofar as I am speaking,
and here there is an act which is always new, even if repeated,
an act whose "meaning" is always unprecedented; but upon
reaching its destination, this I is received by my interlocutor as
a stable sign, product of a complete code, whose contents are
recurrent. I n other words, the I of the one who writes I is not
the same as the I which is read by )IOU. This basic dissymmetry
of language. explained by Jespersen and Jakobson by the notion
o f shifter or an overlapping of code and message, is finally
beginning to disturb literature by showing it that intersubjectiv
ity. or rather interlocution, cannot be accomplished simply by a
pious wish about the merits of "dialogue," but only by a deep,
patient, and often circuitous descent into the labyrinth of
meanmg.

S. Diathesis
There remains to be discussed one last grammatical notion
which may illuminate the activity of writing at its very center,

Malena m dlreitl
.8 From Science to Luera/uTe
since it concerns the verb to write itself. It would be interesting
to know at what moment this verb began to be used intransitively,
the writer no longer being the one who writes something, but
the one who writes-absolutely: this shifl is certainly the sign of
an important change in mentality. But does it really involve
intransivity? No writer. of whatever period, can be unaware that
he always writes something; we might even say that it is
paradoxically at the moment when w write seems to become
intransitive that its object. under the name book or texl, assumes
a special importance. Hence, it is nOl, at least primarily. on the
side of intransivity that we must look for the definition of the
modern verb to write. Another linguistic notion may give us the
key: that of diathesis or, as the grammar books put it, "voice"
(active, passive, middle). Diathesis designates the way in which
the subject of the verb is affected by the action; this is obvious
for the passive; and yet linguists tell us that, in Indo-European
at least, the diathetical opposition is not between active and
passive but between active and middle. According to the classic
example given by Meillet and Benveniste, the verb to sacrifice
(ritually) is active if the priest sacrifices the victim in my place
and for me, and it is middle voice if, taking the knife from the
priest's hands, I make the sacrifice for my own sake; in the case
of the active voice, the action is performed outside the subject,
for although the priest makes the sacrifice, h e is not affected by
it; in the case of the middle voice, on the contrary, by acting,
the subject affects himself, he always remains inside the action,
even if that action involves an object. Hence, the middle voice
does not exclude transitivity. Thus defined, the middle voice
corresponds exactly to the modern state of the verb to write: to
write is today to make oneself (he center of the action of speech,
it is to effect writing by affecting oneself, to make action and
affection coincide, to leave the scriptor inside the writing-not as
a psychological subject (the Indo-European priest could perfectly
well be overflowing with subjectivity while actively sacrificing
for his client), but as agem of the action. We can even take the
diathetic analysis of the verb to write a little further. We know

C JPYnghied mater1
To Writt: An Intransitivt VtTb1 '9

that in French cenain verbs have an active meaning in their

simple form (alltT-to go, amvtT-to arrive, rmtrtT-to return,
sortir-to leave) but take the passive auxiliary (itre-to be) in
forming the perfect tense (jt sUll alii, jt SUll arrive): in order to
explain this bifurcation peculiar to the middle voice, Guillaume
distinguishes between what he calls a diriment perfect (with the
auxiliary avoir-to have), which supposes an interruption of the
action due to the speaker's initiative (jt marcM, je m'arrite de
marchtT, j'ai march.e-I walk, I stop walking, I have walked), and
an inttgrant perfect (with the auxiliary itre-to be), peculiar to
the verbs which designate a semantic whole, which cannot be
delivered by the subject's simple initiative (je SUll sarti, it est mort
I have left, he has died-do not refer to a dirimmt interruption
of leaving or dying). To write is traditionally an active verb
whose paSt is diriment; but in our literature the verb is changing
status (if not form): to write is becoming a middle verb with an
integrant past, precisely insofar as to write is becoming an indi
visible semantic whole; so that the true past, the "right" past of
this new verb is not j'ai icrit but je sUll ierit-as one says jt sull
Ri, it est mort, etc., expressions in which. despite the verb etre.
there is no notion of the passive, since without forcing matters
we cannot transform jt SUll ierit-l am written-into on m'a
ierit-someone has written me.
Thus, in the middle voice of to write, the distance between
seriptor and language diminishes asymptotically. We could even
say that it is the writings of subjectivity. such as romantic writing,
which are active, for in them the agent is not interior but anunor
to the process of writing: here the one who writes does not
write for himself, but as if by proxy, for an exterior and
antecedent person (even if both hear the same name), while. in
the modern verb of middle voice to write, the subject is constituted
as immediately contemporary with the writing, being effected
and affected by it: this is the exemplary case of the Proustian
narrator. who exists only by writing, despite the reference to a
pseudo-memory.

C:lpynghled malenal
'0 From Scimct to LiUralure

6. The instance of discourse

These observations suggest that the central problem of modern
writing exactly coincides with what we might call the proble
matics of the verb in linguistics: just as temporality, person, and
diathesis define the positional field of the subject, so modern
literature is trying, by various experiments, to establish a new
position for the agent of writing in writing itself. The meaning
or the goal of this effort is to substitute the instance of discourse
for the instance of reality (or of the referent), that mythic alibi
which has dominatedstill dominates-the idea of literature.
The field of the writer is only writing itself, not as pure "form,"
conceived by an aesthetic of art for art's sake, but much more
radically as the only possible space of the Q1ie who writes.
It seems to me necessary to remind those who accuse such
investigations of solipsism, formalism, or scientism that by
returning to the fundamental categories of language, such as
person, tense. and voice, we place ourselves at the heart of a
problematics of imeriocution. for such categories are precisely
the ones where we may examine the relations of 1 and of what
is deprived of the mark of I. Inasmuch as person. tense, and
voice (so properly named) imply these remarkable linguistic
beings known as shifters, they compel us to conceive language
and discourse no longer in terms of an instrumental and
consequently reified nomenclature, but as the very exercise of
discourse: for example. the pronoun. which is doubtless the
most dizzying of the shifters, belongs structurally (I insist) to
discourse; this is. one might say, its scandal, and it is on this
scandal that we must work today, in linguistics and in literature;
we are trying to sound the depths of the "pact of speech" which
unites the writer and the other. so that each moment of discourse
is both absolutely new and absolutely understood. We can even,
with a certain temerity, give this research a historical dimension.
We know that the medieval Septenium. in its grandiose classifi
cation of the universe. prescribed two great sites of exploration:
on the one hand the se<:rets of nature (quadrivium), on the other

,
To Write: An Intransitive Vt'TM ..

the secrets of discourse (trillium: grammatica, rlutorica, didlectica);

this opposition was lost between the end of the Middle Ages
and our own time, language being considered only as an
instrument in the service of either reason or the hean. Today,
however, something of that ancient opposition is reviving: to
the exploration of the cosmos corresponds, once again, the
exploration of language, conducted by linguistics, psychoanal
ysis, and literature. For literature itself is a science-no longer
of the "human hean," but of human discourse; its investigation.
however, is no longer addressed to the secondary forms and
figures which constituted the object of rhetoric, but to the
fundamental categories of language: just as, in our Western
culture, grammar was born only long after rhetoric, so it is only
after having made its way for centuries through Ie beau littb-aire
that literature can raise the fundamental problems of language
without which it would not exist.

Colloquium at Johns Hopk.ins University, 1966

,
 Reflections on a Manual

I should like to ofTer some simple, even simplistic observations

suggested by a recent reading or rereading of a manual of
French literary history. While rereading or reading this manual,
which closely resembles those I remember from the lycie, I asked
myself this question: Can literature be anything else for us than
a childhood memory? I mean, what is it that continues. what is
it that persists, what is it that speaks of literature after the lycee?
If we were to make an objective inventory, we would answer
that what abides (from literature) in adult, current life is: certain
crossword puzzles, some televised quiz shows. the posters of the
centenaries of some writer's birth, some writer's death. a few
paperback titles, some critical allusions in the newspaper we're
reading for altogether different reasons-looking for something
altogether different from these allusions to literature. All of
which has a lot to do, I believe, with the fact that we French
have always been accustomed to identify literature with the
history of literature. The history of literature is an essentially
academic object which in fact exists only because it is taught; so
that the title of our conference, "The Teaching of Literature,"
is for me almost tautological. Literature is what is taught, period.
It is an object of teaching. It is generally agreed that at least in
France no major synthesis-say of the Hegelian type has been
produced on the history of our literature. If this French
literature is a childhood memory-and that is how I am taking
it-I should like to see this will be the object of a very limited
and quite banal inventory-what elements this memory consists
of.
First of all, this memory consists of certain objects which
recur, which continually repeat themselves, and which we might
..

B
Reflectibns on a Manual
almost call monemes of the meta-literary language or the
language of literary history; these objects are of course the
authoTS, the schools, the movements, the genres, and the cen
turies. And then, around these objects, there is a certain
actually very limited-number of features or predicates which
find a place and combine with each other. If we were to read
the manuals of literary history, we should have no difficulty in
determining the paradigmatics, the elementary structure of
these features, which appears to be that of couples in opposition
with an occasional mixed term; this is an extremely simple
structure: for instance, there is the archetypal paradigm of our
whole literature, romanticism-classicism (though French romanti
cism, on the international scale, seems a relatively poor thing),
occasionally amplified into romanticism-reaism-symbolism
l (for the
nineteenth century). As you know, the law of combinative
operations permits, with very few elements, the immediate
production of an apparent proliferation: by applying cenain of
these features to certain of the objects I have mentioned, we
produce cenain individualities, or cenain literary individuals.
This is how the manuals always present the centuries themselves:
in a paradigmatic fashion. Actually, it's odd how a century comes
to have a kind of individual existence, but it is precisely our
childhood memories which accustom us to make the centuries
into individuals of a sort. The four great centuries of our
literature are strongly individuated by our literary history: the
sixteenth is overflowing life; the seventeenth is unity; the
eighteenth is movement; and the nineteenth is complexity.
Other features are added which again can very nicely be set
in opposition, paradigmatized. Here is a random sampling of
these oppositions, these predicates which are fastened onto
literary objects: there is "exuberant" opposed to " restrained" ;
there is "lofty an" or "deliberate obscurity" opposed to "expan
siveness"; "rhetorical coldness" to "sensibility"-which overlaps
the familiar romantic paradigm of cold and wann-or again the
opposition between "sources" and "originality:' between "labor"
and "inspiration." What we have here are the rudiments of a

,
 From Science to Literature
little roster of this mythology of our literary history, one which
would begin by establishing those mythic paradigms of which
French textbooks have always been so fond, perhaps because
this was a good method of memorization or perhaps, on the
contrary, because a mental structure that functions by contraries
has a high ideological yield (we need an ideological analysis to
tell us). It is this same opposition that we encounter, for instance,
between Conde and Turenne, the great archetypes of two French
temperaments: if you put them together in a single writer
Oakobson has taught tht the poetic act consists in extending
a paradigm into a syntagm), you produce an author who
reconciles, for example, "formal an and extreme sensibility" or
who manifests "a witty nature concealing a tragic sense" (such
as Villon). What 1 am saying here is simply the sketch of what
we might imagine as a kind of little grammar of our literature,
a grammar which would produce stereotyped individuations:
authors, movements, schools.
Second element of this memory: French literary history
consists of dismissals we need to explore. There is-as we know,
as has already been said-a whole olher history which would be
precisely the history of such dismissals. What are these " censor
ships"? First of all, the social classes; the social structure which
underlies this literature is rarely found in manuals of literary
history, we must turn to more emancipated, more highly de
veloped critical works in order to find it; when we read these
manuals, references to class structure may sometimes exist, but
only in passing and as aesthetic oppositions. Actually, what the
manual sets in opposition are class atmospheres, not realities;
when the aristocratic "spirit" is opposed to the bourgeois and
folk spirit, at least for previous centuries, it is the distinction of
a refined taste which is opposed to good humor and realism.
We also find, even in recent textbooks, sentences of this son:
"A plebeian, Diderot lacks tact and delicacy; he commits faults
of taste which affect the sentiments themselves with a cenain
vulgarity . . ." Thus, class exists, but as an aesthetic or ethical
atmosphere; on the level of the instruments of knowledge, these

B
Reflections on a Manual
manuals betray the flagrant absence of any economics or soci
ology ofour literature. The second "censorship" would obviously
be mat of sexuality, but I shall not discuss it here, because it
overlaps the much more general censorship which our entire
society brings to bear upon sex. A third "censorship"-for my
part, I regard it as a censorshiwould be that of the very
concept of literature, which is never defined as a concept,
literature in these manuals being an object which is selfunder
stood and never interrogated in order to define, if not its being,
at least its social, symbolic, or anthropological functions; whereas
in fact we might reverse this omission and say-in any case, I
personally should be glad to say-that the history of literature
ought to be conceived as a history of the idea of literature, and
that such a history does not seem to exist, for the moment.
Finally, a fourth "censorship," and not the least important, bears
on "languages," as always. A language is a much more important
object of censorship, perhaps, than all the rest. By which I mean
a manifest censorship, the kind these manuals bring to bear on
states of language remote from the classical norm. This is a
well-known phenomenon: there is a vast censorship of preciosity,
which notably in the seventeenth century is described as a sort
of classical inferno: every French person, through the teaching
of our school system, has the same judgment and the same view
of preciosity as Boileau, Moliere, or La Bruyere. This one-way
indictment is repeated for centuries-and this despite what a
real history of literature would readily make clear, i.e., the
enormous and persistent success of preciosity throughout the
seventeenth century, since even in 1663 a voluminous collection
of patsies galantel by the Comtesse de Suze went into fifteen
printings. Hence, there is a point to clarify here-a point of
censorship. There is also the case of sixteenth-century French,
what is called Middle French, which is rejected from our
language, on the pretext that it consists of ridiculous novelties.
halianisms,jargon, baroque audacities, etc., without ever raising
the question of what it is we have lost today in the great
traumatism of classical purity. We have lost not only means of

C :lpynghled malenal
.6 From Science to Lien-ature
expression, as they are called, but mentaJ structures as well, for
language is a mental structure. Here again, there is perhaps an
indictment to be brought. one which should obviously begin
with a condemnation of"c1assico-centrism," which in my opinion
still marks our whole literature, specifically in regard to lan
guage. Once again, we must include these problems of language
in the problems of literature; we must raise the great questions:
When ds a language begin? What ds to begin mean for a
language? When does a genre begin? What does it mean when
we are told of the first French novel, for instance? It is evident
that there is always. behind the classical idea of the language, a
political idea: the language's very being. i.e., its perfection and
even its name, is linked to a culmination of power: the Latin
classic is Lati n or Roman power; the French classic is monarchic
power. This is why it must be said that, in our teaching, we
cultivate, orwe promote, what I should call the paternal language
and not the mother tongue-particularly since, let me say in
passing, we do not know what spoken French is; we know what
written French is because there are grammars of good usage;
but no one knows what spoken French is; and in order to know,
we should have to begin by escaping our c1assico-centrism.
Third element of this chi.ldhood memory: this memory is
centered. and its center is as I have just said-classicism. This
c1assico-centrism seems anachronistic to us; yet we are still living
with it. Even now, we pass doctoral theses in the Salle Louis
Liard, at the Sorbonne, and we must inventory the portraits in
that hall; they are the divinities which preside over French
knowledge in its entirety: Corneille. Moliere, Pascal, Bossuet.
Descartes. Racine under the protection-this is an admission
of Richelieu. This c1assico-centrism goes far. then, since it always
identifies literature-and this even in the discussions of the
manuals-with the king. Literature is the monarchy. and invin
cibly the academic image of literature is constructed around the
name of certain kings: Louis XIV, of course. hut also Franois
I. St. Louis. so that, ultimately, we are presented with a kind of
shiny image in which king and literature reflect each other.
There is also. in this centered structure of our literary history,

GJPYnghtcd maknal
Refltctions on a Manual '7
a natUmal identification; these manuals of history perpetually
advance what are called typically French values or typically
French temperaments; we are told, for instance. that JoinviUe
is typically French; what is French-General de Gaulle has
provided one definition-is what s
i "regular, 1Wr7M1, national."
This is obviously the range of our literature's norms and values.
From the moment that this history of our literature has a center,
it is obvious that it s
i constructed in relation to this center; what
comes after or before in the structure s
i presented as harbinger
or desenion. What is before classicism heralds dassicism
Montaigne is a precursor of the classics; what comes after
classicism revives or betrays it.
A last remark: the childhood memory I invoke borrows its per
manent structuration, down through these centuries, From a grid
which is no longer a rhetorical gTid in our teaching, for that was
abandoned around the middle of the nineteenth century (as Gh
ard Genette has shown in a splendid article on the problem); it s
i
now a psychological grid. All academic judgments rest on the
conception of form as the subject'S "expression." Penonality is
translated into style: this postulate nourishes all judgments and
all analyses concerning authors; whence, ultimately, the key value,
the one most often invoked to judge authors: sincerity. For in
stance, du Bellay will be praised for having produced cenain sin
cere and personal cries; Ronsard had a sincere and profound
Catholic faith; Villon, a cry From the heart. etc.
These remarks are simplistic, and I am uncertain as to their
value in a discussion, but I should like to conclude them with a
last observation. To my sense, there is a profound and irredu
cible antinomy between literature as practice and literature as
teaching. This antinomy is serious because it is attached to what
is perhaps the most serious problem we face today, the problem
of the transmission of knowledge; this is doubtless. now, the
fundamental problem of alienation, for if the gTeat structures
of economic alienation have been more or less revealed, the
structures of the alienation of knowledge have not; I believe
that in this regard a political conceptual apparatus is not enough
and that there must be, precisely, one of psychoanalytic analysis.

C JPYnghied mater1
 From Science w Literature
Hence, it is for this that we must work., and this will have many
suquent repercussions on literature and on what can be done
with it in teaching, supposing that literature can subsist in
teaching, that it is compatible with teaching.
Meanwhile, we can indicate certain points of provisional
correction; within a teaching system which retains literature on
its program, I see three immediate ones. The first would be to
reverse c1assico-centrism and to "do" literary history baclcwards:
instead of envisioning the history of literature from a pseudo
genetic point of view, we should make ourselves the center of
this history, and if we really want to "do" literary history,
organize this history starting from the great modem break.;
thus, past literature would be dealt with through present-day
disciplines, and even in present-day language: we should no
longer see first-year lycie students obliged to study a sixteenth
century whose language they scarcely understand. on the pretext
that it comes before the seventeenth century, itself beset by
religious disputes unrelated to their present situation. Second
principle: to substitute lext for author, school. and movement.
The text, in our schools. is treated as an object of explication,
but an explication of the text is itself always attached to a history
of literature; the text must be treated not as a sacred object
(object of a philology), but essentially as a space of language, as
the site of an infinite number of digressions, thereby tracing,
from a certain number of texts, a certain number of codes of
knowledge invested in them. Finally, a third principle: at every
opportunity and at every moment to develop the polysemic
reading of the text, to recognize finally the rights of polysemy,
to construct a sort of polysemic criticism, to open the text to
symbolism. This would produce, I believe, a considerable de
compression in the teaching of our literature-not, I repeat. as
teaching is practiced-that depends on the teachers-but as it
seems to me to be codified still.

Colloquium at the Centre culture!

international de Cerisy-Ia-Salle. 1969

GJPYnghtcd maknal
 Writing Reading

Has it never happened, as you were reading a book, that you

kept stopping as you read, not cause you weren't interested,
but because you were: because of a How of ideas. stimuli,
associations? In a word, haven't you ever happened to read while
/oohing up from JOUT boo?
It s
i such reading, at once insolent in that it interrupts the
text, and sminen in that it keeps returning to it and feeding on
it, which I tried to describe. In order to write it, in order for
my reading to become in its tum the object of a new reading
(that of the readers of Sill, I obviously had to try to systematize
all those moments when one loolcs up. In other words. to
interrogate my own reading was to try to grasp the form of all
readings (form: sole site of science). or again: to devise a theory
of reading.
I therefore took a shon text (this was essential to the detailed
scope of the enterprise), Balzac's Sarrasine. a little-known tale
{but isn't Balzac defined precisely as Inexhaustible. the author
no one ever reads all of. except by some exegetic vocation?},
and I kept stopping as I read this text. Criticism ordinarily
functions (this is not a reproach) either by microscope (patiently
illuminating the work's philological, autobiographical, or psy
chological details) or by telescope (scrutinizing the great histor
ical space surrounding the author). I denied myself these two
instrumenu: I spoke neither of Balzac nor of his time, I explored
neither the psychology of his characters nor the thematics of
the text nor the sociology ofthe anecdote. Recalling the camera's
first feau in decomposing a horse's trot, 1 too attempted to
"film" the reading of Sarrasint in slow motion: the result, I
suspect. s
i neither quite an analysis (I have not tried to grasp

'9

I...1 1 1
 From Scienct to Liuraturt
the stcrd of this strange text) nor quite an image ( l don't think
I have projected myself imo my reading; or if I have, it is from
an unconscious site which falls far short of "myself"). Then
what is SIZ? Simply a text, that text which we write in our head
when wt look up.
Such a text, which we should be able to call by a single word,
text-as-reading, is little known because for centuries we have
been overly interested in the author and insufficiently in the
reader; most critical theories try to explain why the author has
written his work, according to which pulsions, which constraints,
which limits . . . This exorbitam privilege gramed to the site
the work comes from (person or Story), this censorship applied
to the site it seeks and where it is dispersed (reading) determine
a very special (though an old) economy: the author is regarded
as the eternal owner of his work, and the rest of us, his readers,
as simple usufructuaries. This economy obviously implies a
theme of autlwrity: the author, it is believed, has certain rights
over the reader, he constrains him to a certain meaning of the
work, and this meaning is of course the right one, the real
meaning: whence a critical morality of the right meaning (and
of its defect, "misreading"): we try to establish what tht autlwr
meant, and not at all whaJ tIu rtader understands.
Though certain authors have themselves notified us that we
are free to read their text as we choose and that they are not
really iDlerested in our choice (Valery), we still find it hard to
perceive how the logic of reading differs from the rules of
composition. These, inherited from rhetoric, are still taken as
referring to a deductive, i.e., rational model: as in the case of
the syllogism, it is a matter of compelling the reader to a
meaning or an issue: composition chanruls; reading, on the
contrary (that text we write in ourselves when we read), dispersts,
disseminates; or at least, dealing with a story (like that of the
sculpwr Sarrasine), we see dearly that a certain constraint of
our progress (of "suspense") constantly struggles within us
against the text's explosive force, its digressive energy: with the
logic of reason (which makes this story readable) mingles a logic

I...1 1''''''
Writing Rtading 31

of the symbol. This lauer logic is not deductive but associative:

it associates with the material text (with each of its sentences)
othtr ideas, other images, Qthtr significations. "The text, only the
text," we are told, but "only the text" does not exist: there is
immediattly in this tale, this novel, this poem I am reading, a
supplement of meaning for which neither dictionary nor gram
mar can account. It is this supplement whose space I wanted to
explore in writing my reading of Balzac's Sarrasint'o
I have not reconstituted a reader (you or myself) but reading.
I mean that every reading derives from trans-individual forms:
the associations engendered by the letter (but where is that
letter?) are never, whatever we do, anarchic; they are always
caught up (sampled and inserted) by certain codes, certain
languages, certain lists of stereotypes. The most subjective
reading imaginable is never anything but a game played ac
cording to certain rules. Where do these rules come from?
Certainly not from the author, who does nothing but apply
them in his own way (this can be inspired, as in Balzac's case);
visible apart from him. these rules come from an age-old logic
of narrative. from a symbolic form which constitutes us even
before we are born-in a word, from that vast cultural space
through which our person (whether author or reader) is only
one passage. To open the text, to posit the system of its reading,
is therefore not only to ask and to show that it can be interpreted
freely; it is especially, and much more radically, to gain aclmowl
edgment that there is no objective or subjective truth of reading,
but only a ludic truth; again, "game" must not be understood
here as a distraction. but as a piece of work-from which,
however, all labor has evaporated: to read is to make our body
work (psychoanalysis has taught us that this body greatly exceeds
our memory and our consciousness) at the invitation of the
text's signs, of all the languages which traverse it and form
something like the shimmering depth of the semence.
I can easily imagine readable narrative (the one we can read
without declaring it "unreadable': who does nof understand
Balzac?). As one of those articulated lay figures that painters

N , 1 1"';'
3' From Science to Litn"aturt
use (or used to use) in order to "catch" the various postures of
the human body; reading. we too imprint on the text a certain
posture, and it is for this reason that it is alive; but this posture,
which is OUT invention, is possible only because there is a
governed relation among the elements of the text, in shoTt a
proportion: I have tried to analyze that proportion, to describe
the topological disposition which gives the reading of a classical
text both its contour and its freedom.

U Figaro lil/baire, 1970

 On Reading

I should like, first of all, to thank the Writing Conference of

Luchon for wdcorning me here. Many things unite us, beginning
with that shared question which each of us asks from his own
position: What is reading' How does ont read' Why does ont muir
One thing. however, separates us, which I shall not atlempt to
disguise: I have. for a long while, ceased to engage in any
pedagogical practice: school, lyeit. and college are today un
k.nown to me, and my own teaching practice-which coums fOT
a great deal in my life-at the Ecole des Hautes Etudes, is very
marginal. very anomie, even within university teaching, Now,
since this is a congress, it seems to me that each of us should
make his own voice heard-the voice of his practice; hence. I
shall not compel myself to join, to mimic, a pedagogical com
petence which is not my own: I shall abide by a particular
reading (as any reading is?)-the reading of the subject I am,
whom I believe myself to be.
I am, with regard to reading, in a great doctrinal confusion:
as for a doctrine of reading. I have none; on the other hand, a
doctrine of writing is gradually taking hape. This confusion
sometimes goes so far as to become a doubt: I do not even know
if one must have a doctrine of reading; I do not know if reading
is not, constitutively, a plural field of scattered practices, of
irreducible effects, and if, consequently. the reading of reading.
meta-reading, is not itself merely a burst of ideas, of fears, of
desires, of delights, of oppressions about which we should speak
in fits and starts, blow by blow. in the plural image of the many
and various workshops which constitute this congress.
I shall not attempt to reduce this confusion (moreover, I have
no means of doing so), but only to situate it, to comprehend

C ;.pvnghted malaria
34 From Scil!nct 10 LittratuTi!

the excess of which the notion of reading is evidently the object

in myself. Where to start? Well. perhaps with what has permiued
modern linguistics to get under way: with the notion ofpertinence.

1. Pertinence
Pertinence is---or al least was-in linguistics the point of view
from which one chooses to consider, to question, to analyze an
ensemble as heterodite, as disparate as language: it was only
when he had made up his mind to regard language from the
point of view of meaning. and from that point of view alone,
that Saussure stopped fumbling. left panic behind, and was able
to establish a new linguistics; it was by deciding to consider
sounds alone within the pertinence of meaning alone that
Trubetskoy and Jakobson launched the development of pho
nology; it was by consenting. at the expense of many other
possible considerations. to see in hundreds of folk tales only
silUations and stable, recurrenl roles-in short, forms-that
Propp founded the structural analysis of narrative.
If, then, we could determine a pertinence within which to
interrogate reading. we might hope to develop. gradually. a
linguistics, a semiology. or simply an analysis of reading-from
anagnosis: an anagnosology: why not?
Unfortunately. reading has not yet encountered its Propp or
its Saussure; that desired pertinence, image of the scholar's
alleviation, has not been found-at least not yet: the old perti
nences do not suit reading, or at least reading overflows them.
l. In the field of reading, there is no pertinence of objects:
the verb to read, apparent1y much more transitive than the verb
to speak, can be saturated, catalyzed by a thousand complements
of objects: I read texts, images, cities, faces, gestures, scenes.
etc. These objects are so varied that I cannot unify them within
any substantial nor even formal category; I can find only one
intentional unity for them: the object I read is founded by my
intention to read: it is simply legmtium, to be read, issuing from
a phenomenology, not from a semiology.

Malena :>m d reitl al

On Reading 35

2. In the field of reading-and this is more serious-there is

not only no pertinence of leveiJ, there is no possibility of
describing levels of reading, because there is no possibility of
closing the list of these levels. Of course, there is an origin of
graphic reading: this is the apprenticeship to letters, to written
words; but, on the one hand, there are readings without
apprenticeship (images)-at least without technical, if not cul
tural apprenticeship and on the other hand. once this uchne
is acquired, we do not know where to halt the depth and the
dispersion of reading: at the apprehension of a meaning? Which
meaning? Denoted? Connoted? These are artifacts, I shall call
them ethical artifacts, since denoted meaning tends to pass for
the simple, true meaning and to found a law (how many men
have died for a meaning?), while connotation permits (this is its
moral advantage) positing a law with multiple meanings and
thereby liberating reading: but how far? To infinity: there is no
structural obligation to close my reading: 1 can just as well extend
the limits of the ,eadable to infinity, decide that everything is
finally readable (unreadable as this Sttms), but also, conversely,
I can decide that in the depths of every text, however readable
its conception, there is, there remains a certain measure of the
unreadable. Our knowing fww to read can be determined. verified
at its inaugural stage, but it very quickly becomes a knowledge
without basis, without rules, without degrees, and without end.
This difficulty in finding a pertinence, from which to establish
a coherent analysis of reading, we must assume we are spon
sible for simply because we lack genius. But we can also suppose
that non-pertinence is somehow congenital to reading: something,
statutorily, comes to blur the analysis of the objects and levels
of reading, and thereby checkmates not only any search for a
pertinence in the analysis of reading, but even, perhaps, the
very concept of pertinence (for the same thing seems to be
happening in the realm of linguistics and narratology). This
something I believe I can name (in a quite banal fashion,
moreover): it is Desire. It is because every reading is steeped in
Desire (or Disgust) that anagnosology is difficult, perhaps im-

Matena dlreitl
 From Science to Literature
possible-in any case, that it is likely to be achieved just where
we do not expect it, or at least not exactly where we expect it:
by-recent-tradition, we expect it in the realm of structure;
and no doubt we are partly right: every reading occurs within
a structure (however multiple, however open), and not in the
allegedly free space of an alleged spontaneity: there is no
"naturdl," "wild" reading: reading does not ovtiflow structure;
it is subject to it: it needs structure, it respects structure; but
reading perverts structure. Reading is the gesture of the body
(for of course one reads with one's body) which by one and the
same movement posits and perverts its order: an interior
supplement of perversion.

2. Repression

I a m not inquiring, strictly speaking, about the avatars of the

desire to read: notably, I cannot answer this irritating question:
Why don't Frenchmen today wanl 10 read? Why, it appears, do
fifty percent of them not read? What we might consider is the
trace of desire or of non-desire-that Ihere is within a reading,
supposing that the will-ta-read has already been assumed. And
first of all, the rtpressioru of reading. Two of which come to
mind.
The first results from all the constraints social or interiorized
by a thousand relays-which make reading a dury, in which the
very act of reading is determined by a law: the act of reading,
or better still, the act of having read, the almost ritual trace of
an initiation. Hence, I am not speaking of the "instrumental"
readings necessary to the acquisition of a specific kind of
knowledge, of a technique, and according to which the gesture
of reading vanishes beneath the act of learning: I am talking
about "free" readings, which you "must" nonetheless have
performed: you "mwt" have read (The Princm of CtroeJ, Anti-Otd
iPW). Where does the law come from? From various instances,
each of which is instituted as a value, as ideology: for the avant
garde militant, you "mwl" have read Bataille, Anaud. For a long
time, when reading was narrowly elitist, there were duties of

Malena :>m d reitl al

On Reading 37
universal reading; 1 suppose mat the collapse of humanist values
has pOl an end to these duties' of reading: for them have been
substituted certain private duties, linked to the "role" the subject
acknowledges he has in today's society; the law of reading no
longer comes from, an eternity of culture, but from a bizarre.
or at least enigmatic. instance located between History and
Fashion. What I mean is that there are group laws, microlaws,
from which one must be entitled to liberate oneself. Or again:
freedom to read. whatever its price. is also freedom not to read.
Who knows if certain things are not transformed, who knows
if cenain imponant things do not happen (in work, in me
history of the historical subject) not only by the effect of readings
but also by the effect of reading's omissions (forgettings): by
what we might call the uncorutTaints of reading? Or again: in
reading, Desire cannot be detached, whatever the cost to our
institutions, from its own pulsional negativity.
A second repression s
i perhaps that of the Library. No
question, of coune. of contesting the institution, or of ignoring
its necessary development; but a question of acknowledging the
trace of repression in this fundamental and inevitable feature
of the public (or simply: collective) Library: its facticitj. Facticity
in itself is not a road to repression (Nature has nothing panic
ularly liberating about it); if the Library's facticity produces a
failure in the Desire to read, it is for two reasons.
1 . By status, whatever its dimension. the Library s
i infinite,
insofar as it is always (however well-conceived it may be) bom
short of and in excess of demand: tendentially, the book you
want is never there, while anomer book is offered to you: the
Library is the space of substitutes for desire; confronting the
adventure of reading, it is reality, in that it calls Desire to order:
to derive pleasure, satisfaction, gratification from a Library, the
subject must renounce the effusion of his own imagerepertoire;
he must have dent his Oedipus that Oedipus 1 must not only
"do" at the age of four, but every day of my life that I desire.
Here in the Library, it is the very profusion of boolu whkh is
the law, castration.
2 . The Library is a space one visits, but not that one inhabits.

Malena :>m d reitl al

 From Scitnct 10 Littralure
We should have in our language. versatile as it is said to be. two
different words: one for the Library book. the other for the
book-athome (let us use hyphens. producing an autonomous
syntagm whose referent is a specific object); one for the book
"borrowed"-usually through a bureaucratic or magistral me
diation-the other for the book grasped, held. taken up as if it
were already a fetish: one for the book-as-object of a debt (it
must be returned), the other for the book-as-object of a desire
or an immediate (without mediation) demand. Domestic (and
not public) space deprives the book of any function of social,
cultural. institutional appearance. Of course the book-at-home is
not a pure fragment of desire: it is (generally) traversed by a
mediation which has nothing particularly clean about it: money:
we have had to buy it. which means not having bought others;
but things being what they are, money is itself a liberation
which the Institution is not: in the Fourierist utopia, books are
worth virtually nothing. but they nonetheless experience the
mediation of a few pennies: they are covered by an Expense.
whereupon Desire functions: something is released.

3. Desire
What is there of Desire in reading? Desire cannot be named.
not even (unlike Demand) expressed. Yet it is certain that there
is an eroticism of reading (in reading. desire is there with its
object, which is the definition of eroticism). Of this eroticism of
reading. there is perhaps no purer apologue than that episode
in Proust's novel where the young Narrator shuts himself up in
the Combray bathroom in order to read, so as not to see his
grandmother suffer when she has been told, as a joke. that her
husband is going to drink cognac . . . ): ". went up sobbing to
the very top of the house. to the room next to the schoolroom,
under the roof, a little room smelling of iris, and also perfumed
by a wild currant bush sprouting between the stones of the wall
outside and which thrust a flowering branch through the open
window. Intended for a more particular and vulgar use, this

Matena dlreitl
On Rtading 39

room, from which one had a view, during the day, aU the way
to the donjon of Roussainville-Ie-Pin, long served me as a refuge,
doubtless because it was the only place I was allowed to lock
myself in, for all those occupations of mine which required an

inviolable solitude: reading. reverie, tears. and pleasure."

Thus, a dtsiring reading appears. marked with two institutive
features. By shutting himself up to read. by making reading
into an absolutely separated. clandestine state i n which the whole
world is abolished, the reader is identified with two other human
subjects-actually quite close to each other-whose state also
requires a violent separation: the amorous subject and the mystic
subject; Theresa of Avila specified reading as a substitute for
mental prayer; and the amorous subject, as we know, is marked
by a retreat from reality. he releases himself from the outer
world. This certainly confirms that the reader-subject is a subject
entirely transposed into the register of the image-repertoire;
his whole economy of pleasure consists in nursing his dual
relation with the book (i.e., with the Image), by shutting himself
up alone with it. fastened to it. like the child fastened to the
mother and the Lover poring over the beloved face. The iris
smelling bathroom is the very closure of the Mirror. the site of
the paradisiac coalescence of subject and Image of the book.
The second feature from which a desiring reading is consti
tuted-as we are told explicitly by the bathroom episode-is
this: in reading, all the body's emotions are present, mingled,
coiled up: fascination, emptiness, pain. voluptuousness; reading
produces an overwhelmed body. but flot parceled out (otherwise.
reading would not issue from the image-repertoire). Yet some
thing more enigmatic is presented for us to read, to interpret
in the Proustian episode: reading-the delight of reading-has
some relation with anality; one and the same metonymy connects
reading. excrement. and-as we have seen-money.
And now-without leaving the reading room-this question:
Are there different pleasures of reading? Is there a possible
typology of these pleasures? It seems to me that there are, in
any case, at least three types of pleasure of reading or, to be

Matena dlreitl
 From Science to Literature
more specific. thrtt ways by which the Image of reading can
capture the reading subject. According to the first mode, the
reader has a fetishist relation with the text being read: he takes
pleasure in the words, in certain words, in certain arrangements
of words; in the texts, certain areas, certain solates
i are formed
and in their fascination the reader-subject is lost. ruined; this
would be a kind of metaphoric or poetic reading; to enjoy this
pleasure. is there any need of an extended linguistic culture?
This is not certain: even the very young child, at the stage of
prattle, knows the eroticism of the word, an oral and aural
practice available to pulsion. According to the second mode,
which is just the contrary, the reader is drawn onward through
the book's length by a force always more or less disguised,
belonging to the order of suspense: the book s
i gradually
abolished, and it is in this impatient, impassioned erosion that
the delectation lies; a matter, chieRy, of the metonymic pleasure
of all narration, without forgetting that knowledge itself can be
recounted, subjected to a movement of suspense; and because
this pleasure is visibly linked to the observation of what is
unfolding and to the revelation of what is hidden, we can
suppose there is some relation to the discovery of the primal
scene; I want to surprise, I am about to faint from expectation:
a pure image of delectation, in that it ds not belong [0 the
order of satisfaction; we should also question, conversely, the
blockages, the distastes of reading: Why don't we go on with a
book? Why cannot Bouvard, deciding to take up the Philosophy
of History, "finish Bossuet's celebrated Discours"? Is this Bou
vard's fault, or Bossuet's? Are mere universal mechanisms of
attraction? Is there an erotic logic of Narration? Here the
structural analysis of narrative should raise the problem of
Pleasure: it seems to me that it now has the means to do so.
Then there is a third adventure of reading (I am calling adventure
the way in which pleasure comes to the reader): that of Writing;
reading is a conductor of the Desire to write (we are now sure
that there is a delectation of writing, although it is Hill very
enigmatic for us); not that we necessarily wanted to write 1iJr.e

Malena :>m d reitl al

On Reading

the author we enjoy reading; what we desire is only the desire

the scriplOr' has in writing. or again: we desire the desire the
author had for the reader when he was writing, we desire the
love-me which is in all writing. This has been very clearly put by
the writer Roger Laporte: "A purt reading which does not call
for another writing is incomprehensible to me . . . Reading Proulll,
Blanchot. Kafka. Artaud gave me no desire to write on these
authors (not even, I might add. like them). but to wrilt." In this
perspective. reading is a veritable production: no longer of
interior images, of projections, of hallucinations, but literally of
WOTA:: the (consumed) product is reversed into production. into
promise, into desire for production, and the chain of desires
begins to unroll. each reading being worth the writing it
engenders. to infinity. Is this pleasure of production an elitist
pleasure, reserved only to potential writers? In our society, a
society of consumption and not of production, a society of
reading, seeing, and hearing, and not a society of writing,
looking. and listening. everything is done to block the answer:
lovers of writing are scattered. clandestine, crushed by a thou
sand-even internal-constraints.
This is a problem of civilization: but, for me, my profound
and constant conviction is that it will never be possible to liberate
reading if. in the same impulse. we do not liberate writing.

4. Subject
There has been a great deal of discussion, and long before the
advent of Structural Analysis. of the different points of view an
author can adopt to tell a story----or simply to produce a text. A
way of connecting the reader to a theory of Narration, or more
broadly to a Poetics, would be to consider him as himself
occupying a point of view (or several in succession); in other
words. to treat the reader a.s a character. to make him into one
of the characters (not even necessarily a privileged one) of the
fiction and/or the Text. Greek tragedy affords an example: the
reader is that character who is on stage (even if clandestinely)

B
4' From Science to LileraluTt
and who hears what each of the partners of the dialogue d5
not hear; his hearing is double (and therefore vinually multiple).
In other words. the reader's specific site is the paragram, as it
obssed Sau55ure (did he not feel he was going mad. this
scholar, from being souly and compktely the reader?): a "true"
reading. a reading which woul!i assume its affirmation, would
be a mad reading, not because it would invent improbable
meanings (misconstructions), not because it would be "delirious,"
but because it would n:eive the simultaneous multiplicity of
meanings, of points of view, of structures, a space extended
outside the laws which proscribe contradiction ("Text" is the
very postulation of such a space).
This imagination of a total-i.e., totally multiple, paragram
matic-reader may be useful in that it permits us to glimpse the
Paradox of the reader: it is commonly admitted that to read is
to decode: letters, words. meanings. structures, and this is
incontestable; but by accumulating decodings (since reading is
by rights infinite). by removing the safety catch of meaning. by
putting reading into freewheeling (which is its structural voca
tion). the reader is caught up in a dialectical reversal: finally,
he does not decode, he overcode.s; he does not decipher, he
produces, he accumulates languages, he lets himself be infinitely
and tirelessly traversed by them: he is that traversaL
Now, ths
i s
i the very situation of the human subject, at least
as psychoanalytic epistemology tries to understand him: a subject
who s
i no longer the thinJcing subject of idealistic philosophy, but
rather devoid of all unity, lost in the double misreading of his
unconscious and of his ideology, and remembering only a
whirligig of languages. I mean by this that the reader is the
complete subject, that the field of reading is that of absolute
subjectivity (in the materialistic sense which this old idealistic
word can now have): every reading proceeds from a subject,
and it is separated from this subject only by rare and tenuous
mediations, the apprenticeship of letters, a few rhetorical pro
tocols, beyond which (very quickly) it is the subject who redis
covers himself in his own, individual structure: either desiring,

B
On Reading 43

or perverse, or paranoiac, or imaginary, or neurotic-and of

course in his historica1 structure as well: alienated by ideology,
by the routines of codes.
This is to indicate that we cannot reasonably hope for a
Science of reading, a miology of reading, unless we conceive
the possibility, some day, of-a contradiction in terms a Science
of the Inexhaustible, of infinite Displacement: reading is precisely
that energy, that action which will seize-in this text, in this
book-the very thing "which refuses to be exhausted by the
categories of Poetics"; reading, in shon, is the permanent
hemorrhage by which structure-patiently and usefully de
scribed by Structural Analysis-<ollapses, opens, is lost, thereby
consonant with any logical system which ultimately nothing can
dose-leaving intact what we must call the movement of the
subject and of history: reading is the site where structure is
made hysterical.

u FrallfaU aujourd'Jnu. 1976

Oswiilld Ducro( and Tlvetan Todorov, Diltionrtairt tnCJC IW scU-ntft

dw I4ngagt (Paris, 1972), p. 107.

B
 Freedom to Write

Flaubert's last novel is missing a chapter on orthography. In it

we would find Bouvard and Pecuchet ordering from Dumouchel
a whole little library of spelling manuals. at first delighted, then
astounded by the comminatory and contradictory character of
the rules prescribed, finally working each other up and endlessly
arguing: Why this particular written form? Why write Cam, Paon.
Lampe, Vmt. Rang when the vowel sound is the same in each
case? Why Quatrt and Cail/e. since these two words have the
same initial consonant? Whereupon Pecuchet would inevitably
conclude, bowing his head: "Spelling might be a hoax!"
This hoax, as we know, is not an innocent one. or course, for
a historian of the language, these accidents of French onhog
raphy are explicable: each one has its reason, analogical, ety
mological, or functional; but the sum of these reasons is unrea
sonable, and when this unreason is imposed, by means of
education, upon an entire nation, it becomes culpable. It is not
the arbitrary character of our orthography which is shocking,
it is the fact that this arbitrariness is statutory. Since 1835, the
official orthography of the Academie Franr;aise has had the
value of law in the eyes of the State; from the very first classes
of the young French citizen, "spelling mistakes" are punished:
how many lives spoiled for a few spelling errors!
The first effect of spelling is discriminatory; but it also has
secondary effects of a psychological order. If orthography were
free-free to be simplified or not, according to the subject'S
desire-it might constitute a very positive practice of expression;
the written physiognomy of the word might acquire a properly
poetic value, insofar as it emerged from the 5criptor's phantas
matics, and not from a uniform and reductive law; just think

44

,
Freedom to Write 45
of the kind of intoxication, of baroquejubilation which explodes
in the orthographic "aberrations" of old manuscripts, of
texts by children and the letters of foreigners: might one not
say that in such efflorescences as these the subject seelu his
freedom: to trace, to dream, to remember, to understand?
Are there not occasions when we encounter particularly
"happy" spelling mutakes-as if the Scr1ptor were obeying not
academic law but a mysterious commandment that comes
to him from his own history-perhaps even from his own
body?
Conversely, once spelling is made uniform, legalized, sanc
tioned by state means, in its very complication and its irration
ality, it is obsessional neurosis which is instated: the spelling
mistake becomes Transgression. I have just sent off a letter of
application for a job which can change my life. But have I
remembered to put an s on that plural? Was I careful to put
two p's and just one l in apptin'? I worry. I am in agony, like
the vacationer who can't remember if he turned off the gas and
the water back home, and if a fire or a flood will be the result.
And just as such worry keeps our vacationer from enjoying his
vacation. legalized spelling keeps the scn'ptDT from enjoying
writing, that euphoric gesture which permits putting into the
tracing of a word a liule more than its mere intention to com
municate.
Reform spelling? It has been tried several times. it s
i tried
periodically. But what is the use of rerqaking a code. even an
improved one. if it is once again in order to impose it. to legalize
it, to make it a specifically arbitrary instrument of selection? It
is not spelling which should be reformed. but the law which
prescribes its minutiae. What can be asked is this: a certain
"Iaxism" of the Institution. If I enjoy writing "correctly", i.e ..
"in conformity." 1 am quite free to do so, as 1 am to enjoy
reading Racine or Gide today: statutory spelling is not without
its charm. it is not without perversity; but let "ignorances" and
"blunders" be penalized no longer; let them cease to be perceived
as aberrations or debilities; let society agree at last (or once

,
 From Scitnct to Literature
again) to relea writing from the state apparatus to which it
belongs today; in shon, let us stop excluding "for reasons of
spelling,"

C ;.pvnghted malaria
 2
FROM WORK TO mer

B
 The Death of the Author

In his tale Sarrasine. Balzac. speaking of a castrato disguised as

a woman, writes this sentence: "She was Woman, with her
sudden fears, her inexplicable whims, her inSlinctive fears, her
meaningless bravado, her defiance, and her delicious delicacy
of feeling." Who speaks in this way? Is it the hero of the tale.
who would prefer not to recognize the castrato hidden beneath
the "woman"? Is it Balzac the man, whose personal experience
has provided him with a philosophy of Woman? Is it Balzac the
author. professing certain "literary" ideas about femininity? Is
it universal wisdom? Romantic psychology? We can never know,
for the good reason that writing is the destruction of every
voice, every origin. Writing is that neuter, that composite. that
obliquity into which OUT subject flees, the black-and-white where
all identity is lost. beginning with the very identity of the body
that writes.

No doubt it has always bttn so: once a fact is rtCounted-for

intransitive purposes. and no longer to act direct1y upon reality,
i.e., exclusive of any function except that exercise of the symbol
itself-this gap apars, the voice loses its origin. the author
enters into his own death. writing begins. However. the affect
of this phenomenon has been variable; in ethnographic societies,
narrative is never assumed by a T50n but by a mediator.
shaman. or reciter. whose "rformance" (i.e . his mastery of
the narrative code) can be admired, but never his "genius,"
The author is a modern character, no doubt produced by our
society as it emerged from the Middle Ages, inflected by English
empiricism. French rationalism. and the rsonal faith of the
Reformation. thereby discovering the prestige of the individual.

49

,
50 From Work to Text
or, as we say more nobly, of the "human person." Hence, it is
logical that in literary mauers it should be positivism, crown
and conclusion of capitalist ideology, which has granted the
greatest importance to the author's "person." The author still
reigns in manuals of literary history, in biographies of writers,
magazine interviews, and in the very consciousness of litterateurs
eager to unite. by means of private journals, their person and
their work; the image of literature to be found in contemporary
culture is tyrannically centered on the author, his person, his
hislory, his tastes, his passions; criticism still largely consists in
saying that Baudelaire's oeuvre s
i the failure of the man Bau
delaire, Van Gogh's is his madness, Tchaikovsky's his vice:
aplanaJion of the work is still sought in the person of its producer,
as if, through the more or less transparent allegory of fiction,
it was always, ultimately, the voice of one and the same person,
the author, which was transmitting his "confidences."

Though the Author's empire is still very powerful (the new

criticism has quite often merely consolidated it), we know that
certain writers have already tried to subvert it. In France,
Mallarme, no doubt the first, saw and foresaw in all its scope
the necessity to substitute language itself for the subject hitherto
supposed to be its owner; for Mallarme, as for us, it is language
which speaks, not the author; to write is to reach, through a
preliminary impersonality-which we can at no moment identify
with the realistic novelist's castrating "objectivity"-that point
where not "1" but only language functions, "performs": Mal
larme's whole poetics consists in suppressing the author in favor
of writing (and thereby restoring, as we shall see, the reader's
place). Valery, entangled in a psychology of the ego, greatly
edulcorated Mallarmean theory, but led by a preference for
classicism to conform to the lessons of Rhetoric, he continued
to cast the Author into doubt and derision, emphasized the
linguistic and "accidental" nature of his activity, and throughout
his prose works championed the essentially verbal condition of
literature, as opposed to which any resort to the writer's inte
riority seemed to him pure superstition. Proust himself. despite

,
The Death of the Author 5'
the apparently psy,hological character of what is called his
analyses. visibly undertook to blur by an extreme subtilization
the relation of the writer and his characters: by making the
narrator not the one who has seen or felt, or even the one who
writes, but the one who is going to wnu (the young man of the"
novel-but, as a matter of fact, how old is he and who is he?
wants to write but cannot, and the novel ends when writing
finally becomes possible), Proust has given modern writing its
epic by a radical reversal, instead of putting his life into his
novel, as is so often said, he made his life itself a work of which
his own book was the model. so that it s
i quite clear to us that
it is not Charlus who imitates Montesquiou, but Montesquiou,
in his anecdotal, historical reality, who is only a secondary,
derived fragment of Charlus. Finally Surrealism. to keep to this
prehistory of modernity. could doubtless not attribute a sover
eign place to language, since language s
i system, and what this
movement sought was, romantically, a direct subversion of the
codes-an illusory subversion. moreover, for a code cannot be
destroyed, only "flouted"; yet. by constantly striving to disap
point expected meanings (this was the famous surrealist "shock"),
by urging the hand to write as fast as possible what the head
was unaware of (this was automatic writing), by accepting the
principle and the experiment of collective writing. Surrealism
helped desacralize the image of the Author. Last, outside
literature itself (in fact, such distinctions are becoming quite
dated), linguistics furnishes the destruction of the Author with
a precious analytic instrument, showing that the speech-act in
its entirety is an "empty" process, which functions perfectly
without its being necessary to "fill" it with the person of the
interlocutors: linguistically, the author is nothing but the one
who writes, just as 1 is nothing but the one who says I: language
knows a "subject," not a "person," and this subject. empty
outside of the very speech-act which defines it. suffices to "hold"
language, i.e . to exhaust it.

The removal of the Author (with Brecht, we might speak

here of a veritable distancing. the Author diminishing like a

B
 From Work to Text
figure! at the far end of the literary stage) is not only a historical
fact or an act of writing: it utterly transforms the modern text
(or-which is the same thing-the text is henceforth produced
and read so that the author absents himself from it at every
level). Time, first of all, is no longer the same. The Author,
when we believe in him. is always conceived as the past of his
own book: book and author are voluntarily placed on one! and
the same line, distributed as a before and an after: the Author is
supposed to fud the book, i.e., he lives before it, thinks, suffers,
lives for it; he has the same relation of antecedence with his
work that a father sustains with his child. Quite the contrary,
the modern scriptor is born at tht .same time as his text; he is not
furnished with a being which precedes or exceeds his writing.
he is not the subject of which his book would be the predicate;
there is n o time other than that of the speechact, and every
text is written eternally lure and now. This is because (or it
follows that) writing can no longer designate an operation of
recording. of observation. of representation. of "painting" (as
the Classics used to say), but instead what the linguists, following
Oxfordian philosophy, call a performative, a rare verbal form
(exclusively found in the first person and in the present), in
which the speech-act has no other content (no other statement)
than the! act by which it is uttered: something like the I dtclare
of kings or the I .sing of the earliest poets; the modern .scriptor,
having buried the Author. can therefore no longer believe,
according to the pathos of his predecessors, that his hand is
slower than his passion and that in consequence, making a law
of necessity, he must emphasize this delay and endlessly "elab
orate" his form; for him, on the contrary, his hand, detached
from any voice, borne by a pure gesture of inscription (and not
of expression), traces a field without origin--or at least with n o
origin but language itself, i.e., the very thing which ceaselessly
calls any origin into question.

We know now that a text consists not of a line of words,

releasing a single "theological" meaning (the "message" of the

,
The Death of tJu Author 53
Author-God), but of a multi-dimensional space in which are
married and contested several writings, none of which is original:
the text is a fabric of quotations, resulting from a thousand
sources of culture. Like Bouvard and Pecuchet, those eternal
copyists, at once sublime and comical, whose profound absurdity
precisely designates the truth of writing, the writer can only
imitate an ever anterior, never original gesture; his sole power
is to mingle writings, to counter some by others, so as never to
rely on just one; if he seeks to express himself, at least he knows
that the interior "thing" he claims to "translate" is iuelf no more
than a ready-made lexicon, whose words can be explained only
through other words, and this ad infinitum: an adventure which
exemplarily befell young Thomas Ik Quincey, so versed in his
Greek that in order to translate certain absolutely modern ideas
and images into this dead language, Baudelaire tells us, "he had
a dictionary made for himself, one much more complex and
extensive than the kind produced by the vulgar patience of
purely literary themes" (Les Paradis artiflCitls); succeeding the
Author, the scriptQT no longer contains passions, moods, senti
ments, impressions, but that immense dictionary from which he
draws a writing which will be incessant: life merely imitates the
book, and this book itself is but a tissue of signs, endless imitation.
infinitely postponed.

Once the Author is distanced, the claim to "decipher" a text

becomes entirely futile. To assign an Author to a text is to
impose a brake on it, to furnish it with a final signified, to close
writing. This conception is quite suited to criticism, which then
undertakes the important task of discovering the Author (or his
hypostases: society, history, the psyche, freedom) beneath the
work: once the Author is found, the text is "explained," the
critic has won; hence. it is hardly surprising that historically the
Author's empire has been the Critic's as well, and also that (even
new) criti.cism is today unsettled at the same time as the Author.
In multiple writing, in effect, everything is to be disentangled,
but nothing deciPhered, structure can be followed. "threaded"

C:lpynghled malenal
54 From Work to Text
(as we say of a run in a sloelting) in all its reprises, all its stages,
but there is no end to it. no bottom; the space of writing is to
be tTaveTd, not pierced; writing conslandy posits meaning.
but always in order to evaporate it: writing seeks a systematic
exemption of meaning. Thereby. literature (it would be beuer,
from now on, to say writing). by refusing to assign to the text
(and to the world-as-text) a "secret," i.e., an ultimate meaning,
liberates an activity we may call countertheological, properly
revolutionary, for to refuse to halt meaning is finally to refuse
God and his hypostases, reason, science, the law.

To return to Balzac's sentence. No one (i.e., no "person") says

it: its source, its voice is not the true site of writing, it is reading.
Another very specific example will help us here: recent inves
tigations a.-p. Vernant) have shed some light on the constitu
tively ambiguous nature of Greek tragedy, whose text is "woven"
of words with double meanings. words which each character
understands unilaterally (this perpetual misunderstanding is
precisely what we call the "tragic'); there is, however. someone
who understands each word in its duplicity. and funher under
stands, one may say, the very deafness of the characters speaking
in his presence: this "someone" is precisely the reader (or here
the listener). Here we discern the total being of writing: a text
consists of multiple writings. proceeding from several cultures
and entering into dialogue. into parody. into contestation; but
there is a site where this multiplicity is collected. and this site s i

not the author. as has hitheno been claimed. but the reader:
the reader is the very space in which are inscribed. without any
of them being lost, all the citations out of which a writing is
made; the unity of a text is not in its origin but in its destination.
but this destination can no longer be personal: the reader is a
man without history. without biography, without psychology;
he is only that someone who holds collected into one and the
same field all of the traces from which writing is constituted.
That is why it is absurd to hear the new writing condemned in
the name of a humanism which hypocritically claims to champion

C ;.pvnghted malaria


The Dealh of tJu Author 55

the reader's rights. Classical criticism has never bttn concerned
with the reader; for that criticism. there is no other man in
literature than the one who writes. We are no longer so willing
to the dus of such antiphrases, by which a society proudly
recriminates in favor of precisely what it discards, ignOTe5,
muffles, or destroys; we know that in order to restore writing
to its fmure, we must reverse the myth: the birth of the reader
must be requited by the death of the Author.

Manteia. 1968

C :lpynghled malenal
 From Work to Text

A change has lately occurred, or is occurring. in our idea of

language and consequently of the (literary) work which owes to
that language at least its phenomenal existence. This change is
obviously linked to the present development of (among other
disciplines) linguistics. anthropology, Marxism, psychoanalysis
(the word linlr is used here in a deliberately neutral manner: no
determination is being invoked. however multiple and dialecti
cal). The transformation of the notion of the work does not
necessarily derive from the internal renewal of each of these
disciplines, but rather from their intersection at the level of an
object which traditionally procttds from none of them. We
might say. as a maller of fact, that inwdisciplinary activity. today
so highly valued in research, cannot be achieved by the simple
confrontation of specialized branches of knowledge; the inter
disciplinary is not a comfonable affair: it begins effectively (and
not by the simple utterance of a pious hope) when the solidarity
of the old disciplines breaks down-perhaps even violently,
through the shocks of fashion-to the advantage of a new object,
a new language, neither of which is precisely this discomfort of
classification which permits diagnosing a certain mutation. The
mutation which seems to be affecting the notion of the work
must not, however, be overestimated; it is part of an epistemo
logical shift, more than of a real break of the kind which in fact
occurred in the last century upon the appearance of Marxism
and Freudianism; no new break has occurred since. and we
might say that for the last hundred years we have been involved
in a repetition. What History, our History, allows us today is
merely to displace, to vary, to transcend, to repudiate. Just as
Einsteinian science compels us to include within the object

56

,
From Wor. to Tat 57
studied the rlaJivily of refertnC points, so the combined action of
Marxism, Freudianism. and structuralism compels us, in litera
ture, to relativize the rtlations of scriptor. reader, and observer
(critic). Confronting the work a traditional notion, long since.
and still today, conceived in what we might call a Newtonian
fashion-there now occurs the demand for a new object, ob
tained by a shift or a reversal of previous categories. This object
is the Tat. I know that this word is fashionable (I mystlf am
compelled to use it frequently), hence suspect in some quarters;
but this is precisely why I should like to review the main
propositions at whose intersection the Text is located, as I see
it; the word proposition must here be understood more gram
matically than logically: these are speech-acts, not arguments.
"hints," approaches which agree to remain metaphorical. Here
are these propositions: they concern method. genres, the sign.
the plural. filiation. reading, pleasure.

