Foucault. Structuralism and Literary Analysis

Structuralism and Literary Analysis
Michel Foucault
Translated by Suzanne Taylor and Jonathan Schroeder
Ladies and Gentlemen,

I believe we’re here mainly to have a discussion, which means that I
shouldn’t be talking at all; but, after all, I suppose that in order for you
to be able to exercise your right to pose questions, which will be a right
to examine and a right to critique, it’s necessary that I expose myself to
your blows. So I’m going to present some slightly untidy remarks, on
the basis of which I hope you yourselves will have an occasion to express
your opinions.
I’ve chosen my subject basically without knowing beforehand to whom
I’d be speaking—luckily, as it happens, because if I had known, I think I’d
have given up speaking altogether! On the one hand, the audience is com-
prised of extremely intimidating people who are my colleagues and who,
consequently, know more than I do. On the other, there are many students
who know me already and who have already seen me go through my paces.
So all this is a little intimidating and embarrassing for me. Anyway, without
knowing too much about my audience, I thought that I could speak about
the problem of the relations between structuralism and literary analysis.
Some months after the publication of The Order of Things (1966), Michel Foucault ob-
tained a leave from the University of Clermont-Ferrand to go teach philosophy at the Univer-
sity of Tunis. He moved to Sidi Bou Saïd in the autumn of 1966, and he would not leave Tu-
nisia until two years later, in October 1968. The “Structuralism and Literary Analysis” talk
was given at Club Tahar Haddad in Tunis on 4 February 1967. The present transcription was
made from the recording preserved in the Bancroft Library at the University of California,
Berkeley (Reel 81–A, B, and C). The original French text was established by Henri-Paul
Fruchaud and Daniele Lorenzini. We would also like to thank Richard Neer and Lorenzini
with their help with this translation.—Trans.
Critical Inquiry 45 (Winter 2019)

© 2019 by The University of Chicago. 00093-1896/19/4502-0006$10.00. All rights reserved.
531
Critical Inquiry 2019.45:531-544.
Downloaded from www.journals.uchicago.edu by Tulane University on 12/18/18. For personal use only.
532 Michel Foucault / Structuralism and Literary Analysis
As you are perfectly aware, I obviously have no competence to speak
about the problem of the relations between structuralism and literary anal-
ysis. In fact, if I’ve chosen this subject, it’s mainly because it’s currently the
nest, the site, of a number of ambiguities. You all know, from having heard
echoes at least, about the debate around what is called la nouvelle critique.
Hiding underneath this debate, I believe, are a number of concepts that are,
ultimately, rather ill-defined. And it’s with a search for definitions in mind
that I’d like to orient what must be the crux of this meeting, which is to say,
the earlier debate.
In general, I think we can say the following. This debate has been drag-
ging on for several years now not just in France but in other countries as
well, and the impression is that it opposes a number of things and peo-
ple—for example, a critique of, shall we say, a scientific type to one of an
impressionistic type. There’s also the impression that it opposes the sup-
porters of content and meaning to the supporters of pure form—and
the further impression that it opposes historians to those who are only in-
terested in the system and the synchrony of works. Lastly, there’s the im-
pression that, in the end, it’s also a conflict between persons or even social
groups: on one side, the supporters of an old and outdated French univer-
sity; on the other, the supporters of, well, a sort of intellectual renewal that
has to come from outside the university.
I’m not sure that this manner of characterizing the debate is completely
accurate. It isn’t true that the most reactionary participants in the debate
around the nouvelle critique necessarily come from within the university—
and the university, which doesn’t always have many reasons to be proud
of itself, can boast that it does not count among its members a number of
people who are the supporters of precisely this old-style critique. Nor is it
true that analyses like those of Jean-Pierre Richard entirely ignore the mean-
ing of a work in order to speak only of its content.1 It isn’t true that the cur-
rent tendencies in literary analysis deny history in favor of pure system and
synchrony. Therefore I don’t believe that all these assessments and analyses
allow us to situate the debate correctly.
1. See Jean-Pierre Richard, L’univers imaginaire de Mallarmé (Paris, 1961).
Mic h el Fou ca ul t , acknowledged as the preeminent philosopher of France

in the seventies and eighties, continues to have enormous impact throughout
the world in many disciplines. He died in 1984. S u z a n n e T ay l o r is a visiting
assistant professor at Hamilton College. J o na t h an S ch r o e d er is an assis-
tant professor at the University of Warwick.

