love and strife in empedocles cosmology

8/3/2019 Love and Strife in Empedocles Cosmology

http://slidepdf.com/reader/full/love-and-strife-in-empedocles-cosmology 1/41

Love and Strife in Empedocles' Cosmology

Author(s): F. SolmsenSource: Phronesis, Vol. 10, No. 2 (1965), pp. 109-148Published by: BRILLStable URL: http://www.jstor.org/stable/4181763 .

Accessed: 18/12/2010 18:50

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp. 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/action/showPublisher?publisherCode=bap. .

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 a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

BRILL is collaborating with JSTOR to digitize, preserve and extend access to Phronesis.

http://www.jstor.org

http://www.jstor.org/action/showPublisher?publisherCode=bap

http://www.jstor.org/stable/4181763?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp



http://www.jstor.org/page/info/about/policies/terms.jsp

http://www.jstor.org/stable/4181763?origin=JSTOR-pdf




Loveand Strife n Empedocles'osmology*

F. SOLMSEN

In Heracitus and Parmenidesassumptions which form the basis ofour interpretation are subject to frequent reexaminations andrevisions. With Empedocles matters are different. Here large

hypotheses have for a long time remained unchallenged and are now

near the point of hardening into dogmas. In particular the recon-struction of a dual cosmogony in his "cycle", originallya theory whichhad to contend with others, is now often regarded as established,treated as though it were a fact, and used as premise for furtherinferences. The only full scale interpretation of the evidence whichbacks up this theory is Ettore Bignone's Empedoclel;yet whatever themerits of this book, it can hardly be denied that in the fifty years sinceits publication we have learned many new lessons regarding therelative value of testimonies and fragments, the trustworthiness of

Aristotle's reports on his precursors, and other questions of vitalbearing on the reconstruction of a Presocratic system. A recent textbook which seeks to fit the material into the framework of two cos-mogoniesdoesnot inmy opinionsucceed nstrengtheningthis position2;

* Interpretations here included were presented to the Conference on AncientPhilosophy held at Amherst College in August 1964. After completing the paperI received through the kindness of H. Diller and U. Holscher a manuscript of thelatter's article: "Weltzeiten und Lebenszyklen. Eine Nachpriifung der Empe-dokles-Doxographie" (now published in Hermes 93, 1965, pp. 7ff.). It was

most encouraging to see how often we had reached the same conclusions. Aswe start from different &aopjLaxnd proceed along different lines, there is littleoverlapping of the arguments; all I could do was to shorten my paper in a fewplaces, while adding references to Holscher's study. The paper has been read andcommented upon by Alexander Mourelatos and Gregory Vlastos, and I havealso received helpful suggestions from Diller and from my colleague JuliusWeinberg. To all of them I am most grateful.lTorino, 1916. See esp. ch. III and Appendix 2.8G. S. Kirk and J. E. Raven, The Presocratic Philosophers (Cambridge, 1957)ch. 14. For the reason stated in the text I have more often expressed dis-agreement with Raven than with any other scholar favoring a dual cosmogony.

Raven is far less prone than Bignone to use "parallels" in other Presocratics forthe reconstruction of Empedocles' scheme. See also for the theory of twocosmogonies etc. Ferdinand Dummler, Akademika (Giessen, 1889) pp. 217ff.;

109



on the contrary it may be said that difficultieswhichwere essapparent

as long as the discussion confined itself to individual fragments orgroups of fragments become more visible when the entire scheme isworked out and presented. Perhaps the wisest course would be toadmit ignorance on crucialpoints. If I, nevertheless,prefer to offer analternative reconstruction - in essential aspects a revival of vonArnim's3 - my hope is that, whether right or wrong, it will serve agood purpose f it showsthat opinionscurrentlyaccepted arenot firmlygroundedin the evidence at our disposal. I have made no methodicalcommitment except to keep the KaEOp[uoLut of the discussion of InEp'L

cpu's . Similar or identical motifs, like the fundamental importanceof Love and Strife, the kinship of all living beings, are clearly presentin both poems but to argue from recurringmotifs to an identity orsimilarity of doctrine is nothing less than a petitio.There are too manyunknownfactors. The time interval may have been long or short. Thequestion of priority has not been settled.4 We cannot assume thatEmpedocles' mind was of a rigidly dogmaticcast incapableof respond-ing to new experiencesand impressions (nor can we know what theseexperiences may have been). What we do see is that his attitude to

"reality"differsin the two works. Surely the place for a comparison safter the reconstructionof the poems,not priorto or in the course of it.

Burnet, E.G.P. (4th ed.) pp. 234ff.; Cornford, C.A.H. IV, 566ff. (and elsewhereincidentally); Harold Cherniss, Aristotle's Criticism of Presocratic Philosophy

(Baltimore, 1935) p. 195; W. K. C. Guthrie, In the Beginning (Ithaca, N.Y.,1957) p. 42ff.; C.E.Millerd, On the Interpretation of Emp. (Chicago, 1908)pp. 28,45 and pass.; E. L. Minar, Phronesis 8 (1963), 127 ff.3 Festschrift TheodorGomperz (Wien, 1902) pp. 16ff. Other scholars who favored

a single cosmogony are Zeller, Die Philos. d. Griechen (5th ed., Leipzig, 1892)p. 778 ff. (cf. 6th ed., Erster Teil... Zweite Halfte ed. by Wilhelm Nestle, Leipzig,1920, 975ff.) and Tannery, Pour l'histoire de la science Hell. (Paris, 1887) pp.

304ff., esp. 308ff. See also Diels, SBBA 1898, 414f., Jaeger, The Theology of the

Early GreekPhilosophers (Oxford, 1947) p. 142. I realize that it might have beenof interest to trace some of the problems back to the form in which they werediscussed in the 19th century and in particular to compare earlier collections of

the fragments with that of Diels; for if my views are approximately correct, itwould seem that the arrangement of the fragments in Diels is not invariablybetter than that of Simon Karsten in the second volume of his PhilosophorumGraecorum VeterumReliquiae (Amsterdam, 1838).

4 For an important argument suggesting the priority of the physical poem seeWilamowitz, SBBA 1929, 643. Cf. also K. Reinhardt, C.P. 45 (1950), 170ff.

(= Vermachtnis der Antike, Gottingen, 1960, 101ff.).

110



The Workmanship / Love

Fragment 35 may set us on our road:

'ulrp Zy& 7U(XXvopaoGEXDZUaO[Xa< n6pov 4tVGV,

'OV Otp6?pOV X CEXec, ?,6you ?X6yov 0oXers'X?:L V ?7rp NXOCT [UE ?VpucVtXX 3Jo<xsZvov, Oetd Nr:ZxoqLev9v6p-rwvovxerope6voo8Evi;, 'v gi ,t DL?6tX'rrpO(P&LyyL yew1Y7TC,

5 EMvtjj oBe 7t&Vro auVepXzetOt 9v 6OVOV VLV,t,

Qux Xocp, aXO Oe?IAX uVla' ULeVJ&X?)oOev &XOCa.

TCt)Vae? re tULL5Op?eVwvX zr el"Vec [LUVpOCOv-1-(?V.

7rOXXO'A' nasXrc ?QXE XCpOLoCL6voLatvwX,Z6aa' 91 NeZxoC ?puxe 0?e'opaL.ov oV yap a,uF?cp

10 rCov n)&ov E`Ee'GrXeVe?r ?aztro 'rzppXa XU,XXOU,

OXXOC TrOC ev Xr eVe?0tpV i1exewV, roc 'r I?Xet.

6caov 8' OCMeVn=?X7po06oet, 'O6aovOctvC7?tL

L6pO)~L?XO,T7nOq O({Lz[kYpSOq 4~Lpo,7oq O4p[iA,.

ax14 8d Ov4I' 9pyov-7o r& tpLvFa&OovO&vo&r'tVOCI,

15 ~op& r a 7rptv OCXpqrOC,&cXXB ov'rx xe?1Oouq.

,rovU e ~Ltyo~i.6v&v xeZ&Oveo [?up(ocOv-qr6v,

VtOLL~Lq~aL ',rp4-, OOC3pcMaOac.

Returning to the "path of the songs" - evidently after an excursus -

Empedoclesat once defines the point of the cosmicprocess at which hefinds himself: Strife is at the bottom of the whirl, Love coming to itscenter.' Her power is on the increase and she manifests herself in hercharacteristicway. All things come together gv Lo6vovvax (5). Howeverthis gv has to be understood as the final goal of the developmentsinitiated by Love. For the time being we are still far from the conditionin which all things are "one".6 Neikos is keeping up resistance and

yielding only by and by as it withdraws to the circumference.7Weclearly have a situation in which the four elements are in part mixedand in part unmixed. The world condition here described includesconsiderableamounts of earth, water, air and (the heavenly) fire, each

6 B 35.3ff. For the "whirl" motif see below n. 45. The subjunctive in v. 4seems impossible (yiveto a-rpocp&XLyytVan Groningen).

6 In v. 5 I accept without hesitation Simplicius' reading &v'rj (= kv OL?X6'nyT&;

cf. B 21,8 auv 8' gliq &vDtX6-TJL); 8' is a reminder that this operation of TLX6rnqis familiar from earlier statements. 'r68cndivrxmight refer to the four physical

elements as present in the Cosmos but equally well to all things to be seen inthe world.7 See 9-12; the reading at the beginning of 10 is uncertain.

1ll



of them by itself - aloof as it were (erM'paLov) and holding out against

the pull of Philotes, yet there are also mixed forms which the poetidentifies as gOvea up(o OVIv.8 Without doubt all mixtures are"mortal", temporary formations and as such the opposite of theimmortal elements. The mortal beings which issue from the mixingprocess are in the final line said to be nvavrotovo naw aprp6aT.

Neither navrolcusnor Om5tme'aOatuggests of necessity creatures ofsuch strangenessas the Pouyeviwv9p67tpypxnd others that belong tothe second of the so called four evolutionary stages.9 After a periodcharacterizedby the existence of only the four elements, each in its

unmixed state, the first emergenceof living beings with their forms somanifold and ingeniously varied (in contrast to the monotonoussamenessof each element) is indeed "a wonder to behold".

Our present world certainly shows large compact aggregates ofearth, water, air and fire; it also shows, so to speak between them andparticularlyhere around the centre, a great variety of mixed beings,namely plants, land animals, fishes, birds, perhaps also stones. Wehave learned in the beginning of the poem that in our present worldcondition Love and Strife are both of them powerfuland that they are

contendingwith one another. B 35 is the only fragmentthat describesthe origin of such a condition. May it neverthelessrefer to a differentcosmicphase?We are not yet in a positionto form a judgment.

Simplicius n his commentaryto de caelo10 efersfragments35 and 71to the same world condition, to wit the present, v5v. If we read his

argument carefully, we can see that he felt surer in the case of 71.11

Our next step should therefore be to compare these fragments inde-

8 Vv. 7,16. V. 7 has been called into question ("Ubertragung aus Vers 16"

Diels-Kranz) but the words &?Lewr' and xepacLoldvoLaLv in 8 make it very desirable

to keep the line in this place. Empedocles' habit of repeating lines (or groups oflines) is well known. See also B. A. Van Groningen, La composition littdraire

archaique grecque (Verhandelingen Koninklijke Nederlandse Akademie, n.r. 65,2,

Amsterdam, 1958) 216.B 60f.

10 528,29; 529,28 Heiberg.YI529,28. Actually Simplicius throughout his exegesis of de caelo 295a29

(528,3-530,26) discusses "this Cosmos" and Empedocles' views concerning this

Cosmos, referring a considerable number of fragments to the present condition.

He describes his procedure 530,11 as follows: 'oaur'AX 6),(yOv r&vr.06q 7rpoa-

7rea6v-rcov&17tiv&voc0&tkvoq npeOft[rv. Should he not have noticed that some-

where between these passages Empedocles went from one cosmic phase toanother? It is perhaps good method at this point not to exclude the possibilityof such an error.

112



pendently of Simplicius' opinion and to consider whether they relate

to the samne osmic phase. For that B 71 describesphenomena of thepresent world is fortunately certain: the v5v of its last line is as goodevidence for us as it was for Simplicius:

EL n aOL7MP G)5Vav76TXo? 970s?eo ntornq,

7r6 580taroq oc-q -re xocttl L?pOq extu -re

xLpVOC~.wL eta re yevoXtoO XpOLatre Ovqt6v

tc6aa' 6aa viv yeyoarccauvmappoaev'r'AyppoUlrM....

