[In the following essay, Knight discusses Barthes's notion of Utopia as presented in several of his works, stressing that it “is a central—and highly conscious—preoccupation” for him.]

Civilization/Harmony: thus did Charles Fourier, archetypal Utopian socialist, polarize wretched contemporary society and unutterably blissful new world. The binary works well for Barthes's part-political, part-ethical, part-aesthetic analyses of his own society and culture; indeed I wish to explore Utopia as the meeting point of his lifelong concern with the problems of history, language, literature, criticism and power. Apart from such obviously relevant texts as his essay on Fourier and Empire of Signs, a surprising proportion of Barthes's writings make their points through a vocabulary of Utopia as both adjective and proper noun. The scope of Barthes's preoccupation with Utopia is indicated in his 1977 inaugural lecture: ‘Utopia, of course, does not save us from power. The Utopia of language is recuperated as the language of Utopia—a genre like the rest’ (1983: 467). On the one hand Barthes refers positively to a ‘Utopia of language’ which embraces an ideal conception of literature and writing. On the other ‘the language of Utopia’, Utopia as a recuperable genre, seems here to be treated with suspicion. It is the aim of this paper to suggest nevertheless that Barthes will find, in overtly Utopian discourse, a solution to some persistent ethical and theoretical dilemmas.¹

The earliest conception of a Utopia of language is quite simply Literature with a capital L. This idea is launched in the heavily existential closing paragraph of Writing Degree Zero:

literary writing carries at the same time the alienation of History and the dream of History; as a Necessity, it testifies to the division of languages which is inseparable from the division of classes, as Freedom, it is the consciousness of this division and the very effort which seeks to surmount it. Feeling permanently guilty about its own solitude, it is none the less an imagination eagerly desiring a felicity of words, it hastens towards a dreamed-of language whose freshness, by a kind of ideal anticipation, might portray the perfection of some new Adamic world where language would no longer be alienated. The proliferation of modes of writing (écritures) brings a new literature into being in so far as the latter invents its language only in order to be a project: Literature becomes the Utopia of language.

(1970: 93-4)

By the unaugural lecture Barthes's social vision has certainly changed. The transparency of social and linguistic relations no longer has anything to do with universality—indeed the latter has been gradually replaced by an affirmation of infinite, irreducible difference. However Barthes is still aligning the Utopian function of literature with the ‘forces of freedom’. These are located not in literature's content but again in its language, more precisely in ‘the labour of displacement’ that the writer brings to bear on language. This labour is equated with a salutory trickery: since language and speech equal power and aggression, and since there is no exit from language, the writer's solution is to use literature to cheat speech of its meanings.

Barthes's view of man as condemned to meaning depends upon his Saussurean view of language as a sign system which forces all would-be statements about the world to pass through the stage of mental representation, that is through the relay of the signified.² It is at this stage that man gets caught up in the warring systems of meaning that Barthes sees as alienating social relations. As mythologist Barthes accepts the political necessity of engagement with this war of meanings, an engagement that takes the well-known form in his work of unmasking the historical interests that masquerade as natural states of affairs. But Barthes always yearned—and all of his writing testifies to this—for a Utopian realm beyond meaning, and Utopia is normally his word for this realm. The beyondness of the realm is crucial, for Barthes is careful to distinguish the projected après-sens from a regressive nostalgia for some pre-semiological Golden Age. To abolish or ignore meaning would be semiologically and politically naive—instead Barthes describes the desired outplaying of meaning in terms of its suspension and occasionally of its theft. Thus an essay on Robbe-Grillet toys with literature's potential for ‘unexpressing’ and ‘designifying’ the world that we discover already imbued with meaning, and his fascination with the uncertain irony of Flaubert's writing is fixed on its solution to his own uneasy relationship with the doxa—his concern with how to outplay the codes without proclaiming oneself exempt from stupidity. This is indeed a problem for Barthes, which he struggles to find a way through:

it is possible to enjoy the codes even while nostalgically imagining that someday they will be abolished: like an intermittent outsider, I can ‘enter-into’ or ‘step-out-of’ burdensome sociality, depending on my mood—of insertion or of distance. (1977a: 131) whence a double tactic: against Doxa, one must come out in favour of meaning, for meaning is the product of History, not of Nature; but against Science (paranoiac discourse), one must maintain the utopia of suppressed meaning.

