Richard Woodfield, 'Expression, Form and Kunstwissenschaft' more

Expression, Form and Kunstwissenschaft* Richard Woodfield From time to time, one has an experience which throws into relief fundamental questions of understanding and self-understanding. This happened to me recently at a performance of Bartok's Bluebeard's Castle at the Opera House in Budapest. Being completely unfamiliar with the story I was completely baffled. It wasn't just that I couldn't understand the words; it was that I couldn't make any sense at all of an activity which was going on under my nose. Visitors to art galleries rarely suffer such bafflement. It's not necessary to speak Chinese to experience the delights of looking at a Southern Sung landscape painting. It is this ability of the spectator to appreciate the depiction of things which led to the thought of painting as a universal language. Pope Gregory I famously declared: `What Scripture is to the educated, images are to the ignorant, who see through them what they must accept; they read in them what they cannot read in books.' 1 In the history of aesthetics, Addison set the agenda for debate by suggesting that `[c]olours speak all-Languages, but Words are understood only by such a People or Nation'.2 Jonathan Richardson went further: Words paint to the Imagination, but every Man forms the thing to himself in his Own way: Language is very Imperfect: There are innumerable Colours, and Figures for which we have no name, and an infinity of other Ideas which have no certain Words universally agreed upon as denoting them; whereas the Painter can convey his Ideas of these Things Clearly, and without Ambuiguity; and what he says every one understands in the Sense he intends it. And this is a Language that is Universal; Men of all Nations hear the Poet, Moralist, Historian, Divine, or whatever Character the Painter assumes, speaking to them in their own Mother Tongue.3 This is, to say the least, a bold claim but it's one which lies at the heart of the old practice of Kunstgeschichte als Geistesgeschichte and the current `New Art History'. Works of art are treated as the expression of an epoch or class and read for the kind of information that they can offer. Some prying may be necessary, but there is no denying what is under one's nose. It goes without saying that different things are obvious to different people. Gombrich has reminded us of an observation made by the ancient sage Apollonius of Tyana: We know from Philostratus' life of Apollonius that there was an archaic statue of one Milo in Olympia, standing on a disc with his two feet close together; in his left hand he grasped a pomegranate; the fingers of his right hand were extended and held tightly together. `The * Published in Aesthetic Matters: Essays presented to Göran Sörbom on his 60 birthday edited by Lars-Olof Åhlberg and Tommie Zaine, Uppsala 1994. 1 `St. Gregory the Great to Bishop Serenus of Marseille', in C Davis -Weyer, Early Medieval Art 300 - 1150 Englewood Cliffs, 1971), p. 48. 2 Interestingly, he went on to add that painting was more imperfect than speech `because it is impossible to draw the little connexions of Speech, or to give the Picture of a Conjunction or an adverb' (Addison, Steele and Others, The Spectator, No. 416, London, 1973, p. 291). 3 J. Richardson, Snr, An Essay on the Theory of Painting (London, 1725), pp. 3-4. th people of Olympia thought that these features showed Milo to have been so inflexible and firm that he could never be induced to budge from the spot where he stood; and that is the meaning of the clenched fingers ... and why they look as if they could not be separated ... however much you struggled'.... Apollonius knew better. He told his guides that these puzzling traits were due to the archaic style of sculpture.4 At one level we could say that this example illustrates the possibility of simple misunderstanding, like misunderstanding a person's behaviour. At another level, it shows a fundamental failure to grasp a cultural difference which prevented the possibility of a correct understanding. At both levels interpretation played a role: the people of Olympia knew that they had to make sense of the sculpture and so did Apollonius. The difference was that Apollonius recognised the importance of mediation in the production and perception of works of art. Such mistakes were not confined to the people of Olympia, however. At the time when the teenage Gombrich was inducting himself into the mysteries of art historiography5 the artists and art historians of the Expressionist era were busy using works of art to penetrate the depths of the human psyche. Max Dvorak was instrumental in leading Gombrich into pursuing art history as an academic discipline. Looking at gothic art revealed the conflict between realism and nominalism in the conflict between idealism and naturalism. Looking at The Burial of Count Orgaz revealed `something hard and at the same time magnificent about these faces, a hint of ascetic egotism and traditional love of