Have we any reason for referingthis fragment to a cosmic condition

different from that described in B 35? Again Aphrodite-Philotesis thecreative power, againit is her function to fashionliving entities througha mixing of the four elements; and again thereis a great variety of suchentities. While the other fragment speaks of countless Oveofurnishedwith nacvotoa ae'Lx,e here read of "so many forms and colors".12 Inboth instances the emphasis is on a great numberand diversity whichmust be astonishing if it has sprung from basic principlesthat are nomore than four. Surelythe conceptionis one and the same, and in viewof the visv n v. 4 we may now with somewhat greater confidence

consider B 35 as a description of the process by which the presentworld condition came into being. The only point of importance notrepeated in B 71 is the simultaneous persistence of elements in anunmixed state. Yet there is nothing in B 71 to militate against thisassumption. In fact Empedocles himself must have made it since hesurely was aware that the condition v-uvncludes these large aggregatesof air, earth etc. (If a proof for this obvious point is needed it may befound in B 21 where these aggregates bear witness to his doctrine offour physical elements).

CuriouslyB 21 and B 71 begin in similar fashion: if you still misssomething, i.e. if you are not yet convinced. B 21 proceeds to supplythe (so far) lacking conviction, and there can be no seriousdoubt thatB 71 likewise led up to details corroborating he basic view. Pausaniasmight indeed wonderhow from the mixing of four elements so great avariety of forms could emerge.Within limits we can see how Empedoclesproved his thesis. He did not take up animal species one by one butconcentrated on "principlesof composition."

B 73 followedB 71 jier'6?Byx. t presents Cyprisas busying herself(?)

12 -6aa' 8aocDiels-Kranz. Simplicius' Mss. have 'roEo sic) 6ax. Karsten's 'r6aa'6aoc eems preferable to Wilamowitz's sro;'otc.

113



about 28em.What 71 promised is here implemented.Other fragments

continue the story. If we compareB 73clG(X4?) 8r tote XO6voou'pL, ?'rsLE' a8tLvev V6O4PP(

?28EM 7rOL7nVCOUaOOOO 7rPL8)UX XPXTUV(X...

with B 96

-1 8 XO'V ?tLVjPo0VzuCepvotCxocxvoLaLv8) 8aUo -v 6XwrVep'wvX&xs a'tL8OG ocyXnq,

'reaaopo8' 'H T oot 8' XSae??ux&yevovwo

'Ap ovE-% 6WnaL 4p1p6X Ocantea[O0v.

we have the impression of remaining in the same context of thought.It is perfectly possible but not essential for our argument that 96followed closely upon 73. The same elements (earth, water, fire) arespecified in both passages and in both Cypris (DLXG'nr,Harmonia) is

similarly active.'3 There is nothing in Simplicius or anywhere else toprevent us from arrangingthe fragments in the order here proposed;in fact we have Simplicius'explicit assurance that B 96 had its placein Book I and that B 35 precededB 9814 which no one has yet removedfrom the neighborhood of B 96). In Diels-Kranz B 96 is placed later

than a fragment (B 62) definitely attested for Book II. Needlessto say,our reconstructionmust be carefulto avoid such glaringconflicts withthe ancient evidence.

Actually B 96 whichdescribesthe nature orcompositionof the bonesis by no means unique in showing Cypris at work. The close similaritybetween it and B 98 could not escape editors and other students ofEmpedocles. In the latter fragment Cyprisaccomplishesher "perfect"(t6eX?Lov)work: flesh and blood represent the ideal mixture 1:1:1:1.

From Aetius' account (A 78) we gather that Empedoclesspecifiedthe

ratio of mixture also for the vi5poc.15 e may assume that the tenor of

1' This must have been realized by Karsten who put B 96 after B 73. Onewonders whether Diels was well advised when he gave up Karsten's sequence.The more general terms of B 71 are in 73, 96 and other fragments presently tobe considered replaced by more specific and concrete descriptions of Cypris'activity: cf. my paper "Nature as craftsman in Greek thought", Journal of theHistory of Ideas 24 (1963), 476f.14Simplic. in phys. 300,20; 32,11 (the passages are quoted in DK before the textof B 96 and 35 respectively).16 AMtius, it is hardly necessary to observe, has no interest in Cypris' doings.

The doxographers are satisfied with the bare bones of the "doctrine". - Nails,tears, and sweat (Aet. ibid. V,22.1) are likely to have been treated as "secondarydevelopments" not directly resulting from Cypris' &p[6etLv.

114



this statement resembledB 96 and 98 and that its place was close to

these fragments.What other fragments inform us about Cypris as mixing and

creating parts of the animal body? B 86 and 87 speak of her asfashioning the eyes. For the material employed we have only theindication PK v at the beginningof 86; however from B 85 which hasbeen very reasonably assigned to the same context we learn that fireand earth contributed to their formation.

All fragments (and doxographic statements) discussed in thepreceding paragraphsillustrate the creation through mixture of 8r6L

present in the 90vecxOvnTj-v.hey give us some idea of how Cypriswentto work. What is not yet accounted for is the astonishing diversity ofbiological forms, the pupwpLocnd ntcvToZLnf B 35 or the -roaacof B 71.However since Empedocles was a comparativephysiologist we have aright to assume that a little could go a long way in his scheme. 'raut&

Tp(xey xL i xomL ov1Wv W T'P& tuxv& xOc'LXent8e;... (B 82). If theseBEc= are identical, one statement concerningtheir nature or compo-sition would cover all of them; and even if Empedoclesin this instancedid not offer a combinationformula,16his audience would understand

that the nxv-roZov35.17) was a matter of appearancerather than ofreality. Another fragment leads us farther and is particularly relevantto the present phase of our inquiry because it once more refers toCypris'creativity:

xxv 6' aG' [L6 7Wvxv&, 7oc 8' 9XTOO& .xV&te'jyS,

KU'7pL8oqv 7rmX&[aLflL7Xa8-% TOlr5a& ruX& .. (B 75).

The arrangementof the firmandloose partshere indicated is character-istic of man and many other living entities but in the case of turtles,snails and certain "inhabitants of the sea"

which have shells, i.e. ofcrabs, lobstersetc., Cyprismust have seen fit to proceed n the oppositeway; for they have the bony structure on the outside:

v' 6betL X06vo Xpw-6z '7'pToTo VoecTioaomV. (76.3).17

16 In view of Aetius' report (cf. Note 15) I should regard it as probable thatEmpedocles stated "ratios of mixture" only for a limited number of tissues.17 The line is puzzling because bone according to B 96 contains only 2/8 ofearth in its composition; however with B 75 to guide us I should not doubt thatsnails etc. illustrate the arrangement of Cypris by which the nwuxv&ame to be

on the outside, the oav&n the inside. Note how Aristotle used these importantthoughts in de part. anim. 653b38-654a9; 679b31-34; 684bl7ff. (see mypaper "Tissues and the soul", Philos. Review 59, 1950, 464).

115



Thus diversity is brought about by arranging he same tissues in more

than one way: flesh may be outside or inside; so may bone. How manymore brilliant insights of this kind Empedocles put forward is aquestion which we must perforce eave unanswered. On the whole it isnot likely that a very large number of similar thoughts should havedisappeared without leaving a trace in the accounts of Aristotle,Plutarch and the doxographers. Inasmuch as Aristotle took up andcarried farther Empedocles' essays in comparative physiology, thehope of recovering from him additional thoughts for Empedocles'comparative scheme should be kept alive. However it is probably

unwise to postulate that Empedocles gave an exhaustive accountcoveringa largevariety of biologicalforms. He may well have thoughtit sufficient to be specific about the most important tissues and a feworgans,and after this to suggest that the tissues entered into a varietyof combinations and that items seemingly heterogeneous were yetidentical in substance. A few striking examples would create 7artL4;

and since exact proofs could not be given, it was important to inspirethe imagination of the audience and turn it in the right direction.Evenfrom the little that is left to us we can see how varied forms the

mixture of the elements could take and what fruitful possibilitieswerelatent in the idea of "mixture".Moreover here were good reasonsforfocusing on "rational" rnixtures, while indicating that there wasconsiderable latitude for departuresfrom the "norm"."8

We have proceded on the supposition - hardly too bold - that afterB 71 Empedocles went on to supply the still lacking 7rtLTtL for thecreation of many diverse forms out of the four eternal elements."'Unless we have gone completely astray, even the fragmentarymaterialgives us some idea of how Cypris went about this task and how

Empedoclespresentedher activity. The passagesjust consideredmakeus visualize and believe what B 71 presents as an idea sufficientlyfamiliarin its generalterms but not yet established in concreto. In allinstances Philotes is the creator, a fact fully in harmony with what

1 See B 98,4 in combination with A 86, lOf.19For colors, the fragments just studied teach us little; note however xeuxacB 96,3 in the description of the bones. This whiteness must be due to the pre-valence of fire in the composition; cf. B 21,3 where the "white" Sun representsthe fire. For colors as associated with the tissues see P1. Tim. 74d4, 80el f.,83 c f. (no specific indebtedness to Empedocles can be claimed for these passages;

yet it is well to remember that with his theory of the tissues in the TimaeusPlato finds himself in the Empedoclean or "Sicilian" tradition. See the papercited in n. 17, 446ff.).

116



B 35 leads us to expect. There is evidence in the fragments that

Empedocles thought of her creative activity as materializing at aspecific time: TO'TE73,1 (cf. the similar use of Us 8i X6X'B 84,7 and

K7tPL8O4 bV 7rC0UXM'VT.LV're 01 7rp&s' iyuOVTO 95). The passagesdiscussedare not al all conceived in an evolutionary spirit; they allowus to suspect that Empedocles placed all of Cypris'creative acts at thesame point in the "cycle". This point should be near to the eventsdescribed in B 35. For since B 71 refers to the living beings v5vandsince B 73,96 and the other fragments just studied specify the &psu64av

of tissues, forms, and parts belonging to the animal kingdom now in

existence, we may now with greater confidence regard B 35 as intro-ducing us to the origin of the living beings that are a part of ourpresent world. The living beings whose origin this fragment describesare just as much the product of mixture as the living beings and their"forms"explainedin the other fragments.If B 35 were to be related toanother phase of the cycle, Cypris would play an identical role inopposite phases, a result which can hardly be welcome to scholarsfavoringa dual cosmogonyand zoogony.20As for Strife, we know fromthe early fragments of 7tep'LU'aecohat in the world around us it

contends with Love. B 35 shows this condition of antagonism. AsStrife seeks to keep the elements apart, it may well be responsibleforthe disintegration or "death" of the mixed forms, while Love isresponsible for their origin. As B 20 explains, Love "integrates" theliving organisms in this world,and Strife causes them to disintegrate.

Our ancient authorities inform us that Strife has built up ourCosmos, and that it has separated the elements, congregatingeach ofthem by itself.2' When this process is completed, Heaven, air, land,and sea have come into being; we have the cosmic pattern which is

familiar under the name of "four concentric layers".22It is into this20 See B. 17, 3-5, verses of crucial importance for the theory of a dual cosmogonyetc. We shall discuss them below p. 138. Burnet p. 240 thinks Love can createbones, flesh etc. even in our world of Strife "because it is by no means banishedfrom the world yet, though one day it will be". I doubt whether he has thoughtthis idea through and realized what difficulties arise if Love creates perfectmixtures (B 98) after Strife has produced a radical separation of the elements.Moreover Burnet has paid too little attention to B 71 and its relation to B 35.'l See A 30,37,49; Aet. 11,6,3 (Burnet p. 236, Raven p. 332).

22 I use this term as an approximate designation, remembering that the spatial

relations of earth and water are not well described by it and not wishinggratuitously to import these difficulties into Empedocles' scheme. "Fourconcentric spheres" are assumed also by Raven (346); however I am puzzled

117



Cosmos, and into the 8lv- somehow connected with its origin or early

phases, that Love makes her entry in B 35, whereStrife is still holdingon to the elements in their state of separation.As we have said before,our world consists of Heaven, air, land, and sea plus the multitude oforganicbeings that populate it. I know of no reason why B 35 shouldnot convey a picture of this world.

Two rathertechnicalpoints may brieflybe taken up at this juncture.As far as I am aware, no recent interpreter of Empedocles has comeforwardwith a flat statement that the flesh, the bones, the eyes etc.whose creationwe find set forth in the fragmentsare not the flesh, the

bones, etc. of living beings now in existence. Commonsense would beoutraged by this idea. Bignone who went as far as to formulate thehypothesis that the tissues, organs and forms in our world might bedifferent,dismissed it in favor of an alternative theory. In his view thefragments which we have discussedbelong to the opposite phase of thecycle, since in the present periodStrife, not Love should be in ascend-ance; however, the "forms"of the present phase being identical withthose of the opposite, Empedocles when he came to deal with thepresent world condition "si referi alla descrizione gia data per il

periododell' Amicizia"23 a good scholarly procedure,to be sure, andI would not deny that Empedocles could say CU np6T&povovre}Xc,but should we not expect somewhat more variation especially if onegenesis is supposed to be the work of Strife, the other that of Love,and one to materialize through a &LocpxeagOoc,he other through aau pXaOMxL?4

The other technical item relates to the work of Strife. It may not besuperfluousto observe that when Strife has built the Cosmos,havingseparated the elements into four homogeneousmasses, it has reached

the peak of its power. For the process of 8aLtaOaL has been completed,and we have no right to ask for more. I could be brief on this point,regardingit as settled, if it were not the case that a passage in Sim-plicius seems to imply a furtherdevelopment: Neikos, Simpliciusheresays, is the creatorof the physical CosmosZ0',tov'7XprXTPO TXzZ at,

Tiq a&mXpEaccogTOv XO6aCOV TOUTOV 7roLoi3v.25 If Strife were to attain the

since he does not state whether or not they constitute the Cosmos. See alsoHolscher pp. 18 f.28 Bignone, op. cit. p. 556 n. 1.

