1 All references to the Critique are to the standard first and second edition pagination. I have used Schmidt's, R. edition of the Kritik der reinen Vernunft (Hamburg: Felex Meiner 1954Google Scholar). Translations are from Norman Smith's, Kemp Kant's Critique of Pure Reason (New York: St. Martin's 1929Google Scholar), except where an alteratinn has been made and indicated by a ‘T’ after the page citation.

2 I discuss the work by Sellars, Heidemann, Ameriks, and Kitcher in the notes be· low. See also Weldon, T.D. Kant's Critique of Pure Reason 2nd ed. (Oxford: The Clarendon Press 1958Google Scholar).

3 It should be noted that Metaphysical Neutrality by itself does not entail Noumenal Ignorance. We might think it possible to present an epistemology which involves no commitment to a position on what kind of thing thinks, without at all thinking that nothing can be said about that issue, as a separate topic.

4 Cf. Kritik der Urteilskraft, in Kants gesammelte Schriften, Kiiniglich Preussischen Akademie der Wissenshaften (Berlin, Leipzig: de Gruyter 1922) (AA, hereafter), vol. 35, 406; Critique of Judgment, trans!. Bernard, J.H. (New York: Hafner 1968), 254Google Scholar.

5 One of the first suggestions about such an approach can be found in Wilfrid Sellars, ‘Metaphysics and the Concept of a Person,’ in Lambert, Karel ed., The Logical Way of Doing Things (New Haven: Yale University Press 1969), 219-52Google Scholar. See also Kitcher, Patricia ‘Kant's Real Self’ in Wood, Allen ed., Kant on Self and Nature (Ithaca: Cornell University Press 1985), 111-45Google Scholar; and, for a much broader point of view, Dennett, Daniel ‘Artifical Intelligence as Philosophy and as Psychology,’ in Brainstorms (Cambridge: MIT Press 1978), 109-26Google Scholar, especially 111 and 122-6. I might note that a fuller demonstration of why Kant cannot be enlisted into Dennett's project would have to take account of what he says in Chapter Five of this book, his argument that the restriction of AI to some form of mechanism is, given Church's theorem, no real restriction. See also p. 83.

6 This relation is quite an important sub-theme in Ingeborg Heidemann's Spontaneitiit und Zeitlichkeit (Köln: Kölner Universitats-Verlag 1958), although at times she makes Kant far more of a ‘dialectical’ and ‘historical’ thinker than he is. Cf. 13-14, 86, and 260. The issue pursued here – Kant's case against the possible phenomenality of the transcendental subject, and his denial that such a subject could be a causal system – is also an important element of her discussion, but is couched in terms of what she calls the necessary ‘duality’ of ‘temporality’ and ‘spontaneity,’ what she eventually calls the ‘Distanz der Spontaneitat zur Zeitlichkeit,’ 222.

7 Hegel, G.W.F. Gesammelte Werke, vol. 4 (Hamburg: Felix Meiner 1968), 43Google Scholar; The Difference Between Fichte's and Schelling's System of Philosophy, transl. Harris, H.S. and Cerf, Walter (Albany: SUNY Press 1977), 130Google Scholar

8 The original for the key concluding phrase is ‘ein bloss intelligibiler Gegenstand.’ Merrill: New York 1969), 80. Kant is here referring to a distinction he sometimes makes between the spontaneity of the understanding which, however spontaneous must still deal with the content of intuition, and the more unrestricted, purely spontaneous self-legislation of reason. Cf. sections IV and V of the Introduction to the Critique of Judgment.

9 This passage has been the occasion for quite a number of rather speculative interpretations of the Critique on self-knowledge. See Heidemann, all of section #10, 207 ff.; Heinz Heimsoeth, ‘Persönlichkeitsbewusstsein und Ding an sich in der kantischen Philosophie,’ Studien zur Philosophie Immanuel Kants, Metaphysische Ursprüng und Ontologische Grundlagen (Köln, Kölner Universitäts-Verlag 1956), 229-55; and Gottfried Martin, Kant's Metaphysics and Theory of Science, transl. P.G. Lucas (Manchester: Manchester University Press 1955), 176-81.

10 AA, Bd. 4, p. 290. Prolegomena to Any Future Metaphysics, trans. L.W. Beck (Indianapolis: Bobbs-Merrill 1950), 38

11 Reflexion 5413, AA, Bd. 18, 176; see also 5441, 182.

12 AA, vol. 4, 452. Foundations of the Metaphysics of Morals, transl. L. W. Beck (Bobbs Merrill: New York 1969), 80. Kant is here referring to a distinction he sometimes makes between the spontaneity of the understanding which, however spontaneous must still deal with the content of intuition, and the more unrestricted, purely spontaneous self-legislation of reason. Cf. sections IV and V of the Introduction to the Critique of Judgment.

