"All-Crushing" Critic of Metaphysics 257This alone should be enough to show that he did not consider this partic¬
ular skepticism as deserving of an answer.
Kant did not object to being regarded as close to Hume. From his own
point of view, Hume was more an ally or predecessor than an adversary.
Though Kant did not follow Hume in all details and rejected Hume's skep¬
tical conclusions, the differences were not very significant to him. There
are surprisingly few remarks that are critical of Hume in his theoretical
works, and there are none that are hostile. Instead, Kant constantly em¬
phasized the importance of Hume in his published works. In his lectures
he advised his students to read Hume's works "many times." To sum up,
Kant thought that Hume offers the following argument:(1) Assume the causal relation to be rational.
(2) If a relation is rational, it can be thought a priori and on the basis of
concepts.
(3) For objects to be causally related, they must stand in a necessary relation,
such that if one object is posited, the other one must also be posited.
(4) It is impossible to see by reason alone how the existence of one object
necessitates the existence of another.
(5) Therefore, "it is wholly impossible to think such a conjunction a priori
and out of concepts."
(6) Therefore, the causal relation is not rational.
(7) Therefore, it is impossible to understand "how the concept of such an a
priori connection can be introduced."
(8) Therefore, it must have some other source or sources, and the most rea¬
sonable ones are imagination and custom.
(9) But imagination and custom can produce only "subjective" necessity.
(10) Metaphysics requires necessity based on intersubjectively valid concepts.
(11) Therefore, metaphysics is impossible.Kant thinks that the argument ending with (6) as a conclusion is sound.
Hume "proved," he says, "irrefutably: that it is wholly impossible for rea¬
son to think such a conjunction a priori and out of concepts."
What Kant does not accept is (7) and the conclusions founded upon it.
He cannot, if only because this would show that the science of metaphysics
is impossible. In order to save the science of metaphysics, or to show how
it is possible, he must show how it is possible to introduce the concept of
such a connection a priori. There is no reason for Kant to accept (7), in
any case. From the fact that the causal relation cannot be shown by reason
to be a priori, it does not follow that it cannot be shown to be a priori in some
other way, just as it does not follow from the fact that I cannot determine