Richard Woodfield, 'Words and Pictures' more

WORDS AND PICTURES Richard Woodfield THE QUESTION of the relationship between the verbal and visual arts has a long history which, in the western tradition, originated in the classical doctrine of mimesis. Throughout the middle ages, in the Western Empire, the use of images was justified by the claim that they were the bible of the illiterate. The claims of a literary view of painting were systematically advanced by the fifteenthcentury Florentine humanist Alberti in his important treatise De pictura and developed through the sixteenth and seventeenth centuries by subsequent writers. The notion that the construction and analysis of paintings was to be drawn from rhetorical and poetic theory lay at the heart of academic teaching until its dissolution in the twentieth century and it has received a recent reincarnation through the semiotic project. The notion that a science of signs based upon a linguistic paradigm could provide a viable basis for the analysis of the visual arts is becoming increasingly fashionable. The project has not been without its critics. As early as 1949, Sir Ernst Gombrich subjected Charles Morris's Signs, Language and Behaviour to an astringent review arguing that: It will remain for a general iconology to clarify the relations between image and language which Mr. Morris has raised. Mr. Morris hardly distinguishes sufficiently clearly between the linguistic character of 'art' and that of the visual sign. His statements on this question are unusually vague and hesitant. We hear (p. 194) that the arts may 'at least approximate the formative ascriptors of speech and writing' but (p. 195) that 'formators are clumsily handled in other media than speech or writing'. But are there any 'formators' corresponding to 'logical words' in the language of the image? Lessing noticed almost two hundred years ago that negation cannot be expressed through the image. What are the formators which the image can 'approximate'? The question may be formulated in this way- Are there images equivalent to verbal statements?1 Later, in 1961, Richard Bernheimer argued, in The Nature of Representation, that: the current identification of representation with sign function is illegitimate; that while the two modes of apprehension overlap, the field of representation is wider than the area which it shares with its sister function, comprising phenomena which no theory of semantics, however constituted, can ever claim for itself; that representation has an inner structure of its own, akin to but by no means identical with that possessed by vanous categories of signs; and finally that the function most akin to representation is not, as semanticists suppose, that of signification, but the much neglected and little known one of substitution... . Signs do not stand before us, as representamina do, to be examined, studied, enjoyed, weighed and estimated. They do so only to be transcended, for signs are stimuli meant to arouse an immediate appropriate response, and thus are ready to be obliterated in our consciousness as soon as their business is done. The perfect reaction to a sign is to forget that it has occurred the moment the sequel has taken place The perfect sign is one that subsists without individuality, character, and internal qualities, a mere empty shell of its own function.2 1 2 Art Bulletin XXXI (1949), p. 72 Bemheimer, op. cit., pp. 24 & 26. While the journal Semiotica was established in 1969 to pursue the application of semiotics to a variety of fields of research, to the end of 1983 only seven articles concerned themselves with the visual arts. One of the articles, by Oleg Grabar in 1979, asked 'Are pictures signs yet?' and observed: Is it, then, necessary to use a complex and abstract terminology derived from linguistics when all it does is to obfuscate perfectly simple, contextually reasonable, and culturally understandable visual impressions? The answer to this fundamental question can, it seems to me, be given in the m following manner. That visually perceived works of art form a language which communicates according to as yet unidentified rules of perception and creation is self-evident, but none of the existing semiotic theories known to me are entirely adaptable to the visual arts because they have not reached (and perhaps cannot reach) the level of flexible abstraction which can account for the peculiarities of visual perception, the only obvious exceptions being simplistic visual prose like stop signs or airport symbols.3 I am not sure that even stop signs can function as 'simplistic visual prose', but will leave that to philosophical argument. Interestingly, and not coincidentally, the major interest and reservations in the possibility of applying semiotics to the visual arts has come from art historians with iconographical concerns. There is a growing body of literature on relationships between images and texts and a growing selfawareness of the problems encountered in attributing meaning to images. Norman Bryson's recent work on developing a semiotic approach to the figurative arts seems to have totally ignored developments in art history and instead appeals to literary g theory. This paper will be devoted to examining the first chapter of Word and Image in the light of iconographical experience. 