In Barthes's view, we are perpetually caught up, at every moment of our experience, by a mesh of words that prevents us from seeing what is really happening…. [We] perpetually see life in terms of the books we have read, and have quite lost the ability to see physical objects as they actually are. In so far as it ties us down to a predigested version of the way somebody else first saw the world and expressed it for us, this habit prevents us from realising our full potential as free human beings. It is consequently—though here I am extrapolating from Barthes's work, not referring to any formal statement which he has made—the task of the person who writes either about literature or about language to make people conscious of the distortions created by the way verbal communication works. The missionary role thus entrusted to the linguist or literary critic constitutes the most important conclusion which Barthes has drawn from Saussure's insistence on the arbitrary nature of signs, and provides both the central theme linking the whole of his work together and his most significant contribution to the intellectual life of the mid to late twentieth-century. (pp. 136-37)

Barthes is more a philosopher of language than a literary critic, and there is therefore some justification for his work being so difficult to understand. I would nevertheless maintain that to see him as a man determined to free people from preconceived ideas by pointing carefully at the individual strands in the mesh of language holding them captive is to look at his work from the most fruitful point of view. He described himself, in Roland Barthes par Roland Barthes [Roland Barthes by Roland Barthes], as a man who sees language, and each one of his books can be read as an attempt to make people conscious of how completely the way we express ourselves conditions our vision of the world. Le degré zéro de l'écriture denounces the illusion that any form of literary language can be natural, and warns against the trap of assuming that because an account of experience is immediately comprehensible, it is therefore innocent of pre-conceived ideas. Mythologies is shot through with an insistence on the need to avoid 'la naturalité du signe' by constantly emphasising the artificial nature of all communication systems, while Système de la Mode is dominated by the realisation that we can be honest with our fellow human beings and ourselves only by seeing the clothes we wear as consciously expressing the deliberate choice which we make of how we would like other people to see us. Sade, Fourier, Loyola carries Barthes's position in Le degré zéro, Mythologies and S/Z to its logical conclusion by recognising that the role of language is to enable the writer to create his own autonomous world, while Le Plaisir du Texte uses the neologism signifiance to embody the notion that it is only when we have freed ourselves of the illusion that words reflect reality—or even, perhaps, that there is a reality to be reflected—that we shall begin to see language as creating our awareness of meaning by the physical impact which it makes upon our senses. (pp. 138-39)

Barthes does, of course, make a genuinely original contribution to literary discourse by the implications which he sees in the fact that there is no necessary connection between the word 'rose' and the flower it signifies. His ambition to 'battre en brèche la naturalité du signe' ['attack and destroy the idea that signs are natural'] is both laudable and intelligible, and his application of it to the novel is a valuable step in clear thinking about how language functions in a literary context. Frederic Raphael expressed the same notion from the other side when he said that he realised, as a working novelist, that what you say cannot finally be distinguished from the way you say it, and his remark underlines how foolish it would be to think, for example, that Pickwick or Emma Bovary would not change as characters if their adventures were described in different language. (p. 142)

It is nevertheless more difficult, both on logical and on empirical grounds, to follow Barthes in the next step which he takes in applying the idea of the arbitrary nature of signs to literature. For while there is no difficulty in using it to reinforce one's rejection of the naïve notion that Madame Bovary is a good novel because it tells the literal truth about a particular person, it is quite another thing to argue, as Barthes does in S/Z, that there is never any 'content' whatsoever in works of art. Flaubert may not be the master of representational art that some of his theories suggest and that many of his nineteenth-century admirers considered him to be…. But to deny the interest and validity of all attempts at representational art, as Barthes does when he writes that realism 'consists not of copying reality but of copying a painted copy of reality', is to throw out the baby of Flaubert's complex literary achievement with the bathwater of late nineteenth-century theories about the inevitability of realism. (p. 143)

It is one of the more embarrassing peculiarities of Barthes's work that he should inspire … laboured restatements of the obvious, and it is partly for this reason that the refusal to call him a critic is not always entirely a compliment. For it is surely—to invoke a central Doxa [Barthes's term for bourgeoise thinking] of English literary thinking—an openness to as many different kinds of literary experience as possible that characterises the good literary critic and is essential to the great one. One might, of course, explain this aspect of his work by biographical considerations. Barthes constantly evokes, in the remarks which he makes about contemporary French society and the intellectual atmosphere in which he is obliged to live, the vision of a vast conspiracy aimed at perpetuating a whole series of oppressive myths against which he is fighting. The French bourgeoisie, if one is to believe in what he says about it, still uses language to perpetuate its domination over other classes, still maintains a strict hierarchy of linguistic usage to match the social hierarchy which it imposes upon other people, severely censures both the frivolous use of language and any cult of physical pleasure, is intensely suspicious of any intellectual activity … and refuses to entertain the possibility that the link between the thing said and the word which says it is anything but wholly natural. It is perhaps this feeling which he has of being persecuted that explains why Barthes should himself adopt so intolerant a stance as a means of defending himself against his attackers, and there is again, for the English reader, a curious biographical similarity to F. R. Leavis. Yet while both men have the same sense of being cold-shouldered by the academic establishment of their day, there is much less justification for Barthes to see himself as a martyr in the cause of intellectual progress. He has not done all that badly and has in fact, as John Weightman predicted, ended up as a member of the Collège de France—'than which there is no higher consecration here below'. The intolerant strain that is often apparent in Barthes's attitude towards literature cannot therefore be explained away as a justified riposte to genuine oppression, and is, to the conservative foreign observer, a serious flaw in his work. This is even more the case when it is accompanied by the second reason for which one would hesitate to flatter Barthes by calling him a critic: an apparent inability to distinguish between the prescriptive and the descriptive use of language.

