harnad@phoenix.Princeton.EDU (S. R. Harnad) (01/11/90)
kck@g.gp.cs.cmu.edu (Karl Kluge) of Carnegie-Mellon University wrote: > Are we to suppose that you can impose an interpretation on the symbols > such that they form both a coherent conversation in Chinese *and* a > legal and coherent chess game? Your point about there not necessarily being more than one coherent interpretation of a symbol system is correct. (In fact, I've made the point myself in print, about both the alleged "radical underdetermination" of language translation (Quine/Goodman) -- in which it is supposed, without proof, that meanings can be swapped willy nilly while preserving a coherent overall semantic interpretation -- and about "spectrum inversion" -- in which subjective quality is imagined to be similarly swappable, say, green looking red and vice versa, again on the assumption that the psychophysics, discourse and behavior could be coherently preserved under the transformation.) To assume that such swaps are possible (always, sometimes, or even ever) is equivalent to assuming that there exist semantic (or behavioral) "duals," very much like mathematical duals such as not/and vs. not/or in the propositional calculus. There is neither proof nor evidence to support such an assumption except in particular formal cases, such as the nonstandard interpretations of arithmetic. However, Searle's point does not depend on this assumption! He's not claiming that the alternative interpretations would be COHERENT; he's only reminding us that projecting an interpretation is clearly all that's involved -- in either the coherent or the incoherent case. The difference is subtle, but crucial to understanding Searle's (perfectly valid) point. Stevan Harnad -- Stevan Harnad Department of Psychology Princeton University harnad@confidence.princeton.edu srh@flash.bellcore.com harnad@elbereth.rutgers.edu harnad@pucc.bitnet (609)-921-7771