1 Bryson's theory is that works of art may be situated within a continuum of signs which is dominated at one end by discursivity and at the other by figurality. He writes: 'By the "discursive" aspect of an image, I mean those features which show the influence over the image of language ... By the "figural" aspect of an image, I mean those features which belong to the image as a visual experience independent of language its "being-as-image".'4 Another way of defining the same continuum is in terms of the opposition between the signified (`discursivity') and the signifier (`figurality'). He gives substance to his continuum by the use of examples ranging from a stained glass window at Canterbury Cathedral, through Masaccio's Tribute Money, Piero della Francesca's Flagellation, a Netherlands School Still-Life, Vermeer's Young Woman Seated at a Virginal to Jackson Pollock's Enchanted Garden. He concludes his analysis of the continuum with the suggestion that: '... the division of the painterly sign, as with all signs, is the division between the signified and the signifier, which gives rise to the discursive and the figural; and considered at the level of the sign, painting is found to have a history which is not at all the history of emergent realism we find in the natural attitude, or the history of successive visual styles we find in classical art history'.5 Bryson finds it very hard to pin down anyone who has the 'natural attitude', but it is fairly clear that his target is Art and Illusion as he remarks: 'The structures proposed by the sociology of knowledge construct reality in toto: there is no other Reality to be expressed beyond the reality the social process constructs. But in Gombrich's account there is always an ulterior and veridical world which Grabar, op. cit., vol 25, p. 187. N Bryson, Word and Image (1981), p. 6. Hereafter WI. 5 WI, p. 28. 4 3 exceeds the limited, provisional world-picture built from schema and hypothesis'.6 Bryson's denial of the existence of reality would seem to extend to history as well, while he is prepared to accept, though not as an exception, the reality of the physical object which is the picture: ' ... briefly to address the problem of relativism, we can say that the only component of the painterly sign guaranteed to survive is figural: the context of discursivity has to be reconstructed and the accuracy of our archaeological endeavour can never be certain. The problem is part of the whole issue of intention: are we to place the work in relation to the unknowable, irretrievable mind of its producer, or in relation to our own alien time? Dilemma without resolution ...'.7 Bryson offers us a new history to replace the old ones and this can take the form of either re-interpreting existing historical evidence, taking cognizance of such evidence as there is, or bringing new evidence to bear on existing interpretations, and such evidence could be from other disciplinary fields such as linguistics. Either way Bryson's new history has to be open to the possibility of criticism; if it is not, it becomes something other than history. II Bryson starts by telling us that although we may take a disinterested delight in the colours and shapes of a stained glass window at Canterbury Cathedral it 'plays down the independent life of its signifying material, which progressively yields, as we approach it, to a cultivated transparency before the transcendent Scripture inscribed within it. The status of the window is that of a relay or a place of transit through which the eye must pass to reach its goal, which is the Word. Qualities that might detain the eye during that transit are to be carefully restrained.'8 This might be less mysteriously rephrased as the suggestion that in the same way that when one reads a novel one attends to what is said, rather than the type which says it, so mediaeval man read pictures; to quote St Gregory's dictum `painting can do for the illiterate what writing can do for those who can read'. This is a fairly familiar idea, but closer scrutiny of the Canterbury windows shows that their iconography was so complex that it would have taken an educated person reading the inscriptions around the images to grasp their meaning. The windows may have had a didactic function but they had other functions as well. Stained glass windows were the latest fashion when Canterbury Cathedral was being built and they formed an integral part of its decoration along with the sculptures, shrines, reliquary objects, gargoyles, miserichords and architectural embellishments. In separating the window from its larger context, which was the other windows and the Cathedral as a whole, Bryson projects the twentieth-century concept of the work of art as a unique exhibitionable object back on to something which was not conceived in that way. The idea that decoration precludes meaning is also a contemporary notion and an alternative positive view is needed. Mediaeval images had another function besides straightforward didacticism which St Gregory formulated in a letter to Secundinus: `Your request [for images] pleases us greatly, since you seek with all your heart and all intentness Him, whose picture you wish to have before your eyes, so that, being accustomed to the daily corporeal sight, when you see an image of Him you are inflamed in your soul by N Bryson, Vision and Painting (1983), p. 33. WI, p. 109. 8 WI, p. 3. 