162 



PROFESSOR THOMPSON, ON THE GENUINENESS OF 



Academy: and we can have no doubt that Plato regarded such long-drawn chains of dis- 

 tinctions in the light of a useful exercise for his pupils. They became " more inventive" and 

 " more dialectical" — may we not say, clearer-headed — by the process. 



I may add that the Invention of the Divisive Method is traditionally attributed to Plato 

 by the Greek historians of philosophy. Aristotle devotes several chapters of his Posterior 

 Analytics to the discussion of this method: he points out its uses and abuses, and defends it 

 against the cavils of Plato's successor Speusippus, who abandoned the method because, as he 

 alleged, it supposed universal knowledge on the part of the person employing it. The 

 method discuss.ed is that which we have been considering, for Aristotle describes it as 

 Division by contradictory Differentiae^ He also replies to the objection that this process 

 is not demonstrative — that it proves nothing — by the remark that the same objection 

 applies to the counter process of collection or induction. Tiiis defence, I presume, would 

 not in the present day be accepted as satisfactory ; for, as the able translator of the 

 Analytics observes, "This is the chief flaw in Aristotle's Logic: for some more vigorous 

 method than the Dialectical, the method of Opinion, ought to be employed in establishing 

 scientific principles." To shew the superiority of modern over ancient methods of 

 arriving at truth, is a gratifying, if it is not the most profitable employment of the Historian 

 of Ancient Philosophy. At the same time, I must confess my inability to discover the 

 flaw in the principle of dichotomy, as a principle of classification, in cases where the 

 properties of the objects to be classified are supposed to have been ascertained. A Class 

 can exist as such only by exclusion of alien particulars. The Linnean Class Mammalia 

 for instance, implies a dichotomy of Animals into Mammal and Non-Mammal — into 

 those which give suck and those which do not. The distinction may or may not be a 

 natural or convenient one, but in any other which may be substituted, some " differentia," 

 some property or combination of properties must be fixed upon, which one set of species 

 or individuals possesses, and which all others want. And this is all that is essential in " dicho- 

 tomy," or the " method of Division by contraries^." The application of the method will, 



' Anal. Post. ii. c. xm. § 6, and Schol. in loc. So 

 Abelard (Ouvrages Inedits. Op. 5fi!), ed. Cousin : coll. pp. 451, 

 461), distinguishes between those divisions which imply di- 

 chotomy and those which do not : e.g. 



animal. animal. 



man. horse. ox, &c. man. not man. 



Porphyry attributes the latter or dichotomous method to Plato. 

 It could not be " Eleatic," tor each of the contraries would be 

 in that scheme a "non-ens." It is remarkable that a similar 

 Divisio Divisicnum occurs in the Potiticus, p. 287, § 27, where 

 in lieu of the regular dichotomy a rougher form of classi- 

 fication is for once adopted. This Plato, keeping up the 

 original metaphor in the Phtzdrus, describes as a fieXoroixia. 

 Kaxa fieXt) Toivvv auTa^ oiou lepe'iov Siaipufxeda, eireiSri 

 iiX^ (iSuvaTOVfieUt ^** ywp cis tow eyyuraTa OTt fid\i(TTa 

 Tc'/ui/eiK dpiBfjiiv dei. The division he proceeds to make, is a 

 distribution of "accessory arts" avvairioi reX""') into seven 

 co-ordinate groups. A similar relaxation is permitted in the 

 Philebut, p. 16 D: Aei ouv tj^as dfi /ui'av ISeav irepl 



traVTOi eKaV-roT* deflevov^ X.iiTelv...edv ouv [neTa]\d^fi€v, 

 juera pi'iav Suo, et ttws elcrij iTKOireXv, el Si /.it;, t/>c7s 

 »? Til/' dWov dpid fJLOV, Kai twu ev enelvmif SKacnov 

 irdXw ciiTauTa}^ /te')(pnr6p dv to »c«t' dpxf'^^ ^^ M^i ^'Ti ei/ 

 Kai uTretpd tcTt fiuvov iSri -ris, dWd Ktil o-Kuva. 1 understand 

 this passage as conveying Plato's di>tinction between his own 

 method and that of the Eleatics and tlieir Eristic successors, 

 who acknowledged only a ev and an direipov, 



' For the length of the process will evidently depend on the 

 distance, so to apeak, between the Species generalissima and 

 the Species specialissima, between the remote and the proxi- 

 mate class in the tabulation of species. The very brief dicho- 

 tomy in the Gorgias, p. 464, is evidently the same in principle 

 as the long drawn divisions in the Sophisla, as will be seen 

 fiom the following scheme : 



Qepaireia^ 



11 TOV <TtOfiaT0^. 



rill ^vx^^' 



yvfivaiTTtKti, 



laTptKtJ. UOflodeTlKIJ* 



SlKaVTLKlj, 



