Hume's Problem: Induction and the Justification of Belief (review)

Hume Studies Volume 29, Number 1, April 2003, pp. 155-162 COLIN HOWSON. Hume's Problem: Induction and the Justification of Belief. Oxford : Clarendon Press, 2000. Pp. 261. ISBN 0-19-825037-1, cloth, $35. Hume's Problem comprises two main projects: (a) defending Hume's argument about induction against a dozen or so purported answers, and (b) laying out a logic of induction that incorporates Hume's great insight in a formal theory. In this review, I will look at several instances of Howson's defense of Hume; then I will sketch the broad outlines of Howson's own "answer," the details of which are myriad and sometimes technical. I "There is no good reason to suppose that inductive practice should have been successful at all." Such is the upshot of Hume's iron logic regarding the "problem of induction." How, then, do we account for the "undoubted fact that induction not only seemed to work but to work surpassingly well" (10)? Just as mathematicians have practically been able to ignore Gödel's proofs about the limits of completeness and consistency, so scientists have been oblivious of Hume's "problem" with induction. Astronomers' predictions of solar and lunar eclipses are really and truly more reliable than the vaticinations in the daily horoscope. The situation is neatly encapsulated in C. D. Broad's characterization of induction as "the glory of science and the scandal of philosophy." Before we decide that Hume is indeed wrong about induction, we should make sure that we have his argument straight. Here is Hume himself: 'Tis evident, that Adam with all his science, would never have been able to demonstrate, that the course of nature must continue uniformly the same, and that the future must be conformable to the past. . . . Nay, I will go farther, and assert, that he could not so much as prove by any probable arguments, that the future must be conformable to the past. All probable arguments are built on the supposition, that there is this conformity betwixt the future and the past, and therefore can never prove it. (T Abstract, 14; SBN, 651) Most of Hume's readers (then and now) would grant that there is no deductive link between statements about the past and statements about the Hume Studies 156 Book Reviews future, but many bridle at the sundering of even a probable link between the two. Can Hume be right about this? Right about what? He never says that we are wrong to believe in induction, that the future will not resemble the past. He says only that we cannot make sound inductive inferences on the basis of observation alone. On pain of patently begging the very question at issue, we may not use any inductively established premises—premises that give pride of place to some hypotheses over others that are also compatible with the observational data. Howson thinks that we tend to underestimate the force of Hume's argument and exaggerate the devastation it is supposed to wreak. Although Hume's argument is unassailable (an irresistible force, so to speak), it nevertheless leaves intact something of inestimable value; namely, "the logic of science itself" (119). Here, then, by way of anticipation, is Howson's (halfway) positive solution to Hume's problem: There are "demonstrably sound inductive inferences"; but they do not—and must not be claimed to—justify induction. In canvassing the putative solutions to Hume's problem, Howson examines the major 'isms' of twentieth-century philosophy of science: reliabilism, realism, falsificationism, naturalism, and Bayesianism—not to mention the Anthropic Principle, significance tests, and miracles. He also considers eight "quick responses" to Hume's problem, some of which do not fit comfortably under any of the 'isms' just listed. I will briefly sample a few quickies. 1. D. C. Stove and J. L. Mackie contend that Hume equates "probable reasoning " with "reasoning concerning matters of fact and existence." Here is Mackie's verdict: "Reasonable but probabilistic inferences, then, have not been excluded by Hume's argument, for the simple reason that Hume did not consider this possibility" (cited on 13). Howson contends—correctly, I think— that the historical and textual evidence against Mackie, Stove, et al...