The question is Peirce’s, from his “The Principles of Phenomenology.” His answer is one I endorse, but I would quibble with him about it is one born out of phenomenology. Against those who would have qualities depend upon the mind of their observer, Peirce thinks of qualities as dispositions inherent in objects. He does not use the language of disposition, but rather the language of potentiality. But he is clear that by ‘potentiality’ he does not mean potentiality as lack of actuality, but potentiality as a real capacity, and not merely some dormant can-be-actualized-but-that-depends-on-actualization. His whole discussion hooks up with Shaviro’s recent commentary on Molnar’s Powers, and works in support of my account of sensations. Here’s Peirce:
[A quality] is not anything which is dependent, in its being, upon mind, whether in the form of sense or in that of thought. Nor is it dependent, in its being, upon the fact that some material thing possesses it. That quality is dependent upon sense is the great error of the conceptualists. That it is dependent upon the subject in which it is realized is the great error of all the nominalistic schools. A quality is a mere abstract potentiality; and the error of those schools lies in holding that the potential, or possible, is nothing but what the actual makes it to be. (‘The Principles of Phenomenology’, in Philosophical Writings of Peirce, ed. Justus Buchler, Dover, 1955, pp. 84-85)
Peirce argues in the following way: an object that is red in the light is, by common sense, believed to be red when the lights go out. Now, if you believe that they are no longer red in the dark, then you must hold that they assume some other color in the dark, or that they become indeterminate with respect to their properties. Or you may believe that the object is no longer red, but that it has taken on some other determine color in the dark. In the latter case, there is no reason to suppose a specific quality that is not red; under the former, you still maintain the the object has some qualities, in which case you believe that the object possesses qualities that are not dependent on the perceiver.
As to the question of the indeterminacy of unperceived qualities, he reasons thus:
If, however, you hold that the bodies become indeterminate in regard to the qualities they are not actually perceived to possess, then, since this is the case at any moment in regard to the vast majority of the qualities of all bodies, you must hold that generals exist. In other words, it is concrete thins you do not believe in; qualities, that is, generals–which is another words for the same thing–you not only believe in but believe that they alone compose the universe. Consistency, therefore, obliges you to say that the red body is red (or has some colour) in the dark, and that the bard body has some degree of hardness when nothing is pressing upon it. (p. 85)
I’m in agreement with this ideas that red and hardness persist as real qualities even when no one is around to perceive them. It coheres with my idea of what sensation is, as I began to outline in brief here. What I’m keen to defend is the view that sensations really reside in things; that things emanate to radiate sensations, and that an object is in one respect a conspiracy of qualities whose autonomy is embodied in its singular capacity/disposition/power to effect other objects. In the end–and this will constitute my attempt to think objects in their own right–I’m working toward a speculative aesthetics that will try to imagine a world where qualities conspire into objects and exchange sensations without the facilitation of humans or other sentient creatures. Peirce, I think, is an ally in this project, although his phenomenology is not really an object-oriented ontology.