1. The text must not be understood as a computable object.

It would be futile to attempt a material separation of works
from texts. In particular. we must not permit ourselves to say:
the work is classical, the text is avant-garde; there is no question
of establishing a trophy in modernity'S name and declaring
certain literary productions in and aut by reason of their
chronological situation: there can be "Text" in a very old work.
and many products of contemporary literature are not texts at
all. The difference is as follows: the work is a fragment of
substance. it occupies a portion of the spaces of books (for
example. in a library). The Text is a methodological field. The
opposition may recall (though not reproduce term for term) a
distinction proposed by Lacan: "reality" is shown [se montrej. the
"real" is proved [s dimontrel; in the same way, the work is seen
(in bookstores. in card catalogues, on examination syllabuses),
the text is demonstrated. is spoken according to certain rules
(or against certain rules); the work is held in the hand, the text
is held in language: it exists only when caught up in a discourse
(or rather it is Text for the very reason that it knows itself to

B
 From WorA: to Ttxt
be so); the Text is not the decomposition of the work, it is the
work which is the Text's imaginary tail. Or again: the Ttxt is
experienced only in an activity, in a production. It follows that the
Text cannot stop (for example, at a library shelf); its constitutive
moment is traversal (notably, it can traverse the work, several
works).

iii. Similarly, the Text does not stop at (good) literature; it

cannot be caught up in a hierarchy, or even in a simple
distribution of genres. What constitutes it is on the contrary (or
precisely) its force of subversion with regard to the old classifi
cations. How to classify Georges Bataille? Is this writer a novelist.
a poet, an essayist, an economist, a philosopher, a mystic? The
answer is so uncertain that handbooks of literature generally
prefer to leave Bataille out; as a matter of fact, Bataille has
written texts, or even, perhaps, always one and the same text.
If the Text raises problems of classification (moreover, this is
one of its "social" functions), it is because it always implies a
certain experience of limits. Thibaudet used to speak (but in a
very restricted sense) of limit-works (such as Chateaubriand's
Life of Rance. a work which indeed seems to us to be a "text"):
the Text is what is situated at the limit of the rules of the
speech-act (rationality. readability, etc.). This notion is not
rhetorical, we do not resort to it for "heroic" postures: the Text
attempts to locate itself very specifically behind the limit of the
doxa (is not public opinion, constitutive of our democratic
societies, powerfully aided by mass communications-is not
public opinion defined by its limits, its energy of exclusion, its
censorship?); taking the word literally, we might say that the Text
i always paradoxical.
s

3. The text is approached and experienced in relation to the

sign. The work closes upon a signified. We can attribute two
modes of signification to this signified: either it is claimed to be
apparent. and the work is then the object of a science of the
letter, which is philology; or else this signified is said to be secret

B
From Work to Text 59
and final. and must be sought for, and then the work depends
upon a hermeneutics, an interpretation (Marxist, psychoanalytic,
thematic, etc.); in shan, the work itself functions as a general
sign. and it is natural that it should represent an institutional
category of the civilization ofthe Sign. The Text, on the contrary.
practices the infinite postponement of the signified. the Text is
dilatory; its field is that of the signifier; the signifier must not
be imagined as "the first pan of the meaning," its material
vestibule. but rather, on the contrary, as its aftnmath; similarly,
the signifier's infinitude does not refer to some notion of the
ineffable (of an unnamable signified) but to a notion of play; the
engendering of the perpetual signifier (in the fashion of a
perpetual calendar) in the field of the Text is not achieved by
some organic process of maturation, or a hermeneutic process
of "delving deeper," but rather by a serial movement of dislo.
cations. overlappings. variations; the logic governing the Text
is not comprehensive (trying to define what the work "means")
but metonymic; the activity of associations, contiguities, cross
references coincides with a liberation of symbolic energy (if it
failed him, man would die). The work (in the best of cases) is
moderately symbolic (its symbolics runs short, i.e., stops); the Text
is radically symbolic: a work whose integrally symbolic nature ant
conuivts, perceives, and receives is a text. The Text is thus restored
to language; like langqage, it is structured but decentered,
without closure (let us note, to answer the scornful suspicion of
"fashion" sometimes. lodged against structuralism, that the ep
istemological privilege nowadays granted to language derives
precisely from the fact that in it (language] we have discovered
a paradoxical idea of structure: a system without end or center).

4. The Text s
i plural. This does not mean only that it has
several meanings but that it fulfills the very plurality of meaning:
an irreducible (and notjuSl acceptable) plurality. The Text is not
coexistence of meaning, but passage. traversal; hence, it depends
not on an interpretation, however liberal. but on an explosion.
on dissemination. The plurality of the Text depends, as a matter

B
60 From War. to Text
of fact, not on the ambiguity of its contents, but on what we
might call the stereographic plurality of the signi6ers which
weave it (etymologically, the tex.t is a fabric): the reader of the
Text might be compared to an idle subject (who has relaxed his
image-repenoire): this fairly empty subject strolls (this has
happened to the author of these lines, and it is for this reason
that he has come to an intenu: awareness of the Text) along a
hillside at the bottom of which Hows a wadi (I use the word to
attest to a certain alienation); what he perceives is multiple,
irreducible, issuing from heterogeneous, detached substances
and levels: lights, colors. vegetation, heat. air. tenuous explosions
of sound. tiny cries of birds, children's voices from the other
side of the valley, paths, gestures, garments of inhabitants close
by or very far away; all these incidents are half identifiable: they
issue from known codes, but their combinative operation is
unique, it grounds the stroll in a difference which cannot be
repeated except as difference. This is what happens in the Text:
it can be Text only in its difference (which does not mean its
individuality); its reading is semelfactive (which renders any
inductive-deductive science of texts illusory: no "grammar" of
the text) and yet entirely woven of quotations, references,
echoes: cultural languages (what language is not cultural?),
antecedent or contemporary, which traverse it through and
through, in a vast stereophony. The intenextuality in which any
text is apprehended, since it is itself the intertext of another
text, cannot be identified with some origin of the text: to seek
out the "sources," the "influences" of a work is to satisfy the
myth of filiation; the quotations a text is made of are anonymous,
irrecoverable, and yet alrtady read: they are quotations without
quotation marks. The work disturbs no monistic philosophy
(there are antagonistic ones, as we know); for such a philosophy,
plurality is Evil. Hence, confronting the work, the Text might
indeed take for its motto the words of the man possessed by
devils: "My name is legion, for we are many" (Mark 5:9). The
plural or demonic texture which sets the Text in opposition to
the work may involve profound modifications of reading. pre-

B
From W01"A to Text
dsely where monologism seems to be the law: certain "texts" of
Scripture, traditionally adopted by theological (historical or
anagogical) monism, may lend themselves to a diffraction of
meanings (i.e., finally, to a materialist ading), while the Marxist
interpretation of the work, hitherto resolutely monistic, may
become more materialist by pluralizing itself (if, of course,
Marxist "institutions" permit this).

5. The work is caught up in a process of filiation. What is

postulated a a lkttnnination of the world (of the race, then of
History) over the work, a c01ISecution of works among themselves,
and an aPfrropri4titm of the work to its author. The author is
reputed to be the father and the owner of his work; literary
science thus teaches us to respect the manuscript and the author's
declared intentions. and society postulates a legality of the
author's relation to his work (this is the "author's rights," actually
a recent affair, not legalized in France until the time of the
Revolution). The Text, on the other hand, is read without the
Father's inscription. The metaphor of the Text is here again
detached from the metaphor of the work; the lauer refers to
the image of an O1"ganism which grows by vital expansion. by
"development" (a significandy ambiguous word: biological and
rhetorical); the metaphor of the Text is that of the nttUJorA; if
the Text expands, it is by the effect of a combinative operation,
of a systematics (an image, moreover, close to the views of
contemporary biology concerning the living being); no vital
"respect" is therefore due to the Text: it can be brolen (moreover,
this is what the Middle Ages did with two nonetheless authori
tarian texts: Scripture and Aristode); the Text can be read
without its father's guarantee; the restoration of the intertext
paradoxically abolishes inheritance. It is not that the Author
cannot "turn" in the Text, in his text, but he does so, one
might say, as a guest; if he is a novelist, he inscribes himself
the as one of his characters. drawn as a figure in the carpet;
his inscription is no longer privileged, paternal. alethic. but
ludic: he becomes. one can say. a paper author; his life is no

ateria nen al.llorskVIT

6. From Work to Text
longer the origin of his fables, but a fable concurrent with his
life; there is a reversion of the work upon life (and no longer
the contrary); the work of Proust and Genet permits us to read
their lives as a text: the word bio-graph] regains a strong,
etymological meaning; and thereby the sincerity of the spee<:h
act, a veritable "cross" ofliterary ethics, becomes a false problem:
the I that writes the text is never anything but a paper I.

6. The work is ordinarily the object of consumption; I intend

no demagoguery by referring to what is called a consumer
culture, but we must recognize that today it is the work's "quality"
(which ultimately implies an appreciation of "taste") and not the
actual operation of reading which can make differences between
books: "cultivated" reading is not structurally different from
reading on trains. The Text (if only by its frequent "unreada
bility") decants the work (if it permits it at all) from its con
sumption and recuperates it as play, task, production, practice.
This means that the Text requires an attempt to abolish (or at
least to diminish) the distance between writing and reading. not
by intensifying the reader's projection into the work. but by
linking the twO together into one and the same signifying
practice. The distance that separates reading from writing is
historical. In the period of strongest social division (before the
instauration of democratic cultures), reading and writing were
equally class privileges: Rhetoric. the great literary code of that
time, taught writing (even if what was ordinarily produced were
discourses. not texts); it is significant that the advent of democ
racy reversed the watchword: the (secondary) school prides itself
on teaching reading and no longer writing. In fact, reading. in
the sense of consuming. is not playing with the text. "Playing"
must be taken here in all the polysemy of the term: the text
itself "plays" (like a door that "plays" back and forth on its
hinges; like a fishing rod in which there s
i some "play"); and
the reader plays twice over: he plays at the Text (ludic meaning).
he seeks a practice which reproduces it; but, so that this practice
is not reduced to a passive, interior mimesis (the Text being

C JPYnghied mater1
From WorA: to Text
precisely what resists this reduction), he plays the Text; we must
not forget that play is also a musical term; the history of music
(as practice, not as "art") is, moreover, quite parallel to that of
the Text; there was a time when, active amateurs being numer
ous (at least within a certain class), "to play" and "to listen"
constituted a virtually undifferentiated activity; then two roles
successively appeared: fim of all, that of the interprettr, to which
the bourgeois public (though it could still play a little itself: this
is the entire history of the piano) delegated its playing; then
that of the (passive) amateur who listens to music without being
able to play it (the piano has effectively been replaced by the
record); we know that today post-serial music has disrupted the
role of the "interpreter," who is asked to be in a sense the co
author of the score which he completes rather than "expresses."
The Text is a little like a score of this new kind: it solicits from
the reader a practical collaboration. A great novation this. for
who executes the work? (Mallarme raised this question: he wanted
the audience to produce the book.) Today only the critic executes
the work (pun intended). The reduction of reading to con
sumption is obviously responsible for the "boredom" many fed
in the presence of the modern ("unreadable") text, the avant
garde film or painting: to be bored means one cannot produce
the text, play it, release it, molce t
i go.

7. This suggests one final approach to the Text: that of

pleasure. I do not know if a hedonist aesthetic ever existed
(eudaemonist philosophies are certainly rare). Of course, a
pleasure of the work (of certain works) exists; I can enjoy
reading and rereading Proust, Flaubert, Balzac. and even-why
not?-Alexandre Dumas; but this pleasure, however intense.
and even when it is released from any prejudice, remains partly
(unless there has been an exceptional critical effort) a pleasure
of consumption: for. if I can read these authors, I also know
that I cannot 'rewrite them (that one cannot, today, write "like
that" ); and this rather depressing knowledge suffices to separate
me from the production of these works, at the very moment

C ;.pvnghted malaria
 From Work to Tat
when their distancing founds my modernity (to be modern-is
this not really to know that one cannot begin again?). The Text
is linked to delectation. i.e., to pleasure without separation.
Order of the signifier, the Text participates in its way in a social
utopia; before History (supposing that History does not choose
barbarism), the Text fulfills if not the transparency of social
relations, at least the transparency of language relations: it is
the space in which no language prevails over any other, where
the languages circulate (retaining the circular meaning of the
word),

These few propositions do not necessarily constitute the

articulation of a Theory of the Text. This is not merely the
consequence of the presenter's inadequacies (moreover, in many
points he has merely recapitulated what is being investigated
and developed around him). This is a consequence of the fact
that a Theory of the Text cannot be satisfied with a meta
linguistic exposition: the destruction of meta-language. or at
least (for it may be necessary to resort to it provisionally) calling
it into question. is part of the theory itself: discourse on the
Text should itself be only text, research, textual activity, since
the Text s
i that social space which leaves no language safe,
outside, and no subject of the speech-act in a situation ofjudge,
master, analyst. confessor, decoder: the theory of the Text can
coincide only with a practice of writing.

RtlI1U d'eslhitiqw. 1971

GJPYnghtcd maknal
 Mythology Today

Some fifteen years ago, a certain idea of contemporary myth

was proposed. This idea, which on its first appearance was
hardly developed at all (the word retained an openly metaphoric
value), nonetheless included veral theoretical articulations. I .
Myth. dose to what Durkheimian sociology calls a "collective
presentation." can be read in anonymous statements of the
press, advertising. mass consumption; it is a social determinate.
a "reflection," 2. This reflection, however, in accord with Marx's
famous dictum, is invnud: myth consists in turning culture into
nature, or at least turning tbe social, the cultural. the ideological,
the historical inlo the "natural": what s
i merely a product of
class division and its moral. cultural. aesthetic consequences is
preM!nted (stated) as a natural consequence; the quite contingent
grounds of the statement become, under the effect of mythic
inversion, Common Sense, Right Reason, the Norm, Public
Opinion, in a word, the Endoxa (the secular figure of the Origin).
3. Contemporary myth is discontinuous: it is no longer stated
in extended, constituted narratives, but only in "discourse"; at
most, it is a phraseology, a corpus of phrases (of stereotypes);
myth disappears, but the mythic remains, all the more insidious.
4. As speech (this was, after all, the meaning of mutho.J),
contemporary myth ssues
i from a semiology which permits the
"correction" of mythic inversion by decomposing the message
into two semantic systems: a connoted system whose signified is
ideological (and consequently "straight," "noninverted," or, to
be dearer, even by speaking a moral jargon, cynical), and a
denoted system (the apparent literalness of the image. of the
object, of the sentence), whose function is to "naturalize" the
class proposition by giving it the guarantee of the most "inno

GJPYnghtcd maknal
66 From Work to Text
cent" of natures: that of language (age-old, maternal, academic.
etc.).
This was how myth today appeared, or at least appeared to
me. Has anything changed? Not French society, at least on this
level, for mythic history is on a different time scale from political
history: nOT the myths, nor even the analysis; there is still a
great deal of the mythic in our society: equally anonymous,
slippery, fragmented, garrulous, available both to an ideological
criticism and to a semiological dismanding. No. what has changed
in the last fifteen years is the scienu of reading, under whose
scrutiny myth, like an animal long since captured and observed.
nonetheless becomes a difftTent object.
A science of the signifier (even if it is still being elaborated)
has in fact taken its place in the work of the period; its goal is
not so much the analysis of the sign as its dislocation. with
regard to myth, and though this is still a task which remains to
be accomplished, the new semiology-or the new mythology
can no longer (or will no longer be able to) separate so easily
the signifier from the signified, the ideological from the phrase
ological. Not that this distinction is false or ineffectual, but it
has become mythic itself: any student can denounce the bour
geois or petit-bourgeois character of a form (of life, of thought,
of consumption); in other words, a mythological endoxa has been
created: demystification (or demythification) has itself become
a discourse. a corpus of phrases, a catechistic statement; con
fronting which a science of the signifier can only be displaced
and stop (provisionally) farther on: no longer at the (analytic)
dissociation of the sign. but at its vacillation: it is no longer the
myths which must be unmasked (the endoxa now undertakes
that), but the sign itself which must be perturbed: not to reveal
the (latent) meaning of a statement, of a feature, of a narrative,
but to fissure the very representation of meaning; not to change
or to purify symbols. but to contest the symbolic itself. What is
happening to (mythological) semiology is a little like what
happened to psychoanalysis: it began. necessarily. by establishing
lists of symbols (a loosened tooth - the castrated subject, etc.),

GJPYnghtcd maknal
Mythology Today
but today, much more than interrogating this lexicon which,
without being false, is no longer of much interest (except to
amateurs of the psychoanalytic vulgate), it examines the very
dialectics of the signifier; semiology began in the same way by
establishing a mythological lexicon, but the task facing it today
is rather of a syntactical order (which aniculations, which
displacements constitute the mythic fabric of a mass-consump
tion society?); initially, we sought the destruction of the (ideo
logical) signified; now we seek the destTUctio of the sign:
"mythoclasm" is succeeded by a "semioclasm" that s
i much
broader and raised to a higher level. The historical field is
thereby extended: it is no longer French society, but far beyond
it, historically and geographically, the whole of Western (Greco
Judeo-Islamo-Christian) civilization, unified in one and the same
theology (essence. monotheism) and identified by the system of
meaning it practices, from Plato to Fronce-DimanclM.
The science of the signifier contributes a second correction
(or a second extension) to contemporary mythology. The world.
taken obliquely by language. is written, through and through;
signs, constantly deferring their foundations. transforming their
signifieds into new signifiers, quoting each other to infinity.
nowhere come to a halt: writing is generalized. If society'S
alienation still compels us to demystify languages (and notably
that of the myths), the means of this combat is not-is no
longer-a critical decipherment. it is evaluation. Faced with the
world's writing systems, the tangle of various discourses (didactic,
aesthetic. propagandistic, political, etc.), we must determine
levels of reification, degrees of phraseological density. Shall we
succeed in specifying a notion which seems to me essential: that
of a language's comjHUtmu? Languages are more or less tkns;
some the most social, the most mythical-present an unshak
able homogeneity (there is a power of meaning, there is a war
of meanings): woven of habits. of repetitions, of stereotypes, of
obligatory fragments and key words, each one constitutes an
idicuel (a notion which twenty years ago I designated as writing);
today, more than myths, it is idiolects which we must distinguish,

B
68 From Work to Text
describe; mythologies are succeeded by a more formal, and
thereby, I bdieve. more netrating. idiolectology, whose op
erative concepts are no longer sign. signifier, signified, and
connotation, but citation, reference, stereotype. Thus. the den
languages (such as mythic discourse) can be apprehended in
the cross fire of the trans-writing whose still literary "text,"
antidote to myth. would occupy the pole. or rather the region
airy, light, open, spaced, decentered, noble, free-where writing
deploys iudf against the idiolect. i.e., at its limit. and combats
it there. Myth in fact must be included in a general theory of
the language of writing. of the signifier, and this theory.
supponed by the formulations of ethnology, psychoanalysis.
semiology. and ideological analysis, must extend its object to
take in the sentence, or better still, to take in sentences (the plural
of the sentence); by which I mean that the mythic is present
wherever sentences are turned, where stories art told (in every sense
of these expressions): from interior monologue to conversation,
from the newspaper article to the political speech. from the
novel (if there are any left) to the advertising image-all
utterances that can be included in the Lacanian concept of the
image-repertoirt.
This is no more than a program, perhaps in fact no more than
a "desire." Yet I believe that, even if the new semiology-mainly
concerned, recently, with the literary text-is no longer applied
to myths of our time since the last text of Mythologies. in which I
sketched an initial semiotic approach to social speech, it is at least
conscious of its task: no longer merely to reverst (or to correct) the
mythic message, putting it right side up, with denotation at the
bottom and connotation at the top. nature on the surface and class
interest deep down, but to change the object itself, to engender a
new object, point of departure for a new science; to shift-mak
ing due allowance. of course. for differences in importance. and
according to A1thusser's scheme-from Feuerbach to Marx, from
the young Marx to the great Marx.

Esprit. 1971

B
 Research: The Young

Here is a special issue of CommunicatiON: it has not been devised

(0 explore a body of knowledge or to illustrate a theme; its
unity, at least its original unity. is not in its object but in the
group of its authors: th are all students. rently committed
to research; deliberately collected here is the first work. of young
researchers sufficiently free to have determined their research
project themselves and yet still subject to an institution, that of
the third-cyde doctorate. What I shall discuss here is therefore
mainly the research itself. or at least a certain research, research
still linked to the traditional realm of arts and letters. It is solely
with that research that 1 sha11 be concerned.

On the threshold of his work. the student ex.periences a series

of divisions. As a young subject he belongs to an economic class
defined by its unproductiveness: he is neither an owner nor a
producer; he is outside of exchange. and even. one might 5ay.
outside of exploitation: socially. he is excluded from any nom
ination. As an inUlkctual subject. he s
i brought into the hierarchy
of tasks, he is supposed to participate in a speculative luxury he
nonetheless cannot enjoy. for he has not yet mastered it. i.e.
the availability of communication. As a researching subject. he is
dedicated to the separation of discourses: on one side the
discourse of scienti6city (discourse of the Law). and on the
other, the discourse of desire, or writing.

The task (of research) must be perceived in desire. If this

perception does not occur. the work is morose. functional,
alienated. impelled solely by the necessity of passing an exami
nation, of obtaining a diploma. of insuring a career promotion.

6g

B
70 From Work to Tat
For desire to be insinuated into my work, that work must be
marukd of me not by a collectivity seeking to guarantee my
labor aop to gain a return on the loans it grants me, but by a
living collection of readers expressing the desire of the Other
(and not the control of the Law). Now, in our society, in our
institutions. what is asked of the student, of the young re
searcher, of the intellectual worker. is never his desire; he is
not asked to write, he is asked to speak, to "report" (with a view
to regular verifications).
Here the intention has been that the work of research befrom
ts
i inupticm the object of a strong demand, formulated outside
the institution-a demand which can only be the demand for
writing. or course, only a fragment of utopia can be represented
in this issue. for we realize that society is not ready to concede
this happiness broadly. institutionally. to the student. and sin
gularly to the student "of letters": that it is not his competence
or his future function that is needed. but his present passion.

It is perhaps time to dispose of a certain fiction: the one

maintaining that research is reponed but not wriuen: here the
researcher is essentially a prospector of raw materials, and it is
on this level that his problems are raised; once he has commu
nicated his " resuhs," everything is solved; "formulation" is
nothing more than a vague final operation, rapidly performed
according to a few techniques of "expression" learned in sec
ondary school and whose only constraint is submission to the
code of the genre ("darity," suppression of images, respect for
the laws of argument). Yet it is unlikely that. even in the area
of simple tasks of "expression," the student of the social sciences
is adequately prepared. And when the object of research is the
Text (a word to which we shall return), the investigator is faced
with a dilemma-a formidable one: either to speak of the Text
according to the conventional code of inauthentic writing or
icrivance. Le., to remain prisoner of the " image-repertoire" of a
scholar who seeks to be or, worse still, believes himself exterior
to the object of his study and claims, in all innocence, in all

B
ReJearch: The Young 71
assurance, to put his own language in a position of extraterri
loriality; or else to enter the play of the signifier, the infinity of
the speechact, in short "to write" (which does not simply mean
"to write well"), to extract the " ego" from its imaginary hull,
from that scientific code which protects but also deceives, in a
word to cast the subject acTOSS the blank page, not to "express"
it (nothing to do with "subjectivity") but to disperse it: to overflow
the regular discourse of research. It is obviously this overflow,
however slight, which we are allowing, in mis issue of Commu
nications. to come on stage: an overflow variable according to
the authors: we have not sought to reward any one kind of
writing; the important thing is that at one level or another of
his work (knowledge, method, speechact) the researcher decides
not to be imposed upon by the Law of scientific discourse (the
discourse of science is not necessarily science: by contesting the
scholar's discourse. writing in no way does away wim the rules
of scientific work).

Research is done in order to be published, though it is rarely

published, especially in its early phases, which are not necessarily
less imponant than its conclusion: the success of a piece of
research--especially textual research-does not abide in its
"result," a fallacious notion, but in the reflexive nature of its
speech-act; at every moment of its trajectory, a piece of research
can turn language back upon itself and thereby overcome the
scholar's bad faith: in a word, displace author and reader.
However, as we know, the work of students is rarely published:
the third-cycle thesis is in fact a repressed discourse. By pub
lishing fragments of initial research, we hope to combat that
repression and to release not only me author of the article but
his reader. for the reader (and specifically the magazine reader)
is also caught up in the division of specialized languages.
Research must no longer be that parsimonious task performed
either in the researcher's "consciousness" (a painful, autistic
form of monologue) or in the impoverished oscillation which
makes the "director" of a research project its only reader.

B
7' From Work to Text
Research must join the anonymous circulation of language. the
dispersion of the Text.

These studies are research in that they seek to renew reading

(the reading of older texts). To renew reading: nOI to substitute
new scientific rules for the old constraints of interpretation, but
rather to imagine that afree reading might become, finally, the
norm of "literary studies_" The freedom in question is of course
not just any freedom (freedom is in contradiction with 'Just
any"): the claim of an innocent freedom revives a memorized.
stereoty culture (the spontaneous is the immnJ.iau field of
the already said. the dJja-dU) : this would inevitably be the return
of the signified. The freedom "staged" in this issue is the
freedom of the signifier: the return of words. of word games
and puns, of proper names, of citations. of etymologies, of
reftexivities of discourse. of typogTaphies, of combinative op
erations, of rejections of languages. This freedom must be a
virtuosity: the kind which ultimately permits us to read within
the support text. however ancient, the motto of all writing: it
circuJ.au.s.

Interdisciplinary studies, of which we hear so much. do not

merely confront already constituted disciplines (none of which.
as a matter of fact. consents to leave off). In order to do
interdisciplinary work. it is not enough to take a " subject" (a
theme) and to arrange two or three sciences around it. Inter
disciplinary study consists in creating a new object. which belongs
to no one. The Text is, I believe, one such object.

The semiotic studies undertaken in France the last fifteen

years have in fact stressed a new notion which must gradually
be substituted for the nOlion of the work: this is the Text. The
Text-which cannot be allotted to the traditional realm of
"Literature"-was theoretically founded by a certain number of
initiatory writings: first of all. the Text was theory. The studies
(one should like to be able to say. the testimonies) collected here

C JPYnghied mater1
RestQrch; The Young
cospond to that moment when theory must be fragmented
for the sake of particular investigations. What is put forward
here is the passage from theory to research: all these articles
deal with a particular. contingent text belonging to historical
culture. but all are also the product of that preliminary theory
or of the methods of analysis which have prepared it.

With regard to "letters." reflection on research leads to the

Text (or, at least. let us admit today, research is free to lead to
it): hence the Text, equally with research, is the object of this

Issue.
The Text: let us make no mistake about either this singular
or this capital letter; when we say the Text, it is not in order to
divinize it, to make it the deity of a new mystique, but to denote
a mass, a field requiring a partitive and not a numerative
expression: all that can be said of a work is that there is Text
in it. In other words. by passing from text to the Text, we must
change numeration: on the one side, the Text is not a comput
able object, it is a methodological field in which are pursued.
according to a movement more "Einsteinian" than "Newtonian."
the statement and the speech-act, the matter commented on
[the commenu1 and the matter commenting [the commentantJ; on
the other side. there is no necessity that the Text be exclusively
modern: there can be Text in ancient works; and it is precisely
the presence of this unquantifiable germ that makes it necessary
to disturb, to transcend the old divisions of Literary History;
one of the immediate, obvious tasks of new research is to
proceed to such accounts of writing. to explore what Text there
can be in Diderot. in Chateaubriand. in Flaubert, in Gide: this
is what many of the authors gathered here are doing; as one of
them says, speaking implicitly in the name of several of his
comrades: "Perhaps our work merely consisLS in identifying
fragments of writing caught up in a discourse .still guarantd
by the Father." No beuer definition of what. in previous work.
is Literature, and what is Text. In other words: how can this
past work still be read? These young researchers must be credited

GJPYnghtcd maknal
74 From Work to Text
with raising their activity to the level of a critical task: the
present evaluation of a past culture.

All these studies form a collective gesture: it is the very

territory of the Text which is gTadually being drawn, colored
in. Let us follow brieRy, from anicle to article, the collective
hand which, far from writing the definition of the Text (there
is no such thing: the Text is not a concept), describes (de-scribes)
the practice of writing.
First of all. this, which is necessary in order to understand
and to accept the range of articles collected here: the Text
frustrates any cultural typology: to show the limitless character
of a work is to make it a text; even if reRection on the Text
begins with literature (i.e., with an object constituted by the
institution), the Text does not necessarily stop there; wherever
an activity of signifying is staged according to the rules of
combination, transformation, and displacement, there is Text:
in written productions, of course, but also in the play of images.
of signs, of objects: in films, in comic strips, in ritual objects.
Then this: as deployment of the signifier, the Text often
dramatically argues with the signified which tends to recrudesce
within it: if it succumbs to this recrudescence, if the signified
triumphs. the text ceases to be Text, the stereotype within it
becomes "truth" instead of being the ludic object of a second
combinative operation. Hence, it is logical that the Text engage
its operator in what we may call a drama of writing (which we
shall see analyzed here apropos of Flaubert), or its reader in
preliminary critical evaluation.
However, the main and so to speak massive approach that
can be made to the Text consists in exploring all its manifest
signifiers: structures which the linguistics of discourse can
articulate: phonic configurations (puns. proper names), typo
graphic arrangements, polysemies, enjambments. blanks, col
iages, associations, everything which calls into question the book's
substance. will be recognized here, proposed by various authors,
from Flaubert to Claude Simon.

GJPYnghtcd maknal
Rtstarch: The Young 75
Last, the Text is above all (or after all) that long operation
through which an author (a discoursing author) discovers (or
makes the reader discover) the irrtparability of his speech and
manages to substitute it spealu for I spealt.. To know the image
repertoire of expression is to empty it out, since the image
repertoire is lack of knowledge: several studies. here. attempt
to evaluate the image-repertoire of writing (apropos of Cha
teaubriand, of Gide, of Michel Leiris) or the image-repertoire
of the researcher himself (apropos of a research on cinemato
graphic suspense).
It must not be supposed that these various "prospects" help
tneirck the Text; rather, it is to expand the Text that the entire
issue functions. Hence. we must resist trying to organize, to
program these studies. whose writing remains very diverse (I
have been reluctant to acknowledge the necessity of "introduc
ing" this wue of Communications. for thereby I risk appearing
to give it a unity in which the contributors may not recognize
themselves, and lending each of them a voice which is perhaps
not entirely his own: any presentation. by its intention of
synthesis, is a kind of concession to discourse). Ideally. through
out the issue. independent of what precedes and of what follows.
the research of these young scholars should appear both as the
revelation of certain structures of speech-acts (even if they are
analyzed in the simple language of a report) and the critique
(the auto-critique) of any speech-act: moreover it is just when
research manages to link its object to its discourse and to
dispossess our knowledge by the light it casts on objects not so

much unknown as unexpected-it is at just this moment that

research becomes a true interlocution. a task in behalf of others.
in a word: a social production.

C JPYnghied mater1
 The Rustle of language

Speech is irreversible; that is its fatality. What has been said

cannot be unsaid, except by adding to it: to correct, here, is, oddly
enough, to continue. In speaking, I can never erase, annul; all
I can do is say "I am erasing. annulling. correcting," in shon.
speak some more. This very singular annu)ation-by-addition I
shaH call "stammering." Stammering is a message spoiled twice
over: it is difficult to understand. but with an effort it can be
understood all the same; it is really neither in language nor
outside it: it is a noise of language comparable to the knocks by
which a motor lelS it be known that it s i not worlUng properly;
such is precisely the meaning of the misfrre, the auditory sign of
a failure which appears in the functioning of the object. Stam
mering (of the motor or of the subject) is, in shon, a fear: I am
afraid the motor is going to stop.

The death of the machine: it can be distressing to man, if he

describes it like that of a beast (see lola's novel). In short,
however unsympathetic the machine may be (because it consti
tutes, in the figure of the robot, the most serious of threats: the
loss ofthe body), it still contains the possibility of a euphoric theme:
its good furutioning; we dread the machine when it works by
itself, we delight in it when it works well. Now, just as the
dysfunctions of language are in a sense summarized in an
auditory sign, stammering, similarly the good functioning of
the machine is displayed in a musical being: the rwtle.

The rusde is the noise of what s i working well. From which

follows this paradox: the rusde denotes a limit-noise, an impos
sible noise, the noise of what, functioning to rfection, has no

GJPYnghtcd maknal
Tiu Rustle of Languagt 77
noise; to rustle is to make audible the very evaporation of noise:
the tenuous, the blurred. the tTemulous are received as the
signs of an auditory annulation. .
Thus, it is happy machines which rustle. When the erotic
machine, so often imagined and described by Sade, an "intel
lectual" agglomerate of bodies whose amorous sites are carefully
adjusted to each other-when this machine starts up. by the
convulsive movements of the participants, it trembles and rus
tles: in short, it 'WOT",u. and it works well. Elsewhere, when today's
Japanese surrender themselves en masse. in huge halls. to the
slot-machine game called pachinko. these halls are filled with
the enormous rustle of the little balls. and this rustle signifies
that something, collectively, is working: the pleasure (enigmatic
for other reasons) of playing. of moving the body with exacti
tude. For the rustle (we see this from the Sadean example and
from the Japanese example) implies a community of bodies: in
the sounds of the pleasure which is "working." no voice is raised,
guides, or swerves, no voice is constituted; the rustle is the very
sound of plural delectation-plural but never massive (the mass,
quite the contrary, has a single voice, and terribly loud).

And language can language rustle? Speech remains. it seems,

condemned to stammering; writing, to silence and to the dis
tinction of signs: in any case, there always remains too much
meaning for language to fulfill a delectation appropriate to its
substance. But what s
i impossible is not inconceivable: the rustle
of language forms a utopia. Which utopia? That of a music of
meaning; in its utopic state. language would be enlarged. I
should even .say denaluud to the point of forming a vast auditory
fabric in which the semantic apparatus would be made unreal;
the phonic. metric. vocal signifier would be deployed in all its
sumptuosity, without a sign ever becoming detached from it
(ever naturalizing this pure layer of delectation), but also-and
this is what is difficult-without meaning being brutally dis
missed. dogmatically foreclosed, in shon castrated. Rustling,
entrusted to the signifier by an unprecedented movement
 From Work to Text
unknown to our rational discourses, language would not thereby
abandon a horizon of meaning: meaning, undivided, impene
trable, unnamable, would however be posited in the distance
like a mirage, making the vocal exercise into a double landscape,
furnished with a "background"; but instead of the music of the
phonemes being the "background" of our messages (as happens
in our poetry), meaning would now be the vanishing point of
delectation. And just as, when attributed to the machine, the
rustle is only the noise of an absence of noise, in the same way,
shifted to language, it would be that meaning which reveals an
exemption of meaning or-the same thing-that non-meaning
which produces in the distance a meaning henceforth liberated
from all the aggressions of which the sign, formed in the "sad
and fierce history of men," is the Pandora's box.
This is a utopia, no doubt about it; but utopia is often what
guides the investigations of the avant-garde. So there exists here
and there, at moments, what we might call certain experiments
in rustling: like certain productions of post-serial music (it is
quite significant that this music grants an extreme importance
to the voice: it is the voice it works with, seeking to denature
the meaning in it, but not the auditory volume), certain radio
phonic researches; and like the latest texts by Pierre Guyotat or
Philippe Sollers.

Moreover, we ourselves can undertake this research around

the rustle, and in life, in the adventures of life; in what life
affords us in an utterly impromptu manner. The other evening,
watching Antonioni's film on China, I suddenly experienced, at
the end of a sequence, the rustle of language: in a village street,
some children, leaning against a wall, reading aloud, each one
a different rook to himself but all together; that-that rustled
in the right way, like a machine that works well; the meaning
was doubly impenetrable to me, by my not knowing Chinese
and by the blurring of these simultaneous readings; but J was
hearing, in a kind of hallucinated perception (so intensely was
it receiving all the subtlety of the scene), I was hearing the
The Rustk of Language 79
music, the breath, the tension. the application, in shon some
thing like a goal. Is that all it takes-just speak all at the same
time in order to make language rustle, in the rare fashion,
stamped with delectation, that I have been trying to describe?
No, of cour not; the auditory scene requires an erotics (in the
broadest sense of the term), the elan, or the discovery. or the
simple accompaniment of an emotion: precisely what was con
tributed by the countenances of the Chinese children.
I imagine myself today something like the ancient Greek as

Hegel describes him: he interrogated, Hegel says, passionately,

uninterruptedly, the rustle of branches, of springs, of winds, in
shon, the shudder of Nature, in order to perceive in it the
design of an intelligence. And I-it is the shudder of meaning
I interrogate, listening to the rustle of language, that language
which for me, modem man. is my Nature.

Vm um t5thetiquL sans mtnu.'f!$ (U.G.E.), 1975

 3
LANGUAGES AND SlYLE

Malena m dlraitl I


 Rhetorical Analysis

Literature presents itself lO us as an institution and as a work. As

an institution. it colleclS all usages and all practices which govern
the circuit of the thing wrinen in a given society: the writer's
social status and ideology, modes of circulation, conditions of
consumption, sanctions of criticism. As a work, it is essentially
constituted by a verbal, written message of a certain type. It is
the work-as-object that I wish [0 deal with, suggesting that we
concern ourselves with a still Iittle-explored field (though the
word s
i very old), that of rlutoric.
The literary work includes elements which are not special to
literature; I shall cite at least one of these, because the devel
opment of mass communications permits its incontestable rec
ognition today in films, in comic strips, and perhaps in the news
item [the fait-divers], i.e., elsewhere than in the novel: this is
narrative, story, argument, what Souriau has called. apropos of

film. dugesis. There exists a diegetic form common to different

arts. a form we are beginning to analyze today according to new
methods inspired by Propp. However, confronting the element
of fabulation it shares with other creations, literature possesses
one element which defines it specifically: its language; this
specific element the Russian formalist school has already sought
to isolate and to treat under the name of Literaturnost, "literar
iness"; Jakobson calls it potties; poetics is the analysis which
permits answering this question: What is it that makes a verbal
message a work of art? It s
i this specific element which. for my
part, I shall call rhetoric, so as to avoid any restriction of poetics
to poetry and in order to mark our concern with a general level
of language common to all genres, prose and verse alike. My

83
 Languages and Style
question is whether a confrontation of society and rhetoric is
possible, and under what conditions.
For centuries-from antiquity to the nineteenth century
rhetoric has received a definition which is at once functional
and technical: it is an art, i.e., a set of constraints which permit
either persuasion or, subsequently, expressiveness. This de
e1ared goal evidently makes rhetoric into a social institution,
and, paradoxically, the link which unites the forms of language
to societies is much more immediate than the strictly ideological
relation; in ancient Greece, rhetoric is born very specifically in
the property trials which followed the exactions of the Tyrants
in fifth..century Sicily; in bourgeois society, the art of speaking
according to certain rules is both a sign of social power and an
instrument of that power; it is not insignificant that the class
which concludes secondary studies of the young bourgeois in
France s
i called the clasJt de rhitorique. However, it is not this
immediate (and, indeed, quickly exhausted) relation that we
shall linger over, for, as we know, if social need engenders
certain functons, these functions, once they are set in operation,
or, as we say, once they are tUtn'mined, acquire an unforeseen
autonomy and acquire new significations. For the functional
definition of rhetoric, I shall therefore substitute an immanent,
structural definition, or to be still more specific. an informational
definition.
We know that every message (and the literary work is one of
them) includes at least one level of expression, or level of
signifiers. and one level of content, or level of signifieds; the
junction of these two levels forms the sign (or group of signs).
However, a message constituted according to this elementary
order can, by an operation of separation or amplification,
become the simple expressive level of a second message, which
is of the extensive variety; in short, the sign of the first message
becomes the signifier of the second. We are then in the presence
of two semiotic systems imbricated within each other in a regular
fashion. Hjelmslev has called the second system thus constituted
connotative semiotics (in opposition to the meta-language, in which

C JPYnghied mater1
Rhetorical A.nalysis
the sign of the first message becomes the signified and not the
signifier of the second message). Now, as language, literature
is. from all evidence, a connotative semiotics; in a literary text,
a firSl system of signification, which is language (French, for
instance), serves as a simple signifier in a second message, whose
signified is different fr.om the signifieds of the language; if I
read: Foius ovane" us commodiUstk La canvtTSlllion (Bring forward
the comforts of conversation). I perceive a dnwtaud message
which is the order to move the armchairs closer. but I also
perceive a cannotaud message whose signified here s
i "preciosity."
In informational terms, we shall therefore define literature as
a double system, denoted-connoted; in this double system, the
manifest and specific level, which is that of the signifiers of the
second system. will constitute Rhetoric; the rhetorical signifiers
will the connotators.
Defined in informational terms, the literary message can and
must subjected to a systematic exploration, without which we
can never confront it with the History which produces it, since
the historical ing of this message is not only what it says but
also the way in which it is fabricated. Of course. the linguistics
of connotation-which we cannot confuse with the old stylistics.
for the latter, studying means of expression, remained on the
level of speech [parou], while the former, studying codes, takes
its place on the level of the language [Langut}-is not yet
constituted; but certain indications of contemporary linguists
permit us to propose at least two directions to rhetorical analysis.
The first has been sketched by Jakobson, who distinguishes
six factors in every message: a sender, a receiver, a context or
referent, a contact. a code, and finally the message itself; to
each of these factors corresponds a function of language; every
discourse mixes most of these functions. but it receives its marA
from the dominance of one function or another over the rest;
for instance, if the emphasis is put on the person emitting the
message. the expressive or emotive function dominates; if it is
put on the receiver, it is the connotative (exhortative or suppli
cative) function which prevails; if it is the referent which receives

C JPYnghied mater1
86 Languages and Styu
the emphasis, the discourse is denotative (as is the case here);
if it is the contact (between sender and receiver), the phatic
function refers [0 all the signs intended to maintain communi
cation between the interlocutors; the meta-linguistic function,
or function of elucidation, accentuates rIXourse to the code;
last, when it is the message itself, its configuration, the palpable
aspect of its signs which are emphasized, the discourse is poetic,
in the broad sense of the term: this is obviously the case of
literature; we can say that literature (work or text) is specifically
a message which puts the emphasis on itself. This definition no
doubt permits a better understanding of how it comes about
that the communicative function does not exhaust the literary
work, but that the latter, resisting purely functional definitions,
always presents itself in a certain fashion as a tautology, since
the message's intra-mundane functions remain ultimately subject
to its structural function. However, the coherence and decla
ration of the poetic function may vary with History; and further,
synchronically, this same function may be "devoured" by other
functions, a phenomenon which in a sense diminishes the work's
coefficient of literary specificity. Jakobson's definition therefore
involves a sociological perspective, since it permits us to evaluate
both the process of literary language and its situation in relation
to non-literary languages.
Another exploration of the literary message is possible, this
time of a distributional type. We know that a whole portion of
linguistics is concerned today with defining words less by their
meaning than by the syntagmatic associations in which they can
take their place; roughly speaking, words associate among
themselves according to a certain scale of probability: dog is
readily associated with barh but rarely with mew, though syn
tactically there is nothing that forbids the association of a verb
and a subject; this syntagmatic "filling" of the sign is occasionally
called catalysis. Now, catalysis has a dose relation with the special
nature of literary language; within certain limits, which are
precisely those to be studied. the more aberrant the catalysis,
the more patent literature becomes. Of course, if we abide by

C JPYnghied mater1
Rhetorical Analysis
the literal units, literature is not at. all incompatible with a normal
catalysis; in the sky is blue as an orange, no literal association is
deviant; but if we refer to a higher level of units, which is
precisely that of connotators, we recognize the catalytic distur
bance without difficulty, for it is statistically aberrant to associate
blueness with an orange. The literary message can therefore be
defined as a divergence of association of signs (Guiraud);
operationally, for instance, confronting the normative tasks of
automatic translation, literature might be defined as the sum of
the insoluble cases presented to the machine. We can say in
another fashion that literature is essentially a costly system of
infmmation. However, if literature is uniformly luxurious, there
are several luxury economies, which can vary with periods and
societies; in our classical literature of the anti-precieux generation,
syntagmatic associations remain within normal margins on the
level of denotation, and it is explicitly the rhetorical level which
supports the high cost of the information; on the contrary, in
surrealist poetry (to take two extremes), the associations are
aberrant and the information costly on the level of the elemen
tary units themselves. We can reasonably hope, here again, that
the distributional definition of the literary message will cause
certain links to appear between each society and the economy
of information it assigns to literature.
Thus, the very form of the literary message is in a certain
relation with History and with society, but this relation is special
and does not necessarily coincide with the history and sociology
of contents. The connotators form the elements of a code, and
the validity of this code can be more or less lasting; the classical
code (in the broad sense) has lasted for centuries in the West,
since it is the same rhetoric which animates an oration by Cicero
or a sermon by Bossuet; but it is likely that this code underwent
a profound. mutation in the second half of the nineteenth
century, even if, to this very day, certain traditional writings are
subject to it. This mutation is doubtless related to the crisis of
bourgeois consciousness; the problem, however, is not to know
if the one analogically reflects the other, but if, confronting a
88 Languages and Style
certain order of phenomena, history does not somehow inter
vene to modify the rhythm of their diachrony; as a matter of
fact, as soon as we deal with forms (and this is obviously the
case with the rhetorical code), the processes of change are more
on the order of translation than of evolution: there is successive
exhaustion of the possible mutations, and history is called upon
to modify the rhythm of these mutations, not these forms
themselves; there is perhaps a certain endogenous development
of the structure of the literary message, analogous to the one
which governs changes of fashion.
There is another way of appreciating the relation between
rhetoric and society: by evaluating the degree of "frankness" of
the rhetorical code. It is certain that the literary message of the
classical period deliberately paraded its connotation, since the
figures constituted a code transmissible by apprenticeship (whence
the numerous treatises of the period), and since it was not
possible to form a recognized message except by drawing on
this code. Today, as we know, this rhetoric has exploded; but
precisely by studying its debris, its substitutes, or its lacunae, we
can doubtless account for the multiplicity of writings and
recognize, for each of them, the signification it possesses in our
society. We might thus approach quite precisely the problem of
the division between good literature and the others, whose social
importance is considerable, especially in a mass society. But
here, too, we must not look for an analogical relation between
a group of usages and its rhetoric; our task is rather to
reconstitute a gen<;ral system of sub-codes, each of which is
defined in a certain state of society by its differences, its distances,

and its identities with regard to its neighbors: elite literature

and mass culture, avant-garde and tradition, constitute, formally,
different codes simultaneously placed, according to Merleau
Ponty, in a "modulation of coexistence"; it is this set of simul
taneous codes, whose plurality has been recognized by Jakobson,
which should be studied; and since a code is itself merely a
certain way of distributing a closed collection of signs, rhetorical
analysis should derive not from sociology strictly speaking, but
Rlutorical Analysis 89

rather from that socia-logic, or sociology of forms ofclassification

already postulated by Durkheim and Mauss.
Such are, summarily and abstractly presented, the general
perspectives of rhetorical analysis. It is an analysis whose project
is not new, but to which recent developments of structural
linguistics and of information theory afford renewed possibilities
of exploration; but. above all, it requires of us a methodological
analysis that is perhaps new: for the formal nature of the object
it seeks to study (the literary message) obliges us to describe in
an immanent and exhaustive fashion the rhetorical code (or
codes) before setting this code (or these codes) in relation with
the society and the history which produce and consume them.

Goldmann colloquium. 1966

 Style and Its Image

I should like to begin with a personal consideration: for some

twenty years, my investigations have been concerned with lit
erary language, without my being altogether comfortable in the
role either of critic or of linguist. I should like to take advantage
of this ambiguous situation in order to deal with an impure
notion, one which is at once a metaphoric form and a theoretical
concept. This notion is an image. I do not believe, as a matter
of fact, that scientific work can proceed without a certain image
of its object (as we know, nothing is more resolutely metaphoric
than the language of mathematicians or geographers); nor do
I believe that the intellectual image, heir of ancient Pythagorean
cosmogonies, at once spatial, musical. and abstract, can be
divested of a theoretical value which preserves it from contin
gency without exaggeratedly deflecting it toward abstraction.
Hence, it is an image which I seek to question, or, more
specifically, a visian: How do we su style? What is the image of
style which troubles me, what is the one I desire?
Simplifying greatly (the privilege of vision), it seems to me
that style (in the current sense of the word) has always been
part of a binary system, or. if you prefer, of a mythological
paradigm of two terms; these terms have, of course, changed
names and even content, according to periods and schools. Let
us consider two of these oppositions.
The first. the oldest (it is still with us, at least quite frequently
in the teaching of literature), is that of Canlml and Form; it
derives, we know, from one of the first classifications of classical
Rhetoric, the opposition between Res and Verba: on
Res (or
demonstrative materials of discourse) depends Invmtio, or re
search into what one can say about a subject (quaestio); on Verba

go

C :lpynghled malenal
Style aM Its Image 9'

depends Eloculio (or transformation of these materials into a

verbal form); this Elocutio is, roughly, our "style." The relation
of Form and Content is phenomenological: Form is reputed to
be the appearance or garment of Content, which is iu truth or
body; the metaphors attached to Form (to style) are therefore
of a decorative order: figures, colors, nuances; or again, this
relation of Form and Content si experienced as expressive or
alethic: the writer (or the commentator) must establish a proper
relation between content (truth) and form (appearance), between
message (as content) and its medium (style); between these two
concentric terms (the one being insUk the other) is assumed a
reciprocal guarantee. This guarantee gives rise to a historical
problem: can Form disguise Content, or must it be subservient
to it (so that there can no longer be a "coded" Form)? It is this
argument which sets in opposition, down through the centuries,
Aristotelian (later Jesuit) rhetoric and Platonic (later Pascalian)
rhetoric. This vision subsists, despite terminological change,
when we consider the text as the superposition of a signifud and
a s;gniflLT, the signified then being inevitably experienced (I am
speaking here of a more or less assumed vision) as a secret
hidden behind the signifier.
The second, much more recent opposition, of a more sden
tific aspect and largely tributary to the Saussurian paradigm
Langue I Parole (or Code I Message), is that of Norm and Deviance.
Style is then seen as the exception (though coded) to a rule;
it is the (individual. yet institutional) aberration of a current
usage, sometimes perceived as verbal (if we define the norm by
the spoken language), sometimes as prosaic (if we set Poetry
in opposition to "something else"). Just as the opposition
Form I Contn1t implies a phenomenological vision, so the
Nonn I Deviance opposition implies an ultimately moral vision
(under cover of a logic of endoxa): there is a reduction of the
systematic to the sociological (the code is what is statistically
guaranteed by the greatest number of users) and of the socio
logical to the normal, where social nature begins; literature, the
space of style, and because it is specifically this space, then

C:lpynghled malenal
92 Languages and Style
assumes a shamanic function, which UvjStrauss has well de
scribed in his Introduction to the Work of Marcel Mauss: it is the
site of (verbal) anomaly, as society establishes, recognizes, and
assumes it by honoring its writers, in the same way that the
ethnic group establishes the supernatural in the person of the
witch doctor (the way an abscess marks the limits of a disease),
in order to recuperate it in a process of collective communication.
I should like to stan from these two visions, less to attack
them than to complicate them.

Let us take first the opposition of Content and Form, of

Signified and Signifier. No doubt it includes a ceMain, irreducible
portion of truth. Structural analysis of narrative in its achieve
ments and its promises is wholly based on the conviction (and
the practical proof) that we can transform a givm text into a
more schematic version, whose meta-language is no longer the
integral language of the original text, without changing the
narrative identity of this text: in order to enumerate functions,
to reconstitute sequences, or to distribute agents--in short, to
bring to light a narrative grammar which is no longer the
grammar of the vernacular of the text-we must peel off the
stylistic (or, more generally, elocutionary, "expressive") film
from another layer of secondary (narrative) meaning, to which
the stylistic features have no pertinence: they can be varied
without affecting the structure. That Balzac should say of a
disturbing old man that he "kept upon his bluish lips a fixed
and paralyzed smile, as implacable and jeering as the smile of
a death's-head," has exactly the same narrative (or, more pre
cisely, semantic) function as if we were to transform the phrase
and say that the old man had something funereal and fantastic
about him (this seme is irreducible, since it is functionally
necessary to the sequence of the slory).
The error, however-and it is here that we must modify our
vision of Content and Form-would be to stop "peeling off"
style prematurely; what this (possible, as we have just said)
peeling off reveals is not a content, a signified, but another

GJPYnghtcd maknal
Style and Its Image 93

form, another signifer, or if we prefer a more neutral term,

another level, which is never the last (for the text is always artic
ulated around codes which it does not exhaust); signifieds are
forms, as we have known since Hjelmslev and even more dearly
from the recent hypotheses of certain psychoanalysts, anthro
pologists, philosophers. Recently, analyzing a tale by Balzac, 1
attempted to bring to light-without reference to style, with
which I was not concerned, and remaining within the boundaries
of the signified-an interplay of five different codes: actional,
hermeneutic, semic, cultural, and symbolic; the "citations" which
the author (or more exactly the performer of the text) extracts
from these codes are juxtaposed, mixed, superimposed within
one and the same expressive unit (a single sentence, for example,
or, more generally, a "Iexia," or unit of reading), so as to form
a braid, a fabric, or even (etymologically) a tat. Here is an
example: the sculptor Sarrasine is in love with a prima donna
whom he does not know to be a castrato; he abducts her, and
the apparent soprano defends herself: "The Italian woman was
armed with a dagger. 'If you come doser,' she said, 'I shall be
forced to plunge this weapon into your heart.' " Is there, lwhind
the statement, a signified? Not at all; the sentence is the "braid"
of several codes: a linguistic code (the French language), a
rhetorical code (antonomasia, interpolation of an inqut, i apos
trophe), an actional code (the armed defense of the victim is a
unit in the sequence Rape), a hermeneutic code (the castrato
conceals his sex by feigning to defend his virtue as a woman),
and a symbolic code (the knife is a symbol of castration).
Hence we can no longer see the text as the binary structure
of a content and a form; the text is not double, but multiple;
in the text, there are only forms, or, more precisely, the text in
its tola1ity is only a multiplicity of forms-without (a) content.
We can say metaphorically that the literary text is a stereography:
neither melodic nor harmonic, it is resolutely contrapuntal; it
mingles voices in a volume. not in a line, not even a double one.
Doubtless, among these voices (these codes, these systems, these
forms), some are more especially attached to verbal substance,

C :lpynghled malenal
94 Languages and Style
verbal play (linguistics, rhetoric), but this is a historical distinction,
useful only for the literature of the Signified (which is in general
the only literature that we have studied); for we need merely
think of a few modern texts to see that as the (narrative, logical,
symbolic, psychological) signified recedes still further. it is not
possible to sel in opposition (even with the greatest sensitivity
to nuance) systems of Form and systems of Content: style is a
historical (and not universal) concept, which has peninence only
for certain historic works. Does it have. within this older
literature, a definite function? I believe it does. The stylistic
system. which is one system among olhers. has a function of
naturalization, or of familiarization, or of domestication: the
units of the codes of content are in effect subjected to a rough
pigeonholing (actions are separated. characterial and symbolic
notations are disseminated, the march of truth is fragmented,
retarded); language. in the elementary aspects of sentence,
period, paragraph, superimposes upon this semantic disconti
nuity established on the level of discourse the appearance of
continuity; for however discontinuous language itself may be,
its structure is so fixed in the experience of each man that he
recognizes it as a veritable nature: do we not speak of the "flux
of speech"? What is more familiar, more obvious, more natural,
than a sentence read? Style "overspreads" the semantic articu
lations of content; by metonymic means, it naturalizes the story
told. declares it innocent.

Now let us turn to the second opposition, that of Norm and

Deviance. which is in effect the opposition of Code and Message,
since style (or literary effect) is experienced here as an aberrant
message which "surprises" the code. Here too, we must refine
our vision of the opposition rather than destroy it.
The features of style are undeniably drawn from a code, or
at least from a systematic space (this distinction seems necessary
if we want to respect the possibility of a multi-code. or even the
existence of a signifier whose space is governed and yet infinite.
i.e., an unsaturatable paradigm): style is a distance. a difference;

C JPYnghied mater1
Style and Its Image 95
but in relation to what? The reference is most often, implicitly
or explicitly, to the spoken ("current," "normal") language. This
proposition seems to me both excessive and insufficient: excessive
because stylistic codes of reference or difference are numerous,
and the spoken language is always only one of these codes
(which, moreover, there is no reason to privilege as the princeps
language, the incarnation of the fundamental code, the absolute
reference); insufficient because the opposition of spoken and
written is never exploited in all its depth. A word on this last
pomt.

