Ennoia and in the Stoic Theory of Knowledge Author(s): F. H. Sandbach Source: The Classical Quarterly, Vol. 24, No.

1 (Jan., 1930), pp. 44-51 Published by: Cambridge University Press on behalf of The Classical Association Stable URL: http://www.jstor.org/stable/637056 . Accessed: 10/10/2013 03:49
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

.
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 support@jstor.org.

Cambridge University Press and The Classical Association are collaborating with JSTOR to digitize, preserve and extend access to The Classical Quarterly.

http://www.jstor.org

This content downloaded from 134.173.199.1 on Thu, 10 Oct 2013 03:49:04 AM All use subject to JSTOR Terms and Conditions

ENNOIA AND

IIPOAHPI2 IN THE STOIC KNOWLEDGE.

THEORY

OF

THE starting-point of Plutarch's dialogue de communibusnotitiis is a claim made by the Stoics that Providence sent Chrysippus to remove the confusion surrounding the ideas of 'vvowa(conception) and (preconception) before the subtleties of Carneades were brought into play.' wp6g.teL Unfortunately our surviving information on the subject is so much less full than could be desired that it has again returned to an obscurity from which there are only two really detailed modern attempts to remove it. The one, by L. Stein (Erkenntnistheorieder Stoa, pp. 228-276), is most unsatisfactory;2 the other, by A. Bonhbffer (Epiktet und die Stoa, pp. 187-232), though of the greatest value in many ways, is vitiated by the fact that it constructs a system from the use of the words by Epictetus and then attempts to attach this system to the old Stoa in the face of the evidence of the doxographers, which is emended or violently interpreted to suit Epictetus. Even if Epictetus were in general a good authority for in the opinion of H. von Arnim he is not3the technicalities of Chrysippus-and this would not be a sound method of procedure. The only safe way is to take first the statements which can be attached to the Old Stoa, and having obtained our results from these, to see whether Epictetus does in fact agree. Bonh6ffer contends that Kov4 vvota is equivalent to rrpd~X#l (though finally he restricts it to the most general part of a preconception, that actually brought to consciousness in the mind of every human being); and that rrpoXIpubEs, which are 'spermatically' inborn and develop independently of any sense-impressions, only occur in the restricted field of moral conceptions and conceptions of the divine. The unsatisfactory nature of this scheme is most easily shown by a piece of evidence with which he was not acquainted.4 This is from the work of Alexander de mixtione, r JLEwS WrELp~raL TL SVF II. 473: Tb 8 (sc. Chrysippus) Samopa~sdEaL 7, aJrvcas 3E WTLO-TOVO-OTaL cL~ TWjV KOLV&)V EYYOL&W) J~cXLYT~CL KPLT-qPLct r~~ chLX-qElcL~yLcv 7j~ w~ap~i ?ri If and are then equivalent, -acv'aS. Chrysippus AafELv wpoXh5bgEs KOLvaV 'vvoLaL 4o0XrEW did not restrict them, whatever Epictetus may have done,5 to the moral and theological field. There is need, therefore, for a new consideration of this part of the Stoic theory of knowledge, namely the conception, gvvota, and its subdivision rpo'Xrtt. I. The starting-point of the discussion must be the generally accepted identity of KOLV1 .6 The evidence usually quoted consists of two passages vvota and wpo6Xrb which are supposed to show that preconceptions are universal and therefore KOLVCU One is from Seneca, Ep. 117, 6: 'multum dare solemus praesumptioni EvvoLaL. (= rpoXAtet) omnium hominum et apud nos ueritatis argumentum est aliquid omnibus 1
KaL 1059C:

rav
ELS

7repi

rds

7rpoX4?fELS KacL7r9 KGa &opOrcas

zVolas Kc67rrV

rdpaXOZ dc?eXaw

7LZ7radrat

LoEVOS 2 O-t of H. von Arnim, See e.g. the opinions Deutsche Litteraturzeitung, 1888, p. i6; A. BonE. Bevan, h6ffer, Epiktet und die Stoa, IV.-V.; Stoics and Sceptics, p. 6. 3 SVF (=Stoicorum Veterum Fragmenta) XVII., Musonius, Seneca, 'ad Chrysippum Epictetus,

TO OKELOV.

restituendum nullum fere usum praebent.' 4 S VF was, of course, not available when he wrote. 6 Evidence given later will show that the restriction does not hold even for Epictetus. 6 E.g. Zeller, Geschichte der griechischen Philo-

sophie4, III. i. 76.

(All following references to

Zeller are to this volume.) Ueberweg-Praechter, Geschichte der Philosophiel2, I. 418.

