Hostname: page-component-78c5997874-j824f Total loading time: 0 Render date: 2024-11-10T06:39:42.684Z Has data issue: false hasContentIssue false
  • English
  • Français

Not Actually Hume's Problem: On Induction and Knowing-How

Published online by Cambridge University Press:  08 October 2008

Stephen Hetherington
Stephen Hetherington
Affiliation:
University of New South Wales
Get access Rights & Permissions [Opens in a new window]

Abstract

Philosophers talk routinely of ‘Hume's problem of induction’. But the usual accompanying exegesis is mistaken in a way that has led epistemologists to conceive of ‘Hume's problem’ in needlessly narrow terms. They have overlooked a way of articulating the conceptual problem, along with a potential way of solving it. Indeed, they have overlooked Hume's own way. In explaining this, I will supplement Hume's insights by adapting Ryle's thinking on knowledge-how and knowledge-that. We will also see why Hume's ‘sceptical solution’ was a perfectly appropriate response to his ‘sceptical argument’ — rather than (as is often thought by analytic epistemologists) a merely descriptive response patently missing the normative sceptical point so strikingly formulated by Hume.

Type
Research Article
Information
Philosophy , Volume 83 , Issue 4 , October 2008 , pp. 459 - 481
Copyright
Copyright © The Royal Institute of Philosophy 2008

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 Hume's Problem: Induction and the Justification of Belief (Oxford: Clarendon Press, 2000).

2 ‘Epistemology Naturalized’, Ontological Relativity and Other Essays (New York: Columbia University Press, 1969), 72.

3 I will present this version of Humean scepticism in contemporary epistemological terms, talking of epistemic (inductive) justification and of (inductive) knowledge. In §4, I will explain my preferred replacements for such talk in this case.

4 Hume reaches this verdict, of course, having rejected the alternative of some demonstrative a priori reasoning underlying such a belief.

5 It is not a parsing bereft of substantial scholarly support. Two influential advocates of versions of this form of interpretation are Stove, D. C., Probability and Hume's Inductive Scepticism (Oxford: Clarendon Press, 1973)CrossRefGoogle Scholar and Stroud, Barry, Hume (London: Routledge & Kegan Paul, 1977), 50–67CrossRefGoogle Scholar.

6 And this response of his has attracted few epistemological supporters. Is it therefore of no epistemological value? Perhaps epistemologists have not accurately understood what Hume meant by it. I will return to this issue soon, beginning in §3.

7 See The Concept of Mind (London: Hutchinson, 1949), ch. II; ‘Knowing How and Knowing That’, Collected Papers, Vol. II (London: Hutchinson, 1971), 212–225.

8 ‘How to Know (That Knowledge-That is Knowledge-How)’, Epistemology Futures, (ed.) S. Hetherington (Oxford: Clarendon Press, 2006), sec. 2.

9 I did so, as part of defending Ryle against criticism by Jason Stanley and Timothy Williamson, ‘Knowing How’, The Journal of Philosophy 98, No. 8 (August 2001), 411–444.

10 Strictly (as John Williams has pointed out to me), R's falsity might be thought to allow the possibility that a given instance of one's knowing how to F (while doing F) involves propositional knowledge without involving one's knowing how to apply it. But (in what is, I assume, a Rylean spirit) I mean the negation of R to involve only the negation of one's-having-prior-propositional-knowledge-by-which-one-is-being-guided. The different aspects of this complex (‘hyphenated’) property are present entirely, as a group — or not at all. What is in question when we assess intellectualism is propositional knowledge's playing a specific sort of guiding role, not its mere presence, say. See John Williams, ‘Know-How and Propositional Knowledge’, Synthese (forthcoming).

11 This general result is not new with me, as §6 will explain. But because there is something new in my version, I will outline it before discussing (in §6) kindred interpretations by other philosophers.

12 My page references for Hume will be to the Enquiry's second edition, (ed.) L. A. Selby-Bigge (Oxford: Clarendon Press, 1902), a volume that also includes Hume's Enquiry Concerning the Principles of Morals.

13 But it almost is; and §4 will seize upon this kinship.

14 Note, too, that nowhere in this account does Hume deny that an inductive inference can lead to an accurate belief, a true belief. The problem is not one of inferring accurately. Again, it is one of what constitutes the action — the intelligent action — as such of inferring inductively.

15 In calling this stability-knowledge (as I am terming it) a ‘supposition’, though, is Hume distinguishing it from knowledge? Not clearly, because earlier (36; my emphasis) Hume puts the same point in terms of

the similarity which we discover among natural objects, and by which we are induced to expect effects similar to those which we have found to follow from such objects.

