Calligram Essays in New Art History from France Edited by NORMAN BRYSON, CAMBRIDGE UNIVERSITY PRESS Cambridge New York Port Chester Melbourne SydneyContrib Acknowledgen NORMAN BRYSON 1 Art as Semiological Fact JAN MUKAROVSKY 2. Time and the Timeless in Quattrocento Paintin YVES BONNEFOY 3 Giotto’s Joy JULIA Kristeva 4 The Trompe-l'Ocil JEAN BAUDRILLARD To’ z in the Visual Arts: Poussin’s The Arcadian Shepherds LOUIS MARIN 6 Le Menifias MICHEL FQUGAULT 7 The World as Object’ ROLAND BARTHES Ambrosia and Gold MICHEL SERRES 9 In Black and White jean-CLAUDE LEBENSZTEIN 1o Turner Translates Carnot MICHEL SERRES 11 The Wisdom of Art ROLAND BARTHES Index st of illustrations page vii ix xi xiiiINTRODUCTION Norman Bryson irre can be little doubt: the discipline of art history, having for so long lagged behind, having been among the humanities perhaps the slowest to develop and the Jast to hear of changes as these took place among even its closest neighbours, is now unmistakeably beginning to alter. One index of change is the number of new journals that in the past ten years, and strikingly in the past five, have ap- peared on both sides of the Atlantic, journals that explicitly go beyond the discipline’s stats quo: in the United States we have seen the emergence of October and Representations in the United Kingdom, of Art History and World ard Image, along- side the continued flourishing of Block. Another index is the appearance and, more crucially, the institutional acceptance of a generation of new writers on art, writers whose work consciously challenges or modifies prevailing and profes~ sional modes: one thinks here of T. J. Clark, John Barrell, Thomas Crow, Svetlana Alpers, Ronald Paulson, Miehacl Fried, ; To varying degrees, cach of these writers brings art history into relation with another field of enquiry. Although such mapping docs no justice to the complexity of negotiation involved, one can nevertheless say that in almost every case recent innovation has taken place extra-territorially, in the interval between art history and another domain: between ‘historical’ art history and modem art criticism in the case of Michael Fried, and between art history and a certain literary criticism with Ronald Paulson; between art histoty and social and political history with T. J. Clark and Thomas Crow, and these taken together with litcrary history in the writing of John Barrell. In recent years the discipline of history has itself been changing, with the werk of the Annales school culmi- nating in the fascinating achievement of historians such as Braudel and Le Roy Ladurie; and when Svetlana Alpers orNORMAN BRYSON xiv Michael Baxandall writes about the ‘visual skills’ and the “visual culture’ of a society, it is clear that these concepts, and the complex picture of material practices that emerges in the New History, are in close alignment, In a period that has also seen the rapid development of film studies, and more recently of the history of photography, one expects to sec an alliance between art history and such new disciplines, between anal ysis of painting and the kinds of discourse that, for example, made Saren such an important forum for debate in the 19708; and although the interlinks between art history and film stud ies are still rudimentary and in need of further development, that conjunction is part of the context of October in America, of Block in England, and of those aspects of feminist art his tory that derive from feminist film critic What about France? In some respects the landscape looks familiar, at least to the British observer. As elsewhere, in France too a conservative art history dominates the major institutions. At the official centre of French art history onc finds the cult of the great Parisian exposition and of the great Parisian exposition catalogue, whose methods tend to focus and to consolidate the most positivistic aspects of art historical writing: while at the same time they provide some measure of official disapproval of methods and approaches less ap= propriate to the conservative catalogue form (although there have been some notable exceptions}. Surrounding and sup- porting the scholarship, and ideology, of the catalogue, one finds a familiar array of university departments and auction houses, and recognizable jostling of the professional art his- torical article against a surround of trade advertisements. But ‘one also notices features in the French context that are quite singular and interesting Perhaps the first in importance of these features is that major figures in French intellectual life do not hesitate to discuss works of art; and to do so not in the marginal or occasional essay, but in the course of their main or central work. When Michael Foucault, in The Order of Things, anal- yses Velézquez’ Las Menifias, and Jacques Lacan, in The Four Fundamental Concepts, discusses Holbein’s painting of The French Ambassadors, we find important theses being presented across what is to us an entirely unknown and unfamiliar idiom, a form of writing that is not art history as we in the English-speaking world know it (yet if it is no art history, what is it). Iu a similar way Roland Barthes, m his essays on Dutch painting, on Arcimboldo, and on Cy Twombly, felt no apparent uncertainty about his right co discuss paint. ism.ing, and indeed when Barthes moves the subject of his writ ing from the text to the image one feels no change of gear. Again, with Yves Bonnefoy, one senses little absolute dis- tinction