7 6 love to Him whose picture you wish to see. We do no harm in wishing to show the invisible by means of the visible.'9 Images were intended to have an emotional effect on the spectator and to remind him of the existence of the heavenly world by placing him directly in contact with it by 'daily corporeal sight'. This theme was echoed by the Abbot Suger of St Denis, one of the great church builders of the Gothic period, in his Liber de Rebus in Administratione Sua Gestis: ' ... when out of my delight in the beauty of the house of God the loveliness of the many-colored gems has called me away from external cares, and worthy meditation has induced me to reflect, transferring that which is material to that which is immaterial, on the diversity of the sacred virtues: then it seems to me that I see myself dwelling, as it were, in some strange region of the universe which neither exists entirely in the slime of the earth nor entirely in the purity of Heaven; and that, by the grace of God, I can be transported from this inferior to that higher world in an anagogical manner'.10 Suger took the decoration of his church very seriously and justified his delight in the sparkle and glitter of precious materials by an ancient metaphysics of light which had recently, and conveniently, been revived: Whoever thou art, if thou seekest to extol the glory of these doors, Marvel not at the gold and the expense but at the craftsmanship of the work. Bright is the noble work; but, being nobly bright, the work Should brighten the minds, so that they may travel, through the true lights, To the True Light where Christ is the true door.11 As the image of Christ was intended to remind the worshipper of the living presence of Christ so the luminous interiors of Gothic cathedrals were intended to bring heaven down to earth. A moment's reflection makes one wonder about the practicalities of combining ineffability with intelligibility, light with text. Stained glass windows are not terribly legible. They are made out of partitioned segments of coloured glass which are far less easily decipherable than paintings or drawings; they bear more of a relation to bejewelled surfaces, sparkling from the interior light of candles and glowing from exterior sunlight until they get too dirty to see. Suger thought that there was a problem as well: `because the diversity of material [such as] gold, gems and pearls is not easily understood by the mute perception of sight without a description, we have seen to it that this work, which is intelligible only to the literate, which shines with the radiance of delightful allegories, be set down in writing'.12 It would seem that while the sight of light may lead the spectator to God, its muteness needs words for instruction. The spectacular decoration of Gothic, and Byzantine, buildings could obstruct understanding while stimulating devotion. This was a theme taken up by Boccaccio in a comment on Giotto: `the human sense of sight was often deceived by his works and took for real what was only painted. Thus he restored to light this art which for many centuries had been buried under the errors of some who painted in order to please the eyes of the ignorant rather than satisfy the intelligence of experts ...'.13 The new way of understanding the Scriptures promoted in Italy by the Franciscan movement encouraged the growth of naturalistic, non-schematic, imagery and ineffability gave way to intelligibility. Ringbom, Icon to Narrative (1983), p. 13. E Panofsky, Abbot Suger on the Abbey Church of St. Denis and its Art Treasures (1979), pp. 63 & 65. 11 Ibid , pp. 74 & 49. 12 Ibid , p. 63. 13 Decameron, VI, 5. 10 9 It is undoubtedly easier to respond to Masaccio's Tribute Money than the Canterbury windows; size and appearance have a lot to do with this, as well as a much clearer sense of something happening. It would be silly to pretend absolute ignorance of the subject as we know the title, recognize the depiction of Christ and have our bibles at hand to dig out the story; we certainly don't have to be able to read Latin to respond. The situation would have been very different for a contemporary worshipper, to whom all of this would have been common knowledge. For Bryson, Francastel has the ready, popular, explanation of why we respond to the Tribute Money: `Henceforth man will be defined not by the rules of narrative, but by an immediate physical apprehension. The goal of representation will be appearance, and no longer meaning.'14 He continues: `The opposition Francastel erects is Meaning versus Being, and it might seem that the terms "discursive" and "figural" repeat that opposition: that the Canterbury window, where discourse subjugated figure, has subordinated Life to Text, whereas the Masaccio, with its goal of "appearance and no longer meaning", appeals to "universal visual experience". Received opinion would support such an opposition: there remains a vague conception of the image as an area of resistance to meaning, in the name of a certain mythical idea of Life. Vasari's history of the progress of Renaissance painting is built on this "common-sense" view: the progress he sees is an evolutionary liberation of Life from the repression of the textual.'15 It is not clear to me what the expression 'an evolutionary liberation of Life from the repression of the textual' means, but more importantly it would have had no meaning for Vasari or Masaccio's admirer Alberti. For both of them, a painting's importance lay in its historia and increased realism contributed to the more effective rendering of a story told visually; they regarded realism as a means to an end. Alberti wrote that the 'great work' of the painter was the historia16 and that 'Everything the people in the painting do among themselves, or perform in relation to the spectators, must fit together to represent and