Critical Inquiry / Winter 2019 533
To try to do better, I’d like to introduce a notion that’s now completely
familiar and one that at first glance should actually suggest many more
problems than solutions: the notion of structuralism. In general, we can
say that the debate currently turns upon the possibility, the right, the fruit-
fulness of what we call the structuralist method. But what is structuralism?
It’s extremely difficult to define it when we consider that under this word
we designate analyses, methods, works, and individuals as different as, for
example, the history of religions as done by Dumézil, the analysis of my-
thologies by Lévi-Strauss, the analysis of the tragedies of Racine by Barthes,
the analysis as well of literary works as it’s currently done in America with
Northrop Frye, the analyses of folktales that Russians like Propp have
done, the analyses of philosophical systems like those of Guéroult.2 All
of that gets placed under the structuralist label—so it’s perhaps a bit risky
to try to illuminate all these problems with such a confused notion.
Still, I’d like to linger for a bit on structuralism. Clearly, it’s not a phi-
losophy, and what’s more it can be linked to philosophies that are com-
pletely different from one another. Lévi-Strauss has explicitly linked his
structural method to a philosophy of, shall we say, a materialist type.
Someone like Guéroult, on the contrary, has linked his own method of
structuralist analysis to what we can roughly call an idealist philosophy.
Someone like Althusser explicitly uses the concepts of structural analysis
within a philosophy that’s explicitly Marxist. Thus I don’t believe that
we can establish an unequivocal and determinate relation between struc-
turalism and philosophy.
You’ll tell me that all of this is well known, that we know that structur-
alism isn’t a philosophy but a method. It’s exactly here that I’d like to make
an objection. It doesn’t seem to me, in the end, that we can really define
structuralism as a method. First of all, it’s very difficult to see in what
2. See Georges Dumézil, Les dieux des Indo-Européens (Paris, 1952), Gods of the Ancient
Northmen, ed. Einar Haugen (1959; Berkeley, 1973), and La religion romaine archaïque (Paris,
1966). See Claude Lévi-Strauss, The Raw and the Cooked, vol. 1 of Mythologiques, trans. John
and Doreen Weightman (1964; Chicago, 1983) and Structural Anthropology, trans. Claire
Jacobson and Brooke Grundfest Schoepf (1958; New York, 1963). See Roland Barthes, On Ra-
cine, trans. Richard Howard (1963; New York, 1964). See Northrop Frye, Anatomy of Criticism
(Princeton, N.J., 1957) and A Natural Perspective: The Development of Shakespearean Comedy
and Romance (New York, 1965). See Vladimir Propp, Morphology of the Folktale, trans.
Laurence Scott (1928; Austin, Tex., 1971), and Les racines historiques du conte Merveilleux,
trans. Lise Gruel-Apert (1946; Paris, 1983). Martial Guéroult (1891–1976), professor in the Col-
lège de France from 1951 to 1962, had given his chair the name “History and Technology of
Philosophical Systems.” During his life, he devoted systematic studies to, among others, René
Descartes, Baruch Spinoza, Nicolas Malebranche, Gottfried Wilhelm Leibniz, George Berke-
ley, and Johann Gottlieb Fichte.

