Xenophanes' Scepticism Author(s): James H. Lesher Source: Phronesis, Vol. 23, No. 1 (1978), pp.

1-21 Published by: BRILL Stable URL: http://www.jstor.org/stable/4182025 . Accessed: 19/04/2011 05:32
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 support@jstor.org.

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

http://www.jstor.org

Xenophanes'Scepticism
JAMES H. LESHER

Xenophanes of Colophon (fl. 530 B.C.) is thought to have been the first sceptic in the history of western philosophy, but the character of his scepticism was the subject of dispute as early as the fourth century B.C., and the central statement of his position, Fragment 34,1 has been variously

interpreted ever since. Much of recent discussion has concerned the severityand scope of his sceptical thesis, but it would be of equal philosophicalinterestto know what Xenophanes'reasonswere for maintaining scepticism, or what features of Xenophanes' life and times may have contributedto his scepticaloutlook. It is the contentionof this paperthat Xenophanes' scepticism is best understood as a response to traditional religiousand poetic ways of thinking,and is thereforeclosely tied to his criticismof Homericreligion,and thatthe key to a properunderstanding of his sceptical remarkslies in an often mentioned but seldom discussed aspectof his writings:the attackon divination.
I - ANCIENT ACCOUNTS The following extract from Xenophanes' poetry was widely quoted and discussed in antiquity ( B 34):
XaL TO 1VEV 0VVOW1)rS OVTLS&'Ap LMEV OVU8 TLS WT(tL TE XviL E18WS &UGK X\eyW ITEpi -N6t&vmWv &R1 OECOV xoviTX FXXLOTOM TV XvOL TerENXeuyvov L'rTW'v, ya&p S' eFrL 17>r' %&y b OX0S & s~~~~8OL& aWTOS O1aS OVX 0 8'xos rIIOL T'TVX_ eTvxTa. tL et

While a full translationrequiresargumentand cannot be assumedat the outset,it is clear that the basic elementsof Xenophanes'view are (in some sense of these terms):truth(To x0s), knowing ('Cev, d&8,s, oLE), speaking of the real (TETEXE%p.?VOV EVL'6v), and belief or seeming (80xos). Any adequateinterpretation of Xenophanes'scepticism,while it can be supplemented by referenceto other fragmentsand backgroundinformation, must make sense of these notions, and their inter-connections.Sextus mentions two alternative interpretations:in the first,3 Xenophanes is
thought to have held that everything is incomprehensible ('rT&vTa and this is supportedby readinguaawsas 'true'or 'known': &xaTaixn1rrTa), Yet the true and known - at least in respect of non-evident things - no I

human being knows; for even if by chance he should hit upon it, still he knowsnot that he has hit upon it but imaginesand opines. As the contextof Sextus'discussion(VII,46-52) makesclear,the sceptical disputeconcernsnot so much the attainingof true belief, but a criterion are whichbeliefsare trueor whichappearances (xpL vpLov) for determining of a existence veridical. Xenophanes is taken here to be denying the sincehe holds thateven if someonewereto saywhatis realor true, criterion he would not know that'hehad done so, and hence would have only belief or opnion. of the fragmentis implausible.Not only does it But this interpretation restrict,withoutjustification,the scope of the sceptical thesis to what is 'non-evident',but it translatesXenophanes'simple "'heknows not" (ovix oT C"E1414XpEV 0o8) into "he knowsnot that he has hit upon it" (ovx oL8Ev cxvTOfl), and requires that we attribute to Xenophanes the rather subtle doctrinethat knowingthe truthentailsknowingthat one knows the truth, or at least that knowing the truth entails being convinced that what one believes is true. It is doubtful, at least in English, that either entailment holds, but they are not so far fetched as to exclude being attributedto Xenophanes.The difficultyis simply that both are more complex formulations than the original "even if he says what is true, he does not know". Xenophanes'scepticare not synonyms,4 Further,since aot4 and &\XqOs ism could be directed toward 'certain','absolute', or 'sure' knowledge,
rather than knowledge of the truth simpliciter.

