The Origin of Atomism

Bury, R.G. Saum ek The Origin of Atomism ab The Classical Review, XXX - 1916 - In 1-4 nm À Pe Pur Pacs #G. Sarga Sabor # Ens Pand. Classical Review iew {The}, The‘, vol, XXX. 4 No origi de :l'atomismeiR. GBury). |jR. 6 syshane1-4. de Les dovtriniss 0. XX. N°4. L'origine e l'alomisme kin alomisliques ont des rapporls élroil s avec le en N gore Pius qu'avec loul aulre systéme: si elles se rap . Poe ’ pro hese doe ines. des Eléate , s, c'est dans la inesure où ces dernieres Ee t es rapporis avec le système pvlhagoricie a qui tes asser preuvcompt ; dans un texte dd'Aris e de ecelte assertion Arislo le ‘de ide Caelo lole IIL, 309° 303° 8: Cacllil. o | Alomistes ’ parmi philosophies qui rallachent tout aux Nombres : la termi | nolovie lesles philos preuve, De plus a gie des théoriciens de l'alomisme en est une autre des parti . sans be du5, moer ‘Alotmen istes et et Pytha Fr thago goric ricie iens ns se disli isli nguent épalement Qualité des wine i du pluralisme ; ils adopleul également la qui se combine I cl te Caractère invariable des monades primi tives ie com M? er an pas chimiquement mais mécaniquement, Il est vrai 1 : Weremment les monades ; ; pour to me Pytha pour les €83 qu'en renee enciés identi pointsles ques,8, ;n e sont dilfér , tous forma nt goric des iens, figureles s ; : ‚les: atomes diffèrent de s le début 5 ais c e i iv gence de vue ne parait pas irréductible . € ' man delle vers 173568. BURY R.G., The origin of Atomism : PCPhS 100-102 1915 18. Bury défend la thèse de Gomperz selon laquelle il faudrait probanon de blement faire dériver l’Atomisme du Pythagorisme ancien, et ales princip des examen Un encore. croit le PÉléatisme comme on le rer démont 4 cherche es Atomist des logie termino la de et doctrin bien-fondé de cette position.

The Origin of Atomism R. G. Bury Stable URL: http://links.jstor.org/sici?sici=0009-840X%28191602%291%3A30%3A1%3C1%3ATOOA%3E2.0.CO%3B2-G The Classical Review is currently published by The Classical Association. Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/journals/classical.html. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is an independent not-for-profit organization dedicated to and preserving a digital archive of scholarly journals. For more information regarding JSTOR, please contact support@jstor.org. http://www.jstor.org Mon Jan 29 08:11:35 2007

Review FEBRUARY, 1916 ORIGINAL CONTRIBUTIONS THE ORIGIN OF ATOMISM. For a variety of reasons it is not altogether easy to make out the origin and filiation of the Atomictheory. For one thing, although we know pretty definitely the main tenets of the doctrine of Democritus, Leucippus, ‘the white horse,’ remains a dark horse. The standard authorities are divided among themselves, so that while we have Windelband, on the one hand, declaring that ‘ Atomism is an offshoot of Eleaticism’ and that ‘ Leucippus stands in the most marked dependence on the Eleatic teaching,’ on the other hand we have writers like Gomperz and Mr. A. W. Benn protesting against this derivation of Atomism from Eleaticism. Briefly put, Gomperz? maintains (a) negatively that Leucippus was not really ‘a pupil of Parmenides,’ and that the points of contact between the two were mainly points of contradiction; and (5) positively, that ‘the raw material for the Atomistic theory, namely the twin conception of invisible moving particles and invisible vacant interstices, is derived partly from the older Ionians (Heraclitus, Anaximenes, and Anaxagoras) and partly from the Pythagoreans.’ Mr. Benn * agrees with Gomperz on the negative side, in so far as he lays stress on the way in which Leucippus ‘seems to contradict Parmenides at every point,’ and argues that the progress of thought ‘does not pro1 A paper read to the Camb. Philol. Soc. in November, 1915. 