The Evolution of Modern Metaphysics: Making Sense of Things (35 page)

14
This is intended to allow for the possibility that the knowledge involves concepts ‘that can be drawn only from experience’ (B3).

15
Cf.
Prolegomena
, 4:266.

16
Kant may be presupposing a Leibnizian view whereby every judgment is of subject–predicate form (see
Ch. 3
, n. 32). Or he may be presupposing that the extension of his definition to judgments of other forms, like its extension to negative judgments, ‘is easy’ (A6/B10). If the former, then either this is a lapse on his part or some account has to be given of how it consists with what he says at A73–74/B98–99 and B140–141. For an interesting discussion of these matters, see Ian Proops (
2005
), pp. 592–596. (Proops there argues for a third possibility, namely that Kant does not intend his dichotomy to extend to all judgments. I remain unconvinced.)

17
Whether they coincide or not depends on, among other things, whether Leibniz also allowed for truths of reasoning that can be shown, by a finite process of analysis, not to be deniable without violating the principle of sufficient reason: see
Ch. 3
, n. 33. If he did, they do not.

18
For helpful discussions of Kant’s two dichotomies, see Bennett (
1966
), §§2–4, and Gardner (
1999
), pp. 52–55; and for a helpful discussion of the second, see Allison (
1983
), pp. 73–78. For an excellent discussion of the two dichotomies with particular reference to mathematics, see Potter (
2000
),
Ch. 1
.

19
Quine is the best known example: see
Ch. 12
, esp. §4.

20
Frege is the best known example: see
Ch. 8
, esp. §§2 and 3.

21
See e.g. Strawson (
1966
), Pt V, countered by Hopkins (
1982
).

22
This explicitly contradicts Kant’s example at A157/B196. But it also implicitly contradicts the geometrical example that he has already given, because it means that none of the relevant straight lines is ‘the’ shortest.

23
See Einstein (
1920
), §I and Appendix V. Note: my tone in this paragraph may be a little more cavalier than is strictly warranted. For a different, somewhat fuller account of how Kant is superseded by twentieth-century science, see
Ch. 11
, §3. For a reversion to something more cavalier, see
Ch. 12
, §4.

24
Cf. Carr (
1999
), p. 62. This goes some way towards countering an objection that Edward Craig voices to the answer that Kant will eventually give to his own question: see Craig (
1987
), pp. 237–239.

25
Throughout this chapter I use ‘answer to’ elliptically for ‘answer correctly to.’ See also n. 28 for a gloss on how independence is to be understood here.

26
Cf. Bernard Williams’ famous formula that ‘knowledge is of what is there
anyway
’ (Williams (
1978
), p. 64, emphasis in original). This is Williams’ summary way of putting what he describes as ‘a very basic thought,’ namely ‘that if knowledge is what it claims to be, then it is knowledge of a reality which exists independently of that knowledge, and indeed (except in the special case where the reality known happens itself to be some psychological item) independently of any thought or experience’ (ibid.). (We might wonder whether incorrigible knowledge of one’s own sensory states is a counterexample – albeit a counterexample that would do nothing to assuage scepticism about the possibility of synthetic
a priori
knowledge. In fact, however, there is good reason not to count anything of that kind as ‘knowledge’: cf. Wittgenstein (
1967a
), Pt I, §246. Certainly nothing of that kind is included in what Kant calls ‘cognition’: see n. 13.)

27
‘Roughly’ because he actually defines experience as a kind cognition, not as a kind of knowledge: see n. 13.

28
By ‘
x
is independent of
k
’ I mean something that requires
x
to be independent not only of the actual formation of
k
but also (for instance) of the concepts that
k
involves. On this understanding, analytic knowledge does not answer to anything independent of it.

29
Cf. the analogies between Cartesian ‘intuition’ and sensory perception which I emphasized in
Ch. 1
, §4. And see
Ch. 1
, §3, for the rather anaemic sense in which Descartes himself took what was knowable
a priori
to be necessary. For further worries about the Cartesian view, see ‘Letter to Herz’, dated 21 February 1772, in
Correspondence
, 10:131.

30
Cf. Brandom (
2002a
), pp. 22–23, and McDowell (
2007b
), p. 399. Cf. also what Kant says in a famous passage from the letter to Marcus Herz cited in the previous note, referring to an early version of the first
Critique
for which he had made plans: ‘As I thought through the theoretical part, considering its whole scope and the reciprocal relations of all its parts, I noticed that I still lacked something essential, something that in my long metaphysical studies I, as well as others, had failed to consider and which in fact constitutes the key to the whole secret of metaphysics, hitherto still hidden from itself. I asked myself the question: What is the ground of the relation of that in us which we call “representation” to the object?’ (
Correspondence
, 10:129–130).

31
Kant takes it to be a fundamental mark of our finitude that our knowledge involves reception of this kind. An infinite being, Kant holds, could produce what it knew in knowing it (B145). For further discussion of this idea, see
Ch. 6
, §3, and
Ch. 21
, §§2(d) and 2(e). And cf. Heidegger (
1962b
), p. 31.

32
Kant further insists that it is the only possible account (B41; cf. A92/B124–125).

33
In the last sentence of this paragraph I am compressing one strain in the notoriously difficult ‘Transcendental Deduction’, at A84–130 and, differently in the second edition, B116–169. A very helpful and much more accessible account of this material occurs in
Prolegomena
, §§18ff. For a thorough and very interesting discussion of the issues raised in this paragraph, see Guyer (
1987
), esp. Pt II.