Sextus mentions a second, less sceptical, interpretationof this sort: Xenophanesdoes not deny all comprehensionor apprehension(xaTa'Xand &&L&'rTmTov 1ILS) of the truth, but only that which is '1EUTL?OVoLx11V (Bury: "cognitiveand inerrant").Men can apprehendthe truth, even if
they cannot attain it with certainty
-

or without reservations (iwoyyov) -

whatis and we can adopt probablereasoningas a criterionfor determining these "let from Fr. 35: true (VII, 110). This reading gains some support thingsbe believed as resemblingthe truth(1oLXO6r TOlS vIT4LoLU)". Yet we must still assume that Xenophanes' concern was also that of the later sceptics:the existenceof a criterionfor distinguishingbetween truthand falsity(or realityand deceptive appearance)and we are also requiredto attributeto Xenophanessome general notion of apprehensionof which given by knowledge and belief are distinct species. Both interpretations Sextus constituteexpansionsin the languageof a later period,and in the contextof a disputeof which,for all we know,Xenophanesmay have been concludeonly thatXenophanes whollyignorant.So farwe mayreasonably denies that men have knowledge(or perhapsthat men have knowledgeof 2

To

while he allows that men have beliefs which,in some cases, may uawos) resemblewhat is true or real (TV4LOLOL). Later writerstell us that Xenophanescoupled the distinctionbetween knowledge and mere belief with a contrast between divine and human capacities: god knows the truth, but belief is alloted to men,5 and Alcmaeon begins his work with a similar remark, perhaps following Xenophanes:"concerningthe non-evident,concerningthings mortal,the gods have a clear understanding (awvrlvEV), but men merelyconjecture from signs."6 As Snell has shownin detail, therewas by Xenophanes'time a well established poetic traditon contrasting divine knowledge and human ignorance,7and the attributionof this view to Xenophanes is supportedby the frequent,and often disparaging,remarksabout the beliefs of mortals (B 14: 13poTol, B 18, B36: OvyrCotL), and by his conceptionof the one God, superiorto gods and men, unlike them in body and mind (B23),who moves thingsby his mind (B25),and who enjoys,in some sense, a whole or completeseeing, thinkingand hearing(B24).We have so far no reason to think that B34, taken by itself, reflects this religious context (although I shall argue for this in Section III following), but we can reasonably conclude on the basis of the evidence already cited, that Xenophanes'scepticismhas this feature: as Guthrie puts it, "men could have no certainknowledge,that was reservedfor God" (ibid.,p.398).
II- RECENT ACCOUNTS

In Karl Popper's famous "Back to the Pre-Socratics",8 Xenophanes is alleged to have held that "all our knowledgeis guesswork, yet thatwe may nevertheless, by searchingfor thatknowdedge'whichis the better'find it in the course of time." Popper sees Xenophanes, as did the early Greek sceptics, as an early proponent of his own theory of knowledge: "that knowledge proceeds by way of conjectures and refutations",and not accordingto the Baconianmyth of induction.Popperconcedesthat it may soundincredible,but assertsanyway,thatthereis a clearrecognitionof this "theoryof rational knowledge almost immediatelyafter the practice of criticaldiscussionhad begun."(ibid).But even concedingfor the moment that Xenophanesdid thinkof humanunderstanding as progressing toward but neverattainingknowledgeof the 'final truth',we are still veryfar from being told that knowledge is obtained not simply from observationand experiment,but from the constructionand criticismof theories.9According to Diogenes Laertius,Xenophanesheld opinionswhich were opposed to (&vrLbo8&axL) those of Thales and Pythagoras,and we know that he 3