2 Greek Thinkers, E. Tr., vol. I, pp. 316 sgg. 3 The Greek Philosophers, ed. 2, pp. 31 sgg. NO. CCLV. VOL. XXX. ceed either by antithesis or by synthesis, but by almost insensible transition and suggestion.” But (b) on the positive side Mr. Benn stands by himself in his view that the atoms were first suggested to the scientific imagination of Leucippus by the observation of clouds of dust; and that, dust being earth in its finest form, Leucippus is the one and only philosopher who chose earth as his primary substance (and eluded Aristotle, we may add, in doing SO). I will not stop to criticise Mr. Benn’s view, but pass on at once to the constructive part of my observations. What I shall endeavour to show—in so far as so highly speculative a matter is capable of demonstration—is that Atomism is more closely allied to the Pythagorean than to any other system, and that, if Leucippus is to be labelled at all or grouped at all, he should be labelled as an independent Pythagorean rather than grouped with Pluralists or Conciliationists. The usual methods of placing him seem to me, as to Mr. Benn, examples of the ‘post hoc, ergo propter hoc’ fallacy. To begin with Aristotl—whatever he may say elsewhere, to suit his book, here is what he says about the Atomists in de Caelo III. 3032 8—rpórrov yap Twa Kal oùrou mrávra Ta Övra mowdoıv apLOpovs Kat EE àpiôu@v. Ie. the Atomists may themselves be numbered amongst the number- philosophers; they are Pythagorisers of a sort, if not actual Pythagoreans. Let us see how far this

is supported by the terminology of Atomism. Take first the term elòos or idéa. A work mrepl üÖeôv is ascribed to Democritus, and it seems quite likely that the term goes back to Leucippus. Prof, A. E. Taylor (Var. Socr., p. 248) would render the Democritean title, Concerning Primary Bodies, and he argues at length that ‘eidos and idéa, wherever they appear as technical terms, alike in rhetoric, in medicine, and in metaphysics, have acquired their technical character under Pythagorean influence’ (p. 257). The term ëy«os—which is certainly Epicurean in the sense of ‘ atom ’— may probably go back to Democritus and Leucippus. To the Atomists a ‘magnitude’ can only have meant an atom or glomeration of atoms. I conjecture also that the use of the term was common with the mathematical philosophers, from its occurrence in Theaet. 135 A, Parmen. 164 A, and in Zeno’s arguments on motion as reported by Aristotle. 'Avaryen as applied to the atomic law of motion—the dives or Sivn of atoms spinning ## vacuo—is another conception which seems to link Atomism with Pythagoreanism. We find this combination of Necessity with rotatory motion in the Platonic Myth of Er, where the spindle with its whirling odovédvrot revolves Ev Tols ris ’Avdyans yovacw (Rep. 617 B). Adam, in his note ad loc., says ‘ the notion is probably Pythagorean, for Parmenides, who attaches himself to the Pythagoreans in this part of his system, speaks of a central ’Avdysn as the cause of all movement and birth.’ We may note also arworoun as an Atomistic term, and compare it with the technical use in geometry to denote ‘section,’ or cutting off of segments (Meno 85 A), and with Parmen. Fr. 3 où yap amotunfe TO Eöv Tod Eövros &xeodaı, where it is natural to detect a controversial reference to Pythagoreanism. Another term which, to my mind, has a Pythagorean flavour is oxjvos, which Democritus is fond of using for cpa. Where was he more likely to pick up the expression ‘earthly tabernacle’ than from those precursors of Paulinism, the oöya-onua sect? A curious word of the Atomists is mepimarafıs, in the sense of auumAorn, denoting the ‘embrace’ or mutual entanglement of atoms—a word which, by the way, is not given in Liddell and Scott, though éurdAayya is quoted. I conjecture that it too may have Pythagorean associations, although I admit that my ground is not of the most solid. In Phaedo 82 D we have the phrase éxeivot ols Te pédet THS éavrôv Vruvijs GANG py TOpaTL TAATTOVTES Cot in a passage which savours strongly of Orphicism or Pythagorism (preaching «äbapois wuyijs, and the body a dungeon). Here the right reading is maxarrovres (Diels), which combines the sense of close contact and contagion (aluarı Hom.) * I may add here two further notes on Atomistic terminology which are not strictly relevant to my main contention. One of the three differentiae of the Atom, as given by Aristotle, is duadıyn, which is explained to mean ra&ıs. I am inclined to think that the word (inappropriate in itself)? is a blunder for drain (= dıadenıs), since in Met. 4. 19 dıadecıs is defined as 4 tod &xovros uépn Taëis, and Democritus Fr. 9 (Diels) uses d:aOy«n for Òuddeois in the phrase Kata cwparos Òradijknv. Moreover there are variants for Òraduyú in the MSS. and Ae. 1. 