13 Kant's Theory of Form (New Haven: Yale University Press 1982)

14 Kitcher, ‘Kant's Real Self,’ 124

15 Kitcher, 124

16 Moreover, there is fairly persuasive evidence that Kant fully realized after 1785 that he could get no moral mileage out of his transcendental philosophy, even though he continued to use such non-phenomenal terms as Spontaneitiit in the second edition of the Critique. However, establishing that, as a matter of fact, Kant did not argue against the possible phenomenality of the transcendental subject because he thought he thereby would establish some sort of noumenal agency, does not of itself address Kitcher's central and quite appropriate question: why did Kant not identify the transcendental and phenomenal subject, while admitting the difference between transcendental and empirical claims about this subject?

17 Since apperception is a component of consciousness, or, as Kant says, the ‘vehicle’ of concepts, and not itself a mental act accompanying consciousness (or the mental act that is consciousness), Kant is not committed to various claims about the self-conscious transparency of mental life, or its incorrigibility, etc. He is not, therefore, subject to the kind of complaints offered against his theory by Hans Georg Hoppe in Synthesis bei Kant (Berlin: de Gruyter 1983), 113-75. Indeed, the claim for the necessity of such an implicit self-construal in cognitive consciousness opens wide the door to realizing the profoundly mediated (and so corrigible) character of conscious activity, a fact which becomes the methodological fulcrum of Hegel's Phenomenology of Spirit.

18 Strawson, P.F. The Bounds of Sense (London: Methuen 1966)Google Scholar

19 Henrich, Dieter Identitiit und Objektivitiit (Heidelberg: Carl Winter UniversitatsVerlag 1976)Google Scholar. Pothast, Ulrich in iiber einige Fragen der Selbstbeziehung (Frankfurt a.M.: Klostermann 1971Google Scholar) has tried to argue that Kant was simply ambiguous about whether apperception was an explicit self-perception or the implicit form of all thinking. Cf. p. 13. I am claiming that Kant rather clearly intended the latter option, and do not see how the passages Pothast cites as evidence of the former view (A342 and A343) support his reading, since they are clearly preparing the way for what rational psychologists believe.

20 Cf. Anscombe, G.E.M. ‘The First Person,’ in Guttenplan, S.D. ed., Mind and Language (Oxford: Oxford University Press 1975), 45-65.Google Scholar For an excellent counter, see Shoemaker, S. ‘Self-reference and Self-awareness,’ journal of Philosophy 65 (1968), 555-68.CrossRefGoogle Scholar

21 Although in such an account apperception is a necessary condition for conscious (cognitive) perception, one should not conclude that the position thereby limits all sentient interaction to such conscious perception. There can indeed be all sorts of ‘non-apperceptive’ ways in which information of a sort about the external world is received and processed. But, even, if the perception of such information might affect behavior, the position holds that it would not be conscious perception. For an account of non-apperceptive ‘receiving and processing,’ see Patricia Smith Churchland, ‘Consciousness: The Transmutation of a Concept,’ Pacific Philosophical Quarterly 64 (1983), 80-95.

22 AA, vol. XVIII, pp. 318-19. Cf. also the helpful discussion of the passage in Allison, Henry Kant's Transcendental Idealism (New Haven: Yale University Press 1984), 276-8.Google Scholar

23 In his recent Kant iiber Freiheit als Autonomie (Frankfurt: Klostermann 1983), Gerold Prauss has attempted a new, reconstructive interpretation of the theory of activity that Kant, it is claimed, late in his career realized was at the foundation of and unified his accounts of ‘knowing’ (Erkennen) and ‘acting’ (Handeln). For Prauss, that theory (sometimes called a theory of spontaneity) involves a complicated analysis of the intentional, purposive (absichtlich) character of knowledge claims, their possible ‘succeeding’ or ‘failing,’ and the derived status of practical actions which ‘depend’ on the results of such intendings. My interpretation differs from his in presenting the problem of apperception as more fundamental to the issue of spontaneity, and so freedom, than intention, and in arguing that the chief reason Kant was hesitant and often confused about developing such a unified theory explicitly was his realization that it would require massive alterations in his transcendental program itself, and not because of a pre-critical allegiance to the ‘purity’ of theory (cf. p. 110). In general, Prauss, it seems to me, takes off from Kant's claims on the ‘freedom’ or spontaneity of thinking, and omits discussing whether Kant was entitled to such metaphysical sounding claims, and if so, how they would square with transcendental idealism.

