Catherine Elgin

If the point of argument is to produce conviction, an argument tor a foregone conclusion is pointless. I maintain, however, that an argument makes a variety of cognitive contributions, even when its conclusion is already believed. It exhibits warrant. It affords reasons that we can impart to others. It identifies bases tor agreement among parties who otherwise disagree. It underwrites confidence, by showing how vulnerable warrant is under changes in background assumptions. Multiple arguments for the same conclusion show how our beliefs hang together.

Nelson Goodman was one of the soaring figures of twentieth century philosophy. His work radically reshaped the subject, forcing fundamental reconceptions of philosophy’s problems, ends, and means. Goodman not only contributed to diverse fields, from philosophy of language to aesthetics, from philosophy of science to mereology, his works cut across these and other fields, revealing shared features and connecting links that narrowly focused philosophers overlook. That the author of The Structure of Appearance also wrote Languages of Art is not in the end surprising.

I argue that the picture theory provides both a common referential hase and a common logical syntax for languages embodying alternative conceptual schemes. I offer an analysis of depiction, explicating the Tractarian concepts of pictorial structure, pictorial relationship, and representational form. Significant failure of reference and the existence of languages with incompatible ontological commitments show that on the molar level depiction is not required for sense. Using three premises, taken to be axiomatic for Wittgenstein, I show that analysis leads to a base of elementary propositions which must depict in order to be significant. There, the relations between pictorial structure, pictorial relationship and representational form are such that reference is secured and conceptual relativity precluded.