metaphysics, a spirit capable of understanding that which is hidden from the senses'.6 It didn't take Gombrich long to find out that the historical evidence not only failed to confirm, but also invalidated, the expressionists' inferences. It has, however, been a constant source of interest for him to fathom why they were made in the first place. Needless to say, his views are complex, having emerged over a number of years in a string of articles. His earliest forays into the problem turned on what he called the physiognomic fallacy, which is that one might diagnose a mental state from a visible structure, originally a head; his essay, `On Physiognomic Perception',7 opened with Lichtenberg's wicked parody of a physiognomic analysis of a dog's tail. Given the view that there was such a thing as the Spirit of an Age and that works of art were its visible manifestation, then they give the same kind of insight into that age as do a person's clothes into her character. The fallacy involved is obvious: clothes are only a giveaway when they are made to be a giveaway, when there is a choice. The character of the Danish language, for example, is not expressive in itself: it is only sentence utterances within the language which may be treated as being expressive.8 E.H. Gombrich, Art and Illusion (London, 1986), pp. 114-5. Wandlungen in der Kunstbetrachtung (Von Winckelmann bis sur Jetzeit): Hausarbeit des Ernst Gombrich, manuscript, Vienna 1928. I am indebted to Professor Gombrich for allowing me access to manuscript sources. 6 Max Dvorâk, The History of Art as the History of Ideas, trans. John Hardy (London, 1984), p. 97 7 Reprinted in Meditations on a Hobby-Horse, London 1963. 8 Note Gombrich's remarks in `Achievement in Medieval Art', originally published in German in Kritische Berichte in 1937, available in English in Meditations on a Hobby-Horse: Popular psychology may infer from the rough sound of a language to the rough character of those who use it. A scientific appraisal of expression, however, can only be concerned with judgements based upon the understanding of the language and the discrimination and elucidation of the individual expressive features within it (pp. 75-6.). 5 4 The fact of the physiognomic fallacy does not stop us, however, from crediting works of art with expressive characteristics. Human beings are meaning-seekers and given an image which falls to our attention we are propelled into giving it a reading. Töpffer's Law captured this process at work, `[f]or any drawing of a human face, however inept, however childish, possesses, by the very fact that it has been drawn, a character and expression'.9 Gombrich's first research project after leaving university was on the expressive features of the statues of the founders in the Cathedral of Naumburg: These lifelike but imaginary portraits appeared to be so full of expression that a whole drama had been woven around them. Ciceroni had developed the legend that all these figures were participants in a story of conflict and murder.10 There was no doubt that the sculptures had enlivening features, in comparison to earlier sculptures, but `their expression was more complex than clear'.11 Empirical investigation into spectators' responses concluded that there was not sufficient consensus on the interpretation of the expressions on individual sculptured heads to justify the belief that they did have specific expressions. There could be even less justification for the incorporation of the figures into a dramatic plot. As Gombrich put it in his early essay 'Wertprobleme and mittelalterliche Kunst': `The medieval artist may very well simply have accepted the emotional overtones including the facial expressions as they happened to emerge.'12 Without knowing the context of a particular production it is easy enough to assume that an expressive reading is legitimate. In his essay `The Evidence of Images' Gombrich cited the case of an art historian who had falsely assumed an armorial bearer from a piece of furniture to be a statuette of Hercules as a Christian Knight: The eyes gaze into the distance, they stand in a face that bears the marks of hard experiences. This man is no longer a wild adventurer, he is sensitive to the suffering destiny has laid upon him; it is with sorrow that he awaits the next test, though he is sure that he will win through in the end.13 Spurred by a particular reading of the so-called Hercules' physiognomy, the author ascended into the giddy heights of his own fantasies. The fact that he thought it to be Hercules in the first place helped stimulate the reading. When, in a different context, Baxandall suggested of Botticelli's Primavera, `we miss the point of the picture if we mistake the gesture'14 he has simply got the matter the wrong Art and Illusion, p. 287. For the context of this research, which was undertaken with Ernst Kris, see `The Study of Art and the Study of Man' in Tributes (Oxford, 1984), pp. 224 ff. 11 Ibid., p. 226. 