REMARK 2

The following important observation here naturally presents itself: There are three kinds of antinomies of pure reason, which, however, all agree in forcing reason to abandon the otherwise very natural assumption which takes the objects of sense for things-in-themselves, and to regard them, instead, merely as phenomena, and to lay at their basis an intelligible substrate (something supersensible, the concept of which is only an idea and affords no proper knowledge). Apart from some such antinomy, reason could never bring itself to take such a step as to adopt a principle so severely restricting the field of its speculation, and to submit to sacrifices involving the complete dissipation of so many otherwise brilliant hopes. For even now that it is recompensed for this loss by the prospect of a proportionately wider scope of action from a practical point of view, it is not without a pang of regret that it appears to part company with those hopes, and to break away from the old ties.

The reason for there being three kinds of antinomies is to be found in the fact that there are three faculties of cognition, understanding, judgement, and reason, each of which, being a higher faculty of cognition, must have its a priori principles. For, so far as reason passes judgement upon these principles themselves and their employment, it inexorably requires the unconditioned for the given conditioned in respect of them all. This can never be found unless the sensible, instead of being regarded as inherently appurtenant to things-in-themselves, is treated as a mere phenomenon, and, as such, being made to rest upon something supersensible (the intelligible substrate of external and internal nature) as the thing-in-itself. There is then (1) for the cognitive faculty an antinomy of reason in respect of the theoretical employment of understanding carried to the point of the unconditioned; (2) for the feeling of pleasure and displeasure an antinomy of reason in respect of the aesthetic employment of judgement; (3) for the faculty of desire an antinomy in respect of the practical employment of self-legislative reason. For all these faculties have their fundamental a priori principles, and, following an imperative demand of reason, must be able to judge and to determine their object unconditionally in accordance with these principles.

As to two of the antinomies of these higher cognitive faculties, those, namely, of their theoretical and of their practical employment, we have already shown elsewhere both that they are inevitable, if no cognisance is taken in such judgements of a supersensible substrate of the given objects as phenomena, and, on the other hand, that they can be solved the moment this is done. Now, as to the antinomy incident to the employment of judgement in conformity with the demand of reason, and the solution of it here given, we may say that to avoid facing it there are but the following alternatives. It is open to us to deny that any a priori principle lies at the basis of the aesthetic judgement of taste, with the result that all claim to the necessity of a universal consensus of opinion is an idle and empty delusion, and that a judgement of taste only deserves to be considered to this extent correct, that it so happens that a number share the same opinion, and even this, not, in truth, because an a priori principle is (r)presumed to lie at the back of this agreement, but rather (as with the taste of the palate) because of the contingently resembling organization of the individuals. Or else, in the alternative, we should have to suppose that the judgement of taste is in fact a disguised judgement of reason on the perfection discovered in a thing and the reference of the manifold in it to an end, and that it is consequently only called aesthetic on account of the confusion that here besets our reflection, although fundamentally it is teleological. In this latter case the solution of the antinomy with the assistance of transcendental ideas might be declared otiose and nugatory, and the above laws of taste thus reconciled with the objects of sense, not as mere phenomena, but even as things-in-themselves. How unsatisfactory both of those alternatives alike are as a means of escape has been shown in several places in our exposition of judgements of taste.

If, however, our deduction is at least credited with having been worked out on correct lines, even though it may not have been sufficiently clear in all its details, three ideas then stand out in evidence. Firstly, there is the supersensible in general, without further determination, as substrate of nature; secondly, this same supersensible as principle of the subjective finality of nature for our cognitive faculties; thirdly, the same supersensible again, as principle of the ends of freedom, and principle of the common accord of these ends with freedom in the moral sphere.