24 See below pp. 140 f. on B. 17, 3-5. (c.f. also p. 122).25 in Phys. 1124,3. I do not think that Arist., Metaph. 985a25ff. or de caelo295a 30ff. can be used as supporting Simplicius' implications (we may wonder

118



maximum of its influence only after building the Cosmos this could

only mean that the layers of the elements become in the end com-pletely separated from one another - so completely that they do noteven touch. Speculating about this development, one might surmise

that a countermovementof Love is neededto bring the fourelementarymasses into mutual contact, thereby restoring the Cosmos. A hypo-thesis of the kind would in no way interfere with the view of Love's

actions here put forwardon the basis of B 35; yet it probablyis betterto dismiss this speculation as being out of keeping with the balance of

the tradition. An acute friend has pertinently asked what would be in

the space between the elements if they do not touch. The answer(o pudor!) is - void; but TroU ov-'O 8' ou8V XeV?06.26

Love and Strife in theirrespective oles

We have indicated that the reconstruction here put forward runscounter to the prevailingtheory which places the contemporaryworldin a periodof Strife or of "increasingStrife".This theory assumes twoformationsof the Cosmosand the living beings in it, one while Strife,

having brokenup the Sphairos,separates the elements, the other whileLove, as she returns, makes the elements combine and moves towardtheir complete fusion in the Sphairos. Ourpresent worldwould on thisview be that created by Strife. We shall in a later part of our studyexamine the basis of this opinion. But it may be said even now that noancient authority and no testimony of any kind associates the presentworld with "an increase" of NeZxo. What we do find is that theKosmos, i.e. the present Kosmos, is e'ntNeExouv,hat it has been built

up by Strife, that the elements 8&La 'Xet U -6o5 NeBxou4 a statementborne out by more detailed accounts.27). With such information I

about xcop[4n the last passage and it is perhaps not quite fair simply to refer tothe definition of this word in Phys. 226b 22). Zeller (5th ed., p. 783ff.) evidentlybelieved that the total separation of the elements does not allow the formation ofa Cosmos; see against him Tannery, op. cit. p. 310ff. (who in turn errs in allowingLove an influence on the cosmogony and in denying that Strife produces "une

separation complete'), and Millerd p. 51 f. Miss Millerd not without justice chargesZeller with creating confusion. I am tempted to use the same word of her

reasoning p. 52 and n. 2 (why should B 35 be called a "stopgap"?). Cf. alsoBignone 235-238, and especially von Arnim's good and clear statement (p. 18).

26 B 14 (cf. B 13). I owe the question as well as the answer to my colleagueJulius Weinberg. - Cf. also Hoelscher p. 20.27 Arist. de gen. et corr. 334a5 &i,uU xot '6v x6apov 6[LoiCwgXeLv qpqaiv&rl 're ro

119



consider the interpretation here advanced entirely in accord. For on

our reading of B 35 Neikos has indeed built up our Cosmos by sepa-rating the elements and is even now - at the stage of B 35 - continuingto keep them separated, although it has passed the height of its power.Love does not create a Cosmos but fashions living beings. It makes aconsiderabledifferencewhetherin referring o the Cosmoswe mean thephysical frame - Heaven and the heavenly bodies, the air betweenHeaven and Earth, finally Earth itself andsea - orwhether we at thesame time include the living beings contained within this Cosmos.Failure to keep these two meanings distinct seems to be at the root of

much misunderstanding and confusion. In what sense Empedocleshimself used the word x6a[Locs againa different question (to which weshall come back); it is not for reasonsof his own usage but for the sakeof clarity that I propose in the balance of this paper to confine theword "Cosmos" o the former meaning, using it solely of the physicalframe.This means that we can accept the consensusof ancient opinionwhich makes the Cosmosthe product of Strife, can accept Aristotle's

testimony (and argument) to the effect that Empedocles "left out" acosmogony or xoa,oTotoxof Love,28 and need find no fault with

Simplicius' reports, according to which both Strife and Love havecontributed to this world. In fact when looking more closely at

Simplicius'statements we realizethat the builderof the Cosmos s for

him Strife, whereas Love manifests her power in the mixing andfashioning of living beings (which is entirely our opinion). Simplicius,who is thoroughly consistent in his understandingof Empedocles,29

Ne(xous v5v xxl np6repov &7d ~qi4D0,6,qoroq(see below p. 130); Metaph. 985a23

6'rxv l.Lv yap eit TX'cToLXetX 8L(a-r-)rot -6 7t&vxr6 ro5 Ne(xou4; de caelo 295a30:

6re... -T&aroLXcxaateLerate Xcopl5 CtrD6oG NeLxouq... For what happened to the

elements in the diakrisis see A 30,49 and for Strife as the builder of this Cosmossee the evidence in Simplicius to be discussed below n. 29. Incidentally, although

in de gen. et corr. 333b22-334a9 carping criticism dominates so much that it is

at first difficult to make out the doctrines against which it is directed, closer

reading shows that Aristotle throughout refers to the diakrisis of elements from

the Sphairos, his point at a 1ff. being that Empedocles does not consistently use

Neikos in this context as the agent.28 de caelo 301a15f.; see below p. 124.29 I quote the most pertinent passages: in de caelo 528.11f.: {nrb 'ro5 NctxouS

&OaxpLvovroq 'ra aTOLXCel 'rOU5,Ov >ycL y(vc0a0L r6v x6a[ov 6 'E. 6anep tnt6 ^

PDLXLxsaUvOyovaIq xal kvoU6ai acucxr&6v EmLpov; ibid. 590. 19ff. ('E.) acyxpLv6-

,evo... np6-?pov u'7r6 r, (DXota ra' roxcTx 5a-repov vso6 to5 Ne&xou48&MxpLv6Mcvat-6v3e 'r6v x6ctov 7OLetv XkyeL. 591.3 xmxt' 1xeEvnv (sc. OLXLAv)oCuX68 6 x6a5oq 6

otaOiT6q, &XX' 6 vojT6q kytveTo. For the last point Simplicius quotes the de-

120



informsus on several occasions that what Strife creates is the Kosmos,

and what Love creates is the aocppoq r voyr6o x6a[Loqagain a use ofthe term x6a[Loq hich we shall do well to eschew.) If one follows his

argumentone would be hard pressedto find between the votoq x64aloqof Love and the taOr6s xo6apoqf Strife a place for another oajr6q

x6alioqof Love. Firm believers in a Cosmoscreated by Love might tryto discount Simplicius' testimony by arguing that he was so pleased

with Love as creator of an intelligibleworldas to overlookher analogousrole in a physical world30 and as result of this blindness to assign

B 35 to a wrongphase. If there were strong and good evidence for Love

scription of the Sphairos B 27.4. Making allowance for this Platonising identi-

fication of the Sphairos as vo-T6q x6a5Loq,we may additionally quote in Phys.1123,28ff. orOwX octrLm(for E.) ron ,iv vo-1'nxo5r?OvXt vD a& vc ew NO6v

XcpmLpovnoLo5amv (here follows again a reference to E.'s text), ro3 8& taNtroi 'r6

Nrtxor 6'rzv47rLxpxOCT.! X&reXa&g& T-4 BLcxp(aec&4 rTv x6eqov 'O5'ov nOLO5v. Infact Simplicius in Phys. 31.31 knows it as a widely held opinion (ot 7ro?Xolo[iA-ColJGL)that Love alone has built the vo-I)6q,Strife alone the EaL6a0r66a[?o;. Hisown argument ibid. to the effect that Love has a hand also in this (xtbs)

x6a[Loscorresponds to that in de caelo 528.30-530.11; in both instances the

passages adduced show that Philotes' contribution consists in the creation of the

living beings and their parts (in Phys. 32.2 xxt ris ivraix0a 8-%LLOUpytXiq auyxp&-acco- r?v 'A(ppo8[trv %toL'rv 9pLdov oat'rLavpna(v).'O Examining Simplicius' statements from this point of view, I have wondered

whether his misunderstanding in de caelo 293,18ff. of B 17.7-13 discredits him

as authority on the operations of Love and Strife. Yet it would be very rash tosuppose that he "derived" from these verses the information vouchsafed

293,20ff.: 'v 0Dtmxv xal 'r NeNxos xxa& vLpoc knxp0Couv-x 'iv [?iv aUVd&yeLV'rw

ndi&vrotL Evxal 9(peEpL v 'ro5 Nelxou4 x6a,uov xal trot?Tv &kmuoi-'r6v zalpov, '6

8i Nctxoq 7tc?xLv;LXpCVELV 'r&oToLXa xaClroMctvLv rvoLou'rov x6a:,ov. He introducesB 17.7ff. not by 'E. Xkyabut by 'E. a-xLves Xkywv nd the main reason why headduces it is that he wishes to illustrate Aristotle's observations in de caelo

279bl4f.: ot 8' &vaXXM&'e t?&voC)rco. &e8i &)XcogXeCv(scil. Poavl6v x6a:,ov)YOCLp6icov xocl 'Oro &el 8tmareXetv uS',co4arcp 'E. Cf. in Phys. 160,18 ff., where

the almost identical verses B 26,11f. are used for the same purpose. If even soSimplicius reads too much into (though not out of) vv. 17.7-13, due allowanceshould be made for the fact that v. 9 was missing in his copy. This is shown bythe parallel quotations in de caelo 141,1 ff.; in Phys. 157.25ff.; cf. Diels P(oeta-rum) P(hilosophorum) F(ragmenta) ad loc. Simplicius doubtless consideredhimself entitled to read f 8i (instead of W) in the beginning of v. 10, wheresome of his Mss. actually have this reading, and it must be conceded that withthis reading and without v. 9 the meaning of the entire passage comes nearer towhat he finds in it. Miss Millerd (op. cit. p. 48) accuses the Neoplatonic commen-

tators of "almost inextricable confusion." I hope to have shown (pp. 120 f.) thatthis is not fair to Simplicius. Philoponus may be more easily discounted sincethere is no evidence for his acquaintance with Empedocles' text.

121



as building a physical Cosmos we might have to resort to such

speculations.On the view which we find so difficult to accept the present world-

order would have come into being while Strife is gaining upon Love.

Yet we have foundampleevidencethat the living entities of thepresent

world are the productof Love. Would it not be paradoxicalthat Love

while on the retreat and on the way out has so much power - enough

power to fashion perfect mixtures? If there are two xo6atoc,houldnot

in one Strife create the beings, preferablyby causing a atapu L?31But if B 35 does not referto ourworld, Love has the same function of

mixing in opposite cosmic phases. This seems the wrong kind ofsymmetry.32And let us make no mistake: if living beings, no matter

by whom fashioned, are to arise in a periodof increasingStrife, Strife

must have done a good part of its work beforehand.For such beings

need earth and sea to exist, both of which arise late in the separation

of the maxima membramundi.33Love, whom one would suppose at

that point to be reduced to a last ditch defensive position (if not

completely "goneout"), would have to have an astonishing degreeof

initiative and aggressivenessto producesomething r-XeLov.n fact for

Strife too it is late to create ~Cpoc;or the sundering of the elementsinto cosmic masses is by all rational standards a far more advanced

stage of his ascendance than the creation of compounds. Raven has

made the experiment of placing the origin of animal life that belongsto the present world between what he calls "the first stages" of

cosmogonyand the completed "rule of Strife". Yet he actually has for

the "firststages" used up - and probablycould not help using up - all

material available for the completion of the Cosmos and thereby for

the complete rule of Strife. When he comes to the latter subject he

wonders "whether Empedocles ever described the rule of Strife indetail"34and tries desperately to fill the yawning gap by five passages.

Cf. B 17.5 (8 7 r&?sv Lxpuoi.LvcV Op cpOetao 8;tk-rr), a passage regularly

invoked by those holding that Strife created the living beings of the present

phase.'' Cf. von Arnim, loc. cit. p. 22: "die zweite Entstehung wiurde,da sie ebenfalls

die Liebe zur Urheberin hatte, eine einfache Wiederholung der ersten sein".38 See A 49. I assume that Aucher's and Conybeare's Latin rendering of' the

Arabic translation of Philo's de providenta may be trusted for our purposes. The

doctrines look credible. A 48 (Plato, Legg. 889 B) refers to a plurality of thinkers

and is at best a "typical" account of (late) Presocratic cosmogony; I know of nogood arguments for including it among the testimonies for Empedocles.34 Op. cit. 345 f. (for the "first stages" see 332-335).

