pebeEw is one which may exist between A and B and between A and not B, but not between both at once ; but the relation of genus A to differentia B is one which A has at once to B and to not-B. Therefore genus and differentia do not form a unity xara péOeéw.
QI. et S€ Kat petexer. ‘Even if the genus could partake of the differentia, this would not explain how a genus + several differentiae form a unity.’
24. otTw pev yap ...év, ‘Forat that ratea genus and a// its differ- entiae would form a unity.’
27-29. Aristotle says the question must first be considered with re- gard to definitions by division (cf. 10384 34), but he never gets to any other kind. The other kind is that é« rév évurapxdvrwv (B. 998? 13, H. 1043? 20).
10384 5-9. Since the genus does not exist apart from the species or exists only as their matter, it offers no obstacle to the unity of the definition, and accordingly the definition may be considered as if it consisted only of differentiae. This is the first step in the explana- tion of the unity of the definition. The next step is to show that the differentiae in a definition may be reduced to one.
5. Ta ds yevous eidn, cf. A. got 31 n.
7. TA €idn Kal Ta oToLxeia, ‘the species, i.e. the letters’,
Q-10. TH... Svadopd, Prof. Joachim’s emendation of ry... da- opay, seems to be required by the sense.
14. GAN’ 7 KTH., ‘(and in general we can divide “animal with feet” into nothing) except ‘“ with divided feet ” and ‘‘with undivided feet” ’. ddN’ % in effect = ‘but only’. Cf. 1036> 31, I. roo5@12, A.A. 563» 22, 580% 20, Pol. 125721. The idiom is explained by Cook Wilson in Class. Quart. tii. 121-124.
17-18. tote & . . . Svadopats, i.e. there will be as many infimae species of foot or of animal with feet as there are ultimate differentiae.
19. 7 TeAcuTata Siapopa 7 otcia tod mpdypatos €oTat, since it will presuppose all the previous differentiae and finally the genus. From another point of view Aristotle can say (Zop. 139% 29, 142° 27, 143% 18) that the genus is the element in, the definition most expressive of the essence ; it is so because all the other elements presuppose it.
30. Kard ye Td dpOdv, i.e. according to the method in which each differentia is a differentia of the previous one.
33. Tdéis 8 obk eorw ev TH otcia, sc. and therefore what is seen by a perdraéis to be superfluous must have been superfluous before.
35. Thy mpérnv, ‘at the first attempt’, cf. 1037%27n. For the idiom (in which some such word as oddv was originally understood) cf, Ar. Zhesm. 662, Dem. O2. iii. 2, Hdt. iii. 134, i. 153 (the last passage has rv mpdrny eivor). Pol. 12865 is not a parallel, for adeiobw tyv mpornv seems to mean ddeicbw thy rpeTyv povapxtav €mLoKOTrELV.
208 Commentary
No UNIVERSAL IS SUBSTANCE; NO SUBSTANCE CONSISTS OF SUBSTANCES (chs. 13-16).
The universal ts not substance (ch. 13).
1038” 1 (B) (cf. 1028>33). We have discussed two things that are held to be substance—the essence, and the substratum (which we showed to ‘underlie’ in two ways, as the ‘ this’ underlies the accidents, and as the matter underlies the actuality).
6. We now proceed to discuss the universal, which some think to be the truest cause. But it seems that no universal can be a substance, for :
g. (1) If it be suggested that the universal is the substance of a thing, we answer: (a) The substance of a thing is that which is peculiar to it, but the universal is common to many. It must be the substance of all or of none. But it cannot be the substance of all; and if it be the substance of one, this one will de the others, for things whose substance or essence is one are one.
15. (4) It is that which is not predicated of a subject that is substance, but the universal zs predicated of a subject.
16. (2) If it be suggested that the universal is not substance in the sense of essence, but is included in the essence, e. g. ‘animal’ in man, then (a) evidently it is definable (and so there will be an infinite regress).
19. But (4) even if not all elements in substance are definable, the universal will be the substance of something ; as ‘man’ is the sub- stance of the manin whom it is present, ‘ animal’ will be the substance of that in which it is present and to which it is confined. (Thus suggestion (2) turns into (1).)
23. Further, (c) a ‘this’ or substance, if it is composite, must con- sist not of qualities but of substances. Otherwise non-substance will be prior to substance ; but this is impossible; for affections are not prior to substance in definition, in time, or in generation, else they would be capable of existing apart.
2g. Further, (@) in Socrates a substance (animal) will be present as an element, and will therefore be the substance of two things (the class of animals and Socrates).
30. (e) In general, if infimae species are substances, none of the elements in their definition is the substance of anything or can exist apart from its particular instances or in anything else.
34. (3) No common predicate indicates a ‘this’, but only a ‘such’, Otherwise we get the ‘third man’ and other difficulties.
Zats. 1933 2~7 209
1039° 3. (4) A substance cannot consist of other substances existing actually ; for what is actually two is not actually one (e. g. a line which is double another consists only potentially of two halves, for their actualization separates them).
g. Democritus puts our point well when he says that one cannot be produced out of two nor vice versa; he refers to atoms, which he regards as the only substances. Similarly, if number is, as some say, a synthesis of units, the number two is not one, or else contains no units actually.
14. Our result involves a difficulty. If no substance can consist of universals (because-they indicate a ‘such’, not a ‘this’), nor of substances actually existing, all substance will be uncompounded and therefore indefinable. ‘
1g. But it is universally agreed that substance is the only or the chief object of definition. Either, then, all things are indefinable or they are definable in one sense though not in another. Our meaning will appear more clearly later.
