The difference between the Idea and the ordinary subject of defini- tion is brought out in the next words (17-21). ‘ This applicability of the proposed definition to more subjects than “that for which it is pro- posed mus/ exist in the case of the eternal entities (the Ideas), since the genus and the differentia are prior to and parts of the compound, the species, and since further they exist separately. They must exist separately, if the species does; for the genus is described as existing apart from the species, and if the genus does so the differentia must do so too.’
18. mpdtepd y’ ova Kal pépy is a loose accusative absolute, for which cf. Kiihner ii. 2. § 487. 3».
21. Bz. objects to et’ (which purports to introduce a new argument) on the ground that the priority of genus and differentia to species has already been used in the previous argument (]. 18). But Alexander affords no support to Bz.’s suggestion dvadopa éorar, dru, except that he seems not to have read «79. Bz.’s objection is removed if we treat kat rovTo . .. diaopd (17-21) as parenthetical, and <6 dr, &c., as answering to rpGrov pev, 1.15. The second argument then is: Genus and differ- entia are prior in being to species, and what is prior to another is not destroyed by its destruction and therefore cannot be its definition.
22. dvtavaipettar. Aristotle has dvravaipeois in a somewhat similar sense (Zop. 158 33), but his usual word for this relation is ovvavat- pew (e.g. K. 1059) 38).
Alexander had érz instead of émeita (or érera dé) ei, and treats ém... . e€ oy (I. 23) as a Separate argument. But so taken it is exces- sively obscure, while, if we read «i, this clause offers a suitable protasis to what follows. ‘ Again, if the elements of Ideas are Ideas (as they must be, elements being simpler than their compounds), it will follow further that they must be predicable of more than one subject.’ Asc. seems to imply such a connexion (dvéyxn otv karyyopeicbar KrH., 443. 12). The history of Alexander’s reading is doubtless as follows. The first part of ANTANAIPEITAIEIL EITAEI was read as ANTANAIPEITAIETI, and the last part omitted by haplography.
23-27. The argument is: ‘Genus and differentia must be predicable of more that one subject. (.... They cannot form the definition of a single subject.)’ How, Aristotle asks, can genus and differentia merely by combination with one another lose this character of ‘ com- monness ’?
25. Ts yywpioOnceTar; sc., the Ideas being usually apprehended by generalization from several particulars. The Idea is for Plato what answers to a common name (Rep. 596 A 6).
27-2, év tots didiors goes (as Asc. sees) not with ddvvaroy épica- oGax but with AavOdver. It isnot impossible to define eternal individuals any more than other individuals, but the impossibility is less obvious, since the objection arising from the fact of perishability (1039> 27—
1040 7) is not present in their case. Again, 74 povaxad, things which are the only instances of their kinds, are not specially impossible to define, but the impossibility is specially hard to detect. Definitions may be (1) unduly narrow (29~33) or (2) unduly wide (33-> 2). (1) In the definition of A as BC, B or C may be irrelevant. If there are several instances of A, probably there will be some in which B or C is absent and the badness of the definition will be noticed. But where there is one instance only, the badness may easily escape notice (cf. 1036” 6). (2) If you define an individual A as BC, then if there are like individuals the definition will apply to them also and will be seen to be a bad definition of A. But if, like the sun, A is unique, and there are no other things to which BC is applicable, you may easily not realize that there mzght de other such things, and that you have defined not ‘the sun’ but only ‘sun’.
Definition of the individual is inevitably liable to both the errors here described. (1) You may be sure that certain attributes are not essential, but since you do not know the whole history of the individual you cannot be sure whether certain others are essential or not, (2) No series of marks will exhaust the nature of the individual, since every series of marks marks off not an individual but a kind. Thus the individual is indefinable, and if Ideas are individuals, they are indefinable.
27. domep otv elpyrat. Aristotle has not made the remark in ques- tion before, but he has treated the impossibility of definition as less obvious in the case of eternal entities, since he has given several argu- ments to prove it (8-27) over and above the general arguments against the definition of individuals (1039 27104087). Cf. esp. 1040717,
ZI. vuxtixpudés, apparently a coinage of Aristotle’s on the analogy of Parmenides’ vuxripats pas for the moon (fr. 14).
32. 7% pavy. ded may very well be understood without being actually inserted. Asc. has it in his interpretation, but there is no reason to suppose that he read it.
by-2. @\N fv... Zwxpdtns. I.e. the sun though unique is as much an individual-member of a class of suns as Cleon is a member of the class of men.
2-3. émel 81d Ti. . . iS8€as; is tacked on loosely. ‘ If the definition of the individual (and therefore of the Idea) is not open to these objec- tions, why do the Platonists never produce a definition of an Idea?’
Two wrong views about substance (ch. 16).
1040” 5. (1) Most so-called substances are potentialities—the parts of animals (which do not exist separately, and even when separated
218 : Commentary
are merely matter) and the elements; they are not unities but merely aggregates till they are fused into one.
10. One might suppose that the parts of living things and the corresponding parts of the soul are both, existing both actually and potentially, because living things have sources of movement in their joints so that some animals live when divided. Yet they exist only potentially when they are united by nature—not by force or by adhesion, which is a malformation.
16. (2) Since the meaning of ‘one’ answers to that of ‘ being ’, and the substance of what is one is one, and things whose substance is one are one, unity or being cannot be the substance of things, any more than being an element or principle can be so; we have still to ask what the principle is.
21. ‘Being’ and ‘unity* are more substantial than ‘ principle’ or ‘element’ or ‘cause’, but are not substance, because (@) they are common, while substance belongs only to itself and to that which has it. Further (4), one thing cannot be in many places at once, while what is common can. No universal, then, exists separately from particulars.