This content downloaded from 134.173.199.1 on Thu, 10 Oct 2013 03:49:04 AM All use subject to JSTOR Terms and Conditions

F. H. SANDBACH
uideri';
KOLVa.

45
'-~s vvolac
is lost
Ki 7

the other is from Plutarch,

comm. not. io6oa:

ivIVLcLXTR7a r7V aLpErLv . . . (the

w-~ in a lacuna).

7~ rpOh/ELS S

The natural interpretation of these passages is that some wpoX35i/Ea are not universal. For if all WpoXjbEL are universal, why should anyone take the trouble to add 'omnium hominum' or i% KOLVaS? But these additions are not uncommon;
Sextus, reproducing Stoic
dvpirrop v

syllogisms,
do--TL7rl

says
(adv. M.

EH'o-s 6( YE KaRL rs
IX. I24, SVF II.

KoLVa

WrpoXThbEs rrvrov

1017).

EvvoaF KcL Plutarch

at KoLvaL rrpoIEL~s (IO4If, lo73d, Io74f). Epictetus regularly uses the expression de mixtione (SIVF II. (Diss. IV. i. 42) speaks of ra&wpoXAEL~~ r&S KOLvdS. Alexander 475) shows that the theory of KPco-~L &' (XvOYis rapi 7& KOLVcL From such rrpoXpLb,. passages the natural inference is that just as of EvvocaLsome only were universal, so
it was with wpoX'AELs; the non-universal preconceptions whose existence we should infer, we actually find in Plutarch, comm. not. Io84d: XX& rca3ra t pv rapa r& KOLV L
/LadCovacL 7rpoX

eEs

EKELva

(' 8Sy

rrapa

7rs

8la,l

and

Io8Ib:

TEo-r rrCvT)

It might be objected that Plutarch used the word in a popular un-Stoical way. But it is just from Plutarch's use of it in Io6oa and from a similar use in Seneca that the very idea has been drawn that for a Stoic all preconceptions were universal. Further, as far as I know, no ancient author ever suggests that there was any peculiarity in the meaning attached to the word by Chrysippus; if there had been, it would have been a likely ground for an attack in de communibusnotitiis (cf. Io84f:
VvoLcr au;7V
oVrOGv

d~O'todKV

KaL

.
o'OtfcT-ro&v

rTL rrEptrv

av;roL;

Tapc7lTEO-Oap rPOXkbELS.

a ii wRap

vvoios

E'XYXOIIEV OV;TErrWLOvoO

But we have still to reckon with the evidence of Epictetus. Diss. I. xxii. opens with the statement: rrpoXujEL; The context shows that v pdrropLS KOLVaU lrw-'LV El'v.2 he is thinking of the preconceptions of the good, the beautiful and other simple moral 'rb 8&KcLOV KcaXdv ideas. As an example he quotes: That his i(TT Kcl irpTrov.3 &phrase must be limited to such things he shows himself, for in other places he speaks
of
KvvouTo, (III. xxii. I). Epictetus' statement is, then, inconsistent with his own use of the word; there is no reason for believing it to be consistent with that of the Old Stoa.4 We thus see that the evidence by which it is sought to prove the identity of
KOLV) The interpretation of the doxographical evidence, which will opposite conclusion. be considered next, is greatly simplified by the abandoning of this supposed equivalence."

,vvo,;a

;).

cAXXo vvv q rTv aCIpE(oLV and Io73c: VrrOTLrEVraL, wpdLlO-,oLE o OV' KOLVO.S rrpc~ytacUtv ;otLLXu7/CLEVOL EKo-Tpovo'EOaV dVdOVOao'L "rL

wpoXujbEL;which

we

cannot

believe

are universal,

e.g. rrpd6Xryb 70-o

rpd6Xrphband

VVOta is quite

insufficient,

and that some of it points

to a directly

1 Cf. the parallel in Io62a: expression o0 XX& EvvoaS cLtXoo0oV7wo'r&v L6vov r-aph r&s KOLV&S
Kar 7 TitiarS
KUKL iT'JV.

word in its proper meaning.

