University of Virginia Library

Search this document 
Dictionary of the History of Ideas

Studies of Selected Pivotal Ideas

collapse sectionII. 
collapse sectionII. 
collapse sectionII. 
collapse sectionVI. 
collapse sectionVI. 
collapse sectionVI. 
collapse sectionVI. 
collapse sectionIII. 
collapse sectionI. 
collapse sectionVI. 
collapse sectionVI. 
collapse sectionI. 
collapse sectionVI. 
collapse sectionVI. 
collapse sectionVI. 
collapse sectionVI. 
collapse sectionVI. 
collapse sectionIV. 
collapse sectionIV. 
collapse sectionII. 
collapse sectionIV. 
collapse sectionV. 
collapse sectionIII. 
collapse sectionVI. 
collapse sectionIII. 
collapse sectionIII. 
collapse sectionV. 
collapse sectionVI. 
collapse sectionIII. 
collapse sectionIII. 
collapse sectionVI. 
collapse sectionVI. 
collapse sectionVI. 
collapse sectionV. 
collapse sectionV. 
collapse sectionVII. 
collapse sectionV. 
collapse sectionI. 
collapse sectionI. 
collapse sectionV. 
collapse sectionVI. 
collapse sectionVII. 
collapse sectionIII. 
collapse sectionIII. 
collapse sectionIII. 
collapse sectionVII. 
collapse sectionIII. 
collapse sectionI. 
collapse sectionIII. 
collapse sectionVI. 
collapse sectionII. 
collapse sectionVI. 
collapse sectionI. 
collapse sectionV. 
collapse sectionIII. 
collapse sectionI. 
collapse sectionVII. 
collapse sectionVII. 
collapse sectionII. 
collapse sectionVI. 
collapse sectionV. 
collapse sectionV. 
collapse sectionI. 
collapse sectionII. 
collapse sectionII. 
collapse sectionIV. 
collapse sectionV. 
collapse sectionV. 
collapse sectionV. 
collapse sectionII. 
collapse sectionII. 
collapse sectionV. 
collapse sectionV. 
collapse sectionIV. 


Structuralism, as a recognizable movement in the
history of ideas, was born between the two world wars
and is still developing at the present time. No account
of it, therefore, can pretend to be complete or final.
The idea of structure is much older, the term having
entered the vocabulary of biology in the seventeenth
century and of language, literature, and philosophy in
the nineteenth. The closely associated notion of “sys-
tem” is of course older still. It is important at the outset
to get clear the relations between these two concepts,
and this can best be done be relating them both to
a third, namely that of “function.” According to the
standard structuralist account, structures are structures
of systems; systems function, structures in themselves
do not function—but systems function because they
have the structures they do. The system of traffic
signals, for example, has the function of controlling
traffic; its structure is a binary opposition of red and
green lights in alternating sequence. The system may
share its form with other systems having different
functions, but structure is not merely form; form is
something that can be abstracted from matter or con-
tent and considered separately, whereas structure, in
the structuralist sense, is precisely the significative (as
opposed to the material) content of the system.

As a first, although necessarily misleading, definition
it might then be said that structuralism is the view
that structure in this sense is a more fundamental
characteristic of the objects it studies (all, it must be
noted, products of the human mind or of human cul-
ture), than are their physical components, their genetic
origins, their historical development, their function or
purpose, and so on.

This article is intended, however, as an account not
of the idea of structure itself but of a set of perspectives
to which the recent prominence of this idea in a num-
ber of disciplines has given rise. The occurrence of the
term in the writings of a historically important figure
is not enough to make him a structuralist, while its
absence may not prevent him from being recognized
as such by other structuralists. Ferdinand de Saussure,


le structuraliste sans le savoir, as one commentator has
called him (Georges Mounin, Saussure, ou le struc-
turaliste sans le savoir,
Paris [1968]), hardly used the
word “structure” at all, and yet if a single point of
origin had to be found for the movement it could only
be the posthumous publication in 1916 of his Cours
de linguistique générale,
compiled from the notes of
his students, Bally, Sechehaye, and Reidlinger. The
picture is complicated by the fact that there is dis-
agreement among the leading structuralists as to the
scope of the movement—whether it is to be limited
to the methodology of the social sciences (or even of
linguistics only) or whether its extension to literary
criticism and philosophy is legitimate. Some of the
most obvious structuralists in the latter categories (e.g.,
Michel Foucault) have disclaimed the designation.