27. The believers in Forms are right in saying the Forms exist apart if they are substances, but wrong in saying that the one in many isa Form. Because they cannot say what the eternal substances are that exist apart from sensible particulars, they make them the same in kind as the perishable things which we do know, merely adding ‘itself ’,—‘ the horse itself’, &c.
34. But, even if we had never seen the stars, they would have been eternal substances apart from the things we knew; and so, even if we do not Know what eternal substances there are, yet there must be some.
1041® 3. No universal, then, is a substance, and no substance is compounded out of substances.
In this chapter Aristotle abandons his own fourfold list of the things that might be considered to be substance (ch. 3 ad znit.), and returns to criticize the popular views about substance expressed in ch. 2 according to which the parts of living things, and the four elements, are among the things that have the most obvious claim to be substances.
1040” 6, Surdpers eiot, are not actual substances but components capable of contributing to the life of the whole body.
6-8. ‘For none of them is separately existent ; and when they are separated from the living body, then too they are existent, all of them, only as matter” The hand while it is in the body is not separately
existent, and therefore not a substance but simply a material con- stituent, isolable in thought, of a substance. And when removed from the body, then too it is not a substance, with an activity of its own, but a mere collection of épovomepy, skin, bone, and flesh. Cf. 1035 17, G. A. 726 22. 3
8. Kat yf Kal Tp kat dnp. Alexander and Bz. take these words closely with &s vAy, but a comparison with 1028 9, 10 shows that they are co-ordinate with rd te pdpia tov Cawv. Bz. has to suppose that re is answered by paAuora & (1. 10).
8-10. obey yap .. . €v, i.e. a part ofan animal body, e.g. a hand, is unified into a genuine whole only by the form of the living body to which it belongs—otherwise it is simply a collection of diverse materials, cf. De Gen. ef Corr. 321» 31 di Kal reOvedros paddov av ddgeev elvan ere cdpE Kal dorodv 7} xelp Kal Bpaxiwv.
Q. swpéds, which is much better attested than 6 dppos, is Aristotle’s regular example of that which has not organic unity (1o41? ra, H. 1044° 4, 104529, M. 1084) 22), while dppds does not seem to be used in this way. The latter reading is no doubt due to rep6¥, but both the plural airév and the usage of zpiv show that zpiv 7 Krd. goes with oidey yap airav ev éotw, Not with otov cwpds. mérrew, a regular Aristotelian term for maturing or working up (cf. Bz. Zndex 5902 61— 591% 1), is applicable both to the elements (cf. A. 989% 16) and to the parts of the body.
II. mdpeyyus dpow ylyveoOoa. Alexander's interpretation pddiora 8¢ trodd Bor dy Tis TA TOV ewdywv pdpia, ... dpolws 8& Kal THs Yoxns, mapeyyus Tod Kal évepyeia. peTa TOU OAOV ovTa ovoias aira A€yew Kal duvdpet (535. 27), interprets wdpeyyvs in an impossible way. So does Bz.’s ylyvecOan oxed0v dvta dudorépws Kat evredcxelg. kal duvaper (the comma being omitted). mdpeyyvs does not mean cyxeddv. Like ctveyyvs, it is used adjectivally rather than adverbially, and (where it does not mean ‘near’ in place or time) means ‘closely related’ (Top. 167° 5, G.A. 76927, Pol. 127120). Probably the best translation is ‘ above all one might suppose the parts of animate things and the parts of the soul nearly related to them to turn out to be both, i. e, existent both actually and potentially’. An alternative translation is: ‘One might be specially tempted to suppose that the parts of the animals come-to-be more or less on the same level of being (rdpeyyvs) as the parts of the soul, thus possessing, both taken together, a being which is actual as well as a being which is potential’.
12. dvra kal evrehexeta kal Suvdpet, i. e. existing actually, since they can initiate movement by themselves, and potentially, since they are absorbed in the life of the whole soul and the whole body.
12-13. T@ dpxas ... Kaprrats. For the ball and socket joint as the bodily instrument of motion cf. De An. 433” 19-27, M. A. 698? 16-? 7.
1Z. 816 eva LGa Srarpovpeva Ch, i.e. some insects and many other animals doa pa Cwrixd diav <iot, e.g. tortoises, as well as some plants (De An. 411? 19, 41316, PN. 467°18, 468825, 479°3, PLA. 68245, » 30, 1. A. yo7b 2, G. A. 731% 21).
220 Commentary
15. ouppdcer. Aristotle’s use of this word is rather puzzling. Sometimes it means complete normal organic union (Phys. 2139, 224223, A. 101422, K. 1069212, A. 1070411); sometimes as here it is applied to abnormal unions (e. g. atresiae, G.A. 773214, 16, 25, ‘ Siamese twins’, ib. 4, 8). Insects are one cvppioer (De Luv. 468° 9) ; the parts may be said to exist actuallyjust because the whole is not fully one.
16. Td éy héyerar domep kal 7d dv, i.e. these are the widest of all universals, common to all things whatsover, B. 998 21. They cannot be the substance of things, for then all things would be one.
21-22. paddov péev ody. . . aitiov. One of the reasons which Alexander gives for this (536. 35) is doubtless what Aristotle had in mind—that being and unity belong to a thing in itself, while it is a principle, element, or cause only in relation to something else.
25-27. ét. TO év ... xwpts. ‘Further, that which is truly one cannot be in many places at once, while that which is common (like being or unity) is in many places at once. No universal, then, can be an individual existing apart from its particulars.’
33. Tavtas, as if Aristotle had said rats POaprats in the previous clause.