Chs. 13-16 form a separate section of the inquiry into substance, the main upshot of which is summed up in 10412 3-5: ‘no universal is substance, and no substance contains substances as its parts’. The section begins (ch. 13) by discussing and rejecting the claim of the third claimant to the title of substance—the universal, and implicitly also that of the fourth claimant—the genus. From this follows (ch. 14) the rejection of the claim of the Ideas to be substance. Further it is argued (ch. 15) that the individual is indefinable. This is argued partly on the ground that the individual is subject to destruction and change ; but the connecting link with ch. 13 lies in the other argument, that since the universal is never a substance, never a ‘this’, always a ‘such’ (13. 10392 15, 16), definition, which is an enumeration of universal marks, can never adequately express the nature of an individual. Next a corollary is inferred from the principle which was made the basis of one of the proofs in ch. 13 (1039 3, 16), viz. that sub- stance cannot be composed of actual substances; it is argued that the material parts of substances are not actually substances (16. 1040 5— 16). And finally a further attack is made on the Platonic tendency to identify substance with the universal (1040 16—1041° 5).
1038" 2, 3. This is a reminiscence of 1028) 34; genus, however, is now omitted (as coming under the universal), and 76 ék rovtwv, the concrete individual, is added.
4. mept rod ti Hv etvor, chs. 4-6, 10-12; Kal Tod droKerpevou, ch. 3.
5. ott SixGs bmdxertat, cf. 10297 2 n.
7. trow, the Platonists.
210 Commentary
10. I have adopted here the reading which sense and idiom seem to require ; it accounts well for the various readings found in the manu- scripts and the commentators.
13-15. The argument is not clear but may be interpreted as fol- lows : ‘ What will the universal be substance of? Either of all its particulars or of none (for there is no-reason why it should be sub- stance of one any more than of the others) ; but it cannot be substance of all (since, as we have just seen, 1. 10, the substance of a thing is something peculiar to it. It follows, then, that it is the substance of none of its particulars). Ifwe try to avoid this conclusion and treat it as the substance of one of them, then (since (the universal will be no less the substance of its other particulars, and) things that have the same substance are identical) this one will de the others; which is absurd,’ ;
18-23. In answer to the suggestion now put forward, that the universal is not the substance of things in the sense of essence, but is a substance because it is an element present in their essence, Aristotle replies that the universal can be defined, and seems to have meant to add that in that case it will itself contain a generic or universal element, and so substance will be contained in substance, ad zufinifum. But, he observes, we need not make this assumption, that everything that is part of the substance of something can be defined. Quite apart from this assumption, the universal will be the substance of something, as ‘man’ is of the man in whom it is present; so that the new view that the universal is a substance present 7m the essence of its particulars turns into the old one which has already (9-16) been refuted, that the universal zs essence. For the universal, e. g. animal, will be the sub- stance, not indeed of man but of that in which it is present as some- thing peculiar to it; i.e. of the class including all animals. This interpretation enables us to treat ll. 19-23, 23-29, 29-30 as giving three distinct arguments (as éru in ll. 23, 29 shows that they do), not (as Bz. does) as giving parts of one argument.
22, Bz.’s suspicions of otaia in this line are amply confirmed by its absence in A> and in Asc.c. The vulgate reading is due to the conflation of alternative readings, ovata éxe‘vov and éxeivov ovata.
23-29. The argument seems to be as follows: Universals having been -shown not to be substances (Il. gQ—16), they must not be regarded as constitutive elements of substance (as the view under consideration regards them, 16-18), since then non-substance would be prior to substance.
27-28. olte byw yap ore xpovw ore yevéoet .. . mpdtepa. Aristotle nowhere else distinguishes between xpdvw mpdrepov and yevéce mpo- tepov, nor is any possible distinction apparent. Prof. A. R. Lord has suggested yvwoe. for yevéoe, and this derives some support from 1028 32. But (1) the Greek commentaries as well as the manu- scripts read yevéoe, and (2) xpévm would not be likely to be put between Adyw and yvwoe, which are at least very near one another in sense (in @. 104917 they seem to be identified), It is better there-
fore to keep yevéoe and to suppose that it is added as a synonym of xpove.
29. The variety of readings here seems to be due to Alexander’s 7 Ywxpdre odcia dvr evurrdpser ovoia, in which ovota dvr is no doubt added to bring out the point. ‘In Socrates, himself a substance, there will be present as an element a substance (sc. animal), which will therefore be the substance of two things (sc. of the class of animals and also of Socrates).’
30. 6 dvOpwros, i.e. the individual man, Alexander thinks. But as Bz. observes, we should in that case expect something like 6 cis avOpwros ; and we should expect in 1. 33 twa dvOpwrov rapa Tovs Twas. It seems therefore that Aristotle means the infima species, which according to one of his lines of thought he regards as substance; cf. P. A. 644% 23. It is the substantiality not of man but of animal that he has been attacking in ll. 16-30; the infima species is at any rate more substantial than the genus.
32. év @AXw, not, as Alexander interprets it, ‘in the Idea’, The words are quite general.
33. Ta tTwd, ‘the particular species of animal’.
34. &k te... toUTwy is answered by xal dru (‘ and because’).
1039%2. 6 tpitos avOpwros, cf. A. g90? 17 n.
6. 4 Simdacta, sc. ypaypyy, says Asc., doubtless rightly. We may cf. 7 eveta, » qrodiaia, and A. rorg* 8 n., ©. 10488 33n.