Joshua Gert

This article attempts to do two things. The first is to make it plausible that any adequate dispositional view of color will have to associate colors with complex functions from a wide range of normal circumstances to a wide range of (simultaneously) incompatible color appearances, so that there will be no uniquely veridical appearance of any given color. The second is to show that once this move is made, dispositionalism is in a position to provide interesting answers to some of the most challenging objections that have been pressed against it. It explains why colors do not seem to come on when the lights come on, and why the right thing to say seems to be that the light causes the objects to reveal their colors. It explains why the content of visual experience need not be problematically circular. It eliminates the problem of intrapersonal variation in color appearance. And it even provides resources for dealing with the problem of interpersonal variation. It is true that in moving to the more complex view, the dispositionalist becomes subject to a new objection, which has been called the problem of unity. The paper offers a solution to this problem as well.

Ruth Chang has argued for the existence of a fourth positive value relation, distinct from betterness, worseness and equality, which she calls “parity.” In an earlier article I seemed to criticize Chang's suggestion by offering an interval model for the values of items that I claimed could accommodate all the phenomena characteristic of parity. Wlodek Rabinowicz, offering his own model of value relations, endorsed one central feature of my proposal: the need to distinguish permissible preferences from required ones. But he, along with Chang, correctly pointed out that my own model suffered from serious formal problems. The present article offers an improved interval model, but the main point is not the model itself. Rather, it is the derivation of the model from a response-dependent account of practical rationality. This derivation helps explain why we have a concept for which Rabinowicz's model is appropriate, and therefore supplements Chang's existence arguments. My conclusion is that we can take Chang's view, Rabinowicz's and mine as providing a reinforcing set of views

Response-dispositional accounts of value defend a biconditional in which the possession of an evaluative property is said to covary with the disposition to cause a certain response. In contrast, a fitting-attitude account of the same property would claim that it is such as to merit or make fitting that same response. This paper argues that even for secondary qualities, response-dispositional accounts are inadequate; we need to import a normative notion such as appropriateness even into accounts of such descriptive properties as redness. A preliminary conclusion is that the normativity that appears in fitting-attitude accounts of evaluative properties need not have anything to do with the evaluative nature of those properties. It may appear there because evaluative properties—or at least thosefor which fitting-attitude accounts are plausible—really are so much like secondary qualities that it might well be appropriate to think of them as a subclass of secondary qualities. In the second half of the paper I discuss the views of three of the philosophers who have been most influential in discussions of response-featuring accounts of evaluative notions and who explicitly distinguish response-dispositional accounts of value from fittingattitude accounts: John McDowell, Simon Blackburn, and Crispin Wright. I highlight some of the theoretical temptations that can be associated with the assumption that the response-dispositional/fitting-attitude distinction parallels the secondary quality/evaluative property distinction.