34
I shall return briefly to this example in §10.

35
Kant therefore distinguishes between the ‘transcendental’ and the ‘transcendent’ (see A296/B352–353 and
Prolegomena
, 4:373, n.). We could say, in what is admittedly something of a caricature, that whereas immanence belongs to what is inside our sensory bubble, i.e. to objects of our experience, and transcendence belongs to what is outside our sensory bubble, i.e. to things in themselves, transcendentality belongs to the bubble itself, or to its film. (Two caveats. First, there is much talk, mostly in the first edition of the first
Critique
, of ‘transcendental objects’, where what seems to be intended is something at the level of things in themselves: see e.g. A379–380. This is connected with complications in the system to which we shall return in §§8–10. Second, this caricature completely abstracts from the idea that the immanent is related to the transcendent as appearance to reality.)

36
There is an issue about Kant’s transcendental idealism which I should mention here, though I do not propose to address it. It concerns the contrast between the knowledge of appearances that we can have and the knowledge of things in themselves that we cannot have. That Kant accepts such a contrast is clear. The issue is about what exactly it comes to. One view would be that the two kinds of knowledge are distinguished by their subject matter. The fact that we can have the one and cannot have the other is, on that view, akin to the fact that we can know about events inside our light cone but cannot know about events outside it. (See Matthews (
1982
) and Allison (
1983
) for two of the many notable attempts to oppose this view, in apparent opposition to Strawson (
1966
), though we should beware that they may be as guilty of misrepresenting Strawson as they take him to be of misrepresenting Kant.) A quite different view is that the two kinds of knowledge are two kinds of knowledge about the same things. The fact that we can have the one and cannot have the other is then more like the fact that we can have historical knowledge about events of which we cannot have eyewitness knowledge. Both views cast knowledge of things in themselves as free of any ‘human’ perspective. But on the first view, unlike the second, this is dictated by the very subject matter of the knowledge. On the second view, the phrase ‘things in themselves’ should strictly speaking only ever be used syncategorematically, in tandem with some suitable epistemic expression: to say, for instance, that things in themselves are not coloured is an improper way of saying that our knowledge of the colours of things is not knowledge of things in themselves (is not free of ‘human’ perspective). With the possible exception of the caricature in the previous note, I have presented Kant’s transcendental idealism in such a way as to evoke the second view, but I certainly do not take myself to have refuted the first. For material conducive to the first view, see
Prolegomena
, 4:318, and 3rd
Critique
, 5:195. For material conducive to the second view, see Bxviii–xix, n.; A45–46/B62–63;
Prolegomena
, 4:289; and
Groundwork
, 4:450–452. For a third view, to which I cannot hope to do justice but which I should certainly mention, see Bird (
2006
), esp. Chs 1 and 30. For some remarks relevant to the role that transcendental idealism will play in our narrative, see the Appendix to this chapter.

37
Note that this use of the term ‘intuition’ is very different from Descartes’ (see
Ch. 1
, §4). For an excellent discussion of Kant’s use, see Hintikka (
1969
).

38
Just as empiricists have missed this distinction by effectively trying to make do with differences of degree among our intuitions, so too, Kant alleges, Leibniz missed it when he spoke of monads representing the world more or less distinctly (see
Ch. 3
, §3) and effectively tried to make do with differences of degree among our concepts (A44/B61–62 and A270–271/B326–327).

39
See
Ch. 3
, n. 15: there is a comparison to be drawn here with Leibniz. But Kant believes that Leibniz’ view is vitiated by the error to which I referred in the previous note (A275–276/B331–332).

40
That we have the
a priori
intuitions we have and that we have the
a priori
concepts we have are, for Kant, brute facts about us (see e.g. B145–146 and
Prolegomena
, 4: 350–351). To be sure, what Kant says at B148 might be interpreted as meaning that any being that is given objects in intuition must (can? will?) use the same
a priori
concepts as we do to think about those objects. But what Kant surely means is rather that, for any being that is given objects in intuition,
we
must (can? will?) use these concepts to think about those objects.

41
For discussion, see Bennett (
1966
), Chs 11 and 15; Strawson (
1966
), Pt Two,
Ch. 3
; Allison (
1983
),
Ch. 10
; and Guyer (
1987
),
Ch. 10
.

42
For further discussion, including discussion of other contrasts that Kant recognizes between metaphysics and mathematics, see Moore (
2010b
), esp. §2. For reservations about the idea that we can determine conditions that must obtain in order for us to enjoy experience of the kind we do, see
Ch. 21
, §2(e).

43
Or, as Kant says at one point, they seek the ‘
unconditioned
, which reason necessarily and with every right demands in things in themselves for everything that is conditioned’ (Bxx, emphasis in original).

44
At B7 Kant says that the ‘unavoidable problems of pure reason … are
God
,
freedom
and
immortality
,’ and adds that ‘the science whose final aim in all its preparations is directed properly only to the solution of these problems is called
metaphysics
’ (emphasis in original; cf. A798/B826).

45
There is an issue here about the Causal Principle which Kant takes himself to have proved. Why does that not yield a negative answer to the question about free will, at least given something else that Kant holds, namely that any attempt to reconcile the Causal Principle with our possession of free will by maintaining that our exercises of free will have a distinctive type of cause (any attempt of the kind that Hume made: see Hume (
1975a
), §VIII) is ‘a wretched subterfuge’ (2nd
Critique
, 5:96)? We shall return to this issue in the next section.