For present purposes we may agree that a movement
exists if, at a more or less precise historical juncture,
some conceptual development attracts the interest of
a number of thinkers, who interact (even if sometimes
negatively) in virtue of this common interest, claim
common intellectual ancestors, etc. It is not a question
of their belonging to the movement, of adhering to
it in any formal sense (as e.g., in the case of surrealism);
intellectual relations need not be accompanied by
social or professional ones. Also the title given to the
movement may not be the clearest possible indication
of its intellectual content. All these remarks apply to
structuralism, and this may account in part for the
extraordinary range of opinion about it among critics
and practitioners alike: it has been dismissed as a mere
fashion, and hailed as a fundamental and irreversible
change in the pattern of human thought. Whatever
the merits of these extreme positions, recognizably
structuralist trends are now to be found in half a dozen
disciplines in the social sciences and humanities,
collectively the sciences humaines for the French,
among whom the movement has chiefly taken hold.


The master discipline of structuralism, to which all
its practitioners constantly revert, is linguistics. Scien-
tific linguistics has a long history, but two distinct
stages—the comparative and the structural—can be
recognized in its modern development. The first begins
with J. J. Rousseau's Essai sur l'origine des langues
(published posthumously, 1817) and was given its
greatest impetus by the rediscovery in the West of
Sanskrit in the early nineteenth century. The impor-
tance of the latter was to suggest a common origin
and an earlier common structure at least for Indo-
European languages, and also to draw attention to the
details of the evolution of these languages from earlier
forms. But comparative and evolutionary (diachronic)
studies remained fragmentary in the absence of a sys-
tematic theory of language as a synchronic entity.

Ferdinand de Saussure, in his courses in Paris and
Geneva, put forward and developed the view that a
language, as actually spoken by a linguistic community
at a given time (i.e., viewed synchronically), forms a
self-contained system, each element of which has its
place and its own relations (grammatical, etymological,
etc.) to the other elements. His work was devoted to
describing the structure of this system, although as
remarked earlier he did not make use of the term in
this sense. In the English version of the Cours de
linguistique générale
(Saussure, 1959—referred to from
now on as CGL) the word “arrangement” comes closest
to it in meaning. Saussure distinguishes between “ex-
ternal” and “internal” linguistics with the famous ex-
ample of the game of chess:

In chess, what is external can be separated relatively easily
from what is internal. The fact that the game passed from
Persia to Europe is external; against that, everything having
to do with its system and rules is internal. If I use ivory
chessmen instead of wooden ones, the change has no effect
on the system; but if I decrease or increase the number
of chessmen, this change has a profound effect on the
“grammar” of the game. One must always distinguish be-
tween what is internal and what is external... everything
that changes the system in any way is internal

(CGL, pp.

We might now read “structure” for “system” in this
last sentence. The novelty of Saussure's method was
its resolute adherence to internal questions (in contrast
to comparative and other earlier methods).

Some of the chief concepts of contemporary struc-
turalism are borrowed directly from Saussure, and it
will be well to introduce them in the context of his
work. “Synchrony” and “diachrony” have already been
referred to; the opposition between them is reflected
in the opposition langue/parole (rendered in CGL as
“language” and “speaking”). Both are special cases of
langage, the general human faculty of language, of
which speech is the primary manifestation, writing
being an anomalous and, from the point of view of
the linguist, unhelpful addition. Langue is the
synchronic, social reality apart from its individual
manifestations, which constitute parole. Langue there-
fore is the system whose structure at a given time and
in a given community is the object of linguistic investi-
gation; parole is constituted of the particular verbal
acts (controlled by the conventions of langue) which
are performed daily and forgotten, which give lan-
guage its empirical and historical reality and are its
diachronic medium of evolution. Without langue, pa


role would be a series of isolated and meaningless
utterances; without parole, langue would be an abstract
and empty system.

The function of the system of language is significa-
and its elements are signs. The sign, however,
is a complex entity, whose two aspects—the signans
and signatum of the Stoics—are called by Saussure
signifiant and signifié (“the signifier” and “the signi-
fied”). Both are psychological: the signifier is a “sound-
image,” the signified is the concept associated with this
sound-image. The sign, for Saussure, is arbitrary—that
is to say, there is no internal connection between its
aspects which would make a given signifier the natural
or necessary vehicle for a given signified. What gives
the sign its linguistic value is the system of differences,
on the one hand between signifiers, on the other be-
tween signifieds (or significata, some writers in English
preferring “significatum” to “signified” as a noun). “In
language there are only differences.... A difference
generally implies positive terms between which the
difference is set up; but in language there are only
differences without positive terms” (CGL, p. 120). This
quotation may be taken as a key to structuralism in
general. Thought on the one hand, language on the
other, are not antecedently segmented, they do not
exist as separable atomic units; the structuring activ-
ity creates the units, bringing definiteness to both
sides simultaneously. The same would apply to other
significative systems; Saussure himself envisaged a
new science of “semiology” that would study all
systems of signs and of which linguistics would be only
a part.