2 The same thing is implied in II. xi. 2 in a similar context. Bonhbffer also quotes, in a different connexion (p. 220), Nemesius 203=de
Xi'yoev

5 We may notice that when the context makes the meaning clear i irpb6Xu~s is used to mean u comm. not. 1o75e: KOL1V rp6Xu9tas, e.g. Plutarch,
irpds rv 'E'riKovpov o0h8v ~droXEi7rovcL 7rG"v rpayxCiro&w Wyttenbach) 1o6 io0 pei qeO (dv oihevi, ypa/incirwv E s av 7y iv 7riV Oediv irp6Xutv (cf. /oov"r oUyXov7ra This is exactly paralleled by the very Io75a). KOLV vvoLta (rap& 7riv frequent use of i7 gvvota= o vvodoiv EO7LY 1073d, Io77a, Io77e, etc., 7rs repi OEV dvvolas Kai /L3iceoOact o-XyrXta iroteyv rdv 'EriKovpov It is Io82e). ivvolaS this usage of 7r irp6Xujts that accounts for 0lo76a,

anima hominis, Migne, Vol. XL. 661, ivvolas rcs d~ds&iKT5Os irUL irpooodaas, eiat Oebv. But the thoughts here are not so much

hi voULK'& dis 7d

Stoic as Platonic; the point under discussion is tlvod/zr/o't . 3 Cf. IV. i. 44. 4 Bonh6ffer attempts (p. 198) to explain away the inconsistency. He is not very intelligible, but seems to be driven here as elsewhere to the assumption that Epictetus did not always use the

Xdoovo'

probably the persistence of the view tions are universal.

that all preconcep-

This content downloaded from 134.173.199.1 on Thu, 10 Oct 2013 03:49:04 AM All use subject to JSTOR Terms and Conditions

46

F. H. SANDBACH

II. Diocles Magnes (in Dio. L. VII. 54) defines rpd6Xrp Ls as a 'natural concepIt is thus tion of the general characteristics of a thing' (i'vvoLa #vo-LKrj r-v KaLXov).1 which was, according to the Stoics, the essence of a opposed to the 1lov rd6'oorLs, is one distinguishing mark of the preconception; it is an definition.2 This, then, as opposed to the thought-out definition. For example the undeveloped conception, Stoics said that all men preconceived the gods as immortal and blessed and benevo^ovdO4varov XoyLKbV 7XELOV ; lent;3 but their definition was a development of this:
voEpov Ev EV&pLovli. KK~KO1
wravTO",

dVETWIEKT7LKOV WpOVOy?TLKOV
KOO'&Ouv

7E

KRLL

7V

(SVF

Preconception is limited by Diocles to conceptions that are natural. What these are we have to try to gather from a mutilated passage of Aetius (Plac. IV. 1= T 6 LvOpo(o, O'(TrrtEp 7pO II. 83): 6'ray SVF EXEq}fOVLK)V /x yE.vvyeO7 ES
XapVrhv
'pJoJ

II. 10o2I sqq.).

V, K00L9)"/

,3 rLEXOOrVTo

EVEpyov El. dvaypcLs 7r ,

drrTOYpa7V. ow "rp4 tv bLv7JJLJv EXov yap

0,t . , T

rv 'ral'

1070ro r J

Jlav

EKdO"Trv

owv O/LOELSELS

r'ov

EvvoLv

EvarroypaT'rat.

alo'O 8E

" ao'aOlEvo, OVa rVrTwXv KL

yc;p fat WrUO

TLvos

otoV

rwoXXaLfLvj

E i'XELV ELLgrEplav [LWEple at /LEV 4VOLKOJ~ lVOV7c(L *7fLETipas 8&XOKaXlaK

ETL

^vo6/oELv 7TW

7EvouiVaL * ***

EVKOV, d fLEY TOTE EVVOLWv E

v 7 8V

TOVU

E p'17/LEVOVS 7pOrrOVS

VErLTEXVTWS

r)
KaL

KLL ErL/LEXELLS cVTcaL 1LLV OYV EVVOLCLL KcLXoVV7LL KclO OV E71pocr yop?urvLEdcL
XOYLKOL

AdvOV,

c EKE VaL

WpOXqjLELs.

0 6E XdyOS

EK 7TV

7tpOX 7EwV

(Tp.X~r pO%(T

XEElJaL

KLand it is supposed that originally the passage resembled one after rhXOos, fallen out

77-V r-p&'rTv

~0popda.

The

plural

Ipdrovs

shows yap

that
voovLvEv

something cLv

has

in Diocles
7rrpl~?rTocrv Ka7O
AK

Magnes (Dio. L. VII. 52=SVF

II. 87): r1v

7~~
Ka,

KaO' 4L0L6np)a (then

Ta

bE

KcL-3 cLvcLXoyW.V Ta 6 KaL

KcL7L ,LEa8crLOLv [JE7C(rCO'IV KcXTO E'OLV KLT(T Kcl

i-

aL-V T(
CS 1rvc OLOY

d
T , VOE'LV KLL
0
* TtLVOELCLL TOS

follow
&KCLLOV

Kat

WcO

to 4LsrEpla in Aetius,4 and the following modes corresponds aXELp. Here as -ros to those there rwEpurrmoTo' rpdnrovs.5 The two passages omitted, but summarized