(87)

The enthusiasm for Zen Buddhism that permeates Empire of Signs is totally comprehensible in the context of this search for an after or a beyond of meaning. Indeed the very rigours of the Zen training—the long meditations on absurd riddles and anecdotes to find a way beyond language and logic—testify to Barthes's conviction of the enormous difficulty of the task:

All of Zen, of which haiku is merely the literary branch, thus appears as a vast set of observances (une immense pratique) destined to halt language, to jam that kind of internal radiophony constantly sending in us, even in our sleep (perhaps this is the reason the apprentices are kept from falling asleep), to empty out, to stupefy, to dry up the soul's incoercible babble; and perhaps what Zen calls satori, which Westeners can translate only by certain vaguely Christian words (illumination, revelation, intuition), is no more than a panic suspension of language, the blank which erases in us the reign of the Codes, the breach of that internal recitation which constitutes our person; and if this state of a-language is a liberation, it is because, for the Buddhist experiment, the proliferation of secondary thoughts (the thought of thought), or what might be called the infinite supplement of supernumerary signifieds—a circle of which language itself is the depository and the model—appears as a jamming: it is on the contrary the abolition of secondary thought which breaks the vicious infinity of language.

(1982: 74-5)

Not only does Barthes find in Zen an ideal in keeping with his ethic of the sign, but he finds in haiku a literary solution which seems to surpass all others in his eyes, and versions of which he tries out himself, notably in the Japan book itself, in Roland Barthes by Roland Barthes (in the section of recollections (anamnèses), and in the Nouvel Observateur chronicles (chroniques) of 1978 and '79. The attraction of haiku, that Barthes chooses to restress in an interview on Empire of Signs, is that it achieves its exemption from meaning through a perfectly readerly discourse (whereas Western modern art tends to attempt this by making discourse incomprehensible). Haiku is celebrated for its lack of Western description and definition, for its delicate designation of objects and events without attribution of meaning, for its ability to outstrip connotation. But for Barthes, and for Western discourse in general, this remains at best a Utopian aspiration.

I find it particularly interesting that Barthes's vision of Japan as a semiological Utopia should be published in the same year as S/Z (thus turning 1970 into a landmark year in Barthes's career). If the appeal of Zen is its intuition of something beyond connotation, beyond the codes, and beyond what Barthes calls ‘the vicious infinity of language’, it might seem paradoxical that the contemporary S/Z should take up position within the codes, and more specifically within a deliberate multiplication of connotations. An important section of the essay ‘Writers, Intellectuals, Teachers’, published the following year in 1971, might serve as a comment on this parallel publication of two works almost complementary in their extreme difference from each other. Barthes starts from two sorts of typing error: one which produces a non-existent nonsense word, and one which produces an unintended but recognizable word which thereby creates strange reverberations in the sentence that contains it. From this he draws two types of criticism, and pauses to reflect on the different historical stake of each. In so doing he poses clearly the political problem of one sort of Utopia of language—can language really be liberated ahead of social relations, and, if so, what is at stake in jumping ahead of history in this way?

The first has in its favour the right of the signifier to spread out where it will (…) Why stop? Why refuse to push polysemy as far as asemy? In the name of what? Like any radical right, this one supposes a utopian vision of freedom: the law is lifted all at once, outside of any history, in defiance of any dialectic (hence the finally petit bourgeois aspect of this style of demand) (…) By liberating reading from all meaning, it is ultimately my reading that I impose, for in this moment of History the economy of the subject is not yet transformed and the refusal of meaning (of meanings) falls back into subjectivity.

(1983: 395)

Attempts to escape meaning through self-indulgent and limitless expansion of the signifiers are apparently thrown over in favour of what Barthes identifies as a historically more correct move, a move that is clearly that of S/Z:

Thus the second type of criticism, that which applies itself to the division of meanings and the ‘trickery’ of interpretation, appears (at least to me) more historically correct. In a society locked in the war of meanings and therefore under the compulsion of rules of communication which determine its effectiveness, the liquidation of the old criticism can only be carried forward in meaning (in the volume of meanings) and not outside it. In other words, it is necessary to practise a certain semantic enterism. Ideological criticism is today precisely condemned to operations of theft: the signified, exemption of which is the materialist task par excellence, is more easily lifted in the illusion of meaning than in its destruction.