At the end of CGL Saussure raises the question of
a possible link between linguistics and anthropology,
but concludes that the former can be of little help to
the latter. He was thinking of course of the content
of linguistics as a datum for anthropology, rather than
of its method as a model; at the time, as Claude Lévi-
Strauss remarks (Anthropologie structurale, hereafter
AS, p. 39, n.); “the founders of modern linguistics
placed themselves resolutely under the patronage of
the social scientists. It was only after 1920 that Marcel
Mauss began... to reverse the trend” (trans. P. C.).
In fact a good deal had to happen in linguistics before
its position as an exemplar for the social sciences be-
came clear. Not the least of these developments in
historical importance was the explicit emergence of
the term “structure” in the sense in which it is now
understood, which occurred in a programmatic docu-
ment prepared for an international congress of lin-
guists at the Hague in 1928, by Roman Jakobson, S.
Karcewski, and Prince Nicolas Troubetzkoy.

Troubetzkoy was a pioneer of the science of
phonology, and his Grundzüge der Phonologie (1939)
contains a detailed treatment of the concept of opposi-
which, while worked out in terms of phonological
systems, forms an essential basis for technical analysis
in other domains. In an earlier article (Troubetzkoy,
1933) he suggested a distinction which was to be cru-
cial for later structuralism between conscious and
unconscious levels of structure in language. What is
actually pronounced (which is studied by phonetics)
does not necessarily exhibit directly the system of
spoken language
(which is studied by phonology); the
laws of the latter have to be established by induction
and hypothesis on the evidence provided by the former.
Troubetzkoy himself, in his later work, placed less
weight on the opposition conscious/unconscious than
some of his readers (notably Lévi-Strauss, as noted
below), but he would have been prepared to admit
the general point that the true structure does not
always appear on the surface. The notion of structure
here is extended to the domain of parole, and, with
an explicit reference to “teleological elements,” to the
diachronic. The article referred to ends with a general
remark: “The epoch in which we live is characterized
by the tendency of all scientific disciplines to replace
atomism by structuralism...” (Troubetzkoy [1933], p.
246; trans. P. C.).

What might be called a structuralist explosion was
touched off in France by the anthropological works
of Lévi-Strauss (see below) and has since been the most
visible manifestation of the movement. But linguistic
structuralism was to take on a new direction and a
new importance with the work of Noam Chomsky.
Chomsky was one of the first to draw serious attention
to a comparatively neglected aspect of linguistic be-
havior, namely its creative character (although he rec-
ognizes a precursor in Wilhelm von Humboldt). The
structures dealt with by most grammarians are surface
generative grammar, Chomsky's main con-
tribution to technical linguistics, takes its theoretical
point of departure from deep structures, basic gram-
matical forms from which the whole variety of surface
structures can be generated. The existence of deep
structures explains the ability of all speakers of a lan-
guage to utter sentences that may never before have
been uttered, and to be understood at once by other
speakers of the same language. A language then is not
an actual but a potential (and potentially infinite) set
of utterances, governed by the laws of its deep struc-

The interest of this work for the structuralist move-
ment is the suggestion that languages whose surface
structures are widely different may share the same deep
structure, and that this may reflect in some funda-
mental way the structure of mind. Chomsky speculates


... the linguistics of the next generation will reveal that
each natural language is a specific realization of a highly
restrictive schema that permits grammatical processes and
structures of a very limited variety, and that there are
innumerable “imaginable” languages that violate these re-
strictions and that are, therefore, not possible human lan-
guages in a psychologically important sense, even though
they are quite able, in principle, to express the entire
content of any possible human language

(Chomsky [1967],
p. 8).

It should be remarked that what gives Chomsky
grounds at this time for this expectation is “the new
understanding of recursive mechanisms and the nature
of algorithms that has developed in the last thirty
years” (i.e., up to 1967); the structuralist hypothesis,
at least in the hands of its more responsible proponents,
is no mere speculation but rests on painstaking obser-
vation and vigorous analysis. It must be said further
that Chomsky himself has serious reservations about
the extension of linguistic structuralism to other do-
mains, even by comparatively careful thinkers like
Lévi-Strauss (Chomsky [1968], p. 65). While we may
admit with Lévi-Strauss that the success of linguistics
in structural analysis is encouraging, it seems best to
confront anthropological and other data on their own
terms and not to assume that every significative system
is in fact a kind of language.