EApy'~vov, together give a perfectly consistent account of a preconception as the first conception of a thing, arrived at without special mental attention, and derived either directly or 1 A difficult phrase; Bonhbffer tries to connect not derived from rd& Kca6Xov with conceptions experience, as being less definite than those that are. This is connected with his theory of the to the moral field, restriction of preconceptions Stein which is, as we shall see, certainly false. 'die rendering, (p. 511) gives the impossible iiberall waltende Weltordnung.' 2 For the general idea of this distinction cf. Galen in SVF II. 229: Pdpj@LE' o0v acldts a-b rwv
p/lYEvELEw 'TrXov 8pwv oi)s 0oV' Ca/ULEV &VVO7lzCTLKa yU'C$KOUCLY . .. o. o& Wv UravrES ?vLOpWeot 7rep L broypag8r Io5Ie. V

and seems to have meant 'direct experience,' as appears from the non-philosophical examples quoted in Liddell and Scott: Hippocrates 26, i,
KaraTcLV (certain
Tdv

f XoyLtcYdv

6v

rEp

LK 71EptLrT&LOS

Plutarch, rodl'rat r7v CipXYv,

QN. 918c, ro6TWY8&

KraL

use them in case of need). To separate the philosophical examples from these, and to make
the subject
irplrrrwcrs necessary, though done by Bonhoffer (' Anfall an die Sinne') and in the latest edition of Epicurus (' by the fitting of one experience on to another'

ErpirTWroS

orTe <&?actWXtCa irOEV 0o76e>) wrpca that yyovPEv acTro0s (the wild beasts

of the verb wreptrirW implied in not the man but his cavraoLa, is un-

S r7rpoJ0yoplas
7e

o06' 6povus dtoaov, 01E.LVO2 dXX'

comm. not. d6VOtdi SCf. Plutarch, irroT7r3ELtS' Io75e and Stoic. rep. is rather an obscure word; who-

Kal

C. Bailey,

introduced it to Greek philosophy, it did everrepLzrr~cs'

plural rp6rovr has also been explained as meaning twgjlgv KcdltLprEplaV (Diels, Doxogra hi
Graeci, p. 400). Aetius is then committed to a

5 The

Epicurus, p. 415).

KocL &vaXoyiav

Kai 6t0o6r177-t Kal cr6veoftv,

cvJfpaXXoLdvov

as if it were a generally accepted notion (adv. M. VIII. 56). Cicero appears to translate it by' usu' (Fin. III. 33, see below). It must be connected in the neutral sense ' meet with,' with
repairmrroW

r 70roXoycoipo, and Sextus

uses it

statement of obvious incompleteness. The explanation given above supposes no inaccuracies, and is in accord, for what that may be worth, with Cicero, Ac. II. 30 (= Antiochus, see Zeller, 619, n. 2): 'cetera (sc. uisa) autem similitudinibus (= KaO' construit (sc. mens) quibus 6tot6rb77a) efficiuntur notitiae rerum quas Graeci tum ivvolas
tum 7rpoXCt/Ers uocant.'

This content downloaded from 134.173.199.1 on Thu, 10 Oct 2013 03:49:04 AM All use subject to JSTOR Terms and Conditions

ENNOIA AND IIPOAHXPIE

47

by some simple and unconscious mental operation from the data given by the senses. This straightforward interpretation could not be adopted by anyone who believed preconceptions to be universally held. There must be many conceptions that cannot possibly be universal, yet arise naturally from experience, and ought therefore, Thus the champion of the rule that according to Aetius, to be preconceptions. preconceptions are universal must assume, as Bonh6ffer does, that Aetius entirely On our explanation no misses the difference between conception and preconception. such assumption is required. III. Bonh6ffer had two other reasons for not accepting the evidence of Aetius: he supposed preconceptions to have been limited to fundamental ideas of morals and religion, and he thought that they were inborn in potentiality, or 'spermatically' as he put it. There seems, as it happens, to be no case in which it can be shown that a member of the Old Stoa spoke of a preconception of anything but the gods. For preconceptions of natural objects we have to go to Epictetus, who uses the phrases
T-KTOVOS, 4<L0O-4ov, a pure accident, the word 'vvoLa is used where the due to the fact that in our fragments ,YLELro more definite term might have been used if accuracy had been consulted. -pdrlXs An example is supplied by the passage from Alexander, de mixtione, quoted in the second paragraph of this article. Here the KOLvaL~vvoLaL concerned with mixture cannot be anything but preconceptions, since the words )fLs rap~i v 4vo-Lvhap/Ev which on the undoubted evidence of Diocles ra4ras show that they are cvo-LKcL ;'VVOtat, (Dio. L. VII. 54) are equivalent to wpohXgEbs.1 Plutarch not only says of many pieces of Stoic doctrine outside the field of moral conceptions that they are contrary to preconceptions,2 but he also states that the Stoics attacked the Academy as philosophizing against the preconceptions on the problem of change and identity a subject that lies outside that field-namely, not. o83a-b). (comm. I Indirect evidence, too, speaks against the alleged restriction. Preconception, according to the Stoic theory of knowledge, made possible the search for and discovery of new knowledge.3 That is, having a general idea of the characteristics of a thing, we have an indication of the lines to follow in a search for more definite If preconceptions were limited in the way Bonh6ffer suggests, they knowledge. would be of very little use for this purpose. Again, a man's reason is said to be formed from his preconceptions.4 Moral preconceptions would form a very insufficient basis. Bonh6ffer realizes this, and adds as a note (p. I93, n. 2), 'wobei jedoch nicht bloss die 4vOLKat (sc. rpohbEss) gemeint sind,' but he does not say what is meant.5 IV. A more difficult problem is presented by the origin and formation of preconceptions which are universal. We may take as true in general the account gained by combining Aetius and Diocles, and yet doubt whether there was not something more behind some universal preconceptions, something besides the winning of an idea from material provided by the senses. These preconceptions are those of the moral sphere.