way the method of analysis of folktales by Propp can resemble the method
of analysis of philosophical systems by Guéroult—in what way the analysis
of literary genres by Frye in America can resemble the analysis of myths by
Lévi-Strauss.
In fact, it seems to me that the word structuralism really designates a set
of disciplines—perhaps not even disciplines, but preoccupations, a num-
ber of analyses that share, at bottom, an object. And so, paradoxically
enough, I’d define structuralism and the different structuralisms by the
commonality of their object. Structuralism is currently the set of attempts
by which we try to analyze what we might call the documentary mass [masse
documentaire]: the ensemble of signs, traces, or marks that humanity has
left behind and that humanity continues, every day and in greater and
greater numbers, to form around itself. What is this mass of traces made
of—this mass of signs that have been deposited and sedimented in the his-
tory of the world, that have been entered into a universal archive that is con-
tinually being formed? It’s made of all the strictly verbal traces, all the writ-
ten traces: literature, of course, but more generally all the other things that it
has been possible to write, print, circulate, everything that has been said and
that, in one way or another, has been preserved in the memory of men,
whether that memory be psychological or the material memory of some re-
cord. Just so, it is made of all the marks that man has been able to leave
around himself: works of art, architecture, cities, and more. All of which
means that the objects that man has made conform not only to the pure
and simple laws of production but also to systems that constitute them
as marks and precisely as marks of what man himself has made.
I believe that what we’re in the process of discovering right now is the
autonomy of that aspect by which, and under which, we can analyze every-
thing that man can make. This aspect is not that of the economic produc-
tion of these objects, things, signs, marks, and so on; it’s the aspect by
which these marks and these signs are consistent with one another as
marks, as signs. It’s a matter of finding the system for determining the doc-
ument as a document. Drawing on etymology (which is not my strong
suit!) I imagine that we should be able to start from the Greek verb deik-
numi and call this discipline of the document as document something like
“deixology.” It would be the general discipline of the document as docu-
ment—which is, basically, what structuralism is currently in the process
of constituting, an analysis of the internal constraints of the document as
such. It’s from here, I think, that we can understand the seemingly diffuse
character of structuralism. In effect, structuralism deals with everything.
It deals with philosophy; it deals with advertising; it deals with cinema; it
deals with psychoanalysis; it deals with works of art; and so on.

Secondly, this explains, I think, the importance that structuralism can
hardly fail to attach to something like linguistics, insofar as linguistics is,
precisely, at the very heart of all those documents that man leaves around
himself. La langue is, after all, the most general form in which the human
document in general appears.
Thirdly, it explains, I believe, the conflicts that the method, or let us say
the concepts of structuralism—that structural descriptions—[provoke].
Such conflicts arise around structural description in disciplines that do
not study the document specifically as a document but rather as some-
thing that it was possible to produce in a system of the economic type in
the broad sense. That is to say, faced with everything that has been
sedimented in the history of humanity, we can take essentially two atti-
tudes. Either we can seek the series of processes that have permitted these
different objects created by humanity to be produced (and this is the search
for those laws of production that I’d call, broadly speaking, economics), or
we can try to study the set of remnants, the set of marks that define the ob-
jects created by humanity as, and only as, documents. This second aspect,
I think, characterizes what we could call a deixological analysis of those same
objects, over and against economics, the economic analysis of production.
The distinction between these two forms of analysis is a bit tricky, and
you’ll understand why it poses problems. But, after all, we have a model
right before our eyes. Quite simply, the natural sciences offer a model.
We’ve known for a good thirty years now that the old analysis practiced
in the nineteenth century, the analysis of processes of energy, is no longer
able to account fully for a number of phenomena: physical, chemical, and
above all biological. In addition to energy processes, it’s also necessary to
analyze what we call information processes. Nowadays, we can no longer
do biology without envisaging the perpetual interaction of the energy pro-
cesses and information processes that make all biological phenomena pos-
sible. The definition of the relations between energy processes and infor-
mation processes evidently poses many problems, but the analysis of these
relations can only proceed insofar as we have distinguished the two levels:
the energetic level and the informational level.
It seems to me that the problem is more or less the same with regard to
human phenomena. The phenomena called human must be analyzed at
two levels: at the level of their production, which is the economic level,
and at the level on which they conform to the very laws of the document
as document, that is to say, the deixological level. It’s true that it will be
necessary one day to try to study the interference between these two lev-
els—an interference that is the very substance, the very object, of history.
Still, it will only be possible to define this interference once we’ve properly