We know that the object of linguistics, what determines at

once its task and its limits, is the sentence (however difficult to
define): beyond the sentence, there is no linguistics, for here
discourse begins, and the combinative rules of sentences are .
different from those of monemes; but, short of this, there is no
linguistics either, for then we can expect to find only shapeless,
incomplete, "unworthy" syntagms; only the sentence, we feel,
guarantees organization, structure, unity. Now, the interior
language (the language of thought) is essentially a sub-sentence
language; of course, it can include complete sentences, but the
code of the genre does not require this for the success and
advantage of the communication: we constantly talk without
finishing our sentences. Listen to a conversation: note how many
sentences there are whose structure is incomplete or ambiguous,
how many clauses are subordinated with no main clause or with
no clear antecedent, how many subjects lack predicates, how
many adversatives lack correlatives, etc. To the point where it
is abusive to continue speaking of "sentences," even "incomplete"
or "defective" ones; better to speak, more neutrally, of syntagms
whose "congregation" remains to be described. But if we open

a book, there will not be one sentence which is not ended, by an

overdetermination of operators structural, rhythmic, and
punctuational.
Whence, by rights, two autonomous linguistics: a linguistics
of the syntagm and a linguistics of the sentence, a linguistics of
the spoken and a lingubtics of the written. By carrying this
96 Languages and Style
distinction to its conclusions, we shall only be following the
recommendations of philosophy, which today assigns speech
and writing different ontologies; it is, philosophy says, by a
paradoxical abuse that linguistics deals only with the written
(language in sentences), while claiming that the canonical form
of language is speech, in respect to which writing is only
. .
" transcnptlon. "
We lack, obviously, a grammar of the spoken language (but
is this grammar possible: is it not the very notion of grammar
which would be eliminated by this division of communication?),
insofar as we have only a grammar of the sentence. This lack
determines a new distribution of languages: there are languages
of the sentence and all the other kinds. The first are marked
by a constraining character, an obligatory rubric: the completion
of the sentence. Style is obviously one of these written languages,
and its generic feature (which attaches it to the genre of the
written but does not yet distinguish it from its neighbors) is its
requirement to complete its sentences: by its finitude. by its
"neatness," the sentence declares itself written, en route to its
literary state: the sentence is already, in itself, a stylistic object:
the absence of smudging, by which it fulfills itself, is in a sense
the first criterion of style; we see this clearly in two properly
stylistic values: simplicity and contour: both are effects of neatness,
one litotic, the other rhetorical: if this sentence of Claudel's
("The night is so calm it seems salted") is both simple and
"contoured," it is because it completes the sentence in its
necessary and sufficient plenitude. This can be related to several
historical phenomena: first of all, a certain gnomic inheritance
of the written language (divinatory maxims, religious formulas,
whose typically sentential closure assures polysemy); then, the
humanist myth of the living sentence, effluvium of an organic
model, at once closed and generative (a myth discussed in the
treatise On the Sublime); last, the attempts though as yet inef
fectual, so closely is literature, even subversive literature, linked
to the sentence to explode the limits of the sentence (Mal
larme's Coup de des, hyper-proliferation of the Proustian sen-
Stylt and Its Image 97

tence, destruction of the typographic sentence in modern po

etry).
The sentence, in its closure and its "neatness," seems to me,
then, the fundamental determination of writing. From which
many written codes are possible (though not yet always identi
fied): learned, academic, administrative, journalistic writing,
each describable in terms of its clientele, its lexicon, and its
syntactical protocols (inversions, figures, c1ausulae, all features
marking the identity of a collective writing by their presence or
their absence). Among all these kinds of writing, and even
before speaking of style in the individual sense in which we
ordinarily understand this word. there is literary language, a
truly collective writing whose systematic features should be
itemized (and not only its historical features. as has been done
hitherto): What is it, for instance, which is permitted in a literary
text but not in an academic article? Inversions. order of com
plements, syntactical license. archaisms. figures, lexicon? What
we must grasp first is not the idiolect of an author but of an
institution (literature).
This is not all. Literary writing must be located not only in
relation to its closest neighbors but also in relation to its models.
I mean by models not sources, in the philological sense of the
word (let us note in passing that the problem of sources has
been raised almost exclusively on the level of the content), but
syntagmatic patkmS, typicaJ fragments of sentences, formulas,
if you like. whose origin is not identifiable but which make up
pan of the collective memory of literature. To wriU is to let
these models come to one and to transform them (in the sense
this word has acquired in linguistics).
I shall point out. in this regard. three phenomena, taken
virtually at random from a recent experiment. Ttoe first is
personal testimony: having worked for some time on a tale by
Balzac, I often catch myself spontaneously carrying over into
the circumstances of daily life fragments of sentences, formu
lations spontaneously taken from the Balzacian text; it is not
the memorial (banal) character of the phenomenon which

ateria IlIln al.llorskVIT

98 Languages and Style
interests me here, but the evidence that I am writing daily life
(it is true, in my head) through these formulas inherited from
an anterior writing; or again, more precisely, life is the very
thing which comes already constituted as a literary writing:
nascent writing is a past writing. The second phenomenon is an
example of external transformation: when Balzac writes: "I was
plunged into one of those profound reveries which overcome
everyone, even the frivolous. in the midst of the most tumultuous
festivities," the sentence, if we except its personal mark ("I was
plunged'). is merely the transformation of a proverb: Amidst
tumultuow ftstivities, profound rtvtries; in other words. the literary
speech-act refers, by transformation. to another syntactic struc
ture: the first content of the sentence is another form (here, the
gnomic form), and style is established in the effort of transfor
mation applied not to ideas but to forms; it remains, of course.
to identify the chief stereotypes (such as the proverb) from
which literary language is invented and generated. The third
phenomenon is an example of internal transformation (which
the author generates from his own formula): at a certain moment
of his stay at Balbec, the Proustian narrator tries to engage the
young elevator boy of the Grand Hotel in conversation. but the
boy does not answer him, Proust says, "whether from astonish
ment at my words. attention to his work. concern for etiquette,
hardness of hearing, respect for the place. fear of danger.
mental torpor, or orders of the director"; the repetition of the
same syntactical formula (a noun and its complement) is ob
viously a game, a "turn," and style then consists in (I) trans
forming a potential subordinate clause into a nominal syntagm
(becawt he did not hear wtll becomes his hard1USS of hearing); ()
repeating as often as possible this transformational formula
through different contents.
From these three random and virtually impromptu remarks,
I should merely like to draw a working hypothesis: that we
consider stylistic features as tTans/fJTJflOJiuns, derived either from
collective formulas (of unrecoverable origin, literary or pre
literary) or, by metaphoric interplay, from idiolectal forms; in
both cases, what should govern the stylistic task is the search

Mat(trlal neny autorsk

StylL and Its Image 99

for models, for pauems: sentential structures, syntagmatic dichb,

divisions and clausulae of sentences; and what should animate
this task is the conviction that style is essentially a citational
procedure, a body of formulas, a memory (almost in the
cybernetic sense of the word), an inheritance based on culture
and not on expressivity. This permits situating the lTansfonnation
to which I allude (and consequently the transformational stylis
tics I can desire): it cenainly has some affinity with transfor
mational grammar, but it differs from it on one fundamental
point (where linguistics, inevitably implying a cenain vision of
language, once again becomes ideological): stylistic "models"
cannot be identi6ed with "deep structures," with universal forms
derived from a psychological logic; these models are only the
depositaries of culture (even if they seem very old); they are
repetitions, not foundations; citations, not expressions; stereo
types, not archetypes.

To return to that vision of style to which I alfuded at the

beginning: in my opinion, it must consist today in seeing style
within the plurality of the text: a plurality of semantic levels
(codes). whose "braiding" forms the text, and a plurality of
citations deposited in that code we are calling "style" and which
I should prefer to call. at least as a 6rst object of study. literary
language. The problem of style can only be treated in relation
to what I shall call the layered quality of discourse; and, to
continue the alimentary metaphor, J shall sum up these few
remarks by saying that, if hitherto we have sun the text as a
fruit with its pit (an apricot, for instance), the flesh being the
form and the pit the content, it would be better to see it as an
onion, a superimposed construction of skins (of layers, of levels.
of systems) whose volume contains, 6nally, no heart, no core,
no secret, no irreducible principle, nothing but the very in6nity
of its envelopes-which envelop nothing other than the totality
of its surfaces.

Bellagio coUoquium, IgGg

Mater1 neny a
 Pax Culturalis

To say that there is a bourgeois cuhure is wrong. cause it is

OUT entire culture which is bourgeois (and to say that our culture
s
i bourgeois is a tiresome truism, one which is mouthed in all
our universities). To say that culture is in opposition to nature
is dubious. because we are not sure where the limits of each
are: Where is nature in man? I f one is to describe himself as
man, that man must have a language, Le., culture itself. In the
biological? Today we recognize in the living organism the same
structures as in the speaking subject: life itself is constructed as
a language. In shoTt. everything is culture, from garment to
book, from food to image. and culture is everywhere. from end
to end of the social scale. This culture, certainly. is a very
paradoxical obje<:t: without contours, without oppositional term,
without remainder.
Let us even add, perhaps: without incident-or at least without
schism, subject to a tireless repetition. Here, on television, an
American spy serial: cocktails on a yacht. and the characters
indulging in a kind of worldly banter (flirtations, double mean
ings, worldly interests); but this has already been 5un Qr said: not
only in thousands of popular novels and films, but in earlier
works belonging to what might pass for another culture, in
Balzac, for instance: one might suppose that the Princess de
Cadignan has simply changed place5, that she has left the Fau
bourg Saint-Germain for the yacht of a Greek shipowner. Thus,
culture s
i not only what returns, it is also and especially what
remains in place, like an imperishable corpse: it s
i a bizarre toy
that History nt'VtT brtalu.
, A unique object, since it never sets itself in opposition to
,
anything. an eternal object, since it never break5--in short, a

>00

C :lpynghled malenal
Pax Culturalis 101
peaceable object. in whose bosom everyone is gathered without
apparent conflict: then where is culture's reflexive task where
are its contradictions. where is its inadequacy?
To answer. we must. despite the epistemological paradox of
the object. risk a definition. the vaguest imaginable of course:
culture is a field of dispersion. Of what? Of languages.
In our culture. in the Pax culturalis to which we are subject.
there is an inveterate war of languages: our languages exclude
each other; in a society divided (by social class. money. academic
origin). language itself divides. What portion of languages can
I. as an intellectual. share with a salesman in the Nouvelles
Galeries? Doubtless. if we are both French. the language of
communication; but this is an infinitesimal share: we can exchange
pieces of information and truisms; but the rest. i.e the enor
.

mous volume. the entire play of language? Since there is no

subject outside language. since language is what constitutes the
subject through and through. the separation of languages is a
permanent grief; and this grief. if it does not occur only when
we leave our "milieu" (where everyone speaks the same lan
guage). is not only the material contact of others. coming from
other circles. other professions. which lacerates us it is precisely
that "culture" which. in good democracy. we are all supposed
to share: it is precisely when. under the effect of apparently
technical determinations. culture seems unified (an illusion
rather stupidly reproduced by the expression "mass culture")
that the division of cultural languages is excruciated. Spend an
evening at your television set (to keep to the commonest forms
of culture); you will receive despite efforts of general banali
zation undertaken by the producers several different lan
guages; it is impossible that all of these languages will respond
not only to your desire (I use this word in its strongest sense)
but even to your intellection: there is always. in culture. a share
of language which the Other (hence I myself) does not under
stand; my neighbor is bored by this Brahms concerto. and I
regard this variety sketch as vulgar and that soap opera as
idiotic: boredom. vulgarity. stupidity are various names for the
102 Languages and Style
secession of languages. The result is that this secession not only
separates men from each other, but each man, each individual
is in himself lacerated ; each day in myself, there accumulate,
without communicating, several isolated languages: 1 am frag
mented, severed, scattered (which in other circumstances passes
for the very definition of "madness"). And, even if 1 manage to
speak the same language all day long, how many different
languages 1 am compelled to receive! That of my colleagues, of
my postman, of my students, of the sports commentator on the
radio, of the classical author 1 read in the evening: it is a
linguist's illusion to consider on equal status the language spoken
and the language heard, as if they were the same; here we must
return to the fundamental distinction proposed by jakobson
between active grammar and passive grammar: the first is
monotonous, the second is heteroclite that is the truth of
cultural language; in a divided society, even if it manages to
unify its language, each man struggles against the explosion of
listening: under cover of that total culture institutionally offered
to him, the schizophrenic division of the subject is imposed
upon him every day; culture is in a sense the pathological field
par excellence, in which is inscribed the alienation of contem
porary man.
Thus, it seems, what is sought by each social class is not the
possession of culture either to obtain or to preserve it for
culture is there, everywhere and for everyone; instead, il is the
unity of language, the coincidence of speech and listening. How
then, today, in our Western society, divided in its languages and
unified in its culture, how do the social classes, the ones Marxism
and sociology have taught us to recognize how do they look
toward the language of the Other? What is the (alas, very
disappointing) interlccutory play in which, historically, they are
involved?
The bourgeoisie in principle possesses all of culture, but it
has been a long time (I am speaking for France) since it has
possessed any cultural voice of its own. Since when? Since its
intellectuals, its writers have dissociated themselves from it; the
Pax Culturalis 103
Dreyfus Affair seems to have been. in our country. the instituting
shock of this dissociation; this was. moreover. the moment when
the word intellectual appeared: the intellectual is the "clerk" who
seeks to break with the good conscience of his class if not his
class of origin (that a writer is "working class" changes nothing
about the problem). at least his class of consumption. Here.
today. nothing is invented: the bourgeois (owner. boss. executive.
administrator) no longer accedes to the language of intellectual.
literary. artistic research. because this language contests him;
he resigns in favor of mass culture; his children no longer read
Proust. listen to Chopin. but maybe Boris Viano pop music.
However. the intellectual who threatens him is no longer trium
phant for that; try as he will to posit himself as a representative.
as a procurator for the proletariat. as an oblate for the socialist
cause. his critique of bourgeois culture can use only the old
language of the bourgeoisie. which is transmitted to him by
university teaching: the idea of contestation itself becomes a
bourgeois idea; the public of intellectual writers may have shifted
(though it is certainly not the proletariat which reads them). but
not the language; of course. the intelligentsia seeks to invent
new languages. but these languages remain enclosed: nothing
has changed in social interlocution.
The proletariat (the producers) has no culture of its own; in
so-called developed countries. its language is that of the petite
bourgeoisie. because this is the language offered it by mass
communications (popular press. radio. television): mass culture
is petit-bourgeois. Of the three typical classes. it is the inter
mediate class perhaps because this is the century of its historical
promotion. which today is seeking to elaborate an original
culture. one that would be its own: it is incontestable that
important work is being done on the level of so-called mass
culture (i.e petit-bourgeois culture}--which is why it would be
.

ridiculous to hold aloof from it. But by what means is such work
being done? By the already known means of bourgeois culture:
it is by taking and degrading the models (the patterns) of
bourgeois language (its narratives. its types of reasoning. its


104 Languages and Style

psychological values) that petit-bourgeois culture creates and
implants itself. The idea of degradation may seem "moral," the
product of a bourgeois which regrets the excellence of past
culture; I am giving it, quite the contrary, an objective, structural
content: there is degradation because there is no invention;
models are repeated on the spot, banaliz.ed, because petit-bourgeois
culture (censored by the state) excludes even the contestation
the intellectual can contribute to bourgeois culture: it is immo
bility, submission to stereotypes (conversion of messages into
stereotypes) which defines such degradation. One can say that
in petit-bourgeois culture, in mass culture, it is bourgeois culture
which returns to the stage of History, but as a farce (Marx's image
will be recalled).
A game of hunt-the-slipper thereby seems to govern the
cultural war: the languages are indeed separated, like partners
in the game, sitting beside each other; but what is passed from
hand to hand is always the same object, the same culture: tragic
immobility of culture, dramatic separation of languages such
is the double alienation of our society. Can we trust socialism
to undo this contradiction, at once to "ftuidify," to pluralize
culture, and to put an end to the war of meanings, to the
exclusion of languages? Certainly; what hope elsewhere? Yet
we must not blind ourselves to the threat of a new enemy which
lies in wait for all modern societies. Indeed, it seems that a new
historical entity has appeared, has established itself, and is
developing pathologically, complicating (without outdating)
Marxist analysis: this new figure is the state (here, moreover,
was the enigmatic point of Marxist science): the state apparatus
is tougher than revolutions and so-called mass culture is the
direct expression of this state presence: in France, for example,
the state now seeks to dissociate itself from the university, to
turn its interest elsewhere, to concede the institution to the
communists and the protesters, for it realizes that it is not here
that a conquering culture will be created; but for nothing in the
world will it release its hold of television or radio; by possessing
these means of culture, it governs real culture, and, by governing
Pax Culturalis
it, makes it into its oum: a culture within which must gather
together the intellectually "resigning" class (the bourgeoisie),
the promotional class (the petite-bourgeoisie), and the silent
class (the proletariat). Thus, we understand why on the other side,
even if the problem of the state is far from being settled, the
People's Republic of China has named the radical transformation
of society it has undertaken a "cultural revolution."

The Times Literary Supplnnent, 197 1

 The War of guages

While out walking one day in my region, which is southwestern

France, a peaceable terrain of retired minor officials, I had
occasion to read, within a few hundred yards, on the doors of
three villas, three different signs: Vicious Dog, Dangerous Dog,
Watchdog. This region, evidently, has a very lively sense of
property. But that is not of such interest as this: these three
expressions constitute one and the same message: Do Not Enter
(or you will be bitten). In other words, linguistics, which is
concerned only with messages, could say nothing about them
but what is very simple, very banal; it would by no means
exhaust the meaning of these expressions, for this meaning is in
their difference: Vicious Dog is aggressive; Dangerous Dog is philan
thropic; Watchdog is apparently objective. In still other words,
through one and the same message, we read three choices, three
commitments, three mentalities, or again, three image-reper
toires, three alibis of ownership; by the language of his sign
by what I should call his discourse, since the language is the same
in the three cases the villa's owner is sheltered and reassured
behind a certain representation, even a certain system of own
ership: here fierce (the dog, i.e., the owner, of course, is vicious),
here protective (the dog is dangerous, the villa is armed), here
finally legitimate (a dog is guarding the property, a statutory
right). Thus, on the level of the simplest message (Do Not Enter) ,
language (discourse) explodes, fragments, diverges: there is a
division of languages, for which no simple science of communi
cation can account; society, with its socio-economic and neurotic
structures, intervenes, constructing language like a battle
ground.
Of course, it is the possibility of saying one and the same
1 06

I
 The War of Languages 1 07

thing in several ways, it is synonymy, which permits language

to divide itself up; and synonymy is a statutory, structural,
"natural" datum of language; but the war of languages is not

"natural": it occurs where society transforms difference into

conflict; one might say that there was a convergence of origin
between the division of social classes, symbolic dissociation, the
division of languages, and neurotic schism.
I have deliberately chosen my example a minimo, out of the
language of a single class, that of small landowners, whose
discourse sets in opposition certain nuances of appropriation. A
fortiori, on the level of social society, if I may put it that way,
language appears divided into great masses. However, one must
be convinced of three things which are not simple: the first is
that the division of languages does not match the division of
classes term for term: from one class to the next, there are
skids, borrowings, grids, relays; the second is that the war of
languages is not the war of subjects: it is linguistic systems which
are in opposition, not individualities sociolects, not idiolects; the
third is that the division of languages is marked against an
apparent background of communication: the national idiom;
more specifically, I should say that, on the scale of the nation,
we understand each other but we do not communicate: putting
things at their best, we have a liberal practice of language.
In contemporary societies, the simplest division of languages
bears on their relation to Power. There are languages which
are articulated, which develop, and which are marked in the
light (or the shadow) of Power, of its many state, institutional,
ideological machineries; I shall call these encratic languages or
discourses. And facing them, there are languages which are
elaborated, which feel their way, and which are themselves
outside of Power and/or against Power; I shall call these acrahc
languages or discourses.
These two major forms of discourse do not have the same
character. Encratic language is vague, diffuse, apparently "nat
ural," and therefore not easily discerned: it is the language of
mass culture (popular press, radio, television) and it is also, in


1 08 Languages and Style

a sense. the language of conversation. of public opinion (of the

doxa); encratic language is both (a contradiction which constitutes
its strength) clandestine (it is not easily recognizable) and trium
phant (it is inescapable): I shall say that it is sticky.
ACTatic language. on the other hand. is separate. severed from
the doxa (hence. it is paradoxical) ; its schismatic strength derives
from the fact that it is systematic. it is constructed around thought.
not around ideology. The most immediate examples of this
acratic language would be. today. Marxist discourse. psychoan
alytic discourse. and. to a lesser degree but statutorily percep
tible. structuralist discourse. But what is perhaps most interesting
is that. even in the acratic sphere. there occur. once again.
certain divisions. regionalities. and antagonisms of language:
critical discourse is fragmented into dialects. enclaves. systems.
I am tempted to call these discursive systems Fictions. adopting
Nietzsche's word, and to regard intellectuals as those who form,
again according to Nietzsche, the sacerdotal class, the caste re
spol}sible for elaborating, as artists, these Fictions of language
(has not the priestly class. for a very long time. been the owner
and the technician of formulas. i.e., of language?).
Whence certain relations of force between the discursive
systems. What is a strong system? It is a system of language
which can function in all situations. and whose energy subsists,
whatever the mediocrity of the subjects using it: the stupidity
of certain Marxists, of certain psychoanalysts, or of certain
Christians in no way jeopardizes the force of the corresponding
systems, of the corresponding discourses.
How do we account for the aggressive force, the power of
domination of a discursive system, of a Fiction? Since the old
Rhetoric. now permanently alien to our world of language. no
applied language has yet revealed the weapons of linguistic
combat: we do not know for certain the physics or the dialectic
or the strategy of what I shall call our logosphere though not a
day passes when each of us is not subject to the intimidations
of language. It seems to me that these discursive weapons are
at least of three kinds.
The War of Languages "'9
I. Every strong system of discourse is a representation (in the
theatrical sense: a show), a staging of arguments, aggressions,
retorts, formulas, a mimodrama, in which the subject can engage
his hysteric gratification.
2. There cenainly exist figures of system (as we used to say
figures of rhetoric), paniai forms of discourse, set up with a view
to giving the sociolect an absolute consistency, to closing the
system, to protecting it and to excluding the adversary from it
irremediably: for example, when a psychoana1yst says: "The
rejection of psychoanalysis is a resistance which itself relates to
psychoanalysis," that is a figure of system. In a general way,
figures of system aim at including the Other in discourse as a
simple object, the better to exclude him from the community
of subjects speaking the strong language.
3. Last, going still fUTlher, we can speculate whether the
sentence, as a practically closed syntactic structure, is not itself,
already, a weapon, an operator of intimidation: every complete
sentence, by its assertive structure, has something imperative,
something comminatory about it. The disorganization of the
subject, his timorous subjection to the masters of language, is
always translated by incomplete sentences, with vague contours,
undecided being. As a matter of fact, in ordinary, apparently
free life, we do not speak in sentences. And conversely, there
is a mastery of the sentence which is very dose to power: to be
strong is first of aU to finish one's sentences. Does not grammar
itself describe the sentence in terms of power, of hierarchy:
subject, subordinate, complemtnl, etc.?
Since the war of languages is general, what are we to do? By
"we" I mean we intellectuals, writers, practitioners of discourse.
Everything suggests that we cannot escape: by culture. by
political choice, we must be committed, engage in one of the
panicular languages to which our world, our history compels
us. And yet we cannot renounce the gratification-however
utopian-of a de-situated, dis-alienated language. Thus, we
must hold in the same hand the twO reins of commitment and
gratification. must assume a plural philosophy of languages.

Malena :>m d reitl al

 Languages and Style
Now, this eluwhere which remains, so to speak, within, has a
name: it is the Text. The Text, which is no longer the worlr, is a
production of writing, whose social consumption is certainly not
neutral (the Text is little read), but whose production is su
premely free, insofar as (Nietzsche again) it does not respect
the Whole (the Law) of language.
Only writing, as a matler of fact, can assume the fictional
character of the most serious, even the most violent dialects,
can replace them in their theatrical distance; for example, 1 can
borrow psychoanalytic speech in its wealth and its extent, but
make use of it in petto as of a language of fiction,
Moreover, only writing can mix languages (psychoanalytic,
Marxist, structuralist, for example). can constitute what is called
a heterology of knowledge, can give language a festive dimension,
Last, only writing can be deployed wthout
i a sile of origin; only
writing can bame every rhetorical rule, every law of genre, every
arrogance of system: writing is atopic; in relation to the war of
languages, which it does not suppress but displaces. writing
anticipates a state of reading and writing practiCes where it is
desire which circulates. not domination.

Le Conferenze deU'Associazione
Culturale luliana . 1973

Malena m dlreitl
 The Division of Languages

Is our culture divided? Indeed not; everyone, in our France

today, can untUrstawi a television broadcast, an anicle in FTonce
Soi,., the arrangement of a banquet; moreover, one can say that.
aside from a little group of intellectuals, everyone consumes
these cultural products: objective participation is total; and if
we define the culture of a society by the circulation of symbols
which occurs there. our cuhure appears as homogeneous and
united as that of some little ethnic society. The difference is tbat
it is only consumptiun which is general in our culture. not
production: we all understand what we hear in common, but we
do not all speak the same thing that we hear; "tastes" are divided,
sometimes even opposed, quite inveterately: I like this broadcast
of classical music which my neighbor cannot endure. while J
cannot bear the boulevard comedies he adores; each of us turns
on his television precisely when the other turns it off. In other
words, this culture of our time, which seems so general, so
peaceable, so communal, rests on the division of two activities
of language: on one side a-national-Iistening, or, if one
prefers, acts of intellection; on the other, if not speech, at least
creative participation, and. to be still more specific, the Ianguagt
of dtsirt. which remains divided: 1 listen on one side, I love (or
do not love) on the other: I undn-stand and I'm bortd: the unity
of mass culture corresponds, in our society, to a division not
only of languages but of language itself. Certain linguisu-
concerned, however. only with language and not with dis
course-have had a presentiment of this situation: they have
suggested-without being followed, so far-that we distinguish
two grammars: an activt grammar, or a grammar of the language
as it is spoken, emitted, produced, and a passive grammar, or a

, . .

Malena :>m d reitl al

, "

grammar of simple listening. Raised. by a trans-linguistic mu

tation, to the level of discourse, this division would nicely account
for the paradox of OUT culture, unitary in its code of listening
(of consumption), fragmented in its codes of production, of
desire: "cultural peace" (no apparent conflict on the level of
cuhuTe) refers to the (social) division of languages.
Scientifically. till now this division has been relatively uncen
sored. Of course, linguists know that a national idiom (French.
for example) includes a cenain number of species; but the
specification which bas been studied is geographic (dialects,
patois), not social; doubtless such a thing is postulated, though
minimized, reduced to "fashions" of expression (argots, jargons,
pidgins); and in any case, it is assumed, idiomatic unity is
reconstituted on the level of the speaker, provided with his own
language, an individual constant of speech known as an idiolect:
species of language are only intermediary, floating, "amusing"
states (derived from a kind of social folklore). This construction,
which originated in the nineteenth century, nicely corresponds
to a certain ideology-from which Saussure himself was not
exempt-which sets, on one side, society (idiom, language) and,
on the other. individual (idiolect, style); the tensions between
these two poles can only be "psychological": the individual is
supposed to struggle for recognition of his language or to
avoid being completely smothered by the language of others.
Yet the sociology of this period could not grasp the conflict on
the level of language: Saussure was more of a sociologist than
Durkheim was a linguist. It is literature which anticipated the
division of languages (even if it remained psychological), more
than sociology (we shall hardly be astonished: literature contains
all knowledges, though in a non-scientific state: it is a Mathesis).
The novel. once it became realistic. inevitably encountered
the copy of collective languages; but in general the imitation of
group languages (socio-professional languages) has been dele
gated by our novelists to secondary characters, to supernumer
aries responsible for "fixing" social realism, while the hero
continues speak.ing a timeless language whose "transparency"
TJu Diviswn of Languages
and neutrality are supposed to match a psychological universality
of the human soul. Balzac, for instance, has an acute awareness
of social languages; but, when he reproduces them, he frames
them, a little like set pieces, rhetorically produced; he marks
them with a folkloric, picturesque index; these are caricatures
of languages. such as the jargon of Monsieur de Nucingen.
whose phonetism is scrupulously reprodued. or the concierge
language of Madame Cibot, Cousin Pons's housekeeper; yet
there is another mimesis of language in Balzac. a more interesting
one. first of all because it is more naive, then because it is more
cultural than social: this is the mimesis of the cOlhs of public
opinm which Balzac often adopts as his own. when he is
commenting incidentally on the story he is telling: if, for
example, Balzac slips the figure of Brantome into the anecdot
(inSur Catherine tU Mtdicis), Brantome will speak of women
exactly as public opinion (the doxa) expects Brantome to honor
his cultural "role" of "specialist" in stories about women
without our being able to swear, alas. that Balzac himself is quite
conscious of his own procedure: for he believes he is reproducing
Brantome's language. whereas in fact he is copying only the
(cultural) copy of that language. This suspicion of naivete (some
will say. of vulgarity) we cannot attach to Flaubert. who does
not include himself in reproducing simple tics (phonetic. lexical.
syntactical); he uses imitation to capture the mon subtle and
diffuse linguistic values and to grasp what we might call figures
of discourse; and above all. if we consider Flaubert's most "pro
found" book. Bouvard and ptcucJut, the mimesis is without basis
or prop: the cultural languages-languages of sciences. of
technologies. of classes too: the bourgeoisie are cited (Flaubert
does not take them for ready money). but, by an extremely
subtle mechanism, one which only today is beginning to be
discovered. the author who copies (contrary to Balzac) remains
unrecoverable. insofar as Flaubert never gives us a sure means
of knowing whether he puts himself definitively outsUk the
discourse he is "borrowing": an ambiguous situation which
makes somewhat illusory the Sartrian or Marxist analysis of

Matena dlreitl
"4 Languags and Style
FlauTt'5 bourgeoisie; for if Flaubert, a bourgeois, speaks the
language of the bourgeoisie, we never know from what site this
speech-act functions: A critical site? A distant one? Or an
associated one? In truth, Flaubert's language is ulopic, and this
is what constitutes its modernity: are we not in the process of
learning (from linguistics, from psychoanalysis) precisely that
iangtUlge is a sitt with no almor? After Balzac and F1aubert-to
mention only the greatest-in order to confront this problem
of the division of languages, we can cite Proust, because we find
in his work a true encyclopedia of language; without returning
to the general problems of signs in Proust-which Deleuze has
treated so remarkably-and remaining on the level of articulated
language. we find in this author every state of verbal minusis,
i.e., characterized pastiches (the letter from Gisele, which mimics
academic jargon, the Goncourts' Journal), idiolects of character,
each panicipant in the Sarch for Lost Tim having his simulta
neously characterial and social language (Charlus the medieval
seigneur. Legrandin the snob), a clan language (the jargon of
the Guermames), a class language (Franoise and the "folk,"
though one reproduced here mainly by reason of its allegiance
to the past), a catalogue of linguistic anomalies (the distorting,
"outlandish" language of the manager of the Grand Hotel de
Balbec), the scrupulous collection of phenomena of acculturation
(Franoise contaminated by her daughter'S "modern" language)
and linguistic diaspora (the Guermantes language's "swarms"),
a theory of etymologies and of the founding power of the name
as signifier; there is even, in this subtle and complete panorama
of the types of discourse, a (deliberate) absence of certain
languages: the narrator, his parents, Albertine do not have a
language of their own. Whatever advance literalUre has made
in the description of divided languages, one sees the limits of
literary mimesis: on one hand, the language reported does not
manage to emerge from a folklorist (one might say, colonial)
view of exceptional languages; the language of the Other is
framed, the author (except perhaps in Flaubert's case) speaks it
in a situation of extraterritoriality; the division of languages is

GJPYnghtcd maknal
Tlu Division of Languages
often recognized by these "objective" authors with a perspicacity
socia-linguistics might well envy. but it remains external to the
describer: in other words. contrary to the acquisitions of modern.
relativist science. the observer does not utler his place in the
observation; the division of language stops at th one who
describes (if he does not denounce) it; and on the other hand.
the social language reproduced by literature remains univocal
(still the division of grammars denounced earlier): Franoise
speaks by herself. we understand her. but no one, in the book.
answers her; the language observed is monologic. it never
participates in a dialectic (in the proper sense of the term); the
result is that the fragments of language are in fact treated as so
many diolects-and
i not as a total and complex system of pro
duction of languages.
Hence, let us turn to the "scientific" treatment of the question:
How does (socia-linguistic) science see the division of languages?
The postulation of a link between the division of classes and
the division of languages is obviously not a recent insight: the
division of labor engenders a division of lexicons; it can even
be said (Greimas) that a lexicon is precisely the outline imposed
upon the semantic mass by the practice of a certain labor: no
lexicon without a corresponding labor (there are no grounds
for making an exception for that general. "universaJ" lexicon,
which is merely a lexicon "outside labor"); socia-linguistic in
vestigation would therefore be easier to conduct within ethnic
societies than in our historical and "developed" societies. where
the problem is extremely complex; for us, in effect, the social
division of languages sms blurred both by the weight, the
unifying force of the national idiom. and by the homogeneity
of so-called mass culture, as has been suggested; a simple
phenomenological observation suffices. however, to attest to the
validity of linguistic separations: one nd merely emerge for a
minute from one's own milieu and to have the wk. if only for
an hour or two, not only to listen to other languages besides
one's own but also to panicipate in the conversation as actively
as possible, in order to perceive, always with embarrassment,

GJPYnghtcd maknal
 Languages and Style
sometimes with laceration, the extremely hermetic nature of
languages within the French idiom; the failure of these lan
guages to communicate (except about "the weather") occurs not
on the level of language, understood by all, but on the level of
discourse (an object which is beginning to join linguistics); in
other words, lack of communication is not strictly speaking of
an informational order but of an interlocutory order: from one
language to another, there is indifference, lack of curiosity: in
our society, the language of the same suffices us, we have no
need of the Other's language in order to live: to each his own
language suffrus. We lock ourselves into the language of our
own social, professional cell, and this sequestration has a neurotic
value: it permits us to adapt ourselves as best we can to the
fragmentation of our society.
Obviously, in the historic conditions of sociality, the division
of labar is not refracted directly, as a simple mirror-image within
the division of lexicons and the separation of languages: there
is compkxiuztion, overdetermination or contrariety of factors.
And, even in countries of relatively equal development, differ
ences, generated by history, can persist; I am convinced that,
compared to other countries no more "democratic" than ours,
France is particularly divided: there is in France, perhaps by
classical tradition, an intense awareness of the identities and
propmies of language; the other's language is perceived accord
ing to the most extreme qualities of its otherness: whence the
frequent accusations of 'Jargon" and an old tradition of irony
with regard to closed languages which are quite simply othn
languages (Rabelais, Moliere, Proust).
Confronting the division of languages, do we possess a scheme
of scientific description? Yes, and it is obviously socio-linguistics.
Without wanting to lodge a contestation of this discipline, we
must nonetheless acknowledge a certain disappointment: socio
linguistics has never dealt with the problem of the social language
(as a divided language); on the one hand, there have been
certain rapprochements (though episodic and indirect) between
macro-sociology and macro-linguistics, the phenomenon "soci-

C :lpyngh!ed malenal
The Division of Languages " 7

ety" being put in relation with the phenomenon "language"; on

the other hand, and one might say at the other end of the scale,
there have been few attempts at a sociological description of
speech communities: language of prisons. parishes, formulas of
politeness. baby talk; socio-Iinguistics (and it is on this poim that
we can register our disappointment) refers to the separation of
social groups insofar as they are struggling for power; the division
of languages is not conceived as a total fact, involving the very
roots of the economic system, of culture. of civilization, even of
history, but only as the empirical (and not at all symbolic)
attribute of a half-sociological. half-psychological arrangement:
the desire for promotion-a narrow view. to say the least, and
one which does not correspond to our expectations.
Has linguistics (and, no longer, sociology) done any better? It
has rarely brought languages and social groups into relation,
but it has ventured to make historical investigations into voca
bularies, into lexicons endowed with a certain social or institu
tional autonomy: we may instance. here, the work of Meillet on
the IndO-European religious vocabulary; of Benveniste on Indo
European institutions; of Matore, who attempted, some twenty
years ago, to establish a veritable historical sociology of the
vocabulary (or lexicology); more recently, of Jean Dubois, who
has described the vocabulary of the Commune. The attempt
which best shows the interest and the limits of socio-historical
linguistics is perhaps that of Ferdinand Brunot; in Volumes X
and XI of his monumental Histoire th la langue frarlfaise dis
origines a 1900, Brunot studied in great detail the language of
the French Revolution. The interest of his work is this: what is
studied is a political language, in the full sense of the word; not
a group of verbal tics intended to "politicize" language from
outside (as often occurs today), but a language elaborated in
the very movement of a political praxis; whence the more
productive than representative character of this language: words.
whether discarded or promoted. are linked almost magically to
a real effectiveness: by abolishing the word, one believes one is
abolishing the referent; with the banning of the word noblesse,

C ;.pvnghted malaria
. .8 Languages and Style

it is apparently the nobility which is being banned; the study of

this political language might furnish a fine context for an analysis
of our own political (or politicized?) discourse: affectivized words,
marked with a taboo or a countertaboo. cherished words (Nation,
Law, Patm, Constitution), execrated words (Tyranny, Aristoc
racy. Conspiracy), exorbitant power of certain words, however
"pedantic" (ConstilUtion, Federalism), terminological "transla
tions," substitutive creations (cltTgy-+ "prttraillt, " religion-fa
naticism, religious objtcHTeligiow baubus, enemy soldiers-vii sal
ellius of the de.spoo, taxn-+cOfllribulio,n snvant-+"/wmme de wnfIaTlU,"
informers-police agents, "comidiens"_artists, etc.), unrestrained
connotations (revolutionary ends by signifying prompt, acttlerattd).
As for the limit, it is the following: the analysis bears only on
the ltxicon; it is true that French syntax was little affected by (he
Revolutionary shock (which in fact made every effort to govern
it and to maintain classic "good usage"); but perhaps it would
be better to say that linguistics does not yet have the means to
analyze that fine structure of discourse which is located between
grammatical "construction" (too loose) and the vocabulary (tOO
limited), and which doubtless corresponds to the region of
frozen syntagms (for instance: "the pressure of the Revolution
ary masses"); the linguist is then obliged to reduce the separation
of the social languages to phenomena of the lexicon--even of
fashion.
Hence, the most interesting situation, Le., the very opacity of
the social relation, seems to escape traditional scientific analysis.
The basic reason, it seems to me, is of an epistemological order:
confronting discourse, linguistics has remained, one might say,
at a Newtonian stage: it has not yet experienced its Einsteinian
revolution; it has not theorized the linguist'S place in the field
of observation. It is this relativiz.ation that must first be postu
lated.

It is time to give a name to these social languages outlined

within the idiomatic mass, and whose hermeticism, however
existentially we have experienced it, follows, through every

C :lpynghled malenal
The Division of Languages "9

conceivable nuance and complication, the division and the

opposition of classes; let us call these group languages socwkcts
(in evident opposition to the idwkct or jargon of a single
individual). The main character of the sociolectal field is that
no language can be exterior to it: all speech is inevitably included
in a certain sociolect. This constraint has an important conse
quence for the analyst: he himself participates in the interplay
of sociolects. It will be said that in other cases this situation in
no way prevents scientific observation: this is the case of the
very linguist who must describe a national idiom, i.e., a field
which no language (including his own) escapes; but precisely:
the idiom is a unified field (there is only one French language),
the one who speaks it is not obliged to situate himself within it.
Whereas the sociolectal field is defined by its division, fts
inveterate secession, and it is within this division that the analysis
must take place. It follows that sociolectal research (which does
not yet exist) cannot begin without an initial, founding action
of tvaluation (1 should like this word to be understood in the
critical sense Nietzsche gave it). This means we cannot pour all
sociolects (all social jargons), whatever they are, whatever their
political context, into a vague undifferentiated corpus, whose
very lack of differentiation, whose equality, would be a guarantee
of objectivity, of scientificity; here we must refuse the adiaplwria
of traditional science, we must accept-a paradoxical order, in
the eyes of many-that the types of sociolects govern the analysis,
and not the converse: typolngy is antericr to definition. Let us
specify further that evaluation cannot be reduced to appreciation:
quite objective scholars have accorded themselves the (legitimate)
right to appreciate the phenomena they were describing (this is
precisely what Brunot did with the French Revolution); to
evaluaJe is not a subsequent but a founding act; it is not a
"liberal" order of behavior, but on the contrary a violent one;
sociolectal evaluation, from the start, experiences the conflict of
groups and of languages; by positing the sociolectal concept, the
analyst must immediately account both for social contradiction
and for the fragmentation of the scholarly subject (1 refer here

Malena :>m d reitl al

 .20 LAnguages and Style
to the Lacanian analysis of the "subject supposed to know"
sujtt suppose savoir).
Hence, no scientific description of the social languages (of the
sociolects) without a founding political evaluation. Just as Aris
tot1e, in his Rhetoric, distinguished two groups of proofs: proofs
within the uchni (entechnoi) and proofs oulsUU the techni (atechnoiJ,
I am suggesting we distinguish from the start two groups of
sociolects: discourseswithin power (in the shadow of power) and
discourses outside power (or without power, or even in the light
of non-power); resorting to pedantic neologisms (but how else
to proceed?), let us call the former, tneratic discourses, and the
latter, acratic discourses.
Of course, the relation of a discourse to power (or to exclusion
from power) is very rarely direct, immediate; cetaioly the law
forbitb, but its discourse is already mediatized by a whole juridical
culture. by a ratio almost universally admitted; and only the
fabulous figure of the Tyrant could produce a speech which
would instantaneously adhere to his power ("the King ordered that
. "). As a matter of fact, the language of power is always

. .

furnished with structures of mediation, conduction, transfor

mation. inversion (as is ideological discourse. whose inverted
character. in relation to bourgeois power, Marx bas indicated).
Similarly, acratic discourse does not always stand declaratively
against power; to take a specific and current example, psychoan
alytic discourse is not directly linked (at least in France) to a
critique of power, and yet it can be classified among the acratic
sociolects. Why? Because the mediation which intervenes be
tween power and language s
i not of a political order but of a
cultural order: adopting an old Aristotelian ootion, that of the
rkxa (public opinion, the general, the "probable," but not the
"true," the "scientific"), we shall say that the dtJXa is the cultural
(or discursive) mediation through which power (or non-power)
speaks: encratic discourse is a discourse that conforms to the
ckxa, subject to codes which are themselves the structuring lines
of its ideology; and acratic discourse always speaks out, to
various degrees, against the ckxa (whatever it is, acratic discourse
is paradoxial). This opposition does not exclude nuances within

Matena dlreitl
The Division of Languages '21

each type; but, structurally, its simplicity remains valid as long

as power and non-power are in their place; it can be (provision
ally) blurred only in the rare cases where there is a mutation of
power (of the sites of power); thus, in the case of the political
language in a revolutionary period: revolutionary language
issues from the preceding acratic language; in shifting over to
power, it retains its acratic character, as long as there is an active
struggle within the Revolution; but once this struggle dies down,
once the state s
i in place, the former Revolutionary language
itself becomes doxa, encratic discourse.
Encratic discourse since we have subjected its definition to
the mediation of the doxa-is not only the discourse of the class
in power; classes out of power or attempting to take power by
reformist or promotional means can borrow it--<>r at least
receive it consentingly. Encratic language, supponed by the
state, is everywhere: it is a diffused, widespread, one might say
osmotic discourse which impregnates exchanges, social rites, lei
sure, the socio-symbolic field (above all, of course, in societies
of mass communication). Not only does encratic discourse never
describe itself as systematic, but it always constitutes itself as an
OPPOSition to system: alibis of nature, of universality, of good sense,
of c1arity-the anti-intellectualist resistances---become the tacit
figures of the encratic system. Further, it is afull discourse: there
is no room in it for the Other (....hence
. the sensation of smothering,
of stifling which it can provoke in someone who does not
participate in it). Finally, if we refer to the Marxian schema
("Ideology is an inverltd image of the real"), encratic discourse
as fully ideological-presents the real as the reversal of ideology.
In short, it is a non-marlced language, producer of a masked
intimidation, so that it is difficult to assign it morphological
features-unless we manage to reconstitute with rigor and pre
cision what is something of a contradiction in terms: the figures
of the maslced. It is the very nature of the doxa (diffuse. full,
complete, "natural") that makes an internal typology of encratic
sociolects difficult; there is an atypia of the discourses of power:
this genus knows no species.
The acratic sociolects are doubtless easier and more interesting

Malena :>m d reitl al

.22 Languages and Style
to study: they are all the languages which are elaborated outside
the doxa and are consequemly rejected by it (ordinarily under
the name ofjargons). By analyzing encratic discourse, we know
more or less in advance what we shall find (which is why, today,
the analysis of mass culture is visibly marking time); but acratic
discourse is by and large our own (that of the researcher, the
intellectual, the writer); to analyze it is to analyze ourselves
insofar as we speak: always a risky operation and yet one that
must be undertaken: what do Marxism, Freudianism, structur
alism, or the science of the so-called human sciences-insofar
as each of these group languages constitutes an acratic
(paradoxical) .sociolect-what do they think of their own dis
course? This question, which is never assumed by the discourse
of power, is obviously the founding act of any analyst who
claims not to exteriorize himself from his object.
The principal advantage of a sociolect (outside the advantages
which the possession of a language gives to any power one seeks
to preserve or to gain) is obviously the security it affords: like
any closure, that of a language exalts, reassures all the subjects
imide, rejects and offends those outside. But how does a sociolect
act outside itself? As we know, there is no longer, today. an an
of persuasion. there is no longer a rhetoric (at least no longer
one not ashamed to assume itself openly); it will be noted in
this regard that Aristotelian rhetoric. being founded on the
opinion of the greatest number. was by rights, and one may say
voluntarily, declaratively, an eodoxal, hence encratic rhetoric
(which is why, by an only apparent paradox, AristOlelianism can
still furnish very good concepts to the sociology of mass com
munications); what has changed is that. in modern democracy,
"persuasion" and its ttchni are no longer theorized, because the
systematic is censored and because, under the effect of a strictly
modern myth, language is reputed to be "natural," "instrumen
tal." We can say that by a single impulse our society rejects
rhetoric and "forgets" to theorize mass culture (a flagrant
oversight in Marxist theory posterior to Marx).
As a matter of fact. the sociolects do nm derive from a tuhnt

Matena dlreitl
The Divisiun of LAnguages
of TSuasion, but they all include figures of intimidation (even
if acratic discourse seems more brutally terrorist): fruit of social
division. witness to the war of meaning, every sodolect (encratic
or acratic) aims at keeping the Other from saking (this is also
the fate of the liberal sociolect). Hence. the division of the two
great tys of sociolects merely sets types of intimidation, or, if
you prefer, modes of pressure, in opposition: the encratic
sociolect acts by reprtssion (of endoxal superfluity, of what
Flaubert would have called Bitist); the acratic sociolect (being
outside power. it must resort to violence) acts by
subjection, it
mounts offensive figures of discourse, intended to corutrain
rather than to invade the Other; and what sets these two
intimidations in opposition is once again the acknowledged role
of system: declared recourse to an elaborated system defines
acratic violence; the blurring of the system, the inversion of
thought into "experience" (and non-Lhought) defines encratic
repression: there is an inverted relation between the two systems
of discursivity: patent I hidden, overt I covert.
A sociolect has an intimidating character not only for those
excluded from it (by reason of their social, cultural situation):
it also constrains those who participate in it (or. rather, who
receive it as their lot). This results, structurally, from the fact
that the sadoled, on the level of discourse. is a true language;
following Boas, Jakobson has nicely remarked that a language
is defined not because it permits saying . . . but because it compels
saying . . . ; so every sodolect involves "obligatory rubrics," great
stereotyped forms outside which the clientele of this. sociolect
cannot speak (cannot think). In other words, like every language,
the sociolect implies what Chomsky calls a competence, within
which variations of performance become structurally insignifi
cant: the encratic sociolect is not broached by differences of
vulgarity established between its locuLOrs; and on the other side,
everyone knows that the Marxist sociolect can be spoken by
imbedles: the sodolectal language is not called upon to change
according to individual accidents, but only if there occurs in
history a mutation ofdiscursivily (Marx and Freud were themselves

Matena dlreitl
1 24 Languages and Style

such mutants, but since them, the discursivity they founded is

merely repeating itself).
To conclude these remarks, situated ambiguously between
the essay and the research program, may I observe that, as I
see it, the division of social languages, sociolectology, if you like,
is linked to an apparently unsociological theme which has
hitherto been the special realm of theoreticians of literature;
this theme is what is today called writing. In our society of
divided languages, writing is becoming a value worthy to institute
a continuing debate and a constant theoretic exploration, be
cause it constitutes a production of undivided language. Having lost
every illusion, we know today that the writer cannot speak the
"language of the people," as Michelet nostalgically claimed; nor
can we align writing with the language of the greatest number,
for in an alienated society the greatest number is not the
universal, and to speak such a language (as in mass culture,
where one is on the statistical hunt for the greatest number of
listeners or telespectators) is still to speak a special language-e-
even if a majority language. We know that language cannot be
reduced to simple communication, it is the whole human subject
that is committed to speech and is constituted through it. In
modernity's "progressive" attempts, writing holds an eminent
place, not as a consequence of its (extremely reduced) clientele,
but as a consequence of its practice: it is because writing attacks
the relations of the subject (always social: is there any other?)
and language, the outdated distribution of the symbolic field
and the process of the sign, that writing appears as a practice
of counterdivision of languages : probably a utopian, in any case
a mythic image, since it rejoins the old dream of an innocent
language, the lingua adamica of the first romantics. But does not
history, in Vico's splendid metaphor, move in a spiral? Must we
not renew (which does not mean repeat) the old images in order
to give them new contents?

From Hommagt a Gtorgts Friedmann (Gallimard), 1973

FROM HISTORY
TO RFAI JIY


Copyrigh,cd moterial
 The Discourse of History

The formal description of groups of words superior to the

sentence (which will for convenience's sake be called discourse)
is not of recent date: from Gorgias to the nineteenth century,
it was the specific object of the old rhetoric. Recent developments
of linguistic science nonetheless give it a new actuality and new
means: a linguistics of discourse may henceforth be possible; by
reason of its effects on literary analysis (whose importance in
teaching is familiar to us), it even constitutes one of the first
tasks of semiology.
This second linguistics, at the same time that it must seek out
the universals of discourse (if they exist), in the form of units
and general rules of combination, must obviously decide if
structural analysis permits retaining the old typology of dis
course, if it is indeed legitimate still to oppose poetic discourse
to fictional discourse, fictive narrative to historical. It is on this
last point that I should like to offer some reflections: the
narration of past events, commonly subject in our culture, since
the Greeks, to the sanction of historical "science," placed under
the imperious warrant of the "real," justified by principles of
"rational" exposition-does this narration differ, in fact, by some
specific feature, by an indubitable pertinence, from imaginary
narration as we find it in the epic, the novel, the drama? And
if this feature or this pertinence exists, in what site of the
discursive system, at what level of the speech-act, must we locate
it? In order to answer this question, we shall observe here, in a
free and by no means exhaustive fashion, the discourse of
several great classical historians, mainly Herodotus, Machiavelli,
Bossuet, and Michelet.
 From History ro RaLity

1. Speech-act
And first of all, under what conditions is the classical historian
led-or authorized-to designate. in his discourse, the very act
by which he utters it? In other words, what are. on the level of
discourse-and no longer of language-the shifters (in the sense
Jakobson has given this word) which assure transition from
statement to sech-act (or conversely)?
It seems that historical discourse involves two regular types
of shifters. The first ty we might call shifters of listening. This
category has been observed, on the level of language, by
Jakobson, under the name testimonial and under the formula
CCalCa2: besides the event reponed (C"), the discourse mentions
both the act of the informant (cal) and the sech of the "writer"
who refers to it (Ca2). This shifter therefore designates all
mention of sources, of testimony, all reference to a listening of
the historian, collecting an elUWMr of his discourse and speaking
it. Explicit listening is a choice. for it is possible not to refer to
it; it relates the historian to the ethnologist who mentions his
informant; we therefore find this shifter of listening abundant
in such historian-ethnologists as Herodotus. The forms they
employ vary from interpolations of the ty as I hav luard, to
our Jmowkdge, to the historian's present (a tense which attests to
the saker's intervention) and to any mention of the historian's
rsonal exrience; this is Michelet's case, who "listens" to the
History of France starting from a subjective illumination (the
July Revolution of 1830), and accounts for it in his discourse.
The shifter of listening is obviously not rtinent to historical
discourse: we find it frequently in conversation and in certain
artifices of the novel (anecdotes recounted as "heard from"
certain fictive informants who are mentioned).
The second type of shifter covers all the declared signs by
which me "writer," in this case the historian, organizes his own
discourse. revises it. modifies it in me process of expression; in
short, arranges explicit references wimin it. This is an important
shifter, and the "organizers" of discourse can receive many

B
The Discourst of Hislory
different expressions; they can all be reduced, however, to the
indication of a movement of the discourse in relation to its
substance, or more precisely throughout this substance, some
thing like such temporal or locative deictics as voici I voi/d; hence
we have, in relation to the flow of the speech-act: immobility (as
we have said earlitT), harking back (altius repeltre, replicart da pitt
alto [tulgo), the return (ma ritornando air ordine nostro, dico come
. ). the hah (on this point, we shall say no more), the declaration
. .

(here are the othtr memorable actions he performed during his reign).
The shifter of organization raises a notable problem. which we can
only mention here: it is generated by coexistence or. to put it
better. by the conflict of two time spans: the time of the speech
act and the time of the material stated. This conftict gives rise
to important phenomena of discourse; we shall cite three. The
first refers to all the acceleration phenomena of history: an
equal number of "pages" (if such is the crude measure of time
in the speech-act) cover varying lapses of time (time of the
material stated): in Machiavelli's History of Florence. the same
measure (a chapter) covers several centuries here and some
twenty years there; the closer we come to the historian's own
time, the more powerful the pressure of the speech-act becomes.
and the more history slows down; there is no isochrony-the
result of which is implicitly to attack the linearity of discourse
and to reveal a possible "paragrammatism" of historical speech.
The second phenomenon also suggests, in its way, that the
discourse, though materially linear, when confronted with his
torical time apparently determines to explore this time, pro
ducing what we might call zigzag history: thus, with each
character who appears in his History. Herodotus goes back to
the newcomer's ancestors, then returns to his point of depanure,
in order to continue a little funher-and to begin all over again.
Finally, a third phenomenon of discourse, and a considerable

Followin g J. Kristen ("Bakhrine, Ie IDOl, Ie dialogue et Ie roman." CriIiqut.

no. t!9, April 1967), we shall designate as JIa'CIIl'alll4lisl

a lt (derived from
UI5Ure'l Anagnml) the double writings which contain a dialogue of the teXt
with other texts and postulate ill new Iogic::.

,
 From History 10 Reality
one, attests to the destructive role of the shiflers of organization
in relation to history's chronicle time: this involves inaugurations
of historical discourse, places where the beginning of the ma
terial stated and the exordium of the speech-act are united.
The discourse of history knows, in general, two forms of
inauguration: first of all, what we might call the pnformative
opening, for in it speech is actually a solemn act of foundation;
the model of this is poetic, the I sing of the poets; thus, joinville
begins his history by a religious appeal ("In the name of God
Almighty, I, jehan, Sire de joinville, cause to be written the life
of our Holy King Louis"), and even the socialist Louis Blanc
does not disdain the purifying introit,t so difficult does the
inception of speech remain-or so sacred, let us say; subse
quentJy, a much more common unit, the Preface, a characteristic
speech-act, prospective when it announces discourse to come,
or retrospective when it judges that discourse (as in the great
Preface with which Michelet crowned his Histojre de Franct once
it was completely written and in fact published). Our review of
these units tends to suggest that the entrance of the speech-act
into historical statement, through shifters of organization, has as
its goal not so much to give the historian a chance to express
his "subjectivity" as to "complicate" history'S chronicle time by
confronting it with another time. that of discourse itself, a time
we may identify as paper time; in short, the presence, in historica1
narration, of explicit speech-act signs tends to "de-chronologize"
the historical "thread" and to restore, if only as a reminiscence
or a nostalgia, a complex, parametric, non-linear time whose
deep space recalls the mythic time of the ancient cosmogonies.
it too linked by essence to the speech of the poet or the

The exordium (of any discourse) raises one of the most interesting problems
of rhetoric, insofar as it is a (:odification or the breaks in silen(:e and a struggle
against aphasis.
t "Before taking up my pcn I have questioned myself closely, and since I
,

diKerned neither p;lrtisan affections nor implaable hatreds, I have decided

that I could judge of men and things without neglecting justice and without
betraying the tTUth."-louis Blanc. Hist4iTt ,u da t1fU (Pam, 184)

B
T Discount of History '3'

soothsayer; in effect, the shifters of organization attest-if only

by certain apparently rational detours-to the historian's pre
dictive function: it is insofar as he knows what has not yet been
recounted that the historian, like the agent of myth, needs to
double the chronic splitting of events by references to the actua'i
time of his speech.
The signs (or shifters) we have just mentioned bear uniquely
on.the speechact's actual process. There are others which no
longer concern the speechacl but, in jakobson's terminology,
its protagonists (T"), addressee, or "writer." It is a notable and
rather enigmatic fact that literary discourse very rarely includes
signs of the "reader"; we might even say that what specifies it
as literary discourse is that it is-apparently-a discourse without
you. though in reality the whole structure of this discourse
implies a "subject" of the reading. In historical discourse, the
signs of reception or destination are commonly absent: we find
them only when History gives itself out as a lesson; this is the
case with Bossuet's Histoirt univtrstlk, a discourse nominally
addressed by the tutor to the prince, his student; yet this schema
is possible. in a sense, only insofar as Bossuet's own discourse is
supposed to reproduce homologically the discourse God Himself
offers men precisely in the form of History He gives them: it s
i
because the History of men is Scripture that Bossuet, mediator
of this Scripture, can establish a relation of destination between
the young prince and himself.
Signs of the "writer" (or sender) are obviously much more
frequent; here we must lis all the fragments of discourse in
which the historian, an empty subject of the speech-act. gradually
fills himself with various predicates intended to establish him as
a ptrson. provided with a psychological plenitude. with a coun
ttnanct. We shall indicate here one particular form of this
"filling," which relates more directly to literary criticism, This
occurs when the historian intends to "absent himself" from his
discourse and where there is, consequently. a systematic absence
of any sign referring to the sender of the historical message:
history seems to ttU itstlf. This accident has had a considerable

B
'3' From History to Reality
career. since it corresponds in fact to so-called objective historical
discourse (in which the historian never intervenes). As a matter
of fact, in this case, the speaker annuls his emotive person, but
substitutes for it another person, the "objective" person: the
subject subsists in his plenitude, but as an objective subject; this
is what Fuslei de Coulanges called, significantly (and rather
naively), the "chastity of History." On the level of discourse,
objectivity-or lack of signs of the "speaker"-thus appears as
a special form of imagerepenoire, the product of what we
might call the Tt/ern1tial illusion, since here the historian claims
to let the referent speak for itself. This illusion is not proper to
historical discourse: how many novelists-in the realistic pe
riod-imagine they are being "objective" because they suppress
signs of the I in the discourse! The combination of linguistics
and psychoanalysis has increased our lucidity with regard to a
privative speech-act: we know that the absence of signs has a
meaning. too.
To conclude with the speech-act, we must mention the special
case-anticipated by Jakobson, on the level of language. in the
grid of his shifters-in which the speaker (or writer) of the
discourse is at the same time a participant in the process spoken
(or written), in which the protagonist of the text is the same as
the protagonist of the speech-act (Te I 'P), in which the historian,
an actor at the time of the event, becomes its narrator; thus.
Xenophon participates in the retreat of the Ten Thousand and
becomes their historian after the fact. The most illustrious
example of this conjunction of the spoken I and the speaking I
is doubtless the M of Julius Caesar. This famous M belongs to
the statement; when Caesar becomes explicitly the "writer," he
shifts to we (ut supra demorutravimus). The Caesarian M seems at
first glance swamped among the other participants of the spoken
process and. on this account, we have seen it as the supreme
sign of objectivity; it seems, however, that we can formally
differentiate it; how? by observing that its predicates are con
sistently selected: the Caesarian M supports only certain syn
tagms which we might call syntagms of the leader (to give orders, to

B
The DiscOUT5e of History 1 33

hold meetings. to visit. to have done. to cGngTatuialt. to explain. to

thinlt). quite close, as a matter of fact, to certain performatives.
in which speech is identified with action. There are other
examples of this he. past-tense actor and present-tense narrator
(notably in Clausewitz): they show that the choice of the a
personal pronoun is merely a rhetorical alibi and that the true
situation of the "writer" is manifested in the choice of syntagms
with which he surrounds his past actions.