prp6JrLs

JLOUVLKOV

(IV.

viii, io),

(II.

xvii,

9).

But

this is

1 Cf. SVF II., p. 32 note, 'vUOTLKa~L vvotac apud

Chrysippum eadem sunt quae

E.g.

Io59e:

7rertoTw yovOV
0

rpoX45pES.' and lo73d: rdV rp6Xrp/LV

SSee the last sentence of Aetius quoted above. SVF II. 84i, calls it an Chrysippus, -vvoLov rT7tYVWV KCLL 7tpOXjVEWV
5 The only other kind of irp6X?tLs he mentions peculiar among Stoics is the rp6Xbrlts &'qpOpwpvr, to Epictetus, which is properly not a preconception at all, for it is obtained &' Ouer/dpas&LcaOKaXiaS Also if the field of preconception Kal ILreXelas. is restricted, so must be that of 'articulated' or developed preconceptions.
(0potO/cL.

ot &da~ VVOlGS rTS UUOLK~S Cli7i,7C1L. rfs zToarL

"7Ep S 7rpOX'?i Es. a7tTapdTTOToa de Bernardakis fr. anima, VII., 6, a Plutarch, II. Io4, under the heading 'r( p. 29=SVF tl'ropov 6vrWS eL Or6v re ?7TcEV KCd ElpiOKELV doS eV MVOwLt well-known (the problem, p. 8oe), rpofiX?17rat dT&s KOcV

LKdV CUOT

iTi

XTV

X j0TTOV

This content downloaded from 134.173.199.1 on Thu, 10 Oct 2013 03:49:04 AM All use subject to JSTOR Terms and Conditions

48
rrpoX?)Eov (SVF III. 69).

