Vim and rigor

ArtForum, May, 1999 by W.J.T. Mitchell

Nelson Goodman's work touched on so many fields - philosophy, of course, but also the arts, the sciences, and psychology - that it is difficult for anyone to appreciate, much less summarize, what it has all meant for us, or even to specify the "us" that will continue to have some stake in his work. More appreciated in Europe than America, invoked as an authority in fields from cognitive science to artificial intelligence to art criticism to analytic philosophy, Goodman has had a massive yet unobtrusive influence on contemporary thinking in a wide variety of disciplines. He never achieved the kind of cult status sometimes accorded to philosophers in the art world. His thought moved with quiet wit and authority, rarely engaging in the pyrotechnics that make a philosopher quotable and therefore reducible to a few slogans. Goodman does not leave behind an industry built on key words like "difference" or "episteme" or "mirror stage." The vocabulary of his theory of symbols is austere and technical. Terms like "compliance," "inscription," "character," "replete," "articulate," "exemplification," "denotation," the distinctions between a symbol "scheme" and a symbol "system," between "autographic" and "allographic" inscriptions do not set one's pulse to pounding. They are clearly and rigorously defined concepts within an analytic framework of remarkable breadth, embracing both the arts and the sciences, the symbols of mass media and the artifacts of the plastic arts. Goodman's only concessions to stylistic flashiness appeared in pithy, often alliterative phrases like "score, script, and sketch," which condensed an entire theory of notation, conundrums like "the sound of pictures," which crystallized a theory of expression, or dialectical pairings like "pictures and paragraphs," which adumbrated a wholesale reorientation of the classic differentiations of verbal and visual symbols.

For me, Goodman was always first and foremost the author of the 1968 Languages of Art. His project as the most rigorous philosophical nominalism since that of Thomas Hobbes, his work as the "irrealist" epistemologist of Ways of Worldmaking left me impressed but not transformed, except perhaps in ways that are too subtle to measure. I suspect that my experience is not untypical, and that this is why there are not many card-carrying "Goodmanians" in the academy or the art world. But Languages of Art was to me a revelation. It took on the great, central questions of aesthetics and the theory of representation and produced a novel and systematic critique of the differentiations among media, symbolic forms, sign types, modes of expressivity, and referentiality. Perhaps the most dazzling thing about this book was its refusal to enter into an Oedipal relationship with its philosophical predecessors (chiefly C.S. Peirce and Ernst Cassirer), much less to troll through contemporary theories of representation looking for rivals and competitors, of which there would have been no lack among structuralist and poststructuralist theorists.

For Goodman, this sort of thing would have been of "merely historical interest." He concentrated on the main task, a "general theory of symbols" that "ranges beyond the arts into matters pertaining to the sciences, technology, perception, and practice." One could feel in his text a convergence of the Barthian triad, "Image/Music/Text," staged as a dialogue between structural linguistics and theories of nonverbal representation. He tacitly resisted the linguistic imperialism of semiotics, taking pains to insist that the word "language" in Languages of Art was merely a vernacular convenience (the precise word would have been "symbol"). His theory paid as much attention to musical and dance notation as to linguistic signs, and it was centered, most famously, on the notoriously difficult question of the visual image.

Goodman's account of pictorial representation is among the most radical and innovative treatments of the subject in modern philosophy or criticism. Decisively refuting "copy" theories of representation based in likeness, resemblance, similitude, or "iconicity," he seemed to flirt with the reduction of pictures to just another form of language. More than one unwary commentator was taken in by his stress on the "analogy between pictorial and verbal description," which he saw as merely a "corrective and suggestive" move. His real aim, however, was to redescribe the difference between "depiction" and "description," "paintings" and "poems," as the distance between what he called "dense" and "differentiated" symbol systems. A differentiated system is one with a finite number of characters (such as an alphabet or a graduated thermometer); it is also "disjunct" and "discontinuous" in that there are no meaningful signs in the "gaps" between two characters - no letters between a, b, and c in the alphabet, for instance. A dense system, by contrast, is one in which there are an infinite number of characters or meaningful marks, and every tiny variation in the inscription of a mark is potentially significant. The letter "A" designates the same thing no matter how it is written; a drawing of an A-frame house, on the other hand, takes its significance from the innumerable details of its graphic execution. Every difference makes a difference.