2. Statement
The historical statement must lend itself to a figuration destined
to produce units of content, which we can subsequently classify.
These units of content represent what history 5fHaiu about; as
signifieds. they are neither pure referent nor complete dis
course: their totality is constituted by the referent discerned,
named, alady inteUigible. but not yet subjected to a syntax.
We shall not undertake to explore these classes of units here,
such an effort would be premature; we shall limit ourselves to
a few preliminary remarks.
Historical statement, like sentential statement. includes "ex
istents" and "occurrents," beings. entities. and their predicates.
Now. a first inspection suggests that the former and the latter
(separately) can constitute relatively closed, consequently con
trollable lists, in a word, coUtctions whose units ultimately repeat
themselves in obviously variable combinations; thus, in Her4r
dotus, existents are reduced to dynastUs, princes. generols. 50ldien.
fHofJus, and places. and occurrents to actions such as to devastate.
to subjugate, to alliances. to ma4t an txfHdition. to Trip. to employ
a stratagem. to con.sult tM oracle. etc. These collections. being
(relatively) closed, must be accessible to certain rules of substi
tution and transformation. and it must be possible to structure
them-a more or less easy task. obviously, depending largely
on a single lexicon. that of warfare; we must determine whether,
in modern historians, we must expect more complex associations
of different lexicons, and if, even in that case, historical discourse

Malena :>m d reitl al

'34 From History to Reality
is nOl always based, finally, on "strong" collections (beneT to
speak of colucliJms, nOl lexicons, for we are here uniquely on the
level of coment). Machiavelli seems to have had an intuition of
this structure: at the beginning of his History of Flmmce, he
presents his "collection." i.e., the list ofjuridical, political, ethnic
objects which will subsequently be mobilized and combined in
his narration.
I n the case of more fluid collections (in historians less archaic
than Herodotus), the units of content can still receive a strong
structuration. not from the lexicon, but from the author's
personal thematics; such thematic (recurrent) objects are nu
merous in a romantic historian like Michelet; but we can quite
easily find them in so-called intellectual authors: in Tacitus,
fama is a personal unit, and Machiavelli bases his history on a
thematic opposition, that of mantenere (a verb which refers to
the fundamental energy of the man of government) and ruinare
(which, on the contrary. implies a logic of the decadence of
things). It follows that, by these thematic units, generally con
fined in a single word, we discover units of discourse (and no
longer of content alone); here we touch on the problem of the
nomination of historical objects: the word can economize a
situation or a series of actions; it favors structuration insofar as,
projected into content, it is itself a little structure; thus, Machia
velli employs conspiracy to economize explicitation of a complex
datum designating the only remaining possibility of struggle
when a government triumphs over all openly declared enmities.
Nomination, by permitting a strong articulation of the discourse,
reinforces its structure; strongly structured histories are sub
stantive histories: Bossuet, for whom the history of human
beings is structured by God, makes abundant use of successions
of substantive shortcuts.

EXillmple: Here we see the innocen,e iIInd the wisdom of young Joseph
. . . his mY'teriolU dreillms . . . his jealous brothen . . . the Jelling or this gTfilt
man . . . the 10Yilllty he maintained 10 his maSter . . . his iIIdmirable ,hutity: the
persecutions it drew upon him: his prison iIInd his cOO$tancy . ."-Bouuel.
.

DUcoun 511,. I'hisloirt IInivnulu

Malena m dlreitl
T DisCOUTSt of History 1 35

Thse remarks concrn the occurrents as much as th

existents, The historical processes themselves (whatevr
their terminological development) raise--among others this
interesting problem: that of their Statw. The status of a
process can be assertive. negative, interrogative. Now, the
status of historical discourse is uniformly assertive. consta
tive; historical fact is linguistically linked to a privilge of
being: one recounts what has been, not what has not been or
what has been questionable. In a word, historical discourse
does not know negation (or very rarely, in an eccentric fash
ion). This fact may be curiously-but significantly-related to
the arrangement we find in a "writer" quit different from
the historian, the psychotic, who is incapable of subjecting a
statement to a negative transformation; we might say that. in
a sense, "objective" discourse (this is the case of positivi!l
history) joins the situation of schizophrenic discourse; in
either case, there is a radical censorship of the speech-act (in
which feeling alone permits a negative transformation), a mas
sive reflux of discourse toward statement and even (in the
historian's case) toward the referent: no one is thre to assume
the statement.
In order to approach another, essential aspect of historical
statement. we must say a word about the classes of units of
content and about their succession. These classes are, as is
apparent from a first exploration. the very ones we supposed
we could discover in the narrativ of fiction. Th first class
covers all the segments of discourse which refer to an implicit
signified, according to a metaphoric process; thus. Michelet
describes the motley of garments, .the fading of blazons, and
th mixture of arcitectural styles at the beginning of the
fifteenth century as so many signifiers of a single signified,
which is the moral division of the waning Middle Ages; this
class is that of indices, or more precisely of signs (a very abundant
class in the classical novel). The second class of units is constituted
by the fragments of discourse of a reasoning, syllogistic, or more
i almost always imper-
precisely enthymematic nature, since it s

Matena dlreitl
 From History to Reality
feet, approximative syllogisms which are involved. Enthymemes
are not proper to historical discourse; they are frequent in the
novel. where bifurcations of the anecdote are generally justified
in the reader's eyes by pseudo-reasonings of syllogistic type.
The enthymeme arranges, in historical discourse, a non-symbolic
intelligibility, and this is what is interesting: does it subsist in
recent histories, whose discourse attempts to break with the
classical, Aristotelian model? Last, a third class of units-and
not the least-receives what since Propp we have called the
"functions" of the narrative, or cardinal points from which the
anecdote can take a different course; these functions are grouped
syntagmatically into closed, logically saturated series or se
quences; thus, in Herodotus, we frequently find a sequence
Oracle, composed of three terms, each of which is an alternative
(to consult or not, to answer or not, to follow or not), and which
can be separated from each other by units foreign to the
sequence: these units are either the terms of another sequence
and then the schema is one of imbrication--or else minor
expansions (times of information, indices}-and then the schema
s
i one of a catalysis which fills the interstices of the nuclei.
By generalizing-perhaps abusively-these few remarks on
the structure of statements, we can suggest that historical
discourse oscillates between two poles, according to the respec
tive density of its indices and its functions. When, in a historian's
work, indicial units predominate (constantly referring to an
implicit signified), the History is inflected toward a metaphorical
form, and borders on the lyric and the symbolic: this s
i the case,
for instance, with Michelet. When on the contrary it is functional
units which prevail, the History takes a metonymic form, it is
related to the epic: we might cite as a pure example of this
tendency the narrative history of Augustin Thierry. A third
History, it is true, exists: one which, by the structure of its
discourse, attempts to reproduce the structure of the choices

Here is the syllogistK: IChema of a passage in Michelet (HiJlDiu de Mrrym

Age. Vol. Ill. Book VI. chapter II): I. In order to distract the people from
rebellion, they must be: kept occupied. t. Now. the best means is 10 throw them
a man. . Hence. the princes chose old Aubriot, etC.
The Discourse of History
experienced by the protagonists of the process related; in it the
reasonings dominate; this is a reflexive history. which we can
also call a strategic history. and Machiavelli is the best example
of it we k.now.

3. SlgnlflcaHon
For history not to signify, discourse must be limited to a pure
unstructured series of notations: these will be chronicles and
annals (in the pure sense of the word). In constituted historical
discourse. the facts related irresistibly function either as indices
or as nuclei whose very succession has an indicial value; and
even though facts are presented in an anarchic manner, they at
least signify anarchy and refer to a certain negative idea of
human history.
The signifieds of historical discourse can occupy at least two
different levels. There is, first of all. a level immanent to the
material stated; this level retains all the meanings the historian
deliberately gives to the facts he reports (the motley of fifteenth
century garments for Michelet, the importance of certain con
flicts for Thucydides, etc.); such can be the moraJ or political
"lessons" the narrator draws from certain episodes (in Machia
velli, in B055uet). If the "lesson" is continuous, we reach a second
level, that of a signified transcending the entire historical
discourse, transmitted by the historian's thematics. which we are
thereby entitled to identify with the form of the signified; thus.
the very imperfection of Herodotus's narrative structure (gen
erated by certain series of facts withoUi closure) ultimately refers
to a certain philosophy of History, which is the accessibility of
the world of men under the law of the gods; thus again. in
Michelet, the very "strong" structuration of particular signifieds,
articulated in oppositions (antitheses on the level of the signifier),
has as its ultimate meaning a Manichaeistic philosophy of life
and death. In the historical discourse of our civilization, the
process of signification always aims at "filling" the meaning of
History: the historian is the one who collects not so much
facts as signifiers and relates them, i.e.. organizes them in

Malena :>m d reitl al

1 38 From History to Reality
order to eSlablish a positive meaning and to fill the void of
pure senes.

As we see, by its very structure and without there being any

need to appeal to the subslance of the content, historical
discourse is essentially an ideological elaboration or, to be more
specific, an imaginary elaboration, if it is true that the image
repertoire is the language by which the speaker (or "writer") of
a discourse (a purely linguistic entity) "fills" the subject of the
speech-act (a psychological or ideological entity). Hence, we
understand why the notion of historical "fact" has so often given
rise to a cerlain mistrust. Nietzsche has written: "There are no
facts as such. We must always begin by introducing a meaning
in order for there to be a fact." Once language intervenes (and
when does it not intervene?), a fact can be defined only tauto
logically: the noted issues from the notable, but the notable is
since Herodotus, where the word loses its mythic acceptation
only what is wonhy of memory, i.e., wonhy to be noted. Hence,
we arrive at that paradox which governs the entire pertinence
of historical discourse (in relation to other types of discourse):
fact never has any but a linguistic existence (as the term of
discourse), yet everything happens as if this linguistic existence
were merely a pure and simple "copy" of another existence,
situated in an extra-structural field, the "real." This discourse
is doubtless the only one in which the referent is addressed
as external to the discourse, though without its ever being
possible to reach it outside this discourse. Hence, we must
inquire more closely into the place of the "real" in discursive
structure.
Historical discourse supposes, one might say, a double oper
ation, one that is extremely complex. In a first phase (this
decomposition is. of course, only melaphorical), the referent is
detached from the discourse, it becomes exterior to it, grounds
it, is supposed to govern it: this is the phase of res gestae, and
the discourse simply claims to be hi.!toria rerum gestarum: but in
a second phase, it is the signified itself which is repulsed, merged
in the referent; the referent enters into direct relation with the
signifier, and the discourse, meant only to express the real,

,
Tiu DiscouNt of History '39
believes it elides the fundamental term of imaginary structures,
which is the signified. Like any discourse with "realistic" claims,
the discourse of history thus believes it knows only a two-term
semantic schema, referent and signifier; the (illusory) merging
of referent and signified defines, as we know, sui-rtf"mtial
discourses (such as performative discourse); we can say that
historical discourse is a fake performative discourse in which
the apparent constative (descriptive) is in fact only the signifier
of the speh-act as an act of authority.
In other words. in "objective" history, the "real" is never
anything but an unformulated signified, sheltered behind the
apparent omnipotence of the referent. This situation defines
what we might call the reality tffect. The extrusion of the signified
outside the "objective" discourse. letting the "real" and its
expression apparent1y confront each other, does not fail to
produce a new meaning. so true is it, once more, that within a
system any absence of an element s
i itself a signification. This
new meaningxtensive to all historical discourse and ultimately
defining its peninence-is reality itself, surreptitiously trans
formed into a "shamefaced" signifier: historical discourse does
not follow the real. it merely signifies it, constantly repeating
this happnud. without this assertion ever being anything but the
signified wrong .ride of all historical narration.
The prenige of thi.! htJppemd has a truly hinorical importance
and scope. Our entire civilization has a taste for the reality
effect. attested to by the development of specific genres such as
the realistic novel, the private diary, documentary literature.
the news item [fait divm], the historical museum, the exhibition
of ancient objects, and, above all. the massive development of
photography, whose sole peninent feature (in relation to draw
ing) is precisely to signify that the event represented has really
taken place. Secularized, the relic no longer has anything sacred
about it, except that sacred quality attached to the enigma of

Thien has expreued. with great purity and nai\'ete, this referential illusion,
or this merging of rererent and signified, by thul defining the historian's ideal:
MTo be simply true, to be what thing1 are and nothing more than that, and
nothing except that."

B
 From History to Reality
what has been, is no more, and yet offers itself as present sign
of a dead thing. Conversely, the profanation of rdics is in fact
a destruction of reality itself, starting from the intuition that
the real is never anything but a meaning, revocable when history
requires it and demands a veritable destruction of the very
foundations of civilization.-
Since it refuses to assume the real as a signified (or even to
detach the referent from its simple assertion), it is understand
able that history, at the privileged moment when it attempted
to constitute itself as a genre. i.e., in the nineteenth century,
should have come to see in the "pure and simple" relation of
facts the best proof of these facts, and to institute narration as
a privileged signifier of the real. Augustin Thierry made himself
the theoretician of this narrative history, drawing its "truth"
from the very solicitude of its narration, the architecture of its
articulations. and the abundance of its expansions (called, in
this case, "concrete details").t
Thus, we dose the paradoxical circle: narrative structure,
elaborated in the crucible of fictions (through myths and early
epics), becomes both sign and proof of reality. Hence, it will be
understood that the effacement (if not the disappearance) of
narration in contemporary historical science, which prefers to
speak of structures rather than of chronologies, implies much
more than a simple change of school: a veritable ideological
transformation; historical narration is dying because the sign of
History is henceforth not so much the real as the inteUigible.
InfOTTMtion sur UJ sciences soOaUJ, 1967

This is doubtless the meaning, beyond any striedy religious subversion,

which we must give to the act of the Red Guards profaning the temple at
Confucius's birthplace Uanuary 1967): let us recall that the exprelSion ndtural
rftlOllUitm is a very inadequate ttanslatton of Rdestruction of the foundations of
civilization."
t "It has been said that the historian's goal was to recount, nO{ to prove; I do
not know, but I am ceruin that in history the bnt proof, the kind most capable
of arousing and convincing all minds, the kind which pcrmitl the least resisunce
and leaves the fewest doubts, is complete narration . . . "-Augustin Thierry,
RkiU IUS lertIfIJ IIIbwingUru, Vol. II (Paris, 1851)

,
 The Reality Effect

When Flaubert. describing the room occupied by Mme Auhain.

Felicite's employer. tells us that "an old piano supported, under
a barometer, a pyramidal heap orOOxe! and cartons" ("A Simple
Heart," from Three Tales); when Michelet, recounting the death
of Charlotte Corday and reporting that, before the executioner's
arrival. she was visited in prison by an artist who painted her
portrait, includes the detail that "after an hour and a half. there
was a gentle knock at a little door behind her" (Histoire de France:
La Rivolutionhese authors (among many others) are produc
ing notations which structural analysis, concerned with identi
fying and systematizing the major articulations of narrative,
usually and heretofore has left out, either because its inventory
omits all details that are "superfluous" (in relation to structure)
or because these same details are treated as "filling" (catalyses),
assigned an indirect functional value insofar as, cumulatively,
they constitute some index of character or atmosphere and so
can ultimately be recuperated by structure.
It would seem, however, that if analysis seeks to be exhaustive
(and what would any method be worth which did not account
for the totality of its object, i.e., in this case, of the entire surface
of the narrative fabric?), if it seeks to encompass the absolute
detail, the indivisible unit, the fugitive transition, in order to
assign them a place in the structure. it inevitably encounters
notations which no function (not even the most indirect) can
justify: such notations are scandalous (from the point of view
of structure), or, what is even more disturbing, they seem to
correspond to a kind of narrative luxury, lavish to the point of
offering many "futile" details and thereby increasing the cost of
narrative information. For if, in Flauben's description, it is JUSt

'4 '

,
 From History to ReaLity
possible to see in the notation of the piano an indication of its
owner's bourgeois standing and in that of the cartons a sign of
disorder and a kind of lapse in status likely to connOle the
atmosphere of the Aubain household, no purpose seems to
justify reference to the barometer, an object neither incongruous
nor significant, and therefore not participating, at first glance,
in the order of the notable; and in Michelet's sentence, we have
the same difficulty in accounting structurally for all lhe details:
that the ex.ecutioner came after the painter is all that is necessary
to the account; how long the sitting lasted, the dimension and
location of the door are useless (but the theme of the door, the
softness of death's knock have an indisputable symbolic value).
Even if they are not numerous, the "useless details" therefore
seem inevitable: every narrative, at least every Western narrative
of the ordinary sort nowadays, possesses a certain number.
Insignificant notation- (taking this word in its stong sense:
apparently detached from the narrative's semiotic structure) is
related to description, even if the object seems to be denoted
only by a single word (in reality, the "pure" word does not exist:
Flaubert's barometer is not cited in isolation; it is located, placed
in a syntagm at once referential and syntactic) ; thus is underlined
the enigmatic character of all description, about which a word
is necessary: the general structure of narrative, at least as it has
been occasionally analyzed till now, appears as essentially pre
dictive; schematizing to the ex.treme, and without taking into
account numerous detours, delays. reversals. and disappoint
ments which narrative institutionally imposes upon this schema,
we can say that, at each articulation of the narrative syntagm,
someone says to the hero (or to the reader, it does not matter
which): if you act in this way, if you choose this alternative, this
is what will happen (the reported character of these predictions
does not call into question their practical nature). Description

In this brief account, Wf' shall nOl. giVf' f'xampl.,s of "insignificant" notations,
for thf' insignificant can I::c revf'alal only on thf' If',d of an immf'nse StruClUrf':
once alai, a notion is nf'ithf'r significant nor insignificant; it requires an altf'ady
analyzed COntf'xt.

C JPYnghied mater1
TM Reality Effect '43
i entirely different:
s it has no predictive mark; "analogical," its
structure is purely summatory and does not contain that trajec
tory of choices and alternatives which gives narration the
appearance of a huge traffic-control center. furnished with a
referential (and not merely discursive) tempora1ity. This is an
opposition which, anthropologically, has its importance: when,
under the inAuence of von Frisch's experiments, it was assumed
l e these
that bees had a language, it had to be realized that, whi
insects possessed a predictive system of dances (in order to
collect their food), nothing in it approached a description. Thus,
description appears as a kind of characteristic of the so-called
higher languages. to the apparently paradoxical degree that it
is justified by no finality of action or of communication. The
singularity of description (or of the "useless detail") in narrative
fabric, its isolated situation. designates a question which has the
greatest imponance for the structural analysis of narrative. This
question is the following: Is everything in narrative significant.
and if not. if insignificant stretches subsist in the narrative
syntagm. what is ultimately, so to speak. the significance of this
insignificance?
First of all. we must recall that Western culture. in one of its
major currents, has certainly not left description outside mean
ing. and has furnished it with a finality quite "recognized" by
the literary institution. This current is Rhetoric. and this finality
is that of the "beautiful": description has long had an aesthetic
function. Very early in antiquity, to the two expressly functional
genres of discourse, legal and political. was added a third. the
epideictic, a ceremonial discourse intended to excite the admi
ration of the audience (and no longer to persuade it); this
discourse contained in germ-whatever the ritual rules of its
use: eulogy or obituary-the very idea of an aesthetic finality of
language; in the Alexandrian neo-rhetoric of the second century
A.D., there was a craze for ecphrasis. the detachable set piece
(thus having its end in itself. independent of any general
function). whose object was to describe places. times. people. or
works of art. a tradition which was maintained throughout the

C JPYnghied mater1
'44 From History 10 Reality
Middle Ages. As Cuniu5 has emphasized. description in this
period is constrained by no realism; its truth is unimportant (or
even its verisimilitude); there is no hesitation to put lions or
olive trees in a nonhern country; only the constraint of the
descriptive genre counts; plausibility is not referential here but
openly discursive: it is the generic rules of discourse which lay
down the law.
Moving ahead to Flaubert. we see that the aesthetic purpose
of description is still very strong. In Madame Bovary, the descrip-.
tion of Rouen (a real referent if ever there was one) is subject
to the tyrannical constraints of what we must call aesthetic
verisimilitude, as is attested by the corrections made in this
passage in the course of six successive rewritings. Here we see,
first of ali, that the corrections do not in any way issue from a
doserconsideration ofthe model: Rouen, perceived by Flaubert,
remains just the same, or more precisely, if it changes somewhat
from one version to the next, it is solely because he finds it
necessary to focus an image or avoid a phonic redundance
condemned by the rules of le beau 5tyle, or again to "arrange" a
quite contingent felicity of expression; next we see that the
descriptive fabric, which at first glance seems to grant a major
importance (by its dimension, by the concern for its detail) to
the object Rouen, is in fact only a sort of setting meant to receive
the jewels of a number of rare metaphors, the neutral. prosaic
excipient which swathes the precious symbolic substance, as if,
in Rouen, all that mattered were the figures of rhetoric to which
the sight of the city lends itself-as if Rouen were notable only
by its substitutions (the masts liM a forest of netdie5, the islands liM
hugt motionless black [ISh, the clouds like aerial waves silently breaking
against a cliff ); last, we 5tt that the whole description is construckd
so as to connect Rouen to a painting: it is a painted scene which
the language takes up ("Thus, seen from above, the whole

A mechanism distinguished by Valery, in LiltiTalllrt. commenting on Bau

delaire's line "LD strVrlnte au grand CDftIr . . .": 'This line calM to Baudelaire . . .
And Baudelaire continued. He buried the cook out on the lawn, which goes
against the custom, but goes with the rhyme etc. ,"

GJPYnghtcd maknal
The Rtality Elltet 145

landsca had the motionless look of a painting"); the writer

here fulfills Plato's definition of the artist as a maker in the
third degree, since he imitates what s
i already the simulation of
an esnce. Thus, although the description of Rouen is quite
irrelevant to the narrative structure of Madame Bovary (we can
attach it to no functional sequence nor to any characterial,
atmospheric, or sapiential signified), it is not in me least scan
dalous, it is justified. if nm by the work's logic, at least by the
laws of literature: its "meaning" exists. it depends on conformity
not to the model but to the cultural rules of representation.
All the same. the aesmetic goal of Flaubertian description is
thoroughly mixed with "realistic" imratives, as if the referent's
exactitude. superior or indifferent to any other function, gov
erned and alone justified its description, or-in the case of
descriptions reduced to a single word-its denotation: here
aesthetic constraints are steeped-at least as an alibi-in refer
ential constraints: it is likely that. if one came to Rouen in a
diligence, the view one would have coming down the slope
leading to the town would nm be "objectively" different from
the panorama Flaubert describes. This mixture-this interweav
ing-of constraints has a double advantage: on the one hand.
aesthetic function, giving a meaning to "the fragment," halts
what we might call me vertigo of notation; for once, discourse
is no longer guided and limited by structural imperatives of the
anecdme (functions and indices), nothing could indicate why
we should halt the details of the description here and not mere;
if it were not subject to an aesthetic or rhetorical choice. any
"view" would be inexhaustible by discourse: there would always
be a corner. a detail. an inflection of space or color to report;
on the mher hand, by positing the referential as real. by
pretending to follow it in a submissive fashion. realistic descrip
tion avoids being reduced to fantasmatic activity (a precaution
which was supposed necessary to the "objectivity" of the ac
count); classical rhetoric had in a n institutionalized the
fantasmatic as a specific figure. hypotyposis. whose function was
to "put things before the hearer's eyes," not in a neutral,

Malena :>m d reitl al

 From HUlory 10 Rea/it]

constative manner, but by imparting to representation all the

luster of desire (this was the vividly illuminated sector of
discourse, with prismatic outlines: illuslru orato
i ); declaratively
renouncing the constraints of the rhetorical code, realism must
seek a new reason to describe.
The irreducible residues of functional analysis have this in
common: they denote what is ordinarily called "concrete reality"
(insignificant gestures, transitory attitudes, insignificant objects,
redundant words). The pure and simple "representation" of
the "real," the naked relation of "what is" (or has been) thus
appears as a resistance to meaning; this resistance confirms the
great mythic opposition of the true-to-life (the lifelike) and the
intelligible; it suffices to recall that, in the ideology of our time,
obsessive reference to the "concrete" (in what is rhetorically
demanded of the human sciences, of literature, of behavior) is
always brandished like a weapon against meaning, as if, by some
statutory exclusion, what is alive cannot not signify-and vice
versa. Resistance of the "real" (in its written form, of course) to
structure is very limited in the fictive account, constructed by
definition on a model which, for its main outlines, has no other
constraints than those of intelligibility; but this same "reality"
becomes the essential reference in historical narrative, which is
supposed to report "what really happened": what does the non
functionality of a detail matter then, once it denotes "what took
place"; "concrete reality" becomes the sufficient justification for
speaking. History (historical discourse: historia rerum gestarum) is
in fact the model of those narratives which consent to fill in the
interstices of their functions by structurally superfluous nota
tions, and it is logical that literary realism should have been
give or take a few decades-<:ontemporary with the regnum of
"objective" history, to which must be added the contemporary
development of techniques, of works, and institutions based on
the incessant need to authenticate the "real": the photograph
(immediate witness of "what was here"), reportage, exhibitions
of ancient objects (the success of the Tutankhamen show makes
this quite dear), the tourism of monuments and historical sites.
The Reality Effect 147

All this shows that the "real" is supposed to be self-sufficient,

that it is strong enough to belie any notion of "function," that
its "speech-act" has no need to be integrated into a structure
and that the having-bun-there of things is a sufficient principle
of speech.
Since antiquity, the "real" has been on History's side; but this
was to help it oppose .the "lifelike," the "plausible." to oppose
the very order of narrative (of imitation or "poetry"). All classical
culture lived for centuries on the notion that reality could in no
way contaminate verisimilitude; first of all, because verisimilitude
is never anything but opinable: it is entirely subject to (public)
opinion; as Nicole said: "One must not consider things as they
are in themselves. nor as they are known to be by one who
speaks or writes. but only in relation to what s
i known of r.hem
by those who read or hear"; then. because History was thought
to be general, not panicular (whence the propensity, in classical
texts, to functionalize all details, to produce strong structures
and to justify no notation by the mere guarantee of "reality");
finally, because. in verisimilitude. the contrary is never impos
sible, since notation rests on a majority, but not an absolute,
opinion. The motto implicit on the threshold of all classical
discourse (subject to the ancient idea of verisimilitude) is: EJW
(Let tlure be, Jupprue . . . ) "Real," fragmented, interstitial notation,
the kind we are dealing with here, renounces this implicit
introduction, and it is free of any such postulation that occurs
in the structural fabric. Hence, there is a break between the
ancient mode of versimilitude and modern realism; but hence,
too, a new verisimilitude is born, which is precisely realism (by
which we mean any discourse which accepts "speech-acts" jus
tified by their referent alone).
Semiotically, the "concrete detail" is constituted by the dirtct
collusion of a referent and a signifier; the signified is expelled
from the sign. and with it, of course, the possibility of developing
a form of tht lignifud, i.e., narrative structure itself. (Realistic
literature is narrative, of course, but that is because its realism
is only fragmentary, erratic. confined to "details." and because

Malena :>m d reitl al

 ham His/ory /0 Reality
the most realistic narrative imaginable develops along unrealistic
lines.) This is what we might call the referential llusion
i . The

truth of this illusion is this: eliminated from the realist speech

act as a signified of denotation, the "real" returns to it as a
signified of connotation; for just when these details are reputed
to dmott the real directly, all that they do-without saying so
is signify it; Flaubert',s barometer. Michelet's little door finally
say nothing but this: we are the real; it is the category of " the
real" (and not its contingent contents) which is then signified;
in other words, the very absence ofthe signified, to the advantage
of the referent alone, becomes the very signifier of realism: the
reality effect is produced, the basis of that unavowed verisimilitude
which forms the aesthetic of all the standard works of modernity.
This new verisimilitude is very different from the old one, for
it is neither a respect for the "laws of the genre" nor even their
mask, but proceeds from the intention to degrade the sign's
tripartite nature in order to make notation the pure encounter
of an object and its ex.pression. The disintegration of the sign
which seems indeed to be modernity's grand affair-is of course
present in the realistic enterprise, but in a somewhat regressive
manner. since it occurs in the name of a referential plenitude,
whereas the goal today is to empty the sign and infinitely to
postpone its object so as to challenge, in a radical fashion, the
age-old aesthetic of "representation."

Communications, 1968

An illu,ion dearly iIIumated by the program Thieuu5igned to the hi5torian:

''To be simply true, to be what things are and nothing more than that, and
nothing except that.M

Malena m dlreitl
 Writing the Event

To describe the event implies that the event has been written.
How can an event be written? What can it mean to say "Writing
the event"? The event of May '68 seems to have been wrinen
in three fashions, thrtt writings, whose polygraphic conjunction
forms, rhaps, its historical originality.

1 . Speech

Every national shock produces a sudden Howering of written

commentary (press, books). This is not what I want to speak of
here. The spoken words of May '68 had original ascts. which
must be emphasized.
I. Radiophonic speech (that oflhe "peripheral" stations) dung
to the event, as it was occurring, in a breathless, dramatic
fashion, imposing the notion that knowledge of present reality
is no longer the business of print but of the spoken word. "Hot"
history. history in the course of being made, is an auditive
history,. and hearing becomes again what it was in the Middle
Ages: not only the first of the senses (ahead of touch and sight),
but the sense which establishes knowledge (as. for Luther, it
established the Christian faith). Nor is this all. The (reporter's)
informative word was so closely involved with the event, with
the very opacity of its present, as to become its immediate and
consubstantial meaning, its way of acceding to an instantaneous
intelligibility; this means that in terms of Westem culture, where
nothing can be perceived without meaning, it was the event

One recalls Itreets filled with motionless people seeing nothing, looking at
nothing, their eyes down , but their ears glued to u.n,iltor radios, thus
representing a new human anatomy.

'49

C :lpynghled malenal
 From History to Rality
itself. The age-old distance between act and discourse, event
and testimony. was reduced; a new dimension of history ap
peared. immediately linked to its discourse. whereas all historical
"science" had the task to acknowledge this distance. in order to
govern it. Not only did radiophonic speech inform the partici
pants as to the very extension of their action (a few yards away
from them). so that the transistor became the bodily appendage,
the auditory prosthesis, the new science-fiction organ of certain
demonstrators, but even, by the compression of time, by the
immediate resonance of the act. it inAected. modified the event;
in short, wrote it: fusion of the sign and its hearing. reversibility
of writing and reading which is sought elsewhere, by that
revolution in writing which modernity is attempting to achieve.
2. The relations of force bet\Oo'een the different groups and
parties engaged in the crisis were essentially spolcen. in the sense
that the tactical or dialectical displacement of these relations
during the days of May occurred through and by (confusion of
the means and of the cause which marks language) the com
munique, the press conference, the declaration, the speech. Not
only did the crisis have its language, but in fact the crisis was
language: it is speech which in a sense molded history, made it
exist like a network of traces, an operative writing, displacing
(it is only stale prejudice that considers speech an illusory activity,
noisy and futile, and set in opposition to actions); the "spoken"
nature of the crisis is all the more visible in that it has had.
strictly speaking, no murderous, irremediable effect (speech is
what can be "corrected"; its rigorous antonym, to the point of
defining it. can only be death).
3. The students' speech so completely overflowed, pouring
out everywhere, written everywhere, that one might define
superficially-but also, perhaps, essentially-the university re
volt as a Taking ofSpech (as we say Taking of the Bastille). It seems

The insistence wilh which il was repealed, on eilher side, Ihat, whalever
happens, a[/nuItJrtU Gn no longer be like /HffTff doubtless translales, negalh'dy,
the fear (OT the hope) thal in fact aj/m.IIards would become /Hfrm: the event
being speech, il can, mythically, cross itself OUt.

C JPYnghied mater1
Writing tAt Evmt 151

in retrospect that the student was a being frustrated of speech;

frustrated but not deprived: by class origin, by vague cultural
practice, the student has the use of language; language is not
unknown to him, he is not (or s
i no longer) afraid of it; the
problem was to assume iu power, its active use. Hence, by a
paradox which s
i only apparent, just when the students' speech
made iu claims in the sole name of content, it actually involved
a profoundly ludic aspect; the student had begun to wield
speech as an activity, a free labor, and not, despite appearances,
as a simple instrument. This activity took different forms,
which correspond perhaps to phases of the student movement
throughout the crisis.
a) "Wild" $peech, based on "invention," consequently encoun
tering quite naturally the "finds" of form, rhelOrical shoncuu,
the delights of formula, in short felicity of expression; very close
to writing, this discourse (which affected public opinion in
tensely) logically assumed the form of inscription; its natural
dimension was the wall, fundamental site of collective writing.
b) "Missionary" speech, conceived in a purely instrumental
fashion, intended to transport "elsewhere" (to factory gates, to
beaches, into the street, etc.) the stereotypes of political culture.
c) "Functionalist" speech, conveying the reform projects, as

signing lo the university a social function, here political, there

economic, and thereby rediscovering some of the watchwords
of a previous technocracy ("adaptation of teaching to society'S
needs," "collectivization of research," primacy of the "result,"
prestige of the "interdisciplinary," "autonomy," "participation,"
etc.).
"Wild" speech was quite rapidly eliminated, embalmed in the
harmless folds of (surrealist) "literature" and the illusions of
"spontaneity"; as writing, it could only be useless (until it became
intolerable) to any form of power, whether possessed or claimed;
the other two kinds remain mixed: a mixture which rather
nicely reproduces the political ambiguity of the student move
ment itself, threatened, in its historical and social situation, by
the dream of a "social-technocracy."

C ;.pvnghted malana
'5' From History to Reality

2. Symbol
There was no lack of symbols in this crisis, as was often remarked;
they were produced and consumed with great energy; and
above all, a striking phenomenon, they were swtaintd by a
general. shared willingness. The paradigm of the three Rags
(red I black J tricolor), with its pertinent associations of terms
(red and black against tricolor, red and tricolor against black),
was "spoken" (Hags raised, brandished, taken down, invoked,
etc.) by everyone, or just about: a fine agrmenl, if not as to
the symbols, at least as to the symbolic system itself (which, as
such. should be the final target of a Western revolution). The
same symbolic avatar for the barricade: itself the symbol of
revolutionary Paris, and itself a significant site of an entire
network of other symbols. Complete emblem, the barricade
made it possible to irritate and unmask other symbols; that of
property, for example, henceforth lodged, for the French, in
the fact that it appeared much more in the car than in the
house. Other symbols were mobilized: monument (Bourse,
Odeon), demonstration, occupation, garment, and of course
language, in its most coded (i.e., symbolic, ritual) aspects. This
inventory of symbols should be made; not so much because it is
likely to produce a very eloquent list (this is improbable, despite
or because of the "spontaneity" which presided over their libera
tion), but beca.use the symbolic system under which an event
functions is closely linked to the degree ofthis event's integration
within the society of which it is both the expression and the viola
tion: a symbolic field is not only ajunction (or an antagonism) of
symbols; it is also formed by a homogeneous set of rules, a com
monly acknowledged recourse to these rules. A kind of almost
unanimous adherencet to one and the same symbolic discourse

.. For instance: lxkon or rvolutionary work (oommittccs," "oommissions,"

"motions," "poinu or order," etc.), ritual of communication (ndperson
singular rorms, first names, IC.).
t The most important aspect of this invntory would ultimatly be: to discovr
how each group playd or did nox play the symbo!k game: reja:tion of th (rd
or black) flag, refusal or the barricad, etc.

C JPYnghied mater1
Writing the Event '55
ems to have finally marked partisans and adversaries of the
contestation: almost all played the same symbolic game.

3. Violence
Violence, which in modern mythology is linked, as if it followed
quite naturally, with spontaneity and effectiveness violence,
symbolized here concretely, then verbally, by "the street," site
of released speech, of free contact, counterintellectual space,
opposition of the immediate to the possible ruses of all media
tion-violence s
i a writing: it is (a Derridian theme) the trace in
its profoundest gesture. Writing (if we no longer identify it with
style or with literature) is itself violent. It s
i . in fact. the violence
of writing that parates it from speech, reveals the force of
inscription in it, the weight of an irreversible trace. Indeed, this
writing of violence (an eminently collective writing) possesses a
code; however one decides to account for it. tactical or psy
choanalytic. violence implies a language of violence. i.e., of signs
(operations or pulsions) repeated, combined into figures (actions
or complexes), in shon. a system. Let us take advantage of this
to repeat that the presence (or the postulation) of a code does
not intellectualize the event (contrary to what anti-intellectualist
mythology constantly states): the intelligible is not the intellec
tual.

Such at first glance are the orientations that a description of

the traces which constitute the event might take. Yet such a
description risks being inert if we do not attach it. from the
start, to two postulates whose bearing is still polemical.
The first consists in rigorously separating, according to Der
rida's proposition, the concepts of speech and of writing. Speech
is not only what is actually spoken but also what is transcribed
(or rather transliterated) from oral expression, and which can
very well be printed (or mimeographed); linked to the body, to
the person, to the will-to-seize, it is the very voice of any
"revendication," but not necessarily of the revoluL;ion. Writing
is integrally "what is to be invented," the dizzying break with

C :lpynghled malenal
1 54 From History to Reality

the old symbolic system, the mutation of a whole range of

language. Which is to say, on the one hand, that writing (as we
understand it here, which has nothing to do with "style" or even
literature) is not at all a bourgeois phenomenon (what this class
elaborated was, in fact, a printed speech), and, on the other,
that the present event can only furnish marginal fragments of
writing, which as we saw were not necessarily printed; we will
regard as suspect any eviction of writing, any systematic primacy
of speech, because, whatever the revolutionary alibi, both tend
to preserve the old symbolic system and refuse to link its revo
lution to that of society.
The second postulate consists in not expecting written de
scription to afford a "decoding." Considering the event from
the viewpoint of whatever symbolic mutation it can imply means,
first of all, breaking as much as possible (this is not easy, it
requires the sort of continuous labor begun in various quarters,
it must be recalled, some years ago) with the system of meaning
which the event, if it seeks to be revolutionary, must call into
question. The critical aspect of the old system is interpretation,
i.e., the operation by which one assigns to a set of confused or
even contradictory appearances a unitary structure, a deep
meaning, a "veritable" explanation. Hence, interpretation must
gradually give way to a new discourse, whose goal is not the
revelation of a unique and "true" structure but the establishment
of an interplay of multiple structures: an establishment itself
written, i.e., uncoupled from the truth of speech; more precisely,
it is the relations which organize these concomitant structures,
subject to still unknown rules, which must constitute the object
of a new theory.

C"",municali<mS, 1 968
THE LOVER OF SIGNS


Copyrigh,cd moterial
 Revelation

The Berliner Ensemble came to France the first time in 1 954.

Some of us who saw the company then had the revelation of a
new system, one which cruelly dated our whole theater. Such
novelty had nothing provocative about it, and did not borrow
the habitual manners of the avant-garde. It was what could be
called a subtle revolution.
That revolution proceeded from the fact that the playwright
(in this case, Brecht himself) regarded as quite compatible
certain values which our theater had always been reluctant to
combine. Brechtian theater, as we know, is intellectual theater, a
practice elaborated from an explicit theory, at once materialist
and semantic. Longing for a political theater enlightened by
Marxism and an art which rigorously governs its signs, how
could we help being dazzled by the work of the Berliner
Ensemble? Another paradox: this political work did not reject
beauty; the merest blue spot, the most discreet substance, a belt
buckle, a gray rag formed, each time, a scene which never
copied painting and yet would not have been possible without
the most refined taste: this theater which insisted on its "com
mitment" was not afraid to be distinguished (a word we should
release from its usual triviality, so as to give it a meaning close
to Brechtian distancing). The product of these two values gen
erated what we can regard as a phenomenon unknown in the
West (perhaps precisely because Brecht had learned it from the
East): a theater without hysteria.
Finally, an ultimate flavor, this intelligent, political theater of
an ascetic sumptuousness was also according, moreover, to
one of Brecht's precepts an amusing theater: no tirades, no
preaching, never, even, that edifying Manichaeanism which

1 57
 The Lover of Signs

commonly opposes, in all political art, the good proletarians to

the wicked bourgeois; but always an unexpected argument, a
social criticism conducted outside the tedium of stereotypes and
mobilizing pleasure's most recent resource: subtlety. A theater
at once revolutionary, signifying, and voluptuous who could
ask for anything better?
This surprising conjunction, however, had nothing magical
about it; it would not have been possible without a material
datum, one which was lacking and which is lacking still in
our theater. For a long time there has prevailed among us,
inherited from a spiritualist tradition which Copeau perfectly
symbolized, the convenient conviction that one can have excel
lent theater without money: poverty of means thus became a
"sublime" value, converted actors into officiants. Now, Brechtian
theater is an expensive theater, it pays for the unprecedented
attention given to staging, for the elaboration of costumes
whose careful treatment costs infinitely more than the wildest
luxury of our spectacles for the number of rehearsals, for the
actors' professional security, so necessary to their art. This
theater, at once popular and refined, is impossible in a private
economy, where it could be supported by neither the bourgeois
public which makes money nor the petit-bourgeois public which
makes spectators. Behind the success of the Berliner Ensemble,
behind the perfection of its work a thing everyone could
observe we had to see a whole economy, a whole political
determination.
I do not know what the Berliner Ensemble has become since
Brecht's death, but I know that the Berliner Ensemble of 1 954
taught me a great many things and much more than theater.

Le Monde, 1 97 1

 A Magnificent Gift

Jakobson made literature a magnificent gift: he gave it linguistics.

Of course, Literature did not wait to know it was Language; all
of classical Rhetoric until Valery attests to the fact; but once a
science of language was sought (initially in the form of a historical
and comparative linguistics of languages), it was oddly negligent
of the effects of meaning. it too succumbing in that century of
positivism (the nineteenth) to the taboo of specialized realms:
on one side, Science. Reason, FaCl; on the other, Art, nsibility.
Impression.Jakobson was involved since his youth in the amend
ment of this situation: because this linguist has persisted in his
great Jove of poetry, of painting. of cinema, because, at the
heart of his scientific research. he never censored his pleasure
as a cultivated man, he realized mat the authentic scientific
phenomenon of modernity was not fact but relatUmship. At me
origin of the generalized linguistics he outlined was a decisive
opening gesture of classifications, castes, disciplines: these words
lost, with him, their separatist, penal, racist taint; there are no
more owners (of Literature, of Linguistics), the watchdogs are
sent back to their pens.
jakobson endowed Literature in three ways. First of all, he
created within linguistics itself a special department, Potties; this
sector (and this is what is new in his work, his historical
contribution), he did not define from Literature (as if Poetics
stiU depended on the "poetic" or on " poetry") but from the
analysis of the functions of language: every speech-act which
accentuates the form of the message is poetic; therefore, he was
able, startingfrom a linguistic position, to join the vital (and often
the most emancipated) forms of Literature: me right to ambi-

'59

,
 The Lover of Signs
guity of meanings, the system of substitutions, the code of
figures (metaphor and metonymy).
Subsequently. even more strongly than Saussure. he promoted
a pansemiotics. a generalized (and not only general) science of
signs; but here again his position was doubly avantgarde: on
the one hand he maintained a preeminent place in that science
for articulated language (being well aware that language is
everywhere. and not simply close by). and on the other he imme
diately united the realms of Art and Literature to semiotics.
thereby postulating from the start that semiology is a science of
signification-and not of mere communication (thus freeing
linguistics of any risk of technocratic intent).
Finally. his linguistics itself admirably grounds our present
concept of the Text: i.e., that a sign's meaning is only its
translation into another sign. which defines meaning not as a
final signified but as arwther signifying level; and also that the
commonest language involves an important number of meta
linguistic utterances, which attests man's necessity to conceive
his language at the very moment he speaks: a crucial activity
which Literature merely carries to its highest degree of incan
descence.
The very style of his thought, a brilliant. generous. ironic.
expansive. cosmopolitan, Hexible style which we might call
devilishly intelligent, predisposed jakobson to this historical func
tion of opening-of abolishing disciplinary ownership. Another
style is doubtless possible, based at once on a more historical
culture and on a more philosophical notion of the speaking
subject: I am thinking here of the unforgettable (and yet
somewhat forgotten. it seems to me) work of Benveniste, whom
we must never dissociate (and jakobson would agree with me
here) from any homage we pay to the decisive role of Linguistics
in the birth of that other thing operative in our age. ButJakobson,
through all the new and irreversible propositions that constitute
his work of fifty years, is for us that historic figure who, by a
stroke of intelligence, definitively shoved into thepast some highly
respectable things to which we were attached: he converted

B
A Magnlf","" Gift

prejudice into anachronism. All his work. reminds us that "each

of us mUSI realize, once and for all. that the linguist deaf to
poetic function. like the specialist in literature indifferent to the
problems and ignorant of methods of linguistics, is henceforth,
in either case, a flagrant anachronism."

U MONk. '97 '

B
 Why I Love Benveniste

The present emphasis on problems of language irritales some,

who regard it as an excessive fashion. Yet they will have to
resign themselves to the inevitable: we are probably only begin
ning to speak of language: along with other sciences which lend,
today, to be attached to it. linguistics is entering the dawn of its
history: we have yet to discover language, as we are in the
process of discovering space: our century will perhaps be marked
by these two explorations.
Every work of general linguistics therefore answers an im
perious need of today's culture as well as a demand for knowl
edge formulated by every science whose object is to any degree
involved with language. Now, linguistics is difficult to expound,
divided as it is between a necessary specialization and an
explosive anthropological project. Therefore, works of general
linguistics are few in number, at least in French: Martinet's
Ekmmls and Jakobson's Ellays; Hjelmslev's Prolegomena are soon
to be translated. And today there is the work of Benveniste.
This is a collection of articles (normal units of linguistic
research), some of which are already famous (on the arbitrariness
of the sign, on the function of language in Freudian discovery.
on the levels of linguistic analysis). The first group constitute a
description of contemporary linguistics: especially to be rec
ommended here is the splendid anicle on Saussure, who in fact
wrote nothing after his memorandum on Indo-European vowels,
being unable, he believed. to effect that total subversion of past
linguistics he required to construct his own linguistics, and
whose "silence" has the greatness and the bearing of a writer's

.6,

B
Why I Love Bmvmiste
silence. The articles that follow occupy the cardinal poinu of
linguistic space: communication, or again the articulated sign,
situated in relation to thought, to animal language. and to
oneiric language; structure (1 have mentioned the crucial text on
the levels of linguistic analysis: I must also point out the text
fascinating in iu darity-in which Benveniste establishes the
suh-Iogical system of Latin prepositions; if only we had had
such an explanation in the days when we were making our Latin
translations: everything is illuminated by structure); signifutUicn
(for it s
i always from the point of view of meaning that Benveniste
interrogates language); person. to my mind the decisive pan of
the work. in which Benveniste, essentially, analyzes the organi
zation of pronouns and tenses. The work concludes with several
lexical studies.
All of which forms the program of an impeccable scholarship,
answers with darity and power the questions of fact likely to be
raised by anyone with some interest in language. But this is not
all. This book not only satisfies a present demand of culture: it
anticipates it. forms it, directs it. In short, this is not merely an
indispensable book; it is also an important book, an unlooked
for book: it is a very beautiful book.
When the science in which one has specialized is solicited by
the curiosity of amateurs of every kind. it is quite tempting to
defend the specialty of that science rather jealously. Quite the
contrary, Benveniste has the courage deliberately to place lin
guistics at the point of origin of a very wide movement and to
divine in it the future development of a veritable science of
culture. insofar as culture is essentially language: he does not
hesitate to note the birth of a new objectivity, imposed upon
the scholar by the symbolic nature of cultural phenomena; far
from abandoning language on society's doorstep as if it were
merely a tool, he hopefully asserts that "it is society which is
beginning to acknowledge itself as language." Now, it is crucial
for a whole set of investigations and revolutions that a linguist
as rigorous as Benveniste should himself be conscious of his
discipline's powers. and that, refusing to constitute himself as

B
 The Lover of Signs

its owner, he should recognize in it the germ of a new config

uration of the human sciences.
This courage is paired with a profound vision.
this is his success always grasps language at that crucial level
where, without ceasing to be language to the full, it gathers up
everything we were accustomed to consider as external or
anterior to language. Take three contributions, among the most
important: one on the middle voice of Indo-European verbs,
the second on the structure of the personal pronouns, the third
on the system of tenses in French; all three deal in various ways
with a crucial notion in psychology: that of person. Now,
Benveniste magisterially manages to root this notion in a purely
linguistic description. I n a general manner, placing the subject
(in the philosophic sense of the word) in the center of the great
categories of language, showing, on the occasion of extremely
diverse phenomena, that this subject can never be distinguished
from an "instance of discourse," unlike the instance of reality,
Benveniste establishes linguistically, i.e., scientifically, the iden
tity of subject and language, a position at the heart of many
contemporary investigations and of as much interest to philos
ophy as to literature; such analyses may show the way out of an
old antinomy, one that has not yet been liquidated : that of
subjective and objective, of individual and society, of science
and discourse. .
Books of scholarship, of research, have their "style" too. This
one is of a very high order. There is a beauty, an experience of
the intellect, which gives the work of certain scholars that
inexhaustible clarity which also goes into the making of great
literary works. Everything is clear in Benveniste's book, every
thing can be immediately recognized as true; and yet everything
in it is only beginning.

La Quinzaine litte-raire, 1 966

Why I Love Benveniste

Benveniste's place in the concert of great linguists whose influ

ence marks all intellectual work of our times is quite original.
to the point of being, it seems to me, occasionally underesti
mated . Even today, his work is paradoxical twice over: with
regard to the tradition, and with regard to what I shall call the
easy avant-garde, the one which repeats instead of exploring.
What is it, then, that Benveniste tells us? First of all, this: that
language can never be separated from a sociality. This pure
linguist, whose objects of study apparently belong to the appa
ralUS of general, transcendent linguistics, actually never ceases
apprehending language in which we might call its coneomitances:
work, history, culture, institutions; in shon, everything that
constitutes human reality. The Vocabulary of Indo-European Insti
tutions, the studies of agent names, of the verbal prefixes prae
and VOT-, are texts which denature the linguistic discipline,
achieve that subversive moment by which the disciplinary outline
is erased and a new, as yet unnamed, science appears; this is
the moment when linguistics ceases to take a theatrical leadership
and becomes in fact a universal "sociology": the science of a
society which speaks. which is a society precisely because it spealcs.
On this level, Benveniste's work is always critical; a demystifier,
he is tirelessly concerned to reverse scholarly prejudices and
implacably to illuminate (for this man of science is rigorous)
language's social basis. This power Benveniste possesses because
of the precise but today rare, and underrated situation of
his work: he is a linguist of languages, not just of language.
At the other end of the chain (but the hiatus will astonish
only those frivolous minds who impenurbably continue to set
history in opposition to structure). Benveniste has given scientific
body to a notion which has assumed the greatest importance in
the work of the avant-garde: the speech-act. The speech-act
[enoneiation] is not, of course. the statement [enonei]. nor is it (a
much subtler and more revolutionary proposition) the simple
presence of subjectivity in discourse ; it is the renewed act by
166 The Lover of Signs

which the locutor takes possession of the language (appropriates

it, Benveniste says quite accurately): the subject is not anterior
to language; he becomes subject only insofar as he speaks; in

short, there is no "subject " (and consequently no "subjectivity"),

there are only locutors; moreover and this is Benveniste's
incessant reminder there are only interlocutors.
From this point of view, Benveniste considerably widens the
notion of shifter, which jakobson advanced with such brio; he
establishes a new linguistics, which exists nowhere else but in
his work (and, above all, not in Chomsky's): the linguistics of
interlocution, language, and consequently the whole world, is
articulated around this form: I I you. Hence, we understand
Benveniste's insistence on dealing, throughout his work, with
the so-called personal pronouns, with temporality, with diathesis,
with composition (the privileged appropriation of the lexicon).
We also understand why Benveniste could so early establish a
bridge between linguistics and psychoanalysis; why, again, this
specialist in Old Persian was able, without forcing himself, to
understand or at least specifically to keep himself from cen
soring the new investigations of semiology (Metz, Schefer) and
the work of the avant-garde on language. The direct interest of
Benveniste's new book is here: it is the book of the speech-act.
A scholar's intellectual gifts (not what is given to him, but
what he gives us) derive, I am convinced , from a power which
is not only that of knowledge and rigor but also that of writing,
or, to adopt a word whose radical acceptation we now know,
from the speech-act. The language Benveniste appropriates
(since that is his definition of the speech-act) is not altogether
that of ordinary scholars, and this slight displacement is sufficient
to constitute a writing. Benveniste's writing is very difficult to
describe because it is almost neutral; only occasionally does a
word by dint of being accurate, one might say, so much does
accuracy seem to accumulate in him gleam out, delight like a
charm, swept away by a syntax whose measure, proportion, and
exactitude (aU virtues of a cabinetmaker) attest to the pleasure
which this scholar has taken in forming his sentences. Benven-
Why I Love Benveniste
iste's writing thus presents that subtle mixture of expenditure
and reserve which founds the text or, better still, music. Ben
veniste writes silently (is not music an art of intelligent silence?),
the way the greatest musicians play: there is something of
Richter in Benveniste.
Working with him, with his texts (which are never mere
articles), we always recognize the generosity of a man who seems
to listen to the reader and to lend him his intelligence, even in
the most special subjects, the most improbable ones. We read
other linguists (and indeed we must), but we love Benveniste.

La QuinUJine liUbairt, 1974

 Kristeva's Semeiotike

Although recent, semiology already has a history. Derived from

a quite Olympian formulation of Saussure's ("A science is
conceivable which would study the life of signs at the very hean
of social life"), it continues to eXTiment with itself, to split
itself up. to de-situate itself, to enter into that great carnival of
languages Julia Kristeva describes. Its historical role is at present
to be the intruder, the third party, the one who upsets those
exemplary households we hear so much about and which arc
formed, it appears, by History and Revolution, Structuralism
and Reaction, determinism and science. progressisme and the
critique of content. Of this upset. Julia Kristeva's work is today
the ultimate orchestration: it activates its tendencies and supplies
its theory.
Already greatly in Kristeva's d(:bt (and from th(: ginning),
1 hav(:just (:xpc:ri(:nci once more, and this tim(: in iu entirety.
th(: forc(: of h(:r work. Here force means displacement. Julia
Krist(:va chang(:S th(: plac(: of things: sh(: always destroys the
last prejudice. th(: on(: you thought you could reassured by,
could tak(: prid(: in; what she displaces is th(: already-said. [he
dejd-diJ. i.e., the instance of the signified. i.e.. stupidity; what
sh(: subverts s
i authority-the authority of monologic science,
of filiation. Her work is entirely n(:w, exact, not by scientific
puritanism. but cause it takes up the whole of the site it
occupi(:s, fills it exactLy, ohliging, anyon(: who exclud(:s himself
from it to find hims(:lf in a position of T(:sistanc(: or censorship
(this is what we call, with a v(:ry shock(:d (:xpression, t(:TTorism).
Sinc(: I have reach(:d th(: point of spc:aking of a siu of res(:arch.
I(:t rn(: say that for m(: Krist(:va's work constitut(:s this admoni
tion: that we ar(: still going too slowly, that we are wasting tim(:

,68

B
KriskVa's Semeiotike 16g

in "believing." i.e. in repeating and humoring ourselves. that

often a supplement of freedom in a new thought would suffice
to gain years of work. In Julia Kristeva. this supplement is
theoretical. What is theoryr It is neither an abstraction nor a
generalization nor a speculation: it is a reHexivity; it is in some
sense the reversed gaze of a language upon itself (which s i why.
in a society deprived of socialist practice. thereby condemned
to discourse. theoretical discourse is transitorily necessary). It is
in this sense that. for the first time. Julia Kristeva s i giving us a
theory of semiology: Any semiotics must be a crilicism of semiotics.
Such a proposition is not to be taken as a pious and hypocritical
wish ("Let us criticize the semioticians who precede us") but as
the affirmation that. in its very discourse. and not on the level
of a few phrases, the work of semiotic science is shot through
with destructive returns. with countered coexistences. with pro
ductive disfigurations.
The science of languages cannot be Olympian, positive (still
Jess positivist). indifferent. adiaphoric. as Nietzsche says; it is
itself (because it is the language of language) dialogic-a notion
borrowed by Kristeva from Bakhtine, whom she has introduced
to us. The first act of this diaJogism is, for semiotics, to conceive
itself at once and contradictorily as science and as writing
which. I believe, has never been done by any science. except
perhaps by the materialist science of the pre-Socratics, and
which may permit us, let it be said in passing, to escape from
the impasse bourgeois sdmce (spoken) I proktarian sdmce (wriuen.
at least postuJativeJy).
The value of Kristevian discourse is that it is homogeneous
to the theory it enunciates (and this homogeneity is the theory
itself): in it. science is writing. the sign is dialogic. the basis is
destructive; if it seems "difficult" to some, this is precisely because
it is written. Which means what? FiT$t of all. that it asserts and
practices both formalization and its displacement. mathematics
becoming, in short. analogous to the work of dreams (whence
many complaints). Next, that it assumes as theory the terminol
ogical slippage of so-called scientific definitions. Finally. that it

,
'70 Tiu Love'( of Signs
institutes a new type of transmission of knowledge (it is not
knowledge which constitutes the problem. but its transmission):
Kristeva's writing possesses at once a discursivity. a "develop
ment" (we should like to give this word a mc,ciist meaning rather
than a rhewrica! one), and a formulation, a frappe (trace of
shock. and of inscription), a germination; it is a discourse which
functions not so much because it "represents" a thought as
because, immediately, without the mediation of dim ecrivana
(inauthentic writing), it produces thought and aims it. This
means that only Julia Kristna is able to produce semio-analysis:
her discou is nOl propaedeutic, it does not offer the possibility
of a "teaching"; but this also means, conversely, that this
discourse transforms us, displaces us, gives us words, meanin,
sentences which permit us to work and to release in ourselves
the creative movement itself: permutation.
In shon, what Julia Kristeva produces is a critique of com

munication (the first, I believe, since that of psychoanalysis).