explain the "histona"'.17 He liked variety 'provided it is appropriate to what is going on in the picture. When the spectators dwell on observing all the details, then the painter's richness will acquire favour.'18 The image was tied together, as it were, through the central regulative concept of decorum, which dictated the way in which colours were applied to generate surfaces, surfaces were brought together to create volumes, volumes to constitute bodies, bodies moved to indicate states of mind, human forms were shown to indicate states of mind, human forms were shown to indicate age, rank and type, and bodies were organized together to tell stories. One didn't portray Helen with gnarled hands, Venus in a rough soldier's cloak, unspilt bottles in the middle of a fight, nature devoid of natural life, figures standing in regimented poses, expressionless figures, Apostles leaping around like dancers and so on. There was clearly a mixture of concerns: to get the story right, to have figures looking and behaving the part, to have them behaving like people or animals and ultimately to have the image look lifelike. The artist's task was not to recreate reality, for that is impossible, but to give substance to the realities of stories, social life and the appearance of the visual world. In brief, the artist's business was to be a convincing storyteller; that notion was central to Vasari's way of thinking and to the later academic doctrines. As Sir Joshua Reynolds put the matter in the eighteenth century: WI, p. 7. WI, p. 7. 16 L. B. Alberti, On Painting and On Sculpture ed. C Grayson (1972), p. 73. Hereafter AOP. 17 AOP, p. 83. 18 AOP, p. 79. 15 14 'nature herself is not to be too closely copied.... The wish of the genuine Painter must be more extensive: instead of endeavouring to amuse mankind with the minute neatness of his imitations, he must endeavour to improve them by the grandeur of his ideas; instead of seeking praise, by deceiving the superficial sense of the spectator, he must strive for fame, by captivating the imagination'.19 If Boccaccio felt wary about mediaeval glitter, Sir Joshua was wary of Dutch exactitude, but a common concern with 'literary' meaning placed Alberti, Vasari, Sir Joshua and the academics in the same pictorial tradition. Contrary to popular belief, Alberti never maintained that a finished painting was a window on to nature; it was a conspicuous production of artifice and could be marred through neglect of natural appearances. Bryson's theory is that: 'the effect of the real in the image insists on setting up a scale of distance from the patent site of meaning which is read as a scale of distance towards the real. Of course, the real is not involved at all, but when the image purveys to us information which is not tied by necessity to a textual function, that information then constitutes an excess, and it is in this area of excess that we inscribe "real existence". The Masaccio succeeds in creating a threshold between two areas: on one side of that threshold, semantic necessity, and on the other, semantic irrelevance. The Canterbury window never manages to create such a threshold because the visual information it emits is fully exhausted by textuality.'20 He has earlier told us that: 'with Masaccio's scene of the tribute money, all one needs are the component ideas of "apostle", "money", "receiver of money", and the activity which connects them, of "donation". The narrative can be parsed, like a sentence, into its minimally sufficient requirements: apostle donation recipient.'21 And the footnote to this sentence tells us: `The idea of minimal narrative units, as part of a general "narrative grammar", goes back to Propp's Morphology of the folktale: amongst Propp's progeny are Lévi-Strauss's "mythemes" (Mythologiques), and Barthes's "lexias" (S/Z).'22 There are rudimentary textual demands and the information supplied over and above these demands contributes to the 'effect of the re al'. But it seems to me that Bryson has an oversimplified view of textual illustration: there is no essential narrative notion that the Bible records and Masaccio's Tribute Money illustrates, the situation is messier than that. There is a series of events and utterances which the Bible describes, and a biblical text which the Tribute Money illustrates. Christ's utterances were quite as important as the end result and can hardly be parsed out of existence. Bryson's reductionism is particularly suspect in the light of the importance which theologians and politicians attributed to specific details of the text.23 The Tribute Money is a narrative painting which illustrates an episode in Christ's life described in Matthew 57:24-27 and in its context in the Brancacci chapel it was just as importantly about Peter as he was the major subject of the other illustrations. A modern translation will have to serve present purposes: On their arrival at Capernaum the collectors of the temple-tax came up to Peter and asked, 'Does your master not pay temple-tax?' 'He does,' said Peter. When he went indoors Jesus forestalled him by asking, 'What do you think about this Simon? From whom do earthly monarchs collect tax or toll? From their own people, or from aliens?' `From aliens ' said Reynolds, Seven Discourses (1778), p. 69. WI, p. 11. 21 WI, p. 10. 22 WI, p. 255. 23 See A. Molho, 'The Brancacci Chapel. Studies in its iconography and history', Journal of the Warburg and Courtauld Institutes, XL (pp. 19-77). 20 19 Peter. 