The development of linguistics has nevertheless been
intimately connected with that of anthropology. Some
non-Indo-European linguistics indeed seemed for a
long time to be a part of anthropology, since the data
for it came from exotic societies like that of Java, whose
language was studied by Wilhelm von Humboldt in
the early nineteenth century, and above all those of
the Indians of North America, whose extraordinarily
diverse and complex languages provided the initial
stimulus for American linguistics in the late nineteenth
and early twentieth centuries. But just as comparative
linguistics remained fragmentary until the notion of
language as system came to the fore, so comparative
ethnology remained fragmentary (devoted as it was to
the study of tools, pottery, the practice of hunting and
agriculture, etc.) until the parallel notion of society
as system was worked out. This notion had been put
forward earlier on the level of civilized society by
Montesquieu, but the elements of his system were
themselves highly developed institutions requiring
analysis in their turn. The study of kinship provided
for anthropology the paradigm of structure that
phonology has provided for linguistics, and with the
nineteenth-century American ethnographer Lewis H.
Morgan, the idea of kinship systems became firmly
established. Morgan, however, like Saussure, used other
terms than “structure.”

Structuralism as an explicit movement in anthropol-
ogy probably begins with A. R. Radcliffe-Brown, who
added to the concept of function as it is found in Émile
Durkheim and B. K. Malinowski the concept of the
structure of a functioning system, roughly in the form
in which it is outlined in the opening paragraphs of
this essay (Radcliffe-Brown [1952], pp. 179-80). Later
structuralists tend to think of Radcliffe-Brown as a
functionalist (Lévi-Strauss speaks of “that primary form
of structuralism which is called functionalism”; AS, p.
357), and to attribute to him two errors of judgment:
first the belief that every feature of every society has
an explicit function, second the belief that the structure
of society is to be observed directly, as a surface
phenomenon. Lévi-Strauss says, with respect to the first
of these points, “To say that a society functions is a
truism; but to say that everything in a society functions
is an absurdity” (AS, p. 357; trans. P. C.). But Radcliffe-
Brown, speaking of the hypothesis that social systems
have a functional unity, remarks that “the hypothesis
does not require the dogmatic assertion that everything
in the life of every community has a function. It only
requires the assumption that it may have one, and that
we are justified in seeking to discover it” (op.cit., p.
184). On the second point he says: “In the study of
social structure the concrete reality with which we are
concerned is the set of actually existing relations, at
a given moment of time, which link together certain
human beings. It is on this that we can make direct
observations. But it is not this that we attempt to
describe in its particularity... what we need for
scientific purposes is an account of the form of the
structure” (idem, p. 192).

Radcliffe-Brown thus prepares the way for a struc-
turalism which will go beyond the merely descriptive.
It remained for Lévi-Strauss to bring anthropological
structuralism (and with it the movement as a whole)
to the position it now occupies. Lévi-Strauss's early
work was a sequel to that of Marcel Mauss, whose
“Essai sur le don, forme archaïque de l'échange”
(Année Sociologique, N.S. 1, Paris [1932-34]) had pro-
vided a dynamics for social systems in terms of ex-
changes. There is an obvious analogy here to systems
of language; languages signify, but they do so, at least
in part, in order to communicate, and communication
proceeds by the exchange of words. In Les structures
élémentaires de la parenté
(1949, hereafter SEP) Lévi-
Strauss, taking as his starting point the universal prohi-
bition (under some form or other) of incest, interprets
exogamous marriage as a system of the exchange of
women, and adumbrates a general theory in which
gifts, women, and other media of exchange between
groups would behave like signs: “the relations between
the sexes might be regarded as one of the modalities


of a great 'function of communication' which includes
language as well” (SEP, p. 613; trans. P. C.). It is hardly
necessary to point out that Lévi-Strauss does not regard
women merely as signs; on the other hand, the fact
that they are never consciously viewed as such (in our
own or any other society) is not incompatible, for the
structuralist, with their really fulfilling this role.

This unconscious character of the most fundamental
social and psychological structures is (as was remarked
above in connection with Troubetzkoy's work on
linguistic structures) an important component of Lévi-
Strauss's teaching. “It is necessary and sufficient to
arrive at the unconscious structure underlying an insti-
tution or a custom to acquire a principle of inter-
pretation valid for other institutions and other customs,
provided of course the analysis is carried far enough”
(AS, p. 28; trans. P. C.).

The difference between linguistics and anthropology,
so far as the contribution of structuralism is concerned,
lies in the fact that they start from different conditions
of knowledge. In the case of language the function of
the system, or at least a function of it (namely commu-
nication), has been known for a long time; structural
linguistics tells us what structural elements and rela-
tions enable it to fulfill this function. In the case of
kinship the elements and relations have been known
for a long time; what structural anthropology tells us
is how they constitute the structure of a system and
what the function of that system might be (AS, p. 40).
Just as in the case of language, however, structural
characteristics reveal themselves only in a study of
differences among customs, myths, kinship patterns,
and the like: what is constant is not a particular rule
of marriage but a pattern of such rules, any variant
of which will lend stability to a society with limited
numbers by ensuring a cyclic order of exchanges among
its parts.

