Reply to Chirimuuta
Mazviita Chirimuuta is sympathetic to my deflationary construal of scientists’ representational talk when they characterize their models, but she sees objectionable elements of eliminativism in my discussion of representation outside the domain of science. She says of me: “If she has to choose between the ontology of the manifest image, and that of the scientific image, it’s the manifest that will have to go.” But we don’t have to choose. My deflationary account gives both the scientific image and the manifest image their due. What I do reject are central elements of the orthodox philosophical image that results in a problematic mix of the two.
I criticize two distinct aspects of this mix in the book: (1) the widespread (though not universal) presumption that theories that aim to explain our cognitive capacities must characterize them as ineliminably intentional or representational, and (2) the characterization of our ordinary folk psychological practice of attributing propositional attitudes to each other as proto cognitive science, presupposing and anticipating Fodor’s (1979, 1987) representational theory of mind, according to which the attitudes are relations to unobservable internal sentences.
Chirimuuta argues that “… the space of reasons has the right to be left alone by the space of causes.” But application of the two realms to the real world may not be straightforward. They are governed by different principles – the space of reasons by principles of rationality and the space of causes by physical laws. We can’t assume that we can always cleanly pull apart the two elements in our ordinary practices, though I try to disentangle their roles in commonsense psychology in chapter 3 of the book. Commonsense psychology is a framework for attributing causes to behavior, in the service of prediction and explanation, but it is a framework that allows behavior to be evaluated as rational and, moreover, to be shaped to conform to standards of rationality and thereby to become more predictable. The causal processes underlying behavior are modelled by logical and semantic relations that hold among the sentences that we use to attribute propositional attitudes. The space of causes and the space of reasons both play central roles in the practice.
References
Fodor, J.A. (1979), “Propositional Attitudes,” in RePresentations: Essays on the Foundations of Cognitive Science. MIT Press.
Fodor, J. A. (1987), Psychosemantics: The Problem of Meaning in the Philosophy of Mind. MIT Press.