'Why then,' said Jesus, 'their own people are exempt! But as we do not want to cause offense, go and caste a line in the lake; take the first fish that comes to the hook, open its mouth, and you will find a silver coin; take that and pay it in; it will meet the tax for us both'. It is impossible to see how the `minimally sufficient units: apostle-donation-recipient' can actually serve the requirements of the text. The narrative encompasses the following events: the collectors, or collector, addressing a question to Peter, Jesus questioning Peter, Jesus declaring exemption, Jesus instructing Peter to catch a fish and remove a coin from its mouth and finally Jesus instructing Peter to pay the temple-tax. In our age of image-saturation people complain about the effect of the 'film of the novel' where cherished privately cultivated images are shattered by the public objective reality of the film, but one can imagine the reverse situation being true in Renaissance Florence. The realistic effect of the Tribute Money must have come through the way in which contemporary spectators could see the narrative sequence being acted out through the respective gestures of protagonists and assembled audience of which they became a part. While we might find vividness of characterization an obstacle in responding to the film of a novel we know well, renaissance artists used it as a device to capture the spectator's imagination. Vasari remarked on 'the ardour of St. Peter in his request and the attention of the Apostles in their various attitudes about Christ, awaiting His decision with gestures full of life and naturalness. St. Peter in especial, in his efforts to get the money from the fish's body, has his face quite red from bending; more admirable still is the payment of the tribute, including the representation of counting the money, and the satisfaction of the man who is receiving it, who looks at the money in his hand with the greatest delight'.24 Paradoxically, the image violates realism at the same time as enhancing it as the distinct moments in the narrative text coalesce to form one impossible moment. In a sense, the Tribute Money was highly textual in its narrative effect. The devices which Masaccio used to convey the narrative were very closely allied to the methods devised by mediaeval artists to illustrate texts. Cyclical narrative is apparent in the way in which the paintings in the Brancaci chapel were juxtaposed on top of, and next to, each other to recount episodes from the life of St Peter. Continuous narrative is apparent in the tax collector's double presence and St Peter's presence in the centre, left and right of the painting. The central part of the painting seems to me to cope with the successive events of request for tax, reflection on the meaning of the tax and instructions to Peter by a complex system of overlapping gestures: one apparent moment, in naturalistic terms, but a succession of moments in the narrative giving us direct narration. The episodes on the left and right could be interpreted naturalistically as a continuation of the action in the middle but originate in the mediaeval indirect narrative device of separating off thoughts and dreams as amplificatory images. There is, then, not one simple narrative unit but a complex narrative web suitable to an audience familiar with, and enjoying, imagining episodes from the Bible and seeing their artistic representation. The figural dimension of the image does not stand by itself but requires a text to give its composition sense. The complex relationship between image and text ties the Tribute Money into a narrative tradition which extended from the Middle Ages into the Renaissance; the point at which it departs from that tradition was in terms of its realism. Rather than pursue the ambiguities involved in talk of realism and reality it would be rather more useful to remind ourselves of differences between pictures and literary texts instead of trying to straightjacket one into the other. Gombrich has pointed out that: 'The crucial difference between 24 Vasan, Lives... , trans. A. B Hinds (1963), I, p. 268. the two lies of course in the fact that no verbal description can ever be as particularised as a picture must be. Hence any text will give plenty of scope to the artist's imagination. The same text can be illustrated in countless ways. Thus it is never possible from a given work of art alone to reconstruct the text it may illustrate. The only thing we can know for certain is that not all its features can be laid down in the text.'25 The realist text confronts the reader with concrete reality by pinning down details, like Madame Aubain's barometer of Barthesian fame; that is a matter of literary choice. But paintings cannot help but be detailed and this obviously raises questions about the appropriateness of notions of `semantic necessity' and `semantic irrelevance' to visual imagery. A text places certain requirements on its illustration, but can only do so minimally rather than exhaustively whether the image is the Canterbury window or the Tribute Money. It is simply not true that 'the visual information (the Canterbury window) emits is fully exhausted by textuality': the texts make no reference to colour, shape or position, for example, and the amount of possible variation is enormous. All visual artists had to concern themselves with detail: they had to choose to place their shapes and colours one way rather than another. Bryson's figurative dimension runs all the way throughout the spectrum of artistic activity and does not occupy