Communication, she shows. the darling of the positive sciences
(such as linguistics). of the philosophies and the politics of
"dialogue," of "participation," and of "exchange"--<ommuni
cation is mtTchandist. Are we not constantly told that a "clear"
book sells beuer, that a communicative temperament more easily
finds a job? Hence, it is a political task, the very one Kristeva is
performing, to undertake to reduce communication theoretically
to the mercantile level of human relations and to integrate it,
as a simple Ructuating level, to significance. to the Text, an
apparatus outside of meaning. victorious affirmation of Ex
penditure over Exchange, of Numbers over Reckoning.
Will all of this "get anywhere"? That depends on French
inculluf"t, which today seems to be gent.ly lapping, rising around
us. Why? For political reasons, no doubt; but these reasons seem
to affect precisely those who should resist them best: there is a
petty nationalism of the French intelligentsia; one that does not
bear on nationalities. of course (is not Ionesco. after all, the
Pure and Perfect French Petit-Bourgeois?), but on the stubborn
rejection of the othn language. The other language is the one

B
Kristtva',s Semeiotike '7'

spoken from a politically and ideologically uninhabitable place:

an interstitial site, oblique, on the edge-the verge of irregular
action: a cawlin' site, since it traverses, straddles, panoramizes,
and offends. The woman to whom we owe a new knowledge,
from the East and from the Far East, and new instruments of
analysis and commitment (paragram, dialogism, text, productiv
ity, intertextuality, number, formula), teaches us to work in
difference, i.e., above the differences in whose name we are
forbidden to conjugate writing and science, History and form,
the science of signs and the destruction of the sign: it is all these
fine antitheses, comfortable, conformist, stubborn, and self
assured, which the work of Julia Kristeva cuts across, scarring
our young semiotic science with a foreign mark, in accord with
the first sentence of Senuroti4e: "To ma language into W()1'/t, to
work in the malerialiry of what, for society, is a means of contact
and of comprehension-is this not to make oneself, from the
start, foreign to language?"

La Quiruaine UtUrairt, 1970

,
 The Retum of the Poetician

When he sits down in front of the literary work. the poetician

docs not ask himself: What docs this mean? Where does r.bis
come from? What does it connect to? But. more simply and
more arduously: How is this nvukr This question has already
been asked three times in OUT history: Poetics has three patrons:
Aristotle (whose Poetics provides the first structural analysis of
the levels and the parts of the tragic oeuvre), Valery (who
insisted that literature be established as an object of language),
Jakobson (who calls ptHtic any message which emphasizes its own
verbal signifier). Poetics is therefore at once very old (linked to
the whole rhetorical culture of our civilization) and very new,
insofar as it can today benefit from the important renewal of
the sciences of language.
Ghard nette and this defines the personality of his work
masters both the past and the present of Poetics: he is by one
and the same impulse a rhetorician and a semiotician; figure!
are for him logical forms, manners of discourse, whose field is
not only a little group of words but the structure of the text in
its entirety; it s
i therefore appropriate that his written work be
called Figures (I, II, III); for what belongs to the Figure is not
only the poetic image but also, for instance, the form of the
narrative, present object of narratology. Geneue's work thereby
takes its place within a huge and contemporary space: it is a
work at once critical (related to literary criticism), theoretical
(militating for a theory of literature, that object so neglected in
France), practical (applied to specific works), epistemological
(proposing, thanks to the text, a new dialectic of the panicular
and the general), and pedagogic (seeking to renew the teaching
of literature and providing the means to do so).

'7'

B
TM Return of the Pottitian '7S
The poetician: until quite recently, this character might have
passed for the poet's poor relation. But precisely. the poetics
practiced by Geneue has as its object all the praxis of language
or the praxis of all language. Not only does poetics include in
its field the narrative forms (whose analysis is well developed)
and doubtless tomorrow the essay. intellectual discourse-inso
far as it chooses to be um'Utn-but also, turning back to its own
language. it consents, it is compelled to consider itself. in a
certain fashion, as an object of poetics. This return, which is
much more important than a simple expansion. tends to make
the poetician into a writer, to abolish the hierarchical distance
between "creator" and "commentator." In other words, the
poetician accepts the return of the 5ignifier in his own discourse.
At least. this is what happens in Genette's case. I am nO[ here
passing judgment on writing in the name of "style" (though
Genette's is perfect), but on the kind of fanlasmatic power which
makes a scriptcr give himself over to the demon of classifying
and naming, consent to put his discourse on stage. Genette
possesses this power in the guise of an extreme discretion-a
discretion, moreover, sufficiently wily to enjoy such power
(crucial attribute of the pleasure of writing and of reading).
Geneue classifies. vigorously and rigorously (notably the nar
rative figures in PrOUSl, since that is the chief object of his
Figures Ill): he divides and subdivides forms, and this is the first
point where the poetician becomes a poet. for he creates, in the
profile of the work (here Proust's novel), a second tabuau,
deriving less from a meta-language than, more simply, from a
second language (which is not the last, since I myself, among
others, am writing on Genette). Genette's description of the
modes of Proustian narrative reminds me of that text in which
Poe simultaneously describes, discredits, and creates "Maelzel's
Chess-Player": a man is hidden in the automaton, but he is not
seen; the problem (for Poe, and by proxy for Genette) is not to
describe the man (hidden object), or even, strictly speaking, how
he is hidden (since the machine's interior is apparently always
visible), but the subtle shifting of screens, doors, and shutters

B
174 Tlu Lovt'r of Signs
which arranges matters so that the man is never wher one is
IooJUng; in the same way, Genette sees Proust where we are not
looking for him; and from that moment on. it is of lilLIe
importance whether or not he is there: it is not the occupant of
meaning which determines the work, it is his piau; and. also
from that moment on, Proust, the Proustian aroma, returns in
force and circulates in Genette's machine; the quotations pass
in a new light, they engender a different vibrato from the one
to which we had been accustomed by a compact muting of the
work.
Then, too, Genette names what his classification finds: he
argues against received acceptations, he creates neologisms. he
vivifies old names, he constructs a terminology. Le a network
.

of subtle and distinct verbal objecu; now, nrological concern

(or courage) is what most directly establishes what I shall call
the great critical fictivity [romamsque]. To make the work of
analysis into an elaborated fiction is perhaps today's crucial
enterprise: not against truth and in the name of subjective
impressionism, but, on the contrary, because the truth of critical
discourse is not of a referential but of a linguistic order: language
has no truth except to acknowledge itself as language; good
critics, useful scholars will be those who announce the color of
their discourse, those who clearly affix to it the signature of the
signifier. That is what Genette does (his "apr-es-propru' leaves no
doubt as to his intention).
Here is how Genette's project concerns us: what he discerns
in Proust, with predilection (as he himself underlines), are
narrative deviances (by which the Proustian narrative counters
our possible nOlion of a simple, linear, "logical" narrative). Now,
deviances (from a code, a grammar, a norm) are always mani
festations of writing: where the rule is transgressed, there writing
appears as excess, since it takes on a language whkh lJJflj not
fortSen. In short, what interests Genette in Proust is writing, or,
10 be more precise, the difference which separates style from
writing. The term devianc would doubtless be troublesome if
we believed that there exists an anthropological model of

,
Thl Return of the Pattician 1 75
narrative (whose creator "would withdraw"), or even a narrative
ontology (whose "work" would be some monstrous abortion); in
reality, the narrative "model" is itself only an "idea" (a fiction),
a memory of reading. I should prefer to say that nette dips
into the Proustian reservoir and shows us the places where the
story ".sIUds" (this metaphor aims at respecting the text's move
ment, its productivity). Now, a theory of "skidding" is necessary
puci.sel, today. Why? Because we are in that historical moment
of our cuhure when narrative cannot yet abandon a certain
readability, a certain conformity to narrative pseudo-logic which
culture has instilled in us and in which, consequendy, the only
possible novations consist not in destroying th Story, the an
ecdote, but in deviating it: making the code skid while seeming
to respect it. It is this very fragile state of the narrative, at once
conforming and deviant, that nette has been able to see and
to make us see in Proust's work. His work is at once structural
and historical because he specifies the conditions on which
narrative novation is possible without being suicidal.

La Quimain.! littiraire, 1971

,
 To learn and to Teach

Even before the curtain got=s upon his book, Christian Metz gives
us what is inimitable in his voice. Listen to the overture of his lat
est work: "Volume I of this collection. elaborated in 1967 and
published in 1968 (2nd edition, 1971), grouped certain articles
written between 1964 and 1967, published between 1964 and 1968.
This second volume consists of subsequent texts (written between
1967 and 1971, published between 1968 and 19711:), as well a!I twO
unpubljshed texts written in 1971 (texts no. 8 and g).".
These numerical specifications are of coune required by the
scientific-or at least by the scholarly----code of exactitude; but who
could fail to notice that. in the mixture ofinsistence and elegance
which marks the statement, there is something rMTt? What is it?
Precisely, the suhjt's very voice. Dating with any message. Metz
odds on; but what he adds on is neither idle nor vague nor digres
sive nor verbose: it is a matte supplement, the idea's insistence
that it be expressed completely. Anyone who knows Metz in the
triple aspect of writer, teacher, and friend is always struck by this
paradox, which is merely apparent: a radical demand for preci
sion and clarity generates a free, somehow dreamy tone, a tone I
should say sounds almost drugged (did not Baudelaire make
hashish the source of an unexampled precision?): here an enraged
exactitude prevails. Henceforth. we are in Expenditure-and not
in mere knowledge: when Metz gives figures, references, when
he summarizes, when he classifies, when he clarifies, when he in
vents, when he proposes (and in all these operations his labor is
active, tireless, efficacious), he does not merely communicate, he
gives, in the full sense of the term: there is a veritable gift of

loWis stl:1 kI sipifll4liun au cinbM. Vol. II (Paris, 1971).

B
To uam and to Teach '77
knowledge. of language. of the subject insofar as he is concerned
to speak (though his work issues so explicidy from linguistics. does
he not tell us. in his way. that the error of this science is to make us
believe that messages are "exchanged"-always the ideology of
Exchange-whereas the reality of speech is precisely to give or to
take itself back. in short to tkmand?). There are two ways of sub
verting the legality of knowledge (inscribed in the Institution):
either to disperse it or to give it. Metz chooses to give; the way in
which he treats a problem oflanguage and/or of cinema is always
generous: not by the invocation of "human" ideas. but by his in
cessant solicitude for the reader. patiently anticipating his de
mand for enlightenment, which Metz knows is always a demand
for love.

There are perhaps two ways of avoiding mastery (is this not the
stake today of all teaching, of any intellectual "role"?): either to
produce a perforated, elliptical. drifting. skidding discourse; or.
conversely, to load knowledge with an excess ofclarity. This is the
way chosen (savored?) by Mea. Christian Mea is a marvelous di
dacticia n; when we read him. we know everything. as if we had
learned it ourselves. The secret of this effectiveness is not difficult
to find: when Mea teaches a piece of knowledge. a classification,
a synthesis, when he explicates certain new concepts. he always
demonstrates. by the didactic perfection of his utterance, that he
is teaching himselfwhat he is supposed to be communicating to oth
ers. His discourse-th is is his characteristic, his idiolectal virtue
manages to unite two tenses: that of assimilation and that of ex
position. Hence, we understand why the transpartfWj of this dis
course is not reductive: the (heteroclite) substance of knowledge
is clarified before our eyes; what remains is neither a scheme nor
a type, but rather a "solution" of the problem. briefly suspended
before us solely so that we can traverse and inhabit it ourselves.
Metz knows and invents many things, and these things he says
very well: not by mastery (Meu never sets anyone else right), but
by taltnl: by this old word. I mean not some innate disposition but
the artist's or scholar's happy submission to the effect he wants to

B
 The Lovtr of Signs
produce, to the encounter he wanlS to provoke: even to the trans
ference he thus accepu, lucidly, ouuide any scientific image-rep
ertoire, as the very principle ofwriting.

A theoretical work-which is only just beginning-is thus con

structed from a movement (as we say: a movement of the heart);
Metz has sought to break down the fatigue of a stereotype: "Cin
ml(l is a language." What if we were to look and see? What if, sud
denly, we were to grasp the metaphor-nonsensica1 by dint of being
repeated-in the implacable light of the Letter? From this wager,
new and somehow innocent (is not any return to the letter inno
cent?), Metz has drawn a work whose rings uncoil according to an
implacable and flexible project: for, in our time, when sensibility
to language changes so rapidly, Metz follows iu twisu and turns,
iu explosions; he is nOl the man of a semiology (of a grid), but of
an object: the filmic text, a shimmer in which different intentions
may be read, according to the moment of our intellectual dis
course. Such, I think, is Metz's historic place (there is no minor
history): he has been able to give what was (or risked being) merely
a metaphor the plenitude of a scientific pertinence: in this he is a
founder, as is attested by his singular and acknowledged place in
general semiotics and in the analysis of the cinematographic phe
n?menon; yet, having founded, he shiflS ground: now he is at
grips with psychoana1ysis. It is perhaps in this that semiology owes
and will owe him so much: for having conquered for it, in his
chosen realm, a right of mutation. By his work, Metz makes us
realize that semiology is not a science like the others (which does
not keep it from being rigorous), and that it has no intention of
substituting itself for the great epistemes which are our century's
historical truth, but rather that it is their servant: a vigilant serv
ant who, by representing the snares of the Sign, keeps them from
falling victim to what these great new knowledges claim to de
nounce: dogmatism, arrogance, theology; in shon, that monster:
the Last Signified.

Ca, 1975

,
 6
READINGS

 ONE

Cayrol and Erasure

"I'll never have time, if I must Itttp trasing what l have to sa,. "

In all of Jean Cayrol's work., somtont is speaking t4 you. but you

Dever k.now who it is. Is the speaker an individual narrator,
whose individuality is renewed in novel after novel, and does
Gaspard differ from Armand the way Fabrizio does from Julien
Sorel? Is he a single narrator whose voice extends from book.
to book? Is it Cayrol himself, half concealed behind this speaking
Other? The narrator's person. in this entire oeuvre, remains
technically indeterminate; we find neither the narrative duplicity
of the classical novel nor the complexity of the Proustian "."
nor the poet's ego; in literature. ordinarily. the person is a
completed idea (even if it manages to make itsc!'lf ambiguous):
no novelist can begin writing if he has not chosen the profound
person of his narrative: to write is, in shon, to decide (to be
able to decide) who is going to speak.. Now, Cayrolian man is
scarcely a character; he possesses no pronominal cenainty; either
he falls far shon of identity (in the early novels) or, apparently
constituted, he nonetheless k.eeps undoing his person by contin
ual deception of memory and narrative; he is never anything
but a voice (which we cannot even call anonymous, for that
would be to qualify it). and yet this VOlee does not entrust its
indecision about origins to any novelistic technique: neither
collective nor named, it is the voice of .sonumu.
Constantly posited and withdrawn. the narrative's person is
in fact merely the suppon parsimoniously granted to a very
mobile, il/-attach.ed language, which shifts from place to place,

.8.

B
 Readings
from object to object, from memory to memory, while every
where remaining a pure articulated substance. This is anything
but a metaphor; there , in Carrol. a veritable imagination of
the voice substituted for the visual sensibility of writers and
poets. First of all, the voice can rise. emanate from no one
knows where; unsituated, it is nonetheless there, somewhere,
around you, behind you. beside you, but actually never in front
of you; the real dimension of the voice is indirect. lateral; it
takes others obliquely, brushes against them and goes off; it can
touch without telling its origins; hence. it is the very sign of the
unnamed. what is born or what remains of man if we take from
him the materiality of his body, the identity of his face. or the
humanity of his gaze; it is the substance at once most human
and most inhuman; without it, no communication among men,
but with it, [00. the discomfort of a double, insidiously appearing
from a (chthonic or celestial) super-nature, in short from an
alienation; a well-known test tells us that everyone is uncom
fortable hearing his own voice (on tape, for example) and often
fails even to recognize it; this is because the voice, if we detach
it from its source. always establishes a kind of nrange familiarity
which is, ultimately, the very familiarity of the Cayrolian world,
a world which offers itself to recognition by its exactitude and
yet denies itself to recognition by its uprootedness. The voice is
still another sign: that of time; no voice is motionless, no voice
ceases to pass; furthermore, the time the voice manifests is not
a serene time; however smooth and discrete it may be. however
continuous its flux. every voice is threatened; symbolic substance
of human life, there is always at its origin a cry and at its end a
silence; between these two moments develops the fragile time
of speech; fluid and threatened, the voice s
i therefore life itself,
and it is perhaps because a novel by Cayrol is always a novel of
the voice that it is also always a novel of life of fragile
jeopardized-life.
It is said of certain voices that they are caressing. The Cayrolian
voice gives an abusive caress to the or1d. a lost caress. Like the
caress. language here remains on the surface of things; the

,
Cayt'ol and Erasurt
surface is its realm. This superficial description of objects has
been made into a feature common to a certain number of
contemporary novelists; yet, unlike Robbe-Grillet, for example,
Cayrol's surface is not the object of a perception which exhausts
its existence; his way of describing is often profound, it gives
things a metaphorical radiance which does not break with a
certain romantic writing; this is because the surface, for Cayrol,
is not a quality (optic, for instance) but a situation of things. This
superficial situation of objects, of landscapes, of memories, even,
is lew, as we might say of a world seen from floor level; we shall
not find here. on the writer's part, any sentiment of power or
elevation with regard to the things described; the gaze and the
voice which follow them on t !rotl remain captives (and we
with them) of their surface; all the objects (and there are many
of them in Cayrol's novels) are minutely scrutinized, but his
minuteness is a captive, in it something cannot rise, and the very
complete world which the writing caresses remains stricken by
a kind of sub-familiarity; man does not enter completely into
the use of the things he encounters in the course of his life. not
because he sublimates them (as would be the case in a traditional
novel, lapsed into psychology), but on the contrary because he
cannot raise himself to this use-because he remains doomed
to a certain unattainability of objects whose exact altitude he
cannot reach.
This literature at floor !rod (Cayrol himself has already used
the expression) might have the mouse for its totem animal. For
the mouse, like Cayrolian man, deals with things; it omits little
on its way, concerned with everything its oblique gaze, proceed
ing from the ground up, can encounter; a tiny stubbornness,
never triumphant and never discourag, animates it; remaining
on the level of things. it sees them all; the same is true of
Cayrolian description. which in its fragile and insistent way
scrutinizes the countless objects modern life stuffs into the
narrator's existence; this busy, mouselike progless, at once
incidental and continuous, gives its ambiguity to Cayrolian
description (such description is important. for Cayro)'s novels

B
 Readings
are essentially exterior); this description spares nothing, it slides
across the surface of everything, but its sliding lacks the euphoria
of Hight or swimming, it acquires no resonance from the noble
substances of the poetic image-repertoire, the aerial or the
liquid; it is a terrestrial sliding, a sliding across of the Roor,
whose apparent movement consists of tiny jerks, of a rapid and
modest discontinuity: the "holes" in such description are not
even loaded silences, but merely a human impotence to link the
accidents of things: there is a Cayrolian misfortune in not being
able to institute a familiar logic, a rational order among the
phenomena with which time and the journey confront the
narrator. It is here that we rediscover, in a mocking form, the
theme of the caress: in opposition to it, though proceeding from
it, we discover a kind of scratchy perception of things, a grating
touch bestowed upon the world of objects (but silk, [00, can
grate, and often nothing could be more sumptuous, in its
modesty, than a Cayrolian description); whence so many images
of the rough, the nibbled, and the acid, mocking forms of a
sensation which never manages to regain the euphoric continuity
of the caress; the smooth, elsewhere a miraculous theme of the
"seamless," is here an element which "turns," covering itself
with a kind of surficial harshness: the surface of things begins
to vibrate, to grate slightly.
This theme of the rough, of the failed caress, disguises a still
more disturbing image, that of a certain coldness. The touchy is,
after all, merely the active world of the chilly, the susceptibility
to cold. In Cayrol, where seascapes abound, from Dieppe [0
Biarritz, the wind is always sharp; it is faintly cutting, but, more
certainly than deep cold, causes constant shivering, without,
however, altering the progress of events, without astonishing
. . . The world continues, familiar and dose at hand, and yet
one feels the cold. This Cayrolian cold is not that of the great
immobilities, it leaves life intact, even agile, yet fades it, ages it;
Cayrolian man, vulnerable as he may be, is never frozen stiff,
paralyzed; he still walks, but his physical milieu keeps him
cominually on edge: the world is to be warmed up. This sustained

B
.86 Readings
says in La Gaffe. "and yet I know no one more impervious to
pain than you . . . You are impregnable when someone attacks
your secret reserves." This fragile, sensitive world is a resistant
world; beneath the harshness and the piercing wind, behind
the oblivion which fades things. behind that tense footstep.
something (or someone) bums whose reserves nonetheless re
main secret, like a strength which never knows its own name.
This strength is secret because it is not in the hero described
by the book but in the book itself. As a shortcut, we can say
that it is Carrol's own strength, the strength which makes him
write. OUT culture has long wondered what it was that passed
from the author into a work; here we see that, even more than
his life or his times, it is the writer's strength which passes into
his work. In other words, literature is itself a moral dimension
of the book: to be able to write a Story is the final meaning of
that story. This explains how, with an extremely disarmed
world, Cayrol can present a power, even a violence (I am
thinking of Muriel), but this power is not interior to this world,
it is the power of the writer Cayrol. the power of literature: we
can never sever the meaning of a fictive world from the very
meaning of the novel. Hence. it is futile 10 ask by what
philosophy-interior to Cayrolian man but modestly silenced
the default of thisworld can be recuperated, for once literature
takes over "what doesn't work in the world" (as it does here),
the absurd ceases. Led to the brink of the cold and the futile,
every reader of Cayrol also finds himself endowed with warmth
and a sense of being alive which are given by the very sctacle
of someone who writes. Thus. what can be asked of this reader
is to entrust himself to the work, not for what philosophy it
may afford. but for what literature . . .

Just as substances in Cayrol never present themselves except

to a kind of failed caress, to a discontinuous and somehow
slighted perception. similarly Cayrolian time is a devoured time,
insidiously nibbled in places. And when the object of this time
is a man's life (as in Foreign Bodies). something appears which

Malena :>m d reitl al

Cayrol and Erasurt
constitutes the entire Cayrolian novel (this theme will be appar
ent to viewers of Muriel): the bad faith of memory.
Every novel by Cayrol might be called Mmmrs of an Amnesiac.
Not that the narrator makes an enormous effort to recall his
life: it seems to come to his memory quite naturally, as happens
with ordinary recollections; yet the more the narrative develops.
the more perforated it appears; episodes do not connect properly.
something grates in the distribution of actions (one ought to
say, more specifically, with regard to a novel: in their dispatching);
but above all, without our ever being able to catch the narrator
out in his preterition (or lie), the whole of an apparently regular
narrative gradually refers to the sensation of a major forgetting,
settled somewhere in existence and unluckily resonant within
it, devouring it, branding it with a false movement. In other
words. Cayrolian narrative is subject to a montage whose speed
and dispersion designate a very special disorganization of time,
which Cayrol himself anticipated in Lazart and which we find
illustrated in the montage of Muriel. This forgetting. in which
the characters struggle without quite knowing it, is not a
censoring; the Cayrolian universe is not burdened with a con
cealed fault, forever unnamed; confronted with this world,
there is nothing to decipher; what is lacking here are not
fragments of culpable time but only fragments of pure time,
what it is necessary for the novelist not to say in order to separate
a little man from his own life and the life of others, in order to
render him at once familiar and unattached.
Another form of this nibbled time: memories are interchange
able within one and the same life, they constitute the object of
a swap. analogous to that of the dealer and fence Gaspard (a
camembert for an air-chamber riAe): memory is at once a
substance for concealment and for deals; the hero of Foreign
Bodies has two childhoods. which he cites as necessary. depending
on whether he wants to endow himself with a rural origin or
with that of an abandoned child; Cayrolian time consists of
ttansferred-one might even say stolen-fragments, and between
these fragments there is an interplay which constitutes the entire

CJPYnghied mater1
188 Rtadings
novel. Fortign Bodies begins with a review of all the objects which
can enter the body. by negligence or misfortune; but for
Cayrolian man, the real foreign body is ultimately time: this
man is not hewn within tbe same duration as other men; time
is transferred to him, sometimes too short when he forgets,
sometimes too long when he invents. For this unjust (unadjusted)
time must be struggled against, and the entire novel consists in
telling one man's efforts to regain the exact time of other men.
Thus is generated, throughout the Cayrolian monologue (es
pecially in FortignBodie.s), a disclaiming utterance whose function
is not to deny faults but in a more elementary. less psychological
fashion, w erase time. Cayrolian erasure is nonetheless secondary:
the narrator does not try to rub out what exists, to invoke
oblivion of what has been, but. quite the contrary. to repaint
the void of time with bright colors, to paper the holes in his
memory with an invented memory, destined much less to justify
his time (though the collaborator Gaspard desperately needs an
organized time) than to make it rejoin tlu ti1M of others, i.e., to
humaniu time.
For this is basically the great task of the Cayrolian novel: to
say-with all of literature's power of recuperation, of which we
have spoken-how a man is separated from other men, nOl by
the romantic singularity of his destiny, but by a kind of vice of
his temporality. The singularity of tbis Cayrolian world is in
fact that the beings in it are by the same impulse 17Udiocrt and
unwonted, natural and incompuhnuible. Hence, we never know if
the hero of this world is "sympatbetic"-if we can care for him
to the end. All our traditional literature has played on the
positivity of the fictive hero. but here we do not feel alienated
in the presence of a being whose world we know well but of
whose secret lime we are ignorant: his time is not ours. yet he
speaks to us quite familiarly of the places, objects, and stories
we share with him: he is at home with us, yet he comes from
"somewhere" (but from where?). Confronting this ordinary and
singular hero, we experience a sentiment of solitude, but such
solitude is not simple; for when literature offers us a solitary

GJPYnghtcd maknal
Cayrol and Erasure 189

hero, it is his very solitude which we understand, which we love,

and which thereby we put an end to: neither the hero nor his
reader is alone any longer, since they are alone together. Cayrol's
art goes further: he makes us see a solitude and yet keeps us
from participating in it; not only does literature not recuperate
Cayrolian solitude but it goes about purifying it of any positive
complicity: it is not a solitary man whom we see living (in which
case he would not be altogether alone), it is a man who imposes
upon us, in relation to himself. that tenacious instnSibility Cayrol
has spoken of in lazare. Thus. by a final achievement of the
work, the reader experiences the Cayrolian hero exactly as the
latter experiences the world: sensitive and insensitive. installed
in that "parasitical" sympathy characterizing a world in which
we can never love except by proxy.
We know just where this work comes from: from the concen
tration camps. The proof of which is that Lar.are parmi now. a
work which creates the first junction between experience of the
camps and literary reflection. contains in germ. with great
exactitude, all of Cayrol's subsequent work. Pour un romant'sque
la:wrern is a program which is still being carried out today in a
virtually literal fashion: the best commentary on Muriel is Lazare.
What must be suggested, if not explicated, is how such a work
whose germ is in a specific, dated history-is nonetheless entirely
a literature of today.
The first reason for this may be that the mind of the
concentration camps is not dead; there occur in the world odd
"concentrative" impulses-insidious, deformed, familiar-cut off
from their historical model but spread in the fashion of a style;
Cayrol's novels are the very passage from the concentrative
event to the concentrative "life style": in them we rediscover
today, twenty years after the Camps, a certain form of human
malaise, a certain quality of the dreadful, of the grotesque, or
of the absurd, whose shock we receive in the presence of certain
events or, worse still, in the presence of certain images of our
time.
The second reason is that Cayrol's oeuvre, from its beginning.

C ;.pvnghted malaria
. go Radings

has been immediately modern; all the literary techniques with

which we credit today's avant-garde, and singularly the New
Novel, are to be found not only in Cayrol's entire oeuvre but
even, as a conscious program, in Pour un rcnna nesque lazarien
(1950): the absence of anecdote. the disappearance of the hero,
giving way to an anonymous character reduced to his voice or
to his gaze, the promotion of objects, the affective silence which
we cannot call either modesty or insensitivity, the Ulyssean
character of the work, which is always a man's long march
through labyrinthine space and lime. If, however. Carrol's
oeuvre has remained outside recent theoretical discussions con
cerning the novel, it is because its author has always refused to
systematize his work, and also because the technical community
just mentioned is far from being complete; the New Novel
(granting that we can unify it) posits matte descriptions, the
character's insensibility communicating itself to things he speaks
of, so that the New Novel's world (which 1 should be glad to
reduce, as I see it, to the world of Robbe-Grillet) is a neutral
one. Cayro)'s world, on the contrary, even if love is here merely
parasitical (the author's expression), is a world vibrant with
adjectives. radiant with metaphors; granted, objects are pro
moted to a new fictive rank, but man continues to touch them
with a subjective language, he gives them not only a name but
commentary
also a reason, an effect, a relation, an image. It is this
on the world, which is no longer merely uttered but embellished.
that makes Cayrol's oeuvre such an individual communication:
deprived of any experimental intention yet audacious, at once
emancipated and integrated. violent without the theater of
violence, "concentrative" and of the moment, it is an oeuvre
which unceasingly escapes forward, impelled by its own fidelity
to itself, toward the new our times call for.

POSLKript Jean Cayrol's Ln CorfJ5 ttrangtn

to

(U.G.E.). 1964

GJPYnghtcd maknal
Bw, ' 93
blindness which failed to discern, within the bourgeois, the
capitalist. Yet can literature be anything other than indirutly
lucid? In order to constitute his discourse, in order to invent
and develop it within his own truth. the writer can sak only
of what alienates himself, for one cannot write by proxy; and
what alienates the writer. in the bourgeois, is stupidity; bourgeois
vulgarity is doubtless only the sign of a deeper malady, but the
writer is doomed to work with signs, to vary and elaborate them,
not to deRower them: his form is metaphor, not definition.
Hence, Bloy's labor has been to metaphorize the bourgeois.
His disgusts invariably and specifically designate the parvenu
writer as the bourgeoisie acknowledges and delegates him. It
I

suffices to be recogniud by bourgeois institutions (the press. the

salons, the Church) in order to be condemned by an. Sloy's
virulent demystifications therefore aim at all ideologies without
distinction, once they appear priviltgtd. from Veuillot to Riche
pin, from Pere Didon to Renan. Sloy sees no difference whatever
between the populism of Valles and the charities of the Duchess
of Galliera. crudely praised by the press for the fabulous
benefaction of millions which she had only, Sloy says. to reslOrt.
Conversely, none of the (few) writers he supponed were on the
side of Propeny; or more specifically. Bloy's glance at Barbey
d'Aurevilly. at Baudelaire. at Verlaine is a way of galvanizing
these writers. of making them inapt for any bourgeois purpose.
Bloy's discourse does not consist of ideas; nonetheless, his work
is critical insofar as it could discern in the literature of his time
its resistances to order, its power of inadaptation. the permanent
scandal it constituted with regard to collectivities and to insti
tutions. in shon the infinite recession of the questions it raised.
or in still another word: its irony. Because he always saw art as
anti-money, he almost never made a mistake: the writers he
vilified (Dumas fils. Daudet, Bourget, Sarcey) certainly look to
us today like absurd puppets; on the other hand. Bloy was one
of the very first to recognize Lautreamont. and, in Lautreamont,
a singularly penetrating prophecy, the irreversible transgression
of literature itself: "As for literary form. there is none here. It
is a liquid lava. It is senseless, black, and corrosive." Did he not

C :lpynghled malenal
MichLlet, Today 1 97

"good" action, Michelet simply says it is "inexplicable"; this is

be<:ause the tyranny of the predicate produces a kind of defi
ciency of the subject (in logic; but in relation to discourse, the
logical meaning is not far from the psychological meaning: is
not a purely predicative discourse the very discourse of para
noiac delirium?).
Finally, and perhaps this is the most disturbing, it is not only
the concatenation of events which vacillates in Michelet, but the
event itself. What is an evmt1 This is a problem of philosophical
dimensions, the pons asi7lQ1ll:m of historical epistemology. Mich
elet accepts the disorder of the notion. It is not that his history
lacks facts, events, frequently the most precise son; but these
events an never where one expects them to be; or again, it is
their moral resonance which is evaluated. not their exttnt; the
Micheletist fact oscillates between excess of specificity and excess
of evanescence; it never has its exact dimension: Michelet tells
us that on the 18th Brumaire (November 10) stoves were lit in
the great hall of the Orangerie and that in front of the door
was a tapestried drum: but Barras's resignation? the two phases
of the operation? the role of Sieyes, of Talleyrand? No mention
of these facts, or at least no mention which "extracts" from a
strange discourse (strange for our ways of reading history) some
frankly narrative element. In short, what Michelet disorders is
the prqponwn of facts (need we recall that the critique of. relations
is much more subversive than that of notions?). Philosophically,
at least from the viewpoint of a cenain philosophy, it is Michelet
who is right. Here he stands. quite paradoxically, beside Nietzsche:
"There is no fact in itself. What happens is a group of phenom
ena, selected and grouped by a being who interprets them . . .
There is no state-offact in itself; on the contrary, a meaning
must bt introduced even btfore there can be a state offact." Michelet
is, in shon, the writer (the historian) of the even before: his
history is impassioned not because his discourse is rapid, im
patient, not because its author is hotheaded. but because it does
not arrest language at the fact; because, in that enormous
staging of an age-old reality, language precedes fact to infinity: a

C JPYnghied mater1
'98 Readings
proposition outrageous to a classical historian (but, once History
is structuralized, does it not come closer to our present philos
ophy of language?), a proposition promising to the modern
theoretician who thinks that. like any science (and this is the
problem of the "human sciences"), the science of history, not
being algorithmic. inevitably encounters a discourse, and it is
here that roerythingbtgins. We must be grateful to Michelet (among
other gifLS he has given US-giflS ignored or suppressed) for
having represented to us, through the pathos of his period, the
real ccmdilions of historical discourse, and for inviting us to
transcend the mythic opposition between "subjectivity" and
"objectivity" (such a distinction is merely propaedeutic: necessary
on the level of research), in order to replace it with the opposition
between statement [inonci] and spetch act [inonciationJ, between
-

the product of investigation and production of the text.

Criticism of Michelet by many historians and by popular

opinion itself-whose arguments have been ironically summa
rized by Lucien Febvre in his little book Trails ( 1946)-is not
only, of course, a scientific criticism (bearing on the historian's
facts and interpretations) but also a criticism of writing: for
many, Michelet is a bad historian fMcaust he writes, instead of
simply "reporting," "chronicling," etc. Today we no longer
understand writing as the simple product of stylistic mastery.
What makes Michelet a writer (practitioner of writing, operator
of the text) is not his style (which is not invariably first-rate,
being on occasion merely the parading of style) but what we
today call the excess of the signifur. This excess is 10 be read in
the margins of representation. Of course, Michelet is a classic
(readerly) writer: he recounts what he knows, he describes what
he sees, his language imitates reality, he adjusts the signifier to
the referent and produces clear signs thereby (no "clarity"
without a classical conception of the sign, the signifier on one
side, the referent on the other, the former in the latter's service).
Michelet's readerliness, however, is not certain; it is often
jeopardized, compromised by excesses, blurs, breaks. leaks;
200 Readings

of his time, and this "rightness" represents, in his work, the

share which today seems to us correct. Michelet has not "dis
torted" "reality" (or he has done much more than that), he has
located the surfacing point of that "reality" and of his discourse
in an unexpected place; he has shifted history's level of percep
tion; in his historical work, examples abound (phenomena of
collective mentality, mores, ecological realities, material history,
everything which has flourished in subsequent history), but the
example 1 want to cite, of this "perceptual decision," comes
from his natural history (La Mer). Having to describe the terrible
storm of 1 859, by a bold stroke which relates him to the symbolist
poets, Michelet describes it/rom within ; but where he goes further
still is that this "within" is not metaphorical, subjective, but
literal, . spatial: the whole description is made from inside the
room where the storm keeps him confined; in other words, he
describes what he does not see, not as if he were seeing it (this would
be a banal instance of poetic clairvoyance), but as if the storm's
reality were an unheard-of substance, coming from another
world, perceptible to all our organs except that of sight. This is
a veritably drugged perception, the economy of our five senses
being disordered within it. Michelet, moreover, knew the phys
iological stake of his description: the storm provokes him to
make an experiment on his own body, like any taker of hashish
or mescaline: "I went on working, curious to see if this wild
force would succeed in oppressing, fettering a free mind. 1 kept
my thought active, self-controlled. 1 went on writing and ob
served myself. Only in the long run did fatigue and lack of
sleep affect a power within me the most delicate power a writer
possesses, 1 believe the sense of rhythm. My sentences became
inharmonious. This was the first string of my instrument to be
broken." Hallucination is not far off: "[The waves1 . . . affected
me as a dreadful mob, a horrible throng, not of men but of
baying dogs, a million, a billion fierce or mad hounds . . . But
what am 1 saying? dogs, hounds? That was no closer to the
reality. These were hateful and nameless apparitions, beasts
with neither eyes nor ears, only frothing gullets." If we say that
Michekt, Today . 201
it is Michelet's entire history which is hallucinated, it is not his

historical sense which we depreciate but a modem language

which we exalt; that intuition or that courage he has had to
proceed as if our discourse passed through the world and time
to infinity, as if yesterday's hallucinations were tomorrow's
truths, and so on.

There are two means of demystifying a great man: by reducing

him as an individual, or by dissolving him into historical gen
erality, making him the determined product of a situation, of a
moment, the delegate of a class. Michelet was not unaware of
this second means; on several occasions he indicated the links
between Bonaparte and Finance, a procedure which already
moves in the direction of a Marxist critique; but the crux or
the obsession of his demonstration is to depreciate Bonaparte
in his body. The human body it would be better to say the
historical body, as Michelet sees it exists, as we know, only in
proportion to the affections and disgusts it provokes; it is at
once an erotic body (implying desire or repulsion: pulsion) and
a moral body (Michelet is for or against, according to avowed
moral principles). It is, one might say, a body which altogether
abides within the space of a metaphor: for instance, that of
nausea, a physical spasm and a philosophical rejection. Rereading
Michelet after a good number of years, I am struck once again
by the imperious character of his portraits. Yet the portrait can
easily become a tiresome genre, for it is not sufficient to describe
a body in order to make it exist (desire); Balzac, for instance,
never produces an erotic relation between himself (and therefore
ourselves) and his characters; his portraits are deadly. Michelet,
on the other hand, does not describe (at least in the portrait I
am thinking of, that of Bonaparte): in the whole body (labori
ously itemized by Balzac, organ after organ), he briskly checks
off two or three sites and scrutinizes them; in Bonaparte (we
should say on him), it is the hair: brown, but so heavily pomaded
it looks black; the yellow, waxy face: without eyebrows or lashes;
the eyes: gray as a pane of glass; and the extremely white teeth:
,

202 Readings
"But how dark he is, this Bonaparte! , , , He is dark, but what
white teeth he has!" This portrait is striking, but what attests to
Michelet's power (the excess of his text, its transcendence of
any rhetoric) is that we cannot really say why; it is not that his
art is ineffable, mysterious, inherent in a "brushstroke," a "je ne
sais quai," but rather that it is a kind of pulsional art which
directly plugs the body (Bonaparte's and Michelet's) into lan
guage, without the intermediary of any rational relay (by which
we might understand the subjection of the description to a grid,
either anatomical the kind observed by Balzac or rhetorical
traditionally, the portrait derived from a strong code, that of
prosopography), Now, it is never possible to speak directly with
regard to pulsions; all one can do is to divine their locus; in
Michelet, this locus can gradually be situated: in the broad
sense including states of substance, half visual, half tactile it
is color, Bonaparte's colors are sinister (black, white, gray, yellow);
elsewhere outside history, in Nature color is jubilatory; con
sider the description of insects: ", , , charming creatures, bizarre
creatures, admirable monsters, with wings of fire, encased in
emerald, dressed in enamel of a hundred varieties, ;;Irmed with
strange devices, as brilliant as they are threatening, some in
burnished steel frosted with gold, others with silky tassels, lined
with black velvet; some with delicate pincers of russet silk against
a deep mahogany ground; this one in garnet velvet dotted with
gold; then certain rare metallic blues, heightened with velvety
spots; elsewhere metallic stripes, alternating with matte velvet";
the pulsion of multiple color (as it is perceived behind closed
eyelids), which reaches the point of a perceptual transgression:
"I succumbed, 1 closed my eyes and asked for mercy; for my
mind was benumbed, blinded, growing unconscious," And al
ways that faculty of making the pulsion signify without ever
severing it from the body; here, the motley refers to the
inexhaustible profusion of the insects' generating nature; but
elsewhere just the contrary occurs, the bold reduction to an
obsessional color: the chain of the Pyrenees is green: " 1 n the
Pyrenees, the singular water-greens of the torrents, certain
Miclu",. Today

emerald fields . . . the green marble . . ." We must not conclude

that Michelet is a "painter" : color goes far beyond painting (I
refer here to the recent observations of J.-L. Scherer and Julia
Kristeva); color belongs [0 the order of succulence, it belongs
to the detp body; it affords Michelel's text certain zones, certain
reaches available to a reading we might qualify as nulritivt.

Yes, in Michelet the signifier is sumptuous. And yet Michelet

is not read. Perhaps the signifier is too strong (a veritable
poison), if we read Michelet as a historian or as a moralist
which was his public role until he fell into oblivion. OUT
languages are coded. we must not forget: society is forbidden,
by a thousand means, to mingle them, to transgress their
separation and their hierarchy; the discourse of History, that
of moral ideology (or that of philosophy) are to be kept pure
of desire: by not reading Michelet, it is his desire we censure.
Thus, because he blurs the discriminatory law of "genres,"
Michelet fails first of all to be given his place: serious people
conformists exclude him from their reading. But, by a second
displacement, this prince of the signifier is acknowledged by no
avant-garde (or more simply. by no "literature"). This second
exclusion is more interesting, and more contemporary as well;
we must say a word about it. for it is here that we can understand
not only why Michelet is not read by active, productive readers
(by the young, one might say) but also, more generally, what
cenain intolerances of contemporary reading might be.
What we do not tolerate is patlws (it remains to be seen if we
do not have our own). Michelet's discourse is obviously filled
with those apparently vague and sublime words. those noble
and stirring phrases, those pompous and conformist thoughts,
in which we no longer see anything but distant objects, the
rather crude curiosities of French romanticism: a whole vibrato
which no longer moves anything in us (Action, Naturt, Education,
Ptopk. etc.); how receive today a sentence (taken at random)
like: "The Father is for the child a revelation of justice," etc.?
This capitalized language no longer passts. for various reasons,

C JPYnghied mater1
204 Readings
which relate to history and to language at the same time (nothing
is more important and less studied than fashions in words); and,
no longer passing, this language accumulates in Michelet's
discourse and constitutes a barrier: if the book does not fall
from our hands for the signifier is there to enliven it at least
we must continually decant it, split Michelet and, worst of all,
make excuses for him.
This pathetic faU from grace is very extreme in Michelet.
Paradoxically, we might say this: what is sincerest ages fastest
(the reason for this, of a psychoanalytic order, is that "sincerity"
belongs to the realm of the image-repertoire: a realm where
the unconscious is least acknowledged). Further, we must ac
knowledge the fact that no writer ever produces a pure discourse
(one that is irreproachable, integrally incorruptible): work ex
foliates and fragments under the action of time, like a limestone
relief; there are always, in the greatest, the boldest writers, the
ones we like the most, perfectly antipathetic sites of discourse.
It is wisdom to accept the fact (or, less passively, more aggres
sively, it is the very plural of writing which obliges us to do so).
We cannot, moreover, reconcile ourselves to this situation in
Michelet's case with such simple liberalism, we must go further.
These words, whose magic is dead for us, can be renewed.
First of all, these words had, in their time, a living meaning,
sometimes even a fiercely combative one. Michelet used them
with passion against other words, themselves active, oppressive
(language always proceeds in this polemical direction). Here, a
certain historical culture must come to the aid of our reading:
we must divine what language's stake was at the time Michelet
was writing. The historical meaning of a word (not in the narrow
acceptation of philology, but in the much broader one of
lexicology: I am thinking of the word civilization as studied by
Lucien Febvre)-that meaning must always be evaluated dialec
tically: for historical recall sometimes encumbers and constrains
our present reading, subjects it to an untimely equality, and
therefore we must free ourselves from it quite summarily;
sometimes, on the contrary, history serves to revivify a word
Michelet, Today 205
and then we must rediscover this historical meaning as an
enjoyable, not authoritarian element, witness of a truth, but
free, plural, consumed in the very pleasure of a [lCtion (that of
our reading). In short, dealing with a text, we must make use of
the historical reference with cynicism: reject it if it reduces and
diminishes our reading, accept it if it extends that reading and
makes it more delectable.
The more a word has a magic use, the more mobile its
function: it can be employed for everything. This word is
something of a mana-word, a joker-word: it can be blank, it is
true, but it also assumes, at the same time, the highest rank; and
the word's justification is less its meaning than its rank, its
relation to other words. The word lives only as a function of its
context, and this context must be understood in an unlimited
fashion: it is the writer's whole thematic and ideological system,
and it is also our situation as reader, in all its scope and fragility.
The word Liberty is eroded (by dint of having been employed
by impostors)-but history can restore its terrible contempor
aneity; we understand today that liberty, in the meaning this
word has had since the French Revolution, was too abstract an
entity to satisfy the concrete demands of a worker alienated in
his labor and in his leisure; but such a crisis can make us fall
back on the word's very abstraction; this abstraction will once
again become a power, and Michelet once again be readable
(the rise of certain "ecological" dangers may revivify the Mich
e1etist word Nature: this process is already beginning). In short,
words never die, because they are not "beings" but functions:
they merely undergo avatars (in the strict sense), reincarnations
(here again, Febvre's text, published just after the Nazi occu
pation, shows how in 1 946 Michelet's work suddenly reechoed
the sufferings of the French oppressed by foreign occupation
and by fascism).

What separates us from Michelet is, obviously and chiefly, the

intervention of Marxism: not only the accession of a new type
of political analysis but also a whole implacable series of con-
Michelet, Today 207

purgatory we must traverse (whence, in some of us, the revolu

tionary refusal to make the crossing). There is nothing more
tragic, more overwhelming, for Michelet and for us so great
are the difficulties it heralds than this text which concludes a
chapter of one of Michelet's books (Nos Fils, 1869), though full
of pathos: "After the horrible and sinister affair of June 24,
1 848, oppressed, overwhelmed with sufferings, I said to Ber
anger: 'Oh, who can speak to the people? . . . Unless we do so,
we shall die.' That firm, cool mind replied: 'Patience! it is the
people who will write their books.' Eighteen years have passed.
And these books where are they?"
Perhaps this problem, inherited from the old Michelet, is the
problem of tomorrow.

L'Arc, 1972

 Michelet's Modernity

Michelet is not in fashion, Michelet is not modern. The great

historian has himself fallen through History's trapdoor. Why?
This is a severe, even a dramatic question, at least for someone
who loves Michelefs work and yet wants to participate in the
accession of those new values whose offensive constitutes what
we expediently call the avant-garde. This "someone" therefore
believes he is living in contradiction what our civilization, since
Socrates, regards as the most serious trauma a human subject
can receive from others and from himself. And yet: what if it
were not that "someone" who was contradictory, but Modernity
itself? The evident censorship the avant-garde imposes on
Michelet would then turn against Modernity as an illusion, a
negative phantasmagoria which must be explained: can His
tory to which Modernity belongs be unfair, even on occasion
idiotic? It is Michelet himself who has taught us so.
Michelet's Modernity I mean his effective, scandalous mod
ernity, in whose name we would invite him to remain forever
young in the history of French literature blazes forth in at
least three points.
The first concerns historians. Michelet, as we know, established
what is still timidly called the ethnology of France: a way of
apprehending the dead men of the past, not in a chronology
or a Rationality, but in a network of carnal practices in a system
of aliments, of garments, of everyday customs, of mythic rep
resentations, of amorous actions. Michelet reveals what we might
call the sensual side of History: with him, the body becomes the
very basis of knowledge and of discourse, of knowledge as
discourse. It is the example of the body which unifies his entire
oeuvre, from the medieval body that body which tasted of
208
 Michelet's Modernity
tears to the delicate body of the Witch: Nature itself sea,
mountain, animality is never anything but the human body in
expansion and, one might say, in contact. Michelet's work
corresponds to an unknown level of perception which is still
largely occulted by the so-called human sciences. This way of
handling what is historically intelligible remains very odd, for it
contradicts the belief which continues to tell us that in order to
understand we must abstract and, in some sense, disembody
knowledge.
Michelet's second modernity concerns epistemology. All of
Michelet's oeuvre postulates and often achieves a truly new
science, which is still being fought for. We do not yet call it the
science of the unconscious, nor even more broadly a symbolics;
let us call it by the very general name Freud gave it in his Moses:
the science of displacement: Entstellungswissenschaft. How could we
put this (without fear of neologism)? Metabology It is of little
consequence. No doubt certain operations of displacement, of
substitution, metaphoric or metonymic, have pell1lanently marked
the human logos, even when this logos has become a positive
science. But what gives Michelet his high standing in this new
discourse of Science is that in his entire oeuvre perhaps under
the influence of Vico, who, we must not forget, long before
contemporary structuralism, used the great figures of Rhetoric
as ciphers of human History substitution, symbolic equivalence
is a systematic path of knowledge, or, if one prefers, knowledge
is not separated from its means, from the very structure of
language. When, for example, Michelet tells us, literally, that
"coffee is the alibi ofsex," he formulates a new logic which flourishes
today in all knowledge: the Freudian, the structuralist, even the
Marxist all adherents of this science of
feel at home in Michelet's work.
Michelet's third modernity is the most difficult to perceive,
perhaps even to acknowledge, for it presents itself under an
absurd name: that of bias. Michelet is prejudiced how many
critics, how many historians, proudly installed in the comfort of
objective science, have castigated him for it! In order to write,

Michelet's Modernity 211

The present power of a past writer is measured by the detours

he has managed to impose upon the ideology of his class. The
writer can never destroy his ideology of origin, he can only
cheat with it. Michelet could not, or would not, cheat with the
language inherited from the Father: small-time printer, then
manager of a rest home, Republican, Voltairean; in a word,
petit-bourgeois. Now, petit-bourgeois ideology, nakedly ex
pressed, in Michelet's case, is one of those which are not forgiven
today, for it is still broadly our own, that of our institutions, of
our schools; hence, it cannot be taken unseasonably, as can the
"progressive" ideology of the eighteenth-century bourgeoisie.
From a modem point of view, Diderot is readable, Michelet

virtually no longer so. All of Michelet's pathos remains a

consequence of his class ideology, of the idea the fiction, one
might say that the goal of republican institutions was not to
suppress the division between capital and the working class but
to attenuate and in some sense to harmonize the.ir antagonism.
Whence, on the one hand, a whole unitary discourse (today we
would say a discourse of the signified) which must alienate from
Michelet any psychoanalytic reading, and on the other, an
"organicist" conception of History which can only close to him
a Marxist reading.
Then what is to be done? Nothing. Each of us must deal with
Michelet's text as best he can. Obviously, we are not yet ready
for a differential reading, which would agree to fragment, to
distribute, to pluralize, to disconnect, to dissociate the text of
an author according to the law of Pleasure. We are still theo
logians, not dialecticians. We prefer to throw out the baby with
the bathwater, rather than get ourselves wet. We are not yet
"educated" enough to read Michelet.

Revue d'histoire littiraire de fa France, 1974

214 Readings

the logosphere by leaving the bell-headed pins in it, the signs

furbished with their tiny jingle: thus, when we hear a certain
language, we never forget where it comes from, how it was
made: the shock is a reproduction: not an imitation, but a
production that has been disconnected, displaced: which makes
nOISe.

Hence, better than a semiology, what Brecht leaves us with is

a seismology. Structurally, what is a shock? A moment difficult
to sustain (and therefore antipathetic to the very notion of
"structure"); Brecht does not want us to fall under the spell of
another smooth surface, another language-"nature": no positive
hero (the positive hero is always sticky), no hysterical practice
of the shock: the shock is distinct, discrete (and discreet), swift,
repeated if need be, but never established (this is not a theater
of subversion: no great contestatory apparatus). For instance, if
there is a field buried under the smooth layer of the quotidian
logosphere, it is certainly that of class relations; now, Brecht
does not subvert this field (this is not the role he assigns to his
dramaturgy; moreover, how would a discourse subvert these
relations?), he imprints a shock upon it, sticks in a bell-headed
pin: for example, it is Puntila's drunkenness, a temporary and
recurrent laceration, imposed upon the sociolect of the big
landowner; contrary to so many scenes of bourgeois theater and
cinema, Brecht never deals with drunkenness as such (the sticky
tedium of boozer's scenes): drunkenness is never anything but
the agent which modifies a relation, and consequently offers it to
be read (a relation can be read only retrospectively when some
where, at some point, however remote or tenuous, this relation
has altered). Alongside so exact a treatment (exact because kept
to its strictest economy), how absurd seem most films about
"narcotics"! Using the alibi of the underground, it is drugs "as
such" which are always represented, their evil effects, their
ecstasies, their style, in short their "attributes," not their func
tions: does this representation permit a critical reading of some
supposedly "natural" configuration of human relations? Where
is the reading-shock?
Brecht and Discourse "5

Rehearse softly
In his political texts, Brecht gives us a reading exercise: he reads
us a Nazi speech (by Hess) and suggests the rules for a proper
reading of this kind of text.
Thus, Brecht joins the group of Exercise-Givers, of "Regu
lators"; those who give not regulations but regulated means for
achieving a goal; in the same way, Sade gave rules for pleasure
(it is a veritable exercise that Juliette imposes upon the lovely
Countess de Donis), Fourier those for happiness, Loyola those
for communication with me Divine. The rules taught by Brecht
aim at reestablishing the trutb of a text: not its metaphysical (or
philological) trutb, but its historical truth: the truth of a gov
ernmental script in a fascist country: an action-truth, a truth
produced and not assened.
The exercise consists in saturating the mendacious text by
intercalating between its sentences the critical complement which
demystifies each one of them: "Legitimately proud of the spirit
of sacrifice . . ." Hess pompously began, in the name of
"Germany"; and Brecht softly completes: " Proud of me
generosity of those possessors who have sacrificed a Iitt1e
of what the non-possessors had sacrificed to them . . . "

and so fonh. Each sentence is reversed because it is supple

mented: the critique does not diminish. does not suppress,
it adds.
In order to produce the proper supplement, Brecht recom
mends rehearsing the text, the exercise, softly. The critique s
i first
produced in a kind of clandestinity: what s i read is the text for
oneself, not in tself;
i the low voice is the one that concnru me: a
reHexive (and sometimes erotic) voice. producing what is intel
ligible, the original voice of reading. To repeat the exercise (to
read the text several times) is gradually to liberate its "supple
ments"; thus, the haiku compensates for its conspicuous brevity
by repetition: the tiny poem is murmured three times, in echoes;
this practice is so well coded that the amplitude of the supple
ments (the "length of the resonance") hears a name: hibi/ti; as

Malena :>m d reitl al

216 Readings
for the infinity of links liberated by repetition, this is called
utsuri.
What is astonishing, at the endurable limit of the paradox, is
that this refined practice, closely linked to an eTOtics of the text,
is applied by Brecht to the reading of a hateful text. The
destruction of monstrous discourse is here conducted according
to an erotic technique; it mobilizes not the reductive weapons
of demystificalion but rather the caressses, the amplifications,
the ancestral subtleties of a literary mandarinate, as if there
were not, on one side, the vengeful rigor of Marxist science (the
science which knows the reality of fascist spet'1:hes) and, on the
other, the complacencies of the man of leuers; but rather as if
it were natural tolaM pleasure in lhe truth, as if one had the simple
right, the immoral right to submit the bourgeois text to a critique
itself formed by the reading tt'1:hniques of a certain bourgeOis
pasl; and indeed where would the crilique of bourgeois discourse
come from if not from Ihat discourse ilself? Discursivity is, till
now, without alternative.

Concatenation
Because they are concatenated, Brecht says, errors produce an
illusion of Iruth; Hess's speech may seem true, insofar as it is
successive. Brecht questions concalenation, questions successive
discourse; all the pseudo-logic of the discourse--links, transi
tions, the patina of elocution, in short the continuity of speech
releases a kind of force, engenders an illusion of assurance:
concalenated discourse is indeslructible, triumphant. The first
attack is therefore to make it discontinuous-to discominue it:
literally to dismember Ihe erroneous text is a polemical acl. "To
unveil" is not so much to draw back the veil as to cut it to pieces;
in the veil, one ordinarily comments upon only Ihe image of
that which conceals, bUI the OIher meaning of the image is also
important: Ihe smooth, Ihe suswined, the successive; to anack the
mendacious text is to separate the fabric, to tear apart the folds
of the veil.

Matena dlreitl
Brecht and Discourst "7
The critique of the continuum (here applied to discourse) is a
constant one in Brecht. One of his first plays, In the Junglt of
C!Us.
i still seems enigmatic to many critics because in it two
partners take part in a duel incomprehensible not on the level
of each of its peripeties but on the level of the whole, i.e.,
according to a continuous reading: Brecht's theater is henceforth
a series (not a consequence) of cutup fragments deprived of
what in music is called the Zeigarnik effect (when the final
resolution of a musical sequence retroactively gives it its mean
ing). Discontinuity of discourse keeps the final meaning from
"taking": critical production does not wait-it will be instanta
neous and repeated: this is the very definition of epic theater
according to Brecht. Epic is what cuts (shears) the veil. disag
gregates the stickiness of mystification (see the preface to
Mahngonny).

The maxim
Brecht's praise of the fragment (of the scene presented "for
its own sake") is not that or the maxim. The maxim is not a
rragment; first or all. because the maxim is generally the point
of departure or an implicit reasoning, the outset or a continuity
surreptitiously developing in the docile inter-text which inhabits
the reader; then, because the Brechtian fragment never gen
eralizes-it is not "concise." it does not "assemble"; it can be
loose. relaxed. red on contingencies, specifications. dialectical
donntles; whereas the maxim is a statement minus History: it
remains a bluff or "Nature."
Hence, Brecht's unceasing supervision of the maxim. The
Hero is doomed, one might say, because the maxim is his
"natural" language ("Wherever you find great virtues, you can
be sure that something is going wrong"); the same applies to
widespread Custom, ror it is based on gnomic truths: "He who
takes the first step must also take the second": who says this,
and in this rorm? The cultural code, whose raise logic is abusive,
for he who takes the first step does not necessarily have to take
the second. To break the custom is, first of all, to break the
maxim, the stereotype: under the rule, discover the abuse;
under the maxim, discover the concatenation; under Nature,
discover History.

Metonymy
In his speech, Hess constantly speaks of Germany. But Germany,
here. is only the German "possessors." The Whole is given,
abusively, for the part. Synecdoche is totalitarian: it is an act of
force. "The whole for the part"-this definition of metonymy
means: one part against another part. the German possessors
against the rest of Germany. The predicate ("German") becomes
the subject ("the Germans"): there occurs a kind of local Putsch:
metonymy becomes a class weapon.
How to combat metonymy? How, on the Inlel of discQUTSe, to
restore the sum to its parts, how to undo the abusive Name?
This is a very Bre<:htian problem. In the theater, the undoing
of the Name is easy enough, for it is inevitably only bodies that
are represented there. If we must speak of the " People" on the
stage (for this word itself can be metonymic, can engender
abuses), we must divide up the concept: in The Trial of Lucullw,
the " People" is the meeting of a peasant, a slave, a schoolmaster.
a fishmonger, a baker, a prostitute. Bre<:ht says somewhere that
Reason is never what the totality of reasonable people think:
the (invariably abusive?) concept is reduced to a summation of
historical bodies.
However, de-nomination---or ex-nomination-because infi
nitely subversive. is difficult to sustain. It is tempting to exculpate
a Cause, to excuse the errors and stupidities of its partisans,
separating the excellence of the Name from the imbecilities of
its subjecl.5. Berdyaev once wrote a brochure entitled On the
Dignity of Christianity and 1M Indignity of Christians . . . Ah, if we
could similarly purify Marxist discourse of the dogmatism of
Marxisl.5. the Revolution of the hysteria of revolutionaries, and
in a general way the Idea from the neurosis of its supporters!

