No CrossRef data available.
Article contents
Scepticism and the Second Analogy: a modest proposal
Published online by Cambridge University Press: 05 May 2010
Extract
Despite Decades of scholarly attention certain sections of Kant's first Critique have proved recalcitrant to received readings, canonical interpretations are impossible to come by. Nowhere is this more apparent than in the literature on Kant's treatment of causality in the Second Analogy, where there exists a controversy of many years standing about the success of Kant's arguments in favour of what has come to be known as ‘the causal principle’. For example, contemporary Kant scholars of stature no less than Lewis White Beck and W.H. Walsh argue that Kant offers a uniquely persuasive case for causation. Other interpreters claim that the materials of the Second Analogy fail utterly to provide what is needed to vindicate judgements of cause and effect in a way that would satisfy the sceptic about causation.2 Thus the despairing conclusion of one recent review of the literature on the Second Analogy:
It is possible that there lurks somewhere in the pages of the Critique of Pure Reason a convincing reply to Hume's sceptical doubts about the causal principle. But no such reply has yet been brought to light.
- Type
- Articles
- Information
- Dialogue: Canadian Philosophical Review / Revue canadienne de philosophie , Volume 18 , Issue 1 , March 1979 , pp. 27 - 40
- Copyright
- Copyright © Canadian Philosophical Association 1979
References
Note
1 See Beck, L. W., “Once More Unto the Breach,” in Ratio, vol. 9 (1967), pp. 33–37,Google Scholar“The Second Analogy and the Principle of Indeterminancy,” Kant-Studien, vol. 57 (1966), pp. 199–205,Google Scholar and his recent article, “Is There a Non Sequitur in Kant's Proof of the Causal Principle?”, Kant-studien vol. 67 (1976), pp. 385–589.Google Scholar The views of Walsh, W. H. are to be found in “Kant on the Perception of Time,” reprinted in Kant Studies Today (LaSalle: Open Court, 1969), ed. by Beck, Lewis White, pp. 160–180,Google Scholar and the section titled “Kant on Causality,” in his book, Kant's Criticism of Metaphysics (Edinburgh: Edinburgh University Press, 1975), pp. 135–142.Google Scholar Other recent extended defences of the theme ofthe Second Analogy include Wolff, R. P., Kant's Theory of Mental Activity (Cambridge: Harvard University Press, 1963),Google ScholarBird, G., Kant's Theory of Knowledge (London: Routledge, 1962),Google ScholarSchipper, E. W., “Kant's Answer to Hume's Problem,” Kant-Studien, vol. 53 (1961), pp. 68–74Google Scholar and Williams, M. E., “Kant's Reply to Hume,” Kant-Studien, vol. 55 (1965), pp. 71–8.Google Scholar For especially careful analyses of the text of the Second Analogy as well as adjudications of some of the extremes of interpretation see Suchting, W. A., “Kant's Second Analogy of Experience,” Kant Studies Today, pp. 322–340,Google Scholar and Melnick, Arthur, Kant's Analogies of Experience (Chicago: University of Chicago Press, 1973),Google Scholar especially chapter 3.
2 The critical literature is enormous, but any list would have to include Lovejoy, Arthur O., “On Kant's Reply to Hume,” reprinted in Kant: Disputed Questions (Chicago: Quad-rangle Books, 1967)Google Scholar edited with an introduction by Gram, Moltke S., pp. 284–308,Google ScholarStrawson's, P. F. variant of Lovejoy's criticism, The Bounds of Sense (London: Methuen), pp. 133–149,Google Scholar as well as the productive Ratio exchange: L. W. Beck, “Once More Unto the Breach,” op. cit., Murphy, J. G., “Kant's Second Analogy as an Answer to Hume,” Ratio, vol. II (1969), pp. 75–8,Google ScholarWilliams, M. E., “The Breach Again,” Ratio, vol. 11 (1969), pp. 79–81,Google ScholarBeck, L. W., “Rejoinder to Professors Murphy and Williams,” Ratio, vol. 11 (1969, pp. 82–7.Google Scholar Also see Cleve's, James Van useful critical survey, “Four Recent Interpretations of Kant's Second Analogy,” Kant-Studien, vol. 64 (1973), pp. 71–87Google Scholar.
3 Cleve, James Van, op. cit., p. 87.Google Scholar
4 I stress the fact that the Kantian argument pictured in this paper is directed against only a single form of scepticism in order to clearly distinguish my aims from the unfortunate tendency of the last decade to view Kant as responding to every conceivable kind of scepticism. For good evidencelhat this global interpretation of the anti-sceptical thrust of the first Critique is misguided see Wilson, Margaret D., “On Kant and the Refutation of Subjectivism,” Proceedings of the Third International Kant Congress, ed. by Beck, Lewis White (Dordrecht: Reidel, 1972), pp. 597–606,CrossRefGoogle Scholar and Guyer's, Paulreview of Walsh's Kant's Criticism of Metaphysics, The Philosophical Review, vol. LXXXVI (1977), pp. 264–70.Google Scholar Consider as well my “An Anti-Skeptical Argument at the Deduction,” Kant-Studien, vol. 67 (1976), pp. 550–69,Google Scholar especially the last paragraph.
5 Cf. the famous ‘dogmatic slumbers’ passages at the Introduction to theProlegomena to Any Future Metaphysics, trans, by Beck, L.W. (Indianapolis: Bobbs-Merrill, 1950), p. 8Google Scholar and the references in the Critique, e.g. B5, B19–20, A760/B788f.