between poetry, literary criticism (‘poctics’), and art criticism (‘visual poetics’). This is perhaps the most significant feature of such writing in France: the absence of the sense of threshold, of border police ready to pounce. And accompa- nying that, one fecls the absence of the sense of apology with which the writer in England tends to marginalize his work in the visual arts: one thinks of Kenneth Clark, and of his grand refusal to allow the least whiff of the academy to com— promise the pleasures of the cultivated amateur; of the com— parable isolation and sclf-isolation of Adrian Stokes; of the wonderfull essays on art that in England crop up, yet always at the margins of the distinguished eateer clsewhere (for ex- ample, Richard Wollhcim, or Sir Edmund Leach); and of the ephemeral, journalistic form chosen for much of his work in the visual arts by John Berger. Instances such as these spell out what is almost a national fate; and given this tendency towards marginalizing the Vis- ual (while the centre is firmly held by the Literary) the French example can offer us the encouragement of a comparative ease of access to visual experience, of the closeness of the Visual to everything else. ‘A second difference is that, compared with writing in En- glish, French writing on art does not seem to experience a significant dissociation between ‘art history’ and ‘art criti- cism’, One area where such a dissociation is equally unknown to ourselves is of course literature. We do not think it strange if Frank Kermode or Christopher Ricks pronounces upon contemporary authors; that their opinions would be among the first to be canvassed as a matter of course by those seeking (as though for che next Dalai Lama) the next Poet Laureate, highlights the respectability, almost the inevitability of this. We might very well find it strange if Kermode or Ricks had no views about contemporary writing, and might wonder how living a sense of tradition was really evident in the rest of their work, But with the image, in the Anglophone world, things are otherwise. Art history, on the one hand, and writ- ing about contemporary art (‘art criticism’), on the other, take place in two different worlds, with different personnel, modes of funding, journals, and conventions of writing. With. usa figure such as Stephen Bann, who combines the roles, and who can write with equal insight about Couture and Christopher Lebrun, is the exception. To our official art his INTRODUCTIONNORMAN BRYSON tory the past is past. is over and done with; it isa past, rather than a tradition in the sense of T. S. Eliot (or Tung Ch’ Ch’ang), in which the new work can be felt to modify the shape of tradition, and actually to alter the past from which it emerges. There is some irony here. Prevailing art history famously insists on limiting itself to ‘what was possible in the period’: its historicism demands a purity or puritanism of perspective in which ‘leakage’ from the present into the past is viewed with suspicion and alarm. Yet this official art history, which insists on the context of the work's production and thétefore on the work’s present, is in fact largely wnin- terested in its own present, the artistic and crisical present it actually inhabits, "That such a dissociation is less pronounced in France must, among other factors, be ascribed to the far more sophisticated understanding of the rclation of signs to history that appears in the great intellectual movements in France since 1945: ¢x- istentialism and phenomenology, but particularly structur- alism and post-structuralism. Against such a background it is simply hard to support the level of ignorance necessary to maintain an attitude in which art historians can seem profes- sionally bound to disregard the wider intellectual and artistic activity that surrounds them. I think we can take this as the third distinctive feature of new art historical writing in France: its awareness of present debates. Behind all the cssays col- ected here the reader will be able to sense a broader intel- Iectual horizon than we in the Anglo-Saxon world currently allow ourselves. One senses the proximity, in Jean-Claude Lebensztejn and in Jean Baudrillard, of the thought of Jacques Derrida (and notably Derrida’s concept of the supplement). In Michel Serres, painting opens onto the history of science; in Louis Marin, it opens onto the history of shetoric and of subjectivity; and in Julia Kristeva, painting opens onto the field of psychoanalysis. Why do we, in England and America, limit ourselves in this way? When literary criticism, for example, has by con trast become so broad in its horizons, so self-aware in meth- odology, so confident of its right to read from the present, that is from the classic, perspective? One answer must be that for us the image is not yct particularly thought of in terms of signs, as something to be interpreted. Academic art history reacts to the image by secking documentation: that is where it does its reading — in documents (I semetimes have the sense that patronage studies, in particular, will read anything rather than read the painting). The art history reflected in the presentvolume of essays reacts to the image as to any other work of signs. It is naturally hermeneutic, and it knows reading to be as complex and intricate a process as, for academic or Warburg iconology, it is the comparatively simple decoding of emblems and motifs. The status of painting as sign is so fundamental for this alternative