1 Ameriks, Karl, “Recent Work on Kant's Theoretical Philosophy”, American Philosophical Quarterly 19 (1982), 1–24Google Scholar.

2 Ameriks borrows this expression from Walker, Ralph, Kant (London: Routledge and Kegan Paul 1978),Google Scholar viii, who uses it to characterize transcendental idealism and transcendental arguments.

3 Ameriks, “Recent Work”, 4.

4 The most recent and comprehensive account of this position (at least in English) is to be found in Henry Allison's book Kant's Transcendental Idealism (New Haven, CT: Yale University Press, 1983).Google Scholar I certainly do not want to take issue with Allison here, but I will note that, in spite of the impressive acumen and scholarship he displays in his book, some critics still have a hard time understanding exactly what his position is. See, for example, Ameriks' review of the book, forthcoming in Topoi.

5 To my knowledge, the only authors who have applied possible-world semantics to related problems in Kant are Robert Howell and Carl Posy. See, for example, Howell's, Robert “Intuition, Synthesis, and Individuation in the Critique of Pure Reason”, Nous 1 (1973), 207–232;CrossRefGoogle Scholar“A Problem for Kant”, in Saarinen, E., Hilpinen, R., Niini-luotto, I. and Hintikka, M. Provence, eds., Essays in Honour of Jaakko Hintikka (Dordrecht: Reidel, 1979), 331–349; andCrossRefGoogle Scholar“Kant's First-Critique Theory of the Transcendental Object”, Dialectica 35 (1981), 85–125.CrossRefGoogle Scholar See Posy's, Carl “The Language of Appearances and Things in Themselves”, Synthese 47 (1981), 313–352;CrossRefGoogle Scholar and “Dancing to the Antinomy: A Proposal for Transcendental Idealism”, American Philosophical Quarterly 20 (1983), 81–94.Google Scholar However, their concerns and intuitions seem to be vastly different from mine. Howell is primarily interested in using possible-world semantics to understand Kant's “appearing-talk”, not his “appearance-talk”. So he does not apply the semantics to the relation between appearances and things in themselves. Furthermore, he thinks that the world in itself is the actual world, which ultimately accounts for his finding a “problem” in Kant. As for Posy, he is interested in using a possible-world analysis (of the intuitionistic variety) to provide “a linguistic version of transcendental idealism”, which turns it into “an evidential theory of meaning for empirical judgments” (“Dancing to the Antinomy”, 83). Once again, there is no direct question here of the identity (or distinctness) of things in themselves and appearances.

6 More precisely, I conjecture that such explicata are provided by the notions of “an individual as it appears” and “an individual as it is” defined on 11.

7 Strawson, Peter, Individuals (London: Methuen, 1959), 11CrossRefGoogle Scholar.

8 Of course, there is a general sense in which representations are things. But it is also common to contrast representations with (the) things represented, and if one has this contrast in mind (as Kant at times seems to, especially in passages such as the third one below), then it is natural to think that things in themselves and representations should fall in opposite camps.

9 Ermanno Bencivenga, “Knowledge as a Relation and Knowledge as an Experience in the Critique of Pure Reason”, forthcoming in the Canadian Journal of Philosophy.

10 In the end, I claim in “Knowledge as a Relation”, the preliminary statement that follows will turn out to be misleading, since the very notion of what it is to be real (as opposed to “merely” intentional) has been redefined. But such an ultimately misleading statement is very useful (and probably, some such misleading statement is necessary) as an auxiliary means for the formulation of the conceptual shift, in that it allows us to “bracket” our ontological assumptions while we proceed to reconceptualize what it is to have an ontology.

11 In this paper, I simply identify noumena, things in themselves and transcendental objects, by taking seriously passages like the following:

The concept of a noumenon—that is, of a thing which is not to be thought as object of the senses but as a thing in itself, solely through a pure understanding—is not in any way contradictory (A254 B310).

Now whence, I ask, can the understanding obtain these synthetic propositions, when the concepts are to be applied, not in their relation to possible experience, but t o things in themselves (noumena)? (A259 B315).

What matter may be as a thing in itself (transcendental object) is completely unknown to us … (A366).

If, on the other hand, the object is transcendental, and therefore itself unknown; if, for instance, the question be whether that something, the appearance of which (in ourselves) is thought (soul), is in itself a simple being … (A478 B506).

Of course, I know that there are authors who want to distinguish between the three concepts, and I think they have good reasons for doing so. But to my knowledge, they usually conclude that there is a difference between the intensions of the concepts, not between their extensions (which is all I am concerned with here). Henry Allison, for example, says in “Things in Themselves, Noumena, and the Transcendental Object”, Dialectica 32 (1978), 41–76,CrossRefGoogle Scholar that “[transcendental] reflection requires the distinction between two ways of considering an object (as it appears and as it is in itself) and … “noumenon” and “transcendental object” characterize alternative descriptions of an object considered as it is in itself (41).

12 As Kant specifies later (for example, at A255 B311), this distinction is to be taken only in the negative sense, that is, as something we can think of but not experience in the full and “official” sense of the Critique (which requires some intuitional fulfillment). But this possibility of thinking the distinction is all I need here.

13 I want to emphasize again that “taking these passages seriously” means “taking seriously the suggestion they offer as to how to translate Kant's language into our own”. It does not mean that we should take Kant to be committed to the full framework of possible-world semantics, or to the specific application of that framework I will propose here.

