The Parmenides of Plato and the Origin of the Neoplatonic 'One'

E. R. Dodds

The Classical Quarterly, Vol. 22, No. 3/4. (Jul. - Oct., 1928), pp. 129-142.

Stable URL:

http://links.jstor.org/sici?sici=0009-8388%28192807%2F10%291%3A22%3A3%2F4%3C129%3ATPOPAT%3E2.0.CO%3B2-H

The Classical Quarterly is currently published by The Classical Association.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available athttp://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtainedprior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content inthe 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 athttp://www.jstor.org/journals/classical.html.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printedpage of such transmission.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academicjournals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers,and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community takeadvantage of advances in technology. For more information regarding JSTOR, please contact support@jstor.org.

http://www.jstor.orgSat Sep 8 01:40:12 2007

THE CLASSICAL QUARTERLY JULY-OCTOBER, 1928.

T H E PAKIIJENIDES O F PLAT0 -4ND T H E ORIGIN O F T H E NEOPLATONIC 'ONE.'

THElast phase of Greek philosophy has until recently been less intelli- gently studied than any other, and in our understanding of its development there are still lamentable lacu+zne. Three errors in particular have in the past prevented a proper appreciation of Plotinus' place in the history of philosophy. T h e first was the failure to distinguish Neoplatonism from Platonism: this vitiates the work of many early exponents from Ficinus down to Iiirchner. T h e second was the belief that the Neoplatonists, being ' mystics,' were necessarily incomprehensible to the plain man or even to the plain philosopher, T o have encouraged the persistence of this superstition in the nineteenth century is the least pardonable of Creuzer's many sins. The third was the chronological confusion involved in the ascription to Saint Paul's contemporary of the works of the pretended D~OIZJS~ZLS contain fully-A~eopngitn, which a developed Neoplatonic theology. Though the fraud had been exposed by Scaliger, these writings continued down to the beginning of the nineteenth cen- tury (and in certain clerical circles down to our own day') to be used as evidence that the ' N2oplatonic trinity' was an inferior imitation of the Christian one. When this false trail was a t length abandoned the fashion for orientalizing explanations persisted in another guise : to the earliest historians of Neo-platonism, Simon and Vacherot, the school of Plotinus was (in defiance of geographical facts) ' t h e school of Alexandria,' and its inspiration was mainly Egyptian. Vacherot says of Neoplatonism that it is 'essentially and radically oriental, having nothing of Greek thought but its language and procedure.' Few would be found to-day to subscribe to so sweeping a pronouncement; but the existence of an important oriental element in Plotinus' thought is still affirmed by many French and German writers.

This is 'proved' in two ways: In the first place, Plotinus is said by Eunapius and other late authorities after him to have been born in Egypt (though Porphyry knows nothing of this2) ; we do know from Porphyry that he was a student at Alexandria, whose fame as a centre of Greek learning attracted young men from all parts of the world ; that he joined Gordian's

1 Cf, e.g. J ~ l l t r b t ~ c kfiir Philosophie r b . S~ekulative by blood, and probably did not think of him a s Tkeologic, X I I . 483-94 ; XIII. 82-106. an Egyptian in any sense. I n face of this nega-

Porph. Vtt. Piot. I : odrc x e p i T O O -yPuous RE TOO tive testimony of his closest disciple, how much 8 r ~ ~ e i d a r O ~ T Ex e p i weight are we justified in attaching to thef i v e i x e ~ o ofir6 x e p i TGU -you€'wu r1js narp l8os . Further on (iLid., 10) Porphyry evidence of a hagiographer like Eonapius, who twice designates a s ~ b uB i - y L ; n ~ ~ o uthe priest in was not born until three-quarters of a century whose company Plotinus visited the Iseum. after Plotinus' d e a t h ? Under the influence of Since this description serves to distinguish the the fourth-century belief in Egypt as the home priest from Plotinus, we may infer that Porphyry of all wisdom, and in the absence of all positive certainly did not regard his master as an Egyptian information to the contrary, nothing could be

NOS. 111. IV., VOL. XXII. I

130 E. R. DODDS

expedition to the East with a view to studying the philosophy of Persia and India, but failed to get there; and that on one occasion he accepted the invitation of an Egyptian priest to take part in a spiritualistic seance arranged by the priest a t ' the Iseum in Rome.' Add to this the fact that in one passage, dealing with the theory of E e a ~ t y , ~ he expresses his admiration of the Egyptian hieroglyphs; and that (like Plato) he compares philosophy to an initiation into the mysteries-perhaps in his case the Isiac m y s t e r i e ~ , ~ and perhaps not. Even so might an Englishman, educated and perhaps born in India, take advantage of a punitive expedition to study comparative religion on the North- West Frontier, and of an invitation to a Tantrist temple to see something of Indian devil-worship ; he might even praise the sacred sculpture of Renares, and adorn his style with occasional allusions to the car of Juggernaut. We know with certainty that Plotinus' name is Roman, and that he wrote the idiomatic Greek of a native speaker ; he ntny have been an expert in Egyptian religion, but all that he tells us on the subject could have been picked up on a Cook's tour.

The second method of proof is much simpler, since it rests entirely on negative evidence. Certain thoughts and points of view are shared by Plotinus with earlier writers who have been given their passports as ' t rue Greeks '; these are deducted from the sum total of Plotinus' system, and the ~esidzizuizis labelled ' oriental.' Three assumptions are involved in this labelling: That the labeller has a safe criterion for distinguishing the ' true Greeks ' from the half-breeds among Plotinus' predecessors; that he is intimately familiar with the whole of ' t rue Greek' literature, both with what has survived and with what has not ; and, lastly, that Plotinus never invented anything for himself, but composed his works by copying out passages from 'authorities.' Clearly these are large assumptions. I f we are to avoid making them, we must find convincing parallels between specific passages in Plotinus and specific passages of non-Hellenized oriental religious literature. Perhaps the orientalists will one day help us there. Until such parallels are forthcoming4 it seems to me wisest to maintain a position of E ' T O X ; on the whole question, and in the mean- time see what can be made of possible sources nearer home. more natural than that the fact of Plotinus' son in his review of Cumont's paper, Theol. early studies at Alexandria should give rise to Ltterat~rrzeitii~r,a( rgz j ) , No. 21, 485.7. In this t h e legend of his Egyptian birth. T h e value of connexion Rlr. A. D. Nock has called my atten-the further statement that h e was born at Lyco tion to Theo Smyrn. Expos, re?, litath. 14. 18 sqq., seemed doubtful even to Eunapius ( A U K & radrrjv Hiller, where a n elaborate parallelism between duo ,ud~ouuc~~ a i ~ o c 6 B e u ~ i u c o s $rhbuo$os IIop- philosophy they e the Platonic and mysteries is $ I ) P L O S T O G T O OLK dulypa+er paB7~7js T C ai i~oOY E - built on Plato, Phd. 6 9 ~and Phdr. 25oc. Such y e v ; a @ a ~ kdywv, ~ n i rruvca)(o)ra~hrair b v giov i i i 'rav~a metaphors are common from Plato onwards : 9 ri)v ~ X t i a ~ o v e.g. Chrysippus calls discourses about the gods T O ~ T O U[actually only for six years], Vit,p k i l . 45 5). r e X ~ ~ a i(Vet. St. FY. 11. 1008, Arnim). I Ibid. 3 and 10. "EZ. V. viii. 6 . 4 Up to the present we seem to have little o r 3 J . Cochez (in Rev. Nto-Scolastiqite XVIII. no evidence that before they were touched to