distinguished the two levels. I believe that the methodological importance,
the epistemological importance, and the philosophical importance of struc-
turalism lies precisely in this. It was initially a method, and there is no doubt
that it’s as a method that it made, as it were, the breakthrough toward this
new object, toward this stratum, toward this new epistemological domain
that I call by the arbitrary word deixology. And it’s from this methodological
breakthrough that this new object is in the process of being constituted.
From the moment this new object is constituted, inevitably, structuralism
can no longer be defined purely and simply as a method. It becomes the
pure and simple obligation to roam this new domain before us, which is
to say that structuralism has arrived at the point where it must erase itself
and disappear as a method in order to recognize—in some way folding
back on itself at the moment of its erasure—that what it has done was quite
simply to discover an object. We could compare the example of structur-
alism with that of pathological anatomy at the end of the eighteenth cen-
tury. Pathological anatomy at the end of the eighteenth century was quite
simply a medical method used by a number of doctors, and it generated
many arguments and difficulties. In the end, though, the analysis of path-
ological anatomy discovered an unforeseen object: physiology. Physiology
then developed as an autonomous discipline, using pathological anatomy
as a particular method.3 This is probably what is going to happen to struc-
turalism.
So there you have, roughly, the current situation of structuralism.
That’s what I wanted to say about this general meaning, about the general
meaning of this word.
What does literary analysis, strictly speaking, have to do with all of this?
If what I’ve told you is correct, you can see that literary analysis is neces-
sarily part of these disciplines of the document; it’s a matter of studying
those documents that we call literary works in a privileged fashion. In fact,
literary analysis, and structural literary analysis, has always held a some-
what lofty position with regard to the disciplines that I’ve been discuss-
ing—which we have, up to this point, grouped under the name of struc-
turalism. In fact, literary analysis joined this domain of the deixological
disciplines very early on. Why and how?
I think we can summarize the situation very schematically in the follow-
ing manner. Formerly, the function of literary analysis was essentially to en-
able communication, to mediate between the writing—the work proper—
and its consumption, that is, its reading by a public. Literary analysis was in
3. See Michel Foucault, The Birth of the Clinic: An Archaeology of Medical Perception, trans.
Alan M. Sheridan (1963; New York, 1973).

fact a sort of ambiguous act, midway between writing and reading, which
was supposed to allow a number of people to read a text that was written
by one person. This mediating function of literary analysis can be summed
up and described under three headings. To begin with, the function of liter-
ary criticism or analysis was to sort the written texts that ought to be read
from the ones that didn’t deserve to be read. That was how literary criticism
got rid of works like those of Sade or Lautréamont once and for all. This
was its first role. Its second role was to judge works, to tell a potential reader
ahead of time whether a work was worth anything and what it was worth in
relation to other works—thus to position it on a scale. And thirdly, literary
analysis had the role of simplifying the work, anyway of simplifying the op-
eration that reading a work involves. It had to give a sort of schema of the
production of the work itself, explaining how the author wrote, why he
wrote, what he wanted to do, and so on. These three functions—sorting,
judging, and explaining or clarifying—meant that when faced with a writ-
ten work, literary analysis took up the position, as it were, of the ideal reader.
Those who did literary analysis, practicing this absolute, imposing, and ideal
reading, wrote texts that were supposed to be a mediation for the future
reader. They were supposed to authorize, ground, and simplify the future
reader’s reading of the original text. This linear structure of (a) writing,
(b) literary analysis, (c) reading defined the role of what is specifically called
criticism. For literary analysis was criticism, in other words, a censor that
made selections, an aesthetic that offered judgments, and at the same time
a kind of historical inquiry into the production of the work, a detailed de-
scription of the reasons for which it had been produced, a reduction of
the work to those reasons. So there you have, roughly, why all literary anal-
ysis was fundamentally criticism. It’s also why there has been, in all Western-
type societies, that class of curious and formidable personages called literary
critics, the invention of which dates back roughly to Sainte-Beuve (perhaps
it was an unhappy invention, but never mind!).
I believe that over the course of the twentieth century, the position of
literary analysis changed. A completely different configuration replaced
the linear schema that I have tried to represent for you. I believe that liter-
ary analysis has now escaped the track or axis from writing to consumption
that situated it formerly. Literary analysis is no longer a relation of writing
to reading, but of writing to writing. It’s become the possibility of estab-
lishing a new language—a new language that starts from a given language
called the work—in such a way that this second language, acquired from
the first, can talk about the first. The problem of criticism, you see, is no
longer what it was before, in the nineteenth century; how can and how
should readers in general, and the ideal reader in particular, judge the work

