Implicit definition, analytic truth, and apriori knowledge

Nous

Volumn 31, Issue 4, 1997, Pages 423-440

  Horwich, Paul ^a  

^a UNIVERSITY COLLEGE LONDON   (United Kingdom)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 60949466931     PISSN: 00294624     EISSN: None     Source Type: Journal    
DOI: 10.1111/0029-4624.00054     Document Type: Review

References (32)

1. See H. Poincaré, Science and Hypothesis, London, Walter Scott Publishing, 1905
2. H. Reichenbach, H., The Philosophy of Space and Time, New York, Dover, 1968
3. R. Carnap, "Empiricism, Semantics and Ontology" (1950) in his Meaning and Necessity, 2nd ed., Chicago, The University of Chicago Press, 1956
4. A.J. Ayer, Language, Truth and Logic (1936), New York, Dover, 1952
5. See the correspondence between Frege and Hilbert in Frege's Philosophical and Mathematical Correspondence, Chicago, University of Chicago Press, 1986, 31-52
6. G. Frege, "On the Foundations of Geometry, First Series (1903) and Second Series (1906) in his Collected Papers on Mathematical Logic and Philosophy, edited by Brian McGuiness, Oxford, Blackwell, 1983
7. W.V. Quine, "Two Dogmas of Empiricism", in his From a Logical Point of View, Cambridge Mass., Harvard University Press, 1953
8. A.N.Prior, "The Runabout Inference-Ticket", Analysis 21 (1960) 38-39
9. and M. Dummett, Elements of Intuitionism, Oxford, Clarendon Press, 1977
10. A.N.Prior, op. cit
11. Nuel D. Belnap, "Tonk, Plonk and Plink", Analysis 22 (1962) 130-133
12. See B. Russell, The Analysis of Matter, London, Allen and Unwin, 1927
13. R. Carnap, Der Logische Aufhau der Welt, Berlin, Schlachtensee Weltkreis-Verlag, 1928
14. F. Ramsey, "Theories" 1929
15. and D. Lewis, "How to Define Theoretical Terms", Journal of Philosophy, LVII, July 1970, 427-466
16. Thanks to Philip Percival for reminding me of this strategy. But I shouldn't have needed reminding, since I adopt it myself in "How to Choose Amongst Empirically Indistinguishable Theories", Journal of Philosophy, LXXIX(2), 1982, pp.61-77
17. and in "ADefense of Conventionalism," Fact, Science and Morality, edited by G. MacDonald and C. Wright, Oxford, Blackwell, 1986
18. In the case of explicit definition-when one says "Let word "F" have the meaning that "G" already has"-only the fourth of these questions arises: namely, how does it come about that "F" acquires the meaning it does For merely saying "Let it be so" obviously isn't enough. The natural solution to this problem, as in the case of implicit definition, invokes the use theory of meaning. We can suppose that what gives "F" the meaning it has is our using it in accordance with certain regularities-and that this is acheived by our adoption of the rules of inference ....F.... ....G... allowing us to go from any sentence containing "F" to the sentence obtained by it by replacing "F" with "G", and vice versa. (Notice that "F" will not thereby obtain exactly the same meaning as "G"; for the above rule provides the basic regularity for the use of the word "F ", but not for the expression "G", which may be a complex whose use is determined by the regularities governing its constituents. For further discussion, see my "The Composition of Meanings", The Philosophical Review, 1998
19. Paul Boghossian makes this point in his "Does an Inferential Role Semantics Rest Upon a Mistake", Mind and Language 8, 1, 1993, 1-27
20. The present discussion is intended to apply only to scientific assertions. Outside the context of science there is no reason to suppose that what we hold true is governed by the Quinean constraints, and so there may well be cases of apriori knowledge. Consider some of the standard alleged examples of what is known apriori: All bachelors are unmarried. If John is taller than Mary, then Mary is not taller than John. If John persuaded Mary to go, then Mary intended to go. "Snow is white" is true iff snow is white. The chess-bishop moves diagonally. Since these are not scientific hypotheses, we need not suppose that their credibility is hostage to empirical considerations. But then on what does their credibility rest It seems to me that a promising answer is the one suggested by the later Carnap (of "Empiricism, Semantics and Ontology" (1950) in his