(396)

What, then, are we to make of Barthes's parallel elaboration of the concept of Text, frequently presented as a full-blown Utopia, which in his own terms would appear to equal a petit-bourgeois stepping outside of history? (This being, of course, Engels's criticism of Utopian socialism.) Perhaps one might credit Barthes with having paid his historical dues, and done his bit for semantic enterism, with the surely unrepeatable feat of S/Z. Equally we might agree with Annette Lavers's suggestion that Barthes was simply exhausted by the responsibility of forms. More charitably, we might develop Barthes's own cautionary ‘at best’:

At best, one can simply say that this radical criticism, defined by a foreclosure of the signified (and not by its slide (sa fuite)), anticipates History, anticipates a new, unprecedented state in which the efflorescence of the signifier would not be at the cost of any idealist counterpart, of any closure of the person.

(395)

If language is the site of power and of warring meanings, any notion of language freed from connotation and metalanguage could be said to have a peaceable and political dimension from that point of view at least. Indeed once Barthes has defined Text as a social space the appropriateness of the Utopia metaphor becomes more apparent:

As for Text (le texte), it is bound to jouissance, that is to pleasure without separation. Pertaining to the signifier (ordre du signifiant), Text participates in its own way in a social utopia; before History (supposing the latter does not opt for barbarism), Text achieves, if not the transparence of social relations, that at least of language relations: Text is that space where no language has a hold over any other (…) that social space which leaves no language safe, outside, nor any subject of enunciation in position as judge, master, analyst, confessor, decoder.

(1977b: 164)

Thus Barthes attempts to hold the reins of History and Utopia all in one hand, as he explains in a 1973 conference paper:

We have to commit ourselves, to participate in one of the individual languages into which our world and our history force us. And yet we can't give up the jouissance, no matter how utopian, of a displaced and de-alienated language. We must therefore grasp in the same hand the two reins of commitment and of jouissance, we must assume a plural philosophy of languages. Now this elsewhere that remains, I'd like to say inside, has a name: it's called Text (…) Writing is atopical (a-topique); relative to the war of languages, which it doesn't wipe out but which it displaces, it anticipates a particular state of the practice of reading and writing, where it's desire which circulates, and no longer domination.

(1986: 109-10)

If Text and écriture are therefore important parts of a politically grounded Utopia of language, it is the extension of these concepts beyond the written word to the writing and reading of reality that permits the development of more literal Utopias. For example the café is described in a 1975 interview as a ‘complex, stereophonic space’ (1981: 209-10), the site of a textual experience triggering off what Barthes calls the novelistic (le romanesque), while a fragment of the essay on Brillat-Savarin develops the notion of the dinner table as another complex space, one of overdetermined social pleasure, where convivial conversation combines with the anticipated pleasures of eating to introduce jouissance into the very act of communication (1984b: 303). The most developed mini-Utopia is Barthes's celebration of his own teaching seminar in the 1974 essay called, in French, ‘Au séminaire’ (both at the seminar and dedicated to the seminar). Here Barthes develops a parallel with Fourier's phalanstery, describing the intellectual and affective interchange of the seminar as the tracing of a text in its own right—‘the most precious of texts, that which does not pass through the written word’ (1986: 333).

The transition to something like a generically based Utopian writing happens in the book about Japan, from which point Barthes's relentless ethics of the sign is increasingly accompanied by an ethics of social, affective relations, by a heightened interest in the quality of everyday life, and by an escalating obsession with the possible mode of writing that he distinguishes from the novel by calling it the novelistic (roman/romanesque). All this can be read without difficulty in Empire of Signs itself, and is spelled out particularly clearly in a contemporary interview. Japan represents for Barthes a world that is both strictly semantic (nothing escapes signs) and yet strictly atheistic:

The ethical presence of feudalism maintains in this strictly technologized society—not really Americanized—a body of values, a life style (un art de vivre), which is probably rather fragile from an historical point of view, and which must also be linked with the fundamental absence of monotheism. A system which is almost totally immersed in the signifier works in conjunction with a perpetual withdrawal of the signified: this is what I tried to show at the essential level of everyday life (as much for food as for housing (l'habitat), for make-up and for the system of addresses (…) This book is something of an entry, not into the novel, but into the novelistic.

