The Evolution of Modern Metaphysics: Making Sense of Things (33 page)

What, then, does Kant think that we can think about things in themselves? Plenty. (The postulates of pure practical reason are three examples.) This is part of Kant’s view, which he proclaims on numerous occasions, that we can think far more than we can know (e.g. Bxxvi n., B146, B166 n., and A771–772/B799–800).
⁶⁶
However, any thinking that we do about things in themselves must be of an extremely attenuated kind. It must involve us in exercising concepts without intuitions, and Kant famously declares that thoughts in which concepts are exercised without intuitions are ‘empty’ (A51/B75).
⁶⁷
Elsewhere he is more forthright. He says of our
a priori
concepts that their ‘extension … beyond our sensible intuition does not get us anywhere’ (B148, emphasis removed), that ‘our sensible and empirical intuition alone can provide them with sense and significance’ (B149, emphasis removed), and that they ‘are of none but an empirical use, and … have no
sense at all when they are not applied to objects of a possible experience, i.e. to the world of sense’ (A696/B724; cf. A679/B707).
⁶⁸

One very important consequence of all of this is that whether Kant thinks that we can
make sense of
things in themselves depends on how exactly ‘make sense of’ is interpreted. On most interpretations, and certainly on any remotely robust epistemic interpretation, he does not think that we can make sense of them. But on an interpretation weak enough to allow for ‘empty’ thoughts, or to allow for the regulative use of concepts discussed in the previous section, he does think that we can make sense of them – as indeed he had better if his system is not to be a complete sham. In what follows I shall adopt the simplifying assumption that there is a core interpretation of what it is to make sense of something whereby Kant does not think we can make sense of things in themselves, which I shall call the ‘thick’ interpretation, and that there is a core interpretation whereby he does think we can, which I shall call the ‘thin’ interpretation. (This immediately raises a question about ‘transcendental’ sense-making and how it should be classified.
⁶⁹
But let us not forget that any awkwardness attaching to this question may be an awkwardness, not for the simplifying assumption, but for Kant. We shall return to this issue in the next section.)

It is here that the metaphysical importance of Kant’s distinction between what we can know about things in themselves and what we can think about them really lies. To see what this importance is, we must first reflect on the fact that Kant’s project seems to involve drawing a limit to what we can make sense of. But that in turn can seem an incoherent enterprise. More specifically, it can seem self-stultifying. More specifically still, it can seem vulnerable to the following argument, which, because of its recurring significance to the rest of this enquiry, I shall give a name: I shall call it
the Limit Argument
.

First Premise: The Limit-Drawing Principle
: We cannot properly draw a limit to what we can make sense of unless we can make sense of the limit.

Second Premise: The Division Principle
: We cannot make sense of any limit unless we can make sense of what lies on both sides of it.

Conclusion
: We cannot properly draw a limit to what we can make sense of.
⁷⁰

Granted the thick/thin distinction, however, Kant can respond to the Limit Argument as follows. He can accede to the suggestion that his project is a matter of drawing a limit to what we can make sense of
under the thick interpretation
, but he can deny that, under that interpretation, both the premises are true. For instance, he can insist that the Limit-Drawing Principle holds only under the thin interpretation of what it is to make sense of a limit: this gives him license to draw the limit that he wishes to draw without being able to make sense of it under the thick interpretation. Nor is there any reason to suppose that the Limit-Drawing Principle then connects in some other problematical way with the Division Principle. For instance, there is no reason to suppose that making sense of a limit under the thin interpretation requires making sense, under the thick interpretation, of what lies on its ‘far’ side. The threat of self-stultification is, apparently, averted.
⁷¹

Kant himself has a wonderful analogy to illustrate his project. He likens what we can make sense of, under the thick interpretation, to a surface, which, like the surface of the earth, appears flat, so that, given our restricted acquaintance with it, we cannot know how far it extends, though we can know that it extends further than we have managed to travel: however, like the surface of the earth, it is in fact round, and once we have discovered this we can, even from our restricted acquaintance with it, determine both its extent and its limits (A758–762/B786–790). Here, of course, he relies on the important distinction between what we have in fact made sense of and what we can make sense of. He sometimes draws this distinction in terms of what he calls, in the original German, ‘
Schranken
’ and ‘
Grenzen
’ (translated in the Cambridge edition of his works respectively as ‘limits’ and ‘boundaries’ – though it is the latter that corresponds to what I have been calling ‘limits’).
⁷²
The territory covered by what we have in fact made sense of, which is capable of extending over time into what it currently excludes, is marked by
Schranken
; the territory covered by what we can make sense of, which is of an altogether different kind from what it excludes, is marked by
Grenzen
(A767/B795 and
Prolegomena
, §57).
⁷³

But here too some cracks are perhaps beginning to appear. For we can legitimately refer to the limits of a globe only because we have access to a dimension other than the surface’s own two. If we ourselves were two-dimensional beings on the surface, and had no access to any third dimension, then, while we might still acknowledge the surface’s curvature and indeed its finitude, we would have no reason to think of it as having any
limits
(
Grenzen
) at all.
⁷⁴
It is therefore a real question whether, in these glimpses of ours beyond the limit of our own thick sense-making – in this empty play of concepts of ours in which there is sense-making only of the very tenuous, thin kind – there is anything remotely like access to a third dimension of space. If not, then we may not have succeeded in making sense of this limit after all, not even under the thin interpretation, which calls into question whether there
is
any such limit, which in turn calls into question the very distinction between appearances and things in themselves.

