Intentionality

Naturalistic Approaches to Intentionality

Many theorists take intentionality to be an entirely natural, physical phenomenon, and a commitment to reductive naturalism has dominated discussion of mental intentionality in recent years. Reductive naturalism requires one to give an account of mental intentionality in entirely nonmental terms, and the consensus has been that the way to do this is to identify a natural relation that holds between states of the brain and states of the environment when and only when the former are about the latter. Philosophers give the example of tree rings “tracking” the age of trees (an entirely nonmental phenomenon) to ground the sense in which intentionality can be a natural relation. Internal states of the brain are held to track the presence of specific external conditions in a fundamentally similar way and to carry information about the environment in virtue of this tracking relation. Different theories expand on this basic idea, diverging in their more detailed expositions of the tracking relation. The dominant naturalistic theories have been causal or covariation theories and teleosemantic theories.

Problems with Naturalizing Intentionality

The “naturalization project” faces certain well-known difficulties: (1) it must account for the possibility of misrepresentation, (2) it must deal with problems of indeterminacy of content, and (3) it must respond to the worry that, once naturalized, intentionality will turn out to be ubiquitous.

• 1.
  The problem of misrepresentation is (roughly) that if it is necessary, for a mental state S to represent a state of affairs X, that S reliably indicate (or causally covary with) X, then S cannot occur unless X occurs. But this means that there is no room for misrepresentation or mistakes—that is, for a case in which S occurs when X is not the case. But clearly, a mental phenomenon can misrepresent: I can represent a dog as being black even if it is brown, or I can think that someone is Santa Claus even though Santa Claus doesn't exist.
• 2.
  Two difficulties concerning the indeterminacy of content are the “disjunction problem” (the disjunction problem may also be classified as a special case of misrepresentation) and the “stopping problem.” The disjunction problem arises because a mental state S may systematically covary with distinct states of affairs. For example, if S systematically covaries with both sheep on dark nights (=X) and goats on dark nights (=Y), does S represent X or Y, or perhaps the disjunction [X ∨Y] (sheep or goats on dark nights)? The contents of our mental states do not seem indeterminate in this way.

To illustrate the “stopping problem,” consider a subject, Lucy, who is perceiving or thinking about a moose called Mandy. Assuming that Lucy has the appropriate causal connections to Mandy, how does Lucy's experience or thought manage to be about Mandy rather than about the set of Mandy-caused photons impinging on her retinas, or certain other sets of causes on the causal chain leading to the thought? The antireductionist suggestion is that we need to appeal to the character of Lucy's conscious experience to explain how Lucy's experience can be said to be precisely about Mandy rather than about some other stage on the causal chain.

...