Languages of Art at the Turn of the Century

Languages of Art at the Turn of the Century Author(s): Jenefer Robinson Source: The Journal of Aesthetics and Art Criticism, Vol. 58, No. 3 (Summer, 2000), pp. 213-218 Published by: Wiley on behalf of The American Society for Aesthetics Stable URL: http://www.jstor.org/stable/432102 Accessed: 07-07-2017 14:38 UTC JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://about.jstor.org/terms The American Society for Aesthetics, Wiley are collaborating with JSTOR to digitize, preserve and extend access to The Journal of Aesthetics and Art Criticism This content downloaded from 129.137.215.253 on Fri, 07 Jul 2017 14:38:41 UTC All use subject to http://about.jstor.org/terms Symposium: The Legacy of Nelson Goodman Jenefer Robinson Languages of Art at the Turn of the Century I. INTRODUCTION It is hard to overemphasize the impact of Nelson tecedently existing world but help to create new Goodman's 1968 classic, Languages ofArt.1 The ones: book offered a powerful new vision of aesthetics grounded in analytic philosophy of language, Any notion of a reality consisting of objects and events which reframed many of the questions being and kinds established independently of discourse and asked in aesthetics and gave original, ingenious, often eccentric answers to them. Goodman reconceived works of art as symbols in symbol systems, and treated representation and expression in the arts in terms of semantic concepts such as reference and denotation. The book was unrelentingly systematic, highly polemical, and written in a trenchant, witty style. The reader is constantly aware of the author's formidable intelligence as well as his unwillingness to suffer foolish theories gladly. As soon as it was published, Languages of Art was both admired and attacked, and remarkably, it has continued to be admired and attacked until the present day. For more than thirty years no one who has defended an aesthetic theory with any claim to generality has been able to ignore Languages of Art. unaffected by how they are described or otherwise presented must give way to the recognition that these, too, are parts of the story.4 On the other hand, the post-structuralist notion that meanings are constantly in flux, in accordance with the principles of diffrrance and deferral of meaning, is inconsistent with Goodman's nominalism, which holds that terms genuinely refer to objects, events, and kinds. Goodman is not a skeptic about meaning. As I once heard him say, "Derrida deconstructs worlds, whereas I construct them!" Broadly construed, Goodman's view has prevailed. Works of art are now commonly understood as meaningful entities, with cognitive value, which require interpretation rather than passive What is the status of this book at the turn of the appreciation. But Goodman tried to defend his millennium? At the time it was written, aesthetview in terms of a narrow semantic theory, which ics was not a very lively field. Arguably, the only abstracts from the psychology of artists and auenduring masterpiece of analytic aesthetics from diences, and from the historical context of artthis era is Monroe Beardsley's Aesthetics: Probworks, and which is consistent with his nomilems in the Philosophy of Criticism2 from 1958. nalism. Given the constraints that Goodman sets Languages of Art, like its near contemporary, for himself, it is extraordinary how much the Wollheim's Art and Its Objects,3 was concerned theory manages to accomplish. Nor is it surpristo debunk both Beardsley's Dewey-flavored eming that it has been the butt of so much criticism. piricism and the still influential idealism of Croce and Collingwood. For Goodman artworks are II. REPRESENTATION symbols, which refer to the world by virtue of what symbol they are in what symbol system. Like Perhaps the favorite target has been his theor the Saussurean structuralists, he claims that artpictorial representation, or depiction. On Goodworks are signifiers in systems or structures of man's view, if a picture p represents an object o, signs. Also like the structuralists, he argues that then, providing that o exists, p denotes o. If o does language and art do not merely reflect an annot exist, then to say thatp represents o means that The Journal of Aesthetics and Art Criticism 58:3 Summer 2000 This content downloaded from 129.137.215.253 on Fri, 07 Jul 2017 14:38:41 UTC All use subject to http://about.jstor.org/terms The Journal of Aesthetics and Art Criticism 214 p is a picture of a certain kind, an o-representing- Now it is certainly true that Goodman does picture. In this case p does not denote o, but the not speak explicitly about the perceptual abili- phrase "o- denoting-picture" denotes p. Similarly, ties that understanding pictures requires; yet it is a linguistic description of an object o either de- also true that he points out how the