Skip to main content Accessibility help
×
Home

Hume's Defence of Causal Inference

  • Fred Wilson (a1)

Extract

As is well known, the Humean account of causal inference gives a central location to inference habits. Some of these habits one can discipline. Thus, one can so discipline oneself as to reason in accordance with the “rules by which to judge of causes and effects” (173ff.), that is, one can discipline oneself to think scientifically, rather than, say, in accordance with the rules of prejudice (146f.), or of superstition (271). All such judgments, even those of science, are, however, upon the Humean account fallible. Some, or rather very many, have raised the issue about how any such causal judgment or inference is justified. That is, how can such inference habits be justified, shown to be reasonable. Hume meets this challenge by arguing that there are some habits all men do and must have. Because we must have them we ought to have them. They are, therefore, justified.

Copyright

References

Hide All

1 Hume, D., A Treatise of Human Nature, ed. Selby-Bigge, L. A. (Oxford: Clarendon Press, 1888), 175. Hereafter, numbers in parentheses are page references to this text.

2 For the connection between ought implies can and must implies ought, see Wilson, F., “Mill's Proof that Happiness Is the Criterion of Morality ”, Journal of Business Ethics 1 (1982), 5972.

3 Ibid.; cf. Hall, E. W., “The 'Proof' of Utility in Bentham and Mill”, Ethics 60 (1947), 118.

4 We discuss this in greater detail below, in the final section.

5 Hume, Treatise, I, III, iii. Cf. Beauchamp, T. and Rosenberg, A., Hume and the Problem of Causation (Oxford: Oxford University Press, 1981); F. Wilson, “The Lockean Revolution in the Theory of Science”, forthcoming in a festschrift for R. F. McRae; Hacking, I., The Emergence of Probability (London: Cambridge University Press, 1975); Wilson, F., Critical Notice of I. Hacking's The Emergence of Probability in Canadian Journal of Philosophy 8 (1978), 587597.

6 Hume, D., Enquiries concerning the Human Understanding and concerning the Principles of Morals, ed. Selby-Bigge, L. A. (2nd ed.; Oxford: Clarendon Press, 1902), 154.

7 For further on this, see section 3, below.

8 Hume, Treatise, 172. Note that constant conjunction is of species. As Hume puts it in the Enquiries (Enquiry concerning Human Understanding, Sec. XI, ad fin, 148): “It is only when two species of objects are found to be constantly conjoined, that we can infer the one from the other …” (his italics). He makes the same point in the Treatise, 87. Thus Lewis, D.is simply wrong when he suggests (see his “Causation”, Journal of Philosophy 70 [1973], 556567) that law-deduction is not essential to Hume's account of causation; and, for that matter, to Hume's account of the justified assertion of counterfactuals.

9 Cf. Chisholm, R. M., “Law Statements and Counterfactual Inference”, Analysis 15 (1955), 97105. Modern logicians generally have tried to solve the problem of the validity of induction… and generally admit they have failed. They have not, even according to themselves, shown.

10 Prichard, H. A., Knowledge and Perception (Oxford: Clarendon Press, 1950), 154, suggests that Hume thus reduces all causal inferences to irrationality.

11 For further on this, see section 4, below.

12 Compare the discussion of subjective and objective justification in Moore, G. E., Ethics (London: Oxford University Press, 1912), 118121, where it arises in the context of a discussion of utilitarianism.

13 Ewing, A. C., Fundamental Questions of Philosophy (London: Routledge & Kegan Paul, 1951), 168.

14 Clifford, W. K., “The Ethics of Belief”, in his Lectures and Essays, vol. 2 (London:MacMillan, 1879), 163205.

15 Descartes, R., Meditations, in Philosophical Works of Descartes, ed. Haldane, E. S. and Ross, G. R. T., vol. 1 (New York: Dover, 1955), 105.

16 Cf. Lenz, J., “Hume's Defense of Causal Inference”, in Chappell, V. C., ed., Hume: A Collection of Critical Essays (Garden City, New York: Doubleday, 1966), 99128.

17 Hume, Enquiries, 154.

18 Lenz, “Hume's Defense of Causal Inference”, 184–185.

19 I have examined one crucial aspect of Hume's case in Wilson, F., “Hume's Theory of Mental Activity”, in Norton, D. F. et al., McGillHume Studies (San Diego: Austin Hill Press, 1979), 101120.The relevant point is that the mind can monitor its own activities and then adjust its responses as to better achieve goals that it sets itself. Norton, D. F. (David Hume: Common-Sense Moralist and Sceptical Metaphysician [Princeton: Princeton University Press, 1982]) has recently also strongly emphasized that for Hume the mind is not passive but active, and active in controlling and determining its own activities.