122



Two of them do not refer to the operationof Strife but of tyche.35The

third (whichis essentially a critique, not a report) yields no more thanthat the elements were separated by Netxo4.36The two remainingpassages describe the rivalry of Love and Strife in the present worldcondition.37MoreoverRaven's evidence that "the present state of the

world belongs to the transitional phase when Strife is gaining uponLove" is a passage in Aristotle's de generatione et corruptione which

speaks of the condition of the Cosmos 'd 'oi NeLXOuV:V.38I have

already entered my protest against the rendering of this phrase by"duringthe increase of Neikos" or similar turns.

To keep the perspective, let us say here again that the championsof a dual Cosmosand a dual creationof living beings rest their case on

evidence or, to use a moreneutralword,onpassages,whoseexaminationwe have postponed. Yet whatever the evidential value of thesepassages may be, they too must be seen in perspective. If they haveweight this weight must be balancedagainst the testimony not only of

Simpliciusbut of the entire doxographic tradition which knows onlyone Cosmos, to wit the present, and while it has most remarkablethings to report about the formation of the Heaven and two cosmic

hemispheres, about the Sun (if not about two Suns), the moon, thestars, the Earth etc.,39never refers us from the formations, develop-ments, and phenomena of this Cosmos to those in another where

35 de gen et corr. 334a 1; Phys. 196a20. I see no valid reason for separating thesepassages from those adduced p. 332 under the heading "The first stages". Forthe first passage see n. 27.3S Metaph. 985 a23."7 B 17,6-8; 26.3-7. The latter passage says that the elements "running throughone another" form human and other living beings (under whose influence?)

sometimes coming together in Love (rule of Strife?), sometimes moving apartby the action of Strife, until utterly subdued they become One (under whoseinfluence?, see below p. 126). I am equally at a loss to understand how B 17.6ff.may illustrate "the rule of Strife".38 de gen et corr. 334a5 (see below p. 130); Raven p. 339. Raven's comment onB 96 and 98 (p. 335) Empedocles "seems to have been the first of the Presocraticphilosophers to pay much attention to such compounds" is probably correct(cf. Philos. Rev. 59, 1950, 436ff.) but regarding his observation that they are

"essential... in his cosmogony" I must refer to my remarks above (p. 120) about

confusions created by the indiscriminate use of the word "Cosmos". For clearlythese compounds created by Cypris relate to her 4coyovta; see above p. 114.

as 31 A 51, 53-56, 58-61 (Raven pp. 333-335). For the "remaining" of the Earthin its place, we have strictly speaking no doxographic account but depend on

Arist. de caelo 295a13ff.

123



presumablyall things happenedin the reversefashion.40No more does

the tradition include any statement about two distinct Cwoyovtxcunless we assume that any author who refers to a zoogony sil yLMvq

(a perfectly good method and helpful too in the absence of paragraphs

and verse numbers) must know about an alternative CyoyovLa'd

Netxouq. If such references are taken to suggest two CyoyovLaM,should counter this kind of "testimony" by a passage in the Physics

(197b5) where Aristotle calls the mixed forms (f3ouyevi,.e. Aetius'

"second" stage!) adLE4pXs aar'aeL. This can hardly be the normal

Greek for a referenceto the latter stage of developments in another

- and later! - world. Just as no ancient author speaks of a cosmogony&,dor u7r6DLMLM,o none knows anything about a zoogony?dLNeLxouc.

Simplicius,althoughoften criticalof Aristotle'sreportson Empedocles,

readily endorsesthe statement in de caelothat 'Epre8oxXiq rapx7r:S

'rjV SM yLXLoq 6vaLV (in the sense of xoasto7roLEM).42Knowing how

discreditedAristotle's own authority in such matters is, one hesitates

to make much of this passage; nevertheless I venture the comment

that Aristotle, while having no item in the tradition against himself,has reason and a good argument on his side. The Cosmos,he says, Ex

&OCXzP v arqxe... 'coV a'TOLX?eL0v;ut Love's work is uyxpEveLv(not &OCXpLVeLV).43

40 In the myth of Plato's Statesman biological processes in one cosmic phase are

the reverse of those in the other (270dff.). J. B. Skemp (Plato's Statesman,

London, 1952, pp. 90ff.) has tried to define the relations of this myth to Em-

pedocles' dual cosmogony. Proceeding as though he were comparing one text

with another, not a text with an hypothesis, he finds the "differences"... "very

important".41 &itE 'i 0L06-q-r1o, Arist. de caelo 300 b 29 is entirely correct and appropriate;

for the elements would not mix before Philotes comes into operation. Cf. also

de gen. anim. 722b17ff. The interpretation given to A 86.20 (304.34f. DK) byMiss Millerd (op. cit. p. 45) and Burmet p. 235 n. 1 strikes me as biased; r6Te

implies no contrast with v5v but takes up Awl q (DWag.42 de caelo 301 a 15f. (Aristotle continues ov yxp av 486vocoaUa')aCL 'r6sVo0pv6 V...,

and it is generally and rightly understood that the reference is to xoa[oyovEa;cf. 12f.). Cf. Simplicius 590,26-591,7. I do not know whether the exegetes who

according to 591,6 understood Aristotle's nrapxcL?eet E&avxx)ut,u[&vCL(sc.

the cosmogony of Philotes) had in mind the creation of a x6a[Loq r of a aqoatpoq;cf. in Phys. 31,31 and above Note 29. For Simplicius as critical visavis Aristotle's

statements see e.g. in de caelo 587,8ff., 530,12ff."3In a 16-18 Aristotle says (rightly, it seems to me) that it would not have been

possible for Empedocles to make Philotes fashion a Cosmos by synkrisis sincein the Cosmos the elements are &aotxexp[Liva.When he next says that the

Kosmos being &x 8LmxexpL[L.vcovust arise FEkv6q xal aUyXEXpL[LkvOU,t is hardly

124



In any case, Aristotle, whether correct or not, cannot be refuted by

the reasoning by which Raven impugns his testimony: "Aristotle'sremark [in the de caelopassage], that Empedocles passes over thecosmogony of the transition to the rule of Love, is not perhapsstnrctlytrue; not only 464 [= B 35 DK] but also, as we saw, 443 to 446[= 57, 59-61 DK] are all concerned with this phase in the cosmiccycle. But there is no denying that 464 is both vague in outline andobscurein detail... The reason for this obscurity is not hard to guess.Empedocles, by his introduction of the cosmic cycle, has set himself atask which might well overtax the most fertile imagination. He has

imposed upon himself the necessity of describinga cosmogony and aworld that are the exact reverse of the world we know and of thecosmogony that brought it into being. It cannot even be said that thecosmic 'cycle was unavoidable..." (here follows a well meant sug-gestion as to how Empedoclesmight have simplified his scheme)".Evidently the hypothesis of two cosmogonieshas taken such firm holdthat when it leads into trouble the blame must be put on Empedocles.In truth B 57, 59-61, while reasonably placed in the transition to therule of Love, include no word about a cosmogony or a Cosmos (the

caution which we recommended or the use of this word seems justifiedif traps are so plentiful). All four passages deal with the evolution oflving beings. As for B 35, Raven's other item of evidence, we have,I hope, at the beginningof this paper, seen how it is to be understoodbut I may as well ask my readers to look at it once more and to decidewhether imagination, however stretched, may discover anything ofcosmogony in it.4"All that may happen to the Cosmos in the light ofthis fragment is that it will dissolve after a time, giving way to the

possible to say what is reasoning and what is report but the observation wouldcorrectly apply to the separation of the elements by Neikos after they hadbeen united and fused in the Sphairos. - Cf. Holscher p. 21.4" Op. cit. 347f.

'6 I cannot accept Raven's complaint about "vagueness". If there is obscurityit is due to our ignorance of what preceded. I do not profess to know exactly atwhat point of the cosmogony and under what circumstances the dine originated.Raven's remark "the dine described on p. 333 is clearly not the same as this one"(scil. in B 35) is not helpful since neither the term nor some of the most relevantpassages are to be found on pp. 333f.: de caelo 284a24, 295a29ff., 300b3. Thefirst and the last of these include strong indications that the dine is a part of the

present world. Having originated in the formation of the Kosmos throughStrife, it is in operation when Love takes over. Cf. John M. Robinson, The dinein Presocratic Cosmology (Ph. D. diss., Cornell, 1949) 52ff.

125



complete fusion of things, by which Philotes reaches her goal, the

IV p6vov.46

For this prospect as facing our present world we even seem to havea direct and unequivocal statement of Empedocles. The presentcondition of things in which Philotes mixes and creates while Strifecounteracts her is to continue, we read in B 26,7, eLaox9 VUvap9YVV

(subject: the elements) -6 tW&v7evepQey'vi]-rL "till they grow oncemore into one and are wholly subdued" (Burnet's rendering seemscorrect, except for the gratuitous "'oncemore" 7). Yet the ?v and the

auLpcqu6aOoLan only be brought about by Love. It is she whose ac-

tivities are regularly characterized by auv-compounds,like auv6p-xeaot, autLveLV whereas Neikos is associated with 8L&cpgU'aoLand

8&yX. 26, 5-8 do not teach that the cosmic cycle will some day come

to an end in the conditionof the Sphairos (not again to be brokenup?)but that the alternationand competition of Love and Strife that go on

in the world aroundus will continue until the next phase overtakes it.According to B 26,7 this phase must be the Sphairos in which the

elements are "subdued"to the point of being invisible (B 27). Thus,unless we decide to eliminate this tell-tale line from the present

context,48 we have to admit, whether we like it or not, that we aremoving in the direction of the Sphairos, the power of Love being on

the increase.49Of Empedocles' cosmological cycle no passage is preserved which

includes the word xixXo;. The first line of B 38 being unfortunately

corrupt, it is not possible to elicit from this fragment anything about

the starting point of the cosmologicalsection or to infer with confidence

46 V. 5; cf. above 111.

47 E.G.P. 210. Did he read Bywater's cx? Only one manuscript of Simplicius has

Ev,which is now generally and, I have no doubt, rightly accepted. Cf. Diels,Hermes 15 (1880) 163. For 'r6 wravs used adverbially see Aesch. Ag. 175, Suppl.

781, Soph. El. 1009 (xavsca0pou r6 wasv O6XiaOaL); cf. Jula Kerschensteiner,

Kosmos (Zetemata 30, Munich, 1962) p. 127 n. 1. The rendering in DK "zum

Alleinen" seems most dubious. H. Munding Hermes 82 (1954) 142 is almost

correct.48 Wilamowitz (Hermes65, 1930, 246) would eliminate v. 7 as "unverstandlich".

He also uses the argument that the line does not recur in B 17. But it is im-

possible (as Wilamowitz in effect admits) to establish complete agreement in

repeated verse groups. I should add that it is also unnecessary and undesirable.

Were it necessary, I should be tempted to argue that since there is doubtless a

lacuna after 17.8 not only 26.8 but 26.7 and 8 should be inserted there. EvidentlyEmpedocles' argument does progress even where verse groups are repeated.

"9 Cf. for this interpretation of B 26.7 von Arnim. loc. cit. 26f.

126



that he wished to reach the point where aiX' kye'vovo& V5iv aopCo[LEV

Manavat. till the condition of the Sphairoswould be a good apy'.50Weknow that Anaxagoras went from the nokvraO[oi5 o the formation of

the Cosmos and the explanation of differentiated things and maysuspect that also the earlierphysicists, beginning with Anaximander,were not so much concerned with showing in detail how everythingnow visible will be absorbed into the 6tELpOV (orwhatever correspondsto it) as with explaining the processes and the sequence of eventsleading from their (p'pzto 6ax 9nXayeyaxaLv or TM viv GOpFO[?ev

(XCVIOC. In a sense, albeit in a very different sense, even Parmenides

went from the ?v to the account of what is to be found in the presentworld. Thus if we can reconstruct Empedocles' cosmological accounton analogous lines, the hypothesis that he thus procededwould, if byno means firmly established, yet enter the competition with certainadvantages. I should, however, not exclude the possibility thatEmpedocles "ledup" to the Sphairosby describinghow Love broughttogether the elements from all sides to mergethem in the One.5'

In any case we certainly have three and very probably a fourthfragment referingto the Sphairos, and there are two dealing with its

disruptionby Neikos.52For the furtheroperationsand achievementsofNeikos we depend on the doxographictradition, A 30 and what Dielscombined as A 49 providing the most helpful information. And sincethe process which the doxographers call a &LOCxpLaLor a 8a(a'rocaL)fashions besides the other parts of our Cosmos also the Heaven, itseems natural that in this context Empedocles, who is after all notaverse to digressions, put forwardhis views about the shape of theHeaven, its distance from the Earth, the origin and nature of the Sunand other matters easily associated with these.53For what he had to

say about the position of the Earth in the centre and the reasons of itsremainingthereM,he cosmogony of Neikos would similarlyfurnishthe

60 For the hypothesis that in the sequence of the poem the Sphairos precededthe cosmic separation support may be found in Simpl. in de caelo 590,19auyxpv6,Leva yxp np6-epov xt?. (see note 29).