[ rgr r ] 328.40, and A%fi.iila)~ges intellectual life by contactd'Histoive 0,qevts ri with Greek cu l t~ i re Clz Jboelter I . 85-101) claims to have proved this. the peoples of the Near East achie-led anything H e is followed by F. Cumont in ,5fonurne?its deserving the name of abstract thought ; their Piot XXV. 77 sqq. ; but the weakness of their thinking hardly existed outside the myths which case has been effectively exposed by Erik Peter- embodied it (see Th . Hopfner, Oriznt ts. Griecizische

THE PARAlENIDES A N D T H E NEOPLATONIC ' O N E ' 131

This was in substance the advice of Zeller, who called attention to the existence of such sources in Stoicisn~, Neopythagoreanism, and Middle Platonism. They are scattered and for the most part fragmentary. In the last fifty years German scholars like Schmekel and Praechter have done a good deal to illuminate them and bind them together ; but easily the most important contribution to the question since Zellcr is contained in Werner Jaeger's brilliantly written book Nenzesios aon Enzesa-a book which has not yet received in this country the attention it merits, perhaps because it was published on the eve of the war. Jaeger shows, in my judgment convincingly, that some characteristic Neoplatonic doctrines, in particular the notion of a & v 6 e a p o ~ -

the universe as a spiritual co~zti~zztzntz extending through a definite series of media from the supreme God to bare Matter-go back to the Platonizing Stoic source which the Germans have agreed to call Poseidonius. Jaeger indeed would be more precise, and say that most of them went back to Poseidonius' commentary on the Titnaeus-the epoch-making commentary thanks to which, he tells us, the Plato of the Tinzaezts is the Plato of Neoplatonism and of the Renaissance. H e concludes that Poseidonius was the true father of Neo-platonism; had but Poseidonius found a place for the Platonic Ideas there would have been nothing left for Plotinus to do ! l

I t is apparent that Jaeger has here allowed his discovery to carry him too

fast and too far. Poseidonius left out something far more essential to Neo- platonism than the Ideas (which Plotinus might a t a pinch have dispensed with had he not found them in Plato): Poseidonius left out the One. If there is one doctrine more than another which the tradition justifies our accepting as eclzt-Poseido~zischcs it is his definition of God as ' a fiery breath which thinks ' ( ~ v a G p avoephv ~ a l~ v p G 6 e q ~ ) ,which has no shape of its own, but changes into what it chooses and assimilates itself to all things. Poseidonius' highest principle is thus material, immanent (though in varying grades of immanence), and of the same stuff as the human intellect. Eut the Plotinian doctrine of an undifferenced ground of all existence, transcending not only Matter but Mind, creative without will or causality, unknowable save in the unio ttzystica, having no character save the character of being a ground-this is the part of Plotinus' system which has at all times impressed itself most deeply on his readers.

I t is also-and very surprisingly, I think-the part which historians have found most difficulty in accounting for. Zeller called it ' a dialectical develop- ment from S t o i ~ i s m , ' ~ and asserted that it appeared first in Plotinus; "lonrad found it ' oriental ' in doctrine uocc ; 5contrast with the eclzt lzelle~~iscke~z of Vacherot, Guyot, and others derive it from Philo, despite the profound differ- Pliilosophie, pp. 27 sqq. ; Naville, Religiox des results of oriental myth-making with elements auciews kgyptieqls, p. 9 3 ) . Kor is anything really derived from Greek philosophy. analogous to the close reasoning and intellectual 1 Op, cit., p. 70. subtlety of Plotinus to be found even in hybrid 2 Stob. [Aetius] Ecl. I. 2 . 29 [58~]. products like the works of Philo, the Hetwetica, 3 Pilil. dev Grieclr. 1113. 427. 4 Ib id . 435. and the de dfjjsteriis, which are generally recog- 5 )kilos, hlonntslreft XXIST. (1888),p. 186. nized as combining, in whatever proportion, the

132 E. R. DODDS

ence in point of view between Philo and Plotinus, and despite the fact that Philo repeatedly calls his God i;v and vo6q.l Some have thought of Numenius or Alcinous (whom we are now taught to call Albinus); but the God of both these writers remains a superior V O G T , ~and neither of them speaks of him as the One.3 Others, more reasonably, have been reminded of the One and the ~ndeterminate Dyad i n some Neopythagoreans and in Aristotle's version of Plato's metaphysic. But, oddly enough, apart from a passing reference in 1Vhittaker.s book all the professed historians of Neoplatonism whom I have read ignore for some reason the obvious Platonic source.

Think of a principle of unity which so completely transcends all plurality that it refuses every predicate, even that of existence; which is neither in motion nor at rest, neither in time nor in space ; of which we can say nothing, not even that it is identical with itself or different from other things: and side by side with this, a second principle of unity, containing the seeds of all the contraries-a principle which, if we once grant it existence, proceeds to p1t:ialize itself indefinitely in a universe of existent unities. If for the moment we leave fragments out of account and consider only the extant works of Greek philosophers before the age of Plotinus, there is one passage, and so far as I know one passage only, where these thoughts receive connected expression -namely, the first and second 'hypotheses' in the second part of Plato's l'avnle~zides. Plotinus ignored one or two of the more fanciful conclusions reached in these hypotheses ; and to some of those which he adopted he gave a new turn. But how close is the parallelism at many points may be judged from a comparison of the following passages :

First IJybotAesis. 1 (a) Ilxcipov ;pa T& i v . . . ~ t x >l v e v 0x4- I (a) 01%' ou'v ~pbs iiXXo O ~ T E ~pbsn&rb

pa~osbpa . . . ireirrp 0382 plpq EXEC. / ( ~ b . . . 0;'82 ux7jPa TO~VI~V,T E ~ ~ ~ U V T U L EUV) (137~-E.) / ZTC p782 pLpq. (V. V . I I . )

( b ) T O L O C T ~ Y Y E sv ( ~ b2") 0;8apoU i;v E% I (b ) O ~ Kav 6ryoGv ;pa (Ti i v ) . r a i ~ ~osv+

O ~ T C,ap ;V Ilhhy O;T~ Zv ; a v ~ +€6 ( 1 3 8 ~ )/ 068apol:. (V. V. 9.) (c) ~b gv ;pa, &S Eot~sv,oiire ?UT~KEV / (c) 0682 KLVO+EVOV 068' a3 ~ O T & (Ivrr r bOGTE

K L V C ~ T ~ C . ( 13g~ . ) 1 Zv). (VI. ix. 3.)