12 Originally published in Kritische Berichte 1937, translated as `Achievement in Medieval Art' and published to Meditations on a Hobby-Horse, pp. 70-77. 13 E.H. Gombrich, `The Evidence of the Images', in Charles S. Singleton , ed., Interpretation in Theory and Practice (Baltimore, 1969), p. 71. 14 M. Baxandall, Painting and Experience in Fifteenth Century Italy (Oxford, 1972), p. 70. 10 9 way around: we cannot interpret the gesture until we know the meaning of the painting, or rather the text that the painting was intended to illustrate.15 The fact that the figurative image has a multiplicity of functions should make the critic beware of working on the basis of intuition or visual impression. After all, one does not ask the maker of the traditional town-hall lions what he might have thought about their expressive features; they are just lions. At this point one may call into play Bühler's principle of `abstractive relevance': The letters of the alphabet signify through certain distinctive features but in normal contexts their meaning is not affected by the size, colour or font. The same is true of the images which interest the iconographer, be they coats of arms, hieroglyphs, emblems or personifications traditionally marked by certain `attributes'. In every one of these cases there are any number of features which are strictly speaking without a translatable meaning, and only a few which we are intended to read or translate.16 In such cases one has to be quite clear what the expressive characteristics of an image might be and how they might be recognised. It is a well known fact of human behaviour that a movement may be insignificant in one context but deeply significant in another; the same is true of the features of an image. Whether the significance is intended is another matter and depends upon mutuality of understanding. The key word in this account so far has been `understanding' but to attach such a significance to understanding may be disputed in some quarters. Wollheim referred to Gombrich's `cognitive' approach to expression, clearly implying that there were other approaches. Well, there are and Gombrich gave an account of them in his lecture `Four Theories of Artistic Expression':17 three theories of the relation between art and emotion plus his own. Historically, the first was the magico-medical theory of influence or contagion. As we listen to a piece of music or look at the statue of a beautiful woman we fall under its spell; it conjures emotions. As Gombrich noted, this was a theory of art, not artists: how the artist produced the work was irrelevant, what was important was its effect. A later, second, arrival on the scene was the dramatic theory of depiction. We look at a work of art to see the emotional workings of the mind; the more the artist's `description conformed to our own experience, the more his narrative would move us'.18 His concern with expression was one with the representation of emotion. The Romantic movement introduced the third theory that what was important was the artist's expression of his own emotions. Referring to M.H. Abrams' classic study The Mirror and the Lamp, Gombrich noted that the contrast between the notion of art as the mirror and lamp of nature meant that in: The point is admirably demonstrated in what almost stands by itself as an essay in physiognomic interpretation in a footnote to Gombrich's `Botticelli's Mythologies' in Symbolic Images (London, 1972), n. 23, pp. 204-6. 16 'Raphael's Stanza della Segnatura and the Nature of its Symbolism', p. 95 in Symbolic Images. 17 Published in Architectural Association Quarterly, 12 (4), 1980, pp. 14-19. Gombrich does not explicitly say so but the three theories relate to Bühler's three communicative functions: Symptom, Sign and Symbol. On Bühler see Gombrich's `Art History and Psychology in Vienna Fifty Years Ago', Art Journal (1984), pp. 162-4. 18 Ibid., p. 16. 15 the new theory of the Romantic movement, the artist is like a lamp; he sends the rays of his feelings into the world where they are received by the public who will turn to this source of light. His light is his art, whether poetry, painting or music; and when we speak of artistic expression, we mostly mean the expression of the artist which is embodied in the work of art.19 This idea, of art as self-expression, is now taken as the norm by which we think of works of art but, to re-emphasise the matter, it is historically recent. Gombrich's view of the Romantic thesis, which may be generalised to include the other theses, is that art cannot be as simple as that. Invoking magic or medicine is simply not good enough to account for art's arousal function; invoking the notion of the mirror onto nature is not good enough when art simply does not do that and invoking the inspirational thesis does no justice at all to the complexity of the art work. A theoretically adequate account of art's expressive powers should recognise the historical determinants of the theories described so far but at the same time account for the effectiveness of those works