task of deciphering pictures is a task of discriminating char- notes o (if o exists) or is an o-denoting-description. The difference between pictorial representation acters in a system that is syntactically and se- and linguistic description resides only in the dif- mantically dense and relatively replete. Pictures ferent kinds of symbol systems that they are in: cannot be "read" since reading is necessarily in whereas a language is a syntactically disjoint and a language; pictures require acts of perceptual differentiated system (we can always distinguish discrimination. Goodman's theory here, as else- one character from another), a pictorial symbol where, is abstract and schematic. He gets us to system is syntactically dense and undifferentiated rethink such basic aesthetic concepts as repre- (between every two characters there can be a sentation without spelling out all the implica- third), and relatively replete (every aspect of the tions of his view. After all, there is no inherent pictorial symbol-its colors, lines, brushwork, contradiction in believing both that representa- etc.-is constitutive of it as a symbol). tion should be analyzed in terms of reference Goodman's theory has been salutary in show- and that representation requires seeing-in (or ing how simple-minded it is to think of pictur- some such perceptual ability). Gainsborough's ing as mirroring or copying or imitating. Under- picture of his daughters refers to his daughters. standing what a picture or a description refers to How it does this is by getting us to see the daugh- is always a function of the system of symboliz- ters in the picture. Likewise a picture of a unicorn ing within which it functions. Just as the noun is a unicorn-picture, but how we classify it is by "Boot" is a boat-description in German and a seeing a unicorn in the picture. Perceptual theories boot-description in English, so a portrait might be of representation can be construed as filling in the an ordinary-woman-picture in a Cubist symbol gaps of Goodman's very schematic account. system and a grotesque-woman-with-a-serious- On the other hand, because of his nominalist eye-deformity-picture in the symbol system of preconceptions, a picture that "represents a so- late-nineteenth-century academicism. Goodman's and-so" is always ambiguous for Goodman be- view is also salutary in that it emphasizes the tween being a so-and-so-picture and denoting a need to understand pictures, just as we have to so-and-so. The "perceptual" theorists of repre- understand language, and that understanding pic- sentation have the advantage in that they treat tures requires mastery of their pictorial "lan- representation as a univocal concept: if a picture guage." Finally, Goodman is also right to point represents a so-and-so, we (in imagination) see out that a realistic picture cannot be defined as a so-and-so in the picture, whether or not the pic- one that imitates the world particularly closely ture refers to any actual so-and-so's. This seems right. It is true that Gainsborough's portrait of or accurately. After all, there are no fixed criteria for realism; a wide variety of pictorial styles from his daughters picks them out or refers to them, van Eyck to Courbet count as "realistic." But there are many problems. One of the main whereas another picture of two little girls does not refer to anybody, but it does not follow that complaints has been that Goodman ignores what "representing" and "describing" are ambiguous artists and spectators have to do psychologically terms. It is just that sometimes a representation or perceptually in order to grasp what a picture or a description is satisfied and sometimes it is depicts. Spectators are not merely classifying not. The word "represents" means the same thing pictures as two-little-girl-pictures or unicorn- when we say that a picture represents Gains- pictures, or figuring out what a picture refers to borough's daughters and when we say that a pic- (Gainsborough's daughters or nothing); they are ture represents two little girls, although in the first seeing two little girls or a unicorn in the picture case the daughters are denoted and in the second case the little girls are not. Here is one of many (Wollheim), or they are imagining of their see- ing a picture that it is a seeing of two little girls instances where Goodman's nominalist seman- or a unicorn (Walton).5 We understand pictures by (imagining) seeing things in them, not by decoding them. tics leads him to counterintuitive conclusions. Goodman has also been criticized for ignoring the important perceptual constraints on real- This content downloaded from 129.137.215.253 on Fri, 07 Jul 2017 14:38:41 UTC All use subject to http://about.jstor.org/terms Symposium: The Legacy of Nelson Goodman Robinson, Languages of Art 215 ism. For Goodman, realism in painting is a