1 T h e Neopythagorean identification of God mundi 2, 81, although in the same breath h e is with the supreme monad is mentioned by Philo identified with pots ; and r b iiv must be dya80D only to be amended: T C T ~ K T ~ L03u 6 @ E ; S rd Kai (vbs E ~ ~ ~ K ~ L Y ~ ~ T E ~ O Ypovd80s d p ~ ~ y o - ~ a r h KPE~TTOV ~ a i Pv ~ a l r+v povdBa, pBXXov 66 + povhs ~ a r h EvaT Z I Y vdrepov (Vi t . confempl. I , 2 ; cf. Praem. e t pocn. 6 , Bebv ~ 8 syctp d p ~ e p b s Y E ~ T E P O S K ~ ~ , U O U ,As ~ a i 40). Any attempt to extract a coherent system xpbvos, 6 82 Bebs ~rp~u,i3Ljr~pos ~ a i from Philo seems to me foredoomed to failure ;K ~ U ~ U U Bv,utoup~bs (Leg. Alleg. 11. I, 3). S o also Clement of Alex- his eclecticism is that of the jackdaw rather than andria, Prcd I. 8. 71,tells us that God is t v [not the philosopher.

~ T ~ K C L Y ~ L T ~ P a h + v povd8a. Prej.76 f v ] ~ a ? TOG iv6s ~ a i Nurnenius up. Euseb. Ev. XI. 22 ; Both Philo and Clement were of course deeply Alcinous (Albinus), Didascalicus, c. 10.

influenced by Seep:-thagorean speculation, of 3 Some MSS. of Eusebius do make Nunlenius which Alexandria had long been a centre ; but speak once of r b i'v (106. r icit. . i ~ p ~ X ~ r ? j u a ~ p d % q p a , in this matter they were determined to go one i u r ~ rd ilu). But the reading r b Bu has better better than the heathen. Philo's god must authority, and is supported by Plato, Re$. 5 2 4 ~ -similarly be K P E ~ T T W Vjj a676 r b dyaB6v (Dcopif. 5 2 5 A .

T H E PARMENIDES A N D T H E N E O P L A T O N I C ' O N E ' 133

PLATO,Parmcnides. ( d ) 06rw 8$ Z~epdv y f < ra6rbv r b SV O;T'

2v ail+ O ~ T 'BY drip? rzq. ( 1 3 9 % )

( e ) OGTE ;pa dpoiov odre dv6poiov oGO' Gripy o&e iau~-4 gv el? T A i v . ( I ~ o B . ) (f)odre dpa ivbs pErpov pcrixov OGTE

nohhGv O ~ T Edhlyov , odre r b ?rap6xav TO;

a6roG perixov, O U I T E h a w 4 TOT<, &s Z O L K E V , Zurai iuov odre ; X h y a odre a3 pei[ov 0662 Zharrov o$rr iavroG odrr iripov. ( I ~ o D . ) (g) 0682 i v Xp6vy r b napdxav S4vairo Bv

char r b Zv. ( I ~ I A . )

( h ) rb ZV O ~ T €ZV GUTLV ZUTLV. ( I ~ I E . )0 d ~ f

(i) 068' dvopd[crai &pa 0662 h iyrrai 0682 Sofd[erai 0662 y i y v ~ u ~ r ~ a i , 068i ri TGV ZVIWV a&oG al'uOdverai. ( 1 4 2 ~ ~ )

Second Hyfiothcsis.

Cj) i n l x6vra dpa nohha iivra ?j o;ala vev ip t l ra~ ~ a l o6Sevbs a 'nwsare i l r&v 6vrov.

(144B.) ( k ) rb 2v ilpa Bv Zv r i i o r l ~ o v noXXh.~ a l

(145A.) ( I ) ~ a la ~ 7 j p a r o sS7j sivos, i s : O ~ K E , TOLOG-

T O V SV ~ E T ~ X O ~ ( 1 4 5 ~ . )ZV 7b ZV. (m)oi;rw 8$ x e $ v ~ b s rb E"v zp' o 6 ~ a ' v d y ~ q

~ a l~ l v r i ~ e a i iur&vai; ( I ~ s E . )~ a i

( n ) t<aZ &v ra6r6v ye Sei rTvai a6rA i a ~ r r q ~ a l2repov iavroi , ~ a l rois Z;XXois &uairwS ra6rdv re hepov e?vai. ( 1 4 6 ~ . )

PLOTIKUS. ( d ) 86; p;v Y&p r i npb T&VTOV dva i 6irXoGv

TOGTO ~ a l~ c i v r w v ;irepov TGV per' a676, i+' davroG sv, 06 p e , u i y ~ ~ ~ ~ ~ an' a6roi ~ a lTOTS n&Xiv Z T E ~ I O V rP6nov rois dhAois napeival SUV&~€VOV ;v, OGX (;v E ? T ~ 'BV ~ V T W S ~ T € ~ O V ZV .a& 0; +ei3Sos ~ a l rb 2v e?vai. (V. iv. I . )

(e) 06 ykp E V L 0682 ~b OTOV (r$ id),Bry pqS? r b r l . (V. V.6.) (f)06 ykp OiXri (ri) 2v) prr' ~ X X O I IoGre

dvbs odre 6nouo~oGv u ~ v a ~ i O p r i u O a i ,038' Bhws d p i ~ p ~ i o ~ a ~ ' 06 perpo6- pirpov yhp a6rb ~ a i prvoy. (V. V. 4.)

( g ) O ~ KZV x P 6 ~ y (ZUT~ rb i v ) . ( V I . ix. 3.)

( h ) ( rb 2v) ~ a 9 ' 05 +tGSos ~ a l r b 2v eZvai. (V.iv. I . )

Zari 62 0682 rb 2uriv ( K Q T ~ TOG ivds).

( V I . vii. 38.) ( i ) o;re r l rGv n6vrwv ( zur i 7b 2v) 0676

6vopa aGroG, o"ri pq82v K ~ T ' a6~0Vn. (V. iii. 13.)

06 p;Iv a6rb hiYoPrv, 0682 Y v G u ~ v 0682 v6quiv Zxopev a6roG. ( l b i d . 14.)

( j ) ~ o p i u r i o ~ . . , navraxoG. . . e?vac TOG o"vros r b Bv O ~ Kct?roXei?r6pevov iavroG. (VI. iv. I I . )

( k ) roXXd SrC roGro r b 2v e h a i , $v perk r b T ~ V T ~ (VI. vii. 8. )iv .

( I ) uXqp&rwv S$ na'vrwv d+OEvrwv i v r 4 iivri ~ a i ( V I . ii. 21.)x o i 6 r q ~ o s dn&uqs.

(nt) vcpl p2v r b $v T O ~ T W V (SC.u~a 'uews ~ a l~avrjuews) Odrepov i j ctP$6repa d v c i y ~ q .

( V I . ix. 3.) ( n ) € 1 82 xoXhd ( i u r i r b 6v), ~ a 2 irepo'rqs

( Z U T L ) . ~ a l el 2v noXXd, ~ a \ c ravr irqs .

( V I . ii. 15.)

Small wonder that Plotinus9egarded the PIatonic Parmenides as a great improvement on his historical prototype ; that Iamblichus3 considered the Pav~nenidesand the Tiuzaez~sas the only Platonic dialogues indispensable to salvation ; that Proclus4 found in the Parv~e~zides,and there only, the complete system of Platonic theology. Read the second part of the Parl~tenides as Plotinus read it, with the single eye of faith; do not look for satire on the Megarians or on anybody else ; and you will find in the first hypothesis a lucid exposition of the famous ' negative theology,' and in the second (especially if

1 Cf. Enn. V. V. g, o d 6 ~ v J sav dxouraroi ( r b F v ) . 3 Procl. in Tim. I. 13. 15 sq , Diehl ; Proleg. 2 V, i. 8 f i t f . Plat. Phil. 26. ' Thcoi. Plat. I. 7.