9. Sense-Making That Is Neither Straightforwardly Thin nor Straightforwardly Thick

Since §4 we have been suspending our misgivings about the fundamental doctrines on which Kant’s transcendental idealism rests. Even so, there is plenty, as we have just seen, to give pause. In this section I shall rehearse what seem to me to be, in the context of our enquiry, the most serious concerns about where those doctrines have led us.
⁷⁵

I note first that this whole exercise, that is to say the exercise of characterizing synthetic
a priori
knowledge and investigating the possibility, scope, and limits of metaphysics in the light of that characterization, has itself been an exercise in metaphysics. That is, it has itself been a maximally general attempt to make sense of things.
⁷⁶
In Kant’s work we find metaphysics in the service not only of science, ethics, and theology, but also of metaphysics.

We can therefore ask of Kant a question which, in §5 of the previous chapter, looked as though it had the potential to embarrass Hume when asked of
him
: does his own work conform to the views advocated in it? Can Kant himself be seen as pursuing synthetic
a priori
knowledge about how things (must) appear, but not about how they are in themselves?
⁷⁷

There is a problem that threatens to arise here. It is a variation on the problem that we considered in the previous section. Both problems come together in the following question. Can Kant, when he draws a limit to our thick sense-making, do so from anywhere inside that limit, or must he do so from somewhere outside it?
⁷⁸
The reply that I ventured in the previous section, on Kant’s behalf, was that he must do so from somewhere outside it, but that he is exonerated by the fact that he may nevertheless do so from somewhere inside the limit of thin sense-making. (In effect, then, I was suggesting that transcendental sense-making is thin.) The concern about this reply was whether any exercise of thin sense-making can be equal to the task. That concern is now exacerbated by the thought that this task is itself a metaphysical task, whereas metaphysical sense-making, for Kant, must all be thick.
⁷⁹

We can approach the problem that threatens to arise here from a different angle by considering the very judgment that our metaphysical knowledge, like our mathematical knowledge, is synthetic and
a priori
. This must itself, presumably, count as an item of synthetic
a priori
knowledge. And yet, precisely in registering the non-analytic character of the knowledge in question, does it not also have some claim to being, at least to that extent, the very thing that an item of synthetic
a priori
knowledge supposedly cannot be, namely a judgment about things in themselves? For, arguably, there is nothing ‘
from the human standpoint
’ (A26/B42, emphasis added)
to preclude our arriving at our metaphysical knowledge, or our mathematical knowledge, by means of pure conceptual analysis.
From the human standpoint
the various
a priori
conditions of our experience cannot be other than they are. Hence, from the human standpoint, these conditions cannot make a substantial contribution to any of our knowledge. That is to say, they cannot make the kind of contribution that they would not have made if they had been suitably other than they are, the kind that prevents the knowledge in question from answering merely to the concepts involved in it. (Thus even if we need to appeal to intuition to determine that the straight line between two points is the shortest, it is a real question what work this appeal to intuition does that is not likewise done by an appeal to intuition to determine, say, that black is darker than grey, a truth that Kant would presumably count as analytic.) In acknowledging that there
is
a substantial contribution made by the
a priori
conditions of our experience to some of our knowledge, which is what we are doing when we register the non-analytic character of the knowledge, must we not therefore already have taken a step back from the human standpoint? – as indeed Kant all but concedes when he writes:

The proposition: ‘All things are next to one another in space,’ is valid under the limitation that these things be taken as objects of our sensible intuition. If I here add the condition to the concept and say ‘All things, as outer appearances, are next to one another in space,’ then this rule is valid universally and without limitation. (A27/B43)
^80
,
81

(And note that this concern is aggravated when the knowledge is metaphysical, rather than mathematical, by what we saw Kant argue in §5: that in metaphysics, unlike in mathematics, the appeal to intuition is an appeal to the mere possibility of our being given objects in intuition in the various ways we are, so that the knowledge is knowledge ‘from concepts’ (A713/B741).)

There is a similar awkwardness in Kant’s handling of what he sometimes calls ‘the sole fact of pure reason’, which I take to be the fact that we can put pure reason to practical use in accord with the demands of our own freedom, as indicated in §7 (see e.g. 2nd
Critique
, 5:6, 31, 42, 43, 55, and 104, and 3rd
Critique
, 5:468). It is unsurprising that, among all the facts that Kant recognizes – where by a fact here is meant a contingency – this is the one that he is prepared to describe as the sole fact of pure reason. For what this fact is, on Kant’s conception, is a fact about pure reason’s purest
exercise, which is our freely placing demands of pure rational agency on ourselves and freely submitting to those demands. In effect, then, what Kant is prepared to describe as the sole fact of pure reason, in the sense of the sole fact accessible to pure reason, is something that he would also count as the sole fact of pure reason in another sense, the fact that
there is
such a thing as pure reason, capable of being exercised in the purest way, without the aid of any other faculty.
⁸²
But this is a fact about how things are in themselves. It is more like the fact that we have spectacles than like any fact that can be ascertained by looking through those spectacles. There is therefore a certain tension for Kant in supposing it to be accessible to pure reason. In what way accessible? The tension is close to breaking point when Kant says that this fact ‘forces itself upon us of itself as a synthetic
a priori
proposition that is not based on any intuition, either pure [i.e.
a priori
] or empirical’ (2nd
Critique
, 5:31).
⁸³