mat- lows: a work of art a exemplifies a predicate p if ter of the familiarity of the symbol system within and only if p denotes a and a refers to p. Expreswhich the picture is made. Yet, however familiar sion is analyzed as metaphorical exemplifica- some symbol systems become (such as the sys- tion, i.e., the predicate in question denotes the tem of Analytical Cubism), they still do not be- work metaphorically rather than literally. come realistic for us unless they permit us to see As in the case of representation, the concept readily in a picture what it represents. On this of exemplification, though much vilified, has im- issue, too, the "perceptual" theorists seem to come portant virtues. First, it elegantly captures the out ahead. way in which many artworks reveal, show forth, Finally, Goodman's concept of representation or "embody" the very themes and qualities that ignores the historical and cultural context in which they are about. Munch's well-known painting, a picture originates. Goodman argues that The Scream, for example, exemplifies its swirling shapes, lurid colors, dramatic contrasts, and pow- studied quite apart from the acts or beliefs or motives erful brushwork. It also (metaphorically) exemplifies feelings of anguish and alienation. The of any agent that may have brought about the ... refer- picture portrays the chief character as anguished the characteristics and functions of symbols ... can be ential ... relationships involved.6 and alienated; at the same time the picture itself incorporates qualities of anguish and alienation. In saying this, Goodman is a child of his times. Thus the composition of the picture-with the In 1968, when a formal theory was used to model chief character squashed into the lower right parts of natural language, it would normally abstract from speakers' intentions and study syntac- away from the figures at the other end of the tic and semantic regularities that are independent of contextual factors. However, there are aspects corner, pushed up against the picture plane, turned bridge, and separated by the lines of the bridge from the life of the town below-exemplifies the of linguistic meaning and reference that cannot very qualities of anguish and alienation that this be determined independently of context. Similarly, in determining the meaning of artworks, genetic character is depicted as suffering. Moreover, the and other contextual factors are often crucial. ing mouth, so that the whole picture seems to swirling brushwork echoes the character's scream- The very same picture in the same pictorial sym- scream: it exemplifies screaming. Similarly, Shake- bol system may be a quite different kind of pic- speare's lines ture depending upon how and why it came into being. For example, Wollheim cites a Terborch Like as the waves make towards the pebbled shore, painting commonly known as The Parental Admonition, which, he says, in fact depicts an aspirant whore and her potential client.7 The picture can be interpreted in either way; a correct interpretation depends upon knowing about the So do our minutes hasten to their end;8 context in which the painting originated, including the painter's intentions. which they refer. Goodman claims that exempli- III. EXEMPLIFICATION AND EXPRESSION density, and repleteness. Certainly, it has often exemplify the advance and retreat of the waves in line one, and the headlong rush of time in line two. The lines exemplify the very qualities to ficationality is one of four "symptoms of the aesthetic," along with syntactic and semantic been claimed that one of the marks of art is that By contrast with the reams of pages devoted to it "shows" what is "says," as in the Munch and Goodman on representation, the concept of exemplification has been relatively neglected, yet Goodman regarded this concept as one of the main Shakespeare examples. important innovations in Languages of Art. An exemplified property is one that is both possessed and referred to by an artwork. Since Goodman's nominalism requires that only "labels," such as predicates, are exemplified, rather than properties, he defines exemplification as fol- The concept of exemplification is particularly useful in explaining how abstract arts like non- objective painting and so-called "pure" instrumental music can be meaningful, despite the fact that they do not normally represent anything and cannot describe anything because they are not in a linguistic symbol system. We might say, for example, that Beethoven's Fifth Symphony ex- This content downloaded from 129.137.215.253 on Fri, 07 Jul 2017 14:38:41 UTC All use subject to http://about.jstor.org/terms The Journal of Aesthetics and Art Criticism 216 emplifies a "struggle to victory." A heroic theme in a character or the (implied) author, then the meets obstacles in the form of other themes, key feeling may belong