changeableness of metaphysical systems, and also the con stantly accompanying scepticism, in every age has testified. Yet against the possibility of metaphysics in general this changeableness cannot be urged, for the same thing affects just as much all branches of natural science, chemistry, physics, geology, zoology, &c., and even history has not remained exempt from it. But when once, as far as the limits of human intellect allow, a true system of metaphysics shall have been found, the unchangeableness of a science which is known a priori will yet belong to it; for its foundation can only be experience in general, and not the particular and special experiences by which, on the other hand, the natural sciences are constantly modified and new material is always being provided for history. For experience as a whole and in general will never change its character for a new one.
The next question is: How can a science drawn from experience pass beyond it and so merit the name of metaphysics? It cannot do so perhaps in the same way as we find a fourth number from three proportionate ones, or a triangle from two sides and an angle. This was the way of the pre-Kantian dogmatism, which, according to certain laws known to us a priori, sought to reason from the given to the not given, from the consequent to the reason, thus from experience to that which could not possibly be given in any experience. Kant proved the impossibility of a metaphysic upon this path, in that he showed that although these laws were not drawn from experience, they were only valid for experience. He therefore rightly taught that in such a way we cannot transcend the possibility of all experience. But there are other paths to metaphysics. The whole of experience is like a cryptograph, and philosophy the deciphering of it, the correctness of which is proved by the connection appearing everywhere. If this whole is only profoundly enough comprehended, and the inner experience is connected with the outer, it must be capable of being interpreted, explained from itself. Since Kant