in question? Now the issue is: what transformation must one perform on
the language of a work to make the language thus transformed speak about
the work, manifest something about the work? To the extent that this is
what criticism or literary analysis has become, you can understand how
and why it isn’t at all interested in the production of the work, in how
its birth was possible. Rather, it’s going to be interested in the work as doc-
ument, that is to say, as it is made with that form of document that we call
language; literary analysis is going to concern itself with the work insofar as
it is, fundamentally, language. That’s how literary analysis is going to be-
come like the analysis of myths, and so on—and, along with them, a sort
of deixology.
Secondly, this explains why literary analysis, to the extent that it trans-
forms a given language into a new one that must speak about that given
language, is now tied, quite closely, to the problem of linguistics. It also ex-
plains how and why it’s tied to the problem of logic, that is, to the problem
that centrally concerns the transformation of statements. Finally, you see
how and why literary analysis—no longer being that mediation between
writing and reading—cannot but abandon that old function of sorting, cri-
tique, and judgment that used to belong to it. From now on, literary anal-
ysis will suspend all judgment of the work, will suspend every function of
sorting for the reader. There will be no more sacred works, no more works
instantly valorized for literary analysis. The role of the critic, which con-
sisted in sorting and judging works, will no longer be anything but the role
of a sort of surveyor of literature. Relative to literary analysis, reviews like
the ones we read in the newspapers are, so to speak, nothing but a kind of
rump, and it’s well known that at the very extremity of this rump is planted
the quill of Pierre-Henri Simon.4
You see as well why and how history—historical analysis insofar as it is
the study of the production of a work—can no longer be the essential and
primary theme of literary analysis. For literary analysis no longer concerns
itself with how a work might have been produced but rather with how a
work can make space for another language in which the work manifests
itself or some [of its] aspects: the language of analysis.
I think that’s how we might explain the presence, firstly, of this new dis-
cipline called literary analysis and then, secondly, the proximity of literary
analysis to disciplines that seem far removed from it, but whose kinship
now becomes clear: all these disciplines deal with the document as docu-
4. Simon (1903–1972) was a literary critic, essayist, and novelist who, starting in 1961,
wrote the literary column for the newspaper Le Monde. Among his works of literary criticism,
see notably Pierre-Henri Simon, Histoire de la littérature française au xx e siècle: 1900–1950
(Paris, 1956) and Le domaine héroïque des lettres françaises: x e-xix e siècles (Paris, 1963).

ment. It could be a purely spoken document, as in the case of psychoanal-
ysis, or a document of the oral tradition, as in the analysis of folktales, or an
analysis of documents like the ones that sociology deals with.
That’s by and large what I wanted to say, quite schematically, to situ-
ate somewhat the problems of the structuralist disciplines and of literary
analysis. Now I’d like (and this would be the third direction for a possi-
ble debate) to situate for you a little—but you know it better than I do—
the current tendencies of structuralism insofar as it’s the form of literary
analysis.
The use of structural concepts in literary analysis poses a curious little
historical problem. As you know, structural analysis in the literary domain
was invented a very long time ago, exactly a half-century ago, in Russia.
Around 1915 the Russian Formalists, who were trained primarily in linguis-
tics, began to apply concepts that were already basically structural concepts
to literary analysis. Literary analysis in a structuralist form was then devel-
oped in Prague, in Czechoslovakia, the US, and England, where a certain
number of Russian Formalists had emigrated. And finally, just after the
war of 1940–1945, we saw this thing that is literary structuralism take shape,
in a very timid fashion, in France. Now, curiously enough, in France, struc-
turalism in the literary domain did not develop originally out of a reflec-
tion on what la langue is. Historically, that is, the linguistic model played
only a very small role, practically none at all, in the formation of the French
nouvelle critique. In fact, the point where the nouvelle critique established
itself in France, its point of entry, was psychoanalysis in the strict sense
of the term: the extended psychoanalysis of Bachelard and the existential
psychoanalysis of Sartre. The nouvelle critique was constituted from these
forms of analysis. It was only afterwards—quite recently, less than ten years
ago, barely seven or eight years—that literary analysis in France discovered
the linguistic model and sort of transferred the methods of the psychoan-
alytic denomination to the linguistic one. The psychoanalytic denomina-
tion was of course relatively lax, quite free in relation to Freudian scripture.
Nevertheless it was along those lines that structuralism could emerge. It
isn’t at all surprising that the structuralism of the nouvelle critique should
have been born out of psychoanalysis. The reason is simple: insofar as psy-
choanalysis is, in the end, the study of a document, the study of the human
parole as spoken by someone in a certain well-defined situation, it cannot
help but be structuralist, if only in the sense that it’s a deixological disci-
pline as well. So it’s no surprise that literary analysis in France has come
to structuralism not by way of linguistics but by way of psychoanalysis. I
think that’s how we should situate the birth of this nouvelle critique histor-
ically.