6 A Treatise ofHuman Nature, ed. by Selby-Bigge, L.A. (Oxford: Oxford University Press, 1888; reprinted 1965), p. 79–80.Google Scholar References to the Critique of Pure Reason, trans, by Smith, N. Kemp (London: Macmillan, 1929)Google Scholar are imbedded in the text of this paper.
7 On this I follow Wolffs, R. P. lucid account in Kant's Theory of Mental Activity, p. 28.Google Scholar For the purposes of this paper I have omitted discussion of whether Hume is properly viewed as a sceptic about causation.
8 See as well, Treatise, p. 10. The idea of a distinct idea or perception is never given any careful working out by Hume, but his treatment ofcausation shows his test for distinctness i s that a difference in propositional objects entails a difference in ideas. Most likely, as is suggested by his treatment of identity later on in the Treatise, mere temporal difference in the description of propositional objects would be sufficient for marking off distinct ideas, but this matter is not crucial here.
9 The sympathetic addition I am thinking of is just the logically true principle that if it is logically possible that B does not have a cause then it is not logically necessary that B be caused.
10 This, I take it, is the force of his distinction between relations of ideas and matters of fact. “The sciences of Geometry, Algebra and Arithmetic” contain only intuitive or demonstrative certainties. “Matters offact… are not ascertained in the same manner; nor is our evidence of their truth, however great, of a like nature with [relations of ideas].” Hume's Enquiries, ed. by Selby-Bigge, L. A. (Oxford: Clarendon Press, reprinted 1963), p. 25. I am using 'justified certainty' and 'sufficient justification' in the sense that '5 has justified certainty in p' (or ‘5 has sufficient justification for p’) entails ‘p is true.’Google Scholar
11 It could be objected that for Hume justified certainty is broader than analyticity. There is some slight evidence in the Enquiry (IV, I) that Hume might treat e.g. “present testimony of the senses” as somehow privileged. Would this textual fact invalidate my ascriptions of SJ to Hume? I think not. For even if Hume is committed to claiming that one can have sufficient justification in remarking on present sense experiences, one could, without prejudgement or artificiality, restrict SJ so that its substitution instances would includeonly those statements that Hume marks off as claims of fact or reason. Furthermore, the mere fact that Hume separates present testimony of the senses from the set of claims about which we can demand justification, will not by itself show us that Hume thought that sufficient justification attaches to first person claims about present sense experience. For Hume's point couldjust as well be that it makes no sense at all to speak ofjustification — to ask how do you know — in this case. I am grateful to one of the anonymous referees of Dialogue for the objection and to Ron Yoshida for the second counter.
12 In speaking of this as Hume's sceptical challenge I do not mean to suggest that it is the only form of sceptical argument one can find in Hume. Indeed I think a case could be made for claiming that Hume's positive analysis of causation with its attendent psycho-genetic account can generate sceptical doubts. See, for example, Price, H. H., “The Permanent Significance of Hume's Philosophy,” reprinted in Human Understanding, ed. by Sesonske, A. and Fleming, N. (Belmont: Wadsworth), p. 16,Google Scholar with his suggestion that Hume's positive account “of Necessary Connection seems altogether too subjective to be tolerable.”
13 See, for example, Treatise, p. 76.
14 At least this seems obvious from the line I have omitted which introduces the passage: “Let us suppose that there is nothing antecedent to an event, upon which it must follow according to a rule (A 194/B239).” A similar reference to what seems to be the causal principle occurs in a clause which I have deleted at the second set ofdots of omission. My deleting these references represents the sympathetic aspect of my reading which I discuss below.
15 Murphy, Jefrie G., “Kant's Second Analogy As an Answer to Hume,” p. 75.Google Scholar The counter objection proposed here is inspired by Murphy's remarks, but I do not claim that is precisely his.
16 This need not seem obviously wrong. True enough, one could take in the perception of an event sequence in one glance, but under the Humean test for distinctness of ideas (i.e., as individuated by propositional objects) one can still speak of several ideas or perceptions in one glance.
17 Of course how we make such discriminations is complicated; the claim here is simply that i t cannot be by isolating before and after relations between one's perceptions.
18 I borrow Jonathan Bennett's fine phrase.
19 This is an admittedly truncated version of an anti-sceptical argument which I have reconstructed from Kant's Deduction of the Categories. See “An Anti-Skeptical Argument at the Deduction,” op. cit., especially pp. 559–564.
20 Sometimes Kant does seem to be tempted by a move of this sort: “If, then we experience that something happens, we in so doing always presuppose that something precedes it on which it follows according to a rule. Otherwise I should not say of the object that it follows. For mere succession in my apprehension… does notjustify me in assuming any succession in the object.” (A 195/B240). But one wants to know how the line which begins with “For” can serve tojustify the two preceding. Lovejoy seems to have passages like these in mind when he criticizes Kant. See, for example, op. cit., p. 302.
21 Clearly the fact thatjustified certainty must be broader than analyticity cannot be sufficient for establishing the existence of synthetic a priori truths, since it will take additional argument to show that these non-analytic but completelyjustified beliefs are not synthetic a posteriori. Nonetheless, an expression of caution is not to be confused with a confession of fault: the main points of this essay do, I think, provide the essential beginnings for an argument to establish that there must be synthetic a priori truths.
22 In making this restriction I mean to rule out probabilistic causation from the scope of the Chalkboard. If you like, the Chalkboard only applies to events at the macro level, and then o t only a significant subset of these.
23 Given the current wave of interest in de re modalities this suggestion should not seem too shocking.
24 I am grateful to Graham Bird and John King-Farlow and my colleague Ron Yoshida for critical comments.