14 For example, after one of the passages referred to on page 423, Kant goes on to say Whatever pure mathematics establishes in regard to the synthesis of the form of apprehension is also necessarily valid of the objects apprehended. All objections are only the chicanery of a falsely instructed reason, which, erroneously professing t o isolate the objects of the senses from the formal condition of our sensibility, represents them, in spite of the fact that they are mere appearances, as objects in themselves, given to the understanding (A166 B206-B207).

That is, from the fact that mathematics would not be necessarily valid of things in themselves, it does not follow that it is not necessarily valid of appearances.

15 For a good treatment of the notions of possible-world slices and cross-world con-tinuants in possible-world semantics, see Salmon's, NathanReference and Essence (Princeton, NJ: Princeton University Press, 1981), especially 108–111Google Scholar.

16 This term was introduced by Kurt Lewin.

17 Note that this analysis is not unrelated to the previous one in terms of numerical and qualitative identity. For we can either accept a notion of (cross-world) numerical identity which does not require qualitative (cross-world) identity (and hence does not obey the Law of Indiscernibility of Identicals [in this sense]), or “shatter” a cross-world continuant into its possible-world slices and claim that they are not “really” identical with one another but only genidentical: in this second case, we can recover the universal validity of the Law of Indiscernibility of Identicals. Part of my claim here is that the logic of Kant's talk oscillates between the two approaches, and hence is best understood in terms of a combination of them.

Also, the above are not the only options. An alternative way of recovering the universal validity of the Law of Indiscernibility of Identicals is by “shattering” properties into their possible-world slices, and claiming that, say, the (cross-world con-tinuant) individual a does not have the property of redness simpliciter, but rather the property of redness-at-w. This alternative would probably be favoured by those commentators who stress the primary of Kant's “appearing-talk” over his “appearance-talk.”

18 It would be possible to argue, on similar grounds (that is, on the ground of Kant's characterization of possibility at A218 B265), that things in themselves are not even possible. However, the possibility characterized at A218 B265 is real possibility, that is, the possibility of being experienced, and all I am claiming here is that things in themselves are logically possible, that is, that their concept is not contradictory and hence they can be thought. For Kant's distinction between real and logical possibility see for example A244 B302-B303 footnote, A596 B624 footnote.

19 It will be apparent that my intuitions go in the same general direction as Hans Vaihing-er's (see his The Philosophy of “As If”, [London: Routledge and Kegan Paul, 1924]).Google Scholar For a more recent proposal along the same lines, see Schaper's, Eva “The Kantian Thing-in-itselfasa Philosophical Fiction”, Philosophical Quarterly 16 (1966), 233–243CrossRefGoogle Scholar.

20 Melnick, Arthur, Kant's Analogies of Experience (Chicago, IL: University of Chicago Press, 1973)Google Scholar.

21 Brittan, Gordon, Kant's Theory of Science (Princeton, NJ: Princeton University Press, 1978)Google Scholar.

22 Brittan himself does not flesh it out in quite the terms I use below, largely because he is not interested in Kant's general conception of experience but only in Kant's conception of science. So he talks about “a deductive relationship [that] obtains between the propositions of physics [emphasis mine] and sense-experience sentences” (Brittan, Kant's Theory of Science, 6). But I think my way of fleshing out his claim is a fairly natural generalization of his own.

23 As Brittan reminds us, different (and more sophisticated) reductions of the same kind were attempted by later empiricists, for example, by Carnap, Rudolf in The Logical Structure of the World (Berkeley, CA: University of California Press, 1967)Google Scholar.

24 Of course, Kant is just right on this. To put the empiricists in a better light one would have to emphasize the largely prescriptive nature of their programme. Strawson does that when he characterizes Berkeley (and to some extent, Hume) as revisionary metaphysicians (in Individuals, 9).

25 Ameriks, “Recent Work”, 3.

26 An alternative way of saying this is the following: the addition of “für sich” to “wirklich” suggests that Kant has discovered a hidden parameter in “wirklich”, that is, that (in the passage in question) he is not just using “actual”, but rather “actual-at-x” (or “actual-from-the-point-of-view-of-X”).

27 See Salmon, Reference and Essence, 116.

28 Additional evidence that Kant does not intend his terminology to be taken too literally can be found in the following passage from the Critique of Judgment, trans, by Meredith, James Creed (Oxford: Clarendon Press, 1980), Part I, 223:Google Scholar

Thus the words ground (support, basis), to depend (to be held up from above), to flow from (instead of to follow), substance (as Locke puts it: the support of accidents), and numberless others, are not schematic, but rather symbolic hypotyposes, and express concepts without employing a direct intuition for the purpose, but only drawing upon an analogy with one, i.e., transferring the reflection upon an object of intuition to quite a new concept, and one with which perhaps no intuition could ever directly correspond.

Note also that I would use a similar strategy to handle the other things Kant says on the relation between appearances and things in themselves. To give only one other example, when he says “We may, however, entitle the purely intelligible cause of appearances in general the transcendental object …” (A494 B522), I would interpret the notion of a “purely intelligible cause” in terms of the unschematized category of causality, that is, as reducing to the notion of a ground underlying (in the sense explained above) the appearances.

29 To be precise, it is genidentity if we think of possible-world slices, and identity proper if we think of cross-world continuants.

30 That is, the position of the so-called contingent identity theorists. Of course, this position has been vigorously challenged by Saul Kripke, but the debate is still open.