15. 8 calls the same differentia Ssatayyn. Further, the examples given of the differentiae in Met. I. 4 run thus—Grapépez yap To uev A rou N oynpate (=pvope), To de AN rod NA rafa, To de Z tod N Oécer. That, at least, is the reading formerly current; but later editors seem to favour Diels’s correction ro 6é Æ rob H déc. Here the question that suggests itself is—why are these particular letters chosen rather than any others ? I have never seen this question put or answered; and even to attempt an answer is perhaps super-audacious. However, what I venture to suggest is 1 Cp. Aelian V.A. 3. II, of mepumarnrukoi gacw ped' muépar Onrevovoar rv puxny TÔ o@parı mepımAereodaı. 2 Ar. uses the vb. of ‘continuous touch’ (see L. and S.). 3 S:aOny7, see L. and S.

this—that the letters here used form part of a familiar term of the Atomistic School, a term connected with the matter illustrated, viz. vaorov = mAïñpes (of which more presently). Admitting this, how does it bear on the textual question? Obviously, all the illustrative letters will most naturally be taken from the same word, so that H, whether upright or on its side, cannot be right, and the letter most likely to be right is the N which appeared in the old text, and which is often confused with H in uncials. My second note concerns another technical term, which may, I think, fairly be ascribed to the earliest Atomists. Aristotle, de An. I. 2 (compare de Caelo III. 4) uses the word mavotreppia with reference to Leucippus. Does this imply that Leucippus spoke of his atoms as omépuara? I think it may, since Lucretius is evidently Latinising an original Greek Atomistic phrase when he calls the atoms semina rerum. Moreover, we may discern a Platonic pun on oéppara in the description of the Atomists (Soph. 247 D) as avroxyOoves Te Kal omaproi avdpes—‘ seedsmen.’ But this is by way of digression. To return now to the main track. There is, I would contend, something of truth in what Diogenes Laert. says (Diels, Vorsokr., p. 366) of Democritus that he Ilvdayopırav Tıvos arodoaı and PirorAd@ ovyyeyovévat. For besides the verbal connexions which I have tried to trace, it is fairly clear thatintheir main tenets the two schools, Atomists and Pythagoreans, agree. Firstly, as distinct alike from monism and pluralism, both are essentially dualistic—each assuming the reality of a pair of concrete opposites, Solid and Void, or Figure and the Figureless. And, secondly, both agree in the invariable character of their primary monads, which do not alter chemically, but only combine mechanically: both alike reduce quality to quantity, and build the world of idéai or oyyuaTa, compounding masses out of concrete minima. There remains, however, a characteristic difference in the conception of the monads; for 1 Cp. Iambl. a2. Diels 54°. 3 while the Pythagorean units, or points, were all alike, and only when pluralised into triangles, squares, circles, etc., acquired shape-variety, the unitary atoms possessed shape-variety from the start; so that the atomic unit might be described as a kind of telescoped or compressed form of the geometric oxîjpa or multiplied unit of the Pythagoreans. What is the explanation of this characteristic difference? I suggest that it is to be explained in connexion with the term vaorov, as part of a larger fact. The peculiarity of the word vaorov caught the attention of Galen, who writes (viii. 931) dprov pév yap Twa vaaròv êkdhovv, où um GAO Ye TH cpa mpos adrôv (de. TÔv “EAAnvor) oUTws @vopacpévoy émioraua. Without going so far as to say that Leucippus was originally an dproxomos—familiar with grains and floating motes of meal, and sieves, and threshing-floors, and whirling mill-stones and kneadingtroughs (all which seem to figure in Atomistic literature)—I would suggest that vaorov really does convey the central notion around which the whole doctrine of Atomism was built up. The real, To öv, presented itself to the scientific imagination of Leucippus as primarily ‘the kneaded.’ J.e. he derived his physical theory in the first instance from a study of the phenomena of compression and of porous or spongy matter. It is the study of porosity that gets us at once to the double notion of “the Full and Empty,’ and it is because these two seem equally fundamental in Leucippus’ theory, and because also he, as an Ionian (whether of Miletus or Abdera), must be presumed to have been primarily occupied with the study of physical phenomena, that I suggest this as the origin of his view. In support of this suggestion I may again quote Aristotle, Phys. IV. 6, where among various Democritean arguments in favour of the existence of Void he mentions this—öru gatveras Eva cuvidvta Kai mdovpeva. My supposition, based on the term vaoröv, is that this argument from wiAnoıs was the original argument of Leucippus, and that it was in this way—from these physical studies—that he first arrived

at what Gomperz calls the ‘twin conception of invisible moving particles and invisible vacant interstices.’ Then, having once got this twin conception, he expanded and developed it under the influence of the kindred dualism of the Pythagoreans. But with this notable difference, that he applied his vaaróv-idea afresh to the oxnuara of the Pythagoreans, ‘kneading’ them into impermeable solids, while retaining the characteristic shapes (triangle, polygon, etc.) in the solid form: ve, briefly, he telescoped or stereoscoped them. In conclusion, what I would emphasise is this—that Atomism in its origin had probably little or nothing to do with any ‘metaphysical conflict,’ and nothing to do with Eleaticism, except in so far as Eleaticism itself may be termed an offshoot of Pythagoreanism. The only dependence of the Atomists on Parmenideanism proper is, as Mr. Benn and Gomperz urge, the dependence of contradiction; which means that if the Eleatics had never existed, Leucippus might equally well have preached his Atoms and his Void; and if, as tradition states,’ ‘he listened to Zeno,’ this argues no more dependence than may be ascribed to Sir Edward Carson when he listens to John Redmond in the House. R. G. Bury. 1 Cp. Galen ap. Diels 545. THE HOMERIC HYMNS. I. THE emendation of the Homeric Hymns with their manifold and manifest errors of tradition, often defended by dull and foolish scholars with the pretentious dogmatism that one corruption of the text is proved sound by the production of another equally bad; or, in other words, that two blacks make one white, continues to attract periodically the efforts, the praiseworthy efforts, of active and inquiring minds. It is well that it should be so, and therefore I welcome with both hands the paper on this subject by Mr. R. J. Shackle in the September number of the Classical Review of this year and the earlier one by Mr. Evelyn White (C.R. Nov. 1914). It is now nearly twenty years since I wrote a review of Monro’s Homeri Opera et Reliquiae (C.R. Nov. 1896), and there suggested a great many textual changes in the Hymns that might with advantage be adopted. One or two of these I would now revise, and notably one that called forth a considerable amount of discussion of what is called Wernicke’s law. The passage is Hymn. Dem. 268 : ep Anuijrnp Tuudoyos, M TE péytoTov äfavärois Ovntoiai 7’ Övevap Kal Xappıa TETUKTAL. To make a tolerable scansion for the degenerate second line I proposed to read adavaroıs Ovntoici T’ Överap Kai mroXd xapıa, with possibly dfavárov (Stoll) for a linguistic and metrical reason. Prof. Platt courteously enough raised the question whether òverap could rightly stand in this position. I took the opposite view, and suggested that the objection was based on a later metrical usage imposed on Homer. Mr. A. Shewan (C.R. Sept. 1915) perhaps goes too far in treating this suggested explanation as absolute dogma. May I refer him to remarks on this question in my review of Leaf's Iliad, xiii.-xxiv. (C.R. 1902) ? In any case I would now treat this passage somewhat differently. The objection to rervrraı is certainly valid. It will be good Greek here, when ‘I who is’ is good English and not before. The bit of black with which this is whitewashed is the very flagrant and obvious interpolation P. 249-51 q.v. Of xxix. 4 nothing need be said. It may be observed first of all that in the early epic period a verse could not begin with ddavaroıs Ovyroîot te. If the tradition be right as to first word, then it must be in the form d@avaro-