F. H. SANDBACH

Chrysippus said of his account of good and evil (&rrew1EaL) EJlrbrToV All depends on what meaningXdXcra we are to give rv to ~4vroS. The orthodox view is that it cannot mean 'inborn,' as this is contrary to all the other evidence. Although i'ptwvrodoes frequently mean 'inborn,' there are a certain number of cases where it means 'ingrown,'' ingrained,' as Isocrates 222A, Euripides fr. 377 3' 4I'vurov), or 'implanted,' as in James I. 21: &ioa'OE rwovroie v'oTro (70-o T-v TV 3VVMEVOV It is not necessary, then, to take Xoyov v. o-rJrat -&s /v3~ I't4vurov 3tp as meaning inborn. Further, as the following paragraphs will show, what Ek'wov0o evidence we have on the origin of ideas of good does not in any way suggest that anything inborn played any part. In spite of this, it is difficult to feel confident that Chrysippus did not mean 'inborn' when he wrote the word. But if he did, it was but a temporary aberration. The other passage advanced in support of the idea that moral preconceptions are in some way a priori and not derived from experience is the sentence of Diocles S?VOE7tL KacLOv (see above): pWo-LKw~ 7L KaLdyJYaO6v. Bonhiffer interprets this as opposing to the preceding modes of conception, which are all ultimately based on sense, a natural mode not so based. This explanation leads at once to a contradiction with Cicero, Fin. III. 33 (SVF III. 72): 'cumque rerum notiones' (= 'vvota) in animis fiant, si aut usu (= KOaRT aliquid cognitum sit aut rEplrrrwoov) coniunctione aut collatione aut similitudine (=KaO' Afotor/7ra) (=Ka-r ao'vO8~Erv) rationis (= iur' JvaXoylav), hoc quarto quod extremum posui boni notitia facta est. cum enim ab iis rebus quae sunt secundum naturam ascendit animus collatione rationis, turn ad notionem boni peruenit. hoc autem ipsum bonum non accessione neque crescendo (= aZr/tK;Ks) aut cum ceteris comparando sed propria ui sua et sentimus et appellamus bonum.' BonhSffer explains that this passage of Cicero refers not to the first conception of the good, to the rrpdA6X7s-rdyaOo^,but to a development of this conception. This is unconvincing in itself, and made impossible by the similar passage in Seneca, Ep. 120, 4 sqq. Here on the question 'quomodo ad nos prima boni honestique notitia peruenerit,' Seneca says 'nobis uidetur obseruatio collegisse (sc. speciem uirtutis) et rerum saepe factarum inter se collatio: In point of fact per analogian nostri intellectum et honestum et bonum iudicant.'2 this passage of Cicero supplies the clue to the word qvo-tLJK in Diocles. First it VO must be noticed that the sentence /o,-LKw9 8~ introduced KaLI is S6VO tIaLdV dyaO6v "r as a kind of postscript to the original list. Now the examples that Diocles gives ELV(iS under the heading Ka7' ivaXoylav are as follows: K~a7'dvXoylav 6 aOVTtKWS ~ 0 HV7LctY~iOS, KcLL 70 KEV73POV 77jS 7?7SKLLT JvaXoylav JLEUO7LKO); (i)S ~ Ttrv~; Kai K-6KXWo, Evorj7O drcb 7rv op aLtpmv. But Cicero expressly guards against supposing that the good twtKpOT'PWV is conceived by any of these methods (' crescendo' corresponds to the example of the Cyclops, 'cum ceteris comparando' to the example of the small balls, and no one would suppose that the good was conceived by 'diminution '). Just so, whoever made the addition to the original list in Diocles noticed that the good, though conceived 'by analogy,' was not covered by any of the examples given; accordingly he supplied the sentence under discussion, meaning by uvo-tKJs what We recognize Cicero calls 'propria ui sua' (and later 'genere non magnitudine'). the good through the force of its own nature.3
1 Whether called an hvvola or rrp6X'pis the z thing meant is the same (cf. Aetius, KEiVtL EKal es, quoted at beginning of Section II.). urpoXkip The translations are those given by von Arnim. The index to SVF gives notitia=iirurjr but y, bear this meaning though it may sometimes Cicero it as a synonym usually employs for notio.
2 It is true that Seneca says: ' natura semina But in all scientiae dedit, scientiam non dedit.' that follows no place is given to any inborn ideas in the formation of the conception of the good. The 'semina' seem to be the facts observed. 3 It may be noted that no trace of innate preconceptions is to be found in Panaetius' account of the formation of moral ideas, reproduced by

This content downloaded from 134.173.199.1 on Thu, 10 Oct 2013 03:49:04 AM All use subject to JSTOR Terms and Conditions