E. R. DODDS

you take it in connexion with the fourth) an interesting sketch of the derivation of a universe from the marriage of unity and existence. What you will find in the remaining hypotheses I cannot so easily predict; even within the Neo- platonic school there were violent differences of opinion about them l-differ- ences which I must. not attempt to discuss here, as they would carry me too far from the main intention of this paper.

Even as regards the first two hypotheses, it is no part of my purpose to argue that the Neoplatonic valuation is an entirely just one; Parmenides' description of his own performance as yvpvaula and T ~ L ~ L C ~ , ~ taken in conjunc- tion with the obvious fallacies in which some of the hypotheses abound, should be sufficient to warn us against assuming that all his conclusions ~tecessnrily found a place in Plato's own system. At the same time, it should not be forgotten that the Idea of the Good, no less than the 'One ' of the first hypo- thesis, is beyond Being, and that, if we are to believe A r i s t o ~ e n u s , ~ the principal conc!usion reached in the Lecture on the Good was bya86v 2 u . r ~ ~ More-Zv.

over, some of the most important discoveries of the later Platonic logic, especially the distinction between absolute and:relative non-Being, appear first in the Parmenidean hypotheses-surely an odd way to publish them, if these speculations are pure fun. However that may be, I have difficulty in under- standing the present position of so distinguished a scholar as Professor A. E. Taylor, who, when he meets with the negative theology in Proclus or the s~hoo l rnen ,~ moment ' of religious takes it seriously as a necessary and salutary ' experience, but when he meets it in the Parntenides, describes it as ' a highly- enjoyable philosophical jest.'= Professor Taylor cannot well have it both ways : what is sauce for all the little Neoplatonic and medieval geese should also be sauce for their parent, the great Platonic gander.

But is Plato indeed the parent, or only the putative father of these theological bantlings ? I t may be urged that the Plotinian interpretation of the Par~nenides is a complete misunderstanding; that important philosophies are not built solely on the misunderstanding of other philosophies, or, if they are, the misunderstanding is not accidental ; that the Neoplatonists notoriously found in Plato whatever they wished to find (' Hic liber est in quo quaerit sua dogmata quisque '1; and that, in fine, the Neoplatonic interpretation of the Parrrze~tides is subsequent to the rise of Neoplatonism, not prior to it-an effect and not a cause.

Now it is of course true that when, for example, Proclus finds his i i ~ ~ a v r o ~ 0 ~ o ldarkly revealed in the Pav~nenides, he is reading into Plato a drsgma originally constructed out of hints i n the so-called Chaldaic Oracles. But this is hardly relevant to the question at issue. For, in the first place, the systematic allegorization of Plato, which enabled Proclus to bring the teachings of that philosopher into complete harmony with the Orphic and Chaldaic

1 See Proclus in Pavm. 1052-64,Cousin. ~ o g ~ b13.

13jc sqq. : 1 3 7 ~ . PYOC.Avist. Sot., N . S . YVIII. , p 632. 3 H m m . El. II., p. 30, hleib. ; cf. Ar. 113ctaph. 5 Plnto : The ;vlun and iris IVork, p. 370.

T H E PARII IEXIDES A N D T H E NEOPLATONIC ' O N E ' 13j

theology, appears to be mainly, if not entirely, the invention of Iamblichus :' there is very little of the kind in Plotinus. Secondly, the Neoplatonic inter- pretation of Plato's rA Zv and 76 I v i;v rests on a literal, not an allegorical understanding of the text, and has commended itself to some able modern critics who are certainly not Neopla tonis t~ .~ Thirdly, this interpretation is, in fact, older than Plotinus. I t seems worth while to develop this last consideration a t some length, not only because it is the decisive one for my immediate argument, but because in tracing back the history of the Neoplatonic interpretation we are a t the same time retracing one of the main currents of thought which traverse earlier Greek philosophy and issue in Neoplatonism.

Plotinus will not help us in this enquiry: he is too much interested in his own views to trouble about recording other people's. T h e commentaries which Porphyry (?), Castricius Firmus (?), Iamblichus, Plutarchus the son of Nestorius, and Syrianus devoted to the Pnv~~tenidesare lost. Our first resort is therefore Proclus. Proclus distinguishes three schools of interpretation of the second part of the Pnvnzenides. The first saw in it either a polemic against Zeno or a logical exercise; the second took it seriously, but did not find in it r h & ~ 0 1 C ; ~ ~ r 6 r ~ p a :r G v 8 0 y ~ d ~ w vfor them the kernel of the dialogue was the doctrine of the Iv iiv, which embraces the Ideas in its unity. The third school are distinguished from the others in that they agree in referring the first hypothesis to the ijnepo6u~ov b ; most of them refer the second hypothesis to V O ~ , to qjrvxlj, but here agreement ends. noand the third Unfortunately names are attached to the first two schools. The first view (which is also that of many modern scholars) is implied in A l b i n ~ s , ~ and is doubtless much older than Albinus: we may plausibly ascribe it to the sceptical New Academy. T h e second or immanentist interpretation (which is much like that formerly held by Professor Taylor? but now abandoned by him) suggests Stoic influence, and we may perhaps think of Antiochus of Ascalon as its possible originator. The third is clearly the Neoplatonic view. In its primitive form Proclus associates it with the name of Plotinus. For evidence of an earlier origin we must turn to another quarter.

Sextus Empiricusa tells us that while some Neopythagoreans derived the material universe from the effluxion of the point, others derived it from two LpXat ,the One and the Indeterminate Dyad. On the basis of this, Schmekel' and others after him distinguish a monistic and a dualistic school of Neo-pythagoreanism. But the distinction in this form does not really hold: for some at least of the so-called dualists posited an ultimate unity (Q or pov&), prior to the derivative unity which with the Indeterminate Dyad generates

1 See K. Praechter in Genctl~linko?iRobeut, Isag., c. 3 ; cf. c. 6, and Diilascalic~~s,c. pp. 120 Sqq. (p. 155Jin., Hermann).

q e e in particular the interesting recent book 6 ' O n the Interpretation of Plato 's Parmeni- of M. Jean Wahl, Etude szrv le Parnit:tzide. des,' i l l i xd , 1896 7, 1903.

3 111 Pavm., p. 635, Cousin ; cf. Theol. Plnt. 6 Ad-t. Phys. 11. 291.2. I. 8 sqq. Philos. d. A f i t t l , Stoa. 403.39.

