8. dwopodct twes. Pythagoreans, says Alexander; and Aristotle’s expression dvdyova1 avra «is Tovs apibpovs (1. 12) (coupled with the distinction between these thinkers and the Platonists, ]. 13) shows that he is right. Cf. Scholia on Euclid, p. 78. 19 Heiberg, of d IIv@ayé- pelou TO mev oneiov avdXoyov éhapBavov povdd., dvad« de THY ypappav. The representation of the point by 1, of the line by 2, of the triangle by 3, and of the tetrahedron by 4, goes back to Philolaus ( Zheol. Ar. p. 62. 17-22). Alexander’s note (512. 20—513. 3) preserves some information about the Pythagorean theory which he probably derived from Aristotle’s lost work Ox the Pythagoreans. The number 2 was, according to them, 76 mp@rov duacrarov, the first product of the di- | remption of the unit, and, quantity being no part of the essence of the line but the matter in which it is embodied, the line should be defined not as roadv ed’ ev diacrardv, quantity dirempted in one dimension, but as 70 mpétov duacrartov.
Zeit. TOCOY 3532 203
13-17. It seems clear that 74 eiSos THs ypappyis is opposed not (as Alexander, Asclepius, and Bz. take it) to r#v dvdda but to abroypaypyy, and that éyva ev ydp, &c., explains the position of those who said that 2 is the form of the line but not the ‘line itself’, The distinction between avToypayyn and 76 <idos THs ypappys is doubtless peculiar to one of the later forms of Platonism ; ot j<v probably includes Plato himself. The two views) are put neatly in H. 1043" 33; the question is whether the line is dvds év pajxer OY duds.
18. Step kat Tots Nubayopetors ouvéBaivey. Thus for instance they identified 4 both with friendship and with justice and with many other things. Cf. A. 987% 27.
19. Whether we print airdedos or airé eidos, adro probably goes with eidos in the sense of ‘Form-itself’ or supreme Form. wévtev = ‘ of all Forms’.
20. Kkaitor ottws ev méyta éotat, since the nature of things depends, for the Platonists, on their Form.
24. i tapoBody * emt Tod Ldou, i. e. the comparison of the relation . of flesh and bones to humanity with that of bronze to circularity (Il. 5-7).
25. LwKpdrns 6 vedtepos, a Socratic and a contemporary of Theae- tetus, mentioned in Zheaet. 147 c, Soph. 2188, Epp. 358d. He is one of the interlocutors in the Polzticus,
30. éxévtwy mos is really irrelevant. Aristotle is insisting that the definition of man must mention the parts of the body. This reminds him, however, that it is not enough to mention the parts of the body ; it must be specified that they are in a certain condition, i.e. alive. We must not forget the matter; but we must equally not forget the form, Wux7 or vitality.
BI. a\X’ q. For the usage cf. 10387 14 n.
32—1037° 5. Alexander thinks this section may have originally gone with the previous discussion on the difference between the relation of the circle to.its semicircles and that of the syllable to its letters (1034» 24—1035% 17), and have been separated by Eudemus ; and Bz. also thinks it out of place. But it seems to be quite appropriately placed. Aristotle has just rejected the comparison of sensible things like ‘the animal’ to intelligible things like the circle (1. 24), and has insisted that in the definition of ‘animal’ its sensible materials must be mentioned, while in that of the circle its sensible materials (bronze, &c.) must not be mentioned, since they are accidental to it. This naturally leads to the further question about the circle, why are its parts in another sense, the semicircles, not mentioned in its definition. The reason is not that these are sensible materials and therefore irre- levant to the circle, which is an intelligible. They are not sensible. But they contain material none the less—intelligible material, and therefore are not parts of the universal ‘circle’. From this Aristotle, naturally enough, returns to the case of ‘the animal’, in 1037 5.
