The needs of understanding: Kant on empirical laws and regulative ideals

International Journal of Philosophical Studies

Volumn 5, Issue 2, 1997, Pages 216-254

  O'Shea, James R ^a  

^a UNIVERSITY COLLEGE DUBLIN   (Ireland)

Author keywords

Causality; Empirical laws; Kant; Reason; Regulative maxims; Understanding

Indexed keywords

EID: 61249174655     PISSN: 09672559     EISSN: None     Source Type: Journal    
DOI: 10.1080/09672559708570855     Document Type: Article

References (70)

1. The literature on the Second Analogy is enormous; the following is a small selection of the works particularly relevant to the dispute here. In criticism of Kant's argument (Lovejoy, Strawson) or alternatively in favour of 'weak' versions of the Second Analogy (to be discussed below), see: A. O. Lovejoy, 'On Kant's Reply to Hume' in Kant: Disputed Questions, ed. Moltke S. Gram (Chicago: Quadrangle Books, 1967), pp. 284-308;
2. Lewis White Beck, 'A Prussian Hume and a Scottish Kant,' in Essays on Kant and Hume (Yale University Press, 1978), pp. 111-29;
3. P. F. Strawson, The Bounds of Sense: An Essay on Kant's 'Critique of Pure Reason' (London: Methuen, 1966), e.g. p. 137;
4. Gerd Buchdahl, Metaphysics and the Philosophy of Science (Cambridge, MA: MIT Press, 1969);
5. 'The Conception of Lawlikeness in Kant's Philosophy of Science,' in Kant's Theory of Knowledge, ed. L. W. Beck (Dordrecht: Reidel, 1974);
6. 'The Kantian "Dynamic of Reason" with Special Reference to the Place of Causality in Kant's System,' in Kant Studies Today, ed. L. W. Beck (La Salle, Ill.: Open Court, 1969), pp. 187-208;
7. Susan Neiman, The Unity of Reason: Rereading Kant (Oxford University Press, 1994), ch. 2;
8. Stephen V. Bagne 'Objects of Representations and Kant's Second Analogy' Journal of the History of Philosophy (1994), 32: 381-410;
9. and Henry Allison, Kant's Transcendental Idealism: An Interpretation and Defense (Yale University Press, 1983), ch. 10.
10. and Idealism and Freedom: Essays on Kant's Theoretical and Practical Philosophy (Cambridge University Press, 1996), ch. 6. I shall focus on Allison's work, referred to henceforth as KTI and I&F respectively.
11. On what Allison calls the 'strong' reading, which I shall defend, see Michael Friedman, 'Causal Laws and the Foundations of Natural Science,' in The Cambridge Companion to Kant, ed. Paul Guyer (Cambridge University Press, 1992), pp. 161-97;
12. Arthur Melnick, Kant's Analogies of Experience (University of Chicago Press, 1973);
13. Patricia Kitcher, Kant's Transcendental Psychology (Oxford University Press, 1990), e.g. p. 176;
14. and Jeffrey R. Dodge, 'Uniformity of Empirical Cause-Effect Relations in the Second Analogy', Kant-Studien (1982) 73: 47-54.
15. For subtle intermediate positions, see e.g. Robert E. Butts, Kant and the Double Government Methodology (Dordrecht: Reidel, 1986), e.g. pp. 239, 247;
16. and Paul Guyer, Kant and the Claims of Knowledge (Cambridge University Press, 1987), ch. 10;
17. see also Guyer, 'Kant's Conception of Empirical Laws,' Proceedings of the Aristotelian Society, Supplementary Volume: (1990) 221-42.
18. Strawson, The Bounds of Sense, p. 137. The objection was originally raised by A. O. Lovejoy in 1906, in 'On Kant's Reply to Hume'.
19. For references, see sections III-V below. I should note in addition that R. E. Butts (Kant and the Double Government Methodology) has argued for the overall coherence of Kant's account of the regulative maxims (see esp. chs. VII-IX).
20. Philip Kitcher in 'Projecting the Order of Nature' (in Robert E. Butts, ed., Kant's Philosophy of Physical Science, Dordrecht: Reidel, 1986) has previously stressed the connection between the Second Analogy and the perplexity regarding the status of the regulative principles of reason. Although I depart from Kitcher's account in certain important respects (see below), his interpretation is insightful and has the substantial merit of developing an independently defensible and broadly Kantian philosophy of science.
21. See also Kitcher, 'Kant's Philosophy of Science,' in A. Wood, ed., Self and Nature in Kant's Philosophy (Ithaca: Cornell University Press, 1983).
