The Evolution of Modern Metaphysics: Making Sense of Things (31 page)

Our spectacles, to repeat, involve both intuitions and concepts. It is because they involve intuitions that Kant is able to reject the Independence Assumption. For synthetic knowledge must answer to
something
. So given that,
qua
synthetic, it does not answer merely to the concepts involved in it, the only way in which it can fail to answer to what is independent of it is by involving something other than concepts.
A priori
intuitions play just this role. (This is related to the principle, on which Kant again and again insists in the first
Critique
, that synthetic knowledge is never possible
without intuitions (see e.g. B16, A62/B87, A155–156/B194–195, B288–289, and A238–240/B297–299). This principle will later prove to be critical to his determination of the limits of metaphysics: see §6.)

The fact that our spectacles also involve concepts violates any counterpart of Hume’s basic empiricist principle that our simple ideas are copied from our simple impressions. It is nevertheless too soon to conclude that Kant distances himself from any direct equivalent of Hume’s semantic empiricism. For it remains to be seen what sort of meaning, if any, he thinks can attach to these concepts when they are disassociated from experience. On the other hand it is not too soon to conclude that Kant distances himself from any direct equivalent of Hume’s epistemic empiricism. His sheer commitment to synthetic
a priori
knowledge ensures that he does
that
. The question we must now broach is how far, if at all, we can aspire to such knowledge in the most general attempt to make sense of things, that is in metaphysics.

5. Good Metaphysics: The ‘Transcendental Analytic’

Kant believes that this aspiration is, up to a point, perfectly legitimate. He devotes at least a third of his first
Critique
, essentially the part entitled ‘Transcendental Analytic’, to the pursuit of it. There is a section entitled ‘Second Analogy’, for example, in which he considers what I called in §2 the Causal Principle, or what he calls ‘the principle of temporal sequence according to the law of causality’ (B232), and in that section he attempts a proof of this principle. If he succeeds – it is beyond the scope of this chapter to address the issue of how far he does
⁴¹
– then the upshot is, precisely, an item of synthetic
a priori
knowledge of a sufficiently high degree of generality to count as metaphysical. In this part of the first
Critique
, then, Kant’s project is to establish various metaphysical results, much as a mathematician’s project is to establish mathematical results.

(Later in the first
Critique
Kant explicitly compares and contrasts these two disciplines (A712–738/B740–766). There is one crucial feature that they have in common. Because metaphysicians and mathematicians are both in pursuit of synthetic knowledge, they must both appeal to intuitions. But this in turn signals the principal contrast between the two disciplines, which is methodological. Mathematicians appeal to the relevant intuitions by actually exhibiting them (A713/B741). For example, a geometrician might begin a proof by constructing a triangle. Metaphysicians, however, are concerned rather with
experience
, and hence with intuitions at least some of which are empirical. So if they exhibited any relevant intuitions, it would compromise their claim to be engaged in an
a priori
exercise. Their appeal to the relevant intuitions is instead an appeal to the sheer
possibility
of our being given
objects in intuition in the various ways we are (A766/B794). It is thus that metaphysics comes to involve the highly distinctive style of proof that Kant labels ‘transcendental proof’ (A786/B814ff.). A transcendental proof is a proof whose conclusion concerns the conditions that must obtain, as a matter of
a priori
necessity, in order for us to be given objects in intuition in the ways we are, or in order for us to enjoy experience of the kinds we do.
⁴²
)

6. Bad Metaphysics: The ‘Transcendental Dialectic’

The ‘Transcendental Analytic’ reveals the scope of metaphysics. It is followed in the first Critique by an even larger part, entitled ‘Transcendental Dialectic’, which reveals its limits. Some of what metaphysicians aspire to do, nay most of what they aspire to do, is not legitimate. This means that their task is twofold: not just to establish metaphysical results, and thereby to attain synthetic
a priori
knowledge, but also to keep in check their own impulses to try to establish metaphysical results where there is no synthetic
a priori
knowledge to be had; that is to say, not just to practise good metaphysics, but to combat bad metaphysics. Both tasks are united in Kant’s delightfully memorable
aperçu
concerning philosophy, which he might just as well have applied to metaphysics, that it ‘consists precisely in knowing its bounds’ (A727/B755).

Metaphysicians attempt to transgress these bounds whenever they attempt to make sense of what is transcendent. In a second, more Kantian formulation, they attempt to transgress these bounds whenever they attempt to attain synthetic
a priori
knowledge without appeal to intuitions. In the first of these formulations – which supplies a direct answer to the Transcendence Question in §6 of the Introduction – I am presupposing a suitably epistemic interpretation of what it is to make sense of something and a suitably experiential interpretation of what it is for something to be transcendent. The two formulations are equivalent because the knowledge that we can attain by appeal to intuitions, whether these be empirical or
a priori
, is the synthetic knowledge that we can attain about objects of a possible experience (for us). Such knowledge pertains to what we are actually given in experience if the intuitions are empirical. It pertains to our spectacles and therefore, indirectly, to what we are capable of being given in experience if the intuitions are
a priori
.

As we have seen, Kant denies that there is any other synthetic knowledge available to us. He is adamant that herein lie the only two epistemic uses to which we can put our spectacles. We can look through them at the world
and see how the world actually appears through them. Or we can reflect on the spectacles themselves and draw conclusions about how the world
must
appear through them. But there is no other synthetic knowledge that we can attain with their aid (e.g. Bxix, A92–94/B124–127, B147–148, B165–166, A139/B178, A238–240/B297–299, and A702/B730). Nor – Kant is just as adamant about this – can we take them off and look at the world directly.