literally, not metaphorically, to both the gesture and the person making it. In Purcell's celebrated lament from Dido and Ae- changes, etc., that try to alter it and prevent its return, yet the theme manages to survive unscathed, thereby exemplifying a triumph over the neas, for example, Dido is represented as literally experiencing grief and despair at being abanobstacles. Many theorists deny that music can doned by Aeneas, and her lament is literally have "extra-musical" meanings, while others grief-stricken and despairing. In a similar way, claim that some "pure" music is indeed able to Tolstoy's treatment of Anna Karenina is literally tell psychological stories of this sort. The consympathetic and compassionate, and the novel cept of exemplification helps to bridge the gap Anna Karenina literally exhibits this sympathy between these views. It shows how music can and compassion. More importantly, an account refer to qualities "outside" the music that are at of such examples would require a psychological the same time qualities that the music itself posanalysis of the emotions in question and how sesses. The struggles and the triumph over obthey can be expressed in life as well as in literastacles are both musical and psychological. Goodture. The sympathy and compassion that Tolstoy man's concept of exemplification gives us an elegant way of talking about how "pure" musical exhibits is the same kind of "life" emotion that we may feel in real life for those whom we perstructure can have "extra-musical" significance. ceive to be suffering misfortune. To understand Goodman analyzes expression as metaphorical this kind of expression we need a psychological exemplification. Strictly speaking, The Scream account of expressive gestures and how they are exemplifies its swirling brushwork, but expresses related to the states of mind that they express. its feelings of alienation and anguish. Presumably, it also expresses its screaming quality, sinceOther examples of this sort are even harder for the picture refers to a scream, but only metaphor- Goodman to deal with. Constable's landscapes and George Butterworth's music often express a love ically screams. One of the advantages of Goodman's view is that it allows for the expression of for the English countryside, but these examples nonemotional properties. Some theorists of music, are not explained by saying that the artworks in particular, have criticized philosophers for con- themselves metaphorically love the countryside. This difficulty is related to another. The reason fining attention to the expression of emotional why we can confidently assert that Constable's qualities. On Goodman's view, however, poems, landscapes express love for the English countrypaintings, and music can all exemplify such nonemotional properties as fluidity, freshness, stormi- side is that we know the context in which they were brought into being. We know that Constaness, and weight; they can scream and snarl and ble loved the landscapes he painted, and we can droop. Goodman gives us a way of talking about see this love in, for example, the vibrant freshthe expression of qualities that are metaphoriness of his windswept skies and the loving attencally possessed by artworks but that are not tion to details of texture and color in the munspecifically emotional. dane rural objects he depicts. In short, as in the Another advantage of Goodman's view is that case of representation, knowing the symbol sysit emphasizes that what a work of art expresses is (partly, at least) a function of the symbol sys- tem an artwork is in may not be enough to determine what it expresses. We may also need to tem in which it is a character. To borrow an example from Ernst Gombrich, Mondrian's painting know something about its genesis in the artist's intentions. 10 Broadway Boogie-Woogie expresses gay abandon partly because of the role it plays within the IV. ONTOLOGY symbol system of Mondrian's oeuvre. If it were, per impossibile, in the pictorial symbol system Goodman's views on the ontology of art have also of Severini's mature style, it might rather express cool aloofness.9 been roundly criticized and widely condemned. Not all artistic expression can be explained in According to Goodman, a musical work is idensemantic terms, however. If we think of an art- tified as the referent of a sequence of characters in a notational system (a score). A notational work or part of an artwork as a gesture expressive of some feeling (or other psychological state) system is syntactically and semantically disjoint This content downloaded from 129.137.215.253 on Fri, 07 Jul 2017 14:38:41 UTC All use subject to http://about.jstor.org/terms Symposium: The Legacy of Nelson Goodman Robinson, Languages of Art 217 and differentiated. In a score every mark or in- cannot be fully identified via a notation. Sec- scription is