Article contents
Confirmation and the Nomological
Published online by Cambridge University Press: 01 November 1978
Extract
We all suppose that it is sometimes reasonable to project common properties, to argue from the premise that each member of a sample possesses a certain property to the conclusion that members of a population from which the sample is drawn also have that property. Even the most ardent anti-inductivist in the philosophy classroom can be found arguing in just this way when he buys a bottle of wine, takes an aspirin or is deciding whether to wear a raincoat.
This fact constitutes a major challenge, for (as everyone knows) instantial confirmation, simple induction, the straight rule - call it what you will - faces serious philosophical difficulties. This paper is concerned with two of these difficulties, namely, the one which arises from the fact that any finite sample exemplifies indefinitely many common properties and the one which arises from terms like ‘grue’ and ‘emerose’.
- Type
- Research Article
- Information
- Copyright
- Copyright © The Authors 1980
Footnotes
This paper extends and modifies ideas first floated in §III of JacksonFrank “Confirmation and the Nomological”, Journal of Philosophy,72 (1975), 113–131. We are indebted to many discussions with friends and colleagues, including Terry Boehm, Len O'Neill, David Stove and, particularly, Brian Ellis.
References
1 In, e.g., Goodman, Nelson Fact. Fiction, and Forecast. 2nd ed., New York, 1965;Google Scholar Scheffler, Israel The Anatomy of Inquiry, New York, 1963;Google Scholar Pollock, John Knowledge and Justification, New Jersey, 1974;Google Scholar Strawson, P. F. Introduction to Logical Theory, London, 1952.Google Scholar
2 It does not matter for the arguments of this paper whether it is, strictly, properties, predicates, or open sentences that are projected when you employ simple induction.
3 For arguments against other interpretations of Goodman's new riddle see “Grue”, op. cit.. §I and §II.
4 Goodman, “Comments”, Journal of Philosophy. 63 (1966). 328–331.CrossRefGoogle Scholar p. 329. The classical approach is complicated by the need to detail the connection between projectible predicates and projectible conditionals and hypotheses, see, e.g., Davidson, D. “Emeroses by Other Names”. Journal of Philosophy, 63 (1966), 778–780,CrossRefGoogle Scholar and Pollock's comments in Knowledge and Justification. op. cit., p. 236. But our discussion is independent of these details, and for instance could, with a slight increase in complexity, have been directed at classical responses which seek to handle the selection problem by dividing hypotheses into the projectible and the nonprojectible.
5 See, e.g., Goodman, Fact, Fiction, and Forecast, op. cit., ch. 4. Goodman sees entrenchment as the basic notion which underlies our Judgements of relative simplicity, see his “Safety, Strength, Simplicity”, Philosophy of Science, 28 (1961).
6 See, e.g., Carnap, Rudolf “On the Application of Inductive Logic”, Philosophy and Phenomenological Research, 8 (1947), 133-47.CrossRefGoogle Scholar
7 See, e.g., Salmon, W. C. “On Vindicating Induction”, Philosophy of Science, 30 (1963), 252–261.CrossRefGoogle Scholar
8 See, e.g., Quine, W. V. “Natural Kinds”, in Ontological Relativity, New York, 1969, pp. 114–138,CrossRefGoogle Scholar and Schlesinger, G. Confirmation and Confirmability, Oxford, 1974, ch. 1.Google Scholar
9 See Pollock, op. cit., ch. 8.
10 For other examples illustrating the same general point see “Grue”, op. cit., and Kelley, M. “Predicates and Projectibility”, Canadian Journal of Philosophy, 1 (1971), 189–206.CrossRefGoogle Scholar
11 D. Davidson, op. cit., see also Pollock, op. cit.
12 Op. cit., p. 237. Similar points apply to troubles arising from hypotheses with disjunctive antecedents, but we will not detail them here.
13 Op. cit., p. 225. See also the papers cited in n. 1. Incidentally, Goodman sometimes appears to make the doctrine true by definition, see, e.g., Fact, Fiction, and Forecast, op. cit. But this, of course, leaves the substantive issue untouched, merely effecting a redescribing of it in terms of the adequacy of such a definition.
14 See e.g., Pap, A. “Disposition Concepts and Extensional Logic”, in Minnesota Studies in Philosophy of Science, Vol. II, ed. Feigl, H. et. al., Minnesota, 1958.Google Scholar
15 Fact, Fiction, and Forecast, op. cit., p. 73.
16 ‘Laying the Raven to Rest’, Journal of Philosophy, 70 (1973). 747-754, seep. 750, our italics.
17 See, e.g., Ellis, B. D. “A Vindication of Scientific Inductive Practices”, American Philosophical Quarterly, 2 (1965). 1–9.Google Scholar
18 ‘So-called’ because such attempts at Justification are best seen not as claims about word usage but as alleging a logical interdependence between SI and the concept of rationality. See, e.g., Pollock, Knowledge and Justification, op. cit., p. 204.
- 11
- Cited by