or New Art History that the present collection has been prefaced by the great essay on signs by Jan Mukarovsky (made an honorary Frenchman for the occasion). Mukarovsky’s remarks on the sign are in a sense prior to everything that follows. But in case the rele- vance of semiology or of the word ‘sign’ should seem ob- scure, a further account of the issues may perhaps be in order. eae ‘The emphasis on sign may seem odd, but what this term must in the first instance displace is the term perception. It is still almost natural for us to think of painting as in some sense, if not completely, the record of perception, perception that ~ if we follow Gombrich on this point ~ is variously con- ditioned by the previous representations of perceptions that come to the artist from his tradition, Our ordinary assump- tions here owe much to Gombrich, and it is not out of place to clarify some of the thinking we almost take for granted when we picture to ourselves what is involved when the representational painter sets out to create a painting. The thinking, as Gombrich is the first to point out, models itself on a certain understanding of obscrvation in science. First there is an initial problem that science is to explore. A trial solution is proposed, as the hypothesis most appropriate to the problem and the one likeliest to lead to its solution. An experimental situation is devised in which the strengths and weaknesses of the hypothesis can be submitted to falsification The resulting situation reveals new problems whose existent or importance were not apparent at the commencement of the process. And so scientific observation continues, con- stantly testing its hypotheses against the observed world, and retesting its scheme of things against perceptual disclosure. In Art and Illusion Gombrich characterizes the work of the painter along just these lines: as a continuous development consisting in what he calls the “gradual modification of the traditional schematic conventions of image-making under the pressure of novel demands’. The pattern for art is the same as that for science. First there is the initial problem: Giotto, for example, sets out to record the appearance of the human face. Tradition suggests a particular formula or schema for INTRODUC 1ONNORMAN BRYSON xviii its transcription onto canvas: let us imagine that it is an early Giotto where the influence of Cimabucis strongly felt. Giotto tests the schema against observation of the face. Observation reveals that here and there the Cimabue-schema is inadequate to the perceptual findings and that the schema must be mod- ified in aceordance with the discrepant data. The modified schema in turn enters the repertoire of schemata and will in due course be subjected to similar tests and claborations as was its predecessor. This-conception of image-making, with its key terms of schema, observation, and testing, might be called the Percep- tualist account, since the essential transaction concerns the eye, and the accommodations the schema must make to new observations coming into the eye, The viewers, for their part, are defined by this Perceptualist account as performing an activity where those terms reappear in more passive guise the viewers confronting a new image mobilize the stock of perceptual memories, bring them to the new work for testing, and their visual schemata are in tum modified by the en- counter between the new image and their, the viewers’, gaze And if we stand back a little and begin to ask questions of the Perceptualist account, we will find that, crucially, it leaves no room for the question of the relationship between the image and power. ‘The account exhausts itself in a description of image-mak- ing that omits or brackets ous the social formation, for in the Perceptualist account the painter's task is to transcribe as ac- curately as he can perceptions, just as it is the viewers’ task to receive those relayed perceptions as sensitively as they can, and with minimal interference or ‘noise’. The painter per~ ceives, and the viewers reperceive, and the form that unites them is a line of communication from one pole ~ replete with perception, the painter's vision ~ to the other pole — the view- ers” gaze, cager for perception. The image is thought of as a channel, or stream of transmission, from a site dense in per= ception to another site, avid for perception. And if social power features anywhere, in this picture of things it is as something that intervenes between the two sites or poles, that interposes itself and makes demands of another kind. Power, social and political power, may utilize this channel and its object of perceptual transmission, the image, in var- ious ways and according to its own ends. Its intervention may be construed as of a positive and supportive nature, as when for example an individual or an institution — the Patron, the Church ~ economically cnables the painter to carry outSs work. Or the intervention by social and political power seay be of a negative and subtractive nature, appropriating be image to a particular ideology of the Church, the State, she patron class. But either way the place of power is on the ide of this inward perceptual activity of painting and mewing. Power seizes, catches hold of, expropriates, and deflects the channel of perception that runs from painter to jewer; perhaps it enables, supports, maintains, finances that channel; but however we view it, power is theorized by the Perceptualist account as always outside