ENNOIA AND llPOAHPJIE

49

It must be admitted that when we come to examine Epictetus he can hardly be interpreted otherwise than as believing in 'inborn' preconceptions, but it is not justifiable to transfer this to the orthodox Stoics of the third century. Between Chrysippus and Epictetus lie the but half-charted waters of the syncretism of the first century, the results of which we find in Cicero. Stoicism affects Platonism, as in Tusc. I. 57, where Cicero speaks in an account of the Phaedo of 'insitas et quasi consignatas in animis (= notiones, quas ivvoCa uocant'; and Ivarr-cpayto-piras) Platonism gave to Stoicism a belief in inborn conceptions, as in N.D. II. 12: 'omnibus enim innatum est et in animo quasi insculptum, esse deos . . Cleanthes quidem noster quattuor de causis dixit in animis informatas deorum esse notiones.' The four causes are various classes of natural phenomena, but the inconsistency between this and 'innatum' Cicero does not observe or else neglects. The effect of Platonism on the mind of Cicero, or on that of the authors of his sources, is seen in the metaphors which he uses in speaking of conceptions, even where he purports to be reproducing Stoic ideas. To Chrysippus a preconception, though it might be incomplete as an account of a thing, was perfectly clear as far as it went.' But Cicero, with memories of used metaphors which imply that dvctLvyo-Ls, the' notio' contains the whole truth but is but dimly seen, through being in darkness or covered up; Tusc. IV. 53: of all the Stoic definitions of bravery, 'quae non aperit notionem nostram quam habemus omnes de fortitudine, tectam atque inuolutam ?' Off. III. 76: 'complicatam notionemn euoluere;' 8I: 'explica atque excute intellegentiam tuam.' V. So far has been treated as if it were a technical term peculiar to the rrpdt6X~b Stoic school. In point of fact we are told by Cicero that the word was invented by This has been unEpicurus for an idea which till then had no name (N.D. I. 44). reasonably called into question, mainly on the ground that there is a difference in meaning between the Stoic and Epicurean terms which makes an independent origin is probable.2 The chief of these supposed differences are that the Stoic rrpdXhbL common to all mankind, and that it is confined to moral and religious conceptions, whereas the Epicurean may be individual and is in no way restricted in subject. From what has been said above, it will be seen that this rests on a misconception of the Stoic Also, had there been a difference in the meaning of the word as rpdkX~b. used by Stoics and Epicureans, some ancient would surely have mentioned it. As it is, the word is constantly used in the disputes between all the schools, without any suggestion that the parties meant different things by the term. What difference
VI. 6 (Schmekel, Die mittlere Stoa, he enquires whether any theory has done greater Polybius This agrees with Cicero and Seneca p. 64 sqq.). outrage to normal feeling than that of the Stoics in making them rest on the observation of good v 70ro? about the good, KaL he continues, rauRa, and evil actions, and ends airy KaXoOKac GLKaiOV IrEpL dyaGiv KaZ KaKWV LEVcKTrw e pe7rGV TKai a TE Innate -rap' VVOLca. dv6pCirOLtKCL7& OiKELWV KaL dOXXorpiOv & OCVXXOV a6eL OEpp v [re] 7rpr~77 r'LV are equally lacking in the Posipreconceptions Kaii/vxpwv XEVKWKV KIa Te /.eXdYV o Oc/txOrEpOV eXELV c donian passage in Aetius, Plac. I. 6, headed Yp E0w V ELOLV caL KeIVWV Tr7V fVdpYetav l~ (SVF II. 1oog). rTb0ev OePv rvvotav eXa/ov Oavrrcitat ras 6reto6&ot, raCra 8' dK rVj &vOpworot aloi'O;cr't Panaetius and Posidonius were both Platonizing -Gv 4v But yIveCvT. oLrhpVurov yXeL UZv dyaOrv r'v of the believed in I am uncertain of the meaning Stoics, who, had their predecessors last innate would have sentence, and uncertain whether Plutarch is scarcely preconceptions, A very similar account here confuting the Stoics out of their own mouths suppressed the notion. or not. of how early man won the conception of divinity 1 It must have been a cpavraaia is given by Dio Prus. XII. 27 sqq. Though the KaraXrTLK'rK: of nature is the only method he observation cf. Antipater's use of the word ivcdpyeca, considered below. mentions, he does not hesitate to call the result2 L. Stein, Epiktet und die Stoa, pp. 248-250. ing conception UC/Vros; cf. especially c. 39:
t4uvrov &iraccvw dvOpdj-rots ~7rivotcav comm. Plutarch, dv'qvl r&3vUpyvJ.

rhV

acr&ov ",yvonot. Io7oc-d,

He also alleges that Cicero is an untrustworthy witness. D

may perhaps be relevant

to this question.

There

This content downloaded from 134.173.199.1 on Thu, 10 Oct 2013 03:49:04 AM All use subject to JSTOR Terms and Conditions

50

F.H.

SANDBACH

there is lies not in the meaning of the word, but in the accounts given of the origin of the thing denoted. For Epicurus only arises by the way of memory, by wp6r-l~ the coincidence of several presentations of the same object: the Stoics, while retainthe other ing this method, also introduced as possible ways of forming -poXrjes simple mental operations, dva)oyla and the rest. spoXg#EL~ In a passage already quoted Plutarch uses the phrase KaL rEp; 7-v 7-s -dS IvvoLaS 7dpaXov deEXwv(comm. not. 1o59c). The nature of this confusion can only be guessed, but the Epicurean origin of the word suggests that its first use by the Stoics may have been in argument against the Epicureans.1 They used 'the preconceptions of all men' to support their own views on the nature of the gods. The first Stoics probably denied that it supported Epicureans, and claimed that in fact it was in accordance with their own views. If this line of argument is first attested for In Chrysippus, it is so obvious that it must have presented itself to his predecessors. this way there would come into use a term, rpXrgS, for which no place had been We can imagine Arcesilaus and his made in the Zenonian theory of knowledge. followers turning on the Stoics and demanding from them definitions of this new term, distinctions between it and the familiar EvvoLa. A quotation from Cleanthes suggests another way in which difficulty may have arisen. Among the causes which he gave to account for the origin of the conception of the gods we find ' tertiam quae terreret animos fulminibus, tempestatibus, nimbis, niuibus, grandinibus, uastitate, pestilentia,' and by many kinds of portents, 'quibus exterriti homines uim quandam esse caelestem et diuinam suspicati sunt.' It seems that Cleanthes is giving an historical account of the origin of the conception of the gods without any idea that this primitive conception may be used as evidence of their Thus he allows that men should think of the gods as terrible and harmqualities. is It significant that no parallel to this argument of Cleanthes is found in any ful.2 other Stoic writer.3 VI. This discussion of the word wrpd6X? may be concluded4 by a consideration , In Chapter 38 of namely, ivipypa. of an idea which has some points of similarity, de Stoicorum repugnantiis, side by side with the 'conception' of the gods, we find an and runs as follows: 'We vapyea appealed to. The fragment is from Antipater will give a brief account of the ivipyea which we have about God. We conceive of a god as a blessed being, incorruptible and beneficent.' This EvdpypEatallies exactly with the wrp6Xsb of a god. The origin of the word appears to be, like that of wrp6dX)b,Epicurean. Epicurus was of opinion that we must trust our gavrao-la because of their clearness.5 By a
transference of meaning
:vipyea

