J.J. Gibson's ecological approach to depiction is compared with Nelson
Goodman's relativist theory of representation. Goodman's commitment t
o radical relativism and Gibson's to direct realism would make these t
hinkers unlikely candidates for comparison if Goodman himself had not
indicated a substantial body of agreement with Gibson in the area of p
icture perception. The present study analyzes this agreement through s
ystematic discussion of the following theses: realism in representatio
n is not a function of geometrical optics, physical similarity to what
is depicted, or deception; pictures differ in density and articulatio
n from words, so that picturing has no explicit vocabulary; and artist
s can teach us new ways to see the world. The agreement between Goodma
n and Gibson has wide-ranging implications for the further development
of what might be called a Gibsonian relativism.