Page 255 - Encyclopedia of Philosophy of Language

Not until the latter part of the twentieth century have there been what could be called 'theories' of pictorial representation. There have been comments and obser- vations about pictorial representation, claims and counterclaims, ever since Plato's charge that because the painter composes an imitation ('mimesis') of an appearance of a physical object, this latter in turn being an imitation of a Form, the painter's com- position is 'at the third remove' from reality. But not until the late twentieth century have there been well- developed theoretical accounts of what pictorial rep- resentation is. Three quite different accounts are avail- able: that by Nelson Goodman, that by Nicholas Wolterstorff, and that by Kendall Walton. All three are subtle and highly qualified; here most of those complexities will have to be bypassed, as will the pol- emics between and among the theories.
1. Goodman's Account
Goodman's account was first on the scene, appearing in his 1968 book, The Languages of Art. His account is a 'symbol' theory of pictorial representation, its fundamental thesis being that a picture is a character in a representational symbol system. The unpacking of this formula is begun with that last concept, the concept of a 'symbol system.' 'A symbol system,' says Goodman, 'consists of a symbol scheme correlated with a field of reference' (1968:143). In turn, a symbol scheme 'consists of characters, usually with modes of combining them to form others. Characters are certain classes of utterances or inscriptions or marks' (1968: 131). A symbol 'system' is then an ordered pair whose first member is a set of characters, that is, a symbol scheme, and whose second member is a function which assigns sets of entities to the characters in the scheme, the members of a set being the 'extension' of the character to which the set is assigned. For a set to be assigned to a character, the character must 'refer' to members of the set; 'reference' is the central unex- plained concept in Goodman's conceptual frame- work.
As to what makes a symbol system 'represen- tational,' Goodman's thesis is that this happens just in case the system is syntactically and semantically dense, and has a relatively replete symbol scheme. 'A scheme is syntactically dense if it provides for infinitely many characters so ordered that between each two there is a third' (1968: 136); and a 'system' is 'sem- antically' dense if it provides for a set of extensions so ordered that between each two there is a third. Thus, for example, the scheme of arabic-fraction characters is syntactically dense, and the system of arabic frac-
tions is semantically dense. But arabic-fraction characters are obviously not pictures. That is because all that differentiates one character in the scheme from another is the shape; size, color, texture, none of those make any difference. Bycontrast, in a representational system, differences with respect to any of a rather large number of properties function to differentiate one character from another.
What is striking about Goodman's view is that, though a character is never a picture as such, but is a picture only as a character in a symbol system of a certain sort, namely,a representational system, none- theless a character's functioning as a picture has nothing to do with any relation between its look and the look of what it represents; Goodman's theory is resolutely anti-iconic.
2. Wolterstorff's Account
By contrast, Wolterstorff's theory, presented in his 1980 book, Works and Worlds of Art, is an iconic theory. An even more fundamental difference between the two theories, however, is that Wolterstorff's theory is an 'agent-action' theory of representation. In Wolterstorff's theory, the fundamental reality is not that of some symbol being a representation of something, but that of someone performing the action of representing something by means of doing some- thing with some design. The phenomenon of an object's, or a design's, representing something is tre- ated as parasitic on the phenomenon of a person's
representing something with that object or design.
J. L. Austin's concept of 'illocutionary' actions, in distinction from 'locutionary' and 'perlocutionary' actions, is by now well known; examples are asserting, asking, and commanding. An important feature of illocutionary actions is that, though they can indeed be performed by performing locutionary actions— that is how Austin introduced them—nonetheless they can be performed in other ways as well. Rather than issuing a command by uttering words, one can do so by presenting a picture. A fundamental thesis of Wolterstorff's theory is then that in representing something, one performs an illocutionary action. More specifically, one takes up an illocutionary stance toward some state of affairs (proposition); in picturing a man on a horse one introduces, and takes up an illocutionary stance toward, the state of affairs of
'there being a man on a horse.'
What remains is to pick out that particular species
of illocutionary action which consists of representing something. It is at this point that Wolterstorff intro- duces iconicity. To pictorially represent a man on a
Pictorial Representation N. Wolterstorff
Pictorial Representation
233