Swamp swamps swamp swamp swamps.
There are two general approaches to the psychosemantics of phenomenal knowledge in normal subjects that are quite easy to see as totally doomed to fail to account for the psychosemantics of Swamp Mary’s phenomenal knowledge. The first approach imposes as a necessary condition phenomenal representation (the representation of phenomenal properties) that it involve a kind of quotation or exemplification whereby, for instance, the concept of a red quale is partially constituted by a red quale (as in Papineau (2002), though he retreats somewhat from the view in his (2007)). On such a view, just as a carpet sample of a red carpet must exemplify the same color and texture of the carpet it samples, so must the phenomenal concept have the same phenomenal character as the experience it is a concept of. The notion of representation by exemplification here is the same as the one discussed by Goodman (1976, pp. 52-67). The second approach grounds phenomenal representation in a causal relation to some actual past occurrence of a red quale. The quotation approach fails for the following reason. Since Swamp Mary lacks a red quale, her phenomenal knowledge cannot be grounded in quoting or exemplifying a red quale. The causation approach likewise fails. Since Swamp Mary need not have any actual past, her phenomenal knowledge cannot be grounded in causal relations to past qualia occurrences.
Let us turn, then, to look elsewhere for a psychosemantics to ground Swamp Mary’s phenomenal knowledge. There are two worth investigating. promising, each of which relate rather directly to the two key features of physicalism described in the previous section.
The first line for exploration is to ground Swamp Mary’s phenomenal knowledge in some kind of nomic or counter-factual relation to qualia, qualia that she need not actually have had. Such a psychosemantics may be modeled on the sort of nomological approach favored by Fodor (Fodor, 1987).Since the physicalists affirm the causal efficacy of qualia, its open to posit that there are counter-factual-supporting laws that subsume qualia. Such laws may very well relate qualia to concepts of qualia (as well as non-conceptual representations of qualia). So, even though Swamp Mary has never had a red quale, she is in a state that would be causally related to red qualia in certain counter factual scenarios. She would, for instance, go into a state of recognition if she later were appropriately causally related to a red quale. Further, the state she is in is of a type tokens of which can be caused by red qualia.
The second line for exploration is to ground Swamp Mary’s phenomenal knowledge in terms of some kind of structural commonality between her knowledge state and red qualia. Instances of this general psychosemantic approach include description theories (e.g., (Russell, 1905), isomorphism-based theories (e.g., (Cummins, 1996), and conceptual role semantics (e.g.(N. Block, 1986; Sellars, 1953). An approach of this sort is available to the physicalist since the physicalist affirms the ontological complexity of qualia. For certain anti-physicalists, the ontological simplicity of qualia leads directly to their ineffability and recalcitrance to analysis. However, for physicalists, it is open to exploit the ontological complexity of qualia and account for Swamp Mary’s phenomenal representations as, e.g., descriptions that capture defining structural features of a red quale.
Let us call the two approaches just sketched Nomological and Descriptive-isomorphism, respectively. They show some initial promise for underwriting Swamp Mary’s phenomenal knowledge. They are consistent with the physicalist part of gappy physicalism. However, to be consistent with the gappy part of gappy physicalism, these approaches have to also be consistent with the claim of Mary’s prerelease ignorance. And this is where the trouble arises for gappy physicalism. In the next post, I argue that neither Nomological nor Descriptive-isomorphism is compatible with Mary’s prerelease ignorance.