was used

to mean

a favracria.6

As far as I know,

the word does not occur in any fragment of any Stoic earlier than Antipater,7 and it is at least possible that he is using the word because he is attacking the Epicureans for denying beneficence to the gods.
1 The idea that it was borrowed to meet the
attacks of the Academy seems less likely; for they were concerned chiefly with the validity of in support of which preconsense-impressions, ception, and particularly the Epicurean preconof ception, based only on the accumulation would have been of little or sense-impressions, no use. 2 Cf. SVF II. 1115: rbv 'EriKovpov It?OXLoTa
diXEraL (sc. T7rV
ElEpyE7LK)olj

7rp6Xprp/s pretation seems to me to have been given by E.

I have nothing to say as a test of truth.

on the subject of The correct inter-

Brdhier, Chrysipfe, p. o103. Ep. ad Men. 123: OEol 5 Cf. ' yVWGet. alb)Tv iYTLv
d'apyfy
203: yhp

/EV

yap Erlv.l

6 Usener,
T.1

Epicurea 247=Sextus,
K) (V K

adv. M. VII.
Zeller,

Vi

pyELaV KaXrEL.

o 3 civaLpovGras 70rp Ka ?p Xp6roL7ros) Os Xopiev rEpt OleGv, rp6VOaJ dCrb 7TV uVVOLWVj
Kai ~LXav pc 7rovU r.LvooOlVTEs. of Fragments

'phe

3 A. C. Pearson, Cleanthes, p. 284.

Zeno

and

n. 3, calls this an 'eigentiimliche III. i. 401,a/XvrcTaav Bezeichnung.' 7 It is not entered at all in the Index to S VF. Yet E. Bevan, Stoics and Sceptics, pp. 35, 37, talks of enargeia in explaining Zeno.

This content downloaded from 134.173.199.1 on Thu, 10 Oct 2013 03:49:04 AM All use subject to JSTOR Terms and Conditions

ENNOIA

AND

HIPOAHPII

51

But at some time or another the Stoics did adopt the word just as they had es; Posidonius uses the adjective hvapy~s(apud Galen H. et P., p. 400, adopted pXpdXy in Plutarch comm.not. Io83c the Stoics appear as of Wp6d&KO0 and vapy da Miiller), 7T It was enough for the Stoics to adopt the word, because by AvipyEa easy 0o70o. Epicurus meant to denote just that quality of a xvrarcla which Zeno denoted by the word KaTaXyrW1LK4, that quality which makes a man feel certain of its truth. Epicurus expressed this actively, making the presentation real enough to command belief; Zeno passively, making it real enough to be capable of being firmly grasped by the mind.'
The words ivapy~s, ;vpyELa provide a good example of the way the schools

borrowed one another's vocabulary, for not only did the Stoics take them from the Epicureans, but the later Academy took them over also; Philo of Larisa agreed that some things had this clearness and were certainly true, though he denied that clearness implied comprehensibility, as the Stoics maintained.2
F. H. SANDBACH.
TRINITY COLLEGE, CAMBRIDGE.

1 This interpretation

is well upheld by E. Brdhier, Chrysippe,pp. 8oT( 7 ioo. The famous phrase ~p6vov rptXv oxVyl is definitely put down to youngerStoics Xa/pciveTrat

of the word KaTraXryrTtK by Sextus, adv. M. VII. 257, and is not applied to every cavracila KacaXT7rTLK?, but to such as 'con-

tain no obstacle.' 2 Zeller, p. 616, n. 2: Cicero, Ac. II. 34.

This content downloaded from 134.173.199.1 on Thu, 10 Oct 2013 03:49:04 AM All use subject to JSTOR Terms and Conditions