The reasoning, however, is inconsequent. If ‘animal’, though a universal,’ cannot be defined without the mention of its necessary material, the circle cannot be defined without the mention of 77s
Bo COMMENTARY
necessary material, though the material is in this case not sensible but intelligible. Aristotle is right in saying that the semicircles are not mentioned in defining the circle, but the reason is not that they are matter but that the definition would be circular. The circle cannot, in fact, be properly defined without referring to~its ‘material’, viz. space.
35. Eotar yap Ay eviov cat py aieOntav. For try vonry cf. #9 n. ~ 1037%1-2. The fuller form given by A» runs rather more naturally than the shorter form given by the other manuscripts, and I have accordingly kept it, though it is quite possible that A»’s reading may be, as Bz. suggests, merely derived from Alexander’s interpretation.
4. Koptoxos, cf. A. 1015) 172.
ei pev Kat H pux Loxpdtys. The insertion of Swxparys, with AP, makes the meaning clearer. As Aristotle had suggested (10368 17) that ‘animal’ may be identified with its form, soul, he here suggests that Socrates may perhaps be identified with his form, his soul. If he may be identified with his soul alone, as an allernative to being identified with his soul+his body, then ‘Socrates’ is durrov, an ambiguous term. For the general sense cf. 10362 16-25, H. 1043) 24.
g. It seems necessary to insert 16, which is read by the Aldine edition, though probably only ex conzectura.
doTep Td ka0ddou [Te] kal Td Ka0” Exacroy, ‘as is the universal (man), so will be the individual’, i.e. a unity of form and matter (cf. 1. 6). There is some probability in Apelt’s suggestion that re has arisen from an abbreviation of ovrw. The sentence would read rather more naturally with otrw, but cf. 1038> 2 f. ve in any case must be wrong. Christ is perhaps right in reading a comma after awA@s. womep..- éxaotov then means ‘ the individual corresponding to the universal ’.
10-20. Jaeger regards this (Avis/, 206 n.) as a section added later by Aristotle, when he began to view the discussion of sensible substance . in Bk. Z as preliminary to the discussion of insensible substance in A. Cf, 1029» 3-12 n.
II. tig GAN, i.e. a quasi-material principle (such as the great-and- small) present in incorporal things. It seems impossible to supply ovaia, as one is tempted to do, as the substantive understood with ts
aXAn-
12. okewréov Uotepov. ‘The reference is to Bks. MN.
13-14. TovTou yap... Svopife. For other indications that the
treatment of sensible substance is preliminary to the true business of metaphysics cf. 102.9% 33, » 3-12.
16. 08 ydp pdvov wept THs UAns KtA. Cf. De An. i. 1.
17. Tis Kata Tov Né6yor, Sc. ovaias.
émt may be defended by a comparison with the passages referred to in Bz. /ndex 268. 31-44, but Brandis’s conjecture éz is very likely right.
20. oxeTttéov Uortepoy, H. 6.
21-22. Ti... eipyrat answers to ch. 4, 22-33. 81 ti. . . UAy-to chs, 10, 11 (Il. 30-32 to ch. 5), 33. 61... 7 to ch. 6. This summary contains no reference to chs. 7-9, and confirms the view that these belong originally to a distinct treatise. Cf. 1032% 12 n.
Pee TO20" 354039) 6 205
27. peta pev yap THs UAns odk €or. Yet Aristotle has said (1036 » 29) that the definition of man must mention his material parts, Of course a definition must not mention prime matter, since that is aopirrov, i.e. nothing definite can be said about it; but in certain cases the proximate matter must be mentioned.
ZI. dis yap év TovTous Smdpfer H fits. These words appear quite irrelevant in this context, and seem to be due to a copyist who had 1030? 32 in his mind.
bs. The simplest emendation of ovdé is 068 et. ‘ Nor is a thing the same as its essence if the thing be a compound of a subject with an accidental attribute.’ Cf. 10318 19-28.
What constitutes the unity of a subject of definition P (ch. 12).
