¹ Translations from the Kritik der reinen Vernunft are taken from the Critique of Pure Reason, trans, and ed. Guyer, P. and Wood, A. (Cambridge: Cambridge University Press, 1998)Google Scholar, though in some cases I have offered amendments; translations from the Prolegomena zu einer jeden künftigen Metaphysik are taken from Prolegomena to Any Future Metaphysics, trans, and ed. Hatfield, G. (Cambridge: Cambridge University Press, 1997)Google Scholar, and translations from the Opus postumum are taken from Opus postumum, trans. Förster, E. and Rosen, M., ed. Förster, E. (Cambridge: Cambridge University Press, 1998)Google Scholar. All other translations from the German are my own. Citations from the Critique are given in the standard A/B format while all other citations of Kant's works refer to volume and page number in the Gesammelte Schriften, Königliche Preussische Akadeinie der Wissenschaften (Berlin and Leipzig: de Gruyter, 1922)Google Scholar.

² Of course, self-affection is mentioned in a number of Reflexionen, and is briefly discussed in a couple of published works, including the Anthropologie (7: 140), the Kritik der Urtheilskraft (albeit only in the discarded first introduction at 20: 223), and the late essay ‘Welches sind die wirklichen Fortschritte, die die Metaphysik seit Leibnizens und Wolff's Zeiten in Deutschland gemacht haben?’ (20: 270).

³ If the taxonomy in the so-called Stufenleiter passage differs from that in other passages it is most likely because in the former Kant closely follows Baumgarten's presentation in §522 of the Metaphysica (reprinted in vol. XVI of Kant's Gesammelte Schriften): ‘Repraesento mihi quaedam ita, ut aliqui eorum characteres clari sint, aliqui obscuri. Eiusmodi perceptio, qua notas claras, distincta est, qua obscuras, sensitiva …’.

⁴ See also R 5661 (1788-90): ‘An empirical representation, of which I am conscious, is perception’ (18: 318).

⁵ See also Prolegomena §20: ‘At bottom lies the intuition of which I am conscious, i.e., perception (perceptio) which belongs solely to the senses’ (4: 300).

⁶ See also Kant's note in his edition of the Critique: ‘Intuition is related to the object, sensation merely to the subject’ (23: 21).

⁷ See Brook, , Kant and the Mind (Cambridge: Cambridge University Press, 1994), chapter 3, passimCrossRefGoogle Scholar.

⁸ Recall Hamann's report that Tetens's book lay open on Kant's, desk as he wrote the Critique (in Cassirer, Kant's Life and Thought (New Haven: Yale University Press, 1981), p. 194n).Google Scholar

⁹ Locke, John, An Essay Concerning Human Understanding, ed. Nidditch, P. H., (Oxford: Oxford University Press, 1979), II.ix.4 (144, second emphasis mine).Google Scholar

¹⁰ Tetens, J. N., Philosophische Versuche über die menschliche Natur und ihre Entwicklung, vol. 1, (Leipzig: Weidmanns, Erben, und Reich, 1777), p. 262.Google Scholar

¹¹ , Tetens, Philosophische Versuche, p. 289.Google Scholar

¹² , Tetens, Philosophische Versuche, p. 281.Google Scholar

¹³ , Tetens, Philosophische Versuche, p. 284.Google Scholar

¹⁴ , Tetens, Philosophische Versuche, p. 290.Google Scholar

¹⁵ In fact, Tetens endorses this Leibnizian doctrine in the third essay, although only with a revealing amendment. Since to endorse the notion of a perception of which we are not conscious would be in conflict with his claim, presented above, that all perception involves consciousness, Tetens instead refers to ‘unperceived representations’ (‘unwahrgenomme Vorstellungen’) (, Tetens, Philosophische Versuche, p. 265).Google Scholar

¹⁶ Leibniz, G. W., New Essays on Human Understanding, ed. and trans. Remnant, P. and Bennett, J. (Cambridge, Cambridge University Press, 1981), II.ix.4 (134)Google Scholar.

¹⁷ See also the Preface of the New Essays, p.54.

¹⁸ cf. A Treatise of Human Nature, ed. Selby-Bigge, L. A., 2nd edn revised by Nidditch, P. H. (Oxford: Oxford University Press, 1978), p. 251 (I.vi ‘Of Personal Identity’).Google Scholar

¹⁹ Treatise, p. 252.

²⁰ With regard to this dispute, the reader is referred to Dieter Henrich's now classic ‘The proof-structure of Kant's Transcendental Deduction’ (in Review of Metaphysics 22 (1969), 640–59)Google Scholar, as well as Allison's, Henry, Kant's Transcendental Idealism (New Haven: Yale University Press, 1983), ch. 7Google Scholar, and Evans, J. C. “Two-Steps-in-One-Proof': the structure of the Deduction of the Categories' (in Journal of the History of Philosophy 28 (1990), 553–70)CrossRefGoogle Scholar.