E. R. DODDS

plurality. This view is ascribed to ' the Pythagoreans' by Eudorus,l a Platonist who is dated about 25 B.c., as well as by Proclus2 and others. Syrianus attributes opinions of this sort to Archaenetus, Philolaos, and B r ~ t i n u s . ~

This type of monism bears evident marks of Platonic influence. That one of its sources was the sixth book of the Republic appears from the statement ascribed by Syrianus to Brotinus, that the supreme principle voir n-avrdq nal

oLalaq 6 v v a p e ~ n a l . r r p e a p e I q 6 . r r e p i ~ e ~ - a n obviousechoof Plato's words in Rep. 509~. But whence come the two Ones, the transcendent and the derivative ? Hardly from primitive Pythagoreanism : for in Aristotle's references to the Pythagoreans there is no trace of any such duplication of the One ; and the antithesis of the One and the Indeterminate Dyad is Platonic, not P y t h a g ~ r e a n . ~ The true source is, I think, made apparent by the following passage from Simplicius :

T a 6 ~ q v6; 7r~p2 6 ~ 6 v o ~ a v ' e b ~ q n f ' v a ~ pkv TGV 7739 6hq9 T ~ V h o k a a ~ v TPGTOL

'EXh7ju0v 01 I ~ U B ~ ~ ~ ~ E L O L ,per8 8' f 'nelvov~ d ITXdrmv, & F nal Mo8iparoq Luropei.

O ~ T O Fyhp nard TO;$ I T ~ e a ~ o ~ e l o ~ q 2v 6 ~ 8 p T) , p8v ~ p G ~ o v 7 6 e2va~ nal n-iiaav

o;aiav &rro+alvera~, ~ E ~ T C P O V 6v nal V O ~ T ~ V ,rb 8; Zu, a m p hark rb ~ V T O F 78 5. ei8? +qulv elvac, rb 88 T ~ ~ T O V , ~ E T ~ X E L V nai TGV &rep f'071T) , + v ~ ~ n 6 v , 706 &~bq

el8Gv, T ~ V88 + 6 a ~ v r ~ j v TGV alaBqrGv o3aav ,.~q8& T O ~ T O V~ e h s v ~ a l a v

periXe~v,&hXd K ~ T ' Zp+au~v& ~ ~ V O VnenoapijaBa~,rijq f'v a;.roi~ 6Xq9 706 p+

~ V T O F~pci3rws~ ' v T @ ~ V T O F unlaaPa ~ a 2 xalTO+ 0 6 a q ~ Z T L p2XXov 6.rro~ePqnvla~

&rrh T O ~ T O V . nal ra6ra 82 d I Iop+dp~o~f'v 76 8evriPY n e p l i;h?q 78 TOG 10. M O ~ C P ~ T O I J ryf'ypa+&v6 7 ~'PovXq8e;~d 6 v ~ a i o ~T ~ ~ ~ T L ~ ~ ~ ~ v o F X6yo9, 47 rod

+qasv d IThdrwv, y i v ~ a ~ v 2avroi, TGV bvrov a v a ~ 7 j a a a @ a ~ , T ~ V &+' nard

This passage was alleged by Vacherots as proving that the Neoplatonic trinity and the Neoplatonic doctrine of Matter were anticipated by Moderatus -a Pythagorean who can be dated to the second half of the first century A.D.

Zellerg replied that it proved nothing of the kind. He pointed out (rightly) 1 Apud Simplic. in Phys. 181. 10.30, especially Mctaph. 800. 32, Bonitz (quoted below, p. 138).

27 sqq. : &s pkv dpx4 ~b Ev, &s 62 aroixeia T A ?v In Metapic. g25b 27 sqq. : ~ a ifr i npd rGv dlio rai 4 6uds, dpxal dp$w 6vra ?rd?.iv' cipxrjv r i v dvtalav airiav xpohrarrov, +)v 'Apxaivc~or ~ ~ P I U T O S 8v Kal 6 i jA0~ 871 dAk0 p h t 'U~tv?V T) d p x i T&U ? r d v 7 ~ ~ , ['ApxL;ras ci. Boeckh] p2v airiav npb airins etval dAAo 62 Fv rd 75 6 ~ 1 6 6 ~ ~ a l @qoi, @iA6Aaos 62 rrjv ndvrwv dpxav e b a i Giiuxupl- dvri~eipavov,8 povddd. raAo0uiv. The words occur in a verbat~ntcitation l era t , Iipor~vos 86 &s voV n a w ~ i s ~ a i ohuias 6uv(ipei from Eudorus. rai rpeuBe1~ b ~ e p . 4 ~ ~ ~ . Cf. 935b 13 sqq.

2 In Tim. 54" [ I . 176. g sqq., Diehl] : npoq- 4 Arist. Metapir. A 6. 987b 25. y e i ~ a i yhp r b Ev dadaqs Qvavr~duews, hs ~ a i oi CUTLY.SC.871 duh,uaros K G ; & T O L ~ S IIuOay6peiol Gaaiv. LAX' dxei ~ a iperk rhv plav 6 ;xdpiue Zeller : fort. kxop+yqoe. a i ~ l a vi Guhs srjv dpxrjv dvedcivq, ~ a il v radrats 7 I n P h ~ s .A 7, 230. 34 sqq., Diels. i povhr rpelrrwv T$S GuCLGos. . . . Cf. The0 Smyrn. 8 Hist. dc l ' c r o l e &Alex. I . 309. Exp. R e v . Math. 19.12 sqq., Hiller ; Damascius, de 9 1113. 126. 2. I n the fourth edition the pvincip 86. 20 sqq., Ruelle [ I I ~ , Kopp] ; and for passage is treated more summarily, and some what seems to be a different way of putting modifications are introduced (111. ii. 143. 1; essentially the same view, Numenius a?. Chalcid. cf. 130. 5). in Tim., c. 293, Mullach, and ps.-Alexander in

THE PARMENIDES AND THE NEOPLATONIC ' O N E ' 137

that if the words O ~ T O FyhP, etc. (1. 3 sqq.), refer simply to the private opinions of Moderatus, they do not show the dependence of Plato on P~thagoreanism, and the yhp is therefore meaningless. H e accordingly supposed that O ~ T O Syhp

tcarh TO& ~ u 8 a Y o p ~ ~ o v F Moderatus in accordance with the Pytha- meant not gorean~, ' but 'Plato according to the Pythagoreans.' H e also pointed out-again quite rightly-that Simplicius is not citing Moderatus at first hand, but only (as the beginning of the third sentence shows) Porphyry's report of what Moderatus said : ' And this also Porphyry has written in the second book of his Essay on Matter, quoting Moderatus.' H e then bracketed as additions by Porphyry ( I ) the words o"mp T A oVvro~ bv teal voq~dvin 1. 4, and o " ~ e p2 a ~ l

$ U X L K ~ V in the next line; (2) the remark about the two kinds of 6v at the end of the same sentence, from r 7 j ~ 2v a 6 r o i ~ down to ~ a 2&rrA T O ~ T O U . I n his earlier editions he also ascribed to Porphyry the whole passage in inverted commas from P o u X ~ O E I Fonwards, reading the aorist participle r r ~ p a 8 & ~ e v o ~in 1. 9 ; but be withdrew this when it was found that the MSS. were unanimous in giving the present, r a p a T L Bi,uClevoc.