Thus he frequently presents, as statements about writers or about literature in general, ideas which are fascinating when read as suggestions as to how a new type of literature might develop or as descriptions of certain books which exemplify this development, but which are patently absurd when read in the highly general form that he gives them. For example, he writes in the essay Ecrivains et Ecrivants,… that since 'écrire est un verbe intransitif' ['writing is an intransitive verb'], 'la littérature est toujours irréaliste' ['literature is always unrealistic'], and there is obviously a way in which this statement is a valuable antidote to the naïve concept of realism which presents books as photocopies of reality. Even if it were possible for a writer to be a camera, the angle from which he took his shots would still mean that he was composing a picture rather than reproducing the world absolutely as it is. But since there is no indication that Barthes is writing metaphorically rather than literally, the other implications of his statement are decidedly odd. The 'toujours' can only mean that Thomas Hardy tells us nothing about rural England before the impact of the industrial revolution, that Jane Austen's novels have no interest as a portrait of the English gentry at the time of the Napoleonic wars, or that the characters in War and Peace bear no resemblance to early nineteenth-century Russian aristocrats. (pp. 145-46)

[In] so far as Barthes's own ambition in the rest of his work can so often be seen as that of trying to 'purify the dialect of the tribe' by inviting his readers to become aware of the prison in which certain forms of language have enclosed them, it might … be a rather Barthesian exercise to see how his views on literature do appear when expressed in a language different from his own. On the other hand, however, since it implies both that language is a tool for expressing ideas and that ideas remain basically the same whatever the language chosen to express them may be, such an exercise could be seen as a complete denial of everything Barthes stands for: One of the clearest statements of his position in this respect is in a reply which he made in March 1957 to a survey organised by the literary newspaper Arts on the question of why Joseph Conrad had chosen to write in English rather than in Polish. It was not, said Barthes, because English was a better medium than Polish for writing about the sea. Indeed, the very act of asking such a question showed a fundamental misunderstanding of the very nature of language. This is not an 'instrument that one chooses as one might choose a weapon in an armoury or a monkey wrench from a tool kit'. Language, and particularly the language used by an imaginative writer, is 'a structure and a mode of awareness' ['une structure et une conscience']. The decision to write in one language rather than another cannot be separated from the choice which one has made of one's whole identity as a person, and from one's fundamental attitude towards experience. When Conrad decided to write in English rather than in Polish, he was acting in accordance with a deeper, existential preference which he had already expressed for what Barthes calls la britannité over what one would, in a comparable neologism, have to call la polonnité. It follows from this that any attempt to 'say what Barthes really means in your own words' is by very definition a betrayal of his whole philosophy of language. It is, to revert to the story about the churchgoing Scottish housewife and her heathen if inquisitive husband, the equivalent of trying to talk about St. Paul while eschewing all use of terms such as 'God', 'grace', 'sin', 'salvation' or 'spirit'. The temptation nevertheless exists, and Barthes's own francité is too intriguing a challenge to the intellectual imperialism of English empiricism to be seriously resisted…. [Barthes] is still sufficiently interesting as what he himself would call an écrivant—someone who does use language instrumentally—for the way he thinks to be discussed in isolation from the way he writes. Although he may be most successful when writing as an écrivain, giving full rein to his zest for language by evoking rather than communicating the world view of the designer Erté or the pleasures of the table enumerated in Brillat-Savarin, he cannot prevent his basic attitude from being sufficiently coherent to be meaningfully discussed with reference to another cultural tradition.

Indeed, the very way in which he chooses to write is a means of ensuring that this will happen. Like Michel Foucault, Jacques Derrida, Julia Kristeva or Jacques Lacan, he deliberately adopts a style of writing which is not immediately comprehensible, and does so precisely in order to make his reader think things out for himself. This is not merely a condescendingly charitable explanation of his often very opaque prose. Instant accessibility is, in his view, a dangerous trick, and Stephen Heath quotes a very characteristic passage from the 1973 text Aujourd'hui, Michelet to illustrate this idea. The concept of clarity [clarté], writes Barthes, can exist only within 'a classical conception of the sign, with the signifiant on one side, and the référent on the other, the first in the service of the second'. The whole of Barthes's career so far has been devoted to overthrowing this idea and replacing it with a vision of language in which man, recognising that he lives, moves and has his being in and through words, can at last begin to enjoy the experience for its own sake. (pp. 148-49)