5' See on B 36 below p. 129.62 For the Sphairos see B 27, 27a, 28, 29; for its disruption B 30, 31. B 27.3 and27a are not entirely above doubt. There is nothing to justify the place whichB 33 and 34 occupy in DK. I should much prefer to have them placed near otherfragments (see above p. 113 f.) describing Cypris' workmanship.

68 I refer tentatively to A 50f., 53-56, 58-61. For the fragments see below n.110.64 See esp. A 67 (for additional information on the 8(vi cf. n. 45). Such relationas A 48 may have to Empedocles' xoao,oyov(awould be of the remotest.

127



appropriate context, but we can of course not know whether all those

cosmological tems were presented on one and the same occasion and inone and the same "excursus".The organization of his material in thesections following the Cosmogony of Neikos eludes us and while it isattractive to think that the excursus from which he calls himself backat the beginning of B 35 dealt with such cosmologicalsubjects, we hadbetter not be positive about this.

Still the right place for the "cosmological" ragments (in particularfor B 40-48) would seem to be before B 35 rather than subsequent toit, where Diels has seen fit to present them; for all of them are most

naturally connected with the creation of the Heaven and the entireCosmosby Strife, even if some may deal with indirect rather than withdirect results of its impact.55As we have said, B 35 is whereEmpedoclesreturns to the main theme. One may gather from this fragment thateven before the point at which we find ourselves in it Strife had lostsome of its power; if the impressionis correct, a development of thekind is likely to have been described somewhere after the actualcreation of the Cosmos was completed. The fragment itself shows therule of Cyprisbeginning, with Neikos withdrawingfrom the "limbs"

(this balances B 30 where Neikos "growswithin the limbs"). HoweverNeikos keeps still ro?oc ut of the mixture: Heaven, air, earth and seaare after all still there. The powerof Love shows itself in the creation ofcountless living beings "furnishedwith manifold forms". To this phaseof the cycle we have related a goodly number of other fragments.,6

Is it then true that "Empedocles by his introduction of the cosmiccycle has set himself a task which might well overtax even the mostfertile imagination" (Raven 348)? He has, I admit, on the recon-struction here offered not covered everything that Raven and others

think he ought to cover. But he has led us from the phase when Loveis in sole and absolute control, i.e. from the Sphairos,throughthe reignof Strife to the phase in which Love returns and begins to display herpower. And if his account came to an end in the present condition ofthings, he would have explained our Cosmoswith Heaven, Sun, moonor earth, as well as the animal forms - very much the same subjectswhich other Presocratic physicists made it their business to explain.The only phase of the "cycle"apparently not covered would be that

65 Some of the cosmological events seem to have needed considerable time and

may have become completed only after Neikos had passed the peak of its power(cf. A 70 about the cosmic condition under which plants originated).56Above pp. 113ff. See also the "Appendix".

128



leading from Cypris'partial powerto her complete and sole power,i.e.

to the restoration of the Sphairos. It may of course be surmised thatthis was covered in a later section "lookingahead"to future develop-ments. I prefer not to make this assumption. The direction of thedevelopment is clear enough even from B 35 where it is said explicitlythat under the sway of Philotes &ov'Cauv6pxs*rct&v i6vovelvoL (5).Empedocles'audience must have been familiar with this idea67and hecould safely leave it to their imagination to realize whither thingswere headed. It need after all not be an accident that we know morePresocratic 86t(xt about the origin of our world than about its de-

struction. Moreover f the "historical"section did not begin with theSphairos but with the creation of the Sphairos, to wit the comingtogether and complete fusion of the elements, the beginning and endof his account would meet and the cycle would be complete enough tosatisfy even the most literal-minded and unimaginativesegment of hispublic. If B 36 suv 8&cnuvepXopvv C' gaysov t=wro Nctxoq is correctlytransmitted in Stobaeus58and has to be rendered "while these cametogether Strife finally moved outside",I should rather relate it to theformation of the Sphairos, placing it shortly before B 27-29, than

suppose that its place was after and at same distance fromB 35.The advantages of letting the genetic section endin the presentworldconditionarepatent. It has alreadybeensuggested that the explanationof 6asa viuveyaxcrLvr &v5vkaopw,uevwas the traditionalconcern of thephysicists. Moreover,concludingthe "cycle"at this point, Empedocleswas in a position to pass on without break or awkwardnessto otherphenomena of the present world, such as sense functions, respiration,reproduction,etc. We have reasons to believe that he turned - moreor less immediately - to the composition of tissues and organs. Theytoo

belongto

the present world. On the reconstructions now in favor67 See our comments on 26.7 above p. 126. It may be well to state that (unlikeRaven p. 348) I consider the xx),oq motif as essential also for the "historical"section. See below p. 142. The discussion in the text relates solely to questionsof presentation.58 Arist. Metaph. lOOOb2seems to have given a very abbreviated quotation(only the last three words). I cannot help accepting FE, rom Stobaeus andunderstanding IaXcov as adverbial in meaning. If Empedocles began by de-scribing the formation of the Sphairos, the analogy with Anaxagoras (abovep. 127) would of course not be complete. While both thinkers make the Cosmos

and all that it contains originate from a [LEi;q in which nothing was 1v8-qXov(Anaxag. B 1, Emped. B 27) the fact that one was much more interested in thecycle may yet have made a difference.

129



Empedocles has to interrupt the account of the cycle - soon after

having started, morepreciselywhen Strifehas brokenup the Sphairos;for this is the point wherehe must explain the phenomenaof the present

world. Here for compellingreasonsthe cycle must be brokenand a full

account given of sense perception,respiration,reproductionand much

else that goes on in our world. Only after he has covered all of these

subjects may the poet take up the story of the cycle again to describe

first the complete rule of Strife, then the creation of a world which is

the reverse of ours, and finally the returnto the Sphairos.Excursuses

there certainlywere in rrepip6as? but an excursuswhich took up half

of the poem is a bit too much of a good thing. "Awkward"would be amild word for this kind of "organization".59

The basis of theallegeddual cosmogony

However LXaULoVoi TroToil ?i)xou Wiets. Even if a particular recon-

struction of the dual cosmogony fails to convince, the hypothesis as

such might be sound and there may be evidence that should not be

dismissed or minimized. Essentially the case for a dual cosmogony

rests on a passage in Aristotle, on certain fragments relating to thedevelopmentof living beings, and on a passagein B 17 which definitely

states that there is aoc' OvnTrCv6'vea, aOL% ' (7toXLcrn+.The passagein

Aristotle seems to point to a cosmogony by Love, the fragments are

refered to a zoogonyby Strife - neither of which has a place in our

reconstruction - and the verses of B 17 may even be, and are indeed,

taken as attesting both. Let us begin with Aristotle. In the course of

an argumentwhichcriticizesEmpedocles'treatment of cosmicmotions

Aristotle comments on the unsatisfactory role of Neikos, adding &CtcF

ai'r&vo'6aov 4tiotco gyXevypna'v?E7 Tr 'oU NeLxoue v5v xcrL7CporEpoV e=r%

rr OctMaq.6"othing further is here said about the Cosmos of Love.

At first glance there seems to be a glaring contradictionbetween this

passageand that in de caelowhereit is said that Empedocles 7capaXpt7rt

59One might imagine what kind of comments would be made if a poem of such

monstrous lack of proportion were preserved from the fifth century. As far as I

see, neither archaic nor classical poetry includes anything comparable. With

Herodotus the situation is different, yet even in him a digression does not fill

one half of the work. Raven places B 35 far too late (p. 346; cf. above p. 125)

but here as elsewhere his reconstruction has the merit of actually trying out whatothers have merely asserted.'I de gen. et corr. 334a5f.; cff. 333 b22ff.

130



the cosmogony of Love.61 Can Aristotle 6um.9mvauxod &'to&xLv? Can

he have it both ways? The interpretersfind it necessary to make theirchoice; which testimony they accept and which they reject depends on

their convictions regardingthe events in the cycle.62To us it seemedthat the statement in de caelohas reason and Simplicius (as well as thebulk of the tradition) on its side. Perhaps however it is not futile toreduce the area of contradictionby observingthat what the statementin de caelo negates is not a Cosmosbut a cosmogony of Love. CouldAristotle find a Cosmos of Love in Empedocles' poem? He could:

&XXo-re4v 0L0crnrtLGo 0pxv' e1 x6crowv.63he translators rightly

avoid rendering x6a[Lovby "Cosmos" ("order" Burnet, Raven,"Ordnung" Diels, "gefiigte Ordnung" Kranz, "unitU d'armonia"Bignone). However, given Aristotle's cavalier attitude to the texts ofthe early philosophers,we cannot exclude the possibility that on thestrength of this or similar passages he regardedEmpedocles as havingcommitted himself to a Cosmos of Love. But where did he find thisCosmos? Inevitably one thinks of the Sphairosyet hesitates to chargeAristotle with so flagrant a misreading of Empedocles' doctrines.Actually it is not so much a misreadingas a misconstruction,and there

is evidence that Aristotle was capable of it. In an argumentof de caeloA he persuadeshimself that Empedoclesand Heraclitus have in truthan eternal Cosmos and that not their Cosmos but only its changingphases (&Loc0e'ae)pass out of existence. On the construction there put

forwardeven the &XaucLqf the world and the mixing of the elementswould be a cosmic phase, or, as Aristotle puts it, the Cosmos "in acertain condition" (?xeLvr &L .rt0ocxwith an adverb64).We may note

61 See above pp. 127 f.62 See e.g. Burnet p. 235 and n. 3; Raven (above p. 125), Bignone 556ff.-564

(whose discussion I find somewhat inconsistent, although it repeatedly comesnear to what I consider the truth), Cherniss, op. cit. p. 195. It will be clear why I

cannot accept Cherniss' contention that in B 35 and 57 "there are descriptions

of the formation (my italics) of a Cosmos during the increase of Love" (n. 210).Against Cherniss' interpretation of Simplicius (ibid.) I must refer to what I have

said above p. 120. Simplicius has no difficulty in saving "Aristotle's consistency"nor is this his primary concern, and although he is quite capable of going off to

the Sphairos he does nothing of the kind in this instance.68 B 26.5.

64 See de caelo 280all-23 in combination with 279bl4-17; the phrases quoted

occur 279b14; 280a20. For the meaning of the argument I refer to Simplicius

ad loc. and Cherniss, op. cit. (note 3) 181 n. 162, with whom I believe myself tobe in complete agreement. The interpretation is complicated by the fact thatAristotle's reasoning is meant to apply simultaneously to Heraclitus and to

131



the similar language (4to(wo Ixetv) in our passage of de generatione,

where we may also observe that Aristotle speaks of the Cosmos"previously" (7rpo'fpov) but not of <v6v> tp6repov x6at.ov. Finaly,

regarding the words 4to,uo gxcev, this impression may rest on thepresenceof identical lines in the descriptionof the Sphairosand of theCosmoscreated by Strife. That there were such repetitions is indeedprobable. Two lines to the effect that Sun, Earth and Sea were not tobe discerned are quoted by Plutarch as characterizing the effects ofStrife, yet the first of these lines (whichis hardly separablefrom thesecond) is by Simplicius said to describe the at$yxpLaL of the elements

in the Sphairos.65 t is true that in the text as given by these authorsthere is a variation in the last two words but this variation is im-material for the cosmic conditions, which would still be 6[uoL.Theremay have been other instances of such repetition. Aristotle read theentire poem, whereas we have to make our guesses on the basis offragments. The explanation here propounded is not offered in theillusion that it is likely to be correct in every point of detail. Its solepurpose is to show whatkind of error,misconstructionor carelessnessmay account for Aristotle's reference to a "condition"of the Cosmos

"previously"66 nder Love, and for the presencein his treatises of twomutually contradictory statements.

We may next turn to the zoogonyof B 57-62. These fragmentsdealwith the origin of living beings in a manner markedly different fromB 96 and the otherpassagesexaminedabove (p. 115) in whichwe foundLove mixing or otherwise fashioning parts of the animal body. Thisdifferencewill engageus later. For the presentwe followthe customary

Empedocles. To be brief (although after a letter from Gregory Vlastos I realize

that the argument calls for a closer exegesis), I take SmX6etvsc. '6v x6at?ov,a 12)= petp4aOmL(a 15) = c-t &Wjx) 'r&vorroqXCeovcuvL6vr6av (a 16) = one of the two

8&a=eq (18f., 20). The condition referredto would in Empedocles' own systembe the Splaairos, yet Aristotle insists that if there are opposite atna (scil. Love

& Strife) there must be in both extreme conditions an orderly, i.e. cosmicarrangement.65 See B 27 with the testimonies as given in DK. Cf. Bignone p. 220f.; 599ff.,and Cherniss in his edition of Plutarch de fac. lun. (926E). What bothers me in

the variation at the end of v. 1 is that the 6xia yuta of the Sun are more ap-

propriate when Strife is separating the elements and Sun, Earth, and Sea have

not yet become consolidated. Conversely &y?a6ovl8oqwould seem in place when

the Sun stands for the element fire (as presumably in the description of theSphairos).6' For "previously" (7Tp6'rcpov) cf. above n. 50.