22. and 'The Unity of Science and the Unity of Nature', in Paolo Parrini, ed., Kant and Contemporary Epistemology (Dordrecht: Kluwer, 1994);
23. as well as his other writings on the philosophy of science, e.g. 'Explanatory Unification', Philosophy of Science (1981) 48: 507-31;
24. and The Advancement of Science (Oxford University Press, 1993).
25. For a clear account of these matters, see Graham Bird's 'Kant's Transcendental Arguments' in Reading Kant, ed. E. Schaper and W. Vossenkuhl (Oxford: Blackwell, 1989).
26. It seems to me that the reading of the Second Analogy I offer here preserves the sharp conceptual distinction between the transcendental and the empirical that Bird has argued is necessary for giving any coherent account of Kant's transcendental idealism (see his Kant's Theory of Knowledge (London: Routledge & Kegan Paul, 1962), for anticipations of many later developments in Kant interpretation along these lines).
27. In particular, the view to be defended here is compatible with the fact cited by Bird (essentially a scope-distinction) that for Kant to 'recognise an event is to presuppose that some causal factor is present, but not to presuppose any particular causal factor', as well as with the point that the principle of the Second Analogy 'needs to be supplemented empirically in order to yield any particular causal law' (Kant's Theory of Knowledge, ibid., p. 162). We do not need the Buchdahl/Allison 'weak' reading of the Second Analogy in order to capture these distinctions.
28. This would seem to be one important respect in which the present account differs from Kitcher's interpretation (see note 7). Kitcher argues that 'the Appendix to the Ideal completes the work of the Second Analogy by explaining how we are justified in distinguishing merely accidental regularities from those that are endowed with counterfactual-sustaining force' ('Projecting the Order of Nature', p. 221; italics added).
29. As I see it, the Second Analogy is already fully successful in justifying the judgment that there exists some counterfactual-sustaining causal law (as opposed to an accidental regularity) covering any given event. The purpose of the regulative principle of systematic unity is not to 'generate' the notion of a counterfactual-sustaining causal law ('The Unity of Science and the Unity of Nature', p. 260)
30. For recent support of this claim see, in particular, the opening sections of Michael Friedman's 'Causal Laws and the Foundations of Natural Science'.
31. (see Friedman's groundbreaking work, Kant and the Exact Sciences (Harvard University Press, 1992)):'. . . ordinary experience at the more commonsensical level is [thus] to become fully objective and determinate [only] as science progresses
32. (Friedman, 'Kant and the Twentieth Century', in P. Parrini, ed., Kant and Contemporary Epistemology, pp. 35, 38).
33. see, e.g., Kant's Lectures on Logic, J. Michael Young tr. and ed. (Cambridge University Press, 1992), pp. 608-9;
34. References to this work ('MFNS'), as well as to Kant's Prolegomena to Any Future Metaphysics ('Prol.') are to James W. Ellington, trans., Kant's Philosophy of Material Nature (Indianapolis: Hackett, 1985), followed by marginal page references to Ak. IV.
35. For important recent work on the place of the MFNS in Kant's system as a whole, see Michael Friedman, Kant and the Exact Sciences, as well as the papers collected in Robert E. Butts, ed., Kant's Philosophy of Physical Science: Metaphysische Anfangsgründe der Naturwissenschaft 1786-1986 (Dordrecht: Reidel, 1986), and further references therein.
36. See MFNS Ak. 476-7.
37. See the accounts of the Second Analogy in Allison (Kant's Transcendental Idealism; e.g. p. 225)
38. and Guyer (Kant and the Claims of Knowledge, e.g. pp. 247-9) for guidance past this and several other mistaken objections to the Second Analogy, as well as better help than most with Kant's difficult 'irreversibility' claim (which I have only briefly treated here).
