Movatterモバイル変換

1. Fodor (1987: 16) calls the conjunction of (1) and (2)therepresentational theory of mind (RTM). However, it seems betterto reserve that name for the more general view that mentalrepresentations exist. One can accept the general view while rejecting(1) and (2). For example, one can say that mental representationsfigure in subpersonal perceptual activity but deny that they figure inthought.

2. Fodor and Pylyshyn (2015: 28) claim that pictures have parts but notconstituents. They argue on that basis against imagistic theories ofthought.

3. Hume analyzed complex ideas inassociationist terms: acomplex idea arises when associative bonds form between simple ideas.In his words:

’tis impossible the same simple ideas should fall regularly intocomplex ones… without some bond of union among them, someassociating quality, by which one idea naturally introduces another.(1739 [1978] I.i.4)

As Fodor (2003: 93)emphasizes, Hume conflatedcomplex ideas withcausalsequences of ideas. For example, there is a fundamentaldifference between your beliefthat John is tall and a causalsequence during which you think about John and then as a result thinkabout tallness. Your belief is evaluable as true or false. The causalsequence is not; it expresses no truth-evaluable content. Associativebonds between simple ideas can give rise to causal sequences of ideas,but they cannot on their own give rise to truth-evaluable mentalstates. In particular, the capacity to form associative bonds does notexplain the capacity forpredication (e.g. the capacity topredicate tallness of John). Thus, the associationist approach seemsunable to explain the compositional mechanisms instantiated by humanthought. The basic point here goes back to Kant (1781 [1998]). Forfurther discussion, see the entry onassociationist theories of thought.

4. Fodor (1975: 174–194) advances some Wittgenstein-inspired objections toimagistic theories of thought. He claims that a picture on its owndoes not express any truth-evaluable content. For example, a pictureof a man climbing up a hill could just as well be a picture of a mansliding down a hill. Based on such examples, Fodor urges that“[p]ictures aren’t the kind of things that can havetruth-values” (1975: 181). Only when we supplement a picture withadditional descriptive information (e.g. the man is climbing ratherthan sliding) does it become truth-evaluable. Since propositionalattitudes are truth-evaluable, Fodor concludes that mentalpictures/images could not serve as the direct objects of propositionalattitudes. The thesis that pictures taken on their own are nottruth-evaluable is also present in Frege (1918 [1997: 326–327]).In more recent literature, it is widely disputed (Burge 2018;Greenberg 2013).

5. Gibson (1979) insists that perceptual psychology should abjure mentalrepresentations. He advances a non-representationalist framework forthe scientific study of perception. Fodor and Pylyshyn (1981) argueconvincingly that Gibson’s non-representational framework failsto explain very basic phenomena, such as perceptual illusions, thatare readily explicable on a representational approach.

6. Let SEM₁ be the thesis that some transitions betweencomputational states are sensitive to semantic properties.SEM₁ concerns computationaldynamics. It should bedistinguished from doctrines that concern computationalindividuation. Let SEM₂ be the thesis thatsemantic properties inform the individuation of some computationalstates. For example, one might hold that certain mental computationsare implemented only by physical systems with appropriate semanticproperties. Let SEM₃ be the thesis that semantic propertiesinform the individuation ofall computational states.SEM₃ entails SEM₂, but not vice-versa: one mighthold thatsome computations are individuated semantically(SEM₂) and that others are individuated in a purelysyntactic fashion (¬SEM₃). Rescorla (2014b) defendsSEM₂ + ¬SEM₃. Shagrir (2020)defends SEM₃. Piccinini (2008) defends¬SEM₂. SEM₁ is compatible withSEM₃ and with SEM₂ + ¬SEM₃.However, neither SEM₂ nor SEM₃ entailsSEM₁: one might hold that semantic properties bear upon thenature of computational states, and hence play an individuative role(SEM₃), but that computational transitions are notsensitive to semantic properties (¬SEM₁). Fodor(1981: 226–227; 1998: 9–11) advocates¬SEM₁ + SEM₃. In the main text, thephrase “semantic conception of computation” is used forSEM₁. That phrase and similar phrases are sometimes insteadused for SEM₂ or SEM₃.

7. In later work, Fodor distinguishes between “innate” and“unlearned” (2008: 144–145).

8. Some memorable passages in Ludwig Wittgenstein’sPhilosophical Investigations (1953:§§243–271) contain what is known asthe privatelanguage argument. In these passages, Wittgenstein considers andrejects the possibility of aprivate language:

The words of this language are to refer to what can be known only tothe speaker; to his immediate, private, sensations. So another cannotunderstand the language. (§243)

For several decades, the private language argument was regarded bysome philosophers as a devastating refutation of the very idea of amental language. From a contemporary vantage point, it is unclear whatargument if any Wittgenstein intended, or whether any cogent argumentcan be extracted from what Wittgenstein says. Wittgenstein’sdiscussion certainly poses little evident threat to LOTH (Fodor, 1975:68–73). A private language as defined by Wittgenstein differsfundamentally from the mental language as postulated by LOT theoristsfrom the medieval era to the present day. Words in a private languagedenote only the speaker’s “immediate, privatesensations”. Mentalese words can denote anything that naturallanguage words can denote: dogs, cities, quarks, numbers, etc. Indenying the possibility of a private language, Wittgenstein does notengage with LOTH.

9. The problem is a version of Quine’s (1951 [1980]) celebratedattack uponthe analytic/synthetic distinction. A version of theproblem also arises for molecular versions of functional rolesemantics (Fodor 1990: ix–xi).

10. One might hold that, in some cases, a Mentalese word has certainaspects of its meaning essentially but does not have its denotationessentially. Consider Mentaleseindexicals. One might posit aMentalese indexicalhere that necessarilyrefers to the thinker’s present location. On this view,heredoes not have its denotation essentially. Thelocation denoted byhere changes as thethinker moves through space. Nevertheless,herehas a kind of meaning essentially. It expressesa fixedcharacter, in Kaplan’s (1989) sense.

11. Prompted by Frege cases, Fodor (1994) recruits Mentalese words toserve as modes of presentation. He individuates the words in formalsyntactic fashion. Fodor’s approach differs significantly fromFrege’s, because it flouts the Fregean doctrine that“sense determines reference”. Semantically permeatedindividuation is closer in spirit to Frege’s approach, since ittakes the Mentalese word to determine a unique denotation.