20 Prichard, Knowledge and Perception, 184.

21 Hume, Enquiries, 154.

22 See the discussion of credulity, Hume, Treatise, 112; of the adverse effects of education, ibid., 116; of imagination, ibid., 123; and of unphilosophical probability, ibid., I, III, xiii, 143ff. We discuss these in greater detail in section 3.

23 Lenz, “Hume's Defence of Causal Inference”, 185.

26 Capaldi, N., David Hume (Boston: Twayne Publishing, 1975), 128.

27 Cf. Wilson, “Mill's Proof that Happiness is the Criterion of Morality”, for the same point with respect to Mill's use of the must implies ought principle.

28 For a discussion of the similar point in Mill, and a discussion of how Mill supplies a specific content (viz., the principle of utility) to the generic rule that one ought to seek happiness, see Wilson, F., “Mill's 'Proof' of Utility and the Composition of CausesJournal of Business Ethics 2 (1983), 135155.

29 Feigl, Cf. H., “De Principiis Non Disputandum …?” in Black, M., ed., Philosophical Analysis (Ithaca: Cornell University Press, 1950), 119–156.

30 Ibid. The idea is already in Pierce, C. S.; see his “The Fixation of Belief”, in his Collected Papers, vol. 5 (Cambridge: Harvard University Press, 1934), 223248.

31 Hume, Treatise, 578, 417, 439. It is true that benevolence does not extend to all mankind in a force of sufficient strength to curb self-interest (cf. Treatise, 481), but it does not follow that a benevolence extending to all mankind plays no motivating role.

32 Hume, Enquiries, 165.

33 Cf. Wilson, “Hume's Theory of Mental Activity”. See footnote 19, above.

34 For more details, see Wilson, F., “Is there a Prussian Hume?Hume Studies 8 (1982 118; also “Hume and Ducasse on Causal Inference from a Single Experiment”,Philosophical Studies 35 (1975), 305310.

35 Cf. Wilson, “Hume's Theory of Mental Activity”.

36 Reichenbach, H., Experience and Prediction (Chicago: University of Chicago Press, 1938), chap. 5; Salmon, W., The Foundation of Scientific Inference (Pittsburgh: University of Pittsburgh Press, 1969).

37 Bergmann, G., “Frequencies, Probabilities, and Positivism, Philosophy and Phenomenological Research 6 (1945), 2644.

38 H. Geiringer, quoted in ibid., 36, from Erkenntniss 8 (1939), 151–176.

39 Kuhn, T., The Structure of Scientific Revolutions (2nd ed.; Chicago: University of Chicago Press, 1970), 126.

40 This makes the proposition metaphysical, in Popper's peculiar sense of that term. (Cf. Watkins, J. W. N., “Between Analytic and Empirical”, Philosophy 32 [1957], 112131;Influentialand Confirmable Metaphysics”, Mind n.s. 67 [1958], 344365.) It is, of course, empirical—though non-falsifiable. For the importance of this latter, see , Wilson, “Is there a Prussian Hume? ”; “Mill on the Operation of Discovering and Proving General Propositions”, Mill News Letter 17 (1982), 114; “Kuhn and Goodman: Revolutionary vs. Conservative Science”, Philosophical Studies, forthcoming.

41 J. S. Mill, System of Logic, Bk. Ill, Ch. V. (Cf. Wilson, “Mill on the Operation of Discovering and Proving General Propositions”.) Given Hume's Rules I and II concerning spatial and temporal contiguity, his version of the Law of Universal Causation comes close to asserting that for every event there is a process law that explains, understanding “process law” in Bergmann's sense (see Bergmann, G., Philosophy of Science [Madison: University of Wisconsin Press, 1958], chap. 2). Process implies continuity. Hume's contiguity does not quite amount to continuity, due to Hume's nominalistic doctrine concerning the separability of events. But his nominalism can be distinguished from his logical atomism (Wilson, cf. F., Review of M. Mandelbaum, The Anatomy of Historical Knowledge, in Philosophical Review 88 [1979], 663668; Weinberg's Refutation of Nationalism”, Dialogue 8 [1969], 460474). It is only the weaker thesis of logical atomism which he needs for his account of causation (see Wilson, F., “Acquaintance, Ontology, and Knowledge”, New Scholasticism 54 [1970], 148). While the nominalism is incompatible with continuity logical atomism is not. Concerning the connection of time and causality in Hume, see Beauchamp and Rosenberg,Hume and the Problem of Causation.

42 Strictly speaking, this is an application of Hume's Rule VIII.

43 Bacon clearly recognizes the need for generic constraints; cf. Novum Organum, Book II, aphorism iv.

44 Compare the view of T. Kuhn on paradigms as guides to research. See his The Structure of Scientific Revolutions, Section IV, which describes paradigms as providing puzzles to researchers, guaranteeing that there is a solution to the problem which it is the task of the researcher to discover. See Wilson, “Mill on the Operation of Discovering and Proving General Propositions”, and “Kuhn and Goodman: Revolutionary vs. Conservative Science”.