criticizedthe accountsgiven by Homer and Hesiod. There is no evidence that he held, as Aristotleclearlydid hold, that inquirywas best conducted by a reviewand criticismof previousdoctrinesand theories.10Nor do the revealthathe implicitlyfollowedthisprinciple;apart remainingfragments (B7) and doctrineof metempsychosis fromone allusionto the Pythagorean a suggestionthathe admiredThalesfor his abilityto predicteclipses(B 19), principleof thereis no sign thathe practiced,muchless preached,Popper's knowledge'. 'rational Yet Popper'saccount raises an issue that has clear relevance for an of Xenophanes'scepticism:how can Xenophanesbe a prointerpretation in time,through ponentof scepticismif he holdsalso (in B18)that "mortals Can one consistentlydeny seeking,discoverwhat is better(or the better)"? the possibilityof knowledgeand at the same time affirmthat discoveryis possible? Further, as Frankel asks, how can we consistently think of of realitywho took delightin and portrayer Xenophanes,"thisinvestigator the gatheringand contemplationof facts"as one who was also "a sceptic,a tired doubteror a deft but unconvinceddialectician,and that he had no (p. 122). real confidence in the reality of the world of appearances?" Neither of these considerations poses an unavoidable dilemma. that men find "thatknowledge Xenophanesdoes not say, as Popperhas it,"1 which is the better",but only that they discover what is better, and as Guthrie explains, the replacementof divine revelationwith human inquiry,which is the full thesis of Fr. 18, may representXenophanes'rejection of a primitive'golden race', and the promotion of a conception of human progressor improvement'both morally and in the conditionsof life' (p. 400). In any event, there is no inconsistencygeneratedby holding that men fall short of certain knowledge,or even knowledgesimpliciter, while also concedingthat men discoverarts,skills,values,or beliefs which are betterthan previousones. Similarly,one need not, in orderto be justly termeda sceptic,affirma universaldoubt, or a rejectionof the evidenceof the senses. Frankel'sargumentrests on the frequentlyadopted but mistakenassumptionthata sceptic,worthyof the name, mustbe a pyrrhonian sceptic, that is, must call for a suspension of belief, or perhaps even a rejectionof all beliefsas false.Therehavehoweverbeen sceptics,as ancient as Carneadesand as recentas Keith Lehrer,who deny that we ever know anything,but insistnonethelessthat muchought to be believedas true,not the least of which are the beliefs about the world based on the obvious thatp is in asserting evidenceof sense experience.Thereis no contradiction should which true,or ought to be believed,even thoughp is not something Consequently,we be claimed to be known, or known with certainty.12 4

cannot hope to show that Xenophaneswas not really a sceptic simply on the groundsthat one form of scepticismwould be incompatiblewith other aspectsof his philosophy. Frtinkel's interpretation is howeverbased largelyon linguisticconsiderationsinvolvingB34 itself,and since Frankel'saccountis shared,at leastin part,by Snell, Guthrie,and Untersteiner, and presentsa detailedexegesis, it deservescarefulconsideration. Frankelviews B34 as a expressionof a 'robustempiricism', and, far from denying the possibility of knowledge, it holds that knowledge which is empiricallygrounded,based on first-handobservationand experience, can be certainand exhaustive(aais). To know, as can be seen in Herodotus' notionof LaTopia, is to have seen, and thisclose connectionof knowingwith seeing is borne out by the etymology of oZ&a, literally"I have seen", but commonlysimply"I know".Thus Xenophanesis not rejectingknowledge, but only the pretensionsto knowledgeof one who has not seen thingsfor
himself (avro's oVixoi&). Since we have no first-hand experience of divine

attributesand operations,we can have no reliableknowledgeof them, but there are plausiblesuppositionsthat can be made. This interpretationmakes some sense, and we are indebted to Frtinkel for his carefulrenderingof the subtle nuancesof many of the termsin the fragment(e.g. rarLuALUTOt TUXOL)which went un-noticed in earliercommentaries,but his renderingis not without difficulties.Let us assume for the sake of argumentthat'l&?v in line I (xai To p.Ev oivv0cK44sOuTLS &vi1p 'C8Ev OV8i TLS EUTML) does mean seeing, and in particular 'seenon-metaphorical ing' - i.e. visualsense perception,13 and also thatthisjustifiesreadingE?iwS in line 2 (rb8&s&Li40OE?V T'rxvi &ovaXiyW 'rCp;L as "designating 1fvTVTWV) only a knowing rooted in vision" (p. 123).14 We must then take Xenophanes'thesis to be that 'whatis clear or precise(auwis)no man has ever perceived,nor will there ever be anyone who knows on the basis of empiricalobservationsabout the gods and about everythingelse of whichI speak'.We can understandwhy Xenophanesmighthave held thatthe gods could not be perceptuallyknown, and hence why it was not possible to have perceptualknowledge about everything of which he spoke, but why shouldhe have thoughtthat no man has ever had perceptualknowledgeof what is oaOks, especiallyif he is willingto allow thatsense perceptionis the sourceof that knowledgewhich is certainand exhaustive? The problemis that while the second line discussesknowledgespecificallyabout the gods and everythingelse of which Xenophanes speaks, the first line is unrestricted:no man has ever seen To mot's. To repairthe interpretation, and to over-rulewhat is the primafacie sense of the passage,we must find some
5