Malena m dlreitl
Brecht and Discourse 219
,

But in vain: political discourse is fundamentally metonymic. for

it can only be established by the power of language. and this
power is metonymy itself. Thus. there reurs in discourse the
major religious figure, that of Contagion, of Fault, of Terror,
i.e., in all these cases, the subjetion by violence of the part [Q
the whole, of the body to the Name; religious discourse is indeed
the model of all political discourse: no theology could acknowl
edge that Faith is merely the entirety of those who believe. Now,
from the viewpoint of Marxist "custom," Brecht is very heretical:
he resists all metonymies; there is a kind of Brechtian individ
ualism: the "People" is a collection of individuals assembled on
the stage; the "Bourgeoisie" is here a landlord, there a rich
man, etc. The theater compels undoing the Name. I can readily
imagine some theoretician, ultimately disgusted with Names yet
reluctant to abandon all language-l can imagine this Brechtian
epigone renouncing his past speehes and resolving henceforth
to write only novels.

The sign
Ys, Brecht's theater is a theater of the Sign. But if we want to
understand how and whereby this semiology can be, more
profoundly, a seismology, we must always remember that the
originality of the Brechtian sign is that it is to be read twice over:
what Brecht gives us to read is, by a kind of disengagement,
the reader's gaze, not diretiy the object of his reading; for this
object reaches us only by the act of intellelion (an alienated
act) of a first reader who is already on the stage. The best
example of this " turn," paradoxically, 1 should borrow not from
Brecht but from my personal experience (a copy is readily more
exemplary than the original; "Brecht-like" can be more Brech
tian than "Brecht").
Here then is a "street scene" of which 1 was a wilOess. The
public beach of Tangier, in summer, is carefully supervised;
one is not permitted to undress there-not out of modesty, no
doubt, but rather to compel bathers to rent the cabanas which
220 Readings

line the promenade i.e., to keep the "poor" (this category exists
in Morocco) off the beach, thereby reserved for the bourgeois
and the tourists. On the promenade, an adolescent boy, solitary,
sad, and poverty-stricken (signs for me, I confess, deriving from
a simple reading, which is not yet Brechtian), is walking along;
a policeman (almost as filthy as the boy) passes him and looks
him up and down, I see his scrutiny, I see it reach and linger
over the shoes; then the cop orders the boy off the beach.
This scene invites two commentaries. The first will accom
modate our indignation provoked by the barricading of the
beach, the grim subjection of the boy, the arbitrary action of
the police, the segregation of money, the Moroccan regime;
now, this commentary would not be Brecht's (though this would
certainly be his "reaction"). The second commentary will estab
lish the mirror action of the signs; it will note first of all that
there is a feature in the boy's garments which is the major sign
of poverty: the shoe; it is here that the social sign explodes in
all its violence (there used to be, not so long ago, in the days
when we had "the poor," a mythology of the cast-off shoe: if
the intellectual rots from his head down, like fish, the poor
man rots from the feet up which is why Fourier, seeking to
invert the civilized order, imagines a corps of flamboyant
cobblers); and, in the realm of the shoe, the extreme point of
poverty is the old slipper, without laces, the upper flattened
beneath the heel, precisely in the fashion exhibited by the boy.
But what this second commentary would especially note is that
this sign is read by the cop himself: it is when his gaze, descending
the body's length, perceives the wretched shoe, that the police
man, with a single impulse, by a veritable paradigmatic leap,
classifies the boy among those to be expelled: we understand
that he has understood and why he has understood. The
action may not stop here: the cop himself is almost as ragged
as his victim: except, precisely, for his shoes! Round, shiny,
solid, old-fashioned, like all policemen's shoes. Whence we can
read two alienations confronting one another (a situation sketched
in a scene from a neglected play by Sartre, Nekrassov). Our
222 Readings
greatest of all: the art of living"; hence, it is less a matter of .
making pictures than furniture, clothes, tablecloths, which will
have distilled all the juice of the "fine" arts; the socialist future
of art will therefore not be the work (except as a productive
game) but the object of use, the site of an ambiguous flowering
(half functional, half ludic) of the signifier. The cigar is a
capitalist emblem, so be it; but if it gives pleasure? Are we no
longer to smoke cigars, to enter into the metonymy of the social
Fault, to refuse to compromise ourselves in the Sign? It would
be hardly dialectical to think so: it would be to throw out the
baby with the bathwater. One of the tasks of a critical age is
precisely to pluralize the object, to separate pleasure from the
sign; we must de-semanticize the object (which does not mean
de-symbolize it), give the sign a shock: let the sign fall, like a
shed skin. This shock is the very fruit of dialectical freedom:
the freedom which judges everything in terms of reality, and
takes signs conjointly for operators of analysis and for games,
never for laws.

L'Autre Scene, 1975

 TW O

F.B.*

1 . Splinters of language

Though F.B.'s texts may well be the premonitory signs of a

concerted, large-scale work, the author puts his reader under no
obligation, and what each of these texts has to tell us is its
fulfillment. What is fulfilled, here, is wr-iting. Of all the work's
substances, only writing, as a matter of fact, can be divided
without ceasing to be total: a fragment of writing is still an
essence of writing. This is why, willy-nilly, every fragment is
finished, from ' the moment it is written; this is why, too, we
cannot compare a broken work to a sustained one; this is why,
lastly, no one can deny the greatness of fragmentary works: the

greatness not of ruin or of promise, but the greatness of the

silence which follows any fulfillment (only scholarship, which is
the contrary of reading, can regard Pascal's Pensees as an
uncompleted work). Because they are written, F.B.'s texts are
neither sketches nor notations nor raw materials nor exercises;
they suggest neither the notebook nor the diary: they are splinters
of language. Poe once claimed there was no such thing as a long
poem; for example, he saw Paradise Lost as "a succession of

Unpublished, this text was written as a footnote to fragments by a young

writer who seems not to have pursued a literary vocation and who published
nothing. A marginal text, then, intended for the person whose enterprise it
examines. Here tone and address are distinctly ludic, which does not prevent
this text-<\uite the contrary-from constituting a system of acute propositions
on a new type of fictive writing. We can recognize in nucleo, as early as '964,
certain features of Barthes's final manner.-Ed.

223
224 Readings
poetical excitements interspersed, inevitably, with corresponding
depressions. . . . " F.B. manages to eliminate these depressions;
his writing is of a luxuriance without loss, i.e., without duration;
it is the writing itself, and not the story, which ceases, here, to
be uneven, hence boring, hence periodically ugly, as happens
in so many beautiful works: everything is referred to writing,
but this delegation has nothing to do with the effort of form;
artisanry is no longer the necessary condition of style; Stendhal
scoffed at Chateaubriand and virtually never "corrected." Here
the writer gives his effort not to the verbal substance but to the
decision to write: everything happens before writing. The least
of F.B.'s texts bespeaks this anterior "transumption"; the tender
and sumptuous luxury of an absolutely free writing, in which
there is not an atom of death, invulnerable by dint of grace,
expresses the initial decision which makes language into the
fragile salvation of a certain suffering.

2. Incidents

The power of writing: these texts are also, in their way, splinters
of a novel. F.B.'s texts show two indestructible signs of the
novel: first, the uncertainty of the narrative consciousness, which
never clearly says he or I; then a cursive manner, i.e., a continuity
which relates writing to the sustained forms of nature (water,
plant, tune); you "sample" nothing from a novel, rather a novel
is "devoured" (which means that the sustained nature of nov
elistic reading derives not from the care you might take in
reading anything but, quite the contrary, from the rapid trajec
tory which makes you forget certain parts of the itinerary;
writing's continuity is a matter of speed, and this speed is perhaps
ultimately no more than that of the hand). Thus with F.B.'s
texts: they, too, are "devoured": a very small space of -1O:ords
encloses here (the pa;'adox of writing) an essence of continuity.
F.B.'s writing, once it is completed (always too soon), nonetheless
has already flowed past: light, profound, luminescent as the sea
it often speaks of, it leads us, gives us at once the idea of a goal
 Readings

accurately the boy is described, the more curious we are about

his essence, the more we are focused on something we must
understand. F.B. thus substitutes for the grammar of the anecdote
a new intelligibility: that of desire. Desire itself becomes story

and intelligence, there is, finally, coincidence of description and

suspense. In a novelistic description, if it isn't too bad, the story
remotely penetrates every detail, making each contribute to a
general meaning (the poverty of a dwelling, the austerity of a
character); here, in the same way, desire makes description
"profound," or, one might say, alienated: desire becomes ratio,
logos: a power it cannot derive from its satisfaction, but only
from speech, whereby all of literature is justified. Just as the
anecdote always overflows toward a certain meaning, which has
long been called destiny, so desire, once recounted, mysteriously
loses its contingency: embarrassment, sadness, lucidity, sleep,
the city, the sea become the names of desire. Whence this new
literature which works at once by metaphor and by narrative,
by the variation of being and by the concatenation of acts:
something like a new La Bruyere or Theophrastus Characters
not of manners but of bodies.

4. Sublimation
Thereby, F.B. silences not only narrative's morality but also its
logic (which is perhaps the same thing); his descriptions are
subversions, they do not lead on, they detach and "exceed."
How? Each text starts like a novel, each text is a simulacrum of
a novel: there are objects, characters, a situation, a narrator, in
short a realistic instance; but very quickly (i.e., both instantly
and imperceptibly, as if we were leaving the ground), this whole
familiarity of the novel begins to move elsewhere: we are lifted
toward another meaning (what will be given of this meaning is
nothing more than this: it is other; a pure alterity, which is the
sufficient definition of the strange): a character arrives at a
railroad station; the station is described, then suddenly it is the
site, or better still, the triumph of desire; now, this identity is
F.B. 22 7
immediate: the station does not become other than itself, there
is no metaphor, no transport of vision; by a special illogicality,
we receive the succession and the coincidence of the two sites.
This very special montage effaces something which has been
very difficult for literature to get rid of: the astonishment of its
own notations; F.B.'s writing is never to any degree accessory
to the effect it produces: . it is a writing without complicity.
Another text begins like an adventure story: a man makes his
way into an airplane hangar and knocks out the pilot sleeping
there; very quickly, an "excessively" amorous description of the
young pilot (everything is in this "excess") alienates this classical
start; the hallucination "takes," and without leaving the frame
work of the traditional narrative, the scene of escape changes
its nature and finds itself an erotic scene. For F.B. the novel is at
discretion; it lends desire its inceptions; the narration is like a
launching pad; but what happens at the end no longer belongs
to the order of the successibility of events, in other words of
suspense, but to the order of essences. In the (real) novel, desire _

is strong by its acts, its effects, the situations it produces; it is

always treated according to a causal logic (which moralizes it at
every turn); in the simulated novels of F.B., everything stops at
desire, everything glorifies it (theologically, glorification is the
manifestation of essence); the novel yields like a curtain parting
in order to show desire in its "glory." A truth of reversals: desire
sublimates reason.

5. Eros
Of course, desire prowls through all literature, ever since
language, having become sovereign, useless, began saying some
thing which has been called beauty; but this written desire has
never hitherto been anything more than an element of moral,
psychological, theological algebra: literature once served to
comprehend desire, in the name of a larger whole; all literature
thus tended to morality, i.e., to an economy of good and evil,
of light and dark: an Eros recounted means something else than
 Readings

Eros. In F.B.'s texts, the movement is reversed: it is Eros which

"comprehends"; here there is nothing which does not proceed
from Eros: boy-love forms a perfect circle outside of which
nothing is left; all transcendence is concentrated ; yet this circle
is a formal one: its closure does not come from society, or even
from an existential choice, as in other works with the same
object: it is only writing which traces that circle; the desire for
boys is never, here, "culturalized," it has the naturalness of what
is without cause and without effect, it is both without freedom
and without fatality. This naturalness has major consequences
for writing (unless, of course, it derives from writing): what is
written does not appeal to something else; both soft and rich,
writing is nonetheless matte; conforming in this to the newest
languages of today, but without their coldness, it denies itself
and us any induction; because there is no ellipsis in them, we
can infer nothing from these texts. Now, the value of an art, in
a crowded world, is defined by the privative operations it has
the audacity to invoke: not in order to satisfy an aesthetic of
constraint (the classical model), but in order to subjugate mean
ing fully, to deprive it of any secondary outlet. We might say
that, coming at the end of a very heavy tradition, a literature
of desire is the most difficult thing of all; F.B.'s does not derive
its erotic essence from the realism of figures, but from an
unconditional submission to Eros, chosen as the sole god of the
work (Satan is eliminated, and hence God). This regnum as
sured, nothing would seem more out of place than an erotic
repertoire of gestures. F.B.'s texts are therefore not in the erotic
tradition (in the current meaning of the term), precisely to the
degree that Eros here is not a collection and a nomination (of
"postures"), but a sovereign principle of writing. Hence, we
must set in opposition to traditional erotics a new eroticism; in the
former case, the writer must work up the description of "what
happened" until he has found in Eros a transcendence God,
Satan or the Unnamed whereas in F.B.'s "incidents," Eros
being the ultimate intelligence, there can be no paroxysm.
Another difference: traditional erotics are heavy, or tense; here
F.B. 2 29

eroticism is light (writing runs over the surface of the encounters

without completing them) and profound (writing is the thought
of things); it is an air, a space, one might say a geometry, since
we now have geometries which subtilize the cosmos; it is present,
without provocation and without complicity: not naive, for Eros
knows everything, it is wise; and perhaps this is the extreme
note of this writing, that in it desire is a figure of sophrosyne.
Grace and wisdom: it is this impossibility which the Ancients hold
to be perfection, representing it in the lovely myth of the puer
senilis, the adolescent master of every human age. How long it
has been that our literature, in the best of cases, could transport
but not seduce; such a chaml, then, is a new thing.

6. General, individual, particular

The tremor of romantic Sehnsucht, consisting of a dreamy
confusion of the sensitive and the sensual, and yet a deep
metaphysical silence: F.B. takes from language that category
of the general only the extreme, particular edge, never inti
mating a judgment, a maxim, never summarizing the description
under that lyric or moral discourse which the old rhetoric had
recognized as epiphoneme: in F.B.'s writing, nothing ever comes
over what is written: a silky and inductile metal. F.B. occupies,
among our various writing, a dangerous situation. Language
being general (and therefore moral), literature is doomed to
the universal; everything that happens in literature is originally
cultural: its puis ions are always born clad in an anterior lan
guage; the generality with which the writer has been credited
for centuries, endlessly congratulated for making the individual
into the human, is in reality a terrible servitude: how can one

be praised for a constraint imposed by the very t;lature of

language? The writer's problem, on the contrary, is to find an
ultimate particularity despite the general and moral instrument
he is given. This is the problem which is treated (but not argued)
in F.B.'s texts; here the author teaches himself (and us) that the
particular is not the individual; quite the contrary, it is, so to speak,

 Readings

the impersonal and uncoUective share of ourselves; hence, we

shall not find in these texts anything dealing with a formed
person, i.e., with a history, a life, a character; but we shall also.
not find here any mirror of humanity. In other words, the
substance of this writing is not experience (experience, "what has
been lived," is banal it is precisely what the writer must combat),
but it is also not reason (a general category adopted under
various pretexts by all ready-made literatures); this famous
conflict, so apparently irreducible in some people's eyes that it
keeps them from writing, is one whose terms F.B. rejects, and
it is by this innocent rejection that he is likely to fulfill the utopia
of a particular language. This action has a great critical conse
quence: though F.B.'s texts can be described as being, nothing
in the world can keep them from becoming: achieved in writing,
the particular struggles here with the work which every society,
being moral, demands of the one who writes.

7. Technique

Literature's substance is the general category of language; in

order to create itself, not only must it kill what has engendered
it, but even, in order to commit this murder, it can use no other
instrument than this very language it must destroy. This almost
impossible reversal constitutes F.B.'s texts: it is that almost which
is the narrow space in which the author writes. This cannot be
done without a technique, which is not necessarily an apprentice
ship, but according to Aristotle's definition, the faculty of
producing that which can be or not be. The goal of this technique
is to describe a world chosen, not as a desirable world, but as
the desirable itself; desire is here not the attribute of a creation
which preexists it, it is immediately a substance; in other words,
again, the author does not discover (by the action of a privileged
subjectivity) that the world is desirable, he determines it desir
able; hence, it is the time ofjudgment, psychological time, which
is eluded here: particular, but not individual, the author does
not recount what he sees, what he feels, he does not reel off the


F.B.
precious epithets he has the fortune of finding, he does not act
as a psychologist who uses a felicitous language in order to

enumerate the original attributes of his vision, rather he acts

immediately as a writer; he does not make bodies desirable, he
makes desire corporeal, inverting substance and attribute by the
very paradox of writing: everything is shifted to objects, in
order to express not what they are (what are they?) but the
essence of the desire which constitutes them, exactly as lumi
nescence constitutes phosphorus; in F.B.'s texts, there is never
any undesirable object. Thus, the author creates a vast metonymy
of desire: a contagious writing which transfers to its reader the
very desire out of which it has formed things .

8. Signum {aeere
The old rhetoric distinguished disposition from elocution. Dispo
sition (taxis) accounts for the work's major units, its general
arrangement, its "development"; elocution (lexis) accounted for
the figures, the turns of speech, what we should call today the
writing [ecTiture], i.e., a class (and not an epitome) of "details."
F.B.'s texts are fully (at least for the moment) texts of elocution.
The unit of elocution has a very ancient name: it is the song.
The song is not a euphony or a quality of images; according to
the Orphic myth, it is a way of keeping the world under one's
language. What sings here is not directly the words, it is that
second writing, that mental writing which forms itself and
advances between the things and the words a kind of anterior
song (as Baudelaire speaks of a vie anteTieure, a previous exist
ence). Vico at one point mentions certain universals of the imag
ination: that is the space where F.B. forms a particular writing,
without tradition and without provocation; neither "noble" nor
"natural," this writing eludes all the models without ever assum
ing the heavy signaletics of originality. Whence, perhaps, its
naked friendliness, severed from any humanism. To read F.B.
is constantly to form in oneself certain adjectives: fresh, simple,
silky, light, sensitive, accurate, intelligent, desirable, strong, rich
 Readings
(Valery: "After all, the artist's sole object comes down to securing
an epithet'" but ultimately these adjectives dislodge each other,
truth is only in the whole and the whole cannot support any
definition; the very function of this writing is to say what we
could never say about it: if we could, it would no longer be
justified. F.B. stands at the precise point of a double postulation:
on the one hand, his writing makes meaning, whereby we cannot
name it, for this meaning is infinitely more remote than our
selves; and on the other, it makes a sign. Signum facere, such might
be the motto of these texts: these sentences, this entirety of
sentences floats in the mind like a future memory, predeter
mining the discourse of the latest modernity.

 The oque Side

French culture has always attached, it appears, a very powerful

privilege to "ideas," or, to speak more neutrally, to the content
of messages. What matters to the Frenchman is having "some
thing to say," what we nowadays designate by a phonically
ambiguous word with monetary, commercial, and literary ap
plications: Ie fond (or Ie fonds or les fonds). With regard to the
signifier (a word we hope can henceforth be used without
apologies), French culture has known for centuries only the
labor of style, the constraints of Aristotelio-Jesuit rhetoric, the
values of "writing well," themselves centered, moreover, with
an obstinate iteration, upon the transparency and distinction of
the ''fond. '' We had to wait till Mallarme for our literature to
conceive a free signifier no longer burdened by the censure of
the false signified, and to attempt writing finally rid of the
historical repression in which the privileges of "thought" im
prison it. Even the Mallarmean project so stubborn is the
resistance can only be, here and there, "varied," i.e., repeated,
in infrequent works which are all works of combat: suppressed
twice in our history at the moment of the baroque impetus
and of Mallarmean poetics French writing is still in a situation
of repression.
Here is a book to remind us that besides cases of transitive
or ethical communication (Pass the cheese or We sincerely desire
peace in Vietnam), there is a pleasure of language, of similar
fabric, similar silk as erotic pleasure, and that this pleasure of
language is its truth. This book comes not from Cuba (no
question of folklore, even Castrist) but from the language of
Cuba, from that Cuban text (cities, words, drinks, garments,
bodies, odors, etc.), which is itself an inscription of diverse
 2 34 Readings
cultures and periods. Here something is happening which
matters to us as Frenchmen: transported into our language, this
Cuban language subverts its landscape: here is one of the very
rare occasions when a translation manages to displace its lan
guage of origin instead of merely joining it. If the verbal baroque
is historically Spanish (Gongoresque or Quevedian), and if that
history is present in Severo Sarduy's text, national and "mater
nal" like any language, this text also reveals to us the baroque
side of the French idiom, thereby suggesting that writing can
do anything with a language, and in the first place give it its
freedom.
This baroque side (baroque is a temporarily useful word, in
that it allows us to provoke the inveterate classicism of French
literature), to the degree that it manifests the signifier's ubiquity,
present at all levels of the text and not, as is commonly said, on
its surface alone, modifies the very identity of what we call a
narrative, without the tale's pleasure ever being lost. Ecrit en
dansant consists of three episodes, three gestes an old French
word which here recuperates the title of Severo Sarduy's first
book and which corresponds as well to the word's masculine
sense (gesture) as its feminine (exploit) but in it will be found
none of those narrative prostheses (personality of the protago
nists, situation of the locales and of the weather, complicity of
the narrator, and God, who sees into the hearts of the characters)

by which we usually mark the abusive (and, moreover, illusory)

right of reality over language. Severo Sarduy is a good narrator
of "something," which draws us to its end and makes for the

death of writing, but this something is freely displaced, "se-

duced" by that sovereignty of language, which Plato in fact sought
to challenge in Gorgias, inaugurating that repression of writing
which marks our Western culture. Thus, we find deployed in
Ecrit en dansant, a hedonist text and thereby a revolutionary one,
the great theme proper to the signifier, sole predicate of essence
it can actually support, metamorphosis: Cubans, Chinese, Span
iards, Catholics, addicts, pagans, performers, traveling from
caravelles to cafeterias and from one sex to the other, Severo
Tht 80ToqtU Side '35
Sarduy's creatures pass back and fonh lhrough the pane of
refined prattle which they fob off on the author, thereby
demonstrating that this pane does not exist, that there s
i nothing

to Ste behi1ld language, and that discourse, far from being the
final attribute and the last touch of the human statue, as the
misleading myth of PygmaJion suggests, is never anything but
its irreducible scope.
However, let humanists be reassured, at least partially. The
allegiance given to writing by any subject, the one who writes
and the one who reads, an act which has no relation with what
classical repression, by a self-seeking ignorance. calls "verbalism"
or more nobly "poetry," suppresses none of reading's "plea
sures," provided we agree to find its correct rhythm, Severo
Sarduy's text deserves all the adjectives which constitute the
lexicon of literary value: it is a brilliant, lively, sensitive, funny,
inventive text, unexpected yet clear. cultural even, and contin
uously affectionate. Yet I fear that, in order to be received
without difficulty in literary good society, it lacks that suspicion
of remorse, that touch of transgression. that shadow of the
signified which transforms writing into a sermon and thus
ransoms it under the name of "fine work," like a piece of
merchandise useful to the economy of the "human." Perhaps
this text does have one thing in excess, which will embalTIl55:
the energy of speech, which suffices for the writer to be:
reassured,

La Quimaine liUbairt, 1967

 What Becomes of the Signifier

Eden, Etkn, Eden. is a free text: free of any subject, of any object,
of any symbol: it s
i wriuen in that recess (that abyss or that
blind spot) where the traditional constituents of discourse (he
who speaks, what he tells. how he expresses himself) are de trap.
The immediate consequence is that criticism, since it can speak
neither of the author nor of his subject nor of his style, can do
notbing with this text: one must "enter" GuYOtat's language;
not believe in it. be the accomplice of an illusion, participate in
a hallucination, but write this language with him, in his place,
sign it at the same time as P. Curotal himself.
To be in language (as we say: to be in on the deal): this is possible
because Cuyamt produces not a manner, a genre, a literary
object. but a new element (why not add it to the four Elements
of the cosmogony?); this element is a sentence: substance of
speech which has the special nature of a fabric, or a foodstuff,
a single seOlence which never ends, whose beauty derives not
from its "report" (the reality to which it is presumed to refer)
but from its respiration, interrupted, repeated, as if it were the
author's business to represent for us not imagined scenes but
the scene of language, so that the model of this new mimt'5u is
no longer the adventure of a hero but the adventure of the
signifier itself: what becomes of it.
Eden, Eden, Eden constitutes (or should constitute) a kind of
upsurge, of historic shock: a whole anterior action, apparently
twofold. but whose coincidence we see more and more dearly,
from Sade to Genet, from Mallarme to Artaud. is collected.
displaced. purified ofiu period circumstances; there is no longer
either Narrative or Transgression (no doubt one and the same).
there is nothing left but desire and language, not the latter

236
What Becomes of the Signifter ' 37
expressing the former, but placed in a reciprocal, indissoluble
metonymy.
The strength of this metonymy, sovereign in GUyOlat's text,
suggests that a strong censorship is likely-a censorship which
will find united here its two habitual quarries, language and
sex; but also such censorship, which can take many forms, will
be immediately unmasked by its very strength: doomed to be
excessive if it censors sex and language at the same time, doomed
to be hypocritical if it claims to censor only the subject and not
the form, or conversely: in both cases, doomed to reveal its
essence as censorship.
However, whatever the institutional peripeties may be, the
publication of this text is important: all critical, theoretical work
will be advanced by it, without the text ever ceasing to be
seductive: at once unclassifiable and indubitable, a new reference
and a departure for writing.

Preface to P. Guyotafs Edna, Edna, Edna (Gallimard), 1970

Malena :>m d reitl al

 Outcomes of the Text

Here is a text by Georges Bataille "The Big Toe."

I shall not explicate this text; I shall merely produce a number
of fragments which will be, in a sense, outcomes of the text.
These fragments will be in a more or less emphatic state of
severance with each other: I shall not attempt to link, to organize
these "outcomes"; and in order to be sure of frustrating any
liaison (any systematizing of the commentary), in order to avoid
any rhetoric of "development," of the developed subject, I have
titled each of these fragments, and 1 have pUi these titles in
alphabetical order-which is, of course, both an order and a
disorder, an order stripped of meaning. the degree zero of
order. It will be a kind of dictionary (Bataille supplies one at
the end of Documro.ls) which will deal obliquely with the suslain
ing text.

Aplatlssement des vaieurs I Deflation of values

There is, in Nietzsche and in Bataille, one theme in common:

that of regret. A certain form of the present is disparaged. a
cenain form of the past is exalted; neither this present nor this
pan is actually historical; they are both read according to the
formal, ambiguous movement of a decadence. Thus is born the
possibility of a non-reactionary regret, a progressive regret.
Dectuhnce is not read, contrary to the word's accepted connota
tion, as a sophisticated, hyperculturaJ condition, but on the

Finl published in Documents in 199: reprinted in the fiTS! volume of

Biluille's Onwres cOllpliw, 1970. Tn.nslilled in 1985 by Alliin Sleek.! ilS lhe
Big Toe," in Vuions 0{ ExuS$: Stucud Wrilillp 0{ Gwrgts B4UJilu, 19'7-1939
(Minnesota University Press).

Malena m dlreitl
Outcomes of lhe Tnt '39
contrary as a deflation ofvaltUJ: return of tragedy as farce (Marx),
clandestinity of festal expenditure in bourgeois society (Bataille),
critique of Germany. disease, exhaustion of Europe, theme of
the last man, of the vermin "that diminishes everything" (Nietzsche).
We might add Micbelet's diatribes against the nineteenth cen
tury-his own-the century of Boredom. In all, the same disgust
provoked by bourgeois deAation: bourgeois man d not
destroy value, he deflateJ it, diminishes it, establishes a system of
the paltry. This is a theme at once historical and ethical: fall of
the world out of the tragic, rise of the petite-bourgeoisie, written
as an advent: the Revolution (Marx) and the Ubermtruch (Nietzsche)
are vital shocks applied to deAation; all of Bataille's heterology
is of the same order: electric. In this apocalyptic history of value.
''The Big Toe" refers to two time frames: an ethnological time
(marked in the text by verbs in the present tense), the time "of
men," "of peoples" who anthropologically disparage the low
and exalt the high, and a historical time (marked by episodes
in the past tense), which is the time of Christianity and of i15
quintessence, Spain. for which the low s
i purely and scrupulously
censured (modesty). Such is the dialectic of value: when it is
anthropological, rejection of the foot designates the very site of
a seduction: seduction is where one Jaoogely conceals, value is
in the savage transgression of the forbidden; but when it is
historical, sublimated in the figure of modesty, condemnation
of the foot becomes a repressed, deflated value which invites
the denial of LaughuT.

Codes du s.3vo/r I Codes of knowledge

In Bataille's text, there are many "poetic" codes: thematic
(high I low , noble I ignoble, light I muddy), amphibological (the
word erection, for instance), metaphorical ("man is a tree"); there
are also codes of knowledge: anatomical, zoological, ethnological,
historicaL Of course, the text exceeds knowledge by value; but
even within the field of knowledge, there are differences of
pressure, of "seriousness," and these differences produce a

GJPYnghtcd maknal
 Readings

heterology. Bataille stages two knowledges. An endnxal knowl

edge: this is the knowledge of Salomon Reinach and of the
membf!rs of the editorial commiltee of Documtnts (the periodical
from which the text under consideration is taken); citational,
referential, reverential knowledge. And a remoter knowledge.
produced by Bataille (by his personal cuhure). The code of this
knowledge is ethnological; it corresponds to what was once
called Le Magasin pittoreque. a collection of linguistic, ethno
graphic "curiosities" ; in the discourse of this second knowledge
there is a double reference: that of the strange (of elsewhere) and
that of the detail; thus is produced an incipient collapse of
knowledge (of its law) by its futilization, its miniaturization; at
the end of this code, there is astonishment ("wide-eyed"); such
knowledge is paradoxical in that it astonishes, de-naturalizes
itself, unsettles the formula "it is self-evident . . ." This search
for ethnological fact s
i certainly very close to the novelistic
search: the novel in fact is a faded matlusis, tending to circumvent
knowledge. This interference of codes-diverse in origin, in
style-is contrary to the monoiogy of knowledge, which conse
crates "specialists" and disdains polygraphs (amateurs). In short,
there occurs a burlesque. kteroclik knowledge (etymologically:
leaning to one side and the other): this is already an operation
of writing (what we have elsewhere called ecrivance or inauthentic
writing imposes the separation of knowledges-as we say: the
separation of genres); proceeding from the mixture of knowl
edges, writing holds in check "the scientific arrogances" (as
Bataille calls them in Documents) and at the same time sustains
an apparent readability: a dialectical discourse which might be
that of journalism, if journalism were not deBated by the
ideology of mass communications.

Commencement I Beginning

The "bf!ginning" is a rhetorician's notion: How to begin a

discourse? For centuries. the problem has bf!en argued. Bataille
raises the question of the beginning where it had never

C ;.pvnghted malaria
 Readings
to the intrusion of a value: noble and ignoble (top and bottom,
hand and foot).

DeJouer I Baffling
Bataille's text teaches us how to deal with knowledge. We need
not reject it. We must even, occasionally, pretend to place it in
the forefront. It did not at all trouble Bataille that the editorial
committee of Documents consisted of professors, scholars, librar
ians. Knowledge must be made to appear where it is not
expected. As has been said, this text, which concerns a part of
the human body, discreetly but stubbornly avoids psychoanalysis;
the (discursive) play of knowledge is capricious, cunning: "high
heels" appear on the text's stage, yet Bataille eludes the expected
stereotype of the heel-as-phallus; and yet again, by a third turn,
Bataille immediately afterwards invokes sexuality, bringing it
on stage by a transition ("furthermore") that seems deceptively
naive. Knowledge is fragmented, pluralized, as if the one of
knowledge were ceaselessly made to divide in two: synthesis is
faked, baffled; knowledge is there, not destroyed but displaced;
its new place is in Nietzsche's word that of a fiction: meaning
precedes and predetermines fact, value precedes and predeter
mines knowledge. Nietzsche: "No fact exists in itself. What
occurs is a set of phenomena selected and grouped by a being
who interprets them . . . There is no such thing as a state of
fact; on the contrary, a meaning must be introduced before
there can be a fact." Knowledge, in short, would be an inter
pretative fiction. Thus, Bataille assures the baffling of knowledge
by a fragmentation of the codes, but more particularly by an
outburst of value (noble and ignoble, seductive and deflated). The
role of value is not a role of destruction, nor yet of dialectization,
nor even of subjectivization, it is perhaps, quite simply, a role
of rest . . . "it suffices for me to know that truth possesses a great
power. But it must be able to do battle, and it must have an
opposition, and from time to time one must rest from it in the
non-true. Otherwise, truth would become tedious for us, without

, ,
Outcomes of the Text 2 43

savor and without strength, and we would become so as well"

(Nietzsche). In sh ort, knowledge is retained as power, but it is
opposed as tedium; value is not what despises, relativizes, or
rejects knowledge, but what keeps it from being tedious, what
rests us from it; value is not opposed to knowledge according to
a polemical perspective but according to a structural meaning;
there is an alternation of knowledge and value, rest from one
in the other, according to a kind of amorous rhythm. And here,
in short, is what writing is, and singularly the writing of essays
(we are speaking of Bataille): the amorous rhythm of science
and value: heterology, delight.

Habllle / Dressed

In ancient Chinese cultures, a husband must not see his wife's

bare feet: "The Turks of Central Asia consider it immoral to
show their bare feet, and even go to bed in stockings." We
should extend the little ethnographic dossier constituted by
Bataille add North American petting parties; the custom of
certain Arab populations where the women do not undress
when making love; the habit reported by a contemporary
author of certain hustlers who remove every garment except
their socks. All of which would lead us to discuss the relation
of clothes and erotic conduct; this is not at all the abundantly
documented problem of striptease; for our society, which
regards itself as "erotic," never speaks of the real practices of
love, of the body in a state of love: this is what we know least
about each other not, perhaps, by ethical taboo, but by a taboo
of futility. In short, we must and this would not be so banal
as it seems we must rethink nudity. As it happens, for us the
nude is a plastic value, or even erotico-plastic; in other words,
the nude is always in a position of figuration (this is the very
example of striptease); closely linked to the ideology of repre
sentation, the nude is the figure par excellence, the figure of
the figure. To rethink the nude would therefore mean, on the
244 Readings
one hand, to conceive of nudity as a historical, cultural, Oc
cidental (Greek?) concept, and on the other, to transfer it
from the Tableau of bodies to an order of erotic practices.
Now, once we begin to glimpse the complicity of the nude
and representation, we are led to suspect its power of delight
(jouissance]. The nude is a cultural object (linked perhaps to
an order of pleasure, but not to that of loss, of delight), and
consequently, in conclusion, a moral object: the nude is not
perverse.

Idiomatique / Idiomatic
How to make the body talk? We Cjln transfer the codes of
knowledge (of that knowledge which deals with the body) into
the text; we can also take into account the doxa, the opinion of
people about the body (what they say about it). There is a third
means, to which Bataille systematically resorts (and which is
interesting from the viewpoint of contemporary work on the
text): this is to articulate the body not on discourse (that of
others, that of knowledge, or even my own) but on language: to
let idiomatic expressions intervene, to explore them, to unfold
them, to represent their "letter" (i.e., their significance); mouth
will lead us to "fire-mouth" (cannibal expression for cannon),
"close-mouthed" ("lovely as a strongbox"); eye will provoke a
complete exploration of all the idioms in which this word occurs;
the same for foot ("flat-footed," "stupid as a foot," etc.). By this
means, the body develops on the level of language: idiomatism
and etymologism are the signifier's two great resources (proof a
contrario: ecrivance, which is not writing [ecriture], but its in
authentic form, ordinarily censures the work of what, in lan
guage, is both its center and its excess; have you ever seen a
metaphor in a sociological study or in an article of Le Monde?).
Bataille engages in textual work of the same type, of the same
productive energy we see in operation, on stage, in Philippe
Sollers's Lois.
 Readings

Bataille's heterology consists in this: there is a contradiction,

a simple, canonical paradigm between the first two terms: noble
and ignoble ("the fundamental division of the classes of men into
noble and ignoble"); but the third term is not regular: low is not
the neutral term (neither noble nor ignoble), nor is it the mixed
term (noble and ignoble). It is an independent term, concrete,
eccentric. irreducible: the term of seduction outsU:le the (struc
tural) law.
The low in fact is a value on twO accounts: on the one hand,
it is what is outside the mimicry of authority;. on the other, it
is caught up in the paradigm high f low. i.e., in the simulation of
a meaning, of a form, and hence it baffles the nature of matter
in itself: ". . . contemporary materialism, by which I mean a
materialism not implying that maHer is the thing in itself.' In
short, the true paradigm is one which confronts twO positive
values (noble f low) in the very field of materialism; and it is the
normally contradictory term (ignoble) which becomes neutral,
mediocre (the negative value, whose negation is not contrariety
but deflation). Nietzsche once again: "What is it that is mediocre
in the average man? He does not understand that the wrong side
of things is necessary," In other words, once again: meaning's
apparatus is not destroyed (prattle is avoided), but it is made
eccentric, it is made insecure, wobbly (the etymological meaning
of "scandalous"). This process is assured by two operations: on
the one hand, the subject (of writing) deflects in extremis the
paradigm: mcxU.sly. for example, is not denied in favor of its
anticipated statutory and structural contrary (exhibitionism); a
third term appears: Laughter, which baffles Modesty, the meaning
of Modesty; and on the other hand, language itself is audaciously

.. "For above aU one must not submit oneself and one's reason to anything
higher, to anything which might give my being, and to the reason which arms
that being, a borrowed authority. My being and i1$ reason (an only submit, in
fact, to what is lown-, to what in any case cannot serve 10 ape a conventional
authority. Low matter is external and alien to ideal human aspirations and
refuses to allow itself to be re-duced to the great ontological machinery resulting
from these aspirations."-Docuonmts

Malena :>m d reitl al

Outcomts of the Ttxt '47
distended: low [bas] is used as a positive, approbative value ("the
low materialism of gnosis"), but its correlative adverb. which
according to language should have the same value as the original
adjective, is employed negatively, disparagingly ("the basely
idealistic orientation of Surrealism"): it is the theme of deflation
which separates-like a violent, severing value the root word
and its derivative.

Quol et qul?lWhat and who?

Knowledge says of each thing: "What is it?" What s

i the big toe?
What is this text? Who is Bataille? But value. according to the
Nietzschean watchword, prolongs the question: What is itforme1
In a Nieuschean way, Bataille's text answers the question:
What is the big toe for me, Bataille1 And by displacement: What is
this text, for me, the reader? (Answer: It is the text I would want
to write.)
Hence it is necessary-and perhaps urgent-to come OUI in
favor of a cenain subjectivity: the subjectivity of the non-subject,
opposed both to the subjectivity of the subject (impressionism)
and to the non-subjectivity of the subject (objectivism). We can
conceive such revision in two forms: first of all, to come out in
favor of the for-me which is in every "What is it?", to demand
and to protect the intrusion of value in the discourse of
knowledge. Second, to attack the who, the subject of interpre
tation; here again. NietzSChe: "We have no right to ask who is
interpreting. It is interpretation itself. a form of the will to
power. which exists (not as a 'being; butas a process, a becoming)
as a passion . . ." "No subject but an activity, a creative invention,
neither 'causes' nor 'effects.' ..

VOCdbles ! Vocables

Value appears in certain words, certain terms, certain vocahus

("vocable" is good, for it means both: appellation and patronage
of a Saint: as it happens, we are concerned with numen-words,
 Radjngs

with sign-words. with judgment-words). These vocables erupt

in the discourse of knowledge: the vocable is that mark which
discriminates writing [icritur] from its inauthentic version [icriv
ana] (as in an expression like "the most revolting filth," which
no "scientific" discourse would tolerate). Doubtless, we ought to
have-and someday shall have-a theory of value-words (of
vocables). We may nOte. meanwhile: vocables are sensuous,
subtle, amorous words, denOling seductions or repulsions; an
other morpheme of value is sometimes italics or quotation
marks; quotation marks serve to frame the code (to denaturalize,
to demystify the word); italics. on the contrary, are the trace of
the subjective pressure imposed upon the ....ord, of an insistence
which substitutes for its semantic consistency (italicized words
are very frequent in Nietzsche). Bataille himself seems to have
had a theoretic consciousness of this opposition between knowl
edge words and value words (names and vocables). But in his
discussion there is a terminological crisscross: "word" is the
element of philosophical analysis. of ontological system, "de
noting properties which permit an external action," while "as
peel" (our "vocable") is what "introduces the decisive values of
things," derives "from the decisive moments of nature."'
Thus, there is in the text (by Bataille and according to Bamille)
a whole fabric of value (by vocables, "graphisms"), an entire
"verbal display". Linguistically, these vocables are what? (Of
course, linguistics does not know and does not wish to know;
linguistics is adiaphorous, indifferent.) I merely indicate some
hypotheses:
I. Contrary to a whole modernistic prejudice which attends
only to syntax, as if language could emancipate itself (enter the
avant-garde) only on this level, we must acknowledge a certain
"erratism" of words: some behave, in the sentence, like erratic
stones; the role of the .....ord (in writing) can be to interrupt the
sentence by its brilliance, by its difference, its fissuring, separating
power, by its fetish-situation. "Style" is more palpable than is
supposed.
2. Bataille used to say: "A dictionary begins once it gives, not

Malena :>m d reitl al

Rtading Brillal-Savarin 25 '
submission of the gustative sensation to time actually permits it
to develop somewhat in the manner of a narrative, or of a
language: lemporalized. taste knows surprises and subt1eties;
these are the perfumes and fragrances. constituted in advance,
50 to speak. like memories: nothing would have kept Proust's
madeleine from being analyzed by B.-S.

Need I desire
If B.-S. had written his book today. he would surely have
included among the perversions this taste (specifically. of food)
which he defended and illustrated. Perversion is. one might say.
the exercise of a desire which serves no purpose, like the
exercise of the body which gives itself up to love with no
intention of procreation. Now. in the schema of food, B.-S.
always marked the distinction between need and desire: "The
pleasure of eating requires. if not hunger. at least appetite; the
pleasure of the table is generally independent of both." At a
period when the bourgeoisie knew no social culpability, B.-S.
sets up a cynical opposition: on one side, natural aptitt, which
is of the order of need; and on the other. appetitt for luxury.
i of the order of desire. Everything is here, of course:
which s
the species needs to procreate in order to survive. the individual
nteds to eat in order to subsist; yet the satisfaction of these two
needs does not suffice man: he must bring on stage, so to speak,
the luxury of desire. erotic or gastronomic: an enigmatic, useless
supplement, the desired food-the kind that B.-S. describes
is an unconditional ....aste
. or loss. a kind of ethnographic
ceremony by .....hich man celebrates his po....er.
. his freedom to
consume his energy "for nothing." In this sense. B.-5.'s book is
altogether the book of the strictly human," for it is desire
(insofar as it is spoken) .....hich distinguishes man. This anthro
pological basis gives a paradoxical cachet to The Physiology of
Taste: for ....hat
. is expressed through the turns of style, the
.....orldly tone of the anecdotes, and the graceful futility of the
descriptions is the great adventure of desire. The question,

C JPYnghied mater1
 Readings

however, remains unbroached as to why the social subject (at

least in our societies) must assume sexual perversion in a crude,
fierce, "criminal" style, as the purest of transgressions, while
gastronomic perversion, as described by B.-S. (and on the whole
it could hardly be described better), always implies a kind of
affable and accommodating acknowledgment which never de
parts from the tone of good breeding.

The gastronome's body

Food provokes an internal pleasure, interior to the body, enclosed
within it, not even beneath the skin, but in that deep, central
zone, all the more original for being soft, confused, permeable,
which is called, in the most general sense, the bowels; although
taste is one of man's five acknowledged. classified senses, and
although this sense is localized (on the tongue and, as B.-S.
describes so well, in the whole mouth), gustative delight is
diffuse, extensive to the entire secret lining of the mucous
membranes; it derives from what we should probably consider
our sixth sense-if B.-S., precisely, did not reserve that place
for the genetic sense-and which is ce1uslhesia, the total sensation
of our internal body. B.-S., of course, acknowledges this diffuse
arrangement of the pleasure of food as the sense of well-Imng
which follows good meals; but, oddly enough, he does not
analyze, he does not scrutinize, he does not "poeticize" this
internal sensation; when he wants to grasp the voluptuous effects
of food, he will seek them out on the adverse body; these effects
are in a sense signs, received during an interlocution: the other's
pleasure is deciphered; sometimes, indeed, if a woman is in
volved, it is spied on, it is surprised as if we were dealing with a
minor erotic rape; conviviality, the pleasure of dining together,
is thus a less innocent value than it appears; in the "staging" of
a good meal there is more than the exercise of a worldly code.
even if that code has a venerable historical origin; around the
table prowls a vague scopic pulsion: we observe in the Other
the effects of food, we grasp how the body works on itself from
Reading Brillat-Savarin 25 3

within; like those sadists who delight in the signs of emotion on

their partner's face, we observe the changes in the body which
is dining well. The index of this pleasure is, according to B.-S.,
a very specific thematic quality: shininess, the physiognomy
brightens, coloring is heightened, the eyes gleam, while the
mind is refreshed and a gentle warmth penetrates the entire
body. Shininess is evidently an erotic attribute: it refers to the
state of a substance which is at once ignited and moistened,
desire giving the body its sparkle, ecstasy its radiance (the word
is B.-S:s), and pleasure its lubrification. The gourmand's body
is thus seen as a glowing painting, illuminated from within. This
sublimity nonetheless includes a subtle texture of triviality; we
perceive this unexpected supplement in the scene of the belle
gou7'mande: she has shining eyes, glistening lips, and she bites
her partridge wing; for all the affable hedonism, which is the
usual genre of descriptions of conviviality, we must read another
index in such shininess: that of carnivorous aggression, exem
plified here, paradoxically, by woman; woman does not devour
food, she bites, and this bite radiates; perhaps, in this rather
brutal illumination, we may perceive an anthropological notion: .
in spasms, desire reverts to its origins and turns back into need,
gounnandise into appetite (transferred to the erotic order, this
reversal would lead humanity back to the simple practice of
coupling). The strange thing is that, in the excessively civilized
range of gastronomic customs which B.-S. continually provides,
the strident note of Nature is sounded by woman. We know
that in the vast mythology men have elaborated around the
feminine ideal, food is systematically neglected; we commonly
see woman in a state of love or of innocence; we never see her
eating: hers is a glorious body, purified of any need. Mytho
logically, food is men's business; woman takes part in it only as
a cook or as a servant; she is the one who prepares or serves
but does not eat. With a light touch, B.-S. subverts two taboos:
that of a woman pure of any digestive activity and that of a
gastronomy of pure repletion: he puts food in Woman, and in
Woman appetite (the appetites).
254 Readings

The anti-drug
Baudelaire rebuked B.-S. for not speaking well of wine, For
Baudelaire, wine is memory and forgetling,joy and melancholy;
it is what permits the subject to be transported oUlside himself,
to make his ego's consistency yield LO certain alienated states; it
is a path of deviance; in short, a drug.
Now, for B.-S., wine is not at all a conducLOr of ecstasy. The
reason for which is clear: wine is part of food, and food, for
B.-S., is essentially convivial; wine cannot therefore proceed
from a solitary protocol: one drinks at the same time one eats,
and one always eats with others; a narrow sociality governs the
pleasures of food ; of course, dope smokers can gather in groups,
like the guests at a fine table; but in principle this is so that each
can withdraw into his own singular dream; now, this gap is
forbidden to the gastronome, for in eating he submits LO a
rigorous communal practice: conversation. Conversation (among
several people) is in a sense the law which protects culinary
pleasure from any psychotic risk and maintains the gourmand
in a "healthy" rationality: talking chatting about one thing and
another while he eats, the guest confirms his ego and protects
himself against any subjective leakage, by the image-repertoire
of discourse. Wine, for B.-S., has no special privilege: like food,
and with it, wine slightly amplifies the body (renders it "brilliant")
bUl does not mUle it. It is an anti-drug.

Cosmogonies
Bearing on transformable substances, culinary practice quite
naturally leads the writer who speaks of it to deal with a general
thematics of matter. Just as ancient philosophies attributed great
importance to the fundamental states of matter (water, fire, air,
earth) and from these states derived various generic attributes
(the aerial, the liquid, the ardent, etc.) which could pass into all
forms of discourse, beginning with poetic discourse, in the same
way food, by the treatment of its substances, assumes a cosmo-
 Readings

technological humanity: the excellence of the tool (as opposed

to the machine), the preeminence of the artisanal over the
industrial, in a word the nostalgia for the Natural.

The search for essence

Scientifically, by the end of the eighteenth century, the mecha
nism of digestion is more or less understood : it is known how
the most varied and heteroclite roster of foodstuffs (all those

which humanity, since the origins of life, has been able to

discover and ingest) produces one and the same vital substance,
by which man survives. With a slight historical delay, starting
in 1825, chemistry discovers the elements. All of B.-S.'s culinary
ideology is armed with a notion at once medical, chemical, and
metaphysical: that of a simple essence, the nutritive (or gusta
tive since, for B.-S., there is in fact no food until it is tasted)
ichor. The completed state of nutriment is thus the juice the
liquid and rarified essence of a piece of food. The reduction to
essence, or quintessence, the old alchemist's dream, greatly
impresses B.-S.: he delights in it as in an astonishing spectacle;
the Prince de Soubise's cook, like a magician out of the Arabian
Nights, actually conceived enclosing fifty hams in a crystal flask
no larger than one's thumb! Miraculous equations: the ham's
Being is in its juice, and this juice itself can be reduced to an
essence of which only crystal is worthy. The alimentary essence,
thus projected, assumes a divine aura, as is proved by the fact
that, like the Promethean fire, outside of human laws, it can be
stolen: an Englishman having ordered a leg of lamb at an inn,
B.-S. steals its juice (for an egg dish); he incises the meat turning
on its spit and makes away with its quintessence by theft (adding,
moreover, a touch of Anglophobia).

Ethics
It has been possible to reveal the physical nature of erotic
pleasure (tension / release), but gustative pleasure escapes any

Reading Brillat-Savarin 2 57
such reduction, and consequently any science (as is proved by
the heteroclite nature of tastes and disgusts down through
history and around the world). B.-S. speaks as a scholar, and
his book is a physiology; but his science (does he know this?) is
merely an irony of science. All gustative delight inheres in the
opposition of two values: the agreeable and the disagreea.ble, and
these values are quite simply tautological: the agreeable is what
agrees and the disagreeable what disagrees. B.-S. can go no
further: taste comes from an "appreciative power," just as, in
Moliere, sleep comes from a dormitive virtue. The science of
taste thus reverts to being an ethic (this is the habitual fate of
science). B.-S. immediately associates his physiology (what else
can he do, if he wants to continue his discourse?) with certain
moral qualities. There are two principles here. The first is
statutory, castrating: it is exactitude ("of all the virtues of a cook,
the most indispensable is exactitude"); here we encounter the
classical rule: no art without constraint, no pleasure without
order; the second is well known to the ethics of Transgression :
it is disceiliment, which permits the separation of Good from
Evil; there is a casuistry of taste: taste must always be alert, must
train itself to be subtle, to be scrupulous; B.-S. respectfully cites
the gourmands of Rome who could distinguish the taste of fish
caught between the various bridges of the City from those taken
from the Tiber downstream; or those hunters who manage to
perceive the special flavor of the leg on which the partridge has
rested in its sleep.

Language
Cadmus, who brought writing to Greece, had been the King of
Sidon's cook. Let us take this mythological feature as apologue
to the relation which unites language and gastronomy. Do not
these two powers employ the same organ? And more broadly,
the same apparatus, productive or appreciative: the cheeks, the
palate, and nostrils, whose gustative role B.-S. remarks and
which are responsible for fine singing? To eat, to speak, to sing

 Readings
(need we add: to kiss?) are operations which have the same site
of the body for origin: cut off the tongue, and there will be
neither taste nor speech.
Plato had compared (it is true, invidiously) rhetoric and
cooking. B.-S. does not explicitly invoke this precedent: for him,
there is no philosophy of language. Since the symbolic is not
his strong point, it is in certain empirical remarks that we must
seek this gastronome's interest in language, or, more exactly, in
tongues. This interest is very great. B.-S., as he reminds us,
knows five languages; thus, he possesses an enormous repertoire
of words from every source, which he takes for his own use, in
different compartments of his mind, quite shamelessly. In this,
B.-S. is very modern: he is convinced that the French language
is poor, and that it is therefore licit to borrow or steal words
elsewhere; in the same way, he appreciates the charm of marginal
languages, such as the language of "the people"; he transcribes
and quotes with pleasure the patois of his region, the Bugey.
Finally, each time he has occasion to do so, however remote
from his own gastrosophic discourse, he notes this or that
linguistic curiosity: "to make arms" means: to play the piano
with exaggerated elbow movements, as if one were smothered
by feeling; "to make eyes" means: to look up to heaven as if
one were about to swoon; "to make brioches" (a metaphor which
must have pleased him) means: to miss a note, an intonation.
His attention to language is meticulous, as the cook's art must
be.
Yet w'e must go further than these contingent proofs of
interest. B.-S. is certainly linked to language as he was to
food by an amorous relation: he desires words, in their very
materiality. He comes up with an astonishing classification of
the tongue's movements as it participates in manducation: there
are, among other oddly learned words, spication (when the
tongue takes the shape of a stalk of wheat) and vel7'ition (when
it sweeps). A twofold delight? B.-S. becomes a linguist, he deals
with food the way a phonetician would (and subsequently will)
deal with vocality, and he sustains this learned discourse in a
 Readings
of fried food is, in a sense, the rape to which substance has
been subjected; second, and especially, because B.-S. condemns
boiling (but not bouillon): boiled meat loses (according to the
period's chemistry) a precious substance (precious for its sa
pidity), naturally attached to red meat. This substance is osma
lOme.
Faithful to his philosophy of essences, B.-S. attributes to
osmazome a kind of spiritual power; it is the very absolute of
taste: a sort of alcohol of meat; like a universal (demoniac?)
principle, it assumes various and seductive appearances; it is
osmazome which produces the roux of meats, the "browning"
of roasts, the bouquet of venison; above all, it is osmazome
which makes juice and bouillon, direct forms of quintessence
(the word's etymology refers to the combined notion of odor
and bouillon).
Chemically, osmazome is a meat principle; but the symbolic
realm does not respect chemical identity; by metonymy, osma
zome lends its value to everything that is browned, caramelized,
grilled: to coffee, for instance. B.-S.'s chemistry (however dated)
allows us to understand the present vogue of grilled food: aside
from the functionalist alibi (rapid preparation), there is a
philosophical reason for the popularity of grilled food, which
unites two mythic principles, that of fire and that of rawness,
both transcended in the figure of the grilled, solid form of the
vital juice.

Pleasure
Here is what B.-S. writes about pleasure: "It was only a few
months ago that I experienced, while sleeping, an altogether
extraordinary sensation of pleasure. It consisted in a kind of
delicious thrill of every particle composing my being. It was a
kind of magical tingling which, from the soles of my feet to the
top of my head, racked me to the marrow of my bones. I seemed
to see a violet flame that played around my forehead."
This lyrical description accounts niceI.y for the ambiguity of
Reading Brillat-Savarin
the notion or pleasure. Ordinari
l y. gastronomic pleasure is
described by B.-S. as a refined and rational sense or well-being;
or course, it gives the body a luster (shininess), but it does not
depersonalize this body: neither rood nor wine has a narcotic
power. On the other hand. there is a kind or limit alleged;
pleasure is dose to toppling over into delight, into ecstasy: it
changes the body. which reels itselr in a state or electrical
dispersion. Doubtless. this excess is laid to the account or dreams;
yet it designates something very important: the incommensur
able character or pleasure. Hencerorth, it is enough to sociaize
l
pleasure's unknown quality in order to produce a utopia (here
again we meet Fourier). B.-S. puts it very well: "The limits or
pleasure are not yet known or posited, nor do we know at what
point our body can be beatified." A surprising remark in an old
author. whose style or thought is generally Epicurean: it intro
duces into this philosophy a sentiment or the historical limit
lessness or sensation, or an unsuspected plasticity or the human
body. which we find only in very marginal philosophies: it
postulates a kind or mysticism or pleasure.

Questions
The object alluded to by a sign is called a referent. Each time I
speak or rood, I emit (linguistic) signs which relate to an aliment
or to an alimentary quality. This banal situation has unramiliar
implications when the object alluded to by my speech-act is a
desirable one. This is obviously the case with Tiu Physiolhgy of
Task. B.-S. speaks and I desire what he speaks about (especially
ir I am in a state or appetite). The gastronomic statement,
because the desire it mobilizes is apparently simple, presents
the power or language in all its ambiguity: the sign calls up the
pleasures or its rererent just when it traces its absence (which
we know each word does, ever since Mallarme said so of the
Hower, "missing from every bouquet"). Language provokes and
excludes. Whereupon the gastronomic style presents us with a
whole series of questions: What does it mean to represent? to

C JPYnghied mater1
264 Readings
figure? to project? to say? What IS It to desire? What is it to
desire and to speak at the same time?

The first hour

Like every hedonist subject, B.-S. seems to have an intense
experience of boredom. And as always, boredom, linked to what
philosophy and psychoanalysis have denoted as repetition, im
plies, by a contrary means (which is that of the opposition of
meaning), the excellence of novelty. Everything deriving from
a primary temporality is endowed with a kind of enchantment;
the first moment, the first time, the freshness of a dish, of a
rite, in shon the beginning, refers to a kind of pure state of
pleasure: where all the determinations of a felicity combine.
Thus with the pleasure of the table: "The table," B.-S. says. "is
the only place where one is not bored during the first hour."
This first hour is marked here by the appearance of new dishes,
the discovery of their originality, the elan of conversations, in
shon, by a word which B.-S. applies to the excellence of the
best fried food: surprise.

