Reply to Gibson

Citation preview

Reply to Gibson Jerrold J. Katz Philosophical Issues, Vol. 4, Naturalism and Normativity. (1993), pp. 174-179. Stable URL: http://links.jstor.org/sici?sici=1533-6077%281993%294%3C174%3ARTG%3E2.0.CO%3B2-6 Philosophical Issues is currently published by Ridgeview Publishing Company.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/journals/rpc.html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is an independent not-for-profit organization dedicated to and preserving a digital archive of scholarly journals. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Fri May 18 08:53:07 2007

PHILOSOPHICAL ISSUES, 4 Naturalism and Normativity, 1993

Reply to Gibson Jerrold J. Katz

1. Gibson defends Quine against the criticisms in chapter 5 of MM. Gibson's first set of comments concerns Quine's indeterminacy argument. He echoes Quine's complaint in "Comments on Katz" (in Perspectives on Quine, Basil Blackwell, 1990, p. 199) that I misconstrue his argument for the indeterminacy of translation and his argument in "Two Dogmas of Empiricism": "The crux of the matter is that when Katz construes Quine's argument as an attempt to prove the thesis of indeterminacy he misconstrues Quine's intent" (p. 170). I hoped that this red herring had been disposed of in the appendix to chapter 5 where I replied to Quine's responses to my "The Refutation of Indeterminacy", The Journal of Philosophy, 1988, pp. 227-252 (from which the material in the body of chapter 5 was taken). I pointed out that I did not claim that Quine's argument is "a proof by cases, but only that it had a form like a proof by cases" (p. 199). Rather than asking for proof -surely an absurd demand in philosophy- I asked only for a reason of the sort that normally counts as grounds for a philosophical claim. My conclusion was that Quine failed to provide even a minimal reason for thinking that sense concepts cannot be made objective sense of or for believing that translation is indeterminate.

What is surprising is that, in his response to me, Quine (1990, p. 198) seems to agree.' He says that, if the definitions of synonymy and other sense concepts in my semantic theory can be shown to meet the usual demands for systematicity, simplicity, and predictive power, then he has no objection in principle either to the theory or to the intensional entities that go along with it. Quine's willingness to leave it to linguistic research to determine whether those methodological demands are met is conceding that his former philosophical objections to the theory of meaning cannot be maintained. What makes this all the more surprising is that, since, as I have argued and Quine does not contest, those objections are necessary for his claim that there are no linguistically neutral meanings. Conceding that there are no such philosophical objections opens up a gap in the argument for i n d e t e r m i n a ~ y . ~ Gibson reiterates Quine's claim that I have the arguments for doubting intensions and for the indeterminacy of translation backwards. Here is Quine (1990, pp. 198-199): "It is the other way around: doubts about intensions come from reflecting on radical translation. They are doubts about how empirical criteria could in general determine what intension is determined by a sentence". Here is Gibson: "The thought experiment of radical translation is meant to strip away any historical or cultural connections between the source and target language, laying bare the objective empirical data for translation. It is the poverty of those objective empirical d a t a which give rise to doubts about translation" (pp. 170-171). Gibson's complaint, like Quine's, does not come to grips with my challenge to radical translation in MM (pp. 201-202). I am charging Quine with begging the question. Quine puts forth radical translation as if it were an unobjectionable idealization of actual translation, one which merely abstracts away from irrelevant complications like "historical or cultural connections". But the "idealization" also 'This is confirmed in t h e correspondence between Quine and Chalmers Clark in Clark, C. (Meaning, Skepticism, and Truth i n the Immanent Naturalism of W.V . Quine, P h . D . Dissertation, Ph.D. Program in Philosophy, Graduate Center, CUNY, 1993). 2 ~ o whaving , made this critical concession, Quine goes on t o express skepticism about whether such a theoretical definition will be scientifically successful -something which, no doubt, remains t o be established. I said t h a t Quine's expressions of skepticism here cannot amount t o more than "personal doubts" because he has never examined t h e theory of meaning as thus far developed t o determine whether it shows signs of success. Since Quine's skepticism is not based on such a n examination, I think, contrary t o Gibson, t h a t t h e judgment of it as just "personal doubts" is not "too harsh".

lacks what Quine calls "independent controls", that is, evidence rich enough to break evidential symmetries. Since the question at issue is whether actual translation involves such controls, Quine cannot assume an "idealization" in which they are absent and then argue from their absence to evidential symmetry and indeterminacy of translation. Therefore, doubts about whether there are sufficient objective data in actual translation to determine intensions cannot be based on radical translation. The issue is not whether in radical translation there are doubts about what intensions sentences express. There have to be, since without "independent controls" evidence is restricted to judgments about referential properties and relations, and hence will be the same for rival translations, no matter what the linguist does (as illustrated in the case of "Gavagai"). The issue is whether in actual translation there are such doubts. Whether Quine's doubts about intensions in radical translation carry over to actual translation depends on the relation between radical translation and actual translation. That, in turn, depends on the nature of actual translation, in particular, whether it involves a rich enough set of "independent controls". To show that it could, I told a jungle story of my own in MM (p. 194). In my story, the field linguist, faced with the choice of translating "gavagai" as "rabbit", "rabbit stage", or "undetached rabbit part", is not restricted to asking the informant about extensional properties and relations. He or she can query the informant about intensional properties and relations, too. For example, the linguist might ask whether "gavagai" bears the same relation to a native expression that "finger" bears to "hand" or "handle" bears to "knife". By asking informants about the ambiguity, antonymy, synonymy, redundancy, superordination, and other sense properties and relations of relevant examples, the linguist eventually obtains a body of objective data that, together with scientific methodology, provide independent controls that, in principle, determine the English translation of "gavagai" . 3 If my jungle story is true, Gibson is wrong to say that there is a "poverty of.. . objective empirical data which give rise to doubts about translation" (p.170-171). Therefore, if doubts about intensions come from reflecting on radical translation, as Quine says, the doubts can only be possible doubts of the sort that one might raise about any scientific result. They cannot amount to a grounds for believing in indeterminacy until we are supplied with a reason to think 3 ~ h reaction e t h a t a question is begged in asking about sense properties and relations is dealt with in MM (pp. 195-196).

