454 w. WUNDT'S LOGIK, n. Turning to Definitions, we find the distinction between those which are real and those which are nominal kept up ; in the sense that the nominal is understood to be the mere substitution of one word for another which may be less known, just as we may translate a sentence into a better known language. The real, on the other hand, looks to the scientific connexion of the words or concepts, and the relation in which the phenomena with which they are connected stand towards each other. The main principles of Scientific Classification are fully and clearly explained. In place of the ordinary distinction between natural and artificial classification is substituted a three-fold one into the descriptive, the genetic, and the analytical. The merely descriptive one takes account solely of the co- existence of the attributes possessed by the things which are to be arranged, without seeking after their inner connexion or their origin. So far we are resorting to Analysis only. As we then proceed to appeal to Synthesis either by active construction or more passive observation, according to the nature of the subject-matter we begin to obtain a genetic classification in which the production of the determining attributes has to be considered. The province of Mathematics of course offers the most wide and familiar illus- tration of this stage, as when, to take a simple example, we arrange the various curves of the second order according to the processes of construction by which they may be obtained. But wherever possible, we want to proceed further in the way of synthesis than this. We seek to know not merely the mode of production of the constituent elements of our concepts, but to obtain the fullest possible expression of the permanent relations of these elements to each other. This involves a fresh appeal to Analysis, and leads to the analytic classification in the narrower sense of the term. It differs from the descriptive in that it does not merely study the arrangement of the attributes but aims at a full account of the laws by which they are mutually connected. Like the preceding it is hardly attainable except where, as in Mathematics and to some extent also in Jurisprudence and Poli- tical Economy, we are able to penetrate to the essence of the notions with which we deal, owing to the extent to which they are of our own construction. The Doctrine of Proof follows the same leading division ; that is to say, the distinction between Analysis and Synthesis is taken as the principle of classification amongst the different kinds of proof. As the consequent arrangement is very different from that which is given in any of the hand-books familiar in England, it may be desirable to describe it somewhat more fully than would otherwise be expedient. It resembles in some respects that which Lotze has given in his Lonik, but is more elaborate ; more- over, on some points Prof. Wundt distinctly rejects the scheme of his predecessor. Starting then from the commencement we divide into direct and indirect proofs, in the former the thesis