132



procedure of relating B 57-62 to the "four stages" in the genesis of

living beings which Aetius (A 72) distinguishes. Briefly - and we canafford to be brief since we agree with the prevailing interpretation -

the single limbs of B 57 correspondto Aetius' first stage of O&ca.u-a,uopLeX,while the coalescence of these limbs and the odd combinations

describedin B 59-61 representhis second stage. B 62 introduces us toAetius' thirdstage, the oiXoqueZqithout sexualorother differentiation.For the fourth stage of Aetius we have no fragmentsbut there is noreason to distrust his report that this stage was characterized bysexual reproduction; the living beings must here have attained a

certainmeasure of developmentand differentiation.It may be pertinent to remark that neither Aetius nor any otherancient writer knows anything about a connection of stages 1-2 and3-4 with opposite phases of the cosmic cycle.67For assigning 1-2 to

Love there is some ancient authority 68 moreover the line muoraphre

XOcTM?dLov4A'aye.o 3octLiOVLad4uv (59.1) is best understood as in-

dicating the advance of Love who at this point comesinto more activeconflict with Strife. This would have to be the "second stage". It canalso be taken as agreedthat stage 4 representsor includesthe conditionof living beings in the presentworld.

With what degree of confidence may it be asserted that B 62, thefragment describing the origin of "wholenatured"forms, belongs tothe period of Strife and that the differentiation of the sexes was an

additionalmanifestationof this power?I confess that if a decisionhad

to be made solely on the basis of B 62 I might find this view attractivebut I would feel very uncomfortablewere I called upon to defend it.There are argumentsbut no strong arguments.The atmosphereof thefragment is not particularlycheerful, and 7oXuxXuroqv. 1), which isprobably meant to characterize &to' XOLVwOmen and women, evokes

associations with Strife rather than with Love.69 However, in nept

cpuaec? Love and Strife are essentially physical agents. The idea that

under the rule of Love there must be a life of perfect bliss is not to be

found in this poem (but imported from the other). Championsof a

67 Burnet p. 242f., Bignone pp. 570ff. 577ff., Raven p. 336ff. Cornford, loc.

cit. (n. 2).I88Arist. de caelo 300b 29; Simplic. ad loc. 587,8ff.; Arist. de gen. anim. 722b17ff.,

de anima 430 a27 (three of these passages are adduced by Diels only in P.P.F.,

not in the sometimes unduly shortened attestation includedin the Vorsokratiker).

69 XoOack 'r xct xxxuaoc tESv &auv'0Oe X&pov (B 118) Empedocles says in the

Katharmoi, it seems with reference to his birth into a world ruled by Strife.

133



dual cosmogony who limit Philotes to stage 1 and 2 would hardly

maintain that the Pouyevn vAp67rpppcxnd similarcreatureslead a lifeof consummatehappiness - if they do, what a pity that so few surviveto enjoy it. We know what Empedocles thinks of human life: xocdaCpLv apo 8eL?o&al S'aO' (B 15.3). Even if the aecLk are due to Strife(which I see no cogent reasonto assume) it would still not follow thatStrife must be the originatorof human life. - When in the next lines70we read of a xpLw60evov 5pand a7sup- O6Xovtpo 6toLov tiaOxt ourfirstthought may againbe that Strifeis here at work. But "secondthoughtsare wiser"; it is rash to trace every attraction between 6uot to Strife

as the only possiblecause. Empedocles'wordssuggest that fireby itselfhad the impulse, and if having stated in v. 2 that fire &v'yayehe seesfit to add an explanatory verse (6) which says no more than "firewished to reach its like," we had perhapsbest accept his wordswithoutspeculating about ulterior causes. In fact some of the scholars whobelieve in a dual cosmogony have shown judicious reserve concerningthe 6uoov npq 64LoLovmotif.7' If this motif becomes neutral for our

problem, the xpLOv6pvovs5p for which the 6?otovmotif is meant to bethe cause, seems to followsuit. For all that Empedoclessays is that the

fire "detaching itself" from the earth pushes up the ou'Xo9Vu?trU7rtOL;his thought is not that it severs itself from the other elements in these.What elements are present in the tutnot water, earth, fire (?) - isunfortunately not clear because e'tgo0 can hardly be the correct

readingin v. 5, but the fire is and remainsthe active, vitalizing powerresponsible forgrowth, respiration,waking, sex, and (by its departure)death.72In a word it is the vital heat which the body needs and which

70 Vv. 2,6.

71 See Cherniss, op. cit. p. 190 n. 193 and Millerd op. cit. pp. 34-37; Cornford,The Laws of Motion in Ancient Thought (Cambridge, 1931) 33 (cf. also Zeller 5p. 794 n. 1 and Zeller-Nestle 1.995 n. 1). Note what B 22,1 ff. says about therelation between the main bodies of the elements and the ,kp....&.o. XXOkvrM

(3). B 110,9 is another instance where elements or other entities (cf. H. Schwabl,WienerStud. 69, 1956, 49ff.) wish to return to their y6vvowithout being impelledby either Strife or Love. We may not be able to account for all movements ortendencies of elements in Empedocles but should allow them the degree offreedom to which they are entitled as alive and divine entities.72 See A 77 (298,36. cf. 70; 296,18f.); 74, 81, 84, I realize that only A 70 and 74refer to "origins" but think it fairly safe to proceed from these passages, even if

details remain uncertain. For the concept of "vital heat" and its early historysee my article "Cleanthes or Posidonius? The basis of Stoic physics" Mede-delingen Nederlandse Akademie, n.r. 24) 274 ff.

134



Empedocles needed if in the "third"and "fourth"stages there was to

be organicdevelopment.About the details of this developmentit is not possibleto be specific.

We do not know whether the oiX)ocpueZecame differentiatedinto thebeings of stage 4 orwhetherthe firegradually producedmoredevelopedforms. Those asserting that the more articulate beings owe theirexistence to a "separation"caused by Strife73make assumptions towhich we are not entitled. Sexual differentiation may be due to theinfluence of hot and cold, which are Empedocles' sex determinants inreproduction"now".7 It may be due to causes which we cannot divine

(for Empedocles had more imagination than most of us). "Differ-entiation" is a modernconcept apt to prejudgethe issue. For the restEdwin L. Minar in a recent article has pertinently asked: "Can weattribute to Empedocles the cynicism required to place the entirephenomenonof generationby sexual reproduction,underthe impulsionof Aphrodite, in the epoch of Strife - and indeed the latter part of thisepoch, when the whole world is approaching disintegration?"75Presumably if Aphrodite is active at such a stage sexual union wouldbe her last device for restoring some of the lost unity. The idea has

certainattractions and may be supportedby reference o Aristophanes'speech in the Symposium (where it is held that Plato makes use ofEmpedoclean motifs).76 But it is hardly necessary to remark howspeculative all this is. Minarclearly has a point. Was sexual pL'LL orEmpedocles the weakest or the strongest manifestation of Cypris'power? We may think that we know the answer. If we prudentlyrefrain from coming forward with it because definitive evidence islacking, it seems equally prudent not to base much on impressionsderived from B 62 and on hypotheses about the sequel of the story.77

"I Raven is very confident about the "process of separation" and the "influenceof Strife" (p. 338). So was Dummler, op. cit. (n. 2) 218ff. who postulated for thehuman beings of this period "Scheidung", for those of the other "Mischung".Even with the qualifications which he makes his theory does not accord with theevidence. If in flesh and blood all elements are mixed (B 96) living beings cannotdevelop through a separation of these elements.74 Cf. B 65 and 67 (again P.P.F. provides fuller information), A 81 (AetiusV,7,1; only his first statement can be used with confidence).7 Phronesis 8, 1963, 143. See also Zeller5 p. 795 n. 1.76 The fact that Aristophanes' story starts with three sexes should keep us from

looking too eagerly for Empedoclean motifs; if such are present they must havebeen thoroughly transformed. Cf. Millerd 70.7 Von Arnim (loc. cit. p. 24) is of the opinion that Strife causes the production

135



Two CcoyovtaL,ne of which is the reverseof the other, presuppose wo

xoapioyovtatut of a second cosmogony there is nec vola nec vestigium.We have dealt with sumises, hypotheses, possibilities,perhapseven

plausibilitieswithout anywhere finding what we need, a firm footing.If this is the upshot, are we really entitled to dismiss - without furtherado, as it has become customary - the suggestion of continuity inAetius' report about the four stages? Is the idea so unreasonablethatwe must at all costs endeavor to replaceit by something "better"? Itwould be well if those who discount the continuity realized and

admitted what their decision implies. Surely the doxographers are

capable of confusing the issues; it may be no injustice to think thatthey ran together what originally was distinct and separate. Yet inthe present case some additional assumptionswould have to be made.As we have seen, Aetius and whoever else represents this traditionknow only one origin of the Heaven, one explanation of Sun (archetypeand reflection), moon, planets and fixed stars, earth and sea, summerand winter, light and lightning. They never refer from phenomenain

one Cosmosto identical, analogous, or reverse in another. Did they,after consistently and complacently drawing their information from

one world, late in the game just once wake up to the existence ofanother? No wonder that this proved too much for them, that they gotit all garbled, made one account out of two, put late what (in the

opinion of Burnet and others, though not of Bignone) ought to be

early in the cycle, and early what ought to be late, and when theycame to the parallel subjects of the originof plants (A 70), thought it

safer to confine themselves again to one Cosmos. And if the tradition

goes back to Theophrastus,as it after all should,arewe to supposethathe took account of both worlds but that later all references to the

Cosmos of Love were expunged except in this one instance whereinformationrelating to that Cosmos was preserved yet the comments

distinguishing between the two worlds were excised (to say nothingagain of the u'cepovntp6tepovand the confusion thereby worse con-

founded)? However we try to work it out, r6voq7r6vy7n6vov ppL;

severe demands are made on our sorely tried 7da't;. Since Aetius'account as it stands is not inherentlyabsurd it is perhapsbetter not to

of the oi6opuets '67rO by the xpLv6t?evovip but that Love "eingreifen musste"

to unite the isolated limbs with the trunk. The details of his argument do not

convince me but I would not exclude the possibility that in the "third" stage,as in the "second", &Idaycvo &4[Lovt 8Aa4uLv Love with Strife). We are on very

uncertain ground.

136



manipulate it. I suspect that what for Aetius is the first and second

stage was in Empedocles' own presentation a prelude - treated withrelative brevity78 - to the real genesis. With B 62 the tale (tOuOog,f

not rather <o>uOoq,v. 3) of biologicalorigins began in earnest.To combine impressionsderived from B 62 withthe passage affirming

a dual genesis (B 17,3-5) and to build on this combinationan elaboratetheory of two cosmogoniesetc. may have been justified as long as it

was done experimenticausa. The mistake began when the experiment

turned into a dogma, or somethingvery close to it, with the result thatthe evidence to the contrarywas disregardedor forced into conformity.

But the combination and construction, while ingenious, prove weakwhen confrontedwith the arguments massed on the other side. Thereis the evidence for Love as creator of biological forms. There is the

consensus of Simplicius with the balance of the tradition about oneCosmos brought into being by Strife. Thereis the accord between B 35and B 71 plus its following - an accordwhich if accepted proves Loveto have created the C(Oaf our world, if rejected produces identitywhere there ought to be antithesis. Finally there is the affirmationinB 26,7 that we are moving toward the ?v and toward a cr6?eaO. Yet

there are no fragments relating to another Cosmos, no statementsregarding a CcpoyovExy Strife. Must we still speak of the misinter-pretation through which a cosmogony of Love was read into B 35 orabout the awkwardnessof having our world and all that belongs to it,includingourselves,come into existence soon after the cycle starts andspread out comfortably through half of the poem - then to be wipedout by the progressof Strife (althoughof this appallingcatastrophenofragment, no testimony, not even any of the later writers interested in

science fiction transmits a sound)? We have seen in more than one

instancewhat embarrassingdifficulties arise for a courageousattempt

to work out the details of the dual cosmogony. It seems we havereached the point whereeven postulates that Empedocles "must"have

taught the same in the =?pt pu=a and the KmOapEoannot save the

78 I infer this not so much from the observation that the same few lines are

quoted repeatedly and independently as from noticing how little in the way of

concrete detail Simplicius in Phys. 371,33-372,8 found in Empedocles' text. He

evidently did find a statement that the beings composed of mutually fittingparts 4IANLvbut even for the idea of compatibflity he had to invent the illus-

trations (that the liver turns food into blood was unknown long after Em-

pedocles; cf. Aristotle de part. an. 666 a 25 f.). Incidentally, I consider reproductionas out of the question for the "second" stage; for this is a characteristic and new

development of the "fourth".