Now it seems to me that the key to the understanding of this passage lies in the fact, which neither Vacherot nor Zeller recognized, that the first eight lines refer to the interpretation of Plato's Parmenides. This should be obvious to anybody who knows his Proclus, or even his Plato. The first, second, and third ' Ones ' are the three Ones which are posited in the first three hypotheses of the Pnv~ne~zides, and the interpretation here given to them is the same which was current in the school of Plotinus. In what follows (11. 6-8) the sensible world is analyzed into two elements, viz. ' reflections' (6,u+du~le~e)of the Forms and material GXq : the latter is absolute t~?jdv, and is a shadow of the relative p ~ )8v [=Neoplatonic ' intelligible Matter '1 which is implied by the plurality of the Forms (2v T+ rroa@ ~ V V T O F ) . This corresponds to Proclus' explanation of the fourth and fifth hypotheses of the Parrnenide~.~ Finally, the words &c T O ;

+ q a ~ vd IIXLTWU(1. 10) can best be justified if we take them as an allusion to the genesis of plurality from the self-diremption of the Cv OYu [2v~aioq Xdyoc] in the second hypothesis, in combination with the t!~,uayaiovof the Ti~naez~s inter-preted as relative Cl;l 6v.

What we have before us, then, is an interpretation of the Parr~ze~zides. Whose interpretation is it ? Not just that of Simplicius or of Porphyry. Even if O ~ O F r l [ ~ ~ a ~ ~ ~ ~ ~ o u ~xarh T O ~ F means ' Plato according to the Pythagoreans,' we are still dealing with an interpretation not first invented by the Neo-platonists, but taken over by them, at least in part, from an earlier source. But Zeller's objection to referring O&OF to Moderatus, namely that the yhp is without meaning, has now disappeared. Simplicius' argument may now run : ' This conception of Matter goes back to Plato, and ultimately to the Pytha- goreans, as Moderatus relates : for (yhp) Moderatus shows that the Pnr?zelzides is to be interpreted on Pythagorean lines ( ~ a r hT O ~ FIIu8ayoPelou~),and that

Ift Pavm. 1064: s+v 6 t rcrdprqv xspl rGv writers had found the same topics in other dvdAuv . . . s+v 66 a4prrsnv xspl iihqs. Earlier hypotheses (ibid. 1052-9).

138 E. R. DODDS

when so interpreted it is found to contain this conception of Matter.' Every-thing down to A r b r06rov (with the possible exception of the two $ r e p clauses) will in that case be genuine Moderatus. That this is the right way of taking the passage seems to me probable, frorn the following considerations :

(a) 01%-OFis most naturally referred to the nearer name, which is Moderatus. (This, of course, is not by itself decisive.)

( b ) Zeller's way of taking it makes Ylato talk about the first, the second, and the third 'One,' and about the identity of the second ' One ' with the Ideas: which he does not do in the Pavllzenides or anywhere else. Zeller refers to the passage in the Second Letter1 about the three grades of reality: but they are not called ' Ones,' and there is no mention of the Ideas. An itzter- preter of the Pavvzenides, on the other hand, could very well find these doctrines implied though not expressed in it.

(c) The use of 7 6 n-ou6v (1. 8 ) or (1. 11) as a T O U ~ T ~ F description of the element of plurality in the intelligible world is genuinely Neopythagorean : e.g. Theon of Smyrna, in a passage which seems to be based upon hloderatus himself, defines number as 76 2v v o ~ r o i p n - 0 u 6 v . ~ Hence it is not unreason-able to suppose that the substance of 11. 6-8 and 10-14 goes back to h l ~ d e r a t u s . ~

(d) Lastly, the impression that the Neoplatonic interpretation of the Partfzetzides is in its basis Neopythagorean is strengthened by a remark of pseudo-Alexander : 0; &, $ u r e P B p o ~ i v o p6 r I v 0 a y 6 p e ~ o ~ ,I I h d ~ o v ~ a k +aukv

$ 7 ~~ t AyaBt)v) rt) b 2 u r ~~ a lo ; u l o r a ~ i v r@ $v e b a ~ (ift ;li!etnph. 800. 32, Bonitz). The subtle doctrine that the super-essential One, o ; u l o r a ~ f'v 76 gv

szua~, 'has become essence in so far as it is One,' can hardly come from any other source than the Par~te~zides,~ whence we must suppose it to have passed into the Pythagorean apocrypha. The attribution of such a doctrine to the

312E. 70; 6vralou A6you voodpevov roc xdvras robs Xbyous 2 Exfos . rer. math. 19 . 15, Hiller. Theon P E ~ L E L A ~ + ~ T O I .7Gv tjvrwv Qv xapa6elypard 6uri

18. 3-9 + 19. 8-9, 12-13, reproduces almost word r?js TGY U W ~ ~ T W U 3v ~ a i BA~s , airrilv aoubv Kai 703s for word a fragment of Moderatus preserved by IIuOayopeious ~ a i ~ a A ~ i vrdv IIAcirwva h AE y E v , od Stobaeus, Ecl. I. i . 8 [ I ~ H ]; while the continua- rd 3 s ETBOS X O U ~ Y ,dAAh rd ~ a r h KTA. (Theu ~ k p ~ u r v tion in Theon 19. 13-20. 11 is an expansion of remainder of the quotation describes the nature the next sentelttia in Stobaeus, Ecl. I. i. 9. This of OX7 on orthodox Neoplatonic 1ines.j Here second selttentia is d6kuaoros in our MSS. of the repeated +?ui seems to mark the introduction Stobaeus, but Theon must have found the two of Porphyry's comments on Moderatus' concep- juxtaposed ; and while he may possibly have tion of intelligible aou6r7r ;while tAeysv takes us come upon them in some doxographical writer back to the statement attributed to Moderatus a t afterwards used by Stobaeus, it is simplest to the beginning of the passage. T h e words in suppose with Wachsmuth that h e read both of inverted commas a re from the Tiniatus ( ~ I A ,B ; them in Moderatus--presumably in his work 52;). aepi rGv dp10pGv (Porph. vit . Pyth. 48). 1 4 2 ~ - E: Ev ei huriv, bpa oi6v re aJrd eiuar pt'v,

a T h e continuation of Simplicius' citation oDuias bP pi) ~ B T E I X E L Y ; . . . 76 T E yap iiv rd iiv &el from Porphyry runs as follows (231. 12-24) : Qai luxei at rd Bv rd Ev. &UTEcivciy~v 6v' ciei yiyvi)p~vov r a d r ~ s~ O L K E , 7 u i . T?S a o u i ~ ~ ~ o sIIXdrwv T& p78haore ?v ~Tvar. Cf. Chalcidius i , ~i) T i m . , c. 293, aAeiw 6v6para ~arvyop+juar ' xav8ex+ ' K G ~civei6eov illullach: '(Numenius ait) nonnnllos Pythagoreos

. . . putasse dici etiam illam indeterminatam e t immensam duitatem a b una singularitate insti- tutam, rrcsdrntc a nal~tva slta si~~gttlavitateet if$ diritutis habitltm n~igrante.'

T H E PARlWENIDES AND T H E NEOPLATONIC ' O N E ' 139

historical Brotinus, who lived a t the end of the sixth or beginning of the fifth century before Christ, is clearly impossible.