this relay of the visual image: power is an external that moves in, and the forceful- ness of power is measured by the degree to which it penetrates and overtakes the private transmission of percepts, where the essence of power manifests exactly in its exteriority Built into the Perceptualist account, whose fullest state- ment is Gombrich’s Art and Illusion, is the idea of power as alien to the making of images, and accordingly a direction of enquiries into the relation of power to the image, away from the canvas and into institutions outside painting, and in particular into the history of patronage. Perceptualism and the claboration of a history of patronage go hand in hand for y. and are best understood as simultaneous devel- opments of the same Perceptualist position. Each is the other's implication. Yet the connections between the image and power become instantly mysterious, for one is by definition outside the other: both are positioned in a mutual exteriority. A mystic simultancously arises in which it becomes accept able practice to draw up two separate but nevertheless darkly interrelated columns, one of social events, and one of paint- ings painted. Clearly this ‘two-colummn’ approach is inade- quate, so Iet us go back to the fundamentals of image-making and this time examine it from the other side - from the viewer's gaze. Ie may indeed be the case that when | look at a particularly lifelike representation, I, the viewer, reexperience at one re- move the original vision, retinal or imaginary, of its creator, the artist. There might be absolute congruence between the two mental fields of artist and spectator. Yet the recognition of a painting can hardly involve such congruence as a nec- essary criterion. While it might be possible for the painter to know that his image corresponds to his original vision or intention, no such knowledge is available to the viewer. Rec- ognition, here, is not at all an act of comparison between two mental fields, or cross-referral of perceptions from one end of the channel to another. It might well be true that when I INTRODUCTION xixNORMAN BRYSON xx look at a particular canvas | obtain a set of perceptions I can. obtain from this canvas and no other, but the set of percep- tions in the viewing gaze can’t of itself provide criteria of recognition, This is clear enough if we think of sign-systems other than painting. With mathematics, for example, | may have a vivid picture in my mind of a certain formula, bue the criterion of my knowing that the picture was a formula and not simply a tangle of numbers, would be my awareness of its mathematical application. The test of whether or not I had understood the formula would consist not in the examination of my private mental field, or even the comparison between my mental field and its counterpart in the mind of whoever produced che formula, but in seeing if I could place the for- mula in the general context of my knowledge of mathematical techniques, in my ability to carry out related calculations, and so forth; in short, in my executive use of the formula, ‘Again, in the case of a child learning to read, itis hard to devermine the sense of the question, ‘which was the first word the child read? The question seems to appeal to an inward accompaniment to the physical progress of the cye through the chain of characters, an accompaniment that at a particular point takes the form of a ‘now I can read!” sensation. Yet the criterion for right reading cannot be this. The child might indeed have such a sensation, yet be quite unable to read correctly; where teading, like the activity of mathematics, and like the recognition of an image, can be said to take place only when the individual is able to “go on’; not to reveal to- the world a secret event of the interior, but to meet the executive demands placed upon the individual by his ot her world. T hope the implications are becoming clear. Perceptualism, the doctrine whose most cloquent spokesman is undoubtedly Gombrich, describes image-making entirely in terms of these secret and private events, perceptions and sensations occur Hing in invisible recesses of the painter's and the viewer's mind. It is as though understanding in mathematics had been reduced to the occurrence of ‘now I see it!” experiences, or the test of whether or not someone read aright were whether he or she experienced a now I can read!” sensation, The point is that mathematics and reading are activities of the sign, and that painting is, also. My ability to recognize an image neither involves nor makes necessary inference towards the isolated per- ceptual field of the smage’s creator. It is, rather, an ability that presupposes competence within social — thatis, socially constructed — codes of recognition. And the erucial differencebetween the term perception and the term recognition is that the latter is social It takes one person to experience a sensation; it takes (at least) two to recognize a sign. And when people look at representational painting and recognize what they see, their recognition docs not unfold in the solitary recess of the sen sorium, but through their activation of codes of recognition that are learnt by interaction with others, in the acquisition of human culture. One might put this another way and say that whereas in the Perceptualist account the image is said to span an arc that runs from the brush to the retina, an are of inner vision or perception, the recognition of painting as sign spans an arc that