Yet that is what metaphysicians most deeply aspire to do. They seek, as Kant would put it, synthetic
a priori
knowledge of things in themselves.
⁴³
They want to establish whether or not there is a God sustaining all that we experience and directing its various operations; whether or not, in the multifarious episodes that make up our lives, there are any exercises of pure free will; whether or not we each have a soul, persisting through all the vicissitudes of our physical existence and beyond.
⁴⁴
(Each of our protagonists so far has had something to say about each of these.) These are the great questions of metaphysics. Kant himself insists that such questions are of vital concern to us. He regards the three concepts of God, freedom, and immortality as the three most important and most potent concepts of mainstream Christianity (e.g. 2nd
Critique
, Pt One, Bk II,
Ch. II
, §
VI
). But such questions, at any rate on Kant’s understanding of them, are questions about what is transcendent. We cannot hope to answer them.
⁴⁵

In the ‘Transcendental Dialectic’ Kant supplements his general account of what is wrong with attempts to answer such questions, and of why we nevertheless feel the urge to make these attempts, with a case-by-case rebuttal of the various specific attempts that metaphysicians have made, their various forays into transcendent metaphysics, as we might say. Kant himself comments, in the very last sentence of the ‘Transcendental Dialectic’:

It was not only necessary to carry out an exhaustive examination of the vain elaborations of speculative reason in their entirety down to its primary sources, but also – since dialectical illusion is here not only
deceptive for our judgment but …, owing to the interest we take in these judgments, is also alluring and natural, and so will be present in the future too – it was advisable to draw up an exhaustive dossier, as it were, of these proceedings and store it in the archives of human reason, so as to prevent future errors of a similar kind. (A703–704/B731–732)

Kant divides the misguided efforts of metaphysicians into two broad classes, according to whether their questions are ill-conceived or well-conceived (A740–741/B768–769).
⁴⁶
There is an echo in this division of a division that Hume would recognize between violations of his semantic empiricism and violations of his epistemic empiricism. To see how Kant understands the division, we need first to see what he means by an ‘idea of reason’. By an idea of reason Kant means one of the twelve fundamental
a priori
concepts that constitute part of our spectacles or else a concept that can be defined in terms of these twelve, freed of whatever apparatus allows it to be applied to objects of possible experience (A320/B377 and A408–409/B435). So freed, it can be applied to things in themselves. And Kant believes that the questions addressed by metaphysicians in their misguided efforts to attain synthetic knowledge of what transcends experience always involve some idea of reason. Such a question is ill-conceived if it involves a confused amalgam of an idea of reason with some concept that can be applied only to objects of possible experience. It is well-conceived if it involves ideas of reason without any such distortion. In the former case the question has no answer.
⁴⁷
In the latter case the question has an answer, but only at the level of things in themselves. The problem with a question of the latter kind, for those metaphysicians trying to answer it, is simply that they (we) lack the resources to do so. The three questions mentioned earlier, concerning God, freedom, and immortality, are of this latter kind. The three concepts of God, freedom, and immortality (suitably understood) are undistorted ideas of reason. We may speculate about whether they are instantiated among things in themselves. But we can never know whether they are.

Not all metaphysical questions are of this kind, however. Many are of the former kind, that is to say ill-conceived. Thus metaphysicians in the past, notably Leibniz and his followers on the one hand and followers of Newton on the other hand,
⁴⁸
have debated whether the physical universe is either infinitely old or infinitely big. Kant urges that these questions do not so much as arise unless there
is
such a thing as the physical universe,
as a whole
. But on Kant’s view there is not. The concept of the physical universe
as a whole is a confused amalgam of the concept of unconditionedness, which is an idea of reason, with the concept of physical reality. This requires something that is both physical and all-encompassing. But the only physical things that can exist are objects of possible experience. And no object of possible experience can be all-encompassing. That is, no object of possible experience can encompass the whole of physical reality. The source of our mistake, when we conflate these concepts in this way, is a genuine insight: namely, that there must ultimately be something unconditioned corresponding to anything conditioned, and in particular corresponding to any conditioned physical thing (Bxx
⁴⁹
). What we fail to appreciate,
however, is that such unconditionedness must reside in things in themselves, which physical things are not. We naturally assume that some
physical
thing must be unconditioned, in other words, that there must be such a thing as the physical universe as a whole, finite or infinite as the case may be. Once we drop that assumption, we can acquiesce in the conclusion that every physical thing is part of some other physical thing that is older and bigger – as the earth, for instance, is part of the solar system – although there is no one physical thing of which every physical thing is part (‘Transcendental Dialectic’, Bk II, passim, esp. §§IVff.).

These debates about the age and size of the physical universe are especially significant for Kant, for they illustrate perfectly the ‘battlefield of … endless controversies’ to which he refers at the beginning of the Preface to the first edition of the first
Critique
(Aviii). He holds that, on the assumption that the physical universe does exist as a whole, there are entirely valid reasons both for denying that it can be temporally or spatially infinite and for denying that it can be temporally or spatially finite. It is hardly surprising, then, that metaphysicians in the past have again and again returned to these issues, (unsuccessfully) defending their own views by (successfully) attacking the views of their opponents, with no prospect of reconciliation while the offending assumption is still in place. Kant lays out their arguments alongside one another (as he does arguments concerning the divisibility of matter, the sovereignty of the laws of nature, and the existence of a necessary being) as a way of displaying the dialectic from his own impartial standpoint.
⁵⁰
He
then proceeds to explain how removing the offending assumption can lead to a resolution (see esp. A497/B525ff.).