that actual translation is like Quine's jungle story -where there are no independent controls and hence a poverty of data- rather than like my story -where there are independent controls and hence no poverty of data. So, we have no reason for doubting the determinacy of actual translation, unless Quine has somewhere given us a reason for thinking that radical translation is an appropriate model of actual translation. No such reason is to be found in Word and Object (MIT Press, 1960). Quine (pp. 75-76) does say that thinking of translation as on a par with other sciences "misjudge[s] the parallel" because there is no fact of the matter in the former case, i.e., no language neutral meanings. But this statement does not supply the reason, since Quine gives no argument for it in Word and Object. The natural place to look for a reason to think that there are no independent controls is "Two Dogmas of Empiricism". The existence of independent controls, as is clear from the above, rests on the availability of sense properties and relations, and, in that paper, Quine argues that no objective scientific sense can be made of them. Moreover, "Two Dogmas" is the only other source for such a reason, since later works directly cite it or assume its argument. But since the only reason given in that paper mistakenly assumes that substitution procedures are the only basis for explaining concepts in linguistics, Quine has given us no good reason to accept indetermina~y.~ 2 Gibson has "two worries" about my conception of semantics. T h e first is that he can see no non-question begging way of deciding whether analyticity and other sense properties and relations reflect the speaker's psycho-biology, as Chomsky claims, or "something like Platonic form", as I claim in Language and Other Abstract Objects (Basil Blackwell, 1981). We can remove that worry by making the distinction between grammatical theories in linguistics proper and 4 I agree with Gibson t h a t there is a small exegetical mystery here about what might be Quine's "philosophical motive for arguing for indeterminacy in Word and Object ten years later". There are, I think, various solutions. One is t h a t Quine came t o suspect t h a t his former arguments against synonymy and analyticity did not deliver as strong a criticism of intensionalism as he wanted, and so he set out t o construct a stronger one. Another is t h a t he thought those argument conclusively remove t h e possibility of independent controls in translation and wanted t o delineate the consequences. However, it does not m a t t e r whether t h e mystery about Quine's motives can be solved. If t h e elusive argument is not in "Two Dogmas" and related papers, it doesn't exist a t all. And if it doesn't exist a t all, Quine has provided no reason for thinking t h a t translation is indeterminant.

ontological interpretations of such theories in the philosophy of linguistics. That distinction is parallel to that between number theory in mathematics proper and Platonism, Intuitionism, and Nominalism in the philosophy of mathematics. Chomsky and I might agree linguistically on the value of a semantic theory which makes the analytic/synthetic distinction, but disagree philosophically on how the theory should be interpreted. Chomsky's conceptualism and my realism play no role in decisions about whether grammars ought to mark the analyticity of sentences. In Semantic Theory (Harper & Row, 1972, pp. 171-197) and elsewhere I have proposed,'roughly speaking, that a sentence is analytic just in case the representation of its predicational structure is part of the representation of the structure of one or more of its terms. The evidence for that definition is matter of considerations internal to linguistics proper, for example, that it generalizes analyticity from standard subject-predicate sentences like "Bachelors are unmarried" to relational sentences like "Bachelors marry those whom they wed". Thus, both Chomsky and I can accept the definition, just as both a conceptualist and realist in the philosophy of mathematics can accept the Dedekind definition of real numbers. Nor is there a problem of question begging in connection with the philosophy of linguistics, though Chomsky and I disagree here. As a conceptualist, he claims that analyticity is au fond a fact about the mindlbrain; as a realist, I claim that it is a fact about linguistic types -sentenceswhich are abstract objects. Here Gibson's question would be how to decide which (if either) claim is true, and the answer is that we decide on the basis of philosophical arguments of the sorts we use in, say, the philosophy of mathematics to argue for conceptualism (Intuitionism) and realism. The ontological issue is no different in the philosophy of linguistics. I would be surprised if Gibson's worry were that there might be no non-question begging way to answer that question, since that question is part of both the larger issue in the philosophy of linguistics between conceptualism and Platonism and the general philosophical issue between conceptualism and realism. Presumably, Gibson is not arguing that general philosophical questions such as the existence of abstract objects are unanswerable. Gibson's second worry is really two worries. One is the fear that my conception of semantics leads to grammars with an "unchecked proliferation of senses" because there is no way to know when to stop decomposing senses. T h a t worry disappears when we notice that the problem here is an instance of a general problem about the completeness of theoretical analyses. Ideally, we stop analyzing

when no further decomposition can increase the predictive power, simplicity, or systematicity of the theory. Gibson's other worry is how we can know that a decompositional analysis is correct, since there are cases where speakers can make no clear judgment, or where different speakers' judgments conflict. In Syntactic Structures (Mouton & Co., 1957, pp. 13-14), Chomsky explained why such situations are not obstacles to grammatical analysis: grammar construction does not require that all cases be clear; we can let our best theory determine the treatment of unclear cases. In linguistics, as in other sciences, theories are a way of bringing the evidence from clear cases to bear in deciding unclear cases. Finally, Gibson's claim that I have "yet to explain the identity conditions for senses" is surely an oversight on his part. I have consistently maintained (most recently, I