1037” 8. Let us discuss definition in so far as it has not been dis- cussed in the Analytics. The question there stated is of use to.our inquiries about substance, viz. the question why that, the account of which is a definition, is one.
1g. Why is ‘two-footed animal’ one and not two? ‘Man’ and ‘white’ are two when the one does not belong to the other, one when it does ;
18. but in ‘ two-footed animal’ one element does not share in the other ; the genus does not share in the differentiae, else it would share in contraries at the same time. Even if it does share in its differentiae, the same difficulty occurs, since the differentiae of man are more than one—possessed of feet, two-footed, wingless. Why are these one? Not because they are present in one genus, for then a// the differentiae that belong to a genus will form a unity.
24. But the elements in definition mus? be one, since substance, the subject of definition, is a unity, a ‘ this’.
27. Let us examine, first, definition reached by division. There is nothing in definition but the first genus (e.g. animal) and the differentiae; the lower genera are the first genus + the differentiae (e, g. two-footed animal).
33. The number of the elements of the definition makes no differ- ence; let us reduce them to the genus and one differentia.
10388 5. Now (1) the genus does not exist apart from the species, or, if it does, exists only as matter; therefore definition is the account consisting of the differentiae.
g. But (2) we must at each stage divide by the differentia of the previous differentia; we must divide ‘ possessed of feet’ into ‘ cloven- footed ’ and ‘ whole-footed’, not into ‘ winged’ and ‘ wingless’,
206 Commentary
15. We must proceed so till we come to the indivisible species. There will then be as many kinds of foot, and of footed animals, as there are differentiae. The last differentia will be the substance and definition of the thing. ts
20. If we mention the earlier differentiae as well, we shall repeat ourselves. ;
25. If, then, we take a differentia of a differentia, one differentia— the last—will be the form; but if each differentia is accidental to the previous one, there will be as many differentiae as there are steps ‘in the division, Definition, then, properly consists of the last differentia.
go. If we change the order of the definition, putting ‘two-footed ’ before ‘ possessed of feet’, the latter is evidently superfluous; but there is no order in substance (therefore ‘ possessed of feet ’ must be superfluous even when it stands first). So much for definition by the method of division.
The unity of a definition, and of iis subject, according to Aristotle, lies in the fact that the genus and the differentiae have no existence apart from one another, nor have the successive differentiae. The genus is merely the ‘ matter’ of the definition, and each differentia the ‘matter’ of the next. He accordingly condemns definitions in which any of the differentiae are accidental to the genus or to one another.
The discussion is carried further in H. 6.
1037” 8. For the discussion of definition in the Analytics cf. An. Post. ii. 3-10, 18.
Q. év éxetvors, Av. Post. 92° 29. The dzopia is there raised but not answered.
II. 06 tov Aéyov Spiopdy elvat payev. The distinction between Adyos and épucpos has been drawn in 1030? r4.
14-21. The obvious intention of this argument is (1) to show how man and white form a unity, when they do so (Il. 14-18) and (2) to point out that genus and differentia cannot form a unity in this way (il. 18-21). Bz. argues that the mode of unity referred to in the first section (kara mdéGos) is not the same as that referred to in the second (xara peOcéw). He thinks therefore that Aristotle is arguing that genus and differentia are not one ecther xara maOos or xara wéOcéw, and that it is merely by carelessness that he frames the sentence in a way which suggests that these modes are the same. But in 1030813 peroxy is used in this connexion as synonymous with rd@os. If weréxew be so used here, the argument can have its natural meaning. In 14-18 Aristotle describes a unity xara peroynv Kal rdOos, and in 18-21 he shows that the definition is not a unity of that sort. It is true that in the Zopzcs (1214 11 and elsewhere) 76 peréxew is defined differently, but that account of it does not seem to be in Aris‘otle’s mind here ; he is using perexyew in the Platonic sense alluded to in H. 1045° 14-20, by—9. The argument of !]. 14-21 may be put thus, A unity xara