In his book The Red and The Real, Jonathan Cohen defends a role functionalist version of relationalism about color. In other words, he argues that colors consist in a relation between object and perceiver best described as the role of looking a certain way to a certain subject in certain circumstances. The book is structured around a master argument from perceptual variation, which is set up as Cohen’s main motivation for his view. The argument is presented in chapter 2, and the remaining chapters are dedicated to a defense of the view Cohen forms in response to this argument. After his defense of the claim that the argument from perceptual variation should be taken seriously, Cohen discusses various available views on color and evaluates their potential for meeting the demands of the master argument. His defense of his own view comes in two parts: first (in chapters 4-6) he defends relationalism about color, and then (in chapters 7-8) he goes on to defending role functionalism. Cohen’s arguments are generally deep and thought-provoking and his explanations are systematic and thorough. His style of presentation is effortless and sufficiently lively to keep the reader’s interest throughout. In short, the book was a joy to read. While it is impossible to cover all the issues worthy of discussion, I will go on to highlight a couple of them and present some possible objections.

As I mentioned above, the argument from perceptual variation plays a lead role in the book. Its first step consists in noting that a color stimulus can have many different perceptual effects on subjects of different species, on different subjects of the same species, and even on the same subject in different circumstances. The second step involves the premise that we should be egalitarians when it comes to this perceptual variation; that there is no reason to single out one variant as the correct one while excluding others. In the third step, it is inferred that rather than stipulating one of the variants as correct, we should give an account of color that reconciles the different variants. In the fourth step, Cohen concludes that color relationalism provides the best means of reconciliation.

Colors, according to Cohen, consist in relations between objects, subjects, and circumstances. For instance, the redness of a ripe tomato is really redness for a certain subject under certain conditions. By making the relation sufficiently fine-grained, Cohen can respond to the challenge of perceptual variation. One and the same object can look red to one subject (S₁) while looking orange to another (S₂). Neither relation has to take priority over the other one, so the object simply is red for S₁ while orange for S₂. With the addition of circumstances as parameters, the potential for variation is even increased.

A common objection to the division of properties into red-for-S₁-in-C₁, red-for-S₂-in-C₂, etc. is to point out that such a division makes it impossible for two different subjects to agree or disagree about the color of an object, as they are bound to be talking past each other when they attribute a color to it. Cohen avoids this problem through accounting for another, coarse-grained, level of colors. A coarse-grained color consists in the relation of an object to a wider set of subjects and circumstances. Agreement and disagreement about the color of an object are possible because the coarse-grained colors are the referents of our color terms. Jack and Jill can agree about the ripe tomato’s redness because they both attribute to it redness for “contextually relevant perceivers and contextually relevant circumstances” (p. 111), which are the same in such a case.

While the coarse-grained colors save Cohen’s account from the above problem, they open it up to the worry that the requirements posed by the second and third steps of the argument from perceptual variation are not met. Claiming that some perceivers and some circumstances are contextually relevant while others are not does exclude all other perceivers and circumstances. This seems consistent with our use of color terms; we would hardly attribute the term ‘blue’ to a ripe tomato illuminated with blue light, so disqualifying such circumstances as contextually irrelevant would seem appropriate. But my worry is that if there is, after all, a way to distinguish the normal or contextually relevant subjects and circumstances from the anomalous, the third step of the argument from perceptual variation is under threat, and the role of the fine-grained colors becomes less clear. The point of the fine-grained colors is to account for cases such as when two human perceivers with what we might call normal color vision look at the same color chip in the same circumstances and one sees it as red while the other sees it as orange. If the reference of color terms is coarse-grained, the two perceivers disagree when one attributes redness to the color chip while the other calls it orange (and thereby denies that the chip has coarse-grained redness). If the argument from perceptual variation is to keep its force and the fine-grained colors are to be retained, the response here could be that in this case the terms ‘red’ and ‘orange’ refer to fine-grained, but not coarse-grained, colors. That means that the reference of color terms is coarse-grained, except in certain cases when it is fine-grained. Determining when and why to make those exceptions in a non-arbitrary manner can be problematic.  

This engaging and well-written book is a welcome addition to the literature on color ontology. It should be noted, however, that it may not be very accessible to those encountering the subject for the first time. It is a contribution to a deep-running theoretical debate, and not intended as an introduction to the issues.

© 2010 Eyja M. Brynjarsdóttir

Eyja M. Brynjarsdóttir is a postdoctoral researcher at the Institute of Philosophy at the University of Iceland. The primary focus of her research is the metaphysics of properties, ontological subjectivity and objectivity, and some issues in the philosophy of mind, such as sense perception.