That Pythagoreans should thus take their material from Plato, and that their interpretation of it should influence later Platonists, need not surprise us. Under the Early Empire the two schools were closely associated. Both Numeniusl and the earlier Neopythagorean epitomized by Photius2 saw in Plato what Moderatus saw in him-the popularizer of the Pythagorean philosophy. This view of the relation between Plato and the Pythagoreans is already implied in the story-which in its earliest form goes back at least to the third century B.c.-about the Tiftzaezis being copied from a Pythagorean book. I t was a view agreeable to the Neopythagoreans : and they sought to confirm it in two ways-by emphasizing real or supposed Pythagorean elements in Plato's teaching, and by introducing Platonic elements into their own pseudepigraphic literature. The latter procedure created ' Brotinus ' and his kind ; the former led them to seek in Plato a cosmogony based on the One and the Indeterminate Dyad (which passed for Pythagorean), and to find it in the Pavtnefzides. That their interpretations soon began to influence the revived Platonic school is shown by the fact that Eudorus, one of its earliest known representatives, 'emended ' or falsified a passage in Aristotle's Illeta-Physics3 in order to make Aristotle ascribe to Plato the same doctrine which Eudorus found in the Pythagoreanism of his own day. Later, Plutarch shows clearly the influence of Neopythagoreanism; while in the eclectic Platonist Alcinous (or illbinus) the Neopythagorean transcendence theory appears in hopelessly inconsistent combination with the immanence theory (God =voGp= the sum of the Ideas) which had developed under peripatetic and Stoic influence. In his attempt to connect these divergent views he foreshadows Plotinus: his complete failure to make anything coherent of them is one measure of Plotinus' greatness. In the school of Plotinus himself the works of rnen like Numenius and his pupil Kronius were studied no less than those of orthodox P l a t ~ n i s t s . ~ Longinus, who was in a position to know, regarded Plotinus as the ablest exponent of the Pythagorea~zand Platonic Gpxal: these Gpxal,he tells us, had been expounded earlier by Numenius, Kronius, Mode~,atzbs, and T h r a s y l l u ~ . ~

The resemblances between the theology of Plotinus and those of Philo, of the Hermetists, and of certain Gnostics are most easily explained by the

1 A p . Chalcid. in Tim., c. 293, Mullach. Plato's system the Neopythagorean and Neo-2 Cod. 249, 438b 17, Bekker. platonic monism ; cf. Eudorus apud Simplicium 3 Metaph. 988a 10-11(Aristotle reporting Plato's in Phys. 181. 10, quoted above, p. 8, n. I.

view) : r d y d p ~ 1 6 77 0 0 r i QUTLV a h a r o i s &AXOLE, Harmonizing appears to have been Eudorus' 70;s 8' E I ~ E U L Y 7 6 Cv. Alexander ( in Metaph. 58.31- passion, for his ethic, like that of Antiochus, is a 59. 8, Hayduck) tells us, on t h e authority of blend of Platonic and Stoic (Zeller 111. i4. 634). Aspasius, that Eudorus and Euarmostus read Porph. vzt. Plot. 14. here 702s 6' ~ i 6 6 u ~ ~ r i i i A 7 : and h e 6 Ibid. 20. Moderatus figures also in the list 7 6 ?Y K Z had himself found this reading in some copies. of authors studied in the school of Plctinus' T h e effect of the alteration (which may have great contemporary, Origen the Christian (Porph. been suggested by an accidental dittography of up. Euseb. Hist. Eccles. VI. 19.8. Evidently his the opening words of the next sentence, ~ a i was still a name of some significance even in the r i s jJ bA7) was to introduce into Aristotle's account of third century.

140 E. R. DODDS

assumption of a common source or sources.l W e have seen that one source of the kind required existed in Neopythagoreanism; and that this Neopytha- gorean theology was, in part, a t any rate, shaped by the Parmenides. Who i ts original creators were, remains uncertain. That Moderatus was not its first sponsor is shown by the testimony of Eudorus, which is something like a century earlier (as to pseudo-Brotinus, I know no means of dating him). Schmekel supposes that this wing of the Neopythagorean school was influenced by Antiochus of Ascalon; but his evidence is meagre, and in view of the well- known Stoicizing tendency of Antiochus it seems most unlikely that he is the source of a transcendent theology. I t is more natural to think of the Old Academy, and especially of Speusippus. I could not here attempt a recon-struction of Speusippus' metaphysic, even were the task less desperate than it is. But it is surely significant that his first principle was the One, which according to Aetius2 he distinguished from voirr; that Aristotle3 appears to credit him with the view that the One was i~rrc~oba~ov, or a t any rate dvobarov, as well as with the comparison (so often used by Plotinus) of the One t o a seed ; and that this One was the first of a series of Ap~a l , dXX7v pJv &p~BpGv, dhhqv 82 paryeBcjv, & r e ~ ~ a I t seems to me that with Speusippus we+VX$F.~

are already well started on the road to Neoplatonism ;6 and nobody has yet alleged that Plato's nephew was anything but a ' true Greek.'

T o say that the Enfzeads were not the starting-point of Neoplatonism but its intellectual culmination8 is no disparagement of Plotinus' originality. ~ h k philosophical thinking of the first two centuries after Christ was vague, con- fused and incompetent, as transitional thought is wont to be. Without this thinking the Enneads could not have been written. But Plotinus, after the manner of men of genius, fashioned from this unpromising material an edifice which a few of his predecessors may have seen in their dreams but whose construction had remained altogether beyond their powers. Nowhere is the individuality of his genius more manifest than in the doctrine of ecstasy, which for him is the psychological correlate of the doctrine of the One. A recent German writer7 has even suggested that Plotinus' personal experience of the ulzio lttystica determined his conception of the One. But we have seen that this conception is in substance far older than Plotinus. I t is perhaps truer t o

1 Since we know that Plotinus ha.d read shows that the doctrine has a general cosmo-Numenius, and there is some reason to think logical application, and does not aim merely a t that Numenius had read Philo and Valentinus distinguishing arithmetic from geometry. T h e (Norden, Agnostos Theos , p. ~ o g ) , the possibility dptt'poi are for Speusippus what the Forms a re that one or both of the last-named writers exer- for Plotinus. cised some indivert influence on Plotinus ought 5 I find that the same view is suggested by not to be ignored ; but it will not account for all 0 . Immisch, Agatharchidea (Sitzungsberichte the facts without a great deal of forcing. Tha t Heidelberger Akad. der Wiss., Phi1os.-Hist. Plotinus himself could take either Philo or Klasse, 1919, Abh. 7), p. 37. Valentinus seriously as an authority I find it 6 The common view, that they were both, hard to believe in the light of such passages as appears to be self-refuting ; a t any rate, it flies Entread 11. ix. 6. in the face of all historical analogy.

Af i . St0b. Eel. I, i. 29 [ 5 8 ~ ] . 7 J Geffcken, Der Alrsjiang des Grieclrisch-Ro- iMetafih. N 5, 1092a. 11-15, mischet~ He~dcnt t ims , p. 47.

4 Ibid. Z 2. 1oz8b 21. T h e mention of $vx+

T H E PARhlElVIDES A N D T H E NEOPLATONIC ' O N E ' 141

say that his conception of the One determined, not indeed the personal experience itself, but the interpretation which Plotinus attached to that experience. The concept of the One carz be reached, as P!otinus fully recog- nizes, by a purely dialectical regress ; and the element of personal mysticism is absent, so far as I know, from the fragments of the Neopythagoreans (until we reach Numenius) and of the old Academy. Dialectic, however, as we see in the Pnritzenides, can only tell us what the One is not. This tedious accumu- lation of negatives may content the metaphysician ; but, as Inge says some- where, one cannot worship the alpha privative. Before the Absolute of the philosopher can become the God of the worshipper, it must somehow be made accessible to human consciousness. But it was an accepted doctrine since Empedocles that like is known only by like. Hence the ultimate principle of unity in the universe is accessible, if a t all, only to some ultimate principle of unity in man. Hence, also, such access must be supra-rational : as the cosmic unity transcends the cosmic mind, so must the incarnate unity transcend the incarnate mind. The supreme act of cognition will thus not be strictly cognitive at all, but will consist in the momentary actualization of a potential identity between the Absolute in man and the Absolute outside man.

