Academic journal article South African Journal of Philosophy

Towards Defending a Semantic Theory of Expression in Art: Revisiting Goodman

Academic journal article South African Journal of Philosophy

Towards Defending a Semantic Theory of Expression in Art: Revisiting Goodman

Article excerpt


Nelson Goodman's attempt to analyse the expressiveness of artworks in semantic terms has been widely criticised. In this paper I try to show how the use of an adapted version of his concept of exemplification, as proposed by Mark Textor, can help to alleviate the worst problems with his theory of expression. More particularly I argue that the recognition of an intention, which is central to Textor's account of exemplification, is also fundamental to our understanding of expressiveness in art. Moreover I propose that the recognition of this intention depends on our interpretation of the artwork - an insight Goodman tried to capture with his assertion that our attributions of expressive properties to artworks function metaphorically. The realisation of the context-dependence of our expressive judgements about art and, hence, of the central role interpretation plays in these judgements, I contend, counts in favour of theories of expression like Goodman's that focus on semantic concerns.

What do we mean when we apply emotion terms, like "joy" and "sadness", "anguish" and "rapture", to artworks and other lifeless objects? When calling a rendition of a Brahms intermezzo "melancholy" or a Wodehouse novel "jovial", whose feelings, if anyone's, are we describing? There is a rich literature dealing with these questions and in it a wide variety of answers articulated.1 One eminent answer was given by Nelson Goodman in Languages of Art (1968)2, but his approach has fallen into disrepute since.

Goodman proposes that expression in art amounts to what he calls metaphorical exemplification, where exemplification is a restricted form of inverse denotation, and its metaphorical nature relates to the fact that attributing an emotion to an artwork is an instance of abnormal property ascription. Artworks, Goodman says, express emotions by (a) metaphorically possessing them as properties and (b) symbolising them in the way that samples symbolise (i.e. "exemplify") some of their properties.3 Goodman's theory has, however, rightly been criticised on both counts. Typically, his critics point out separate problems with his view of metaphor and with his notion of exemplification respectively. The objections in both cases are substantial but, I will argue, perhaps not insurmountable.

Recently Mark Textor has shown how, with some modifications, Goodman's notion of exemplification can be rescued from the standard criticisms. Textor (2008) supplements Goodman's version by making it an intention-dependent form of reference. It is doubtful whether Goodman would approve of this strategy, but Textor' s suggestion goes a long way towards making exemplification into a useful theoretical tool where Goodman's own version came up short.

In what follows I take stock of the extent to which a reworked version of exemplification, like Textor' s, can be used to salvage Goodman's theory of expression. The insight gleaned from Textor is that the main difficulties for Goodman's theory - not only with exemplification, but also with his view of metaphor - arise because his theory is rooted in an inadequate account of semantic reference. This might sound like an uncomplimentary stance towards Goodman. But on the contrary I want to argue that, whatever the particular flaws of his semantics, the view that expression in art can be analysed in semantic terms should not be discarded too hastily.

My argument will take the following form. In the first part, I give an exposition of Goodman's original account of expression and in the second, I explain the standard objections against it. Thirdly, I turn to Textor' s analysis of exemplification and the promise it holds for salvaging Goodman's theory. In a fourth section, I argue that (something like) Textor's approach can also help to alleviate the worst of the concerns about the role of metaphor in our attribution of expressive properties to art. A short fifth section contains some concluding remarks. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.