Meaning and Necessity, 2nd ed., Chicago, The University of Chicago Press, 1956): namely, that some beliefs are motivated by purely pragmatic considerations-and these are apriori. Where Carnap went wrong was in ignoring his above-mentioned 1920's insight and attempting to draw this distinction within the domain of substantive scientific hypotheses. As Quine pointed out, this doesn't work. For the pragmatic rationale for adopting a certain hypothesis-namely, convenience-derives from considerations of simplicity. But the epistemological rationale-namely, likely hood of being true-would derive from exactly the same thing. Thus it is impossible to determine objectively whether a given revision is pragmatically or epistemologically motivated. However this difficulty is avoided if we confine ourselves to domains whose purpose is not to construct the simplest overall account of what we observe. In such domains it should be possible to find beliefs whose motivation is wholly pragmatic. Note that to characterize such beliefs as apriori is not to deny that they are revisable. The point is that they are not revisable for empirical reasons, but only for pragmatic reasons. Where does this leave logic, arithmetic and geometry-which are deployed both inside and outside science My own inclination is to say that within science such beliefs are subject to Quine's empirical revisability argument: quantum logic may turn out to be needed in the best overall theory; so the logic, arithmetic, and geometry of science are aposteriori. However, regardless of what science comes up with, it seems likely that in ordinary life, in legal contexts, in game-playing, and other non-scientific areas, we will continue to rely on classical logic, on standard arithmetic, and on Euclidean geometry, My guess is that these commitments can be explained in completely pragmatic terms. They are apriori. Thus we might end up with two logics: an apriori one involving our familiar logical concepts, and an aposteriori one deploying technical versions of "and", "not", "every", etc
21. C. Peacocke, A Study of Concepts (MIT Press, Cambridge, Mass., 1992)
22. and "How Are A Priori Truths Possible", European Journal of Philosophy, August 1993
23. PA. Boghossian, "Analyticity Reconsidered", Noûs 30:3, 1996, pp. 360-391
24. and "Analyticity", A Companion to the Philosophy of Language, edited by C.Wright and R.Hale, Oxford, Blackwell, 1997. As my criticism will indicate, it seems to me that the account of aprioricity that Boghossian develops in these essays conflicts with the above-mentioned insight in his earlier paper (see footnote 12): namely, that what is meaning-constitutive need not be true
25. W.V. Quine, "Reply to Hellman" in The Philosophy of W.V.O. Quine edited by L.Hahn and P.Schillp (La Salle, Ill.; Open Court) 1986
26. For further discussion see my "Chomsky versus Quine on the Analytic-Synthetic Distinction" Proceedings of the Aristotelian Society, 1991, 95-108
27. "Scientific Conceptions of Language and Their Philosophical Import", Science and Knowledge edited by E.Villanueva, Ridgeview, 1993, 123-133
28. and P.A. Boghossian, op. cit
29. This of course offers the most sketchy of sketches of a use theory of meaning. In order to articulate and defend such a theory one would need to do at least the following: (1) specify the terms in which the regularities of use are to be expressed (presumably they should be non-semantic, but need not be individualistic); (2) show how the compositionality of meaning can be accommodated; (3) address the normative implications of meaning; and (4) motivate the deflationary ideas about truth and reference which (as we saw in footnote [9]) are affiliated with the use theory of meaning. For suggestions about how to do these things see my "Meaning, Use and Truth, Mind, Vol. 104, 414, April 1995, 355-368
30. "What Is It Like to Be a Deflationary Theory of Meaning" Philosophical Issues, 5: Truth and Rationality, edited by E.Villanueva, Ridgeview Publishing Company, 1994, 133-154
31. "Deflationary Truth and the Problem of Aboutness" Philosophical Issues, 8: Truth, edited by E.Villanueva, Ridgeview Publishing Company, 1997
32. "The Composition of Meanings" The Philosophical Review, October 1997