How was it developed and in what direction? Roughly, I think we can
say that the essential goal of everything we call nouvelle critique is to define,
firstly, the elements according to which one can divide a given text or a lit-
erary work; secondly, the network of relations that hold together the ele-
ments so defined. You’re going to tell me that all of this is simple, but it
poses problems. It poses problems because the work conforms to a division
into chapters, paragraphs, sentences, words; and this division isn’t the same
as the one that the analysis has to establish in order to show how and why
the work functions.
The first principle of structuralism in literary analysis is to consider that,
contrary to the old schema of the nineteenth century, the work isn’t essen-
tially the product of time; a work doesn’t follow, in its birth and then in
its existence, a linear pathway that would be roughly chronological. The
work is recognized as a fragment of space in which all the elements exist
simultaneously. Given this simultaneity, with the whole work laid out in
this manner, we can divide the work into its elements and establish any
functional relationship that may pertain between those elements. In other
words, it isn’t the diachronic thread of the work that must lead us, it’s the
synchrony of the work with regard to itself. This isn’t to say that we dis-
miss the fact that the work has appeared at a given moment, in a given cul-
ture, or with a given individual. But to define how the work functions, it’s
necessary to recognize that it’s always synchronic in relation to itself.
In general, literary analysis up to the present has established the syn-
chrony of the work in relation to itself in two ways: firstly, in the dimension
of the imaginary; secondly, in the dimension of language. The place in which
the work was initially spatialized and rendered contemporaneous with itself
was the imaginary. People tried to constitute—and you can say that, gener-
ally speaking, a number of works of literary analysis did constitute—a logic
or even a geometry of the imaginary. Such was the work of Bachelard at first.
It established a sort of elementary logic of the literary imagination by taking a
certain number of qualities and opposing them to one another, independent
of both the psychology of the author and the psychology of the reader. Such
qualities would exist objectively of themselves, at the heart of things, so to
speak, and their system of oppositions would give to the work its possibility
and its logic.5 That’s a sketch of the logic of the imaginary. You can find an
essay in the geometry of the imaginary in the work of someone like Poulet.
With regard to the circle, for instance, Poulet showed how the works them-
5. See Gaston Bachelard, Water and Dreams: An Essay on the Imagination of Matter, trans.
Edith R. Farrell (1942; Dallas, 1983) and Air and Dreams: An Essay on the Imagination of
Movement, trans. Farrell and C. Frederick Farrell (1943; Dallas, 1988).

selves—both in what they narrate and in the law that forms their different
parts and different elements—conform to geometrical figures that are at
once represented within the work and representative of it.6 In the same vein,
Starobinski, following Poulet, did a study of Rousseau on the theme of ob-
struction and transparency.7 He showed how, through the entire oeuvre of
Rousseau, you find this curious spatial figure of a sort of opacity that comes
to cover things and to isolate man from things—and then the search for a
transparency that cannot be achieved or anyway that must be achieved
through language as the instrument of the “translucification” of this veil or
wall that separates the individual from things. Language is what refines this
sort of veil and makes it transparent. The work is animated by all this in
its themes, but at the same time the work is precisely this spatial figure, for
it is through his work, through that very work of literature, that Rousseau
wrote. It’s through that work that Rousseau tried in effect to make the world
transparent—this world that had become absolutely opaque and lost to
him ever since childhood and the injustice to which he had been victim
in childhood. Thus the work itself is that sort of spatial configuration and
dynamics of space that gets represented in what it says. It’s in this way that
we might situate quite a few analyses of the logic and geometry of the imag-
inary.
There’s a second approach that’s much more recent: the analysis of the
literary work based on the linguistic schemas that characterize it. I think
this analysis was done for the first time in France by Lévi-Strauss, with re-
gard to a sonnet by Baudelaire, where he showed how “Les Chats” was
wholly structured by the phonetic possibilities available to Baudelaire and
that the poet constructed this sonnet based on a system of redundancies
that he arranged according to the phonetic characters proper to the French
language.8 This study, which remained little known—quite forgotten—for
a number of years, has recently been brought to light again. And now the
efforts of Barthes and Genette are wholly oriented in this direction, except
that the linguistic schemas with which they attempt to define a work are
not phonetic, but syntactical and semantic. Basically they use rhetoric and
rhetorical schemes as a guiding thread for the analysis of works. This as-
sumes, of course, that the literary work itself is nothing other than a kind
6. See Georges Poulet, The Metamorphoses of a Circle, trans. Carley Dawson and Elliott
Coleman (1961; Baltimore, 1966).
7. See Jean Starobinski, Jean-Jacques Rousseau: Transparency and Obstruction, trans. Arthur
Goldhammer (1971; Chicago, 1988).
8. See Roman Jakobson and Lévi-Strauss, “‘Les Chats’ de Charles Baudelaire,” L’Homme 2,
no. 1 (1962): 5–21.