²¹ I should also note that in the following I am heavily indebted to Henry Allison's analysis of the Deduction. While I differ from him on a number of points, which will be noted in due course, my general approach and arrangement of the argument in the second half of the deduction parallel his.

²² Indeed, there is an important justification for Kant's focus on space in the footnote at B160-1 based upon the asymmetry between space and time as pure intuitions. This will be discussed in detail below.

²³ See, for instance, B207: ‘Appearances … are not pure (merely formal) intuitions, like space and time (for these cannot be perceived in themselves)’. Kant's claim in the A edition that 'this synthesis of apprehension must also be a priori, i.e., in regard to representations that are not empirical [for] without it we could have a priori neither the representations of space nor time’ (A99) is not inconsistent with the emphasis in the B edition that ‘time is not perceived’ (B219, B225, B233) since, in the A discussion, Kant is referring to a pure synthesis of apprehension which is closely linked in the next section with the transcendental synthesis of the imagination (‘The synthesis of apprehension is therefore inseparably combined with the synthesis of reproduction’, A102). This is a claim Kant will also make in the B Deduction (B162n).

²⁴ I also take this dependency of apprehended objects upon the representation of space and time to be the thrust of the following confusing remark in the Amphiboly: ‘But since sensible intuition is an entirely peculiar subjective condition, which grounds all perception a priori, and the form of which is original, thus the form is given for itself alone, and so far is it from being the case that the matter (or the things … which appear) ought to be the ground (as one would have to judge according to mere concepts), that their possibility presupposes a formal intuition (of space and time) as given’ (A268/B323-4). On this point, also see , Allison, Kant's Transcendental Idealism, pp. 168–9.Google Scholar

²⁵ On the determination of space as a unity, see also the discussion in Opus postumutn: ‘empirical intuition … represents space itself through the composition of the manifold in appearance a priori into an object of experience as of a synthetic cognition of the object of the senses’. That space is directly represented by the combination of the manifold of appearance (without, as will be seen with time, requiring the additional generation of that manifold) is due to the fact that the ‘the pure intuition of the manifold in space contains the form of the objects in appearance a priori of the first order, that is, direct’ (22: 367, my emphasis).

²⁶ See also R 6349 (1797): ‘The difficulty [with regard to inner experience] concerns only how the subject could itself institute experience. It must not merely perceive sensations in itself, but rather excite [erregen] and connect them synthetically, consequently affecting itself.’ (18:673-4)

²⁷ A reviewer for Kantian Review has charged that I force the view upon Kant that inner sense has no manifold of its own (that this ascription is a commonplace is, of course, no counter-argument). The reviewer disputes this attribution by quoting A99 where Kant claims that ‘every intuition contains a manifold in itself without limiting the claim to empirical (as opposed to pure) intuitions. I would point out, however, that Kant does not make a comparably strong claim in the B-edition Deduction, asserting only at B160 (already quoted above) that space and time are also represented as ‘intuitions … (which contain a manifold)’, without commenting (at that point) on how this manifold is supplied. (Indeed, the differences between the A and B Deductions are especially material on this point, given that the doctrine of self-affection is formally introduced only in the latter.) The reviewer continues, arguing that even if spatial analogies are required i n order to represent time as an object, this does not imply that the manifold of time is itself borrowed from outer sense since this may be a unique requirement of the representation of time. Yet, this line of argument strikes me as faulty - why would an analogy be required for representation when a manifold is readily available for synthesis? In addition to the references provided above, the clearest statement of Kant's view that inner sense has no manifold of its own seems to be B67: ‘the representations of the outer senses constitute the proper material with which we occupy our mind’ (as concerns the secondary literature, one might consult Beatrice Longuenesse's Kant and the Capacity to Judge (Princeton: Princeton University Press, 1998), pp. 228–9Google Scholar, and , Allison, Kant's Transcendental Idealism, pp. 258–63, among others)Google Scholar.

²⁸ Of course, this is not to make the additional, and false, claim that in every act of perception the perceiver is required actually to draw a line. Rather, the example is intended only to illustrate the dependence of inner sense upon outer sense for the determination of the time in which a given percept is placed, which dependence is later expanded i n the first Analogy and the Refutation of Idealism. Indeed, at B155, Kant suggests that attending to motion would also yield the appropriate representation of time (a claim which plays upon the close connection between time and the concept of motion in the Transcendental Exposition of time at B48-9). Thus, the example of a line is intended only as one possible example of time's general limitation to representation through space: ‘we can only represent time to ourselves in that we affect ourselves by describing space and grasping the manifold of its representation’ (, LeningradReflexion, International Philosophical Quarterly 29 (1989), 252–261, cf. 253)Google Scholar.