identifiable as belonging to one and ondly, a performance with one wrong note is no only one character, such as the character for longer a performance of that work, while a per- formance of a Chopin Prelude that blatantly igmiddle C or Ct. and has one and only one refernores tempo and expressive markings such as ent: the sounding tones C or Ct. Each sounding tone in turn is unambiguously denoted by one and andantino or vivace but has no wrong notes, only one character.11 Any sequence of sounding counts as a correct performance.13 tones that complies with the score is an instance Goodman answers these objections by pointof the musical work determined by the score: same- ing out that we can identify a piece of music with ness of spelling is all that counts. A literary work is whatever complies with the score, without thereby a character in a notational scheme (a script), so that committing ourselves to the view that everything any inscription or utterance of the script is thereby aesthetically important about the music is capan instance of the work. Two inscriptions of a tured in the way it is identified. The idea that a novel are instances of the same work because they performance of a work with one wrong note is are spelled in exactly the same way. no longer a performance of that work is strictly Both music and literature are therefore what true without having any important aesthetic conGoodman calls allographic arts: the identity of sequences: we may still treat the performance as both is defined broadly speaking by "sameness a (slightly wrong) performance of the work and of spelling." An exact copy of King Lear is just discuss its aesthetic virtues and vices independanother copy of King Lear. Any performance of ently of its ontological status. Here, as elsewhere, The Rite of Spring is just another performance of we have to remember that Goodman's account is The Rite of Spring. By contrast, paintings, sculphighly schematic. At the same time, it has the cortures, etchings, woodcuts, and lithographs are all responding virtues of systematicity and elegance. autographic arts. Because they are in syntactically The distinction between autographic and allodense and undifferentiated symbol systems, they graphic art forms has been profoundly influenhave no distinct characters that can be spelled and tial. No one before Goodman had much of philoread. Consequently, they are identifiable solely sophical interest to say about forgery. Since by means of their history of production. An au- Goodman there has been a flurry of work on this tographic form like painting is such that "even topic. Goodman focused attention on the vital role the most exact duplication of it does not thereby of history of production in the identification of count as genuine."12 Goodman notoriously argues paintings and sculptures. In doing so, however, that there is an aesthetic difference between a he inadvertently opened the door to a serious obpainting and a forged copy of it even if I cannot jection to the distinction between autographic and presently tell them apart just by looking at them. allographic arts. It turns out that what Goodman It is probably theorists of music who have found thinks of as allographic art forms may, like paintmost to grumble about in Goodman's discussion: ings, need to be identified, at least partly, in terms critics argue that the musical work cannot be iden-of their history of production. A musical work, tified with whatever complies with a score. First for example, may be identifiable not just as a seof all, the score leaves unspecified important asquence of sounds determined by a score, but as pects of music that help to define particular works. that particular sequence of sounds that was put A musical work is not just a sequence of tones in together by a particular composer at a particular a particular order, organized by a particular meter; time. Moreover, Borges's tale of Pierre Menard's it is a sequence of tones with dynamic qualities writing part of Don Quixote (in the twentieth and phrasing, played on particular instruments century under the influence of pragmatism!) with particular timbres, and none of these musishows how sameness of spelling does not comcal dimensions can be adequately captured by a pletely determine the identity of a literary work.14 notation. In shaping a phrase, musicians will make We cannot tell the difference between Cervantes's certain notes last fractionally longer than others; work and Menard's just by following the words they will emphasize some more than others; they on the page. will play more quickly or more slowly, louder or more quietly. If phrasing and dynamics are essential features of a musical work, then the work This content downloaded from 129.137.215.253 on Fri, 07 Jul 2017 14:38:41 UTC All use subject to http://about.jstor.org/terms The Journal of Aesthetics