137



dual cosmogony or remove Love from steering the course of events in

our world.But does not Empedocles himself state in plain and unambiguous

words:

8OL! &COvnT@V y6veacs, Ao 8' &itX?L4L

rn~vuCV ap 7arCvTV UvoAoqXTEL' oeXCL Ew,

X8 '7CXLV &0CcpuO[.L6V)VprepCTMat t'enn . (17, 3-5).79

and does not the same fragment after another eight lines plainly refer

to the cycle? xix?,ov is indeed the last word of the first thirteen lines of

B 17 which form a kind of unit, though it may be doubted whetherRaven was justified to present these lines under the heading "The

cosmic cycle" and to use them as an introduction to this doctrine.80 At

the very best the cosmic cycle is in the offing. For what Empedocles

wishes to establish in B 17,1-13 is something quite different, and what

the entire fragment B 17 establishes is again another tenet. We still

find ourselves in the early section of the poem which introduces the

elements and clarifies their nature. In vv. 1-13 Empedocles is particu-

larly concerned with their relation to "becoming" (yLyvzaOCL), and we

can see how the argument of these verses works up to the conclusionof vv. 9-13. The first two verses...

-6're ,uev yap ev u ,uovov eZVaL

?X kx6o,vw, 6TOTC ' xGCL69t 7rXov K e'Vo6 elVXL

are taken up in 9-12 where on the strength of this thought it is con-

cluded that the elements enter into genesis (yCyvovtau)and are not

eternal.8' This is one aspect of their nature ('rn ?v ... ). Similarly

vv. 6-8 xol -x3ra'&?ocaaovt rtX. are taken up in 12f. to establish the

other part of the conclusion: the elements are, g'aLv; they are eternal(aoW datv balances ytyvov-aouf 11); they are unmoved or unchanging

(&x1v-oLbalances ovi... e[Me3os acx1vof 11) but axLvITOL xXO'TcLxiCX?ov,

i.e. they are "unmoved" because their movements or changes follow

always the same pattern and rhythm. We cannot fail to remember

that &xtLvov, eipreov, and the negation of yLysvaCcOOLre Parmenidean

7" OpuqgOclaa8piTrr Mss. of Simplicius. OpOcp0e-Las Panzerbieter's, ML&TCr

Scaliger's emendation. Although OpuepOctaaas found a defender in Wilamowitz

and the verb is used in a very appropriate context by Plato (Parm. 165b4: the

breaking up of the One!) I think we need OprOveLaao balance t(xrel of 4.80 Op. cit. p. 326f.

81 Perhaps rather "not stable".

138



concepts attaching to his Being.82 By making so large concessions to

genesis - i.e. to the genesis of first principles _83 Empedocles has ineffect surrenderedmore of the Parmenideanposition than one mightexpect and than most interpreters realize;84what he saves he savesonly by introducinga new definition of J&xEv-oq.Onemight even call itan equivocation. It will now be clear in what sense he is interested in axuxxoc85 nd also that his topic in these lines is the behavior of theelements but not cosmic phases. What may be said, although only inparenthesis, is that the behavior of the elements on the cosmic scale,where they are separatedto form a Cosmosand then brought together

again to create 4Fpm,s an extension or projectionof what is here statedabout their characteristic cyclical pattern. The remainder of B 17(whichincludesa repetition of vv. lf.) is preoccupiedwith the identityand nature of the elements as they are and manifest themselves in theworld at large, where even Love can be recognized.86Similarly B 26,a fragmentwhich repeats many lines of 17, 1-13 and which, we happen

82 &xlvqoV Parmen. B 8,26 (in closest connection with "no genesis"), 38;1vc8ov ibid. 30. Cf. Jaeger, Theology p. 137f. and n. 36. Although &xtv-yroLn

the Empedoclean passage seems to have the meaning "changeless", it is off the

point to recall that xtvvns is for Plato and Aristotle the generic term for allchanges, including the qualitative. The oblique reference to Parmenides sufficesto account for Empedocles' use of the word.83 Excluding however Love and Strife (cf. B 16). The converse thesis is thatmortal beings "are" even before birth and after death (B 15). Cf. also Holscherpp. 30f on B 26,3ff.

84 Attention is as a rule too exclusively given to the genesis of mortal things

(e.g. Burnet p. 228, whose reference ibid. to B 17 does not do justice to thecomplexity of the doctrine). In B 9 Empedocles explains that in using wordslike yevcaOaL e follows the convention; in truth there is only ,UEksand itsopposite. Although he here looks at the question from the point of view of the

product (compound formations), the principle would apply also where, as in ourpassage, the elements themselves yEyvov-Tcx.85 "Identity in change" invites a comparison with Heraclitus, which I musthere forgo. It will, however, be evident that the xvxXoshere is Empedocles' onlymeans of preserving at least a fraction of the Parmenidean legacy. If we realizethis we shall hesitate to think that in the physical poem Empedocles might havedone without the x6xXo;and that we have to look to the Katharmoi to find theexplanation for this motif (Raven p. 348).88 See esp. 21-26. It is probably unnecessary to say anything about Em-pedocles' frequent resort to the "empirical" aorist (e.g. 17.1 f.). Has this featureof his language influenced interpreters who relate to the cosmic cycle statements

that apply to things in general? I doubt it. Note the juxtaposition of suchaorists with present tenses in 17,4f. and 21.8. B. A. Van Groningen, In the gripo? the past (Leiden, 1953) 21 deals with closely related subjects.

139



to know, had its place later in the poem, treats of the elements as they

act and behave in the present world. It actually establishes the sameconclusionas 17, 1-13 but is content to use as premise for it vv. 7f. ofour fragment,87not the somewhat more pointed and extreme vv. 1-4.If necessary, the use and context of these lines in B 26 may give us aperspective on their meaning in B 17, and what we learn from themapplies also to the more radical formulation of the same thought inB 17, lf.

Now that we understand the direction of the argumentin 17, 1-13,we are in a better position to appreciate the crucial verses 3-5 (above

p. 138). They are not the conclusion to which Empedocles works upbut a corollary of what is stated in vv. lf. In other passages Empedo-cles thinks it adequate to say that y6veaLs (or cpu'aL))= [A6&vocVTOw =

aLahXLq.88 Here he goes further. Developing the implications ofvv. lf.,89 he realizes that on the way toward the ev there is not onlygenesisbut also, when the processof unificationreaches its logical end,destruction (of the compounds); correspondingly in the process of

aLtwpUaocxLhere is before the complete separation of the elementsgenesis of compounds, i.e. mortal beings. Thus the exploration ofgenesisas it affects the elements throws new light also on the genesisof

Ovqj-i.And as genesisfor a post-Parmenidean philosopher s a problemof the first order of importance t surely was worth-while to present thisadditionalcontribution.As it stands it has the characterof a discoveryin the realm of pure thought, being derived from the basic physicalassumptions. Let us appreciate the contribution as such and let us notread into the lines more than they actually say.

Still, could not a thought of the kind, even if it arises and is pro-pounded as a corollary, be followed up? Surely, it could, but we haveno right to consider Empedocles as committed to this. If the cos-mological cycle is an extension of fundamental physical tenets, onlyinterpretation can show which tenets were extended and applied.WasEmpedocles satisfied with presenting mortal creatures as springingfrom the mixture of the elements and being the products of Philotes90or did he apply the more ambitious theory of a dual genesis? Let the

87 26,5f. On 26,7 see above p. 126.88 See esp. B 8,9. Similarly, while speaking of 8&aqupaOa 17,17 f. he is contentto refer to the elements as the outcome of this process.89

Mcp 2; 8txo0Xvcov 5.90 See e.g. the occurrence of the thought &?XXoteL?v 00&CXInOauvepX6lev' EtL tv

&7rav-x 17,7 (repeated with a variation B 26.5) in 35.5.

140



fragments supply the answer. The championsof a dual cosmogony are

convinced that the living beingsof the presentworldorderareproducedby Strife. But Empedocleshimself says the opposite:

T6aa' 6aoxv5vyeycML auvopp.oaOkvt' 'Appo8trs (B 71.4).

We have found ample additional prooffor Cyprisas creator of organiclife in this dispensation."' If B 35 too, as we maintain, relates to thepresent world and its animals, we have one more piece of evidence. Ifthis fragment, as others think, refers to the opposite phase of the cycle,it proves in combinationwith B 71 and the rest of the evidence that in

both phases genesiswas sALqand the work of Cypris:nec sine te quic-quamdias in luminis orasexoritur92 it is hazardousto quote Lucretiusas shedding light on Empedoclesbut it seems that he speaksthe truth).Strifemay destroy, but it does not create organic life. I should still notsay that Empedocles "overtaxed" his imagination by introducing the

8oL) yva but should prefer to take the verses 17, 3-5 as what they

present themselves, to wit a brilliant and exciting new contributiontothe problemof y'veat. It is entirely our doing - not to say, our mistake-if we sever them from their context and find in them a promise.

Since ourinvestigation has led us to considering he relation betweenthe first part of nepLpiaecog,where Empedocles introduced his prin-ciples, and the cosmologicalsection, it may be desirable to add a fewobservations on this topic. In a rough and schematic way it may besaid that in the first part Empedoclesmakes clear which of Parmenides'revolutionary tenets he adopts and from which he departs. He toorejects y'y eaOo and 6XXuaOo,L,et by positing a plurality of "being"things he is able to save the phenomenalworld. We have seen that healso accepts Parmenides'&xEviqrovor his own principles, redefiningit,however, in such a manner that many processes of the phenomenalworld may be explained and again "saved". Now while to save thephenomena without sacrificingphilosophicaltruth and rigorof methodwas certainly a great achievement, it may nevertheless be held thathis principlesand his pattern of explanation passed their crucial testonly when with their help he could deduce - and, if possible, deducebetter - what earlier physicists had deduced from other principles,namely how 8VX'yevovtort vuivaopwi.?v&7raw'cx.do not wish to pressthis point; someone maintaining that the explanation of things andprocesses in our environment is the essential achievement, the cos-

91 See above pp. 113 f.92 Cf. von Arnim. op. cit. p. 22.

141



mological cycle an opus supererogativum, could not easily be refuted.

One thing however seems certain. Once Empedocles embarked on agenetic account he must for his elements preserve the pattern

&?XotE &ViDLX6-nCL oruvepztcev' tq evM TvOx,

&O x&iLX x=a' yopeOp.?vo Netxeoq9x0et.

For if this were abandoned, his version of the Parmenidean "unmoved",

the &xLvIrroL MCIx xCxXov, would be lost, and the elements would in the

cosmogony and zoogony behave without rhyme or reason. Surely, the

TroT;0Vv... gvfUEfN pfvov Zlvocr avepXCOFv' Es ev &rxvTa led, when

carried to its ultimate length, to the EcpaZpo ovf 7tspLty6 yC v93 andthus to the (traditional) undifferentiated arche of all things; the 8'tX

ypopeZaOL hrough the agency of Strife produced, when projected on

the large scale of world history, the Cosmos, and the vteaOyuaLof the

elements accounted for the countless shapes of mortal beings. But this

was not enough. Just as the eternal elements and as the functions

posited for Love and Strife had to be preserved in the historical account,

so the validity of the &xNvntoLaro-c i)Xkovmotif had to be upheld and

vindicated on this larger historical canvass. Without it the whole

account would have been a dilettantish affair. Far then from being"arbitrary", not "unavoidable" and an unnecessary complication

intelligible only with the help of the Katharmoi,94the xix?,oqwhich

guarantees the &AxvvTovs the idea that gives the historical account

philosophical dignity and stability, saving it from being just another

3po'r&vo'ocdevoid of 7dartqOX-njO r another xO6a[o e7eOv aXI64.

Empedocles knew something far better than the positing of two

{?opcpoLone of them misconceived). His is a philosophically respectable

account which safeguards Being and a rhythmical pattern that he can

equate with the Ox'vC*y-ov.We have come to know some of the reasoningwhich creates ntanLqand establishes the "truth" of the cycle; yet

B 17, 1-13 is merely the last step of his reasoning. To trace the nLta'L

through the early fragments would mean to write another paper; here

it is not possible Xoyou ?o6yoveEoXetUSiLv. Even on a priori grounds it

would be hard to maintain that the aoL OWITCvyev'Ecs. is as vital for

98 I wonder whether "solitude" is not after all the meaning of [io'vL B.28.2). Jae-

ger, Theology 141 f. argues for "rest" and it is indeed attractive to find a thought

which may be linked to Parmenides' 9,vLekv... ,kve (B 8,30). In H.S.C.P. 63

(1958) 277 I supported the idea of "rest" in the Sphairos. I now incline to thinkthat there is no place for rest in the cyclical pattern. Cf. Holscher pp. 1Of.