an oppositional pole to discursivity. This is, of course, in marked contrast to the author, whose script has no bearing on the value of his work. For the Canterbury artist, the shape of the human body had to be sufficient to mark it off from the body of a sheep and identify it as a man or a woman; its precise characteristics depended on his own preferences in drawing and colouring bodies; his skill consisted in making. For Masaccio, by contrast, the shape of the human body had to conform to some notion of anatomical, physiognomic and expressive requirements; his skill came out through making in matching, through testing painted image against selected aspects of the visual world. Masaccio was engaged in a process of refining schemata: the mediaeval artist's technique for drawing people was insufficiently developed to mark off varying types of person; Masaccio's apostles were portrayed as recognizably different individuals, not as a clutch. When the mediaeval artist did want to mark off individuals from each other, he did so in terms of their attributes rather than in terms of specific physiognomic appearances. Of course, the figures in the Tribute Money are fictional; they are not individual characters but appear as if they are. It is ironical to find Berger and Luckmann echoing Vasari's notions on how this could be so: 'as long as my friend Henry is available in the plenitude of expressivity of the face-to-face situation, he will constantly break through my type of anonymous Englishman and manifest himself as a unique and therefore atypical individual to wit, as my friend Henry.' 26 Schematization breaks down through psychologically interactive contact and the precondition of that possibility was the artist's ability to create images which could look as if they were psychologically active. The way in which this ties into the artist's placement of shapes or lines is through Topffer's law': `any drawing of a human face, however inept, however childish, possesses, by the very fact that it has been drawn, a character and an expression'.27 While this was of relative unimportance to mediaeval artists it was of crucial importance to Renaissance artists. They knew that the apostles came from various backgrounds, were of various ages and had various temperaments; this kind of variety was not registered by mediaeval artists but Masaccio made E H. Gombrich, 'Aims and Limits of Iconography' in Symbolic Images (1972), p. 3. The Social Construction of Reality (1984), p. 46. 27 E H Gombrich, Art and Illusion (1968), p. 387. 26 25 moves in that direction. Of course Masaccio could not have known what St Peter looked like, but there was a popularly shared view of what he would not have looked like. The function of decorum was to eliminate unwanted readings of the image, to convince by exclusion but something had to be seen and that would depend upon convention. St Peter was not a young man and given his biblical authority he must have been a magisterial figure; such figures had certain physiognomic characteristics displayed in a particular anatomical type which could be found by observation. The realistic effect was generated by the particularly apt expressions of effort and ardour. In the same way, as Vasari noted, Masaccio took people off the tips of their toes and placed them on their feet by accurate drawing in perspective so he gave them appropriate expressions through controlled use of the lines and shapes which portrayed expressive gesture. I think this gives an added sense to a passage in Bruni's first Dialogue of 1404: 'What would you say of a painter who would claim to have such knowledge of his art that, when he started to paint a scene, people would believe that another Apelles or Zeuxis was born in their age, but when his paintings were revealed it would prove to be laughably painted with distorted outlines? Would he not deserve universal mockery?'28 The accidental expressions given by mediaeval artists to their figures through lack of interest in realistic effect would certainly have caused Masaccio some amusement. The ability to depict emotional states convincingly and consistently depends on a knowledge of anatomical structure. Alberti wondered 'who, without the greatest labour, study and care, could represent faces in which the mouth and chin and eyes and cheeks and forehead and eyebrows all accord together in grief or hilarity? All these things, then, must be sought with the greatest diligence from Nature. '29 There was considerable room for improvement, not just on the way that mediaeval artists portrayed the human body but also Masaccio and his immediate successors. As Vasari observed: 'They did not attain to the zenith of design, because, although they made their arms round and their legs straight, they were not skilled in the muscles, and lacked that graceful and sweet ease which is partly seen and partly felt in matters of flesh and living things, but they were crude and stunted . . .'