In several passages of his Commentary Vaihinger attacks views expressed in my edition of the Kritik. Nevertheless this polemic is directed to too great an extent against individual points, and is too fragmentary and not sufficiently thorough with regard to my fundamental position,—the gradual genesis of the K. d. r. V. and the later insertion in an outline which was already almost in its final form of the opposition between analytic and synthetic,—to require that I should here undertake a refutation. I may, however, be permitted to make just two remarks. 1) Vaihinger holds that several passages of the Kritik to which I have ascribed a somewhat late origin must be attributed to an earlier period (cf. pp. 354, 411, 417, 453). I regard the reasons which he gives as entirely inadequate. The alleged archaic changes of standpoint in these passages are entirely without significance, and depend upon that uncertain determination of notions which is, unhappily, in Kant's writings part of the regular program. Vaihinger himself, according to the opinion expressed on p. 354 and 453-4, note, appears to attach no great importance to this fixing of dates. 2) In spite of Volkelt's conclusions, and of the first volume of the Commentary to which Vaihinger refers me (p. 342, note), I hold fast to the opinion—due to some extent to the influence of Paulsen's Entwurf, and confirmed by my own studies,—that "the rescue of rationalism from Hume's attack," was the historical point of departure in the formation of Kant's system, and that it can only be understood with reference to this end. Nevertheless Kant has by no means always kept this point of departure in the foreground, but has laid the greatest value and emphasis, sometimes upon one side and sometimes upon the other, according to the arrangement of the factors of his thought. So much for the confused image of Kant's own expressions and for the opinions of his expositor! I was not able to give reasons for my opinion in the edition of the Kritik, because the space even for my own explanations was strictly limited. Moreover, I adopt Vaihinger's expression (I, 70), that Kant's system is an organism the members of which are purposively arranged, but I turn the figure against its author. Man, also, is an organism and yet there are in him more and less valuable parts, those which are essential to life and those which are superfluous. One might assign to rationalism for Kant's system the office of the blood which, bearing life with it, flows through the smallest parts and those most remote from the center of life. I agree with Vaihinger in holding that Kant endeavored in many problems to mediate between the two extremes. I deny, however, that this tendency to mediate could become a motive