(1985: 83-4)

Empire of Signs is an appealing and happy Utopia, which builds in the passion for the concrete details of ‘otherness’ that Barthes finds in Stendhal's love of Italy, and which he describes as a sort of opposite of racism (1984b: 334). In that Barthes's representation of Japan achieves a complete reversal of all that chokes him in his own civilization (of the rhetorical modes and everyday myths so famously denounced in Mythologies), it belongs to one particular tradition of Utopian writing, and it might therefore be interesting to explore ways in which Barthes builds in the West as the indirect term of comparison. I am thinking in particular of his wonderful device for connecting the two worlds, the twin pictures of the newspaper item on Barthes himself in Japan and the photo of the Japanese actor—while Barthes is ‘nipponified’ by the poor print of the newspaper, the smooth actor takes on the qualities of a thoroughly Western film star. But this movement from one society to the other is in no way inscribed in a historical progression, and the opening sections of the text spell out clearly enough that it is only because Barthes is a temporary visitor and a non-speaker of Japanese that he is able to maintain his Utopian reading of Japan. Only this special status allows him to write as if in ignorance of the fact that the ordinary citizen of Tokyo will certainly be choking in his own struggles with work, smog and the Japanese doxa. Empire of Signs is a superb actualization of Barthes's personal semiological ethic, but it is one that certainly loses the properly political dimension of Utopia.³

To refind that dimension, which resurfaces so crucially in the essay on Fourier published in Sade, Fourier, Loyola the following year (but first written the same year), I want to turn back the clock and to look at the soul-searching ending to Barthes's 1957 essay ‘Myth Today’, the theoretical afterword to Mythologies. Barthes's backward look at his mythologies raises two important problems, and I should like to suggest that the Fourier essay seems to have found a solution to both of them. The first problem is that Utopia is simply an impossible luxury for the mythologist:

In a sense, the mythologist is excluded from this history in the name of which he professes to act (…) It is forbidden for him to imagine what the world will concretely be like, when the immediate object of his criticism has disappeared. Utopia is an impossible luxury for him: he greatly doubts that tomorrow's truths will be the exact reverse of today's lies. History never ensures the triumph pure and simple of something over its opposite: it unveils, while making itself, unimaginable solutions, unforseeable syntheses. The mythologist is not even in a Moses-like situation: he cannot see the Promised Land. For him, tomorrow's positivity is entirely hidden by today's negativity.

(1983: 147-8)

The second problem is that of the disturbance that ideological demystification introduces into Barthes's relationship with reality:

One last exclusion threatens the mythologist: he constantly runs the risk of causing the reality which he purports to protect, to disappear (…) There is as yet only one possible choice, and this choice can bear only on two equally extreme methods: either to posit a reality which is entirely permeable to history, and ideologize; or, conversely, to posit a reality which is ultimately impenetrable, irreducible, and, in this case, poetize. In a word, I do not yet see a synthesis between ideology and poetry (by poetry I understand, in a very general way, the search for the inalienable meaning of things).

(148-9)

The sense of a solution to both problems can be found in the 1975 Roland Barthes by Roland Barthes, in the section headed ‘The purpose of Utopia’ (A quoi sert l'utopie). The first point made is that Utopia enables Barthes to find a way of speaking his world:

Confronting the present, my present, utopia is a second term that permits the sign to function: discourse about reality becomes possible, I emerge from the aphasia into which I am plunged by the panic of all that doesn't work within me, in this world which is mine.

(1977a: 76)

The second point reveals a crucial move from seeing Utopia as an impossible luxury to the realization that it is a taboo:

Revolutionary writings have always scantily and poorly represented the daily finality of the revolution, the way it suggests we shall live tomorrow, either because this representation risks sweetening or trivializing the present struggle, or because, more precisely, political theory aims only at setting up the real freedom of the human question, without prefiguring any of its answers. Hence utopia would be the taboo of the revolution, and the writer would be responsible for transgressing it; he alone could risk that representation; like a priest, he would assume the eschatological discourse; he would close the ethical circle, answering by a final vision of values to the initial revolutionary choice (the reason why one becomes a revolutionary).

(77)

Once Utopian discourse is conceived as a taboo (as a proscribed discourse), its transgression can be assumed by Barthes as a political virtue in its own right, and Barthes's bravest definition of Utopia in all of his work is perhaps the 1972: ‘Utopia is the state of a society where Marx would no longer criticize Fourier’ (1985: 171)—that is, where Utopia would no longer be a revolutionary taboo.