45 For greater detail on this, see Wilson, “Goudge's Contribution to the Philosophy of Science”, in Sumner, L. W., Slater, J. G. and Wilson, F., eds., Pragmatism and Purpose (Toronto: University of Toronto Press, 1980), 151171.

46 Cf. Wilson, “Mill on the Operation of Discovering and Proving General Propositions”.

47 Cf. ibid. For views similar to Hume's and Mill's on the role of analogy in scientific research, compare Lakatos, I., “Falsification and the Methodology of Scientific Research Programmes”, in Lakatos, I. and Musgrave, A., eds., Criticism and the Growth of Knowledge (London: Cambridge University Press, 1970), 91–169; Hanson, N. R., “Is there a Logic of Scientific Discovery?” in Brody, B., ed., Readings in the Philosophy of Science (Englewood Cliffs, NJ: Prentice-Hall, 1970), 620–633; and, of course, Kuhn, The Structure of Scientific Revolutions.

46 Cf. Wilson, “Mill on the Operation of Discovering and Proving General Propositions”.

47 Cf. ibid. For views similar to Hume's and Mill's on the role of analogy in scientific research, compare I.Lakatos, “Falsification and the Methodology of Scientific Research Programmes”, in I. Lakatos and A. Musgrave, eds., Criticism and the Growth of Knowledge (London: Cambridge University Press, 1970), 91–169; Hanson, N. R., “Is there a Logic of Scientific Discovery?” in Brody, B., ed., Readings in the Philosophy of Science (Englewood Cliffs, NJ: Prentice-Hall, 1970), 620–633; and, of course, Kuhn, The Structure of Scientific Revolutions.

48 Hume, Treatise, 132; this pattern is discussed in Wilson, “Is there a Prussian Hume?”. Compare also Treatise, 105, where Hume discusses another use of the same pattern of inference. This other pattern is discussed in Wilson, “Hume and Ducasse on Causal Inference from a Single Experiment”.

49 One also needs some version of the Principle of Limited Variety; it is not clear that Hume recognized this. Compare Wright, G. H. von, The Logical Problem of Induction (2nd ed.; Oxford: Blackwell, 1957), Ch. IV, sec. 6. In effect, the principle of limited variety, as well as that of determinism, is included in the formula which we labelled (3), above. The limitation of variety, Bacon saw (see footnote 43, above). But Hume does recognize that one is dealing with analogous or parallel cases (see footnote 48, above), and it can be argued that parallelism and analogy can be secured only if generic constraints are presupposed and therefore limited variety.

50 Cf. ibid., sees. 3–5. For the Methods of Agreement and Difference, the point is clear enough. For the Method of Concomitant Variation (Rule VI) and Rule VII, it is only a bit more complicated. Cf. Broad, C. D., Induction, Probability and Causation (Dor-drecht, Holland: Reidel, 1968), 127159.

51 Cf. Wilson, “Hume's Theory of Mental Activity”.

52 Compare Mill's inference that only pleasure is an end, discussed in Wilson, “Mill's Proof that Happiness is the Criterion of Morality”.

53 Hume, Treatise, 96. Hume's account is imagistic. This is the conclusion of a somewhat speculative introspective analysis of the sort proposed by the research programme of associationist psychology. For a discussion of such analysis, See Bergmann, G., “The Problem of Relations in Classical Psychology”, in his The Metaphysics of Logical Positivism (New York: Longmans, Green, 1954), 277299. This is an analysis of an intentional content. One must remember that to analyze is not to analyze away. The intentionality remains.Butler, Cf. R., “Hume's Impressions”, in Vesey, G., ed., Impressions of Empiricism (New York: St. Martin's Press, 1976), 122136. Even if one gives up the imagistic account of belief, almost everything else Hume says would still be defensible.

54 Hume, Treatise, 104–105; cf. Wilson, “Hume and Ducasse on Causal Inference from a Single Experiment”.

55 Cf. Hume, Treatise, Bk. II, Part II, sec. ii, for a similar putting of the psychological hypotheses to the test.

56 Cf. ibid., 439ff.

57 Ibid., 416–417. As self-interest curbs itself and creates the artificial virtues to do that curbing (cf. ibid., 492), so it is self-interest that curbs those passions that lead us to make law-assertions contrary to those that satisfy the passion of curiosity: as Hume puts the point succinctly, “Generally speaking, the errors in religion are dangerous; those in philosophy only ridiculous” (ibid., 272).

