One natural route to the view that intersubjective spectrum inversion is possible is via the plausible claim that we can imaging the occurrence of intrasubjective inversion – a change whereby the different colors systematically look different to a person than they did before, although she is able to make all the same color discriminations as she did before, and sees things as having the same color similarity and difference relationships as before. Whereas the existence of complete intersubjective inversion would be behaviorally undetectable and might seem unverifiable, the occurrence of intrasubjective inversion seems straightforwardly verifiable, both from the first-person and the third-person perspective. Yet if intersubjective inversion is at least a logical possibility, it seems that intersubjective inversion should also be at least a logical possibility.

This line of argument I will call the intra-inner argument. It is an argument I have advanced in several places. As might be expected, it has met with a variety of objections.

First, some maintain that insofar as intrasubjective spectrum inversion is conceivable, it would consist simply in its coming to be the case that the subject systematically misperceives the colors of things – sees red things as green, blue things as yellow, and so on. This sort of inversion is possible on the “intentionalist” view according to which the only introspectively accessible features of color experiences are their intentional or representational features, their being “as of” certain colors.