.30 If the compilers of mediaeval bestiaries had let language determine their understanding of animal behaviour, and thus let convention dictate understanding, Leonardo used l nguage as a means for a 31 developing his understanding beyond the current state of knowledge. Reality had dropped out of sight in the bestiaries but re-emerged in Leonardo's scientific writings and his paintings. Michelangelo's contribution to the study of anatomy operates as a warning, however, to those who would want to understand perception as a consequence of linguistic categorization. The mediaeval study of anatomy is a semiotician's delight. The teacher read a commentary on the organs of the body while his students watched a dissection. It was a classic case of a text adumbrating a sign system which was not confirmed by actual observation but ritual learning; the anatomists of antiquity had actually made mistakes in locating human organs which passed undetected until the Renaissance. Recent research has shown that Michelangelo drew attention, through his anatomical studies, to aspects of human anatomy which are only just beginning to E H. Gombrich, 'From the Revival of Letters to the Reform of the Arts' in The Heritage of Apelles (1976), p. 107. 29 AOP, p. 81 30 Vasari, op cit , II, p . 152. 31 E H Gombrich, 'The Form of Movement in Water and Air' in The Heritage of Apelles. 28 receive linguistic categorization. According to James Elkins- 'the available medical surface anatomy nomenclature of roughly six hundred terms had to be increased to over eight hundred terms in order to encompass the details of the human form that Michelangelo represented. About fifty of the forms were so subtle that they had no names in art anatomy or medical anatomy.'32 However real Michelangelo's anatomical studies were, one look at the Last Judgement would confirm its utterly fictional nature as it is only there that one would see such a collection of figures organized in such a way. For a number of people Michelangelo's overtly displayed understanding of anatomy became an obstacle to their appreciation of his art. This comes out very clearly in Dolce's Dialogo della Pittura of 1557 which considered the relative merits of Michelangelo, Raphael and Titian. Dolce preferred Raphael's world of grace and gentility to Michelangelo's heroic vision arguing that 'a delicate body ought to take precedence over a muscular one'.33 By focusing on particular aspects of the appearance of the world and creating images which are consistent with those aspects artists displayed their style. Belief in 'emergent realism' does not entail belief in the 'Essential Copy' and it is hard to imagine anyone, least of all Gombrich, holding such a view. Refusal to recognize the ways in which artists discovered and represented reality and an unwittingly academic attachment to narration lead Bryson to strange intellectual gymnastics in the case of a Netherlandish still-life painting: 'It is by so thoroughly avoiding the sentence, and by perpetually excluding the narrative verb, rather than by its approach to an Essential Copy, that the still-life appears to possess a vraisemblance which history or textual painting lacks: distance from the textual is interpreted as approximation to the real'.34 One has to admire his heroic efforts to keep his system intact and to ignore the obvious. The rule forbidding movement of naked bodies on stage was based upon the recognition that freezing action was a major obstacle to emotional involvement; from that point of view films are more engaging than paintings, which in turn are more engaging than photographs. But to freeze the momentary illumination of a bowl of fruit and to portray in fine detail its rich variety of textures activates those sensitivities by which, according to Gibson, we see the real world. No other art appeals so much to the sheer delight of visual revelation. As Gibson put it: 'Many of the great classical painters, especially those Dutch painters who worked with magnifying glasses and the' finest of brushes, could simulate velvet, satin, the texture of flower-petals, and even the peculiar sheen of a drop of water on the flower by the precise arrangement of spots of pigment. The microstructure of the paint was quite different from the microstructure of the real fabric, the real petal, or the real water-drop. What the painter could reproduce was the microstructure of the light reflected from these surfaces. Qualities of lustre, softness, hardness, wetness, and the like are very clear in these paintings. 35 Too great a preoccupation with narration is a legacy of the academic tradition. The indiscriminate use of a narrative model in analysing Italian renaissance art can lead to misunderstanding and incomprehension.36 Vasari had difficulty, for example, in applying his Florentine notion of istoria to Giorgione's paintings: 'I cannot discover what they mean, whether 'Michelangelo and the human form his knowledge and use of anatomy', Art History 7 (1984), p. 177. M W. Roskill, Dolce's 'Aretino' and Venetian Art Theory of the Cinquecento (1968), p. 143. 34 WI, p. 24 35 The Perception of the Visual World (1950), p. 65. 36 See C. Gilbert, 'On Subject and Not-Subject in Italian Renaissance Pictures', Art Bulletin 35 (1952); E H Gombrich, 'Tobias and the Angel' in Symbolic Images (1973) and S. Ringbom, op. cit. 33 32 they represent some ancient or modem story, and no one has been able to tell me'.37 Perhaps the explanation is that Giorgione did not paint istoria but poesia: the Tempest does not represent a story but a mood and is a form of visual lyric poetry. Other non-narrative paintings formed the subject of religious devotions and verbal wit, but there is not the time to consider them here. Failure to grasp the historical realities of artistic production, coupled with a desire to be modern if not modernist, in analysing paintings has led Bryson into anachronistic readings to prove non-existent subversions of the medium. Jon Whitely has argued 'Bryson does his case no good by overlooking or dismissing the related evidence in his haste to