and Art Criticism 218 V. CONCLUDING REMARKS that is interesting and true. At the end of the day, Languages of Art is both rich and dense. In it Art has done perhaps more than any other book Goodman discusses every significant topic in to transform aesthetics into a vigorous and rigor- or, rather, at the end of the century, Languages of ous discipline. It is a landmark work of twentiethaesthetics from representation to expression, from forgery to metaphor, from the ontology of art to aesthetic value, and in every case he has some- century aesthetics that has helped to define the field as we enter the next millennium.15 thing lucid, original, and provocative to say. His approach has influenced everyone who writes JENEFER ROBINSON on aesthetics in the Anglo-American tradition, Department of Philosophy whether they like it or not. As I see it, while the University of Cincinnati approach has been remarkably illuminating in Cincinnati, Ohio 45221-0374 all sorts of ways, it suffers from three major drawbacks that recur in different guises in many different places in the book. The first drawback is Goodman's apparent denial of the psychological aspects of art, especially in his accounts of representation and ex- internet: jenefer.robinson @ uc.edu 1. Nelson Goodman, Languages ofArt: An Approach to A Theory of Symbols (Indianapolis: Bobbs-Merrill, 1968; Indianapolis: Hackett, 1976). pression. The second is his pervasive neglect of 2. Monroe Beardsley, Aesthetics: Problems in the Philos- the historical and cultural context in which a work ophy of Criticism (New York: Harcourt, Brace, 1958; Indi- of art originates. Of course, we could give a broad anapolis: Hackett, 1981). 3. Richard Woliheim, Art and Its Objects (New York: sketch of the history of a symbol system as we Harper & Row, 1971; Cambridge: Cambridge University could of a language, and what an artwork ex- Press, 1980). presses or represents will be partly determined by the general history of the symbol system itself. But knowing about the origin of the symbol system is often not enough; we may need to know 4. Nelson Goodman, Of Mind and Other Matters (Harvard University Press, 1984), p. 67. In this book Goodman defends Languages of Art from various critics. 5. See Richard Wollheim, "Seeing-as, Seeing-in, and Pictorial Representation," supplementary essay to Art and Its how individual works within the system origi- Objects, 2nd ed.; and Kendall Walton, Mimesis as Make- nated in order to determine what they mean. This Believe (Harvard University Press, 1990). drawback is linked to the first: one of the many kinds of information that we may need to gather about the origin of a work is information about the psychology of its maker. The third drawback is Goodman's adherence to nominalism, rather than a philosophy of language that could include a richer semantics, not 6. Goodman, Of Mind and Other Matters, p. 88. 7. Richard Wollheim, Painting as an Art (Princeton University Press, 1987), p. 92. 8. William Shakespeare, Sonnet 60. 9. See Ernst Gombrich, Art and Illusion (New York: Pantheon, 1960), chap. 11. 10. We also know that Constable flourished in the art historical period known as romanticism, in which artworks were often conceived of as expressions of the artist's feelings to mention a pragmatic dimension. Goodman's about nature. However, this particular fact about history may, nominalism leads him to announce that "deno- in Goodman's view, simply serve to identify the symbol sys- tation is the core of representation" and then retreat to a position where most representations tem in which Constable was working. 11. This is not quite accurate. Stephen Davies reminds me that Goodman allows for notational redundancies in a score. fail to denote. It also leads him to say that only Thus, for even-tempered instruments, C# and D; are every- labels can be exemplified or expressed, a view where intersubstitutable. that perhaps explains his neglect of the psychological aspects of expression. Finally, it leads him to take as his paradigm of a symbol system a notation, a symbol system that is quite unlike any language in actual use. What is perhaps most remarkable about Languages of Art is that despite these mostly selfimposed drawbacks, it manages to say so much 12. Goodman, Languages of Art, p. 113. 13. There are complications. Some tempo and dynamic markings may fit a notation if their ranges are disjoint. For example, mf andf may have ranges that never overlap. I owe this point to Stephen Davies. 14. Jorge Luis Borges, Labyrinths; Selected Stories and Other Writings, ed. Donald A. Yates and James E. Irby (New York: New Directions, 1962), pp. 36-44. 15. I thank Stephen Davies, John Martin, and Stephanie Ross for their useful comments. This content downloaded from 129.137.215.253 on Fri, 07 Jul 2017 14:38:41 UTC All use subject to http://about.jstor.org/terms