nciples are
synthetic, and here, as will be shown, it is not experience but "pure
intuition" which permits us to go beyond the concept and add a new mark
to it.]

[Footnote 2: The Scholastics applied the term _a priori_ to knowledge from
causes (from that which precedes), and _a posteriori_ to knowledge from
effects. Kant, following Leibnitz and Lambert, uses the terms to designate
the antithesis, knowledge from reason and knowledge from experience. An _a
priori_ judgment is a judgment obtained without the aid of experience. When
the principle from which it is derived is also independent of experience it
is absolutely _a priori_, otherwise it is relatively _a priori_.]

Two sciences discuss the _how_, and a third the _if_ of such judgments,
which, at the same time, are ampliative and absolutely universal and
necessary. The first two sciences are pure mathematics and pure natural
science, of which the former is protected against doubt concerning its
legitimacy by its evident character, and the latter, by the constant
possibility of verification in experience; each, moreover, can point to
the continuous course of its development. All this is absent in the third
science, metaphysics, as science of the suprasensible, and to its great
disadvantage. Experiential verification is in the nature of things denied
to a presumptive knowledge of that which is beyond experience; it lacks
evidence to such an extent that there is scarcely a principle to be found
to which all metaphysicians assent, much less a metaphysical text-book
to compare with Euclid; there is so little continuous advance that it is
rather true that the later comers are likely to overthrow all that their
predecessors have taught. In metaphysics, therefore, which, it must be
confessed, is actual as a natural tendency, the question is not, as in
the other two sciences, concerning the grounds of its legitimacy, but
concerning this legitimacy itself. Mathematics and pure physics form
synthetic judgments _a priori_, and metaphysics does the same. But the
principles of the two former are unchallenged, while those of the third
are not. In the former case the subject for investigation is, Whence this
authority? in the latter case, Is she thus authorized?

Thus the main question, How are synthetic judgments _a priori_ possible?
divides into the subordinate questions, How is pure mathematics possible?
How is pure natural science possible, and, How is metaphysics (in two
sen