of redoubling of linguistic structures on themselves—that the literary work
is, in a way, la langue manifesting itself in its structure and in its virtuality.
Lastly, there would be a third approach—and I’ll stop myself because,
once more, I must give you the merest indications. There is a third ap-
proach that’s currently still more or less unexplored, but of which we
can ask ourselves if it would not be [allowed].9 You know that the people
who have been reflecting on language lately, the linguists on the one hand
and the logicians on the other, have noticed that when we study state-
ments, there is an element or rather a series of elements that are at least
as important as la langue, and this is what we call, roughly, the extralinguis-
tic. Linguists like Prieto and logicians like Austin have shown that, in fact,
the linguistic structure of a statement is by no means sufficient to account
for its total existence. Prieto in particular has shown how the contextual
elements constituted by the very situation of the individual speaking are
absolutely necessary to give meaning to a number of statements and, in-
deed, to a large number of statements.10 In fact, every statement relies si-
lently on a certain objective and real situation, and the statement would
certainly not have the form it has if the context were different. The first
(princeps) example that Prieto takes up is this: when you have a red note-
book on a table and you want to ask someone to pick up that notebook,
you say to him, “pick it up” or “pick up the notebook”; when there are two
notebooks, one red and one green, you ask your interlocutor to pick up
one of the notebooks by saying, “pick up the red one” or “pick up the
one on the right.” You see that these two statements have exactly the same
meaning: an order from A addressed to B to pick up the notebook on the
table. One and the same meaning gives rise to two statements that are en-
tirely different depending on whether the objective context is the first or
the second. It follows that the definition of a statement, the choice of the
form of a statement, is possible only as a function of this context.
On the other hand, and now I’m moving to the research of the logicians,
someone like Austin has shown that the statements themselves cannot be
analyzed independently of the speech act that is actually performed by the
speaker at the moment when he speaks.11 For example, when someone says,
“the session is called to order,” this phrase isn’t at all constative. The speaker
is not remarking that the session is open; at that moment, it isn’t open! Nor
is he giving an order: the session doesn’t obey, it doesn’t open on its own
because it was ordered to. So what is this statement? It’s a statement that
9. Conjecture (inaudible).
10. See Luis J. Prieto, Messages et signaux (Paris, 1966).
11. See J. L. Austin, How to Do Things with Words (New York, 1962).

is grammatically identical to a constative and that is nevertheless neither
a constative nor an affirmation. It’s something that Austin calls a perfor-
mative. Regardless of the name, though, you can see from this simple ex-
ample that the description of a statement is by no means complete when we
have simply defined its linguistic structure.
You see from these two examples—which are just points of reference—
that within the study of language, we are in the process of realizing that the
analysis of discourse can no longer be done solely in linguistic terms. Dis-
course isn’t simply a particular case within la langue. Discourse isn’t a
manner of combining elements together according to the linguistic rules
that are given by la langue itself. Discourse is something that necessarily
overflows la langue. We could thus ask ourselves if literary analysis—that
is to say, the analysis of the special discourse that is a literary work—
shouldn’t take into consideration all the extralinguistic elements that we
are now in the process of discovering in the analysis of language. In general,
I see [two] directions in which we could go.12
Firstly, we could try to define, somewhat in line with what Prieto de-
scribed, what is actually said in statements of literature. Indeed, when
you open a novel, there is no context to this novel. When, for example,
Joyce, beginning Ulysses, says—unfortunately the name of the character es-
capes me—“Descended the staircase,” the staircase is designated by a definite
article but it isn’t next to you. It isn’t like when you say, when I say, for ex-
ample, “the glass.”13 When I say “the glass,” you know perfectly well that it’s
this glass right here. When Joyce in his novel says “the staircase,” nobody
knows what this staircase is; there’s no real context. And yet, Joyce doesn’t
say everything, he doesn’t explain exactly what the context should be that
we should put in place in order to fill, as it were, this empty indication given
by the definite article. It’s the work itself that divides, in some way, in a
nonexistent context, what does and does not need to appear. It’s enough to
compare a description in Balzac, for example, to a description in Robbe-
Grillet to see exactly how, in certain Balzacian works, there are a number
of things that absolutely must be said and that in some sense are the con-
text, the extralinguistic presented in the work itself: the date of the event,
the city where it’s located, the name of the character, his ancestors, what
has happened to him, his past, and so on. But if you take a novel by
Robbe-Grillet: when he begins [In the] Labyrinth by saying “Here,” you will
12. Foucault says “three.”