of art produced in different historical conditions. In this respect there is little difference from the production of a scientific theory which must redescribe and incorporate the successes of earlier theories. Gombrich's own theory subsumes the previous three in a model of greater generality. Characterising the first three theories as centrifugal, with feelings being expressed into symptoms, signals and signs, he called his own `centripetal': `a "theory of feedback" ... which stresses the constant interaction between the feeling and form, the medium and the message.'20 As language offers the poet `the means to shape his feelings or thoughts into an artistic creation'21 so too do visual forms offer possibilities to the figurative artist which may then be exploited. In his search for self-discovery `the true artist will encounter feelings to which he is attuned, emotions he can authenticate by his own heart'.22 The culture into which the artist is born offers a visual resource which he can use, subject to the constraints imposed by that culture. It is, of course, the notion of a use for art within a culture which is the key element in Gombrich's account of expression. In a draft for their unpublished book on caricature, Gombrich and Kris had stated categorically: it is only the general context which allows us to say in certain conditions that here we were meant to laugh, and there to meditate.... It is one of the main theses of this book, which cannot be sufficiently emphasised, that there is no unambiguous `language of form' such as artists and aestheticians have sometimes dreamed of. Any such language only comes into being in an institutional context.23 It is not until one can determine `what is going on' in an image that one can determine its expressive features. At one moment a motif may evoke horror, at the next it may be the subject of fun. In 1939 Ibid., p. 17. Ibid., p. 18. 21 Ibid. 22 Ibid., p. 19. 23 Unpublished MS (1938), `The Caricature Style', p. 3. I am indebted to Sir Ernst Gombrich for giving me access to this material. 20 19 Kris and Gombrich described the horrific nature of Bosch's imagery: `The primary and original meaning of medieval drollery was restored to its threatening and terrifying effect.' In 1967 Gombrich reported the existence of a text dating from 30 July 1517 describing Bosch's paintings as being `so pleasing and fantastic'. He commented, à propos the unpleasant behaviour of Henry III of Nassau, the owner of Bosch's Garden of Earthly Delights: It is perhaps well to remember this background of cruelty and coarse humour if we are to see Bosch as his patrons saw him For the visitors from Italy, at any rate, Bosch's gruesome inventions were noteworthy as so pleasing and fantastic things (cose tanto piacevoli et fantastiche), amusing grotesques which they enjoyed. Antonio de Beatis looked at some of the bizarre details, but does not seem to have searched for the meaning of the whole. The tone of his description suggests entertainment rather than horror or anxiety. He was not alone in this reaction, which represents a frequent attitude to Bosch, who was called inventor of comic monsters (grillorum inventor) in the sixteenth century and more surprisingly der Lustige (the humorous) in the early nineteenth.24 As any number of modern writers might have thought of Bosch's paintings as embodying the qualities of a nightmare or a surreal experience, it comes as a surprise that his contemporaries might have thought of it as good clean fun. Perhaps in future years the popular cartoon book100 Ways of Killing a Cat may undergo such a critical revision. To understand the nature of Bosch's imagery and to appreciate the nature of his inventiveness it is not necessary to seek some correspondence between those images and his state of mind. The fact is that we do not know what his state of mind was and to try to deduce it from his imagery leads into a circular argument. It is rather more to the point to consider Bosch's inventiveness in drawing upon traditional forms of image, particularly the drolleries which exploded in the margins of late medieval texts. Psychological explanations, involving notions of estrangement or mental disorder, have less to do with the matter than wit. And wit, we know from reading Freud, is a product of what is possible in the public domain, available from the resources of language. As such it stands in contrast to the private world of the dream.25 It is not very often that we can discover when an artist thought that he might have gone too far and pulled back. Gombrich has given us two examples in his review of the exhibition catalogue The Great Age of Fresco: Giotto to Pontormo:26 a sinopia for an Annunciation at the Oratory of San Galgano at Montsiepi and Dürer's woodcut of The Agony in the Garden. It is at this point that it would be appropriate to consider Wollheim's objections to Gombrich's account of expression which, I believe, results from a misperception of Gombrich's standpoint. It is clear from Gombrich's own account that it would be a mistake to regard all expressive features of works of art as a product of the artist's self-expression. The notion that self-expression has any bearing at all on an account of the artist's activities is modern, and down to Romanticism. Of course one might say that all human beings express themselves in what they do, but this is trivially true and just as applicable to philosophers, mathematicians, bricklayers, shop assistants and artists. E.H. Gombrich, `The Earliest Description of the Triptych', in The Heritage ofApelles (Oxford, 1976), pp. 80-81. On this point see Gombrich's `Verbal Wit as a Paradigm of Art' in Tributes (Oxford, 1984), p. 105. 