extends from person to person and across interindividual space The changcover from the account of painting in terms of perception to an account of painting as sign is nothing less than the relocation of painting within the field of power from which it had been excluded. In place of the transcendental comparison between the image and perceptual private worlds, stand the socially generated codes of recognition; and in place of the link, magical and illogical, that is alleged to extend from an outer world of things into recesses of in wardness and subjectivity, stands the link extending from individual to individual as consensual activity, in the forum of recognition. The social formation isn’t then something that supervenes or appropriates or utilizes the image, so to speak, after it has been made: rather painting, as an activity of the sign, unfolds within the social formation from the beginning. And from the inside: the social formation is inherently and immanently present in the images, and not a fate of anc: ternal that clamps down on an image that might prefer to be left alone, One of the things I think is currently wrong with the way the dominant art history tends to theorize the relation be- tween the image and social and economic power, is really a matter of topology, of the kind of charting in mental space, or of drawing the boundaries around the concepts image, power, and social formation. So far I have been addressing Per- ceptualism, the notion that artistic process can be deseribed exclusively in terms of cognition, perception, and optical truth. What Perceptualism leads to is a picture of art as apart from the rest of socicty’s concerns, since essentially the artist is alone, watching the world as an ocular spectacle but never reacting to the world’s meanings, basking in and recording xtraterritorial perceptions but apparently doing, so in some INTRODUCTION xxiREE EEE EEE NORMAN BRYSON zone, off the social map. Perceptualism always renders art banal, since its view never lifts above ocular accuracy, and always renders art trivial, since the making of images seems to go on, according to Perceptualism, out of society, at the margins of social concems, in some eddy away from the flow of power. And this poverty of the theory can be eliminated ~ there is really no reason why if it is so reductive we need it our vote. Because painting as an art of the sign, which is to say an art of discourse, is coextensive with the flowof signs through both itself and the rest of the social formation. There is no marginalization: painting is bathed in the same circulation of signs that permeates or ventilates the rest of the social structure, This said, I think it equally important to address what might appear to be the opposite extreme, the position that says that art is to be approached in terms of social history, that art belongs to the superstructure, and that the super- structure cannot be understood without analysis of the social, and in particular the economic, base. You might perhaps have supposed that in the claim for the immanently social character of the sign, a social history of art was necessarily being ad vocated, but that doesn’t follow in any simple sense; and the reason it doesn’t is once again that a strict cconomism is no better placed than Perceptualism was, to follow through the implications of what it means when we begin to think off paintings as signs. The essential model here is inevitably that of base and su- perstructure. Taking the base structure as consisting of the ultimately determinant economic apparatus of the society, and assuming the unified action of productive forces and relations of produetion, then ‘art’, alongside legal and political institutions and their ideological formations, is assigned firmly to the superstructure, If we want to understand paint- ing, then first we must look to the base, to the questions of who owns the means of production and distribution of wealth, to what constitutes the dominant class, to the ide- ology that class uses to justify its power, and then to the arts, and to painting, as aspects of that legitimation and that monopoly, The mistake here is, in interesting ways, the same mistake as that made by Perceptualism, because the question that needs most urgently to be addressed to the base—superstruc- ture model is: in which tier of the model should we place the sign? Social history, in this view, is the expression in the superstructure of real, determinant events occurring in the any longer give xxiieconomic base; legal institutions, political institutions, ide- ological formations, and among these the arts ~ and painting, ~— are said to be secondary manifestations or epiphenomena of base action. Very well: but where shall one allocate the sign? Does the sign belong above, along with ideology, law, and the other derivations? Or is it primary, down there next to the technology, the plant, the hard productive base? It is indeed a crucial question. In the extreme statements of base—superstructure thinking, signs are no more than the impress of base on superstructure, The sign follows the base without deviation, which is also to say that the base determines discourse, that discourse takes its patterns from power and re- peats them in another key, the key of idcologies. Signs and dis- course are assumed to accept the impact of the material base as wax accepts the impact of a seal. The sign and discourse, and painting as a discursive art, arc the expression of the giv- en