A short piece published in 1974 in Italian elaborates this appropriate response to Engels's famous critique of Utopian socialism (so dubbed in opposition to the scientific socialism of Marxism). It emphasizes that the importance of Utopian discourse lies not only in its determination to write future values, but equally in its will to project concrete representations of everyday life in the ideal society, in particular through its fantasmatic attention to tiny details:

Utopia is the sphere of desire unlike politics which is the sphere of need. Whence the paradoxical relations between these two discourses: they complement each other but never understand each other. Need disapproves of the irresponsibility and futility of desire: desire disapproves of need's censoring and of its reductive drive (…) Desire should be reinjected into Politics, which is to say that Utopias are not only justifiable, they are also necessary. It's not the outlines of a future society that we are afraid to draw. That society is already there, in politics itself; it's the details of this society, and it's from these that we deduce Utopia, that we deduce desire. For Utopia, and this is precisely its special feature, imagines times, places and customs in minutest detail; it's novelistic, it's none other than the political form of fantasy. Utopia is always ambivalent. It stresses relentlessly the wrongs of the present world, and, at the same time, with the same force, it invents pictures of happiness. It invents them in their special colours, with their own precision and variations, with their own absurdity; it possesses the rarest sort of courage, that of enjoyment, and this is the courage that two of our greatest Utopian thinkers, Sade and Fourier, have shared (…) Fourier's phalanstery and Sade's castle are literally impossible, but the detailed inflections of the Utopian system return to our world like lamps of desire, of possible exultation. If we could grasp them better, they would prevent politics from solidifying into a totalitarian, bureaucratic, moralizing system.

(1974: 259)

In the first of his Sade essays Barthes claims that both Sade and Fourier's Utopias are measurable less against theoretical statements than against the organization of everyday life: ‘for the mark of Utopia is the everyday; or even: everything everyday is utopian’ (1977c: 17). To understand this last comment it is helpful to look at the important distinction that Barthes draws (in a 1975 interview) between le politique and la politique (the political and politics) (1985: 218). The political, he claims, is a fundamental order of history, of thought, of all that is done and spoken. It's the very dimension of the real. Politics, on the other hand, is the moment of the political's conversion into an endlessly churned out discourse of repetition. His attachment to the political is profound, whereas he is simply irritated by the discourse of politics. It's not, he claims, a case of turning his back on politics, but of trying to find an acceptable way of assuming the discourse of the political.

I should like to suggest that Utopian discourse is precisely the appropriate way for Barthes of speaking this fundamental category of reality that he calls the political—that it avoids the political discourse that he finds oppressive even when it comes from the left, yet permits the unalienated complicity with reality which had seemed an impossibility for the lonely 1950s mythologist. Barthes's frequent references right through the 1970s to the category of the novelistic elaborate this more acceptable relation to reality, a relation that Barthes increasingly dares to assume:

The novelistic is a mode of discourse which isn't structured as a narrative; it's a mode of notation, of investment and of interest in daily life, in people, and in all that goes on in life.

(1985: 222)

In daily life, I feel a sort of curiosity about everything that I see and hear, an almost intellectual affectivity which belongs to the order of the novelistic. A hundred years ago, I would probably have walked through life with a realist novelist's notebook in my hand. But I can't see myself inventing a story with characters with proper names, I can't see myself writing a novel. My problem—a future problem since I should very much like to do something in that direction, will be to find a form that will separate the novelistic off from the novel, but that will assume the novelistic far more profoundly than I have done to date.

(1985: 203)

Does this not spell out that the novelistic, which apparently subsumes the political, is close indeed to Utopian discourse?⁴

If the genre of Utopian writing turns out to be Barthes's Utopian genre, then his dense and suggestive essay on Fourier is surely the master text for understanding this important aspect of Barthes's final period. If the major themes of Barthes's work (history, language, power, the body⁵), intersect in Utopia, all are visibly in play in the Fourier analysis. Let me list (in a feeble pastiche of Barthes's own occasional style), some of the features that Barthes prizes in Fourier. They are, at random: the subsuming of political science in the far more important science of Domestics, the supremacy of pleasure, the refusal of notions of normality, the respect for the irreducible difference of the 810 character types, the eccentric attention to sensual detail, the manic classifications and bizarre but logical combinations, the love of neologism, the uncertain status of the discourse, the dilatory style whereby Fourier keeps promising the really important exposition of his theory for later. If the whole detailed invention of a closed and regulated space recalls for Barthes the happy and even choosable life-style of the sanatorium, all of these features have evident reverberations in Barthes's own idiosyncratic values for living and perhaps more especially for writing.

