Page:Philosophical Review Volume 7.djvu/96

82 and Generalization, the five general methods by which knowledge is attained (pp. 515-16 et passim). The author criticises severely Jevons's theory of reasoning from hypothesis (part ii., chap, xvii), which has also been adopted by Sigwart. His own theory, as he himself admits, is closely allied to that proposed by Mr. Bosanquet (Logic, vol. ii., pp. 155 ff). The latter, however, explicitly ranges himself on the side of Jevons and Sigwart, as opposed to Mill (Logic, vol. ii., p. 160), although he takes pains to show wherein he regards their statements as in need of correction (Ibid., p. 175 f). Mr. Hobhouse has taken the opposite course, emphasizing (and, as it seems to me, exaggerating) the difficulties in the theories he criticises.

The book is evidently the result of solid work and careful thinking on the part of a philosophical scholar who is well acquainted with the literature of his subject. And, whether one agrees with the author's conclusions or not, his book is decidedly one that must be read and reckoned with.

Mr. Selby-Bigge has already twice laid the student of philosophy under considerable obligation. In 1888 he published the only convenient and reliable student's edition of Hume's Treatise of Human Nature which we have, and six years later he published an equally needed edition of the two Inquiries. Both of these editions were on an entirely higher plane than the ordinary carefully edited text-book of this kind, and they are so universally appreciated that any praise here would be almost an impertinence. The editorial work on these two volumes, however, was of a singularly impersonal character. Aside from the short, but excellent, introduction to the later volume which was mainly intended to facilitate a comparison between the earlier and the later form of Hume's philosophy Hume was made to perform the whole task of interpreting himself by means of the two very full and altogether excellent indexes, which the editor had supplied. To the same editor we are now indebted for two goodly volumes, which fill a need only less imperative; but the task undertaken here was in several respects a more difficult one. Mr. Selby-Bigge has