GOODMAN'S STRICTURES ON SIMILARITY

There is a popular belief that if S – say a drawing by Whistler – depicts a butterfly, then it must necessarily resemble a butterfly, in at least some respects. As Jerry Fodor boldly puts it:

One could quibble with this admirably straightforward statement: surely icons are symbols of a sort and surely Fodor has somewhat exaggerated the contrast between icons and symbols whose understanding is mediated by convention. In the last two chapters I have argued at length that the real contrast between (for example) a linguistic symbol system and an iconic one does not consist in the conventional nature of the former and the non-conventional nature of the latter but in the nature of the conventions involved in each.

Nonetheless, what stands out from this passage is Fodor's claim that iconic reference is mediated by resemblance. I scrutinised this claim in chapter i and described its shortcomings in lurid detail, the primary one being that there does not appear to be a respect in which S resembles O for each respect in which S iconifies O. Thus, a picture of David Bowie may depict him as smiling without itself smiling or incorporating any smile-like feature. We looked at various attempts to adjust the resemblance model to reckon with this problem, but all of them seemed futile.