Dreams
Appetite relates to dreaming, for it is both memory and hallu
cination, which is why, moreover, it might be better to say that
it relates to hallucinations. When I have an appetite for food,
do I not imagine myself eating it? And, in this predictive
imagination, is there not the entire memory of previous plea
sures? I am the constituted subject of a scene to come, in which
I am the only actor.
B.-S. has reRected on dreams, then, "a life apan, a kind of
extended fiction. " He has grasped the paradox of dreams, which
can be intense pleasure yet exempt from real sensuality: in
dreams. neither odor nor taste. Dreams are memories or com
binations of memories: "Dreams are only the memory of the
senses." Like a language which is elaborated only on the basis

GJPYnghtcd maknal
Rtading Briliat-Savarin
of certain signs, solated
i vestiges of another language, dreams
are a decrepit narrative, consisting of the ruins of memory.
B.-S. compares them to a reminiscence of melody, of which one
might play only a few notes, without adding harmony to it. The
discontinuity of dreams s i opposed to sleep's consecution, and
this opposition is reAected in the very organization of foodstuff's;
some are somniferous: milk, fowl, lettuce, orange blossoms,
pippin apples (eaten before going to bed); others produce
dreams: dark meats, hare, asparagus, celery, truffles, vanilla;
these are strong foods, perfumed or aphrodisiac. B.-S. discerns
the dream as a marked, one might almost say a virile state.

Science
"Thirst," B.S. says, "is the internal sentiment of the need to
drink." Certainly, the interest of such sentences is not in the
formation they provide (here, none at all). By such tautologies,
evidently, B.-S. tries his hand at science, or at least at scientific
discourse; he produces statements without surprise, which have
no other value than to present a pure image of the scientific
proposition (definition, postulate. axiom, equation): and is there
a more rigorous science than the kind which defines the same
by the same? Here, no risk of error; B.-S. is protected from
that malign power which wrecks science: paradox. His audacity
is one of style: to use a learned tone in order to speak of a sense
reputed to be trivial (because banally sensual): taste.
Science is the great Superego of Tiu Physiology of Task. The
book, it is said, was written in consultation with an official
biologist, and B.-S. strews his discourse with scientific solemni
ties. Thus, he imagines he is submitting the desire for food to
experimental measures: "Each time a dish of distinguished and
well-known Aavor is served, observe the guests closely, and you
will note as unworthy all those whose physiognomy does not
betray pleasure." By his "gastronomic assays," 8.-S., however
preposterous the notion, takes into account two very serious
and very modern factors: sociality and language; the dishes he

C :lpynghled malenal
.66 Readings
presents to his subjects fOT experiment vary according to the
social class (income) of these subjects: a fillet of veal or eggs Ii
Ia mige if one is poor; a fillet of beef or a turbot au naturtl if
one is well-off; tTuffie, quails, vanilla meringues if one is rich,
etc.-which is to imply that taste is modeled by culture, i.e., by
social class; and then, a surprising method, in order to rtad
gustatory pleasure (since this is the goal of the experiment),
B.-S. suggests interrogating. not the (probably universal) ges
tures and facial expressions of the diners, but their language, a
socialized object if ever there was one: the expression of assent
changes according to the speaker's social class: in front of his
eggs Ii La migt the poor man will exclaim "Damnation!", while
ortolans provenrok will wring from the rich man: "My lord, your
chef is an admirable man!"
These witticisms, which include several true intuitions, nicely
express how B.-S. took science: in a fashion at once serious and
ironical; his project of establishing a science of taste, of stripping
culinary pleasure of its habitual signs of triviality, was certainly
close to his heart; but he performs it rhetorically, i.e., with
irony; he is like a writer who puts quotation marks around the
truths he utters, not out of scientific prudence, but for fear of
appearing naive (whereby we can see that irony is always timid).

Sex

There are, it is said, five senses. From the first page of his book,
B.-S. postulates a sixth: the generic, or physical love. This sense
cannot be reduced to louch; it implies a complete apparatus of
sensations. "Let us give to the genesic," B.-S. says, "the seruual
place we cannot deny it, and let us bequeath to our nephews
the task of assigning its rank" (we the nephews have evidently
not shrunk from the task). It is obviously B.-S.'s intention to
suggest a kind of metonymic exchange between the first of
delights (even if it is censored) and the sense whose defense
and illustration he has undenaken, i.e., taste; from the point of
view of sensuality, it signifies taste to put it in the same roster

C ;.pvnghted malaria
Rtading Brillat-Savarin
with erotic pleasure. B.-S. insists then, when he can, on the
aphrodisiac power of certain foods: tTuffles, fOT example, or
fish. which provokes his astonishment (a slight anti-clerical irony
here) that this should be the food eaten during Lent by monks,
dedicated to chastity. Yet try as he will, he can find little analogy
between lust and gastronomy; between the two pleasures, a
crucial difference: orgasm, i.e., the very rhythm of excitation
and its release. Pleasures of the table include neither ravishments
nor transports nor ecstasies-nor aggressions; bliss, if there is
such a thing here, is not paroxystic: no mounting of pleasure,
no culmination, no crisis; nothing but a duration; as if the only
critical element of gastronomic joy were its expectation; once
satisfaction begins, the body enters into the insignificance of
repletion (even if this assumes the demeanor of a gluttonous
compunction).

Sociality
Doubtless, a general ethnology could easily show that eating is
in all places and at all times a social act. We eat together, that
s
i the universal law. This alimentary sociality can assume many
forms, many alibis, many nuances. according to societies, ac
cording to periods. For B.-S., the gastronomic collectivity is
essentially worldly, and its ritual figure is conversation. The
table is in a sense the geometric locus of all the subjects discussed;
it is as if alimentary pleasure vivified them, brought them to a
kind of rebirth; the celebration of food is laicized in the form
of a new kind of gathering (and panicipation): the convivium.
Added to the good food, the convivium produces what Fourier
(whom we always find close to B.-S.) called a comprutt
i pleasure.
The vigilant hedonism of the two brothers-in-law n
i spired them
with this thought, that pleasure must be ovtrdtttrmind, that it
must have several simultaneous causes, among which there is
no way of distinguishing which one causes delight; for the
composite pleasure does not derive from a simple bookkeeping
of excitations: it figures a complex space in which the subject

C :lpynghled malenal
268 Readings
no longer knows where he comes from and what he wants- s-
except to have his voluptuous pleasure -jouir. The convivium
so important in B.-S.'s ethic is therefore not only a sociological
fact; it prompts us to consider (as the human sciences have so
rarely done hitherto) communication as a delight, ajouissance
and no longer as a function.

Social classes
We have seen that, in the game of gastronomic experiments,
B.-S. linked difference in tastes to difference in incomes. The
originality here is not to recognize classes of money (want,
comfort, wealth) but to conceive that taste itself (i.e., culture) is
socialized: if there is an affinity between eggs a la neige and a
modest income, it is not only because this dish costs relatively
little to make; it is also, it seems, by reason of a social formation
of taste, whose values are established not in the absolute but in
a determined field. Hence, it is always by the relay of cultur'e-e
and not by that of needs that B.-S. socializes food. So when
he turns from incomes to professional classes (to what was called
the "states" or "conditions"), establishing that society's great
gourmands are chiefly financiers, physicians, men of letters and
the Church, what he considers is a certain profile of customs
and habits, in short a social psychology: gastronomic taste seems
to him linked by privilege either to a positivism of profession
(financiers, physicians) or to a special aptitude to displace, to
sublimate, or to intimize pleasure (men of letters, of the Church).
In this culinary sociology, modest as it is, the purely social is
nonetheless present: precisely where it is missing from discourse.
It is in what he does not say (in what he occults) that B.-S. most
clearly registers the social condition in all its nakedness: and
what is repressed, quite pitilessly, is the food of the people.
What did such food chiefly consist of? Bread and, in the country,
gruels, the cook using for these a grain she pounded herself in
a mortar and pestle, which spared her having to submit to the
monopoly of mills and communal ovens; no sugar, but honey.
Reading Brillat-Savarin
The essential food of the pOOT was potatoes; these were sold,
boiled or roasted, in the street (as they are still in Morocco), as
chestnuts are today; long despised by people "of a certain rank,"
who relegated them to "animals and the very poor," potatoes
owe nothing of their social elevation to Parmentier, an army
physician whose main interest was in substituting starch for
flour in making bread. In B.-S.'s own time, potatoes, while
beginning their redemption, remain marked by the discredit
attached, socially, to anything "boiled." Consider the menus of
the period: nothing but discreet, separated dishes: the combined
and the thickened are found only in sauces.

Topic
B.-S. has understood that, as a subject of discourse, food was a
sort of grid (a topic, the old rhetoric would have said), through
which he could successfully introduce all the sciences we now
adays call social and human. His book tends toward the ency
clopedic, however summarily. In other words, its discourse is
likely to treat food under several pertinences; in short, it is a
total social phenomenon, around which can be convoked various
meta-languages: those of physiology, of chemistry, of geography,
of history, of economy, of sociology, and of politics. It is this
encyclopedism this "humanism" which, for B.-S., is suggested
by the name gastronomy: "Gastronomy is the knowledge of all
that relates to man, insofar as he subsists on food." This
"scientific" opening nicely corresponds to what B.-S. was in his
own life; essentially, a polymorphous subject: jurist, diplomat,
musician, man of fashion, he was quite familiar with both foreign
parts and the provinces, so food was not an obsession for him
but rather a kind of universal operator of discourse.
Perhaps, to conclude, we should glance at some dates. B.-S.
lived from 1 755 to 1826. He was (for instance) a contemporary
of Goethe ( 1 742-18 3 2). Goethe and Brillat-Savarin: these two
names, set side by side, constitute a riddle. Of course, Werther
was not above ordering peas cooked in butter in his Wahlheim
 An Idea of Research

In the little Balbec train, a solitary lady is reading the Revue des
deux mondes; she is ugly and vulgar; the Narrator takes her for
the madam of a brothel; but on his next journey the little clan,
having invaded the train, informs the Narrator that his lady is
Princess Sherbatoff, a woman of high birth, the pearl of the
Verdurin salon.
This pattern, which conjoins two absolutely antipathetic states
in one and the same object and radically reverses an appearance
into its contrary, is frequent in Proust's novel. Here are a few
examples, noted while reading the first volumes: 1 . Of the two
Guermantes cousins, the more affable is in reality the more
disdainful (the duke); the colder, the more sincere (the prince).
2. Odette Swann, a superior woman in the judgment of her
circle, is regarded as stupid by the Verdurin clan. 3 . Norpois,
pontificating to the point of intimidating the Narrator's parents
and of persuading them that their son has no talent, is utterly
destroyed by Bergotte's single phrase ("But he's an old fool").
4. The same Norpois, a monarchist aristocrat, is entrusted with
extraordinary diplomatic missions by radical cabinets "which
even a reactionary bourgeois would have refused to serve and
whose suspicions should have been aroused by Monsieur de
Norpois's past, his connections, and his opinions." 5. Swann and
Odette pamper the Narrator; yet there was a time when Swann
did not even deign to answer Marcel's "persuasive and detailed"
letter; the concierge in the Swanns' apartment building is
transformed into a benevolent Eumenid. 6. Monsieur Verdurin
speaks of Cottard in two ways: if he believes the professor is
little known to his interlocutor, he glorifes him, but follows the
converse procedure and speaks of Cottard's medical genius

271

 Readings
quite simply if Cotlard is recognized. 7. Having just read in a
great scholar's book that perspiration is harmful to the kidneys,
the Narrator encounters Dr. E., who informs him: "The advan
tage of these hot days we're having, when one perspires so
abundantly, is that the kidney is greatly relieved." And so forth.
These notations are so frequent, they are applied so consist
ently to such different objects, situations, and languages, that
we may identify in them a form of discourse whose very
obsessiveness is enigmatic. Let us call this form, at least provi
sionally, inversion, and let us anticipate (without presently being
able to do so) inventorying its occurrences, analyzing its modes
of expression, the devices which constitute it, and situating the
considerable extensions it seems capable of at very different
levels in Proust's work. This would propose "an idea of re
search"-though without allowing us to entertain any positivist
ambition: Proust's novel is one of those great cosmogonies
endemic to the nineteenth century (Balzac, Wagner, Dickens,
Zola), whose character, at once statutory and historical, is that
they are infinitely explorable spaces (galaxies); thereby, our
critical work is shifted (from any illusion of "result") toward the
simple production of a supplemental writing whose tutelary text
(Proust's novel), if we write up our "research," would be only a
pre-text.

Here, then, are two identities of one and the same body: on
one side, the madam of a brothel; and, on the other, Princess
Sherbatoff, lady-in-waiting to the Grand Duchess Eudoxia. We
may be tempted to see this figure as the banal interplay of
appearance and reality: the Russian princess, ornament of the
Verdurin salon, is only a woman of the coarsest vulgarity. Such
an interpretation is strictly moralistu: (the "is only" syntactic form
is constantly used by La Rochefoucauld, for instance); we would
thereby recognize (as has occasionally been the case) the Prous
tian oeuvre as an alethic project, an energy of decipherment. a
search for essences, whose first effort is to rid human truth of
the contrary appearances which superimpose upon it vanity,

B
An Idea of Restarch

worldliness, snobbery. Yet, by reading the ProuSlian inversion

as a simple reduction, we sacrifice the efflorescences of form
and risk distorting the text. There are three such efflorescences
(truth of the discourse and not truth of the project): 1. Ttmpor
aiity, or more exactly, an effect of time; the two terms of the
contradiction are separated by a period of time, an adventure:
it is not, literally, the saint Narrator who reads the madam of
the brothel and the great Russian lady: two trains separate
them. 2 . Climax: the inversion is effected according to an exact
figure, as if a god-a fatum-were maliciously presiding over
the trajectory which leads the princess to coincide with her
geometrically determined absolute contrary; like one of those
riddles Proust was so fond of: What is a madam's apotheosis?
To be a lady-in-waiting to the Grand Duchess Eudoxia-or vice
versa. 3. Surprise: the reversal of appearances-let us no longer
say of appearance into reality-always affords the Narrator a
delicious astonishment: essence of surprise-to which we shall
return-and not essence of truth, veritable jubilation, so com
plete, so pure. so triumphant (as is proved by the success of its
expression), that this mode of inversion can only derive from
an erotics (of discourse), as if its occasion was the very moment
in which Proust took his delight in writing: studded all through
the great continuum of the search, inversion is the suppkmrnt
bliss of narrative. of language.

Pleasure once found, the subject knows no rest until he can

repeat it. Inversion-as fonn-invades the entire structure of La
ReCMrCM. It inaugurates the narrative itself: the first scene
from which will emerge, through Swann, the entire novel-is
articulated around the reversal of a despair (that of having to
go to sleep without the mother's kiss) into a delight (that of
spending the night in the mother's company); here. in fact, are
inscribed the characteristics of Proustian inversion: nO[ only will
the mother, finally (temporality), come to embrace her son
against all expectation (surprise) but, moreover (clima:c). out of
the darkest despair. the most overwhelming joy will appear. the

B
'74 Readings
stern Father unexpectedly LUrning into the kindly Father (" . . .
and tell Franoise to make the big bed for you, and sleep with
him tonight"). Reversal does not remain limited to the thousand
notations of detail of which we have given a few examples; it
structures the very development of the main characters, subject
to "exact" elevations and falls: from the height of aristocratic
grandeur, Charlus falls, in the Verdurin salon, to the rank of
petit-bourgeois; Swann. habitual companion of princes, is for
the Narrator's great-aunts a colorless figure of no particular
status; the cocotte Odette becomes Mme Swann; Mme Verdunn
ends as -the Princess de Guermantes, etc. An incessant permu
tation animates, overturns the social interplay (Proust's work is
much more sociological than is acknowledged: it describes with
great exactitude the grammar of promotion, of class mobility),
to the point where worldliness can be defined by a form: reversal
(of situations, opinions, values, feelings, languages).
In this regard, sexual inversion is exemplary (but not neces
sarily primary), since it enables us to read one and the same
body as the super-impression of two absolute contraries, Man
and Woman (contraries which Proust defined biologically and
not symbolically: a period feature. no doubt; in order to
rehabilitate homosexuality, Gide proposes examples of pigeons
and dogs); the scene of the hornet, during which the Narrator
discovers the Woman in the Baron de Charius, is theoretically
valid for any reading of the interplay of contraries; whence, in
the whole work, homosexuality develops what we might caU its
enantiology (or discourse of reversal); on the one hand, it gives
rise in the world to a thousand paradoxical situations. misun
derstandings, surprises, climaxes, and tricks, which the novel
scrupulously collects; and on the other, as exemplary reversal,
it is animated by an irresistible movement of expansion; by a
broad sweep which takes up the entire work, a patient but
infallible curve, the novel's population, heterosexual at the
outset, is ultimately discovered in exactly the converse position
i.e., homosexual (like Saint-Loup, the Prince de Guermantes.
etc.): there is a pandemia of inversion, of reversal.

B
An Idta of Rtstarch 275
Reversal is a law. Every feature is required to reverse itself,
by an implacable movement of rotation: endowed with an
aristocratic language, Swann can only, at a certain moment,
invert it into bourgeois language. This constraint is so statutory
that it renders futile, Proust says, any observation of manners:
one can readily dtduct them from the law of inversion. A reading
of reversal is therefore equivalent to knowledge. But we must
be careful: such knowledge does not reveal content, or at least
does not stop there: what is notable (statutory) is not that the
great Russian lady is vulgar or that M. Verdurin adapts his
description of Cottard to his interlocutor; it is the form of this
reading, the logic of inversion which structures the world. Le.,
worldliness; this inversion itself has no meaning, we cannot
retain it. one of the permuted terms is not "truer" than the
other: Cottard is neither "great" nor "small"; his truth, if he
has one, is a truth of discourse. extensive with the entire
oscillation to which the Other's speech (in this case. M. Verdur
in's) subjects him. For classical syntax, which would tell us that
the Princess Sherbatoff is only a madam. Proust substitutes a
concomitant syntax: the princess is also a madam; a new syntax
we should call metaphorical because metaphor, contrary to what
rhetoric has long supposed. is a labor of language deprived of
any vectorization: it moves from one term to another only in a
circular and infinite fashion. Thus, we understand why the ethos
of Proustian inversion s
i Surprise; it is the astonishment of a
Ttturn. of ajunction, of a recognition (and of a reduction): to utter
the contraries is finally to unite them in the very unity of the
text, of writing'S journey. It follows. then, that the great oppo
sition which seems at the outset to animate both the Combray
excursions and the divisions of the novel (Swann's Way I Tiu
Guermantes' Way) is. if not fallacious (we are not within the order
of truth). at least revocable: the Narrator discovers with stupe
faction (the same stupefaction he experiences when he realizes
that the Baron de Charlus is a Woman, the Princess Sherbatoff
a madam, etc.) that the two paths which diverge from the family
house will converge. and that Swann's world and that of Guer-

B
 Reading$

mantes, through a thousand anaSlOmoses, ultimately coincide

in the person of Gilne, Swann's daughter and the wife of
Rort de Saint-Loup.
Yet there is a moment in the novel when the great inverting
form no longer functions. What keeps it from doing so? Nothing
less than Death. We know that all of Proust's characters come
together in the last volume of the work (Time Regained); in what
condition? Not in the least inverted (as would have been
warranted by the great lapse of time at whose end they are
collected at the Princess de Guermantes's luncheon), but on the
contrary prolonged. paralyzed (even more than aged), preserved,
and-the neologism is warranted-"perstvtTed." In reprieved
life, inversion no longer "takes": the narrative has nothing left
to do but come to an end-the book has nothing left to do but
begin.

Paragone, 1971

,
 Longtemps, je me suis couche
de bonne heure . . .

Some of you will have recognized my title, and the passage it

initiates: "Time was, I went to bed early. Some evenings, my
candle was no sooner out than my eyes would dose so quickly
I had no time to think: 'I'm falling asleep.' And half an hour
later, the thought that it was time to go to sleep would waken
me . . ,": it is the opening of In Search of Lrut TifM. Does this
mean I am offering you a lecture "on" Proust? Yes and no. My
subject will be. if you like, Prowt and I. How pretentious!
Nietzsche spared no irony about the Germans' use of that
conjunction: "Schopenhauer and Hartmann," hejeered. "Proust
and I" is worse still. Let me suggest that, paradoxically. the
pretentiousness subsides once I myself take the stand, and not
some witness: by setting Proust and myself on one and the same
line, I am not in the least comparing myself to this great writer
but, quite differently, identifying myself with him: an association of
practice, not of value. Let me explain: in figurative language,
in the novel, for instance, it seems to me that one more or less
identifies oneself with one of the characters represented; this
projection, I believe, is the very wellspring of literature; but in
certain marginal cases, once the reader himself wants to write
a work, he no longer identifies himself merely with this or that
fictive character but also and especially with the actual author
of the book he .has read, insofar as that author wanted to write
this book and succeeded in doing so; now, Proust is the privileged
site of this special identification, insofar as his Search . . . is the
narrative of a desire to write: I am not identifying myself with
the prestigious author of a monumental work but with the

'77

B
 Readings

worker-now tormented, now exalted, in any case modest

who wanted to undertake a task upon which, from the very
stan of his project, he conferred an absolute character. .

Therefore, first of all, Proust . . .

In Search of LOlj Time was preceded by many writings: a book,
translations, articles. The great work was not really launched, it
seems, until the summer of 1909; and from that point on, as
we know, a stubborn race was run against death and incomple
tion. Apparently there was, in that year (even if it is futile to
try dating the inception of a work with any specificity), a crucial
period of hesitation. Certainly, Proust seems to be at the
intersection of two paths, two genres, torn between two "ways"
he does not yet know could converge, any more than the
Narrator knows, for a very long time-until Gilberte's marriage
to Saint-Loup-that Swann's Way meets the Guermantes' Way:
the way of the Essay (of Criticism) and the way of the Novel.
At the time of his mother's death, in 1905, Proust passes through
a period of despondency, but also of sterile agitation; he wants
to write, to create a work, but which? Or rather, which form?
Proust writes to Mme de Noailles, in December 1908: "I'd like,
sick as I am, to write on Sainte-Beuve [incarnation of the
aesthetic values he abhors]. The thing has taken shape in my
mind in two different ways, between which I must choose. Now,
I am totally without will, and without any power to see my way."
I should point out that Proust's hesitation-to which, quite
naturally, he gives a psychological form---corresponds to a
structural alternation : the two "ways" he hesitates between are
the two terms of an opposition articulated by Jakobson: that of
Metaphor and Metonymy. Metaphor sustains any discourse
which asks: "What is it? What does it mean?"-the real question
of any Essay. Metonymy, on the contrary, asks another question:
"What can follow what I say? What can be engendered by the
episode I am telling?"; this is the Novel's question. Jakobson

B
 Rtadings
form of his Starch . . . : novel? essay? Neither one, or both at
once: what I should call a third form. Let us question this third
genre a moment.
If I began these refle<:tions with the first sentence of Proust's
Starch . . . , it is because it opens an episode of some fifty pages
which, like a Tibetan mandala, collects together within its view
the entire Proustian lX:uvre. What does this episode discuss?
Sleep. Proustian sleep has an inceptive value: it organizes what
is original (and "typical") in the novel (hut this organization, as
we shall see, is in fact a disorganization).
Of course, there is a good sleep and a bad. The good kind is
the one begun, inaugurated, permitted. consecrated by the
mother's evening kiss; it is the right sleep, in accord with Nature
(to sleep by night, to act by day), The bad kind is the sleep far
from the mother: the son sleeps by day, while the mother is up;
they see each other only at the brief intersection of the right
time and the inverted time: awakening for one, bedtime for the
other; this bad sleep (under Veronal) can only be justified,
redeemed by the entire novel, since it is at the painful price of
this inversion that Proust's StaTch . . . , night after night, will be
written,
And what is this good sleep (of childhood)? It is a "half
waking" ("I have tried to wrap my first chapler in the impressions
of half waking,") Although Proust speaks on one occasion of
the "depths of our unconscious," this sleep has nothing Freudian
about it; it is not oneiric (there are few real dreams in Proust's
work); rather, it is constituted by the depths of consciousness as
disordtT. A paradox defines it nicely: it is a sleep which can be
written, because it is a consciousness of sleep; the whole episode
(and, consequently, I believe, the whole work which emerges
from it) is thus held suspended in a sort of grammatical scandal:
to say "I'm asleep" is in effect, literally, as impossible as to say
''I'm dead"; writing is precisely that activity which tampers with
language-the impossibilities of language-to the advantage of
discourse,
What does this sleep (or this half waking) do? It leads to a

B
Longtemps, je me suis couche de bonne heure . . . 281

"false consciousness": a consciousness out of order, vacillating,

intermittent; the logical carapace of Time is attacked; there is
no longer a chrono-logy (if we may separate the two parts of
the word): "A man who is asleep [read : that Proustian sleep
which is a half waking] holds in a circle around him the course
of the hours, the order of years and worlds . . . but their ranks
can mingle, can break" [italics mine]. Sleep establishes another
logic, a logic of Vacillation, of Decompartmentalization, and it
is this new logic which Proust discovers in the episode of the
madeleine, or rather of the biscuit, as it is recounted in Contre
Sainte-Beuve (i.e., before his Search . . . ): "I remained motionless
. . . when suddenly the shaken partitions of my memory gave
way." Naturally, such a logical revolution can only provoke a
reaction of stupidity: Humblot, the reader for Editions Ollen
dorf, receiving the manuscript of Swann's Way, declared : " I
don't know if I've gone completely blind and deaf, but I can't
see any interest in reading thirty pages [precisely our mandala] on
how a Gentleman tosses and turns in bed before falling asleep."
The interest, however, is crucial: it will open the floodgates of
Time: once chrono-logy is shaken, intellectual or narrative frag
ments will form a series shielded from the ancestral law of
. Narrative or of Rationality, and this series will spontaneously
produce the third fOI'm, neither Essay nor Novel. The structure
of this work will be, strictly speaking, rhapsodic, i.e. (etymologi
cally), sewn; moreover, this is a Proustian metaphor: the work
is produced like a gown; the rhapsodic text implies an original
art, like that of the couturiere: pieces, fragments are subject to
certain correspondences, arrangements, reappearances: a dress
is not a patchwork, any more than is A la Recherde du temps perdu.
Emerging from sleep, the work (the third fOIm) rests on a
provocative principle: the disorganization of Time (of chrono
logy). Now, this is a very modern principle. Bachelard calls
rhythm that force which aims to "rid the soul of the false
permanence of ill-made durations," and this definition applies
very nicely to Proust's novel, whose every sumptuous effort is
to subtract Time Remembered from the false permanence of
 Readings
biography. Nietzsche, more lapidarily, says that "we must reduce
the universe to crumbs, lose respect for the whole," and John
Cage, prophesying the new musical work, announces: "In any
case, the whole will constitute a disorganization." This vacillation
is not an aleatory anarchy of associations of ideas: "I see," Proust
says with a certain bitterness, "readers supposing that I am
writing, by trusting to arbitrary and fortuitous associations of
ideas, the story of my life." In fact, if we adopt Bachelard's
word, what we are dealing with is a rhythm, and a highly complex
one: "systems of moments" (Bachelard again) succeed each
other, but also correspond to each other. For what the principle
of vacillation disorganizes is not Time's intelligibility but biog
raphy's illusory logic, insofar as it traditionally follows the purely
mathematical order of the years.
This disorganization of biography is not its destruction. In
the work, many elements of personal life are retained, in an
identifiable fashion, but these elements are in a sense shifted. I
shall indicate two of these shifts, insofar as they turn not on
details (of which biographies of Proust are full) but on major
,creative options.

The first shift is that of the discoursing person (in the

grammatical sense of the word person). The Proustian oeuvre
brings on stage (or into writing) an "I" (the Narrator); but this
"I," one may say, is not quite a self (subject and object of
traditional autobiography): "I" is not the one who remembers,
confides, confesses, he is the one who d iscourses; the person
this "I" brings on stage is a writing self whose links with the self
of civil life are uncertain, displaced. Proust himself has explained
this well: Sainte-Beuve's method fails to realize that "a book is
the product of a different 'self' from the one we manifest in
our habits, in society, in our vices." The result of this' dialectic
is that it is vain to wonder if the book's Narrator is Proust (in
the civil meaning of the patronymic) : it is simply another Proust,
often unknown to himself.
The second shift is more flagrant (easier to define): in Proust's
novel, there is certainly "narrative" (it is not an essay), but this
Longtemps, je me suis couche de bonne heure . . . 283

narrative is not that of a life which the Narrator apprehends at

birth and follows from year to year until the moment he takes
up his pen to tell the story. What Proust recounts, what he puts
into narration, is not his life but his desire to write: Time weighs
heavily on his desire, maintains it in a chronology; it (the steeples
of Martinville, Bergotte's phrase) encounters trials, discourage
ments (the verdict of Monsieur de Norpois, the incomparable
prestige of the Goncourts' journal), and ultimately triumphs
when the Narrator, arriving at the Guermantes' party, discovers
what he must write: Time regained, and thereby assures himself
that he is going to be able to write: In Search of Lost Time (though
it is already written).
As we see, what passes into the work is certainly the author's
life, but a life disoriented. George Painter, Proust's biographer,
has accurately seen that the novel is constituted by what he has
called a "symbolic biography," or again, "a symbolic story of
Proust's life": Proust understood (and this is genius) that he did
not have to "recount" his life, but that his life nonetheless had
the signification of a work of art. "A man's life of any worth is
a continual Allegory," says Keats, quoted by Painter. Posterity
proves Proust increasingly right: his work is not only read as a
monument of universal literature but as the impassioned expres
sion of an absolutely personal subject who ceaselessly returns to
his own life, not as to a curriculum vitae, but as to a constellation
of circumstances and figures. More and more, we find ourselves
loving not "Proust" (civil name of an author filed away in the
histories of literature) but "Marcel," a singular being, at once
child and adult, puer senilis, impassioned yet wise, victim of
eccentric manias and the site of a sovereign reflection on the
world, love, art, time, death. I have proposed calling this very
special interest readers take in the life of Marcel Proust "Mar
celism" in order to distinguish it from "Proustism," which would
be merely a preference for a certain work or a certain literary
manner.
If I have emphasized in Proust's work-as-life the theme of a
new logic which permits one in any case, permitted Proust.-
 Readings
to abolish the contradiction between Novel and Essay, it is
because this theme concerns me personally. Why? That is what
I want to explain now. Hence I shall be speaking of "myself."
"Myself " is to be understood here in the full sense: not the
asepticized substitute of a general reader (any substitution is an
asepsis); I shall be speaking of the one for whom no one else
can be substituted, for better and for worse. It is the intimate

which seeks utterance in me, seeks to make its cry heard,

confronting generality, confronting science.

Dante (another famous opening, another overwhelming allu

sion) begins his poem "Nel mezzo del camin di nostra vita . . . " In
1 300, Dante was thirty-five (he was to die twenty-one years

later). I am much older than that, and the time I have left to
live will never be half the length of my life so far. For the
"middle of our life" is obviously not an arithmetical point: how,
at the moment of writing, could I know my life's total duration
so precisely that I could divide it into two equal parts? It is a
semantic point, the perhaps belated moment when there occurs
in my life the summons of a new meaning, the desire for a
mutation: to change lives, to break off and to begin, to submit
myself to an initiation, as Dante made his way into the selva
oscura, led by a great initiator, Virgil (and for me, at least during
this text, the initiator is Proust). Age, need we be reminded?
but yes, we do, so indifferently do we experience each other's
age age is only very partially a chronological datum, a garland
of years; there are classes, compartments of age: we pass through
life from lock to lock; at certain points there are thresholds,
gradients, shocks; age is not gradual and progressive, it is
mutative: to consider one's age, if that age is what we French
call un certain age, is not a coquetry intended to bring forth
kindly protestations; rather, it is an active task: what are the
real forces which my age implies and seeks to mobilize? That is
the question, appearing quite lately, which it seems to me has


Longtemps, je me suis couche de bonne heure . . . 285

made the present moment the "middle of the journey of my

life."
Why today?
There comes a time (and this is a problem of consciousness)
when "our days are numbered": there begins a backwards count,
vague yet irreversible. You knew you were mortal (everyone has
told you so, ever since you had ears to hear); suddenly you feel
mortal (this is not a natural feeling; the natural one is to believe
yourself immortal; whence so many accidents due to careless
ness). This evidence, once it is experienced, transforms the
landscape: I must, imperatively, lodge my work in a compart
ment which has uncertain contours but which I know (new
consciousness) are finite: the last compartment. Or rather, be
cause the compartment is designated, because there are no
longer any "outside-instances," the work I am going to lodge
there assumes a kind of formality, a solemn instance. Like Proust

ill, threatened by death (or believing himself so), we come back

to the phrase of St. John quoted, approximately, in Contre Sainte
Beuve: "Work, while you still have the light."
And then a time also comes (the same time) when what you
have done, worked, written, appears doomed to repetition:
What! Until my death, to be writing articles, giving courses,
lectures, on "subjects" which alone will vary, and so little! (It's
that "on" which bothers me.) This feeling is a cruel one; for it
confronts me with the foreclosure of anything New or even of
any Adventure (that which "advenes" which befalls me); I see
my future, until death, as a series: when I've finished this text,
this lecture, I'll have nothing else to do but start again with
another . . . Can this be all? No, Sisyphus is not happy: he is
alienated, not by the effort of his labor, or even by its vanity,
but by its repetition.
Finally, an event (and no longer only a consciousness) can
supervene which will mark, incise, articulate this gradual silting
up of work, and determine that mutation, that transformation
of the landscape which I have called the "middle of life." Rance,
hero of the Fronde, a worldly dandy, comes from his travels

Preface to Rnulud Camus's Tricks ' 93

lO do is to figure or, better still. to cipher, as economically as

possible, a series of actions which, in any case, elude it. Erotic
scenes must be described sparingly. The economy here is that
of the sentence. The good writer is the one who utilizes syntax
sa as to link several actions within the briefest linguistic space
(we find, in Sade, a whole art of subordinate clauses); the
sentence's function is somehow to scour the carnal operation of
its tediums and its efforts, of its noises and its adventitious
thoughts. In this regard, the final scenes of the various Triclts
remain entirely within the domain of writing.
But what I like best of all in Triclu are the preparations: the
cruising, the alert, the signals, the approach, the conversation,
the departure for the bedroom, the household order (or dis
order) of the place. Realism finds a new site; it is not the love
scem which is realistic (or at least its realism is not pertinent), it
is the social scene. Two young men who do not know each other
but who know that they are about to become partners in a
specific act risk between them that fragment of language to
which they are compelled by the trajectory they must cover
together in order to reach their goal. The Trick then abandons
pornography (before having really approached it) and joins the
novel. The suspense (for these Triclts, I believe. will be read
eagerly) affects not behavior (which is anticipated, to say the
least) but the characters: Who are they? How do they differ
from each other? What delights me, in Triclu, is this juxtaposi
tion: the scenes, certainly, are anything but chaste, yet the
remarks are just that: they say satto voce that the real object of
such modesty is not the Thing ("La Chose, toujours La Chose,"
Charcot used to say. as quoted by Freud) but the person. It is
this passage from sex to discourse that I find so successfully
achieved in Triclts.
This is a form of subtlety quite unknown to the pornographic
product, which plays on desires. not on fantasies. For what
excites fantasy s
i not only sex, it is sex plus "the soul." Impossible
to account for falling in love or even for infatuations. simple
attractions or Wertherian raptures, without admitting that what

C :lpynghled malenal
 Rtadings
is sought in the other is something we shall call, for lack of a
better word and at the cost of great ambiguity. the person. To
the person is attached a kind of homing device that causes this
particular image, among thousands of others, to seek out and
capture me. Bodies can be classified into a finite number of
types ("That's just my type"), but the person is absolutely
individual. Renaud Camus's Triclu always begin with an encoun
ter with the longed-for type (perfectly encoded; the type could
figure in a catalogue or in a page of personal want ads); but
once language appears. the type is transformed into a person,
and the relation becomes inimitable, whatever the banality of
the first remarks. The person is gradually revealed, and lightly.
without psychologizing. in clothing, in discourse. in accent, in
setting, in what might be called the individual's "domesticity,"
which transcends his anatomy, yet over which he has control.
All of which gradually enriches or retards desire. The trick is
therefore homogeneous to the amorous progression; it is a
virtual love, deliberately stopped short on each side, by contract;
a submission to the cultural code which identifies cruising with
Don Juanism.
The Triclts repeat themselves; the subject is on a treadmill.
Repetition is an ambiguous form; sometimes it denotes failure,
impotence; sometimes it can be read as an aspiration, the
stubborn movement of a quest which is not to be discouraged;
we might very well take the cruising narrative as the metaphor
of a mystical experience (perhaps this has even been done; for
in literature everything exists: the problem is to know where).
Neither one of these interpretations, apparently, suits Triclu:
neither alienation nor sublimation; yet something like the me
thodical conquest of happiness (specifically designated, carefully
boundaried: discontinuous). The Aesh is not sad (but it is quite
an art to convey as much).
Renaud Camus's Triclts have an inimitable tone. It derives
from the fact that the writing here imitates an ethic of dialogue.
This ethic is that of Good Will, which is surely the virtue most
contrary to the amorous pursuit, and hence the rarest. Whereas

C :lpynghled malenal
Preface to Renaud Camus's Tricks 2 95
ordinarily a kind of harpy presides over the erotic contract,
leaving each party within a chilly solitude, here it is the goddess
Eunoia, the Eumenid, the Kindly One, who accompanies the
two partners; certainly, in literary terms, it must" be very agree
able to be "tricked" by Renaud Camus, even if his companions
do not .always seem aware of this privilege (but we, the readers,
are the third ear in these dialogues: thanks to us, this bit of
Good Will has not been vouchsafed in vain). Moreover, this

goddess has her retinue: Politeness, Kindness, Humor, Generous

Impulse, like the one which seizes the narrator (while tricking
with an American) and causes his wits to wander so amiably
with regard to the author of this preface.
Trick the encounter which takes place only once: more than
cruising, less than love: an intensity, which passes without regret.
Consequently, for me, Trick becomes the metaphor for many
adventures which are not sexual; the encounter of a glance, a
gaze, an idea, an image, ephemeral and forceful association,
which consents to dissolve so lightly, a faithless benevolence: a
way of not getting stuck in desire, though without evading it;
all in all, a kind of wisdom.

Preface to Renaud Camus's Triclcs (Persona), 1979

One Always Fails in Speaking of What One Loves 297
it occurs without any apparent reason. Music, for Stendhal, is
the symptom of the mysterious action by which he inaugurated
his transference the symptom, i.e., the thing which simulta
neously produces and masks passion's irrationality. For once
the opening scene is established, Stendhal constantly reproduces
it, like a lover trying to regain that crucial thing which rules so
large a share of our actions: the first pleasure. "I arrive at seven
in the evening, tormented with fatigue; I run to La Scala. My
journey was justified," etc.: like some madman disembarking in
a city favorable to his passion and rushing that very evening to
the haunts of pleasure he has already located.
The signs of a true passion are always somewhat incongruous,
the objects of transference always tending to become. tenuous,
trivial, unforeseen . . . I once knew someone who loved Japan
the way Stendhal loved Italy; and I recognized the same passion
in him by the fact that he loved, among other things, the red
painted fireplugs in the Tokyo streets, just as Stendhal was mad
for the cornstalks of the "luxuriant" Milanese campagna, for
the sound of the Duomo's eight bells, "perfectly intonate," or for
the pan-fried cutlets that reminded him of Milan. In this erotic
promotion of what is commonly taken for an insignificant detail,
we recognize a constitutive element of transference (or of
passion): partiality. In the love of a foreign country there is a
kind of reverse racism: one is delighted by Difference, one is
tired of the Same, one exalts the Other; passion is Manichaean
for Stendhal, the wrong side is France, i.e., Ia patrie-for it is

the site of the Father and the right side is Italy, i.e., Ia matTie,
the space in which "the Women" are assembled (not forgetting
that it was the child's Aunt Elisabeth, the maternal grandfather's
sister, who pointed her finger toward a country lovelier than
Provence, where according to her the "good" side of the family,
the Gagnon branch, originated). This opposition is virtually
physical: Italy is the natural habitat, where Nature is recovered
under the sponsorship of Women, "who listen to the natural
genius of the country," contrary to the men, who are "spoiled
by the pedants"; France, on the contrary, is a place repugnant
 Readings
"to the point of physical disgust." All of us who have known
Stendhal's passion for a foreign country (this was also my case
for Italy, which I discovered belatedly, in Milan, at the end of
the nineteen-fifties then for Japan) are familiar with the in
tolerable annoyance of encountering a compatriot in the adored
country: "I must confess, though it goes against the national
honor, that finding a Frenchman in Italy can destroy my
happiness in a moment"; Stendhal is visibly a specialist in such
inversions: no sooner has he crossed the Bidassoa than he is
charmed by the Spanish soldiers and customs officers; he has
that rare passion, the passion for the other or to put it more
subtly: the passion for that other which is in himself.
Thus, Stendhal is in love with Italy: this is not a sentence to
be taken metaphorically, as I shall try to show. "It is like love,"
he says: "and yet I am not in love with anyone." This passion is
not in the least vague; it is invested, as I have said, in specific
details; but it remains plural. What is loved and indeed what is
enjoyed are collections, concomitances: contrary to the romantic
project of Amour fou, it is not Woman who is adorable in Italy,
but always Women; it is not a pleasure which Italy affords, it is
a simultaneity, an overdetermination of pleasures: La Scala, the
veritable eidetic locus of Italian delights, is not a theater in the
word's banally functional sense (to see what is represented); it
is a polyphony of pleasures: the opera itself, the ballet, the
conversation, the gossip, love, and ices (gelati, crepe, and pezzi
duTi). This amorous plural, analogous to that enjoyed today by
someone "cruising," is evidently a Stendhalian principle: it
involves an implicit theory of irregular discontinuity which can be
said to be simultaneously aesthetic, psychological, and meta
physical; plural passion, as a matter of fact once its excellence
has been acknowledged necessitates leaping from one object
to another, as chance presents them, without experiencing the
slightest sentiment of guilt with regard to the disorder such a
procedure involves. This conduct is so conscious in Stendhal
that he comes to recognize in Italian music which he
principle of irregularity quite homologous to that of dispersed
One Always Fails in speoJcing of what One Louts '99
love: in performing their music, Italians do not observe tnnpo;
tempo occurs among Germans; on one side, the German noise,
the uproar of German music. beating out an implacable measure
("the first lempi.5ti in the world"); on the other. Italian opera,
summa of discontinuous and untamed pleasures: this is the
natural. guaranteed by a civilization of women.
I n Stendhal's Italian system, Music has a privileged place
because it can replace everything else: it is the degree zero of
this system: according to the needs of enthusiasm. it replaces
and signifies journeys, Women. the other arts, and in a general
manner any sensation. Its signifying status, precious above all
others. is to produce effects without our having to inquire as to
their causes, since these causes are inaccessible. Music constitutes
a kind of primal state of pleasure: it produces a pleasure one
always tries to recapture but never to explain; hence, it is the
site of the pure effect. a central nOlion of the Stendhalian
aesthetic. Now. what is a pure effect? An effect severed from
and somehow purified of any explicative reason, i.e ultimately.
.

of any responsible reason. Italy is the country where Stendhal.

being neither entirely a traveler (a tourist) nor entirely a native,
is voluptuously delivered from the responsibility of the ciliun;
if Stendhal were an Italian citizen. he would die "poisoned by
melancholy": whereas, a Milanese by affection rather than civil
status, he need merely harvest the brilliant effects of a civilization
for which he s
i not responsible. I have been able to experience

the convenience of this devious dialectic m}'self: I used to love

Morocco, I often visited the country as a tourist, even spending
rather long vacations there; therefore, it occurred to me to
spend a year there as a professor: the magic vanished; con
fronted by administrative and professional problems, I plunged
into the ungrateful world of causes and allegiances. I surren
dered Festivity for Duty (this is doubtless what happened to
Stendhal as consul: Civitavecchia was no longer Italy). I believe
we must include within Stendhal's Italian sentiment this fragile
status of innocence: Milanese Italy (and its Holy of Holies, La
Scala) is literally a Paradise, a place without Evil, or again-let

C ;.pvnghted malaria
goo Rtadings

us put matters positively-the Sovereign Good: "When I am

with the Milanese and speak their dialect, I forget tbat men are
wicked, and the whole wicked aspect of my own soul instantly
falls asleep."
This Sovereign Good, however, must express itself: hence, it
must confront a power which is nOI at all innocent, language.
This is necessary. first of all, because the Good has a natural
force of expansion, it constantly opens out toward expression,
seeks at all costs to communicate itself, to be shared; then,
because Stendhal is a writer and because for him there is no
fulfillment from which the word is absent (and in this his Italian
delight has nothing mystical about it). Now, paradoxical as it
seems, Stendhal is uncertain how to express Italy: or rather, he
speaks it, he sings it. he does not represent it; he proclaims his
love, but he cannot express it, or, as we say nowadays (a metaphor
from driving), he cannot negotiate it. This he knows, this he
suffers from, this he complains of; he constantly observes that
he cannot "render his thought," and that to explain the differ
ence his passion proposes between Milan and Paris "is the height
of difficulty." Hence, fiasco lies in wait for lyric desire as well.
All his accounts of Italian travels are strewn with declarations
of love and of failures of expression. The fiasco of style has a
name: platitude; Stendhal has at his disposal only one empty
word: beautiful ("beau," "belle';: "In all my life, I have never
seen so many beautiful women together; their beauty made me
lower my eyes." "The most beautiful eyes I have ever encoun
tered I saw that e\'ening; such eyes are as beautiful and their
expression as celestial as those of Mme Tealdi . . . " And in
order to vivify this litany, he has only the emptiest of figures,
the superlative: "The women's heads, on the contrary, often
reveal the most impassioned finesse, uniting the rarest beauty,"
etc. This "etc." which I am adding, but which emerges from
our reading, is important, for it yields up the secret of this
impotence or perhaps, despite Stendhal's complaints, of this
indifference to variation: the monotony of Italian travel is quite
simply algebraic: the word, the syntax, in their platitude, refer

C :lpynghled malenal
Om Always Fails in Speaking of What Om Loves
expeditiously to a different order of signifiers; once this refer
ence si suggested, Stendhal moves on to something else, i.e.,
repeats the operation: "This is as beautiful as Haydn's liveliest
symphonies." "The men's faces at the ball that night would have
afforded magnificent models to sculptors of busts like Danneker
or Chamrey." Stendhal does not describe the thing, he does not
even describe the effect; he simply says: there, there is an effect:
I am intoxicated, transported, touched, dazzled, etc. In other
words, the platitudinous word is a cipher, it refers to a system
of sensations; we must read Stendhal's Italian discourse like a
figured bass. Sade employs the same procedure: he describes
beauty very poorly, platitudinously and rhetorically; this is
because beauty is merely the element of an algorithm whose
goal is to construct a system of practices.
What Stendhal wanu to construct is, so to speak, a non
systematic set, a perpetual How of sensations: that Italy, he says.
"which is in truth merely an occasion for sensations." From the
point of view of discourse, then. there is an initial evaporation
of the thing: "I am not claiming to say what things are, I am
describing the snuation the}' produce upon me." Does he really
describe it? Not really; he says it, indicates it, and asserts it
without describing it. For it is just here. with sensation, that the
difficulty of language begins; it is not easy to "render" a sensation:
you recall that famous scene in Jules Romains's play Knoell in
which the old peasant woman, ordered by the implacable doctor
to say how she feels, hesitates in her confusion between "it
tickles" and "it itches" (fa me chatouilu I fa me gratouilUj. Any
sensation, if we want to respect its vivacity and its acuity, leads
to aphasia. Now Stendhal must go quickly, that is the constraint
of his system; for what he wants to note is "the sensation of the
moment"; and the moments, as we have seen apropos of tempo.
occur irregularly, refractory to measure. Hence, it is by fidelity
to his system, fidelity to the very nature of his Italy, "a country
of sensations," that Stendhal seeks a rapid writing: in order to
proceed quickly, sensation is subjected to an elementary stenog
raphy. to a kind of expedient grammar of discourse in which

GJPYnghtcd maknal
Ont Always Fails in Speaking of What Ont Lavts
preamble o f TJu Charltrhotue, which Stendhal was quite right to
retain. despite Balzac's reservations: festivity, i.e., the very
transcendence of egotism.
In short. what has happened-what has transpired-between
the travel journals and The Charttrhowe. is writing. Writing
which is what? A power. probable fruit of a long initiation,
which annuls the sterile immobility of the amorous image
repertoire and gives its adventure a symbolic generality. When
he was young. in the days of R ome. Napus. Florenct. Stendhal
could write: ". . . when I tell lies. I am like M. de Goury. I am
bored"; he did not yet know that there existed a lie. the lie of
novels. which would be miraculously-both the detour of truth
and the finally triumphant expression of his Italian passion.

Ttl Qwl. 1980

I...1 1 1


o>Schiitz ".
 Writers. Intellectuals. Teachers

What follows depends on the idea that there s i a basic link

between teaching and speech. This is a very old observation
(has not the whole of OUT teaching emerged from Rhetoric?),
but we can frame it differentJy. nowadays, from the way we
used to; first of ail, because there is a (political) crisis in teaching;
then. because (Lacanian) psychoanalysis has exposed the twists
and turns of La parole vide; finally, because the opposition between
speech and writing has become so obvious that we must begin
deducing its effects.
As opposed to the teacher, who is on the side of speech. let
us call any operator of language on the side of writing a writtr;
between the two, the intellectual: the subject who prints and
publishes his speech. There is virtually no incompatibility be
tween the teacher's language and the intellectual's (they fre
quently coexist in one and the same individual); but the writer
is alone, separate: writing begins at the point where speech
becomes impossibu (as the word is used about a child),

Two constraints
Speech is irreversible: a word cannot be retraced except precisely
by saying that one retracts it. Here, to cancel is to add; if I want
to erase what I have just said, I can do so only by showing the
eraser itself (I must say: "or rathtT . . ," '" txprtsstd mystlf badly
.

.. ''); paradoxically, it is ephemeral speech which is indelible.

.

not monumental writing. There is nothing to be done with

speech but add on more. The corrective and perfective move
ment of speech is a stammer, a fabric that frays out as it knits
itself together. a chain of augmentative corrections which is the

I...1 1''''''
 Environs of the Image
chosen abode of the unconscious part of our discourse (it is no
accident that psychoanalysis is linked to speech, not to writing:
a dream is not wriuen): the eponymous figure of the speaker is
Penelope.
Nor is this all: we cannot make ourselves understood (properly
or poorly) unless we maintain, as we speak, a certain spud of
delivery. We are like a bicyclist or a film---doomed to keep
riding or turning if we are to avoid falling or jamming; silence
and hesitation are both denied us: articulatory speed subjugates
each point of the sentence to what immediately precedes or
follows it (impossible to "venture" the word in the direction of
odd or alien paradigms); context is a structural datum not of
language but of speech; now, context by its very status is
reductive of meaning, the spoken word is "clear"; such clarity,
the banishment of polysemy, serves the Law: all speech is on the
suu of the Law.
Anyone preparing to speak (in a teaching situation) must
become conscious of the staging imposed,by the use of speech,
by the simple effect of a flllu
l Tal determination (which derives
from its physical nature: that of ariculatory breathing). This
staging develops in the following way. Either the speaker chooses
the role of Authority in all good faith, in which case it is sufficient
to "speak well," i.e., to speak in accordance with the Law inherent
in all speech: without stammering, at the right speed; or again
clearly (this is what is asked of good pedagogical discourse:
clarity, authority); the distinct sentence is a sentence indeed,
srntnltia, a penal speech. Or else the speaker is hampered by
the Law which his speech will introduce into what he wants to
say; of course, he cannot spoil his delivery (which condemns
him to "clarity"), but he can apolfJgiu for speaking (for laying
down the Law): in this case he uses the irreversibility of speech
to breach its legality; he corrects himself, adds on, stammers,
enters into the infinitude of language, superimposes upon the
simple message expected of him a new message which spoils
the very notion of message, and by the shimmer of blemishes
and mistakes with which he accompanies his speech line, he

N , 1 1"';'
Wrilm, InuUectuals, Ttachers
asks us to believe with him that language is not reduced to
communication. By all these operations. which bring his stam
mering closer to the Text. the imperfect orator hopes to
attenuate the ungrateful role which makes every speaker a kind
of policeman. Yet. at the end of this effort to "speak badly."
there is still a role which is imposed upon him: for the audience
(nothing to do with the reader), trapped in its own image
repertoire. receives this fumbling as so many signs of weakness
and offers him the image of a human. all too human master: a
liberal one.
The choice is grim: conscientious functionary or free artist.
the teacher escapes neither the theater of speech nor the Law
staged within it: for the Law is produced. not in what he says. but
in tJu fact that Iu sptalu at all. In order to subvert the Law (and
nO[ simply to get around it). he would have to dismantle all
vocal delivery, the speed and rhythm of words. until he achieved
an altogether different intelligibility-or else not speak at all; but
then he would be back in other roles: either that of the great
silent mind. heavy with experience and reserve, or that of the
militant who in the name of frraxis dismisses all discourse as
trivial. No help for it: language is always on the side of power;
to speak is to exercise a will to power: in the space of speech,
no innocence, no safety.

The summary
Statutorily. the teacher's discourse is marked by this character
istic: that it can (or may) be summarized (this is a privilege it
shares with the discourse of politicians). In French schools, for
example, there is an exercise known as tnt reduction; an expres
sion which nicely expresses the ideology of the summary: on
one side there is "thought," object of the message. element of
action. of knowledge, a transitive or critical force, and on the
other there is "style," an ornament associated with luxury,
idleness. hence with the trivial; to separate thought from style
is in a sense to rid discourse of its sacerdotal garments. to

B
3" Environs of tJu Image
secularize the message (whence the bourgeois identification of
teacher and politician); "form," it is assumed, is compressible,
and such compression is not regarded as essentially detrimental:
from a distance, Le., from our Western promontory, is there really
such a difference between the head of a living Jivaro and a
shrunken Jivaro head?
It is difficult for a teacher to see the "notes" taken during his
lectures; he has no desire to do so, either out of discretion (for
nothing is more rsonal than "notes," despite the formal
character of this practice) or else, more likely, out of fear of
discovering himself in a reduced version, dead yet substantial,
like a Jivaro "treated" by bis fellow tribesmen; not knowing if
what is taken (siphoned) out of the How of speech is erratic
statements (formulae, sentences) or the gist of an argument; in
either case, what is lost is the .supplement, just where the stake of
the language is advanced: the summary is a disavowal of writing.
Conversely, therefore, we may call a "writer" (this word always
designating a practice, not a social value) any sender whose
"message" (thereby immediately destroying its nature as mes
sage) cannot be summarized: a condition which the writer shares
with the madman, the compulsive talker, and the mathematician,
but which it is precisely writing's task (i.e., the task of a certain
practice of the signifier) to specify.

The teaching relationship

How can the teacher be identified with the psychoanalyst? It is
precisely the contrary which occurs: it is the teacher who is
psychoanalyzed.
Imagine that I am a teacher: I speak, endlessly. for and
before someone who does not speak. I am the one who says I
(the detours of one or Wt, of the impersonal sentence, are
insignificant), I am the one who, under the cover of an exposition
(of something known), proprut.s a discourse, without ever Jcnowing
how it iJ rutivtd, so that I can never have the reassurance of a
definitive (even if damaging) image which might corutitute me:

,
Writm. Inle[kctuaLs. Ttachen
in the npose, more aptly named than we assume. it s
i not
knowledge which is exposed. it is the subject (who exposes himself
to painful adventures). The mirror is empty: it reflects only the
defection of my language as it emerges. Like the Marx Brothers
disguised as Russian aviators (in A Nighi at the Opera-a work I
regard as an allegory of many textual problems), 1 begin my
expose decked out in a big false beard; but, gradually flooded by
the rising tide of my own speech (a substitute for the pitcher of
water on the rostrum of the Mayor of New York from which
the mute Harpo greedily drinks), 1 feel my beard peeling off in
front of everyone: no sooner have I made the audience smile
by some "clever" remark, no sooner have I reassured it by some
"progressive" stereotype, than I feel the complacency of such
provocations; I regret the hysterical pulsion, I'd like to retract
it. preferring (too late) an austere discourse to a "smart" one
(but, in the contrary case, it is the "severity" of the discourse
which seems hysterical to me); should some smile answer my
remarks, or some assent my intimidation, I am immediately
convinced that such manifestations of complicity are produced
by fools or flatterers (I am here describing an imaginary process);
though it is I who want a response and take steps to provoke
it. once the response comes. I am suspicious; and if the discourse
1 offer chills or postpones any response at all, I do not feel I
am any more on pitch on that account, for then 1 must glory in
the solitude of my speech, furnish it the alibi of missionary
discourse (science. truth, etc.).
Hence, in accordance with psychoanalytic description (Lacan's,
whose perspicacity here an)' speaker will confirm), when the
teacher speaks to his audience. the Other is always there.
puncturing his discourse; and his discourse, though sustained by
an impeccable intelligence. armed with scientific "rigor" or
political "radicality," would still be punctured: it suffices that I
speak, that my speech flows, for it to How away. Of course,
though every professor is in the posture of an analysand, no
student audience can claim the advantage of the converse
situation; first, because there is nothing preeminent about

B
Writers, Intellectuals, Teachers 3'5

whom you cannot place with regard to the image he asks of you
becomes disturbing); (11:) to act as a relay. to extend him. to
spread his ideas. his style, far afield; (3) to let himself be seduced,
to assent to a loving relationship (granting all the sublimations,
all the distances, all the forms of respect consonant with social
reality and with the anticipated futility of this relationship); last,
to permit him to honor the contract he himself has entered into
with his employer, i.e., with society; the peT50n taught is part
of a (remunerated) practice, the object of a craft, the substance
of a production (however delicate to define).
On his side. here. peUmell. is what the person taught asks of
the teacher; (I) to guide him to a favorable professional inte
gration; (11:) to fulfill the roles traditionally devolving upon the
teacher (scientific authority, transmission of a body of knowl
edge, etc.); (3) to reveal the secrets of a technique (of research,
of examinations, etc.); (4) under the banner of that secular saint
Method, to be an initiator of ascesis, a guru; (5) to represent a
"movement of ideas," a School, a Cause, to be ilS spokesman;
(6) to admit him, the student, into the complicity of a private
language; (7) for those obsessed by the thesis (a timid practice of
writing. both disfigured and defended by its institutional final
ity), to guarantee the reality of this fantasy; (8) last. the teacher
is asked to lend his services: to sign forms, recommendations,
etc.
This is simply a Topic, a set of choices which are not all
necessarily actualized at the same time in one individual. Yet it
is al the level of contractual totality that the comfort of the
teactling relation is worked out: the "good" teacher, the "good"
student are those who philosophically accept the plurality of
their determinations, perhaps because they know that the truth
of a relationship of speech is elsewhere.