One of the reasons for the enormous popularity of
the structuralist movement (and hence at least in part
for its extension to other fields) is to be found in two
books by Lévi-Strauss which depart from the purely
scientific concerns of his earlier and later works. These
are Tristes tropiques (1955) and La pensée sauvage
(1962). The former is part autobiography, part
travelogue, part philosophical reflection; it is not
explicitly structuralist at all, but some of its themes
nevertheless throw significant light on what might be
called the moral component of structuralism. It makes
clear the influence of Marxism and psychoanalysis on
Lévi-Strauss's intellectual development (in another
place he speaks of having “borrowed the notion of
structure from Marx and Engels”; AS, p. 364), provides
a vivid background for his later work on mythology,
and reveals his preoccupation with three basic
sociologico-philosophical problems: the place of the
individual in the collective, the relation between the
collective and the natural world, and the role of the
ethnographer with respect both to the culture he
studies and to his own culture. As for Chomsky, so for
Lévi-Strauss the structures he studies are rewarding
insofar as they reflect the structure of mind.

In La pensée sauvage he takes up this theme
explicitly, examining as he puts it “mind in its natural
state.” One of the things that stand in the way of our
grasping it in this state is of course the conditioning
to which our own minds have been subject, so that
we superimpose on the products of other minds a grid
of logical or utilitarian expectations. Hence we are
inclined to think that the primitive has an inadequate
science and an unformed sense of logical propriety.
Lévi-Strauss argues, however, that what he calls the
“science of the concrete”—the complex systems of
nomenclature and association that are to be found in
virtually all primitive cultures—is just as theoretical,
in its way, as our own science, that the heterogeneous
structures of myth and totemism follow rules just as
intelligible and rigorous as those of our more neatly
engineered systems.

Engineering may be taken as a paradigm of our
aspirations: everything new, shiny, detachable, precise.
The primitive is a bricoleur rather than an engineer,
and bricolage has become one of the basic ideas of
structuralism. It means, in essence, the use for purposes
of construction of anything that comes to hand, partic-
ularly of elements or remnants of former constructions.
So the myth will be built up from fragments of earlier
myths, as language is built up from fragments of earlier
languages. This heterogeneity of content is perfectly
compatible with rigor of structure, and that rather than
content is the important thing. But it is much less
readily seen in a product of bricolage than in a piece
of engineering, and therefore has to be sought for by
a process of analysis. The structure that emerges from
this process, however, being in some sense indifferent
to the materials in which it is realized, is likely to
reflect more accurately than a piece of engineering the
natural contours of the mind that gave rise to it.

The process of analysis in the case of myth, to take
the best-known example, consists in showing how one
myth can be seen as a transformation of another, or
both as transformations of a third that may not even
occur in the corpus of myths; it is a search for the
deep structure which explains and generates, in the
epistemological and ontological orders respectively,
the surface structure exhibited by actual myths. The
deep structure is never extracted in order to be shown
by itself, it emerges from the repeated contrasts and
oppositions of variants, complementary versions and


so on. The works in which Lévi-Strauss has carried on
this enterprise are the various volumes of his collection
Mythologiques, and the scale of the project can be
gauged from the fact that in the first volume alone,
Le cru et le cuit (1964, hereafter CC) no fewer than
187 separate myths are analyzed. Structuralist analysis
does not proceed in a straight line, from a clear-cut
problem to a definitive solution, but works over its data
until they yield up their own intelligibility. No myth
is more basic than another—“the world of mythology
is round,” as Lévi-Strauss remarks in Du miel aux
(p. 7)—although one or another myth may be
chosen as a key myth or myth of reference for purposes
of organization. The aim of the work is consistent with
that of the earlier scientific researchers, namely to
“draw up an inventory of mental patterns, to reduce
apparently arbitrary data to some kind of order, and
to attain a level at which a kind of necessity becomes
apparent, underlying the illusions of liberty” (CC, p. 10).


The basic tenet of structuralism at the substantive
rather than the methodological level might be rendered
as follows: what makes anything intelligible to man
is a coincidence of structure between it and him. Na-
ture is not intelligible except insofar as we are able
to formulate its workings in theories of our own con-
struction (that we are able to do this so readily may
be, as C. S. Peirce suggested, due to the fact that we
ourselves are products of the same laws as nature is).
Structuralism applies chiefly to cultural domains, in
which everything is directly or indirectly a product
of mind, so that a coincidence of structure is not to
be wondered at. (Lévi-Strauss's observations on the
prohibition of incest—an apparently cultural rule
which seems to have the universality of a natural
law—gave one of the first impulses to his development
as a structuralist.) But the coincidence may not be
evident at first, or not fully so; work may be required
to bring it to light. It can become intelligible in its
turn only through a second-order coincidence of struc-
ture, and it is obvious that there can be no last step
in this process, which explains why in the end intel-
ligibility must be shown rather than argued.

