In lieu of an abstract, here is a brief excerpt of the content:

Hume's Conception of Science j. P. MONTEIRO HUME'S CONCEPTION OF SCIENCE has been interpreted as a kind of "observationalism ." This term was coined by Randall, who accuses Hume of reducing science to the discovery of laws describing observed relations, and of holding that relations between objects are known only by means of observation.' But we find in Ayer the same interpretation: Hume would accept as scientific, besides formal propositions, only those that describe events of the same kind as events that have actually been observed; all general hypotheses would be nothing but generalizations from observation." Also, according to Passmore, Hume would conceive science as the mere discovery of" the observable behavior of thingsP And Basson asserts that Hume forbids, in science, any reference to entities that are not possible objects of common experience? Finally, Popper finds in Hume, as in Locke and Berkeley, the thesis that the only valid nonformal propositions are those that can be verified by the evidence of the senses: Hume would conceive scientific theories as a kind of "digest" of observationsP This is a widely held interpretation: for Hume, the limits of science would be those of generalization from observation; any reference to unobservables would go beyond those limits, and could have no place in any Humean scientific enquiry. If Hume's theory of causal inference, or inductive inference, 6 was also his John Herman Randall, Jr., The Career of Philosophy, 2 vols. (New York: Columbia University Press, t966), 1: 636, 637. 9 A.J. Ayer, Language, Truth and Logic (Harmondsworth: Pelican, 1972), p. 181; Introduction to British Empirical Philosophers (New York: Simon and Schuster, n.d.), p. 13. s John Passmore, Hume's Intentions (London: Duckworth, 1968), p. 50. 4 A. H. Basson, David Hume (Harmondsworth: Pelican, t958), p. 29. 5 Karl Popper, Objective Knowledge (Oxford: Clarendon Press, 1973), p. 127; Conjectures and R~rutations (London: Routledge & Kegan Paul, 1974), p. 46. e Hume himself did not employ these terms, but they are widelyemployed by his interpreters . For causal inference, see Norman Kemp Smith, The Philosophy of David Hume (London: Macmillan, 196o),pp. t loft.; John Lenz,"Hume's Defense of Causal Inference," in Hume, V. C. Chappell, ed. (London: Macmillan, t97o), pp. t69ff.; T. E.Jessop, "Some Misunderstandings of Hume," in the same volume, p. 42. For inductive inference, see H. H. Price, "The Permanent [327] 328 HISTORY OF PHILOSOPHY theory of science, this interpretation would certainly be undisputable. Inductive inference, as defined by Hume, is strictly bound within the limits of the observable: causal relations can be established only between observables. Causes, as well as effects, must be observable objects or events: "All belief of matter of fact and real existence is derived merely from some object, present to the memory or senses, and a customary conjunction between them and some other object.": If we define a "Humean cause" as one which is conformable to Hume's theory of causal inference, 8 then every Humean cause must be an observable species of object or event. 9 Inductive inferences transcend observation, but only in the sense that they go beyond observed instances, not beyond observable events. Causal relations established by induction give us assurance "concerning objects which are removed from the present testimony of our memory and senses"; ,~ but only the observed is thus transcended, not the observable. If from previous observation I have been able to infer that fire causes heat, and if tomorrow I go into a room and feel a sensation of heat, I have no need to look at the fire to know it is there, causing my sensation. Inductive inferences allow us to discover causes before they "discover themselves" to our senses: the inference acts as a substitute for direct observation. But this presupposes the previous observation of a regular conjunction between two observable species of objects or events, to which belong the observed particular effect and the cause inferred from it. Observability is a necessary condition for a cause to qualify as a term in an inductive inference. The situation remains essentially the same even if we accept, with Davidson , that Hume sometimes seems to hold that causes are not exactly events (or objects, or entities), but...


Additional Information

Print ISSN
pp. 327-342
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.