prove that what appears to him as formalism cannot also have discursive relevance. The picture within a picture in Vermeer's Lady seated at her Virginals has not been "altogether cancelled" as Bryson states, but has been identified by Baburen as a lutanist with a procuress and is so conspicuous that it must, in some sense, allude to the main theme of music making. The claim that Boucher never illustrated "actions or events" is particularly inappropriate when placed, as it is here, against the Wallace Collection Venus and Mars surprised by Vulcan, while the suggestion that the eyes of Mars, staring in fright at the sudden appearance of a very nasty Vulcan, are expressing ecstasy, is not what most people would imagine in front of this gripping narrative. A modicum of research into Gros's Eylau would have told Bryson that the figures of the wounded and dying soldiers in the foreground do not, as he believes, undermine the official purpose and threaten the discursive with formalist reality but were included to make sense of an official request for a picture of Bonaparte visiting the "innumerable victims" on the battlefield. It is not a picture celebrating victory, as Bryson thinks, but a picture of Imperial compassion. 38 Anita Brookner, while expressing admiration for an 'excellent, beautifully written, well argued, and distinctive' book and concluding that his system is more intelligent than any other she knows used the bulk of her review to lay waste to his conclusions. Her observations on Watteau deserve quoting: 'Bryson announces the necessity of "deconstituting" the Watteau myth, but proceeds to do so in what seems to mean unnecessarily heroic manner. He refers to Watteau's "discursive strategy" of posing textual questions and not answering them, of providing dramatic clues and leaving their solution to the spectator. The example he chooses to illustrate his point is the Louvre Meeting in a park, in which he sees all manner of elliptical meanings and intentions. My reading of this example is entirely different. I see it as a pure landscape, on to which Watteau has grafted a random scattering of figures, at a later date, and with a much dirtier, more negligent brush. Whatever the date of this picture, it is clear that the method has become a kind of shorthand, and possibly the element in which Watteau had the least interest, but which had proved popular and marketable. Of course, this line of reasoning will not answer questions about the random placing of figures, their pantomime of desire, their extreme rotational movements, and because these elements continue to intrigue the spectator, Bryson's solution is the more satisfying. If Watteau can be proved to be a strategist, albeit a subversive one, then the historical and logical order is preserved. There does, however, lurk in my mind the notion that as Watteau was not trained in the Académie, he may not have absorbed its teachings, and that his originality (which was seen by his 37 38 Vasari, op. cit., II, p. 170. 'The Trouble with Words', Art Book Review 1 (1982), p. 39. contemporaries as eccentricity) may have been the sheer absence in his life and work of academic conditioning and indeed of sustained reasoning.'39 Which reminds me of the saying that what the French regarded as English diplomatic deviousness was, more often than not, the product of bumbling incompetence. Not that Watteau was an incompetent painter, but that it would be false to judge his work by intentions it is very likely he did not have. I have argued that Bryson was wrong to think that the textuality of the Canterbury window effaced its materiality. At the other end of the spectrum I would suggest that he has got it wrong about the pure figurality of Pollock's Enchanted Garden and that there is a germ of truth in Tom Wolfe's remark that 'Modern art has become completely literary: the paintings and other works only exist to illustrate the text'. While art criticism provided powerful support for developments in the avant-garde in the late nineteenth and early twentieth century, literary theorizing turned into a stimulus for artistic development culminating in the rapid stylistic developments in America in the post-war period culminating in minimal and conceptual art and the 'Art of the Real'. Without its critics to explain what it was about, contemporary abstraction would be utterly baffling, signs without meanings if indeed they could be called signs at all, 'not easily understood by the mute perception of sight without description'. III Lurking underneath Bryson's sophistication there is a fundamental naivety in his attachment to a continuum dominated at one end by discursivity and at the other by figurality. Preoccupation with visual appearance is at the heart of artistic practice, through history and across cultures. In constantly associating discursivity with textuality, Bryson is an unwitting victim of the academic tradition. Perhaps the word `discourses' would be the best to describe the various social practices which have given meaning to the visual arts: written texts, oral traditions, dramatic representations, pictorial conventions, modes of thought about subjects of social concern. While there is no text which Jackson Pollock's Enchanted Garden illustrates, there are discursive strategies which legitimate its existence as art and which give it meaning. This paper was read at the Nineteenth Annual Conference of the British Society of Aesthetics held in London in September 1985 39 Burlington Magazine CXXV (1983), p. 763