Barthes's reading of Fourier through a newly invented binary system/systematics (système/systématique) is surely a comment on the desired relationship between system-building and system-writing in Barthes's own work, in particular the way he tries to account in retrospect for his high-structuralist era: Fourier's ‘system’ is described as that part of his ‘systematics’ which ‘plays with the system in an imaginary way’ (1977c: 110) (qui joue imaginairement au système), ‘imaginary’ presumably to be taken in the Lacanian sense. Thus Barthes introduces this particular section of his essay by aligning systematics and system with those distinctions already established in S/Z between novelistic and novel, essay and dissertation, production and product, structuration and structure. Barthes's crucial point is developed in some detail:

The system being a closed (or monosemic) one, it is always theological, dogmatic; it is nourished by illusions: an illusion of transparency (the language employed to express it is purportedly purely instrumental, it is not a writing) and an illusion of reality (the goal of the system is to be applied, i.e. that it leave the language in order to found a reality that is incorrectly defined as the exteriority of language); it is a strictly paranoid insanity whose path of transmission is insistence, repetition, catechism, orthodoxy. Fourier's work does not constitute a system; only when people tried to ‘realize’ this work (in phalansteries) did it become, retrospectively, a ‘system’ doomed to instant fiasco; system, in the terminology of Marx and Engels, is the ‘systematic form’, i.e. pure ideology, ideological reflection: systematics is the play of the system; it is language that is open, infinite, free from any referential illusion (pretension); its mode of appearance, its constituency, is not ‘development’ but pulverization, dissemination (the gold dust of the signifier); it is a discourse without ‘object’ (it only speaks of a thing obliquely, by approaching it indirectly: thus Civilization in Fourier) and without ‘subject’ (in writing, the author does not allow himself to be involved in the imaginary subject, for he performs his enunciatory role in such a manner that we cannot decide whether it is serious or parody). It is a vast madness which never shuts (ne ferme pas), but which permutates. (…)

(Fourier puts the system to flight—cuts it adrift—by two operations: first, by incessantly delaying the definitive exposé until later: the doctrine is simultaneously highhanded and dilatory; next, by inscribing the system in the systematics, as dubious parody, shadow, game (…))

(109-10)

System and systematics are picked up yet again in Roland Barthes by Roland Barthes:

Is it not the characteristic of reality to be unmasterable? And is it not the characteristic of any system to master it? What then, confronting reality, can one do who rejects mastery? Get rid of the system as apparatus, accept systematics as writing (as Fourier did).

(1977a: 172)

What has happened to the language-reality split here? Not only is detail prized for its fantasmatic status (‘Perhaps the imagination of detail is what specifically defines Utopia (opposed to political science); this would be logical, since detail is fantasmatic and thereby achieves the very pleasure of Desire’ (1977c: 105)), but Barthes quite explicitly reconfronts his old problem of the relationship between ideological demystification and Utopia. His idea of the novelistic is glossed in relation to a newly invented category which he calls the ‘marvellous real’ (le merveilleux réel):

this [Fourier's] is no longer a denunciatory, reductive reading (limited to the moral falsehoods of the bourgeoisie), but an exalting, integrating, restorative reading, extended to the plethora of universal forms.

Is the ‘real’ the object of this second reading? We are accustomed to considering the ‘real’ and the residue as identical: the ‘unreal’, the fantasmatic, the ideological, the verbal, the proliferating, in short, the ‘marvellous’, may conceal from us the ‘real’, rational, infrastructural, schematic; from real to unreal there may be the (self-seeking) production of a screen of arabesques, whereas from unreal to real there may be critical reduction, an alethic, scientific movement, as though the real were at once more meagre and more essential than the superstructions (sur-structions) with which we have covered it. Obviously, Fourier is working on a conceptual material whose constitution denies this contrast and which is that of the marvellous real. This marvellous real is contrasted with the marvellous ideal of novels; it corresponds to what we might call, contrasting it directly with the novel, the novelistic. This marvellous real very precisely is the signifier, or if one prefers, ‘reality’, characterized, relative to the scientific real, by its fantasmatic train.

(353-4)

This novelistic perception of a ‘marvellous real’ recalls the deliberate confusion of a real and a fantasized Japan at the beginning of Empire of Signs. It is reminiscent, too, of Barthes's refusal to choose between the real and the fictional when he invokes his seminar as a Utopian space: ‘Is it a case of a real place or a fictional place? Neither one nor the other. An institution is handled in the utopian mood (est traitée sur le mode utopique): I trace a space and I call it: seminar’ (1984b: 369). The crux of Utopian discourse seems to lie in its modal (as much as its temporal) relationship with reality, situated somewhere between the subjunctive of uncertain status and the indicative of positive affirmation, a mode of perception that maintains reality as the ‘marvellous real’.

