3 - Chrysippus' dog as a case study in non-linguistic cognition

Summary

INTRODUCTION: LANGUAGE AND THOUGHT

Do non-linguistic creatures think? Debate over this question tends to calcify into two extreme doctrines. The first, espoused by Descartes, regards language as necessary for cognition. Modern proponents include Brandom (1994, pp. 145–157), Davidson (1984a, pp. 155–170), McDowell (1996), and Sellars (1963, pp. 177–189). Cartesians may grant that ascribing cognitive activity to non-linguistic creatures is instrumentally useful, but they regard such ascriptions as strictly speaking false. The second extreme doctrine, espoused by Gassendi, Hume, and Locke, maintains that linguistic and non-linguistic cognition are fundamentally the same. Modern proponents include Fodor (2003), Peacocke (1997), Stalnaker (1984), and many others. Proponents may grant that non-linguistic creatures entertain a narrower range of thoughts than us, but they deny any principled difference in kind.

An intermediate position holds that non-linguistic creatures display cognitive activity of a fundamentally different kind than human thought. Hobbes and Leibniz favored this intermediate position. Modern advocates include Bermúdez (2003a), Carruthers (2002, 2004), Dummett (1993, pp. 147–149), Malcolm (1972), and Putnam (1992, pp. 28–30). Proponents may grant that our lower-level cognition resembles the mental activity of languageless creatures, but they insist that we also manifest higher-level cognition unavailable to such creatures. The main challenge facing such a view is to describe non-linguistic cognitive processes that differ in a principled way from higher-level human thought.