16 Here is where, crucially, most contemporary epistemologists will read more into Hume than is required, strictly speaking. His phrase, ‘reasoning or process of the understanding’, will be assumed to denote something normative, not merely psychological. However, the simplest interpretation of Hume here is as denoting a faculty of reason, rather than as denoting and approving of it.

17 But what kind of (inductive) knowledge is thereby at stake? In §4 I revisit this question. (Here is a hint of what we will find there, though. I said that the ‘relevant contrary of ignorance, I assume, is knowledge.’ However, a more careful, still Humean, reading would replace ‘is knowledge’ with ‘is at least true belief’. So, what we gain from inductive habit or custom is at least (inductive) true belief. And if we cultivate the right habits or customs, this process of gaining (inductive) true beliefs can be done well. In which case, the result of doing this is an inductive skill — a skill at forming true beliefs inductively.)

18 I do not call it simply ‘pragmatism’, because that term is used so widely and heterogeneously. Still, my practicalism is one possible restricted application of pragmatism.

19 We might not want to deny that any knowledge-how is also knowledge-that; nor need Hume have denied this. In knowing how to form a true belief that the sun will rise tomorrow, we may well wish to grant, you know that the sun will rise tomorrow. (Certainly we might wish to grant both such knowledge-how and such knowledge-that, if you know well how to form a true belief that the sun will rise tomorrow.) But maybe this is because the knowledge-that is merely knowledge-how. I explain, and argue for, that thesis in ‘How to Know (That Knowledge-That is Knowledge-How)’, op. cit. All I am insisting on right now, on Hume's behalf, is that the successful inductive thinker is forming beliefs that are at least manifestations of inductive knowledge-how.

20 I intend the phrase ‘usually true’ to reflect some sort of reliabilist precisification. This is just one of the possible precisifying paths we could take here. And so an externalist interpretation of the knowledge-how is available to us, all else being equal. In contrast, the intellectualist picture is clearly internalist. It is what Hume was considering, in envisaging an inductive thinker who uses her inductive evidence self-consciously and deliberatively. This is not to say that Hume, as interpreted here, is endorsing externalism and rejecting internalism. But he is rejecting one especially intellectualist version of internalism.

21 You might not have this associated inductive stability-knowledge anyway, if Hume's supposedly sceptical argument is right in its claim about an infinite regress.

22 Thus put, I am denying one notable instance of a pertinent variation on the controversial KK-principle. The latter claims that one's knowing that one knows that p is a necessary condition of one's knowing that p. The instance of the variation I am denying is something like this: One has inductive knowledge of how to predict accurately that p, only if one knows that one does.

23 One elaborative line of thought may be David Papineau's: ‘Reliabilism, Induction and Scepticism’, The Philosophical Quarterly 42, No. 166 (January 1992), 1–20. He conceives of inductive justification as reliabilist justification, and he offers a rule-circular justification of induction — an inductive defence of induction. This accords with my conception of induction as a skill that can be strengthened by being used repeatedly.

24 Hume's Enlightenment Tract: The Unity and Purpose of An Enquiry Concerning Human Understanding (Oxford: Clarendon Press, 2001). All otherwise unattributed page references in this section are to Buckle's book. On Hume's naturalism, more generally, see Buckle's ‘Hume's Sceptical Materialism’, Philosophy 82, No. 322 (October 2007), 553–578.

25 Of course, as part of this denial, Hume is denying that we can have (what we might now call) deductive knowledge-that of the world's underlying causal links — either directly, via reason alone, or indirectly, via premises reporting sensory experience.

26 In this vein, note Hume's assertion, in a letter to John Stewart (February, 1754) that

I never asserted as absurd a proposition as that any thing might arise without a cause: I only maintained, that our certainty of the falsehood of that proposition proceeded neither from intuition nor demonstration; but from another source.

(See Hume's An Enquiry Concerning Human Understanding and Other Writings, (ed.) S. Buckle (Cambridge: Cambridge University Press, 2007), 210.) We may also note Hume's remark, in ‘A Letter from a Gentleman to his Friend in Edinburgh’ (ibid., 157), that

you may judge of the candour of the whole charge, when you see the assigning of one kind of evidence for a proposition, instead of another, is called denying that proposition….