24 Kant's discussion in the Critique should be supplemented by what he says in his Anthropologie, AA, vo!. VII, section #7 and #24. See also the account by Allison, 255-71, the extensive discussion by Karl Ameriks, Kant's Theory of Mind (Oxford: Clarendon Press 1982), 241-55, and my Kant's Theory of Form, 172-82.

25 Cf., for example, Kant's remarks about complete causal determinism for all appearances, B573=A545, and his references to the present and future, hoped for status of empirical psychology as a science, B874=A846- B877=A849. His position on this issue is not without its ambiguities, however. See Michael Washburn, ‘Did Kant Have a Theory of Selfknowledge?’ Archiv fUr die Geschichte der Philosophie 58 (1976), 40-56.

26 Wilfrid Sellars, ‘ … this I or he or it (the thing) which thinks …’ Proceedings of the American Philosophical Association 14 (Sept. 1971), 20

27 Sellars, 14-15

28 This is the summary version of her thesis presented in ‘Kant's Rea1 Self,’ 113. For a full exposition of her interpretation, see ‘Kant on Self-Identity,’ Philosophical Review 91 (January 1982), 41-72.

29 To be sure, this issue of Kant's positive philosophy of subjectivity was widely influential in the German tradition, even beyond Kant's idealistic successors. Its legacy is apparent in any number of contexts. Cf., for example, this typical claim by Alfred Baeumler: ‘Nach der Kritik der reinen Vernunft gibt es keinen mundus intelligibilis mehr. Aber es gibt eine rein intellektuelle Synthesis. Der Substanzenbegriff des noumenon hat sich in einer Funktionsbegriff verwandelt. Im Begriff der reinen intellektuellen Synthesis, dem Urbild der Kategorien, erkennen wir die letzte Sublimierung der Verstandeswelt im Gegensatz zur Sinnenwelt.’ Kants Kritik der Urteilskraft. lhre Geschichte und Systematic (Band I: Das Irrationalitiitsproblem in der Aesthetik und Logik des 18. Jahrhunderts bis zur Kritik der Urteilskraft) (Halle 1923), 346. I think the basic point expressed here is correct, but that Baumler (and Heidemann) greatly confuse their own case by adopting a ‘noumenalist’ language to refer to the subject's activity, and so they perpetuate the notion that post-Kantian idealism requires a pre-critical metaphysics.

30 Reflexion 6001, AA, vol. XVIII, 420-1

31 Strawson, 248 ff.

32 For a fuller discussion of this distinction, see Allison, Chapter 13, especially 292-3. My differences with Allison on this issue turn on an ambiguity in his overall account of apperception: his willingness on the one hand, to concede that apperception provides us with something that ‘cannot be accounted for in terms of the mechanism of nature or the empirical character of the subject’ (323), and, like Kant, his continuing insistence on strict metaphysical neutrality (323-4). For a fuller discussion, see my review of his book, forthcoming in Kant-Studien.

33 Cf. AA, vol. 5, p. 31; Critique of Practical Reason, trans!. L.W. Beck (Indianapolis: Babbs Merrill 1956), 31.

34 Sellars, ‘Metaphysics and the Concept of a Person,’ 252

35 Sellars, 252

36 Sellars, 248, my emphasis

37 Heidemann is a typical example of someone who connects, somewhat quickly and carelessly, I think, the issues of ‘Spontaneitiit and ‘noumenale Existenz,’ 212 and 214, for example.

38 The best summary discussion of these passages can be found in Ameriks, 193-203.

39 This is relevant to a point that Prauss often makes: that Kant ought to have altered his formulations about impulses, needs and desires simply determining behavior causally in heteronomous action, because his own theory commits him to the view that, for an impulse, to affect my behavior, I must freely ‘allow it to,’ and that what the desire is for can be understood only if the free self-determination of purposeful conduct is introduced (cf. 224-5).

40 Ameriks, 195; Allison, 310-29

41 See the remarks at the end of the Critique of Judgment on the ‘causality of man’ (Kausalitiit der Menschen), AA, Bd. 5, 484; CJ, 337.

42 Prauss vigorously denies this, and argues that the post-Kantian idealist tradition profoundly misses some of Kant's deepest insights (cf. 308-26). I think he is wrong that that tradition ignored the purposive (and fallible, finite) character of human reason. Cf. Hegel's remark in the Jena Phenomenology that reason is ‘purposive activity,’ Gesammelte Werke Bd. 9 (Hamburg: Felix Meiner 1980), 20. And, with more space, I would disagree strongly with the way he brushes aside Hegel's insistence on the inherently ‘negative,’ self-correcting and self-transforming nature of thought. Such remarks seem to me to illuminate precisely what Prauss finds hidden in Kant.