²⁹ Thus, again in the Opus postumum, Kant contrasts the ‘direct’ appearance which is given through outer sense (see note 25, above) with the ‘indirect’ appearance generated through self-affection: ‘The composition of the perception of appearance in the subject for the purpose of experience is in turn appearance of the so-affected subject as it represents itself, therefore, indirectly and it is appearance of the second order … that is, appearance of the self-affecting subject’ (22: 367).

³⁰ In addition to Tetens's use, presented above, of the notion of attention, see , Wolff, Psychologia empirica (reprint Hildesheim: Georg Olms, 1968), §237Google Scholar; , Baumgarten, Metaphysica, §529Google Scholar. Baumgarten's student, Meier, G. F., even identifies attention as one of two ‘Hauptvermogen’ in his Metaphysik (2nd edn, Halle, 1765) §506Google Scholar.

³¹ As, for instance, in Wolff's claim in the Psychologia empirica that ‘Videmus adeo attentionis directionem pendere ab arbitrio nostro …’ (§256). Compare Kant's similar claim in the Anthropologie (7:131).

³² Along these lines we might note an ambiguity in Kant's statement at B155: ‘Motion, as action of the subject (not as determination of the object), consequently the synthesis of the manifold in space, if we abstract from this manifold in space and attend solely to the action in accordance with which we determine the form of inner sense [wenn wir von diesem abstrahiren und bloss auf die Handlung Acht haben, dadurch wir den inneren Sinn seiner Form gemass bestimmen], first produces the concept of succession at all’ (my emphasis). The ambiguity concerns the nature of the second clause and whether it further determines the nature of the action, that is, we attend to the action of determining inner sense (this is suggested by the Guyer-Wood translation), or whether it introduces a consequence of the attention to the action taken to be the synthesis of the manifold in space, that is, ‘we attend to the action [of synthesis in drawing the line, etc.], through which [attention] we determine inner sense in accordance with its form’.

³³ See also the late essay, ‘Welches sind die wirklichen Fortschritte’ where Kant notes that ‘we affect inner sense by means of attention [wir … den innern Sinn …vermittelst der Aufmerksamkeit afficiren]’ (20: 270).

³⁴ Translation altered. See note 32 above.

³⁵ Thus A142/B182: ‘I generate time itself in the apprehension of the intuition.’

³⁶ Kant asserts this parallelism at B156: ‘hence if we admit about [objects in outer sense] that we cognize objects by their means only insofar as we are externally affected, then we must also concede that through inner sense we intuit ourselves only as we are internally affected by our selves’.

³⁷ , Allison, Kant's Transcendental Idealism, pp. 266–7.Google Scholar

³⁸ , Allison, Kant's Transcendental Idealism, p. 267.Google Scholar

³⁹ Kant probably foregoes introducing the notion of attention explicitly here since the discussion of the synthesis of the imagination, with which attention is linked, would not properly belong in the Aesthetic.

⁴⁰ On the passivity of inner sense in self-affection, see also B69, B156, B429-30 and R 6354 (1797,18: 680).

⁴¹ See Kant's similar presentation of this argumen t in R 6354 (1797): ‘Inner sense is not yet cognition of my self, rather we must first have appearances through it, immediately after which we make a concept of ourselves through reflection on these appearances, which thereupon has as a consequence the empirical cognition of my self, that is, inner experience’ (18: 680).

⁴² Such is Guenter Zoeller's attempt to reconcile the contrary poles in Allison's accusation of a deep incoherence, in ‘Making sense out of inner sense: the Kantian doctrine as illuminated by the Leningrad Reflexion’ (in International Philosophical Quarterly, 29 (1989), 263–70CrossRefGoogle Scholar, cf. esp. 267-8). Manfred Baum is apparently of a similar view, claiming that another ‘kind of self-affection is necessary for becoming conscious of the representations of my mind’ ( Baum, Manfred, ‘Cosmological Apperception’, International Philosophical Quarterly 29 (1989), 281–9, cf. 283, my emphasis)CrossRefGoogle Scholar. According to my interpretation, however, there is one type of self-affection distinguished only by what occasions it; thus, it is occasioned either by the attempt to perceive a given external object, or by the attention to a mental state with the aim of cognizing the mind and its contents. Since both cases involve apprehension, both cases must also require, as shown above, the synthetic activity of the understanding upon the inner sense in order to generate and determine the manifold and, therefore, both require self-affection understood in precisely the same way.

⁴³ On this point see also , Allison, Kant's Transcendental Idealism, pp. 160–2.Google Scholar

⁴⁴ See especially Kitcher, Patricia, Kant's Transcendental Psychology (Oxford: Oxford University Press, 1990), pp. 91–116.Google Scholar