39. See also Susan Neiman, The Unity of Reason, ch. 2, for a similarly vigorous defence of the Buchdahl interpretation. As far as I can see, the considerations I offer here in response to Allison's objections address most of Neiman's concerns as well. In general, Neiman offers a powerful synoptic view of the importance of reason across all of Kant's critical philosophy, much of which I can agree with despite the present disagreement (concerning which, if I am right, she underestimates the justificatory role of the understanding in relation to reason's ideals).
40. On induction, see Kant's Lectures on Logic, pp. 625-7 (Ak. IX: 131-3);
41. and cf. the Appendix to the Dialectic, e.g. A649/B677, A653/B681.
42. Buchdahl, Metaphysics and the Philosophy of Science, p. 650;
43. quoted in Allison KTI, p. 231.
44. Patricia Kitcher, Kant's Transcendental Psychology, p. 178.
45. For helpful recent discussions of the Appendix focusing on Kant's philo- sophical methodology of science, see Thomas E. Wartenburg, 'Reason and the Practice of Science,' in Guyer, ed., The Cambridge Companion to Kant, pp. 228-48;
46. and Kathleen Okruhlik, 'Kant on Realism and Methodology,' in Butts, ed., Kant's Philosophy of Physical Science, pp. 307-32.
47. See also Margaret Morrison, 'Methodological Rules in Kant's Philosophy of Science', Kant-Studien (1989) 2: 155-72. Note that I am arguing for a stronger thesis than that the regulative maxims of reason are necessary for the possibility of systematic empirical science in Kant; they are necessary for the possibility of any understanding at all.
48. Norman Kemp Smith, A Commentary to Kant's 'Critique of Pure Reason' (London, 1923), p. 547.
49. See Jonathan Bennett, Kant's Dialectic (Cambridge University Press, 1974), for a similarly negative assessment of the coherence of Kant's Appendix chapter.
50. Rolf-Peter Horstmann, 'Why Must There Be a Transcendental Deduction in Kant's Critique of Judgement"?' in Kant's Transcendental Deductions (Stanford University Press, 1989), p. 259 n9.
51. Horstmann, Kant's Transcendental Deductions, ibid., p. 259 n10 and p. 166. Horstmann refers to another article by Wartenburg (noted below) who, pace Horstmann, correctly attributes this (allegedly) 'rather un-Kantian' idea to Kant.
52. Paul Guyer, 'Reason and Reflective Judgment: Kant on the Significance of Systematicity' Nous (1990) 24: (17-43): 39.
53. Thomas E. Wartenburg, 'Order Through Reason: Kant's Transcendental Justification of Science', Kant-Studien (1979) 70: 417n. Wartenburg's detailed and insightful exposition of the argument of the Appendix correctly stresses the objective side of Kant's principle of systematicity. I agree with much of his analysis, although I put more stress on the role of the understanding in the justification of the principle, as well as emphasizing the key part played by the notion of an indeterminate objective validity.
54. As the preceding has hopefully made clear, the texts do not support Guyer's claim that in the First Critique Appendix 'there is no hint that systematicity is a necessary condition for any successful use of understanding at all' ('Reason and Reflective Judgement', p. 33).
55. Compare the title of Patricia Kitcher's work, Kant's Transcendental Psychology (Oxford, 1990).
56. Even if this revolution is a form of idealism, it is in the first instance a methodological idealism. (See Henry Allison, Kant's Transcendental Idealism (Yale University Press, 1983), p. 25: '[T]ranscendental idealism must be characterized primarily as a metaphilo-sophical or methodological "standpoint", rather than as a straightforwardly metaphysical doctrine about the nature or ontological status of the objects of human cognition.') It is a further question whether this implies some kind of ontological idealism, according to which things are ontologically dependent on minds.
57. For the emphasis on the metaphysical dimension of Kant's transcendental inquiry, see Martin Heidegger, Kant and the Problem of Metaphysics (Frankfurt a. M., 1973).
58. For a similar view, see Allison, Kant and the Problem of Metaphysics , op. cit. note 6, chs 1-2.