Clearly language is the most fundamental and most
universal cultural product. Its own proper structure
is the concern of linguistics, but linguistics, even the
linguistics of parole, deals with language on a universal
level. Language in actual use gives rise to a number
of kinds of artifact that escape merely linguistic analy-
sis, among which are two of particular interest for
structuralism: what is made with language, i.e., litera
ture, which is the concern of criticism, and what makes
with language, i.e., the subject, which is the
concern of psychoanalysis.

One of the most striking examples of structuralist
criticism—an analysis of Charles Baudelaire's sonnet
“Les Chats,” which appeared originally in L'Homme,
an anthropological journal—is again due to Lévi-
Strauss, in collaboration not with a literary critic but
with a linguist, Roman Jakobson (Jakobson and Lévi-
Strauss, 1962). Literature seems to Lévi-Strauss to oc-
cupy an area of intersection between linguistics and
ethnology: “The linguist discerns in poetic works
structures that bear a striking analogy to those found
in myths by ethnologists. In their turn ethnologists
cannot help recognizing that myths are not merely
carriers of concepts: they are also works of art” (idem,
p. 5). Art is a cultural product; literature is art in the
medium of language. The analysis of Baudelaire's
sonnet consists simply in laying out the oppositions that
occur in it, on the various levels of phonology, phonet-
ics, syntax, prosody, semantics, and so on; there is
nothing critical about this process if by criticism is
meant bringing to bear external canons of form or style.
But the outcome of it is precisely a “making intel-
ligible” of the work on levels that are not accessible
to the casual reader, or even to the informed reader
who lacks the technical resources of the linguist.

This is a far cry from criticism as it is practiced by
professional critics, but is does not exhaust the potential
contribution of structuralism to their discipline. Works
of literature are linguistic constructions, but like myths
they too have conceptual content, as well as thematic,
social, psychological, and aesthetic aspects. Struc-
turalism recognizes intelligibility wherever it is to be
found, and its task with respect to literature is to
explore the significance of these aspects also. Struc-
turalist criticism in France has come to be classed with
psychoanalytic, Marxist, and existentialist approaches
as “New Criticism” or “ideological criticism” because
of its refusal to observe the customs and pieties of the
literary establishment, and because of its insistence on
taking the work as an object in the world—a text, the
product of écriture or “writing,” a human or social trace
comparable to a myth, or on another level to a temple
or a city, open to archeological reconstruction or
deciphering—rather than as the masterpiece of a par-
ticular author or an element of a chauvinistic literary
heritage. Lévi-Strauss's remark that “myths have no
authors,” if extended to the classics, is certainly cal-
culated to enrage traditional critics who are as con-
scious of their classical antecedents as the French, and
the polemic over structuralist criticism has been espe-
cially bitter where it has concerned Racine, as seen in
Roland Barthes, Sur Racine (1963), Raymond Picard,


Nouvelle critique ou nouvelle imposture? (1965), and
Barthes' reply to this, Critique et vérité (1966).

Barthes is the leader of the structuralist movement
in criticism; he has also attempted (in his Éléments de
1964) to summarize what can reasonably
be said at this point about the system of signs in gen-
eral, as a first step towards the science envisaged by
Saussure as well as by Lévi-Strauss which would deal
with language but go beyond it, treating it as only one
among the many significative systems available to us.
Not surprisingly, the only developed technical appara-
tus he is able to present comes precisely from linguis-
tics, and while it is certainly correct to treat language
as the principal and most obvious structured system,
there may be some danger in assuming that it is
paradigmatic of all of them. Other examples of Barthes'
diverse and resourceful brand of structural analysis are
to be found in his Système de la mode (1967)—which
is not, as many people have supposed, a book about
fashion, but rather about the language to which it gives
rise—and in his commentaries on various aspects of
French culture, such as the plates of Diderot's
Encyclopédie, the Eiffel Tower, etc., and an earlier
series of short essays on everyday phenomena collected
in Mythologies (1957).

In any discussion of significance, from C. S. Peirce
to the present, the question: To whom do the signs
signify what they do? is bound to arise. To this we
might answer, man, the conscious subject, etc., thus
giving a name to the problem but not doing anything
to solve it. Now it is remarkable that the subject does
not in general simply take in significance, but charac-
teristically responds to it, and this most often not by
action (which might be an appropriate response to
stimuli received from the natural world) but by dis-
Significance evokes significance in its turn (in
Sur Racine Barthes alludes to “the silence of the work
which speaks, and the speech of man who listens”).
This leads naturally enough, but by a new route, to
the hypothesis that mind itself is a system with the
structure of discourse. Instead of inferring the structure
of mind in general from the structure of its large-scale
causal consequences (language, cultural artifacts, and
so on), what it suggests is that we might infer the
structure of particular minds, i.e., personalities, from
their immediate causal consequences, the syntagmatic
spoken chain that makes up the speech of each indi-
vidual person. Structuralism thus has an intimate bear-
ing on psychoanalysis, and this connection has been
worked out in the writings of Jacques Lacan, a
Freudian, who in spite of early disagreements with the
official organization of psychoanalysts in France has
continued through his own group (l'École freudienne
de Paris
) to exert a great influence on their practice.

