What is the relevance of the existence of qualia for theories of perception? In his contribution to the Gendler and Hawthorne (eds.) Perceptual Experience , Tim Crane (”Is There a Perceptual Relation?”) answers: little to none. This is (I think) a surprising view, and one I disagree with. I’ll start with how I would answer the question before turning to Crane’s reasons for thinking the “question of qualia is not an important question” for the philosophy of perception.
There is no consensus on a definition for “qualia”. But most would at least agree that, if they exist, they are mental properties, they are properties of subjects or of the experiences had by subjects, and they constitute (fully or in large part) the phenomenal character of a subject’s experience. They are also often said to be “intrinsic” and to be “non-representational”. With regard to the latter, I think we should understand this quite narrowly. That is, qualia are not identical to representational properties, but we might allow that an experience, in virtue of involving particular qualia, will necessarily have some representational content or other. That is, qualia may determine representational content. I’ve been inclined to defend a view of this sort (though not explicitly in these terms), as have others.
Straight away the existence of qualia would seem to rule out naive realism (including disjunctivism). Naive realism tells us that the phenomenal character of an experience is constituted by external properties and objects. The having of an experience with a particular phenomenal character is to be understood relationally rather than in terms of the experience having certain intrinsic phenomenal properties (such as qualia). To treat a particular clear example, where a qualia theorist introspects and takes herself to find a red quale (or better, a red’ quale, following Peacocke), the naive realist introspects and simply finds the redness of some external object. This strikes me as quite a fundamental difference, and again, it seems that the existence of qualia would largely rule-out naive realism and disjunctivism.
As I understand both those who believe in qualia and naive realism, they agree with something like what Robinson calls the “phenomenal principle” as applied to veridical perception–that when it perceptually appears to a subject that something is F, there is something F that the subject is aware of. But crucially, disjunctivists either cannot accept this principle when applied to hallucination or illusion, or they must insist that in hallucination it does not “appear to a subject that something is F”, but merely seems to the subject as if things appear to her in that way.
What is distinctive about intentionalist theories of perception is their denial of the phenomenal principle. What one is aware of in having a perceptual experience is the representational content of one’s experience. So when one has an experience of appearing F, it need not be the case that one is actually aware of an instance of F.
Qualia, it seems to me, are most naturally thought of as sense-data “de-objectified”. By that I mean that where sense-data are a special sort of object that bear the phenomenal properties that constitute the phenomenal character of an experience, qualia just are phenomenal properties, and they are said to belong to either subjects or experiences (rather than to a special sort of object). But otherwise, much of what sense-data theorists said about sense-data and their role in perception can translate easily into a theory of perception from within the qualia theory. [Although I agree with Crane that there may be an interesting way to think of qualia theories as compatible with a form of adverbialism as well.]
And just as naive realism, disjunctivism, and intentionalism can be seen as resisting the existence of sense-data, the most ambitious forms of these views can also be seen as resisting the existence of qualia. For the naive realist, qualia simply aren’t needed; being appropriately related to external objects and their properties is sufficient for us to have perceptual experiences of the sort that we have. Likewise, prominent intentionalists such as Tye and Dretske have aimed to deny qualia and to account for all aspects of phenomenal character in terms of the properties represented by experiences.
So why does Crane think that the question of qualia is barely relevant to the philosophy of perception? It appears that his central reason is the claim that the issue of qualia does not “spring from the debates at the heart of the philosophy of perception” (p. 143). Now if what I said above is right, then even if the question of qualia doesn’t spring from the philosophy of perception, it is far from irrelevant (given that the existence of qualia is incompatible with many theories of perception). Indeed Crane agrees that neither the disjunctivist nor the form of intentionalism favored by Tye has a role for qualia. But I would also argue that the question of qualia at least can be arrived at from the central questions concerning perception. Again, we can turn to the phenomenal principle, which is an essential premise in the argument from hallucination. If one denies the phenomenal principle one can endorse intentionalism and the rejection of qualia. But if one accepts the phenomenal principle, one must then “locate” the phenomenal properties in question. The naive realist locates them out in the external world, but flounders when considering hallucinatory experiences. The disjunctivist makes the (to many, quite bold) claim that hallucinatory experiences are not of the same fundamental kind as veridical perceptual experiences. On some versions of this view, this amounts to the denial that hallucinations have genuine phenomenal character that is the same as in perception. But if one takes seriously the idea that hallucinations can be phenomenally just like perceptions, then one will be forced to locate phenomenal properties somewhere else. The qualia theory can be arrived at in this way, locating the properties that we are aware of in perceptual (including hallucinatory) experience in the mind.
This “solution” to the problem of perception is, again, analogous to the sense-datum theory. But it might be seen as preferred because it is at least slightly less metaphysically extravagant. For one, it might be claimed that the phenomenology of experience supports only the positing of phenomenal properties (qualia) and not of a special kind of object that has those properties. Second, the qualia theory does not face certain sorts of puzzling questions that vexed sense-datum theorists, such as whether or not sense-data have properties other than the ones they appear to have. This question simply does not arise for the qualia theory.
Nice post, Brad. As you know, I agree with what you say, though I tend to get confused when disjunctivism enters the picture. (I’m one of the dummies who wants to read Byrne and Logue’s “Disjunctivism for Dummies” for next time.) I want to ask a further question of the Crane piece–and since it seems to concern a general move by disjunctivists, it is a question for all of them.
Disjunctivists seem to let the phenomenology of experience do very selective work. In response to Tye’s complaint that disjunctivism goes against common sense, Crane says “the view is attempting to preserve what seems to be one of the most obvious or common sense features of perception–its relationality–in the face of the argument from hallucination. The price it pays for this is that it cannot count subjectively indistinguishable states of mind as states of the same fundamental kind, and therefore it imposes limits on what can be known from the subjective perspective…” (141).
My worry is that the “common sense” support of the relational nature of experience draws upon the very subjective perspective that disjunctivism seems to undermine. After all, the phenomenological–and I take it, common sense–data seems to support relationality as much in the hallucinatory case as it does in the veridical case. If one thinks justice can be done to that “subjective point of view” in the hallucinatory case without positing a direct relation to an object, why not think so in the veridical case as well? It seems to me, a self-acknowledged tyro in this debate, that the disjunctivist uses the phenomenological data when arguing against the indirect realist, but then denies that same data when responding to the exactly parallel criticism from the full out intentionalist.
Perhaps this gets to a more general question: what is the more general principle the disjunctivist is using to slice up the pie of perceptual componants? E.g. what is the real basis for saying that veridical perceptual states are not of the same “kind” as hallucinatory states? It cannot be the physical substratum, because these are admitted to have commonalities, and it cannot be purely phenomenological, because the phenomenology is the same in both cases. So…what is it, and why prefer that way of individuation kinds?
Left by Robert Howell on February 21st, 2007