reality, and — so to speak ~ its negative profile. First there is the original matrix of cconomic reality, then out of that ma- trix there appears the inscription, the writing into art of what is happening in the base. But as soon as this picture is fully drawn in, we can see how difficult it is to understand how the model is to work in practice. The base-superstructure conception posits a material base that of itself engenders the sign, at its every point of change. The picture proposes a mystery of spontaneous generation of signs directly out of material substance. Yet it is clear that the economic or ma- terial base never has produced meaning in this uncanny sense: the world does not bear upon its surface signs that are then read there, as though matter itself were endowed with elo- quence. And while the base-superstructure model may seem to lead to a social history of art, and to concede the social character of the sign on which I was earlier laying stress, in fact the ironclad pronouncement that the sign belongs to the superstructure omits its social history. It is in matter ~ in the prior contour of material reality ~ that the sign is said to arise, as its negative relief, or stencilled echo. Yet the sign’s own materiality, its status as material practice, is sublimed or va- porized just as drastically as in the Perceptualist account. The global body of signs, discourse, is said to be part of the cloud of ideas and ideologies hovering over and obscuring the real material base, as though discourse were the transcendental accompaniment, floating and hazy, to a real material world. What the economist position is forced to deny is that the sign, that discourse, is matcrial also, and entails material work and claboration as much as the activities of the alleged base. INTRODUCTION ‘ xxiiiNORMAN BRYSON In the case of painting, the material character of the sign is far more evident than it is in the casc of language, and it is therefore perhaps easier to think of the image in nonidealise terms than it is to think of the word in nonidcalist terms The problem here is that although the material character of painting cannot be ignored, that materiality tends ta be equated with substance, pigment, with the brush and the canvas. And if one sets side by side the image of a factory turning out machines and a studio turning out paintings, it will seem as though all the power is in the factory and none in the studio, and that the social history of art must first describe the hard reality of production, ownership, capital, and dominant and dominated classes, and then trace the re~ percussions of this hard reality in the atelier Once again, painting is off the map or at least relegated to the margin, just as it was in Gombrich. But figurative painting isn’t just the material work of brush and pigment on canvas. Non= figurative painting may tend in that direction, but as long as the images one is dealing with involve recognition, as long as they are representations, they are material signs, and not simply material shapes. And as signs, as complex statements in signs and as material transformations of the sign, paintings are part of a flow of discourse traversing both :he studio and the factory, Discourse doesn’t appear spontancously out of matter: it’s a product of human labour. It is an institution that can’t simply be derived from the alleged economic tase. Like eco- nomic activity, discursive activity is nothing less than the transformation of matter through work, and though the eco~ nomic sphere and the discursive sphere may interact, and in fact can hardly be conceived outside their interaction, to think of discourse as a floating, hazy, transcendental cloud hovering above the machinery amounts to a mystification of the ma- terial operation of ideology. To put this amother way: to theorize the image as a nebulous superstructure accompani- ment to a hard and necessary base is to deny the institution of discourse as a cultural form that interacts with the other — legal, political, economic — forms in the social world The crucial reformulation to be introduced into the social history of art is to break the barrier between base and su- perstructure that in effect places the sign in exteriority to the social formation = an exteriority that merely repeats, in a different register, the Perceptualist separation of the image from social process, What is needed is a form of analysiss sufficiently global to include within the same framework bolt the economic practices that Historical Materialism assigns to the base, and the signifying practices that are marginalized as superstructural imprint. And the topology must be clear. The base-superstructure model can’t cope with the question of the sign, and the problems that arise as soon as ene tries to work out which tier of the model the sign is supposed to fit into are so enormous that the concept of the sign emerges as really a powerful political idea: it prises the model open and finally breaks it apart. Above all, it makes clear the need for a form of analysis in art history dialectical nough, and subtle enough, to comprehend as interaction the relationship among discursive, economic, and political practices. In discussing the visual arts at the moment the need is, | think, an urgent one. In one dominant theorization of paint- ing - Perceptualism — the social formation has little part to play except as intervention or utilization. The inherently so- cial character of the painterly sign is eclipsed