27 Naturally, this will not prevent the inductive knowledge's being fallible. Hume was well aware of that fallibility. And conceiving of inductive knowledge as at least knowledge-how accommodates this, precisely because our fallible skills, even our cognitive ones, are routinely fallible without ceasing to be skills. (‘But when is a Humean habit an instance of knowing-how, a real skill — rather than a mere habit?’ It is knowledge-how insofar as it is a successful habit. ‘How successful must it be?’ That is the question of how good knowledge-how needs to be. Yet this question is not automatically directing us to a problem of philosophical principle for the current suggestion; for in general skills inherently allow for this possibility of being more or less good, even vaguely so, while still being present as genuine skills. It can be indeterminate where exactly to distinguish between an exercise of a particular skill and an exercise not of that skill — without there thereby being no exercises of the skill. Talking of skills enables us to accommodate this point of epistemological principle. And in making this paper's interpretive point about Hume, this is all that needs to be noted about this larger epistemological issue.)

28 Locke had a similar approach, in his Essay Concerning Human Understanding. See I, i, 5:

The Candle, that is set up in us, shines bright enough for all our Purposes.

See also I, i, 6:

’Tis of great use to the Sailor to know the length of his Line, though he cannot with it fathom all the depths of the Ocean. … If we can find out those Measures, whereby a Rational Creature put in that State, which Man is in, in this World, may, and ought to govern his Opinions, and Actions depending thereon, we need not be troubled, that some other things escape our Knowledge.

See, too, Craig's, Edward argument, in The Mind of God and the Works of Man (Oxford: Clarendon Press, 1987)Google Scholar, for ‘one of the most basic points of [Hume's] philosophy’ being the thesis (85)

that where philosophers thought they saw the operations of [a faculty of] reason, the divine spark at work in man, they were watching nothing more than a mundane mechanism and its natural effects in the mind.

29 That contemporary focus could well be needlessly narrow anyway. We should always bear in mind that there might be much more — or indeed much less — to the epistemic world than is currently discussed in epistemology. For some further remarks on this general theme, see my ‘Introduction: Epistemological Progress’, Epistemology Futures, op. cit.

30 M. R. Ayers describes it as ‘the memory of repeated perceptions of similar objects’. See his ‘The Foundations of Knowledge and the Logic of Substance: The Structure of Locke's General Philosophy’, Locke's Philosophy: Content and Context, (ed.) G. A. J. Rogers (Oxford: Clarendon Press, 1994), 63.

31 See also Beauchamp, Tom and Rosenberg, Alexander, Hume and the Problem of Causation (New York: Oxford University Press, 1981), ch. 2Google Scholar; Noonan, Harold W., Hume on Knowledge (London: Routledge, 1999), 110–131Google Scholar; and Beebee, Helen, Hume on Causation (London: Routledge, 2006), 36–45Google Scholar.

32 Cognition and Commitment in Hume's Philosophy (New York: Oxford University Press, 1997), ch. 4.

33 Hume's Reason (Oxford: Oxford University Press, 1999), especially ch. 6. See also his ‘Hume's Doubts About Probable Reasoning: Was Locke the Target?’, Hume and Hume's Connexions, (eds.) M. A. Stewart and John P. Wright (Edinburgh: Edinburgh University Press, 1994).

34 Owen (132–134) explains why, in his view, Garrett's interpretation is mistaken in detail, even if correct in generic thrust. I am not attempting to adjudicate between these two philosophers. My claim right now is just that, no matter which (if either) of them is right, my analysis is general enough to assist in explaining that correctness.

35 The following sort of objection (suggested to me by Paul Snowdon) then needs to be countered:

Maybe the way in which our inductive beliefs come into existence is vital for this paper's interpretation. In particular, is the formation of an inductive belief an action at all? If inductive beliefs are not formed voluntarily, instances of forming them might well not be genuine actions. And if the formation of an inductive belief is not an action, it is also not an exercise of knowledge-how.

Hume himself did not regard belief-formation as subject to the will (48). Would he therefore not have deemed an instance of belief-formation to be an action, an exercise of knowledge-how? Certainly we need not deny this status to such belief-formation. After all, belief-formation like that can be quite skilful, considered purely as a way of registering accurate opinions — no matter that it might be done with no overarching awareness of one's being skilful. In that sense, it can just happen skilfully, given how one is placed within the world. That was Hume's point, too, in talking of custom or habit, and in acknowledging that we do not understand ‘the ultimate reason of such a propensity’ (43). As he realized, customs or habits can help one out as a believer, even if one does not notice, let alone understand, how this is occurring. (And so note 20's point about externalism returns. Even if inductive belief-formation's epistemic merits are to be understood externalistically, this does not entail that skills are not involved. The action in question need not be a consciously planned and controlled action — explicable in intellectualist terms. But this does not entail its not being an action. It can be skilful. Knowing-how is apt.)

36 Recall note 17 above.

37 I am grateful to Stephen Buckle and Paul Snowdon for careful and extensive comments on drafts of this paper. An audience at the University of Sydney was also very helpful.