94 Raven 348.

142



the entire scheme, including the cosmological cycle, as the Oaxvij'oL

xoclrrtx0ov. However we have thought it better to make our decisionnot on a priori groundsbut on the basis of the empiricaldata. It shouldperhapsstill be emphasizedthat in B 26 wheresome of the thoughts ofB 17 are repeated and in a fashion summarized the 8oL' Ovwly&V6vec6

does not reappear. Empedocles may here - shortly before starting onthe "cycle"? - pull together what is essential. &xtvnyrotxo&r xtxXov,

being one of the essential ideas, is repeatedin B 26.

Appendix

It seems fitting to add a few words about the sequenceof the fragmentsyet the scope of our remarks must remain limited and we must keepclose to the arguments put forwardin the main part of the paper. Forpassages which have entered into our reconstruction Simpliciusfurnishesthree important "leads"which we must on no account forsake.B 35 precededB 98. B 96 had its place in the First Book, B 62 in theSecond Book.95We need not here discuss the reasons for which Dielsplaced 62 prior to 96; that a future collection of the fragments mustdepart from him in this point is obvious and has been repeatedlystated. 96 96 clearly carries 98 with it - whether also 97 and 99 - 102, asKranzsuggests, is more of a question.97The real affinities of 96 and 98are with the other fragments specifying Cypris' workmanship inmixing and compounding biological forms; of such we have found acertain number.98On the other hand, I fail to see a close link betweenB 96, 98 etc. and physiological processes like respirationand smelling(B 100-102). However the problem on which we are here touching israther complex. For it may well be felt that the account which treats

96 See the testimonia before these passages in DK. Simplicius' statement aboutthe "relative" place of B 73 is helpful too. - Kranz, Empedocles (Zurich, 1949,142, 144) places B 35 in the Second Book, B 96 and 98 in the First. I think itbetter not to ignore Simplicius' indications.96 See e.g. Wilamowitz Hermes 65 (1930) 245 and Kranz in DK I. p. 308.I wonder how Raven who assures us (p. 322) that "the arrangement of Diels isnow generally accepted" would justify the place which he gives B 35 (p. 346)at the end of the cycle and long after B 62 (p. 338; cf. p. 339) and B 98 (p. 335).What is fatal for this reconstruction is not the departure from the "generallyaccepted" order, on which Bignone and Kranz (in his monograph) had tried toimprove, but the conflict with Simplicius' testimonies.

97 DK I p. 308. Note how different the description of physiological conditionsin B 100,1 ff. is from the passages put together below n. 111.98 See above pp. 113 f and also below note 111.

143



seeing as the leaping forth of a flame fromthe eye (B 84) would forma

good sequel to the formationof the eye described n B 85-87. Similarly- and even more persuasively - the fragments that relate to thinkingand to the blood as organ of thought might attach themselves to the

definition of the blood as the equal (or almost equal) mixture of allfour elements;99for it is evidently the operation and mixture of these

elements which determine the quality and adequacy of our thinking.Here we do not get beyond the point of discerningpossibilities.

As indicated, the fragmentsdealingwith mixture and compositionin

living organismsappearto cohere morefirmly. In the light of what has

been argued above (p. 113) we may regard them as following closelyupon B 71, which they implement, and may look upon 71 itself as notfar separatedfrom 35. One and the same idea of Cypris'workmanshipdominates all of these fragments. We shall come back to them.

B 62 is quoted by Simplicius from Book II where he found it iTpo

'n Tv Vapetcov xal yUVaxLXELWVaw[oc-v &xapOpGae&).We do havefragments dealing with sexual differentiation100 ut must not hasten

to the conclusion that their placewas shortlyafter62; forthe fragmentsclearly refer to reproductionand the sex determinants now, a topic

different from the original formation of the sexes in Aetius' "fourthstage". It is true that in this stage reproductionbegan, and since the

"evolutionary" account here reached its end Empedocles may have

gone on from the first origin of reproductionto the laws that governit now, i.e. from a genetic to the systematic treatment of this topic.This idea has undeniableattractions but for the time beingit representsno more than a possibility. Canwe be sure that Empedoclesdid not

keep all physiological processes - sense functions, respiration,repro-

duction, perhaps also nutrition and digestion - together in his

treatment?If we insist on keeping B 62 in Book II, is it possible to say what

preceded it? Those who assign stages 1-2 and 3-4 of the biologicalevolution to opposite phases of the cycle must, if they wish to be

**B 98; 107f. Theophrastus' report A 86,11 creates a presumption that differ-

ences of man's thinking capacity were explained as a sequel to B 98. This would

have its bearing on B 103, 104; yet 105-109 may also have had their place in

this context. It is not easy to remove B 109 far from B 107 f. Everything relating

to 76pot and 'U76ppoLaLs likely to belong to a different, i.e. later part of the work.

100See esp. B 63, 65, 67 (A 81). The arrangement of the fragments in DK, while

not above criticism regarding the sequence 62, 63ff., has the merit of puttingembryological material (B 68-70) close to B 63, 65, 67. On the place of B 64

I do not venture a suggestion.

144



consistent, separate57-61 by a considerabledistance from62 (between

them Empedocles would have described how power and creativeinitiative in the Cosmos passed from Love to Strife). On the re-constructive here presented B 57-61 would find their place shortlybefore 62; they would form a relatively brief prelude to the develop-ments whose description begins in B 62 and ended with the origin ofsexually differentiated beings of the kind that populate our world.This would mean that B 57-61 had, like 62, their place in Book II(or, if B 62 was the beginningof Book II, at the end of Book I).

If we did not know from Simpliciusthat B 96 formedpart of Book I

and B 62 of Book II, our inclination would probably be to keep thefour evolutionary stages close to the phases of the cosmic cycle andto allow for some distance and the accommodation of other materialbetween the cycle and the fragments dealing with the composition ofbones, flesh, etc. Natural as this inclination is, it must be resisted.Simplicius' indications point in the opposite direction, and if we letourselves be guided by him we can understandwhy Empedocleswoulddiscuss the compositionof tissues etc. shortly ("shortly"being of coursea relative term) after taking us through the "cycle". In fact we only

need to recall some of our earlier conclusions.'0' The cycle led throughthe creation of the Cosmos by Strife to that of the living beings byLove, and may well have ended at this point or soon afterwards.'02The creationof the living beings was a mixing of the elements. Havingdescribed it in such terms, (B 35) Empedocles had good reasons todevelop this motif: If (havingheard now, and also before this'03)thatthe living beings,numerous and varied as they are, have all been fittedtogether from the elements by Aphrodite, you still are not convincedand fail to see how this is possible, (listen to my detailed account).104

In the context of his scheme, origin through mixture was the moreimportant idea, its clarification more urgent than "evolutionary"origins. We may also remember how essential the concept of mixturehad been in Parmenides' genetic account from which Empedoclesborrowedthe idea.105If Parmenidestoo knew the formationof animals

101 See above pp. 112-116, 127-130.102 See above p. 129.L08B 35,5ff., 15ff.; cf. B 21-23. (The parentheses in the text are meant toexplain, not to supplement the thought).104 B 71.

105 Parm. (Vorsokr.28) B 12-18; A 37. I would not go so far as to say that inParmenides too the Cosmos came into existence through a "separation" of the

145



from single limbs,106one would think that this approach remained

peripheral even in the "Way of Opinion", where "mixture" was ofcentral importance. Still evolutionary speculation about the origin oflife had ever since Anaximander'l07ad its placein Presocraticthought.We can understand that Empedocles wished to give his own versionof evolutionary developments but also that he kept them apart fromthe - presumably more recent - motif of mixture, whose inherent

possibilitieshe wished to carry beyond the point to which Parmenideshad gone with it. Since we do not read an "introduction"to hisevolutionary account we cannot say preciselyhow the two approaches

connected, contrasted,or supplementedone another, in his own mind.That he indicated for both of them the cosmic phase to which theybelonged was no more than natural.'08Although we must not imposeour insights on Empedocles,we may as well observe that the accountof Love's mixings is less "evolutionary" han the so called fourstages;for what the formeractually does is to analyze composition,givinguschemical formulasor other definitionsof specificmixtures. This is thereason why Plato and Aristotle,both of them not evolutionarythinkersin the Presocraticsense of the word,availedthemselvesof this account,

continuing physiological analysis along its lines.'09 The successivestages could not be adapted to their biological system and were as amatter of fact discarded.

It was necessary at the beginning of this "Appendix"to leave im-

portant questions concerning the organization of Empedocles' poemunanswered. Yet as soon as we turn to the actual fragments there is

no reason for despondency. What is important - and also quitefeasible - is that the cosmological fragments be kept together, being

two basic and opposite tLopgot;yet it stands to reason that they are separatebefore the daimon mixes them (B 12). If not only the Ovqr but e.g. the moon

too originates through the daimon's mixing - a possibility suggested by A 37 -,the function of Cypris would in this respect differ. Even so there remains enoughsimilarity including, especially on our reconstruction, the sequence of xoaso-yovia and 4coyovta. In fact according to Simplicius in de caelo 559.26 Parmenides

carried his account of genesis ?iXpl rCov toptev 'riv ~44cv which suggests a

sequence corresponding to Empedocles: 1) x6apoq and its parts, 2) Cx, 3) the

.L6ptc (or ?W8i) of the latter.

106 One wonders how much reliance should be placed on Censorinus' report to

this effect (de die nat. 4, 7, 8.; 28 A 51 DK).

107 See 12 A 11,6; 30; see also 21 B 33 (Xenophanes).108 See B 59,1; B 73,1.

109 Cf. "Tissues and the Soul" (n. 17) 445ff., 459ff.

146



given a place before B 35,110and that the passages relating to Cypris'

workmanshipbe likewise treated as a unit,'1' instead of being inter-spersed between others of quite different complexion. Pieces dealing

with the sense functions, with reproduction "now", with respiration

and related physiological topics should be removed from the two

groups just mentioned."L2 Moreover our investigation suggests that

the First Book included beside the exposition of general physical

principles- a subject treated with no hurry and without qualms about

repetitions - the entirecosmiccycle and the accountof Love's mixtures.

Bearing in mind that each of these subjects is a large one, and that

there were excursuses in the cosmogony, we should when asking wherereproduction, respiration, sense functions etc. found their place give

first consideration to Book II. These subjects may well have been

treated with less constant referenceto Love and Strife; their discussion

introducesnew concepts and relies on mechanicalmotifs and technical

analogies (without using Cypris as the technician)."3 If we think of

the physical principles,the genetic cycle and the mixtures as the coreof the work and as essential for its first conception and first presen-

110 In DK they are found after B 35. The fragments in question would includeB 39-49 although we should always allow for the possibility that one or the other

of them did not occur in the context which prima facie seems the most likely)

and probably also B 52-56. See above p. 127 for an excursus prior to B 35. B 50

is of dubious authenticity. On B 51 I should not wish to commit myself. I am

aware that in this and the next note I am more than once returning to ar-

rangements found in Karsten's or Stein's collection of the fragments.

111 It has been said above n. 52 that B 33 f. belong to this group. Other fragments

of the kind are B 71, 73, 75f., 82 (cf. p. 115), 83 (?), 85-87, 96, 98. B 97 is out

of place (whether or not it should be close to the "evolutionary" fragments is

difficult to say; faute de mieux it might be placed after B 62, if it is not better to

assign it to the A section). An interesting problem is presented by B 95. In spiteof appearances I should not include it in the group of fragments specifying

Cypris' mixtures and compositions but regard it as a reference back to this

context made in a later section whose topic was vision.112 In DK B 89-94 are interlopers in the record of Cypris' workmanship to

which the fragments preceding and following them belong. B 89 looks like an

introduction to the theory of sense perception, a subject to which B 101 and 102

must also be related. B 74 and 79-81 are not well placed.113 B 84 and B 100 provide obvious illustrations, although I have for the former

of them indicated an alternative possibility (above pp. 143 f.) and must say that

it would in some respects be more satisfactory if we could regard Cypris as the

grammatical subject of the main clause (7ff.; cf. 7 with 73.1). As the fragmentis now printed in DK this is hardly possible; yet in vv. 7 f. not everything is as

smooth as it has been made to look.

147



tation, we may as well leave open the possibility that the topics in the

latter half of 7?p' puare were worked out by and by, as the spirit orthe Muse moved Empedocles. Being aware that we are entering therealm of hypotheses, I do not wish to pursue these thoughts further.We cannot know whether or not this epic is one of those which "grewby accretion" 114 my plea would be that we do not rashly exclude thispossibility.

Universityof Wisconsin,Madison, Wisconsin.

114 Van Groningen's book (cited above n. 8) has opened fascinating perspectivesregarding this subject.

148

love and strife in empedocles cosmology

Documents