by the picture of the artist alone in his studio, immersed in the privacy of his perceptions, his only link with the outer world consisting of the optical contact he has with the surface of things, and his only major difficulty being the accommodation of his schema to the influx of new sensations. In the major alter native art history presently has as its disposal - social history = the same sequestration of art from the public domain is reinstated, for although the social history of art wants the atelier to come into contact with the rest of society, the con= tact can now be seen as narrowly economic, Out there, in the social base, an economic apparatus is generating dominant and dominated classes, is organizing the means of production and distribution of wealth, and is forging the determinants over the superstructure. In here, in the hush of the studio, the painter passively transcribes onto canvas the visual echo of those far-off events. Or let us say that economism is less ambitious, and that it examines instead the more local relation that exists between painter and patron or patron class. This is certainly an improvement on Perceptualism’s relegation of the painter into social limbo, and hardly less of an improve- ment on the attempt by a dogmatic Historical Materialism to transform the painter into an echo of the distant rumble of history. The lines of capital that link the painter to his patron or patron class are real and of enormous importance. But they are not the only lines that link the painter to the rest of the social world, for there is another flow that traverses INTRODUCTION XXVNORMAN BRYSON the painter, and the patron class, and all those who participate in the codes of recognition: the flow of signs, of discourse, of discursive power. It is a flow in two directions, for the painter can work on the discursive material that comes to him, can elaborate it, transform it through labour, and return it to the social domain as an alteration or revision of the society's discursive field, I stress this because neither Perceptualism nor the economism underlying the ‘two-column’ social history of art has much to say about creativity, or innovation, or more simply the work-of the sign. If the task of the artist is, in Gombrich’s words, the ‘modification of the schema under the pressure of novel [visual] demands’, then the effort 0 image-making consists in making and matching against what is already and preexistently there. The problem for the image becomes a matter of catching up with reality, of discarding those cl ments within the schema that occlude the limpid registration of the world. The image doesn’t have the power to inau- gurate, to commence, to molest the given structures. And again with a strict economism in its full, base—superstructure expression, the image can only at best repeat the larger and truer events of history. Capital flows into the atelier, power flows in, but the flow is in one direction, and ir becomes difficult, if not impossible, to conceive of the reverse of this process, in which the image could be seen as selfempowered and out-flowing, or as an independent intervention within the social fabric. ‘A cardinal virtue of considering the visual image as sign is that having relocated painting within the social domain, inherently and not only as a result of its inst-umental placing there by some other agency, it becomes possible to think of the image as discursive work that returns info the society The painter assumes his society's codes of recognition, and performs his activity within their constraints, bur the codes permit the claboration af new combinations of the sign, fur ther evolution in the discursive formation: the result of paint ing’s signifying work, these are then recirculated into society as fresh and renewing currents of discourse. The configura~ tion of signs that constitutes a particular image may or may not correspond to configurations in the economic and polit- ical spheres, but they need not have first been read there, or match events that only by an act of arbitrary election are privileged as the truth of social history. Ieis usually at this point that one encounters the objection that the power of the image to intervene in the social fabricis severely limited, that the image possesses built-in strategic inadequacies, and that unless images articulate their local acts of innovation with the stronger, the major movements and activities within the social formation, they are insignificant (where that word operates as a term of quantity, a measure of instrumental efficacy). No one is so misguided or so out of the world as to claim otherwise. If we think of even rev- olutionary moments in painting, the impact of the image on its surrounding world may seem hopelessly curtailed. Géri- cault’s portraits of the mad did nothing to modify the juridical status or treatment of the insane, nor did the appearance of Olympia at the Salon of 1865 do much, so far as we can tell, to change the interesting nocturnal economy of Paris. Yet this is only a truth of logistics, of administration of the image; and the danger is that this obvious truth, this platitude of instrumental inadequacy, conceals, makes it difficult to think through, the subtle and far more important truth of topology Instrumentally, an Olympia at the Salon of 1865 may do little to affect the status of prostitution; but the essential point is that these collisions of discursive forms ~ its juxtaposition of Odalisque and