Research
What is a piece of "research"? To find out, we need some idea
of what a "result" might be. What does one find? What does
one want to find? What is missing? In which axiomatic field will

B
 Environs of the Image

the phenomenon be isolated. the meaning revealed. the statistical

discovery placed? No doubt, in each case this depends on the
particular science solicited. But once a piece of research concerns
the text (and the text reaches much fanher than the work), the
research itself becomes text, production: any "result" is literally
im-pertinent. "Research," then, is the discreet name which, under
me constraint of certain social conditions, we give to me work
of writing: research is on the side of writing. it is an adventure
of the signifier. an excess ofexchange; it is impossible to maintain
the equation of a "result" against a piece of "research." This is
why the discourse to which a piece of research must be submitted
(in teaching it) has a purpose beyond its parenetic function
("Write!"): it is intended to recaU"research" to its epistemological
condition: whatever it seeks, it must not forget its nature as
language-and it is this which ultimately makes an encounter
with writing inevitable. In writing. the speech-act "deceives" the
message by the effect of the language which produces it: this
adequately defines the critical, "progressive." unsatisfied, pro
ductive element which even common usage grants to "research."
Such is the historical role of research: to teach the scholar he
spealcs (but if he knew this. he would write-and the whole idea
of science, 5cientifzciry itself. would be changed thereby).

The destruction of stereotypes

Someone writes me that "a group of revolutionary students is
preparing a destruction of the structuralist myth." The expres
sion delights me by its stereotypical consistency: the destruction
of the myth begins. from the statement of its putative agents.
with the finest myth of all: the "group of revolutionary students"
is quite as good as "war widows" or " old soldiers."
Usually the stereotype is a sad business because it is constituted
by a necrosis of language. a prosthesis used to plug a hole in
writing; but at the same time it can only provoke a huge burst
of laughter: it takes itself seriously; supposes itself closer to the

,
Writers, Intellectuals, Teachers
truth because indifferent to its nature as language: it is both
corny and solemn.
To distance the stereotype is not a political task, for political
language itself is constituted by stereotypes; but a critical one,
i.e., one seeking to put language in a state of crisis. Initially,
this permits us to isolate that speck of ideology contained in all
political discourse, and to attack it like an acid capable of
dissolving the fats of "natural" language (i.e., of the language
which pretends not to know it is language). Then, it is a way of
breaking with the mechanistic conception which makes language
the simple response to stimuli of situation or of action; it sets
the production of language in opposition to its simple and
fallacious utilization. Further, it disturbs the Other's discourse
and constitutes, in short, a permanent operation of pre-analysis.
Last of all: the stereotype is at bottom a form of opportunism:
one conforms to the ruling language, or rather to what, in lan
guage, seems to rule (a situation, a privilege, a struggle, an
institution, a movement, a science, a theory, etc.); to speak in
stereotypes is to side with the power of language; it is this
opportunism which must (today) be refused.
But can one not "transcend" stereotypes instead of "destroy
ing" them? Such a solution is unrealistic; the operators of
language have no power except to empty what is full; language
is not dialectical: it allows only a two-stage movement.

The chain of discourses

It is because language is not dialectical (permitting the third
term only as a formula, a rhetorical assertion, a pious hope)
that discourse (discursivity), in its historical development, moves
by fits and starts. Each new discourse can emerge only as the
pa radox which reverses (and often opposes) the surrounding or
preceding doxa; it can be generated only as difference, distinc
tion, standing out against what sticks to it. For instance, Chom
skyan theory is built up against Bloomfieldian behaviorism; then,
once linguistic behaviorism has been liquidated by Chomsky, it
Writers, Intellectuals, Teachers 3 '9

The danger of Method (of a fixation upon Method) comes

from this: research work must satisfy two demands; the first is
a demand for responsibility : the work must increase lucidity,
expose the implications of a procedure, the alibis of a languag'e-
e-
in short, must constitute a critique (let us recall once again that
to criticiu means to call into crisis); here Method is inevitable,
irreplaceable, not for its "results" but precisely or on the
contrary because it realizes the highest degree of consciousness
of a language which does not forget itself; but the second demand
is of a very different order: it is the demand for writing, for a
space of desire's dispersion, where Law is dismissed; hence, it
is necessary, at a certain moment, to turn against Method, or at
least to regard it without any founding privilege, as one of the
voices of plurality: as a view, in short, a spectacle, mounted
within the text the text which is, after all, the only "true" result
of any research.

Questions
To question is to want to know something. Yet, in many
intellectual discussions, the questions which follow the lecturer's
remarks are in no way the expression of a "want," but the
assertion of a plenitude. In the guise of questioning, I mount
an aggression against the speaker; to question then takes on its
police meaning: to question is to interpellate. Yet the interpellated
subject must pretend to answer the letter of the question, not
its "address." So a game is set up: although each side knows just
what the other's intentions are, the game demands a response
to the content, not to the way that content is framed. If I am
asked, in a certain tone, "What's the use of linguistics?" thereby
signifying to me that it's of no use whatever I must naively
pretend to answer: "Linguistics is useful for this, for that," and
not, in accordance with the truth of the dialogue: "Why are you
attacking me?" What I receive is the connotation; what I must
give back is the denotation. In the space of speech, science and
logic, knowledge and argument, questions and answers, prop-
320 Enviroru of the Image
ositions and objections are (he masks of the dialectical relation.
OUT intellectual discussions are as encoded as the old scholastic
disputes; we still have the stock roles (the "sociologist," the
"Goldmannian," the "TelqueJian," etc.), but contrary " to the
disputatio, where these roles would have been ceremonial and
have displayed the artifice of their function, our intellectual
"intercourse" always gives itself "natural" airs: it claims to
exchange only signifieds, not signifiers.

In the name of what?

I speak in the name of what? or a function? Of a body of
knowledge? Of an experience? What do I represent? A scientific
capacity? An institution? A service? As a matter of fact, I speak
only in the name of a language: it is because I have written that
I speak; writing is represented by its contrary, speech. This
distortion means that in writing about spee<:h I am doomed to
the following aporrhoea: denouncing spee.:h's image.repertoire
through writing's unreality; thus, at present. ! am not describing
any "authentic" experience. I am not photographing any "real"
teaching, I am opening no "academic" dossier. For writing can
tell the truth about language. but not the truth about reality
(we are at present trying to learn what a reality without language
might be).

The standing position

What could be a more dubious situation than speaking for (or
before) people who are standing or uncomfortably seated? What
is bdng exchanged here? What is such discomfort the price of?
What is my speech worth? How could the hearer's discomfort
fail to lead to questions about the validity of what he is hearing?
Is not the standing position eminently critical? And is this not
how, on another scale. political consciousness begins: in discom
fort? Listening refers me to the vanity of my own speech, its

Matena dlreitl
Writers. InUlkctuals. Ttachers
pritt, for willy-nilly I am placed in a circuit of exchange; and
linening is also the position of the peT$On I address.

Familiarity
Occasionally-the wreckage of May '68-a student will speak to
a teacher using the familiar tu form of address. This is a strong
sign, a full sign, which refers to the mon psychological of
signifieds: the will to contestation or closeness-muscle. Since
an ethic of the sign is here imposed. it can be contested in its
turn and a subtler semantics preferred: signs are to be dealt
with against a ntutral background. and. in French. that background
is the vow form of address. The tu form can escape the code
only in cases where it constitutes a simplifICation 0/ grammaT (for
example. in addressing a foreigner who speaks French poorly);
in such cases. it is a matter of substituting a transitive practice
for a symbolic behavior: instead of trying to signify jwt who I
think the other is (and therefore just who I think I am), I simply
try to make myself understood by him. But this recourse. too.
is ultimately devious: the tu form becomes another form of
escape behavior; when a sign displeases me. when the signifi
cation bothers me, . shift toward the operational: the operational
becomes a censorship of the symbolic, and thus a symbol of
asymbolism; many political and scientific discourses are marked
by this shift (on which depends, notably, the entire linguistics
of "communication").

An odor of speech
As soon as you have finished speaking, the vertigo of the image
begins: you exalt or regret what you have said, the way you
have said it, you imagine YOUTStlf (you consider yourself as an
image); speech is subject to remanence. it smells.
Writing has no mull: produced (having accomplished its
process of production). it/ails. not like a collapsing souffle but
like a meteorite disappearing; it will trawl far from my body

B
Writers, IntLlkctuals, Tachm

in hiJ final plau. Then we should see that what must be made
reversible are not the social "roles" (""hat is the point of quibbling
over "authority," over the "right" to speak?) but the regions of
speech. Wher is speech? In locution? In listening? In the returns
ofeach? The problem is not to abolish the distinction offunctions
(the tacher, the student: after all, as Sade has taught us, order is
a guarantee of pleasure) but to protect the instability and. so to
speak. the vertigo of the sites of speech. In the teaching space,
no one should anywhere be in his place (1 am reassured by this
constant displacement: were I tofind my plac, 1 should not even
go on pretending to teach, I should give up).
Yet does not the teacher have a fixed place that of his
rtmuneration, the place he has in the economy, in production?
It is always the same problem we keep coming back to: the
origin of a speech does not exhaust it; once the speech is spoken,
a thousand adventures happen to it, its origin becomes confused,
all its effects are not in its cause; it is this redundancy which we
are exploring.

Two criticisms

The mistakes that can be made in typing a manuscript are so

many meaningful incidents, and these incidenu, by analogy,
help shed light on the behavior we must adopt with regard to
meaning when we discuss a text.
Either the word produced by the mistake (if a wrong letter
disfigures it) signifies nothing, finds no textual contour; the
code is simply interrupted: an asemic word is created, a pure
signifier; for example, instead of writing "officer," I write
"offiver," which is meaningless. Or else the erroneous (mistyped)
word, without being the word I wanted to write, is a word which
the lexicon permits identifying, which means something: if I
write "ride" instead of "rude," the new word exists--the phrase
keeps a meaning, however eccentric; this is the path (the pith?)
of puns, anagrams, of signifying metathesis, of spoonerisms:

B
 Environs of tJu lmagt
there is a skid within tlu codt.s: meaning subsists. but pluralized.
faked. without law of content. of message. of truth.
Each of these two types of mistake figures (or prefigures) a
type of criticism. The first type dismisses all meaning of the
support text: the text must lend itself only to a signifying
emorescence; it is its phonism alone which is to be treated. but
not interpreted: we aS5OCiate, we do not decipher; reading offiVtr
instead of Offtctr. I gain access, by the mistake. to a right of
association (I can make "offiver" explode in the direction of
"offer," of "olive," etc.): not only does the ear of this first criticism
hear the cracklings of the phono-pickup, but they are all it
wants to hear-it will make them into a new music. I n the
second type of criticism, the "reading head" rejects nothing: it
perceives both the meaning (s) and its cracklings. The (historical)
stake of these two criticisms (I should like to be able to call the
field of the first signifwsis, and that of the second, signifUlnct) is
obviously different.
The first has in its favor the signifier's right to spread out
where it likes (where it can?): what law and what meaning
from what soured-would be entitled to constrain it? Once the
philological (monological) law has been relaxed and the text
opened to plurality, why stop? Why refuse to take polysemy to
the point of asemy? In the name of what? Like any radical right,
this one supposes a utopian vision of freedom: we revoke the
law immtdiattly, outside of any history, in defiance of any dialectic
(which is why this style of demand can seem ultimately petit
bourgeois). However. once it evades all tactical reason while
nonetheless remaining implanted within a specific (and alien
ated) imellectuaJ society, the signifier's disorder reverts to hys
terical wandering: by liberating reading from all meaning, it is
ultimately my reading which I impose, for in this moment of
History the subject's economy is not yet transformed, and the
rejection of meaning(s) is turned into subjectivity. At best, we
can say that this radical criticism, defined by a foreclosure of
the signifier (and not by its escape), anticipates History, anticipates
a new, unprecedented state in which the signifier's efflorescence

B
Writm, /nlellectuals, TeacMrj

poverty and folklore-the plenitude of the other discourse, within

which the proletariat itself is compelled to live, seeking survival,
diversion. etc. : this discourse is that of culture (it may be that,
in Marx's day, culture's pressure on the proletariat was weaker
than it is today: there was as yet no "mass culture" because
there was no "mass communication"). How to attribute meaning
for the struggle to what does not directly concern you? How
could the proletariat determine. within il.self. an interpretation
of Zola. of Poussin. of pop art, of the Sunday spons supplement,
or the latest news item? In order to "interpret" all these cultural
relays. the proletariat requires representatives: those whom Brecht
calls the "artists" or the "workers of the intellect" (the expression.
at least in French, is mischievous one. the intellect being so
nearly off the top of the head), those who have at their command
the language of the indirect, the indirect as language; in short,
oblates, who dedicate themselves to the proletarian interpretation
of cultural phenomena.
But then begins, for these procurators of proletarian meaning,
a real headache, for their class situation is not that of the
proletariat: they are not producers--a negative situation they
share with (student) youth. an equally unproductive class with
which they usually form an alliance of language. It follows from
this that culture. from which it is their responsibility to derive
a proletarian meaning. ultimately refers them to themselves and
not to the proletariat: How to evaluate culture? According to its
origin? It is bourgeois. According to its goal? Bourgeois again.
According to dialectics? Although bourgeois, this does contain
"progressive" elements; but what, on the level of discourse, distin
guishes dialectics from compromise? And then, with what in
struments? Historicism, sociologism, posiiivism. formalism,
psychoanalysis? Every one of them embourgeoisi. Some people
will ultimately prefer to skip the headache: to dismiss all
"culture." which obliges them to destroy all discourse.
As it happens, even within an axiomatic field apparently
clarified by the class struggle. the tasks are diverse, sometimes
contradictory. and above all established according to different

B
 Environs of the Image
tenses. The axiomatic field consists of several individual axio
matics: cultural criticism develops successively. diversel,. and si
multanequsiy by opposing the New to the Old, sociologism to
historicism, economism to formalism. logico-positivism to psy
choanalysis. then again. by arwthtr tum. monumental history to
empirical sociology. the alien to the New, formalism to histori
cism. psychoanalysis to scientism, etc. Applied to culture, critical
discourse can only be a cross-hatching of tactics, a tissue of
elements now past, now circumstantial (linked to contingencies
of fashion). and now frankly utopian: to the tactical necessities
of the war of meanings is added the strategic conception of the
new conditions which will be applied to the signifier when this
war is over: cultural criticism, as a matter of fact, must be
impatient, because it cannot be carried on without desire. Hence,
all the discourses of Marxism are present in its writing: apologetic
(to exalt revolutionary science), apocalyptic (to destroy bourgeois
culture), and eschatological (to desire, to call for the indivision
of meaning, concomitant with the indivision of classes).

Our unconscious
The problem we set ourselves is this: how can the two great
epistemts of modernity, i.e., materialist and Freudian dialectics,
be made to converge. intersect, and produce a new human
relation (nor is it to be excluded that a third term lurks in
the interdiction of the first two)? Which is to say: how to as
sist the inter-action of these two desires: to change the economy
of the relations of production and to change the economy of
the subject? (Psychoanalysis temporarily seems to us the power
better suited to the second of these tasks; but other topics are
imaginable, those of the East, for example.)
This generalized task comes up against the following question:
What relation is there between class determination and the
unconscious? By what displacement does this determination
"slip" between subjects? Certainly not by "psychology" (as
if there were mental contents: bourgeois I proletarian lin-

B
Writtrs, Inlelkctuau, Teachers
tellectual, etc.), but obviously by language, by discourse: the
Other-who speaks, who is aU speech-is social. On one side,
though the proletariat is separated, it is still bourgeois language,
in it! degraded. petit-bourgeois form, which unconsciously
speaks in the proletariat's cultural discourse; and on the other
side, though the proletariat is mute, it speaks in the discourse
of the intellectual, not as a canonical, founding voice, but as an
unconscious: it suffices to see how the proletariat striAa all our
discourses (explicit reference of the intellectual to the proletariat
in no way prevents the proletariat from occupying, in OUT
discourses, the place of the unconscious: the unconscious is not
"lack of consciousness"); only the bourgeois discourse of the
bourgeoisie is tautological: the unconscioU5 of bourgeois dis
course is indeed the Other, but that Other is a different bourgeois
discourse.

Writing as value
Evaluation precedes criticism. There is no calling into crisis
without evaluating. Our value is writing. This stubborn refer
ence, beyond the fact that it must often irritate, seems in some
people's eyes to involve a risk: the risk of developing a certain
mJstiq1u . The reproach is mischievous, for it inverts point by
point the bearing we attribute to writing: that of being, in this
little canton of our Western world, the materialist fuld par el
lmce. Though deriving from Marxism and from psychoanalysis,
the theory of writing tries to shift-without breaking with-its
place of origin; on one hand, it rejects the temptation of the
signified, i.e., deafness to language, to the return and redun
dancy of it! effects; on the other, it is opposed to speech in that
speech is not transferential and escapes-partially at least, in
very narrow, even particularist social 1imits-the traps of "dia
logue"; in this theory of writing is the germ of a mass gesture;
against all discourses (speeches, inauthentic writings, rituals,
protocols, social symbolics), only the theory of writing, today,
even in the form of a luxury, makes language into something

GJPYnghtcd maknal
To the Seminar 333
and its members. Even in this sense, however, this relation is
not certain: I do not say what I know, I set forth what I am
doing; I am not draped in the interminable discourse of absolute
knowledge, I am not lurking in the terrifying silence of the
Examiner (every teacher-and this s
i the vice of the system-is
a potential examiner); I am neither a sacred (consecrated) subject
nor a buddy, only a manager, an operator, a regulator: the one
who gives rules, protocols, not laws. My role (if I have one) is
to dear the stage on which horizontal transferences will be
established: what matters, in such a seminar (the site of its
success), s
i not the relation of the members to the director but
the relation of the members to each other. That s
i what muS[
be said (as I have learned by dim of discovering the discomfort
of overcrowded groups, where each member complained of
knowing nobody else): the famous "teaching relation" is not the
relation of teacher to taught, but the relation of those taught to
each other. The space of the seminar is not Oedipal but
novtlistic (an offshoot of the novel:
Phalansterial, i.e., in a sense,
in Fourier's work, hannonia1t discourse ends in snatches of a novel:
this is the amorous NroJ W01'ld); the novelistic is neither the false
nor the sentimental; it is merely the circulatory space of subtle,
ftexible desires; within the very artifice of a "sociality" whose
opacity is miraculously reduced, it is the web of amorous
rdations.
The third space is textual: either because the seminar assumes
responsibility for producing a text, for writing a book (by a
montage of writings); or because, on the comrary, it regards
its own-non-functional-practice as already constituting a
text: the rarest text, one which does not appear in writing. A
certain way of being together can fulfill the inscription of
significance: there are writers sans book (I know some), there
are texts which are not products but practices; it might
even be said that the "glorious" text will someday be a pure
practice.
Of these three spaces, none is judged (disparaged, praised),
none prevails over its neighbors. Each space is, in its turn, the

C JPYnghied mater1
 Environs of tlu Image
the beginning of a dislocation-separate discourse from the
body: precisely those three-tenths whose fall defines style, ac
cording to the actor Zeami Oapan, fourteenth century): "Move
your mind to the ten-tenths, move your body to the seven
tenths."

Giddiness
The etymology of giddiness is uncertain, but it seems to derive
from the presence of the god in wine. We should not be surprised,
then, if the seminar is somewhat "giddy," too: displaced beyond
meaning, beyond the Law, abandoned to a certain minor
euphoria. ideas being generated as though by chance, indirectly,
from a flexible listening, from a sort of swing of the attention
(they want to "speak up." hut it is "listening up" which intoxicates,
displaces, subver15; it is in listening that the Law's defect is to
be found).
In the seminar, there is nothing to represent, to imitate;
"grades," massive instrument of record, are out of place here;
what is recorded, at an unpredictable rhythm, is only whatever
traverses our listening, what is generated by a "giddy" listening.
"Grades" are detached from knowledge as a model (a thing to
copy); what is recorded is writing. not memory; grades are in
production, not in representation (showing tlu grade).

Practlces
Let us imagine or remember---three educational practices.
The first is teaching. A (previous) knowledge is transmitted by
oral or written discourse, swathed in the flux of statements
(books, manuals, lectures).
The second s
i apprernhc ship. The "master" (no connotation of
authority: instead, the reference is Oriental) works for himself in
the apprentice's presence; he does not speak, or at least he
sustains no discourse; his remarks are purely deictic: "Here,"
he says, "I do this in order to avoid that . . ." A proficiency is

C JPYnghied mater1
To the Seminar 337
transmitted in silence. a spectacle is put on (that of praxis), to
which the apprentice, taking the stage, is gradually introduced.
The third is mothering. When the child learns to walk, the
mother neither speaks nor demonstrates; she does not teach
walking, she does not represent it (she does not walk before the
child): she supports, encourages, calls (steps back and calls); she
incites and surrounds: the child demands the mother and the
mother desires the child's walking . . .
I n the seminar (and this is its definition), all teaching is
foreclosed: no knowledge is transmitted (but a knowledge can
be created), no discourse is sustained (but a text is sought):
teaching is disappointed. Either someone works. seeks. produces.
gathers, writes in the othen' presence; or else all incite each
other, call to each other, put into circulation the object to be
produced, the procedure to compose, which thus passes fTOm
hand to hand. suspended from the thread of desire like the
ring in round games.

The chain
At either extremity of the metaphor, two images of the chain:
one, abhorred, suggests the assembly line; the other, voluptuous,
refen to the Sadean figure, the ring of pleasure. I n the alienated
chain, objects are transformed (an automobile motor), Jubjects re
peated: the subject'S repetition is the price of the merchandise
produced. In the chain of gratification, of knowledge, the object
is indifferent, but the subjects "pass."
Such would be, more or less. the movement of the seminar:
to pass fTOm one chain to the other. Along the first (classical,
institutional) chain, knowledge is constituted. increased. assumes
the form of a specialty, i.e . of merchandise. while the subjects
persist, each in his place (in the place of his origin, of his
capacity, of his labor); but along the other chain, the object (the
theme, the question), being indirect or insignificant or aban
doned, in any case severed from knowledge, is the stake of no
pursuit, of no market: non-functional, perverse, it is never

GJPYnghtcd maknal
340 Environs of the lmagl!
with them a book in proce$$; let us show ourselves in the speech
act.

The man or statements

The Father (let us continue to 5culate a little upon this
i the Speaker: he who sustains
principle of intelligibility) s
discourse(s) outside of praxis, severed from all production; the
Father is the Man of statements. Hence, nothing is more
transgressive than to surprise the Father in the speech-act; this
is to surprise him in intoxication, in gratification, in erection:
an intolerable (perhaps sacred. in the sense Balaille gave this
word) spectacle, which one of the sons attempts to conceal-for
otherwise Noah would lose his paternity.
The one who shows, the one who states, the one who shows
the statement. is no longer the Father.

To teach

To teach whlJ/. occurs only once-what a contradiction in terms! Is

not to teach, invariably, to repeat?
Yet this is what old Michelet believed he had done: "I have
always been careful to teach only what I did not know . . . I
have transmitted these things as they were then, in my passion
new, lively, blazing (and delightful for me), under love's first
spell."

Guelf I Ghlbelline
This same Michelet opposed Guelf to Ghibelline. The Guelf is
the man of the Law, the man of the Code, the Legist, the Scribe,
the Jacobin. the Frenchman (shall we add the Intellectual?).
The Ghibelline is the man of the feudal link. of the oath sworn
in blood. the man of affective devotion, the German (and also
Dante). If we could extend this great symbolics to such minor
phenomena. we might say that the seminar has a Ghibelline

C JPYnghied mater1
34' EntJ1T07U of the Image
our lives only as comdy, farce. masquerade: culture is accept
able, one might say. only in the second degree-no longer as a
direct value but as an inverted one: kitsch, plagiarism, game,
pleasure, shimmer of a parody-language in which we believe and
do not believe (the characteristic of farce), a fragment of pastiche;
we are condemned to the anthology, short of rehearsing a moral
philosophy of totality.

To the seminar
To the stminar: this expression must be understood as a locative,
as an encomium (like the one the poet von Schober and the
composer Schubert addressed "An du Musik"), and as a dedi
cation.

L'Are, 1974

C ;.pvnghted malaria
 The Indictment
Periodically Lodged . . .

The indictment periodically lodged against inlellecttuds (since

the Dreyfus Affair. which saw, I believe, the birth of the won:l
and of me notion) is an indicunent of magic: the intelltttual is
treated as a witch doctor might be by a tribe of dealers,
businessmen. and jurUu: he is the subject who upsets ideological
interests. Anti-intellectualism is a historical myth. linked no
doubt to the rise oCthe petite-bourgeoisie. Not long ago, Poujade
gavc' this myth iu crudest form ("the fish rots from the head
down"). Such an indictment can periodically excite the gallery,
like any witch trials; yet its political danger must not be over
looked: it is quite simply fascism. whose first objective always
and everywhere is to liquidate the intellectual class.
The intellectual's tasks are defined by these very resistances,
the site from which they emanate; Brecht formulated them on
several occasions: to decompose bourgeois (and petit-bourgeois)
ideology, to study the forces which change the world and advance
theory. These formulas must of course include many practices
of writing and language (since the intellectual is assumed as a
being of language, and since language specifically jeopardizes
the assurance of a world which arrogandy sets "realities" against
"words," as if language were merely the futile decor of human
ity'S more substantial interests).
The intellectual's historical situation is anything but comfon
able. not because of the absurd indictments lodged against him.
but because it is a dialectical situation: the intellectual's function
is to criticize bourgeois language within the bourgeoisie's very
regnum; he must be both an analyst and a utopian, must

343

C JPYnghied mater1
 Environs of the Image
to a true metonymy, the darkness of the theater is prefigured
by the "twilight reverie" (a prerequisite for hypnosis, according
to Breuer-Freud) which precedes it and leads him from street
to street, from poster to poster, finally burying himself in a dim,
anonymous, indifferent cube where that festival of affects known
as a film will be presented.
What does the "darkness" of the cinema mean? (Whenever I
hear the word cinema, I can't help thinking haU, rather than
film.) Not only is the dark the very substance of reverie (in the
pre-hypnoid meaning of the term); it is also the "color" of a
diffused eroticism: by its human condensation, by its absence of
worldliness (contrary to the cultural appearance that has to be
put in at any "legitimate theater"), by the relaxation of postures
(how many members of the cinema audience slide down into
their seats as if into a bed , coats or feet thrown over the row in
front!), the movie house (ordinary model) is a site of availability
(even more than cruising), the inoccupation of bodies, which
best defines modern eroticism-not that of advertising or strip
tease, but that of the big city. It is in this urban dark that the
body's freedom is generated; this invisible work of possible
affects emerges from a veritable cinematographic cocoon; the
movie spectator could easily appropriate the silkworm's motto:
Inclusum labor illwtrat; it is because I am enclosed that I work
and glow with all my desire.
In this darkness of the cinema (anonymous, populated, nu
merous---oh. the boredom, the frustration of so-called private
showings!) lies the very fascination of the film (any film). Think
of the contrary experience: on television, where films are also
shown, no fascination; here darkness is erased, anonymity
repressed; space is familiar. articulated (by furniture, known
objects), tamed: the eroticism-no, to put it beuer, to get across
the particular kind of lightness, of unfulfillment we mean: the
tTOticiwti01l of the place is foreclosed: television doomed us to
the Family, whose household instrument it has become-what
the hearth used to be, Aanked by its communal kettle.

C JPYnghied mater1
uaving 1M MoW Tlualtr 347
In that opaque cube, one light: the film, the screen? Yes, of
course. But also (especially?), visible and unperceived, that
dancing cone which pierces the darkness like a laser beam. This
beam is minted, according to the rotation of its particles, into
changing figures; we turn our face toward th currency of a
gleaming vibration whose imperious jet brushes our skull,
glancing off someone's hair, someone's face. As in the old
hypnotic experiments, we are fascinated-without seeing it head
on-by this shining site, motionless and dancing.

It's exactly as if a long stem of light had outlined a keyhole,

and then we all peered, flabbergasted, through that hole. And
nothing in this ecstasy is provided by sound, music, words?
Usually-in current productions---the audio protocol can pro
duce nofascinaud listening; conceived to reinforce the liftl$$'
of the anecdote. sound is merely a supplementary instrument
of representation; it is meant to integrate itself unobtrusively
into the object shown, it is in no way detached from this object;
yet it would take very little in order [0 separate this sound track:
one displaced or magnified sound, the grain of a voice milled
in our eardrums, and the fascination begins again; for it never
comes except from artifice, or better still: from the artifact-like
the dancing beam of the projector-which comes from overhead
or to the side. blurring the scene shown by the screen Jd wilhouJ
distcnting its imagt (its gtstalt, its meaning).

For such is the narrow range-at least for me-in which can
function the fascination of film, the cinematographic hypnosis:
I must be in the story (there must be verisimilitude), but I must
also be tlsewhtTe: a slightly disengaged image-repertoire, that s
i
what I must have-like a scrupulous, conscientious, organized,
in a word diffICUlt fetishist, that is what I require of the film and
of the situation in which I go looking for it.

The film image (including the sound) is what? A lure. 1 am

confined with the image as if I were held in that famous dual

C JPYnghied mater1
 Environs of the Image
relation which establishes the imagerepertoire. The image is
there, in front of me, for me: coalescent (its signified and its
signifier melted together), analogical, total, pregnant; it is a
perfect lure: I fling myself upon it like an animal upon the
scrap of "lifelike" rag held out to him; and, of course, it sustains
in me the misreading attached to Ego and to imagerepertoire.
I n the movie theater, however far away I am sitting, I press my
nose against the screen's mirror, against that "other" image
repertoire with which I narcissistically identify myself (it is said
that the spectators who choose to sit as close to the screen as
possible are children and movie buffs); the image captivates me,
captures me: I am gllUd to the representation, and it is this glue
which established the naturalmss (the pseuda.nature) of the
filmed scene (a glue prepared with all the ingredients of
"technique"); the Real knows only distances, the Symbolic knows
only masks; the image alone (the imagerepenoire) is clost, only
the image is "trut" (can produce the resonance oftruth). Actually,
has not the image, statutorily, all the characteristics of the
ideological? The historical subject, like the cinema spectator I am
imagining. is also gllUd to ideological discourse: he experiences
its coalescence, its analogical security, its naturalness, its "truth":
it is a lure (QUr lure, for who escapes it?); the Ideological would
actually be the imagerepertoire of a period of history, the
Cinema of a society; like the film which lures its clientele, it
even has its photograms; is not the stereotype a fixed image, a
quotation to which our language is glued? And in the common
place have we not a dual relation: narcissistic and maternal?

How to come unglued from the mirror? I'll risk a pun to

answer: by taAing off (in the aeronautical and narcotic sense of
the term). Of course, it is still possible to conceive of an art
which will break the dual circle, the fascination of film, and
loosen the glue, the hypnosis of the lifelike (of the analogical),
by some recourse to the spectator's critical vision (or listening);
is this not what the Brechtian alienation-effect involves? Many
things can help us to "come out of' (imaginary and/or ideolog-

GJPYnghtcd maknal
Leaving the MoW Theater 349
ical) hypnosis: the very methods of an epic art, the spectator's
culture or his ideological vigilance; contrary to classical hysteria,
the image-repertoire vanishes once one observes that it exists.
But there is another way of going to the movies (besides being
armed by the discourse of counter-ideology); by letting oneself
be fascinated twice DVn', by the image and by its surroundings
as if I had two bodies at the same time: a narcissistic body which
gazes, lost, into the engulfing mirror, and a perverse body,
ready to fetishize not the image but precisely what exceeds it:
the textu of the sound, the hall, the darkness, the obscure
mass of the other bodies, the rays of light, entering the theater,
leaving the hall; in short, in order to distance, in order to "take
off," 1 complicate a "relation" by a "situation." What I use to
distance myself from the image-that. ultimately. is what fasci
nates me: I am hypnotized by a distance; and this distance is
not critical (intellectual); it is, one might say, an amorous
distance: would there be, in the cinema itself (and taking the
word at its etymological suggestion) a possible bliss of discretion?

Communications, 1975

GJPYnghtcd maknal
 The Image

II so happens that this text, prepared in haste a few days ago.

will sttm to copy what has been said subsequently and what you
may recognize in passing. It is a review of persistent themes.
put in a cenain perspective: the perspective of my reality insofar
as it is unreaL

At the origin of everything, Fear. (Of what? Of blows, of

humiliations?) Parody of the Cogito as the fictive moment when,
everything having been "razed," this tabula rasa will be reoccu
pied: "I'm afraid, therefore I'm alive." An observation: Accord
ing to today's moTts (we need an ethology of intellectua1s), one
never speaks of fear: it is foreclosed from discourse, and even
from writing (could there be a writing of fear?). Posited at the
origin, it has a value as metlwd; from it leads an initiatic path.

In Creek, Macke means: combat, battle single combat, duel,

struggle in a contest. "Ludism" of the conflict, of the joust: I
loathe this. The French seem to like it: rugby, "face to face."
round tables. bet.s. always stupid. etc. There used to be a more
penetrating meaning: "contradiction in terms"; i.e logical trap.
.

double bind. origin of psychoses. The logical antonym of Mache

i Acowwhia. natural succession. consequence. exclusive of con
s
flict; this word also has another meaning. to which we shall
return.
Language is the field of Mache: pugna IJtrOOrum. There is a
whole dossier to be collected-a book to be written: that of the
governed contestations of language; they are always regulated:
in language. nothing is ever wild. primitive, everything is coded.
even and especially trials of strength: Sophistics. Disputatio. Hain-

C JPYnghied mater1
352 Environs of the Image
I the one who is limited, inept, am I the one who doesn't
understand?
Confronted with a text 1 cannot read. I am, literally, "bewil
dered"; a vertigo occurs. a disturbance of the semicircular canals:
all the "otoliths" fall on just one side; in my hearing (my
reading). the signifying mass of the text collapses, is no longer
ventilated, balanced by a cultural action.
The status of "unreadabiHty" is "scientifically" (linguistically)
inapprehensible, unless we can refer to certain norms, but these
norms are variable and vague. Which inexorably brings us back
to a situation of language (language in we); linguistics is well aware
that it must now deal with this, otherwise it will perish; but this
means dealing with the whole surface of the world, of the
subject. Unreadability is a kind of Trojan horse in the fortress
of the human sciences.
Yet little by little I recognize in myself a growing desire for
readability. I want the texts I receive to be "readable," I want
the texts I write to "readable," too. How? By work on the
Sentence, on Syntax; I accept the "thetic" (linked to the Sentence
by Julia Kristeva, apropos of the Olophrase), or else I fake it by
means other than syntax. A "well-made" sentence (according to
a classical mode) is clear; it can tend toward a certain obscurity
by a certain use of ellipsis: ellipses must be restrained; metaphors.
too; a continuously metaphorical writing exhausts me.
A preposterous notion occurs to me (preposterous by dint of
humanism): "We shall never be able to say how much love (for
the other, the reader) there is in warA: on the sentence." Charity
of the Thetic. Agape of the syntax? In negative theology. Agape
is steeped in Eros; hence, eroticism of the "readable" Sentence?

I return to the intimidations of language-to language as

Combat, as Mache. A metaphor occurs to me: that of the leech.
J am thinking of complete. systematic languages. the language
of the subjects who have a faith, a certainty. a conviction. and
for me this is a permanent enigma: How can a body sticlt to an
idea-or an idea to a body? There are leech-languages. whose

C ;.pvnghted malaria
354 Environs of the Image
the Revolution has occurred, it has not been able to "change
language," The rejection of language's intimidations therefore
consists. modestly, in drifting within familiar words (without too
much concern if they are outdated); for example: Tolerance,
Ikmocracy, Contract.
Tolerance: we must rework the notion, define a new Tolerance,
since there is a new Intolerance (it would instructive to make
a map of today's world in terms of these new Intolerances).
Democracy: a word saturated with disillusions. to the point of
disgust. sometimes of violence; the lures of bourgeois democracy
have been abundantly demystified. Perhaps we must, nonethe
less, not throw out the baby with the bathwater. I'd like a theory
of History'S "layers of experience": the bourgeoisie is like the
caMh, made up of several layers, some good, others bad; we
must sort outstablish a differential geology. Then. too, we
can have a diffrcuU notion of Democracy: can define it, not as
the realization of a stifling "gregarity," but as "what should
produce aris[()cratic souls" (says one commentator on Spinoza).
ContTact: this word has gathered a whole socia-political dossier,
and a psychoanalytic one as well; let us leave it aside, defining
Contract a minima. as a casual apparatus which keeps the other
(and therefore, conversely, myself) from imprisoning me in the
pincers of a double term: to be either a swine (if I must answer
his blows, his Will-la-power) or else a saint (if I must answer his
generosity); actually, the Contract has this virtue: to dispense
anyone from being a Devil or a Hero (Brecht: "Woe to the
country which needs heroes").

All this implies that in my eyes there is a cowardice, a triviality

in conflicts. and even. I often feel. a kind of "Gallic asininity"
in these wills-ta-conflict, the puerile craving to "start something."
This impression of mediocrity takes the form of an aphorism:
Who wants to be violent has a paltry notion of violence. For
me, the real violence is that of "Everything passes," of ruin, of
oblivion. The violence of effacement is stronger than that of
fracture; death is violent; not so much the death one inBicts,
that one wants to inBict. as the death that comts all by itself (a

C :lpynghled malenal
The Image 355

violence which perhaps cannot be understood before middle

age).

Combat of language-systems: metaphor of the leech. Now let

us return to the combat of Images ("image": what I believe the
other thinks of me); how does an image of myself "take," to the
point where I am wounded by it? Here is a new metaphor: "In
the skillet, the oil spreads, smooth, Hat, matte (barely any smoke):
a kind of materia prima. Drop a slice of potato into it: it is like a
morsel tossed to wild beasts only half asleep, waiting. They all
fling themselves upon it. attack it noisily: a voracious banquet.
The slice of potato is surrounded-not destroyed, but hardened,
caramelized, made crisp; it becomes an object: a French-fried
potato." This is how. on any object, a good language-system
fu.nctions, attacks, surrounds, sizzles, haroens, and browns. All
languages are micro-systems of ebullition, of frying. That is the
stake of linguistic Mache. The language (of others) transforms
me into an image. as the raw slice of potato is transformed into
a pommefriU.

Here is how I become an image (a "French fried") under the

offensive of a quite minor language-system: the dandiacal and
"impertinent" Parisianism apropos of A Lovers Discourse: "Deli
cious essayist, favorite of intelligent adolescents. coliector of
avant-gardes, Roland Sanhes offers us memories which are
anything but. in the tone of the most brilliant salon conversation,
but with a touch of narrow pedantry apropos of 'ravishment.'
Recognizable, among others: Nietzsche, Freud, F1aubert." Noth
ing for it, 1 must pass through the Image; the image is a kind
of social military service: I cannot get myself exempted from it;
I cannot get myself discharged, cahnot desert, etc. 1 see man
sick of Images, man sick of his Image. To know one's Image
becomes a desperate, exhausting seaJ'"Ch (one never achieves it),
analogous to the insistence of a man who wants to know if he's
entit1ed to be jealous ("Misery of my life," Golaud says, vainly
interrogating the dying Melisande).
In order to be immortal (for the body to be immortal. not

CJPYnghied mater1
 Deliberation

I've never kept a joumal--or. rather, I've nevcr known if I

should keep one. Sometimes I begin, and then. right away, I
leave off-and yet. later on, I begin again. The impulse is faint.
intermittent, without scriollSne55, and of no doctrinal standmg
whatever. I guess I could diagnose this diary disease: an insoluble
doubt as to the value of what one writes in it.
Such doubt is insidious: it functions by a kjnd of delayed
action. Initially. when I write the (daily) entry, I experience a
certain pleasure: this i5 simple, this is easy. Don't worry about
finding sometlUng to stl'J! the raw malerial is right here. right now;
a kind of surface mine; all I have to do is bend ovcr-I don't
need to transform anything: the crude ore has its own value.
etc. Then comes the second phase. very soon after the first (for
instance, if I reread today what I wrote yesterday), and it makes
a bad impression: the text doesn't hold up, like some son of
delicate foodstuff which "turns," spoils, btcomes unappetizing
from one day to the next; note with discouragement the
artifice of "sincerity," the artistic mediocrity of the "sponta
neous"; worse still: am disgusted and irritated to find a "pose"
certainly hadn't intended: in ajoumal situation, and precisely
because it doesn't "work"-doesn't get transformed by the action
of work-J is a poseur: a matter of effect, not of intention, the
whole difficulty of literature: is here. Very soon, continuing my
reperusal. get tired of these verbless sentences ("Sleepless
night. and the third in a row," etc.) or sentences whose verb is
carelessly condensed ("Passed two girls in the Place 5t.S."
and try as will to reestablish the propriety of a complete form
(". passed . . ." "I spent a sleepless night"), the matrix of any
journal, i.e., the reduction of the verb, penists in my ear and

359

C :lpynghled malenal
goo Environs of the Image
exasperates me like a refrain. In a third phase. if I reread my
journal pages severa] months. several years after having written
them, though my doubt hasn't dissipated, I experience a certain
pleasure in rediscovering, thanks to these lines, the events they
relate. and even more, the inflections (of light. of atmosphere.
of mood) they bring back. In short, at this point, no literary
interest (save for problems of formulation. Le., of phrasing),
but a kind of narcissistic attachment (faintly narcmistic-Iet's
not exaggerate) to my doings (whose recall is inevitably ambig
UQUS, since to remember is also to acknowledge and to lose once
again what will not recur). But still, does this final indulgence,
achieved after having traversed a phase of rejection. justify
(systematically) keeping a journal? Is it worth the trouble?
I am not attempting any kind of analysis of the "Journal"
genre (there are books on the subject), but only a personaJ
deliberation, intended to afford a practical decision: Should I
keep a journal with a view to publication? Can I make the journal
into a "work"? Hence, I refer only to the functions which
immediately come to mind. For instance, Kafka kept a diary in
order to "extirpate his anxiety," or if you prefer, "to find
salvation." This motive would not be a natural one for me, or
at least not a constant one. Nor would the aims traditionally
attributed to the IntimateJournaJ; they no longer seem pertinent
to me. They are all connected to the advantages and the prestige
of "sincerity" (to express yourself, to explain yourself, to judge
yourself); but psychoanalysis, the Sartrean critique of bad faith,
and the Marxist critique of ideologies have made "confession"
futile: sincerity is merely a second-degree image-repertoire. No,
the journal's justification (as a work) can only be literary in the
absolute, even if nostalgic, sense of the word. I discern here
four motives.
The first is to present a text tinged with an individuality of
writing, with a "style" (as we used to say), with an idiolect proper
to the author (as we said more recently); let us call this motive:
poetic. The second is to scatter like dust, from day to day, the
traces of a period, mixing all dimensions and proportions, from

C :lpynghled malenal
 Environs of the Image
day that has passed ( I can give only one of these, the second
one involving others besides myself).

V , July 13, 1977

Mme , the new cleaning woman, has a diabetic grandson she takes
care of, we are told, with devotion and expertise. Her view of this disease
is confused: on the one hand, she does not admit that diabetes is hereditary
(which would be a sign of inferior stock), and on the other, she insists
that it is fatal, absolving any responsibility of origin. She posits disease
as a social image, and this image is beset with pitfalls. The Mark certainly
appears as a source ofpride and of pain: what it was for Jacob-Israel,
dislocated, disconnected by the Angel: delight and shame of being re
marked.

Depression, fear, anxiety: I see the death of a loved one, I panic, etc.
Such an imagination is the very opposite of faith. For constantly to
imagine the inevitability of disaster is constantly to accept it: to utter it
is to assert it (again, the fascism of language). By imagining death, I
discourage the miracle. In Ordet the madman did not speak, reflJ.Sed
the ga1lulous and perelnptory language of inwardness. Then what is
this incapacity for faithr Perhaps a very human love Love, then, ex
cludes faithr And v!ce versa

Gide's old age and death (which I read about in Mme van Ryssel
berghe's Cahiers de la Petite Dame) were surrounded by witnesses.
But I do not know what has become of these witnesses: no doubt, in most
cases, dead in their tum: there is a time when the witnesses themselves
die without witnesses. Thus, History consists of tiny explosions of life, of
deaths without relays. Our human impotence with regard to transition,
to any science of degrees. Conversely, we can attribute to the classical
God the capacity to see an infinity of degrees: "God" as the absolute
Exponential.
(Death, real death, is when the witness himself dies. Chateaubriand
says of his grandmother and his great-aunt: "I may be the only man in
the world who knows that such persons have existed": yes, but since he
 Environs of the Image
produces the plenitude of an evidence: that it is worthwhile being alive.
The morning el1ands (to the grocer, the baker, while the village is still
almost deserted) are something I wouldn't miss for anything in the
world.

Mother feeling better today. She is sitting in the garden, wearing a

big straw hat. As soon as she feels a little better, she is drawn by the
house, filled with the desire to participate; she puts things away, turns
off the furnace during the day (which I never do).

This afternoon, a sunny, windy day, the sun already setting, I burned
garbage at the bottom of the garden. A complete course of physics to
follow; a1.ned with a long bamboo pole, I stir the heaps ofpaper, which
slowly burn; it takes patience who would have guessed how long paper
can resist the fire r On the other hand, the emerald-green plastic bag (the
garbage bag itself) burns very fast, leaving no trace: it literally van
ishes. This phenomenon might serve, on many an occasion, as a metaphor.

Incredible incidents (read in the Sud-Ouest or heard on the radior

I don't remember): in Egypt it has been decided to execute those Moslems
who convert to another religion. In the U.S.S.R., a French agent was
expelled because she gave a present of underwear to a Soviet friend.
Compile a contemporary dictionary of intolerance (literature, in this
case Voltaire, cannot be abandoned, so long as the evils subsist to which
it bears witness).

July 1 7, 1977
As if Sunday morning intensifUls the good weather. Two heteroclite
intensities reinforce each other.

I never mind doing the cooking. I like the operations involved. I take
pleasure in observing the changingfonns of thefood as they occur (colora
tions, thickenings, contractions, crystallizations, polarizations, etc.). There
is something a little perverse about this observation. On the other hand,
what I can't do, and what I always do badly, are proportions and sched
ules: I put in too much oil, afraid everything will burn; I leave things on
Deliberation
thefire too long, afraid they won't be cooked through. In short, I'm afraid
because I don't know (how much, how long). Whence the security of a
code (a kind ofguaranteed knowledge): I'd rather cook rice than potatoes
because I know it takes seventeen minutes. Thisfigure delights me, insofar
as it's precise (to the point of being preposterous); a round number would
.
seem contrived, andjust to be certain, I'd add to it.

July 18, 1977

Mother's birthday. All I can offer her is a rosebud from the garden;
at least it's the only one, and the first one since we're here. Tonight, M.
is coming for dinner and will do the cooking: soup and a pimento
omelette; she brings champagne and almond cookies from Peyrehorade.
Mme L. has sent flowers from her garden, delivered by one of her
daughters.

Moods, in the strong, Schumannian sense: a broken series of con

tradictory impulses: waves of anxiety, imaginations of the worst, and
unseasonable euphorias. This morning, at the core of Wony, a crystal
of happiness: the weather (very fine, very light and dry), the music
(Haydn), coffee, a cigar, a good pen, the household noises (the human
subject as caprice: such discontinuity alamlS, exhausts).

July 19, 1977

Early in the morning, coming back with the milk, I stop in the church to
have a look around. It's been remodeled according to the prescribed New
Look: now it resembles nothing so much as a Protestant establishment (only
the wooden galleries indicate a Basque tradition); no image, the altar has
become a simple table, no candle, ofcourse. Too bad, isn't it

Around six in the evening, I doze on my bed. The window is wide open,
the gray day has lifted now. I experience a certain floating euphoria:
everything is liquid, aerated, potable (I drink the air, the moment, the
garden). And since I happen to be reading Suzuki, it seems to me that I'm
quite close to the state Zen calls sabi; or again (since I'm also reading
Blanchot), to the "fluid heaviness" he speaks ofapropos ofProust.

 Environs of the Image
July 2 1, 1977
Some bacon, onions, thyme, etc.: simmering, the smell is wonderful.
Now this fragrance is not that offood as it will be seived at table. There
is an odor of what is eaten and an odor of what is prepared (observation
for the "Science of Motley," or "diaplwrology").

July 22, 1977

For some years, a unique project, apparently: to explore my own stu
pidity, or better still: to utter it, to make it the object of my books. In
this way I have already uttered my "egoist" stupidity and my "lover's"
stupidity. There remains a third kind, which I'll someday have to get
down on paper: political stupidity. What I think of events politically
(and I never fail to think something), from day to day, is stupid. This

is a stupidity which I should now utter in the third book of this little
trilogy, a kind of Political Diary. It would take en01mous courage, but
maybe this would exorcise that mixture ofboredom, fear, and indignation
which the Politician (or rather Politics) constitutes for me.

"/" is harder to write than to read.

Last night, at Casino, the Anglet supeimarket, with E.M., we were

filscinated by this Babylonian Temple of Merchandise. It really is the
Golden Calf: piles of(cheap) "wealth," mustering of the species (classified
by types), Noah's ark of things (Swedish clogs to eggplants), predatory
stacking ofcarts. We're suddenly convinced that people will buy anything
(as I do myself): each cart, while parked in front of the cash register, is
the shameless chariot of manias, impulses, perversions, and cravings:
obvious, confronting a cart proudly passing before us, that there was no
need to buy the cellophane-wrapped pitW ensconced there.

I'd like to read (if such a thing exists) a History of Stores. What
happened before Zola and Au Bonheur des dames1

August 5, 1977
Continuing War and Peace, I have a violent reading the
Ikath of old Prince Bolkonsky, his last words oftenderness to his daughter
Deliberation
(Rereading: this bit gave me a distinct pleasure, so vividly did it revive
the sensations of that evening; but curiously, in reading it over, what I
remembered best was what was not wrilten, the interstices of notation:
for instance, the gray of the rue de Rivoli while I was waiting for the
bus; no use trying to describe it now, anyway, or I'll lose it again instead
of some other silenced sensation, and so on, as if resurrection always
occurred alongside the thing expressed: role of the Phantom, of the
Shadow.)

However often I reread these two fragments, nothing tells

me they are publishable; on the other hand, nothing tells me
that they are not. Which raises a problem that is beyond me-e
the problem of "publishability"; not: "is it good or is it bad?" (a
form every author gives to his question), but "is it publishable
or isn't it?" This is not only a publisher's question. The doubt
has shifted, slides from the text's quality to its image. I raise for
myself the question of the text from the Other's point of view;
the Other is not the public, here, or any particular public (that
is the publisher's question); the Other, caught up in a dual and
somehow personal relation, is anyone who will read me. In short,
I imagine that my Journal pages are put in front of "whom I
am looking at," or under the silence of "whom I am speaking
to." Is this not the situation of any text? No. The text is
anonymous, or at least produced by a kind of nom de guerre, that
of the author. This is not at all true of the Journal (even if its
"I" is a false name): the Journal is a "discourse" (a kind of written
record according to a special code), not a text. The question I
raise for myself: "Should I keep a journal?" is immediately sup
plied, in my mind, with a nasty answer: 'Who cares?" or, more
psychoanalytically: "It s your problem."
'

All I have left to do is analyze the reasons for my doubt. Why

do I suspect, from the point of view of the image, Journal writing?
I believe it is because this writing is stricken, in my eyes, with a
kind of insidious disease, negative characteristics deceptive
and disappointing, as I shall try to say.
The Journal corresponds to no mission. Nor is this word
 Environs of the Image
laughable. The works of literature, from Dante to Mallarme,
Proust, and Sartre, have always had, for those who wrote them,
a kind of social, theological, mythic, aesthetic, moral end. The
book, "architectural and premeditated," is supposed to repro
duce an order of the world; it always implies, 1 believe, a monist
philosophy. The Journal cannot achieve the status of the Book
(of the Work); it is only an Album, to adopt Mallarme's distinc

tion (it is Gide's life which is a "work," not his Journal). The
Album is a collection of leaflets not only interchangeable (even
this would be nothing) but above all infinitely suppressible: re
reading my Journal, 1 can cross out one entry after the next, to
the complete annihilation of the Album, with the excuse that "I
don't like this one": this is the method of Groucho and Chico
Marx reading aloud and tearing up each clause of the contract
meant to bind them. But can't the Journal, in fact, be
considered and practiced as that form which essentially expresses
the inessentials of the world, the world as inessential ? For
that, the Journal's subject would have to be the word, and not
me; otherwise, what is uttered is a kind of egotism which
constitutes a screen between the world and the writing; whatever
1 do, 1 become consistent, confronting the world which is not
so. How to keep a Journal without egotism? That is precisely
the question which keeps me from writing one (for I have had
just about enough egotism).
Inessential, the Journal is unnecessary as well. I cannot invest
in a Journal as 1 would in a unique and monumental work
which would be dictated to me by an incontrovertible desire.
The regular writing of the Journal, a function as daily as any
other physiological one, no doubt implies a pleasure, a comfort,
but not a passion. It is a minor mania of writing, whose necessity
vanishes in the trajectory leading from the entry produced to
the entry reread: "I haven't found that what I've written so far
is particularly valuable, nor that it obviously deserves to be
thrown away" (Kafka). Like any subject of perversion (I am
told), subjected to the "yes, but . . . " I know that my text is
futile, but at the same time (by the same impulse) I cannot wrest
myself from the belief that it exists.
Deliberation 37 1
Inessential, uncertain, the Journal is also inauthentic. I don't
mean by this that someone who expresses himself in one is not
sincere. I mean that its very form can only be borrowed from
an antecedent and motionless Form (that, precisely, of the
Intimate Journal), which cannot be subverted. Writing my
Journal, I am, by status, doomed to simulation. A double
simulation, in fact: for since every emotion is a copy of the same
emotion one has read somewhere, to report a mood in the
coded language of the Collection of Moods is to copy a copy:
even if the text was "original," it would already be a copy; all
the more so if it is familiar, worn, threadbare: "The writer, by
his pains, those dragons he has fondled, or by a certain vivacity,
must set himself up, in the text, as a witty histrion" (Mallarme).
What a paradox! By choosing the most "direct," the most
"spontaneous" form of writing, I find myself to be the clumsiest
of ham actors. (And why not? Are there not "historic" moments
when one must be a ham actor? By practicing an antiquated
form of literature to the bitter end, am I not saying that I love
literature, that I love it in a harrowing fashion, at the very
moment when it's dying? I love it, therefore I imitate it but
precisely: not without complexes.)
All of which says more or less the same thing: that the worst
torment, when I try to keep a Journal, is the instability of my
judgment. Instability? Rather, its inexorably descending curve.
In the Journal, as Kafka pointed out, the absence of a notation's
value is always recognized too late. How to transform what is
wrftten at white heat (and take pride in the fact) into a nice cold
dish? It is this waste, this dwindling which constitute's the
Journal's uneasiness. Again, Mallarme (who, moreover, didn't
keep one): "Or other verbiage become just that, provided it is
exposed, persuasive, pensive, and true when one confides it in
a whisper": as in that fairy tale, under the effect of a curse and
an evil power, the flowers that fall from my lips are changed
into toads. "When I say something, this thing immediately and
definitively loses its importance. When I write it here, it also
loses it, but sometimes gains another importance" (Kafka). The
difficulty proper to the Journal is that this secondary importance,
37 2 Environs of the Image

liberated by writing, is not certain: it is not certain that the

Journal recuperates the word and gives it the resistance of a
new metal. Of course, writing is indeed that strange activity
(over which, hitherto, psychoanalysis has had little hold, under
standing it with difficulty) which miraculously arrests the hem
orrhaging of the image-repertoire, of which speech is the
powerful and pathetic stream. But precisely: however "well
written," is the Journal "writing"? It struggles, swells, and
stiffens: am I as big as the text? Never! you aren't even close.
Whence the depressive effect: acceptable when I write, disap
pointing when I reread.
At bottom, all these failures and weaknesses designate quite
clearly a certain defect of the subject. This defect is existential.
What the Journal posits is not the tragic question, the Madman's
question: "Who am I?", but the comic question, the Bewildered
Man's question: "Am I?" A comic a comedian, that's what the
Journal keeper is.
In other words, I never get away from myself. And if I never
get away from myself, if I cannot manage to determine what
the Journal is "worth," it is because its literary status slips
through my fingers: on the one hand, I experience it, through
its facility and its desuetude, as being nothing more than the
Text's limbo, its unconstituted, unevolved, and immature form;
but on the other hand, it is all the same a true scrap of that
Text, for it includes its essential torment. This torment, l believe,
consists in this: that literature is without proofs. By which it must
be understood that it cannot prove not only what it says but
even that it is worth the trouble of saying it. This harsh condition
(Play and Despair, Kafka says) achieves its very paroxysm in the
Journal. But also, at this point, everything turns around, for
out of its impotence to prove, which excludes it from the serene
heaven of Logic, the Text draws a flexibility which is in a sense
its essence, which it possesses as something all its own. Kafkal
whose Journal is perhaps the only one that can be read without
irritation expresses this double postulation of literature to
perfection: Accuracy and Inanity: " . . . I was considering the
Deliberation 3 73
hopes 1 had formed for life. The one which appeared the most
important, or the most affecting, was the desire to acquire a way
of seeing life (and, what was related, of being able, by writing,
to convince others) in which life would keep its heavy movement
of rise and fall , but would at the same time be recognized, and
with a no less admirable clarity, as a nothing, a dream, a drifting
state." Yes, that is just what the ideal Journal is: at once a
rhythm (rise and fall, elasticity) and a trap (I cannot join my
image): a writing, in short, which tells the truth of the trap and
guarantees this truth by the most formal of operations, rhythm.
On which we must doubtless conclude that 1 can rescue the
Journal on the one condition that 1 labor it to death, to the end
of an extreme exhaustion, like a virtually impossible Text: a
labor at whose end it is indeed possible that the Journal thus
kept no longer resembles a Journal at all.

Tel (bul, 1979



LITERATURE I UNGUlsnCS I PHILOSOPHY

"Barthes's career was an exemplary search for understanding

how man creates meaning, a lifelong exploration of man's
definition as homo sigl'lifict1.ns, the maker of meaning in signs.
In anthropology, in linguistics, philosophy, and the discourse
upon literature, this has been a characteristic preoccupation of
our age, and no one addressed himself to it so persistently, so
multifariously, so ingeniously, as Bahes."
-Peter Brooks, The New Republic

"The Rustle of Language collects forty-five of Barthes's essays

on literature and teaching written between 1967 and 1980. It
includes hommage to various authors, from Brecht to Proust,
and to various scholars, from Jakobson to Kristeva. There are
central Barthes statements about the future of semiological
investigation, urging that it go beyond exploring the class
interests underlying discourse, and question discourse making
itself. . . . The typically Barthesian texture of the writing makes
itself felt. That texture delightful to many of us-is composed
of the mutual jostling of many (often mutually incompatible)
registers of discourse. Linguistics, literature, philosophy, . . .
history, semantics, Marxism-these are only the commonest of
the many categories that organize Barthes's thinking. . . . In all of
these essays, the briskness and liveliness of Barthes's style makes
the work interesting."
-Helen Vendler, New York Review of Books

Roland Barthes studied French literature and classics at the University

of Paris, taught French at universities in Rumania and Egypt and
sociology at the Ecole Pratique des Hautes Etudes in Paris. He was a
professor at the College de France until his death in 1980.

Cowr ,usig.. by OuIrks Suggs

UNIVERSITY OF CALIFORNIA PRESS BERKELEY 94720