If Utopian discourse were simply the object of a critical commentary, then the Utopia of language (which abolishes metalanguage), would be largely undermined. To write Utopia oneself is the Utopian way to uphold it, hence, undoubtedly, Barthes's own experiments with the novelistic in his more obviously ‘creative’ writing of the 1970s.⁶ I want finally to recall that Utopia is even inscribed in his last work, Camera Lucida. One of the very few photographs that isn't of a person or group of people, an old house in Grenada, is given the splendidly Utopian caption: ‘That's where I should like to live …’ (C'est là que je voudrais vivre …)

An old house, a shadowy porch, tiles, a crumbling Arab decoration, a man sitting against the wall, a deserted street, a Mediterranean tree (Charles Clifford's Alhambra): this old photograph (1854) touches me: it is quite simply there that I should like to live … I want to live there, en finesse—and the tourist photograph never satisfies that esprit de finesse. For me, photographs of landscape (urban or country) must be habitable, not visitable. This longing to inhabit, if I observe it clearly in myself, is neither oneiric (I do not dream of some extravagant site) nor empirical (I do not intend to buy a house according to the views of a real-estate agency); it is fantasmatic, deriving from a sort of second sight which seems to bear me forward to a utopian time, or to carry me back to somewhere in myself: a double movement which Baudelaire celebrated in Invitation au voyage and La Vie antérieure. Looking at these landscapes of predilection, it is as if I were certain of having been there or of going there.

(1984a: 39-40)⁷

Barthes's final relationship to Utopia appears a total reversal of the position spelled out in Mythologies. Once the ‘impossible luxury’ has been understood as the ‘necessary luxury’, Barthes dares to reverse the normal test of the Utopian system—would one really want to live there?—and to follow Fourier in taking the ‘longing to inhabit’ as his starting point. The apocalyptic vision that the former mythologist was debarred from imagining turns out to be the simple enough desire that everyone should comfortably inhabit everyday reality, that everyone should be able to take their place in it with ease (which is why the power relations of the class war and of the systems-of-meaning war still matter so much). This seems to me in a way to be Barthes's ultimate Utopia, so that all his concern with language, and all his attempts to write values, might be seen retrospectively as tied up in that ideal.

This brief exploration of the ‘place’ of Utopia in Barthes's work needs, certainly, to be set in the context of the vast primary and secondary literature of and on Utopia. In particular, to relate Barthes to the more recent German marxist tradition (the theories of Utopia on which Fredric Jameson's marxist hermeneutic most obviously feeds), would surely throw up important connections—it would at least act as an antidote to the extremist ‘joys of the runaway signifier’ approach which has never seemed to me to fit the real(?) Roland Barthes. On the one hand I hope to have shown that Utopia is a central—and highly conscious—preoccupation in Barthes's writing, and that as such it is an interesting way of drawing together his personal obsessions and of steering an argument through them. On the other hand I want to suggest that a more general case could be made for his continued political importance as a Utopian thinker for our age, one for whom Utopia as gesture and Utopia as writing seem to have equal appeal.

Notes

1. With minor amendments, this is the text of the paper read at the Warwick Barthes conference in 1985. Versions of it have also been read to the French Department research seminars of the Universities of Kent and Reading, and I should like to acknowledge many debts to the discussions that followed on all three occasions. I am currently developing my original argument into a book on Barthes as a Utopian writer. Lines of thought in the paper which now seem to me particularly problematic are picked up (very schematically) in the notes that follow.

2. The points made in this paragraph are developed in a different context in a review article: ‘Roland Barthes: the corpus and the corps’, Poetics Today 5:4 (1984), 831-7.

3. How Barthes's Utopian discourse relates to the discourse of Orientalism as analysed by Edward Said now seems to me a key question. Barthes's Utopias are often projected into the Orient: here Japan, but also China in a very important Utopian text unknown to me at the time I prepared this paper: ‘Alors, la Chine?’ (Well, how was China?), Le Monde, 24 mai 1974. The posthumous and very novelistic Incidents are set in Morocco, as is the autobiographical episode of eating rancid couscous from which Barthes draws his entire essay on Fourier. Does Barthes fall back, in Incidents and Empire of Signs, into the exoticism denounced in Mythologies, for example in his discussion of the film Lost Continent? Or is he, especially in the China essay, working along the lines actually indicated by Said himself: ‘Perhaps the most important task of all would be to undertake studies in contemporary alternatives to Orientalism, to ask how one can study other cultures and people from a libertarian, or a non-repressive and non-manipulative perspective. But then one would have to rethink the whole complex problem of knowledge and power …’ (Orientalism (Harmondsworth, Penguin, 1985 (1978)), p. 24.