26 Republished as `New Revelations on Fresco Painting' in Richard Woodfield, ed., Reflections on the History of Art (Oxford, 1987), pp. 46-53. 25 24 Although Plato had provocative things to say about the activities of the rhapsodes in Ion, Xenophon's Memorabilia made clear that the sculptor's ability to represent `the feelings of bodies' was a product of his ability to produce convincing figures which, in turn, were the product of `pulling down and pulling up' and `squeezed up and opened wide and stretched out and slackened'.27 This was a product of marvellous technical accomplishment. It may be wondered what induced the sculptors to engage in this slackening and squeezing and Gombrich's surprising answer is that it may have been a product of auto-mimesis on their own part. I must hasten to add that he has not said this himself, but it is a conclusion that I draw from his essay on the mask.28In discussing the art of portraiture he said: . it so happens that I had the privilege of listening to Kokoschka when he spoke of a particularly difficult portrait commission he had received some time past. As he spoke of a sitter whose face he found so hard to unriddle he automatically pulled a corresponding grimace of impenetrable rigidity. Clearly for him the understanding of another person's physiognomy took the way over his own muscular experience.29 While it would be quite wrong to think of the Greek sculptors producing a slice of life, as opposed to simulating life, nevertheless it is possible to think of the sculptor as inwardly testing his own capacities of physical expression against a relatively lifeless image that he had seen. The notion that there may be a muscular dimension to this is registered in the work of Franz Xaver Messerschmidt, who was the subject of an early study by Kris.30 But the inducement of a particular kind of muscular behaviour is hardly the same as having the mental state; it is a matter of imagining oneself to be party to a mental state, and that is a matter of role-playing. It is a fairly murderous task to attempt to reply to Wollheim's critique of Gombrich in Art and its Objects, particularly because what he calls the `Gombrich argument' is actually `a reconstructed, and here and there a simplified, version of what is to be found in Art and Illusion and ... Meditations on a Hobby-Horse'.31 I am reluctant to complicate the situation even further by paraphrasing Wollheim, particularly because of the emphasis he appears to lay on artistic self-expression. As I have already suggested to say that all human beings express themselves in what they do is trivially true, as it is to say that they cannot express themselves when they are not allowed to do what they want to do. But when Wollheim talks about choice he does so in such a way as to make it one within a repertoire and a specifiable one at that. In one respect this is helpful and in another it is not. One of Gombrich's favourite analogies is between the traditional artist and the musician. In his essay `Achievement in Medieval Art'32 he wrote: our medieval artist's product should ... be thought of in the way that we today think of the performance of musicians and actors. The text is there, the manner of presentation the I owe my interest in these figures of expression to Göran Sörbom's invaluable study Mimesis and Art (Uppsala, 1966). 28 `The Mask and the Face: The Perception of Physiognomic Likeness in Life and Art' republished in The Image and the Eye (Oxford, 1982), pp. 128-133. 29 Ibid., p. 133. 30 See `A Psychotic Sculptor of the Eighteenth Century' in E. Kris, Psychoanalytic Explorations in Art (New York, 1952), pp. 128-133. 31 Richard Wollheim, Art and its Objects, 2nd. ed., (Cambridge, 1980), p. 56. 32 See note 12. 