ABSTRACT

Pictures are everywhere, and we can usually recognize what they represent with ease. It might be thought, therefore, that we can also explain with ease how pictures represent. Unfortunately, we cannot. For one thing, there are five standard theories of pictorial representation, each of which seems to give obvious and common sense answers to two basic questions: What is a picture? And how do pictures have content? Pictures are representations of some sort; the first question is about how to distinguish them from other forms of representation. Individual pictures are identified by the contents they have; the second question asks how to distinguish one picture from another. These standard theories explain depiction in terms of convention, resemblance, causal relations, mental constructions, and information, respectively. Although each has some intuitive appeal, they cannot all be right. Moreover, there is a tendency to treat them as types in a taxonomy. They may be taken to define the range of categories into which theories of pictorial representation fall. But this standard list is not the best way to divide up the turf. Not only do the categories overlap to some extent, describing them in the traditional way ignores or obscures some important dimensions of theories of depiction. Specifically, it cuts across a basic contrast between perceptual and nonperceptual accounts (Lopes 1996).