As in the case of kinship practices, myths, and works
of art, the speech of the individual man (for the psy-
choanalyst, the patient) is to be seen as moving on two
levels simultaneously. He uses language consciously to
say what he wishes to say, and at the same time uncon-
sciously says something quite different, which it is the
analyst's task to interpret. According to the Cartesian
view, in which the subject (the “I”) is constituted by
the cogito, these two differing and even conflicting
messages would have to be provided by two different
subjects. Given that the conscious, reflective subject
is the “I” of Descartes, who, Lacan asks, is the other
one? His answer is that the Unconscious, which has
the structure of language, is the language of the Other,
who inhabits the “other place” of which Freud speaks.
This Other is genetically speaking prior to the I, which
constructs itself by entering into relations with the
world and with itself, through language and by means
of the body; the unreflective and unconscious becomes
conscious and reflective in a series of quite definite
stages, of which the most important in Lacan's work
has been the stade du miroir, or “mirror stage,” which
in young children represents the first apprehension of
their physical unity and autonomy (in its pathological
form it appears as the neurosis of the “fragmented

The apparent personification of the Other obscures
the importance of this discovery, which is double: first,
the structure of the unconscious is presumably common
to all men, idiosyncrasies of the conscious subject
arising from differences in ontogenetic development;
second, this structure is presumably the same as the
structure found underlying language, kinship systems,
mythology, and literature—it writes, ordains, and so
on, in spite of the fact that men have always believed
that they did.


The discomfort at this anthropomorphizing of the
unconscious can be resolved in one of two ways: either
by accepting it and having recourse to the transcendent
(i.e., theologizing) or by challenging the concept of
man that leads to it in the first place. The latter ap-
proach has been taken by Michel Foucault, whose
doctrine of the “end of man,” worked out in Les mots
et les choses
(1966), has awakened the defenders of
humanism just as the defenders of classical drama were
awakened by Barthes. Foucault, whose earlier works
(Histoire de la folie, 1961, and Naissance de la clinique,
1963) had dealt with the abnormal as a negative
touchstone for the human, shows in this book that the
concept of “man” as it has come to be understood in
contemporary humanism is a comparatively recent
invention, called into being as the subject matter of


the sciences humaines and destined to disappear with
the growing realization that these sciences are dealing
only with surface phenomena whose explanation is to
be found elsewhere.

Once again the theme of a deep unconscious struc-
ture emerges: the search for Man as a possible object
of knowledge has in fact produced linguistic structures,
the unconscious, etc., but never a concept of man
capable of bearing the weight that humanism would
place upon it. For humanism requires a concept that
will give meaning to the totality of things not only
now but for the future, while the condition of man
is such that his meanings always come from outside
himself—from elsewhere, from the Other, from the
system of discourse within which he becomes aware
of the possibility of meaning. No totality can be mean-
ingful in any case, except a finished one (a historical
humanism, an intellectual object, a book, some other
product of écriture), which can only be a part of our
present totality and which necessarily has a closed
structure, whereas the “structurality” of our situation
(to borrow a phrase from Jacques Derrida) is open.

Derrida, in a series of very difficult essays, has begun
the task of bringing structuralist philosophy into rela-
tion with the recent history of Western thought, nota-
bly that of Nietzsche and Edmund Husserl. The
Apollo/Dionysius opposition in Nietzsche becomes
paradigmatic of the opposition between retrospective
analysis in terms of différence, on the one hand, and,
on the other, a current and dynamic presence to the
world, involving the notion of différance (sic), which
by an etymological allusion brings in the notion of
temporality (one of the meanings of différer being “to
defer or postpone”) as well as the activity of making
a difference, engaging in praxis rather than theoria. The
philosophical analysis of this immediacy is phenom-
enology, itself based on a double binary structure of
noesis/noema and formal/material. Derrida sees that
there can be no question of replacing the cogito, either
Cartesian or Husserlian, with structure, but that the
more difficult task of reconciling them has to be under-
taken. The central problem is one of self-reference,
structure being what emerges when language is
employed to raise the question of language, when the
subject raises the question of subjectivity, when man
seeks the essence of man.

Another link with philosophical antecedents, already
claimed at least on a terminological level by Lévi-
Strauss, is provided by Louis Althusser's work on Marx.
For Althusser the concept of ideology plays the key
role; it is the structured, unconscious system through
which men relate to their world and within which they
come to awareness of it. Althusser also makes use of
an essentially structuralist method in his treatment of
the relation between Marxist and Hegelian dialectics,
the former being in his view a structural transformation
of the latter rather than the simple inversion found
in the textbooks. But Althusser himself (like Foucault,
Lacan, and others) strenuously denies that he is a
structuralist, and in conclusion it may be worth coming
back to the problem of whether it is defensible to
include these diverse points of view under a common
heading at all.

The objection that is so strongly felt to the title
“structuralist” by so many of those who have been
identified as leaders of the movement has many com-
ponents, but three stand out. First is the belief, already
alluded to, that structuralism is a school or an ideology,
committing anybody who subscribes to it to a package
of ideas and concepts that must be accepted integrally
or not at all, and excluding other possible points of
view or departure (historical, positivist, existentialist,
etc.). Second, and especially understandable in view
of the national origins of the movement, is the fact
that structuralism has become fashionable; everything
not obviously in open contradiction with it, everything
not fully understood, is likely to be called “struc-
turalist” in the French press, and it is natural to resist
the undiscriminating epithets of the crowd. Third is
the further fact that the notion of structure has a
technical use, especially in mathematics and deriva-
tively in linguistics and anthropology, so that purists
may object to its loose extension into fields like litera-
ture and philosophy. Lévi-Strauss, for instance, has
very little patience with critics who go beyond linguis-
tic structuralism.

The first two of these objections can be answered
reasonably. To the extent that other movements are
ideological or scholastic, structuralism may be incom-
patible with them, but since there is as yet no coherent
and worked-out set of propositions to constitute struc-
turalism “officially,” but only a series of suggestive and
mutually reinforcing conjectures whose empirical
justifications are drawn from a number of different
disciplines, many other positions (although not all) may
prove to be compatible with it. Fashions change, and
the value of the structuralist perspective will survive
these changes.

The last objection is harder to deal with, but it may
be pointed out that linguistics and anthropology, the
most technical of the disciplines involved, have them-
selves borrowed only rather simple forms of combina-
tory mathematics, and that while the question why
there should be intelligible structures in logic and
mathematics may eventually permit a structuralist
answer, for the time being it is simply inappropriate
to invoke the mathematical notion of structure in con-
nection with structuralism in the human sciences.


Structuralism so far is at the descriptive level for the
most part, and this provides ample opportunity for
fruitful and enlightening work in a variety of disci-
plines. In philosophy its theories of the mind and of
meaning are beginning to become clear; its logic and
metaphysics remain to be developed.


L. Althusser, Pour Marx (Paris, 1966). R. Barthes, Sur
(Paris, 1936); idem, Le degré zéro de l'écriture, suivi
de Éléments de sémiologie (Paris, 1965); idem, Critique et
(Paris, 1966). N. Chomsky, “Introduction,” in M. Gross
and A. Lentin, Notions sur les grammaires formelles (Paris,
1967); idem, Language and Mind (New York, 1968). J.
Derrida, De la grammatologie (Paris, 1967); idem, L'écriture
et la différence
(Paris, 1967). O. Ducrot, et al., Qu'est-ce
que le structuralisme?
(Paris, 1968). M. Foucault, Les mots
et les choses
(Pairs, 1966). R. Jakobson and C. Lévi-Strauss,
“'Les Chats' de Charles Baudelaire,” L'Homme, 2, 1
(Jan.-Apr. 1962). J. Lacan, Écrits (Paris, 1966). C. Lévi-
Strauss, Les structures élémentaires de la parenté (Paris,
1949); idem, Tristes tropiques (Paris, 1955); idem, Anthro-
pologie structurale
(Paris, 1958); idem, La pensée sauvage
(Paris, 1962); idem, Mythologiques: Vol. I, Le cru et le cuit
(Paris, 1964), trans. John and Doreen Weightman as The
Raw and the Cooked
(New York, 1969); Vol. II, Du miel
aux cendres
(Paris, 1966); Vol. III, L'origine des manières
de table
(Paris, 1968). A. R. Radcliffe-Brown, Structure and
Function in Primitive Society
(London, 1952). F. de Saussure,
Cours de linguistique générale (Geneva, 1915); trans. Wade
Baskin as Course in General Linguistics (New York, 1959;
reprint 1966). N. Troubetzkoy. “La phonologie actuelle,”
in H. Delacroix, Psychologie du langage (Paris, 1933); idem,
Grundzüge der Phonologie (Prague, 1939).

Translations by the author are designated by P. C.


[See also Criticism, Literary; Language; Linguistics; Myth.]