Abstract
According to orthodox representationalism, perceptual states have constitutive veridicality or accuracy conditions. In defense of this view, several philosophers-most notably Tyler Burge-employ a realist strategy that turns on the purported explanatory ineliminability of representational posits in perceptual science. I argue that Burge's version of the realist strategy fails as a defense of orthodox representationalism. However, it may vindicate a different kind of representationalism.