59. In tracing the development of Kant's thought in the 'silent years' from 1770 to 1781, Wolfgang Carl is emphatic that Kant's epistemology cannot be naturalized: 'Kant's theory of transcendental faculties should not be understood by appealing to an analogy, which it is in any case difficult to follow, with cognitive psychology. Transcendental imagination or transcendental apperception are not psychological faculties, to be understood descriptively; rather, they function as conditions of and elements in the analysis and explication of a certain "kind of knowledge'"; and 'Here [with regard to the transcendental synthesis of the imagination], we are not concerned with this or that specific connection brought about by the imagination as a cognitive faculty. . . . This synthesis and its unity cannot be derived from an empirical inquiry by cognitive psychology': Der schweigende Kant (Göttingen, 1989), pp. 126, 141, my translations.
60. D. Walford and R. Meerbote (eds) Immanuel Kant. Theoretical Philosophy, 1755-1770 (Cambridge, 1992), pp. 373-466. Page references after quotations are to the second volume of the Academy edition.
61. (Kant and the Claims of Knowledge (Cambridge, 1987), pp. 14-15). But if we note that the argument relies on the variability of the receptive faculties of different subjects, we realize why Kant thinks that receptivity implies subject-dependence. This is not, as Guyer thinks, because receptivity is passive, but because receptivity is variable from subject to subject. This also explains why the fact that a representation is created by the mind does not by itself imply an analogous subject-dependence. This is not because such creation is active, but because it may be not variable but 'stable' or 'fixed' across subjects (see below). I return to this point at the end of the next section.
62. Moses Mendelssohn based his rationalist method in metaphysics on the view that any more empirical method 'exposes metaphysics to the attacks of Pyrrhonism'. (Mendelssohn's prize-winning essay of 1764, in Walford and Meerbote, op. cit. note 14, p. 281.) Mendelssohn's essay was familiar to Kant.
63. Sextus Empiricus, Outlines of Pyrrhonism, I. 59 (trans. J. Annas and J. Barnes).
64. He equally considers sceptical attacks on objectivity that are based on the argument from relativity, in particular the attacks of Ernst Platner (pp. 290-2)
65. For a similar understanding of the transcendental line of reasoning, see Allison, op.cit. note 6, e.g. p. 27: '[B]ehind Kant's formal idealism, lies [the] principle . . . that whatever is necessary for the representation or experience of something as an object, that is, whatever is required for the recognition or picking out of what is "objective" in our experience, must reflect the cognitive structure of the mind (its manner of representing) rather than the nature of the object as it is in itself. By object in itself Allison means object 'in abstraction from our cognitive relation to it' (p. 336, n31).
66. For this reading of Sextus, see M. Frede, 'Des Skeptikers Meinungen', Neue Hefte für Philosophie (1979) 15/16: 102-29.
67. Guyer, 'Des Skeptikers Meinungen', Neue Hefte für Philosophie (1979, op. cit. note 15, p. 15.
68. Allison, Neue Hefte für Philosophie , op. cit. note 6, p. 83, emphasizes this logical possibility, but associates it with his reading of außer mir in B38/A23, as meaning not spatially outside me, but rather independent of my mind in some way. I think außer mir must here mean spatially outside me, hence I do not want to associate the logical possibility of alternative forms of sensibility with any particular reading of the außer mir, however crucial this term is for understanding Kant's argument for the claim that the representation of space is a priori.
69. Cf. J. Michael Young, 'Functions of thought and the synthesis of intuitions', in P. Guyer, (ed.) The Cambridge Companion to Kant (Cambridge, 1992), p. 117: 'There may be no contradiction, perhaps, in the notion of an intellect that does not depend on sensibility . . . But we cannot claim to comprehend what such an intellect would be like. Neither can we use the notion of such an intellect as a basis for determining what things must be like "in themselves".'
70. Op. cit. note 4, p. 23, my emphasis.