58 Rational belief as well as irrational belief must receive a causal explanation. There is no difference in this respect between the two: they both receive the same sort of explanation. Empiricists like Feigl have always followed Hume on this point. See, for example, Feigl, H., “Philosophical Tangents of Science ”, 13 (in Feigl, H. and Maxwell, G., eds., Current Issues in the Philosophy of Science [New York: Holt, Rinehart, and Winston, 1961], 119). That there is a difference in kind in the explanation of rational and non-rational (including irrational) processes has recently been argued, but hardly convincingly, by Laudan, L., Progress and its Problems (Berkeley: University of California Press, 1977), chap. 7. The Humean and empiricist position has been forcefully re-stated in David Bloor, Knowledge and Social Imagery (London: Routledge & Kegan Paul, 1976). The thrust of Bloor' s argument ought to be a commonplace. The heat that his essay has produced shows how far Laudan's work is typical of a widespread drift away from empiricism.

59 Cf. Wilson, “Goudge's Contribution to the Philosophy of Science”. The term “imperfect” is Bergmann's; see his Philosophy of Science, chap. 2.

60 Cf. footnote 42, above.

61 As Carnap says, one must take into account the total evidence when making inductive inferences; see Carnap, R., Logical Foundations of Probability (2nd ed.; Chicago: University of Chicago Press, 1962), sees. 44, 45 and 4951.Salmon, Cf., The Foundations of Scientific Inference, 76, 9395.

62 Cf. Wilson, “Mill on the Operation of Discovering and Proving General Propositions”.

63 Francis Bacon, Novum Organum, Book I, aphorism cv; see also Book I, aphorism civ and Book II, aphorism xv. For an excellent discussion of Bacon, see Broad, C. D., The Philosophy of Francis Bacon (London: Cambridge University Press, 1926). The discussion in Leeuwen, H. van, The Problem of Certainty in English Thought (The Hague: Nijhoff, 1963), overemphasizes Bacon's pursuit of certainty. To be sure, Bacon aims at a method that will produce certainty relative to the uncertainty that is the inevitable upshot of induction by simple enumeration. But to say that is not to say Bacon aims at the certainty of the rationalists, as van Leeuwen suggests. All Bacon need be read as aiming at is what later would be called moral certainty, which is to say, fallible knowledge. For a discussion of the empiricist/rationalist clash on this point, see Wilson, “The Lockean Revolution in the Theory of Science”, and also his Critical Notice of Hacking, The Emergence of Probability.

64 Bacon, Novum Organum, Book II, aphorism xv. Von Wright, The Logical Problem of Induction, 206, justly refers to the usual Popperian claims as “exaggerations”.

65 Cf. footnote 10, above.

66 These laws themselves receive a Humean analysis, of course. Cf. Wilson, “Hume's Theory of Mental Activity”.

67 Feigl, “Principiis Non Disputandum… ?”, 136.

68 Feigl, H., “Philosophy of Science of Logical Positivism”, in Feigl, H. and Scriven, M., eds., Minnesota Studies in the Philosophy of Science, vol. 1 (Minneapolis: University of Minnesota Press, 1956), 29.

69 Goudge, T. A., The Thought of C. S. Pe9irce (Toronto: University of Toronto Press, 1950), 193; cf. Wilson, “Goudge's Contribution to the Philosophy of Science”.

70 Cf. von. Wright, The Logical Problem of Induction, chap. 8; Lenz, J. W., “Carnap on Defining Degree of Confirmation',” Philosophy of Science 23 (1956), 230236;Levi, I., “Hacking, Salmon on Induction”, Journal of Philosophy 62 (1965).

71 Cf. the rejection of induction by simple enumeration, Hume, Treatise, 150, discussed above.

72 Madden, E. H., “The Riddle of Induction”, Journal of Philosophy 55 (1958).

73 Cf. Wilson, “Mill on the Operation of Discovering and Proving General Propositions”.

74 E.g., in periods of what Kuhn, The Structure of Scientific Revolutions, refers to as revolutionary science.

75 Mill, System of Logic, Bk. VI, Ch. XII, sec. 2.

76 Ibid., Bk. VI, Ch. I.

77 Or at least, so it is argued in Wilson, “Mill's Proof that Happiness is the Criterion of Morality”.

78 Cf. Capaldi, David Hume, 128.

79 , Sartre, Being and Nothingness, trans. Barnes, H. (New York: Washington Square Press, 1966), 112116. The same point with respect to Mill's use of must implies ought is discussed in Wilson, “Mill's Proof that Happiness is the Criterion of Morality”.

80 Sparshott, Cf. F. E., “In Defense of Kemp-Smith”, Hume Studies 1 (1975), 68.

81 Cf. Wilson, “Hume's Theory of Mental Activity”.

82 Cf. F. Wilson, “Hume's Sceptical Argument against Reason”, Hume Studies, forthcoming.

Hume's Defence of Causal Inference

  • Fred Wilson (a1)

Metrics

Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed