*
   






 
What Is Criticism?

Roland Barthes

It is always possible to prescribe major critical principles in accord with one's ideological situation, especially in France, where theoretical models have a great prestige, doubtless because they give the practitioner an assurance that he is participating at once in a combat, a history, and a totality; French criticism has developed in this way for some fifteen years, with various fortunes, within four major "philosophies." First of all, what is commonly—and questionably—called existentialism, which has produced Sartre's critical works, his Baudelaire, his Flaubert, the shorter articles on Proust, Mauriac, Giraudoux, and Ponge, and above all his splendid Genet. Then Marxism: we know (the argument is already an old one) how sterile orthodox Marxism has proved to be in criticism, proposing a purely mechanical explanation of works or promulgating slogans rather than criteria of values; hence it is on the "frontiers" of Marxism (and not at its avowed center) that we find the most fruitful criticism: Lucien Goldmann's work explicitly owes a great deal to Lukacs; it is among the most flexible and the most ingenious criticism which takes social and political history as its point of departure. And then psychoanalysis; in France today, the best representative of Freudian criticism is Charles Mauron, but here too it is the "marginal" psychoanalysis which has been most fruitful; taking its departure from an analysis of substances (and not of works), following the dynamic distortions of the image in a great number of poets, Bachelard has established something of a critical school, so influential that one might call French criticism today, in its most developed form, a criticism of Bachelardian inspiration (Poulet, Starobinski, Richard). Finally structuralism (or to simplify to an extreme and doubtless abusive degree: formalism): we know the importance, even the vogue of this movement in France since Levi-Strauss has opened to it the methods of the social sciences and a certain philosophical reflection; few critical works have as yet resulted from it, but they are in preparation, and among them we shall doubtless find, in particular, the influence of linguistic models constructed by Saussure and extended by Jakobson (who himself, early in his career, participated in a movement of literary criticism, the Russian formalist school): it appears possible, for example, to develop an entire literary criticism starting from the two rhetorical categories established by Jakobson: metaphor and metonymy.

   As we see, this French criticism is at once "national" (it owes little or nothing to Anglo-American criticism, to Spitzer and his followers, to the Croceans) and contemporary (one might even say "faithless"): entirely absorbed in a certain ideological present, it is reluctant to acknowledge any participation in the critical tradition of Sainte-Beuve, Taine, or Lanson. This last model nonetheless raises a special problem for our contemporary criticism. The work, method, and spirit of Lanson, himself a prototype of the French professor, has controlled, through countless epigones, the whole of academic criticism for fifty years. Since the (avowed) principles of this criticism are rigor and objectivity in the establishment of facts, one might suppose that there is no incompatibility between Lansonism and the ideological criticisms, which are all criticisms of interpretation. However, though the majority of  French critics today are themselves professors, there is a certain tension between interpretive criticism and positivist (academic) criticism. This is because Lansonism is itself an ideology; not content to demand the application of the objective rules of all scientific investigation, it implies certain general convictions  about man, history, literature, and the relations between author and work; for example, the psychology of Lansonism is utterly dated, consisting essentially of a kind of analogical determinism, according to which the details of a work must resemble the details of a life, the soul of a character must resemble the soul of the author, etc.—a very special ideology, since it is precisely in the years following its formulation that psychoanalysis, for example, has posited contrary relations, relations of denial, between a work and its author. Indeed, philosophical postulates are inevitable; Lansonism is not to be blamed for its prejudices but for the fact that it conceals them, masks them under the moral alibi of rigor and objectivity: ideology is smuggled into the baggage of scientism like contraband merchandise.

   If these various ideological principles are possible at the same time (and for my part, in a certain sense I subscribe to each of them at the same time), it is doubtless because an ideological choice does not constitute the Being of criticism and because "truth" is not its sanction. Criticism is more than discourse in the name of "true" principles. It follows that the capital sin in criticism is not ideology but the silence by which it is masked: this guilty silence has a name: good conscience, or again, bad faith. How could we believe, in fact, that the work is an object exterior to the psyche and history of the man who interrogates it, an object over which the critic would exercise a kind of extraterritorial right? By what miracle would the profound communication which most critics postulate between the work and its author cease in relation to their own enterprise and their own epoch? Are there laws of creation valid for the writer but not for the critic? All criticism must include in its discourse (even if it is in the most indirect and modest manner imaginable) an implicit reflection on itself; every criticism is a criticism of the work and a criticism of itself. In other words criticism is not at all a table of results or a body of judgments, it is essentially an activity, i.e., a series of intellectual acts profoundly committed to the historical and subjective existence (they are the same thing) of the man who performs them. Can an activity be "true"? It answers quite different requirements.

   Every novelist, every poet, whatever the detours literary theory may take, is presumed to speak of objects and phenomena, even if they are imaginary, exterior and anterior to language: the world exists and the writer speaks: that is literature. The object of criticism is very different; the object of criticism is not "the world" but a discourse, the discourse of someone else: criticism is discourse upon a discourse; it is a second language, or a metalanguage (as the logicians would say), which operates on a first language (or language object). It follows that the critical language must deal with two kinds of relations: the relation of the critical language to the language of the author studied, and the relation of this language object to the world. It is the "friction" of these two languages which defines criticism and perhaps gives it a great resemblance to another mental activity, logic, which is also based on the distinction between language object and metalanguage.
   For if criticism is only a metalanguage, this means that its task is not at all to discover "truths," but only "validities." In itself, a language is not true or false, it is or is not valid: valid, i.e., constitutes a coherent system of signs. The rules of literary language do not concern the conformity of this language to reality (whatever the claims of the realistic schools), but only its submission to the system of signs the author has established (and we must, of course, give the word system a very strong sense here). Criticism has no responsibility to say whether Proust has spoken "the truth," whether the Baron de Charlus was indeed the Count de Montesquieu, whether Francoise was Celeste, or even, more generally, whether the society Proust described reproduces accurately the historical conditions of the nobility's disappearance at the end of the nineteenth century; its role is solely to elaborate a language whose coherence, logic, in short whose systematics can collect or better still can "integrate" (in the mathematical sense of the word) the greatest possible quantity of Proustian language, exactly as a logical equation tests the validity of reasoning without taking sides as to the "truth" of the arguments it mobilizes. One can say that the criticial task (and this is the sole guarantee of its universality) is purely formal: not to "discover" in the work or the author something "hidden," "profound," "secret" which hitherto passed unnoticed (by what miracle? Are we more perspicacious than our predecessors?), but only to adjust the language his period affords him (existentialism, Marxism, psychoanalysis) to the language, i.e., the formal system of logical constraints elaborated by the author according to his own period.  The "proof" of a criticism is not of an "alethic" order (it does not proceed from truth), for critical discourse—like logical discourse, moreover—is never anything but tautological: it consists in saying ultimately, though placing its whole being within that delay, what thereby is not insignificant: Racine is Racine,  Proust is Proust; critical "proof," if it exists, depends on an aptitude not to discover the work in question but on the contrary to cover it as completely as possible by its own language.

   Thus we are concerned, once again, with an essentially formal activity, not in the esthetic but in the logical sense of the term. We might say that for criticism, the only way of avoiding "good conscience" or "bad faith" is to take as a moral goal not the decipherment of the work's meaning but the reconstruction of the rules and constraints of that meaning's elaboration; provided we admit at once that a literary work is a very special semantic system, whose goal is to put "meaning" in the world, but not "a meaning"; the work, at least the work which ordinarily accedes to critical scrutiny—and this is perhaps a definition of "good" literature—the work is never entirely non signifying (mysterious or "inspired"), and never entirely clear; it is, one may say, a suspended meaning: it offers itself to the reader as an avowed signifying system yet withholds itself from him as a signified object. This disappointment of meaning explains on the one hand why the literary work has so much power to ask the world questions (undermining the assured meanings which ideologies, beliefs, and common sense seem to possess), yet without ever answering them (there is no great work which is "dogmatic"), and on the other hand why it offers itself to endless
decipherment, since there is no reason for us ever to stop speaking of Racine or Shakespeare (unless by a disaffection which will itself be a language): simultaneously an insistent proposition of meaning and a stubbornly fugitive meaning, literature is indeed only a language, i.e., a system of signs; its being is not in its message but in this "western." And thereby the critic is not responsible for reconstructing the work's message but only its system, just as the linguist is not responsible for deciphering the sentence's meaning but for establishing the formal structure which permits this meaning to be transmitted.
  It is by acknowledging itself as no more than a language (or more precisely, a metalanguage) that criticism can be—paradoxically but authentically—both objective and subjective, historical and existential, totalitarian and liberal. For on the one hand, the language each critic chooses to speak does not come down to him from Heaven; it is one of the various languages his age affords him, it is objectively the end product of a certain historical ripening of knowledge, ideas, intellectual passions— it is a necessity; and on the other hand, this necessary language is chosen by each critic as a consequence of a certain existential organization, as the exercise of an intellectual function which belongs to him in his own right, an exercise in which he puts all his  profundity," i.e., his choices, his pleasures, his resistances, his obsessions. Thus begins, at the heart of the critical work, the dialogue of two histories and two subjectivities, the author's and the critic's. But this dialogue is egoistically shifted toward the present: criticism is not an "homage" to the truth of the past or to the truth of "others" —it is a construction of the intelligibility of our own time.

 1963