Such, I take it, is the logical basis of Plotinus' mysticism-the hypothesis whose verification he believed he found in his own inner experience, as other mystics in the like experience have found verification for other hypotheses. In the hypothesis itself I see nothing un-Greek. Starting from the transcendent theology of the Pavinenides and the Republic, it proceeds upon the Platonic principles that like is known by like and that the goal of man Be+is d p o l o c ~ ~

/CUT& 7 ; 6 v v a ~ ~ v . lPlato himself had in the Sy)rtposiz~tn traced the stages of that d t ~ o l o c ~ ~ .Platonic too, though bearing the imprint of his own genius, is the language in which Plotinus essays to express his inexpressible experience. His favourite metaphor of illumination has often been alleged as evidence of ' oriental influence.' Light is a natural symbol of deity, and occurs as such in Judaism and Manichaeanism, as well as in nearly all the religious writers of the Hellenistic period."ut an examination of the passages in Plotinus will show that his use of the metaphor rests partly on the comparison of the sun in Rep. VI., partly on that passage in the Sevefztk Letter (341c), where the sudden moment of insight is compared to ' a light kindled from a leaping fire '; it was also doubtless in part suggested by his own experience, since the like language has been used by mystics of all countries and ages to describe the onset of the state of rapt. T h e notion that such expressions in Plotinus must allude to the vision of luminous shapes offered to the initiate in the Isiac mysteries involves a confusion between two forms of religious experience which are spiritually poles apart. For Plotinus the only ' mystery' was the Platonic philosophy.

1 Theae t . 1 7 6 ~ . The development of the the similarity between the Plotinian and the thought was doubtless influenced by the Stoic Stoic view is pointe.1 out. doctrine that the in man is of one References in J . Ilroll, L e h r c ? ~d's~ ~ Y E ~ O Y L K ~ Y Hermes stuff with the ~ y e p o v i ~ 6 vin the universe ; cf. Tyismegistos, pp. 2 2 sq., and Nock, Su!lltstivs, Iamblichus a t . Stob. Ecl. I . xlix. 37 [856~] ,where p. xcix, n. 10.

142 T H E P A R J I E S I D E S .4ND T H E NEOPLATONIC ' O N E '

His attitude towards ritual is showed in his reply to the churchgoer Amelius, ircelvovq 6s; ~ p b ~ o;lcGp4 ~ ~ ~ X E O B U L ,Gp2 rrp& E' rce lvov~.~

Others have regarded the Plotinian doctrine of ecstasy as oriental on the ground of his supposed dependence on Philo. But Philo, according to the latest Queller~fovscltzt~zg, derives what he has in common here with Plotinus mainly from the Plznedrzu of Plato and from P o s e i d o n i ~ s . ~ And the Plotinian ecstasy is in fact profoundly different from the Philonic. Its distinguishing features are : First, that it comes only as the rare crown of a long intellectual discipline-a discipline which in the supreme moment is transcended but not denied; secondly, that it is clearly distinguished from that condition which Plato called f'vBovaraop6q or tca.rolcwX7j and which we call the mediumistic t r a n ~ e . ~The Plotinian ecstasy, unlike the Philonic, is achieved by a sustained intellectual effort from within and not by a denial of the reason or by a magical intervention from without ; it is presented less as the abnegation of self-hood than as the supreme self-real i~at ion.~ Here as elsewhere, Plotinus appears not as the subverter of the great tradition of Greek rationalism, but as its last constructive exponent in an anti-rational age. I t is true that after Aristotle nearly all the Greek thinkers who counted for anything were in their several ways tainted (or touched to life) by quietism and ' other-worldliness.' Plotinus is no exception to that rule. What makes him exceptional in the third century is his resolute rejection of every short cut to wisdom proffered by Gnostic or theurgist, Mithraist or Christian-his resolute championship of reason as the instrument of philosophy and the key to the structure of the real. T o assume his dependence on Philo because both writers talk about ecstasy is like setting out to derive the ' mysticism ' of a Bradley from the ' mysticism ' of a Madame Blavatsky. If anyone doubts that Plotinus was a man of genius, let him study the efforts of Plotinus' nearest predecessors and followers. Le t him soak for a while in the theosophical maunderings of Philo and the Hermetists, in the venomous fanaticism of Tertullian, in the tea-table tran- scendentalism of Plutarch, in the cultured commonplaces of Maximus, in the amiable pieties of Porphyry, in the really unspeakable spiritualistic drivellings of the de ;Vysteviis--let him do that, and if ever he gets his head above water again, he will see Plotinus in his true historical perspective as the one man who still knew how to think clearly in an age which was beginning to forget what thinking meant. E. R.DODDS.

Wholly different in spirit 2 H. Leisegang, Dcr Heilige Geist, I . i. 163 sqq. is Phllo's teaching, with its insistence on rtjv E'v 3 T h e fact that Plotinus his T ~ U LTO? YEVTTO? ~ ~ ~ Y E L (de somfl. ~ V I . 601 F o r

1 Porph. v i t . Plot. 10fill. 0dK TQ 6v71 i v ~ K E ~ u Y .

very co,nbares O L

ecstasy with the state of oi ~ a i I h i l o human and d ~ v i n e nature a re mutually~ v O o u u i 0 v ~ ~ s K ( i 7 0 ~ 0 1 y ~ v b p ~ v o i(V. iii. 14) should make it exclusive : 6 ~ a v pkv y i p $ 6 ~70' O~iov bsihdpqy, evident that the two conditions a re distinct. To 8liEiai Td dvOprjnrvov, 87av 6' ~ K E ~ V O86777a1, 7 0 0 ~ ' Philo, on the other hand, ecstasy nteafis + EvOeor bua~6Xh~i J ~ O L K ~ { E T ~ Lydp~ ~ I ' U X E L ~ a i . . . pkv Cv ~ a i o ~ u x 7 j pavia ( q l l i ~ jjpiv 6 voV sK ~ T U T Y E ~ ~ C L T O S Ka76T E ~ a i YeY. div. heres 53. 264). 7+v 7 0 ; 6E;oi~ & C $ ~ ~ L Y ,

4 E . g , Enu VI. ix. 11 : +:'EL (e 9uxil) o d ~ ~ i r 82 r+v p ~ ~ a v d u ~ a u r v (quisad7oG ~ ( i h r v ~ i u o r ~ i { ~ i a i dhho, bhh' €is haij77jv, ~ a i o l j~us o l j ~ ?Y &hhy o8ua dv rev. d lv . heres 53, 264.5). od8evi 6u71v, &Ah' <v a u ~ n . 7 d 6' i v airr?j p b v ~ ~ a i