27 charging of that text with meaning is what constitutes the aesthetic achievement.... The analogy is not complete, but perhaps it throws some light on the possibility of achievement within an art where what counts is neither a fresh mastery of natural appearances not the creation of new formal structures.33 Again, and much later, in a 1967 review,34 he wrote: The Renaissance artist who represented The Annunciation or Apollo and Daphne did not expect the beholder to read off the story from the picture. His skill and inventiveness lay in presenting a familiar episode in a novel and convincing way. . . . Hence such representations were frequently judged by contemporaries as we might judge the st'ging of a familiar classic, the way, for instance, in which the difficulties of the Balcony Scene in Romeo and Juliet are resolved by an imaginative producer.35 There was, of course, a world of difference between the possibilities open to medieval and renaissance artists. If the medieval artist's concern was to preserve the motif whilst at the same time demonstrate his ski11,36 a central question must concern the nature of the demonstration of that skill. And that question is not always easy to answer. Take Gombrich's example of the Evangelist in the Schatzkammer in Vienna: if it were based on a motif like the Ébon Evangelist, what would the difference convey? Recognising, again to use Gombrich's example, that `[e]ven someone quite unmusical can see something "appassionato" in the manuscript of Beethoven's Appassionata'37 his note to Schindler about his nephew's shoes38 demonstrates that he simply wrote like that. What can we infer about the drawing of the Evangelist? Without any further evidence, the answer is simply nothing. One can, however, make a distinction between expressive representations and the representation of expression in medieval art. Being essentially pictographic its purpose was to signify events with maximal clarity and insofar as it did that it used expressive gestures that had their roots in ritual.39 Once a narrative fell into the hands of a renaissance artist a concern with the `what' was replaced with one for the `how' and an element of what Gombrich referred to as `tone', `which conveys character and emotion; it can be tense or relaxed, urgent or calm',40 entered into the representation. We are on firmer ground with more recent artists because there is more evidence to draw upon. We know from contemporary documentation which skills were prized and how artists responded to the demands placed upon them. Now this may seem hopelessly `cognitive',41 but it is and it is a demand upon art historical scholarship. Leonardo criticised a contemporary for portraying: Ibid., p. 72. Reprinted in Reflections on the History of Art as `Painted Anecdotes', pp. 152-159. 35 Ibid., p. 155. 36 . An English introduction to the contemporary admiration of medieval artists' skills is C. Barrett, `Medieval Art Criticism', The British Journal of Aesthetics, 5 (1965), pp. 25 -36. 37 `Achievement in Medieval Art', p. 73. 38 Given as illustration 28 in `Watching Artists at Work' reprinted in E.H. Gombrich, Topics of our Time (London, 1991), p. 98. 39 See `Ritualized Gesture and Expression in Art' in E.H. Gombrich, The Image and the Eye (Oxford, 1982). 40 E.H. Gombrich, `Action and Expression in Western Art', in The Image and the Eye, p. 91. 41 Wollheim's expression, though I've lost the source. 34 33 an angel who looked as if in the annunciation he wanted to chase Our Lady out of her chambers with gestures which looked as offensive as one would make towards the vilest enemy, and Our Lady looked as if she wanted to throw herself out of the window in despair.42 If we are to decide that the artist was simply inept, as opposed to being interestingly expressive, we have to see his other paintings; this establishes what Wollheim called the `repertoire'. As Gombrich suggested Leonardo might have felt Botticelli's later style to be possessed; in contrast, Ghirlandaio `a foil of solid immobility to Leonardo's gestures' might have felt that Leonardo's figures `gesticulated like mad'.43 These differences are not unlike the kinds of differences that we may encounter amongst our friends and acquaintances. One needs to get to know people reasonably well before one accepts their behaviour at face value and one can be interested or entertained by nuances of their behaviour. It would be wrong to think that nuances can be located within one scale, even at the simplest level.44 It seems to me that Wollheim has failed to pick up Gombrich's qualification of his use of Osgood's notion of `semantic space' as a crude model. The scale is not two- but multidimensional where a variety of factors may play themselves off against each other. Colours may be interpretable across a spectrum, but when a red is not the red of an apple but an ordinary tomato `our response would change'.45 Innovations can change our perceptions, only to be lost in a moment of time: after looking at an enormous number of abstract expressionist paintings in the late '60s, I was quite moved to see my first hard-edged abstraction. In the early '60s my liking for Piet Mondrian was in tune with my taste for the MJQ. But such things fade and pass away. I suspect that Wollheim would like more certainty in a situation where interpretation is governed by possibilities and probabilities. I find the admission of possible ignorance quite tolerable. Quoted by Gombrich in `Action and Expression in Western Art', republished in The Image and the Eye, p. 92. Ibid., p. 94. 44 Wollheim, Art and its Objects, section 28. 43 42