13. The English text reads: “Halted, he peered down the dark winding stairs and called
out coarsely: / —Come up, Kinch! Come up, you fearful jesuit” (James Joyce, Ulysses [1922;
New York, 1990], p. 3).

never know what this “here” is—if it’s a city, which city it is, in what coun-
try, if it’s an apartment, if it’s a picture, if it’s a real space, if it’s an imaginary
space, and so on.14 You see therefore that the manner in which the extra-
linguistic gets manifested in the statements of the literary work is very dif-
ferent from one epoch to another and from one writer to another. And we
could, in the vein of the linguistic analyses of Prieto, study the role of the
extralinguistic context within the work itself.
Secondly, we could also study somewhat in the line of the logicians,
Austin in particular. We could study the way in which statements are
somehow posed in the very text of the literary work: what is the act that
is actually performed in a given sentence? It’s clear that in a description,
in a reported dialogue, in a reflection by the author on his own character,
in a psychological notation [. . .] you see there an entire formal analysis of
the work, but one that would be done in a way that isn’t at all that of lin-
guistics.15 It would be a structural study of what is extralinguistic in the lin-
guistic statements of the work itself.
I have simply pointed out these possible directions of work to show you
essentially how structuralism, far from being bound to some doctrinal
position, far from being bound to a precise and definitively realised method,
is really much more a domain of research that opens out in a way that is
probably rather undefined. In any case, so long as we have not yet run
through the ensemble of this documentary mass that humanity has depos-
ited around itself, of which literature is a part, so long as we have not used
all possible methods to show what this document as document is, you see
that structuralism—if it’s true that it is simply the science of the docu-
ment—will have a good life. In any event, we should absolutely not identify
structuralism either with a philosophy or even with a particular method.
These are, very roughly, the indications I wanted to give you—unfor-
tunately I’ve gone a little long—as a simple prelude to the questions and
objections that you’d like to make.
14. See Alain Robbe-Grillet, In the Labyrinth, trans. Christine Brooke-Rose (1959; London,
1967).
15. The ellipses in this sentence mark an interruption in the recording.


Foucault. Structuralism and Literary Analysis

Uploaded by

Document Information

Copyright

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Foucault. Structuralism and Literary Analysis

Uploaded by

Copyright:

Structuralism and Literary Analysis

Translated by Suzanne Taylor and Jonathan Schroeder

Ladies and Gentlemen,

Critical Inquiry 45 (Winter 2019)

1. See Jean-Pierre Richard, L’univers imaginaire de Mallarmé (Paris, 1961).

Mic h el Fou ca ul t , acknowledged as the preeminent philosopher of France

Critical Inquiry 2019.45:531-544.

Critical Inquiry 2019.45:531-544.

Critical Inquiry 2019.45:531-544.

Critical Inquiry 2019.45:531-544.

Critical Inquiry 2019.45:531-544.

Critical Inquiry 2019.45:531-544.

Critical Inquiry 2019.45:531-544.

Critical Inquiry 2019.45:531-544.

Critical Inquiry 2019.45:531-544.

Critical Inquiry 2019.45:531-544.

Critical Inquiry 2019.45:531-544.

12. Foucault says “three.”

Critical Inquiry 2019.45:531-544.

Critical Inquiry 2019.45:531-544.

You might also like