34 A. E. TAYLOR : to the Republic, which may serve to bring out the vital connexion between the abstract logical categories of the Parmenides and the more concrete forms of Platonic thought with which we are all familiar. At the beginning of the Philebus there is a striking passage which recalls the paradoxes of the Parmenides so closely that it is almost impossible not to regard it as a piece of conscious self- criticism. 1 At Philebus, 14 c, Socrates is led to attack the problem, which, as he says, " troubles us all," of the re- lation between One and Many. In further specifying the manner in which the question should be discussed, he first (14 D) sets aside, just as he had done in the Parmenides, the vulgar puzzles which arise from the combination of unity and plurality in the sensible world. These puzzles and the paradoxes founded on them are ra davfMaarrwv Trepl TO ev teal TroXXa, eVo? elirelv inro irdvrcav 77877 fj,r) Seiv rwv roiovrwv 7raiSapi(i)8r) KOI paSta rcai a~<f)6$pa rots yoi$ /j,7r6$ia vTToa[j,(3av6vT(i)v yvyvea-Qai (14 D). His concern in the Phi- lebus, as in the Parmenides, is with the ideal unities them- selves. And about these unities there are still, as there were in the Parmenides, two questions : (1) Do they exist ? (15 B, Trp&rov fj,ev el' rti>a<? Sel roiavras elvat /u,oi>aSa? KT). (2) How is their unity compatible with their presence in the countless particulars'? This last question is the standing diffi- culty of philosophy (irepl . . . rwv roiovrfov 77 TTOIJ (nrovSij fjLera &iaip6<TQ)s d/j.<f)ia'l3ijri]O'i<; r yi i yveraL, B. TOUT' ecrri TO, Trepi ra roiavra ev Kai 7roa . . . airacff]^ aTropias atria fj,ij /caXeo? opdXoyrjQevTa real evTropias av /caXai?, c). When it first dawns on a young man he feels that he has discovered a veritable treasure-trove of wisdom, and in his delight in his own skill can never rest from perplexing himself and every one else with his demonstrations that the One is Many and the Many are One (15 E). But the task of the true philosopher, in dealing with the difficulty, is a more laborious one (1(5 D E). He must not content himself with showing that each ideal One is also an indefinite plurality ; his work is to reduce that plurality to order by once more detecting in it subordinate unities as often as may be possible, and so to substitute for the abstract categories of unity and plurality a graduated hier- archy of higher and lower unities. That is, it is not enough to perceive that there is system in the world unless you go on to ascertain the contents of the various systems and their 1 Apelt also notices the similarity, but does not draw from it the inference which I am inclined to do. Beitraye, p. 33.