Prostitute or Géricault’s elision of the social fixity of the portrait with the social placelessness of the insane ~ occur within the social formation, not as echoes or duplicates of prior events in the social base that are then expressed, limpidly, without distortion, on the surface of the canvas, but as signifying work: the cffortful and unprecedented pull- ing away of discursive forms from their normal locations and into this painting, this image. To look for a result in the form of a change in the base, or in the political sphere, is once again to assume that it is only there, in those arbitrarily priv ileged zones, that ‘real" change happens. If your politics is such that the only changes you recognize are those that take place in the economic sphere, and all the rest are mere swirl ings in the cloud of superstructure, you will not find painting a particularly interesting or forceful instrument. And this will be because power is located exclusively in agencies other than discourse: in capital, in the factory, in the production and distribution of wealth. The only revolution and indeed the only change that will then be recognized is in those privileged and limited spheres. Both too narrow and too ambitious in its sense of social change, a dogmatic Historical Materialism will miss where the power is in discourse, and in painting In fact it will be found in every act of looking: where the discursive form of the image meets the discourses brought to bear upon the image by the viewer, and effects a change; INTRODUCTION xxviiNORMAN BRYSON where, in order to recognize the new discursive form that is the image, existing boundaries of discourse ~ the categories and codes of recognition ~ must be moved, turned and over~ turned, in order to recognize what this image is, that is at once Odalisque and Prostitute, socially fixed by the portrait and socially displaced as insanity. If power is thought of as vast, centralized, as a juggemaut, as panoply, then it will not be seen that power can also be microscopic and ciscrect, a matter of local moments of change, and that such change may take place whenever an image meets the existing dis~ courses and moves them over, or finds, and changes, its viewer, The power of painting is there, in the theusands of gazes caught by its surface, and the resultant tuming, the shifting, the redirecting of the discursive flow ~ power not as a monolith, but as 2 swarm of points traversing social stratifications and individual persons. ‘This discussion of signs and of painting as a domain of con- scitutive interpretation will, I hope, help locate tae present eollection of essays in particular, and some of the emphases of the New Art History in general. Writing about art will be seen to have in fact two mandates: archival and hermeneutic. The first of these is the mandate chat governs most art history at the present time: to trace the painting back to its original context of production. Yet the context may now have to be defined in a new way. It cannot be chought of simply as the circumstances of patronage ot commission (im- portant though such factors most certainly are) or as the conditions of original perception and its notation. Original context must be considered to be a much more global affair, consisting of the complex interaction among all the practices tHe make up the sphere of culture: the scientific, military, literary, and religious practices; the legal and political struc qures: the structures of class, sexuality, and economic life in the given society. It is here, in the interactive sphere, that one would locate the theorctical position of Mukarovsky, the work of Svetlana Alpers and Michael Baxandall, the journal Representations, and, among the present essays, those by Michel Serres, Louis Marin, and Roland Barthes (on Dutch painting). The second, hermeneutic, mandate refers to the image as something to be interpreted and read. One of the great weak- nesses of prevailing art history must be its neglect of “reading skills’ and practical criticism. Whereas students of literaturecegularly spend hours in class wrangling over the interpre- sation of texts, the level of reading among students of art history is hardly developed at all, but left somehow to take care of itself. New Art History, at least as we see it in such journals as Block, Word and Image, or October, or in the essays here by Barthes, Baudrillard, Lebensztejn, and Bonnefoy, not only invests far more in the basic act of interpretation (now sadistinguishable from actual recognition): it assumes as so phisticated a level of interpretation as that achieved — after many decades of endeavour = by current practice in literary exiticism. There is no reason for art history to feel this as a shreat from an expanding lit. crit. On the contrary, all art iustory needs to do is to appropriate the advance, take from bterary criticism everything of service to itself, make reading and practical criticism regular components of art historical scaining, and the discipline will be at once more stable, more mature, and more nourishing than before. What must surcly be given up is the unadventurous assumption that strict ar- chival methods, together with a strategy for converting paint- mgs into documents, are all we need to deal with visual representation. That is impoverishment, anda recipe for stag- nation. If the present volume of essays helps to stimulate awareness of other ways of thinking about images, it will have done its work. INTRORUCTION xxix