4. This whole area now seems to me to be riddled with problems. If we accept Barthes's idea that everything everyday is Utopian, what exactly is going on (from the political point of view) once the bourgeois life-style is effectively legitimized under the banner of ‘the political’? Does Barthes simply give up on his political bad conscience and capitulate to revelling in the everyday? Barthes himself would doubtless argue some political justification: ‘Pleasures, too, are finite in number, and if ever we achieve a de-alienated society, it will have to take up again, but in another place, in a spiral, certain fragments of bourgeois savoir-vivre’ (1985: 172-3); ‘he took no part in the values of the bourgeoisie, which could not outrage him, since he saw them only as scenes of language, something novelistic, what he took part in was the bourgeois art de vivre’ (1977a: 45). See too the whole section of Roland Barthes by Roland Barthes on his dual attitude to money (45-6), which alludes to his earlier approval of the positive value placed by Fourier on a luxurious life-style and extravagant expenditure.

5. The politics of sexuality is a central theme of Fourier's writing which is both very important to Barthes and yet totally ignored in this paper. Barthes finds in Morocco and Japan a ‘happy sexuality’ (sexualité heureuse) worthy of Harmony, but what exactly are the sexual politics of Utopia?

6. But most explicitly, of course, in the 1969 text Incidents published only last year.

7. The passage raises numerous problems, however. The presence of memory, albeit in the form here of a fantasized memory, reopens the question of the presemiological Golden Age which I dismissed too easily at the beginning of this paper. Should this sort of memory be related to the novelistic anamnèses of the autobiography? Where exactly does Barthes's escalating obsession with Proust fit in, which extends to childhood recollection? The generally Utopian article ‘La lumière du Sud-Ouest’ (the light of the South-West) picks up Proust through its reference to the ‘memory of the body’ (la mémoire du corps) to conclude that, when it comes down to it ‘childhood is the only Country’ (Au fonds, il n'est Pays que de l'enfance) (1987: 20). Fredric Jameson, in his commentary on Ernst Bloch in Marxism and Form (Princeton, Princeton University Press, 1971), uses Bloch's Utopian The Principle of Hope to argue his own case for Proust as a Utopian writer. As Jameson also suggests there, death constitutes a major stumbling block for any future orientated Utopian thinker, and I am exploring ways of integrating the final melancholy, death-obsessed, womb-nostalgic Barthes into the framework of a still Utopian argument. ‘Barthes beyond the pleasure principle’ seems an appropriate way of posing the problem, and clearly a psychoanalytic reading could open up a new route into Barthes's Utopia.

References to Works by Barthes

Although translations from Le Grain de la voix and Le Bruissement de la langue are my own, I have given page references to the translations published since this paper was first written (The Grain of the Voice and The Rustle of Language). All other references are to published translations as listed; I have occasionally introduced slight modifications.

1970. Writing Degree Zero, translated by Annette Lavers and Colin Smith (London, Jonathan Cape (Paris, 1953)).

1974. Utopia Rivisitata, special number of Almanacco Bompiani (Milan).

1977a. Roland Barthes by Roland Barthes, translated by Richard Howard (London, Macmillan (Paris, 1975)).

1977b. Image-Music-Text, essays selected and translated by Stephen Heath (London, Fontana Paperbacks).

1977c. Sade, Fourier, Loyola, translated by Richard Miller (London, Jonathan Cape (Paris, 1971)).

1982. Empire of Signs, translated by Richard Howard (New York, Hill and Wang (Geneva, 1970)).

1983. Barthes: Selected Writings, edited and with an introduction by Susan Sontag (London, Fontana Paperbacks).

1984a. Camera Lucida: Reflections on Photography, translated by Richard Howard (London, Fontana Paperbacks (Paris, 1980)).

1985. The Grain of the Voice: Interviews 1962-1980, translated by Linda Coverdale (London, Jonathan Cape (Paris, 1981)).

1986. The Rustle of Language, translated by Richard Howard (Oxford, Basil Blackwell (Paris, 1984)).

1987. Incidents (Paris, Seuil).