his scepticismto aaois basisfor thinkingthatXenophaneswishesto restrict province,and Frankelmaintainsthatit is the knowledgeof some particular Apart that is the implicitsubjectthroughout. non-evidentor supersensible from the suggestionthat the precedingcontext might have already supthe only argument that plied a concrete referentand content for yaw$s, in the quotationsby Frankelgives is that the "closelyresemblingpostcripts Areius Didymus and Varrogive a good indicationof the wider context: things"'(p. 128). 'but God has a genuine knowledgeeven of transcendent provideis simplya contrastof divineknowledge Butwhat these postscripts with human opinion, and as Guthrie notes, the dichotomy between transcendentand non-transcendentthings is "not in the originals"(p. 396n.). A similar difficulty for FrAnkel'sinterpretationemerges from the generalityof Xenophanes'conclusionin line 4: b0xos 8' &rrir&at rETWvxTaL: belief (or supposition)is allotted to all things.This remarkis at odds with contention that Xenophanesis concernedto deny only knowlFrAnkel's edge of the supersensibleworld, and counts as well against thinkingthat Xenophanes allows for the possibilty of knowledge based on sense experience.If Xenophanesmeant to deny only knowledgeof the supersensible, one would expect him to claim that belief is allotted to these things (and not all things), and if he did think that men could gain reliable knowledgein some manner,one would not expect him to conclude "but belief is allotted to all things".In short, although lines 2 and 3 might by themselvessuggestthat Xenophanes'scepticismwas directedonly against world,lines I and 4 indicate of the supersensible second-hand'knowledge' a generalscepticismabout the capacity of human beings to see the clear and certaintruth,and a willingnessto concede only that men can attain truebeliefs.Xenophaneswas probablynot the extremescepticthat Sotion took him to be, but he still seems more of a sceptic than Frankelwould have us believe.
III - XENOPHANES AND EARLY GREEK RELIGION

We have so far been thinking of Xenophanes as the originatorof a crypticepistemologicaltheory, and have consideredthat theory through the accountsgiven by laterwriters,both ancientand modern.It is however very unlikely,as CharlesKahn has observed,that either of these ways of philosopher,will providea complete picture: viewinga pre-Socratic sixty-centuryideas, where the The very possibilityof understanding documentation is so sadlylacking,dependsupon our fullerknowledge 6

of the older poetic outlook. It is only by placing the Milesians in between two regionsof light providedby archaicpoetryon one hand and classicalphilosophyon the other- by thusilluminating them,as it were, from above as well as from below - that we may have any hope of seeing a bit deeperinto this darkperiodof transitionand creation.16 It is true that Kahn's inquiryconcernsthe originsof Milesiancosmology, and he may not have intended these remarks to apply to other preSocratics,but thereis amplejustificationfor adoptingthis approachin our investigation of Xenophanes' scepticism. Xenophanes was after all a wanderingpoet who criticizedthe stories about the gods told by Homer and Hesiod (Al, B 11),and testifiedto the extent of Homer'sinfluenceon . . .). saVTs common opinion (B 10: 't &pXpis xO"ODrpov E'rTi rFaiLepxOmat In what ways might the poetic traditionof Homer and Hesiod have influenced Xenophanes' thinking, and especially, how might the religious outlook of the older poets link up with his views on humanknowledge? We have alreadynoted that B34 probablyembodies,or at least is connected with, a traditional poetic contrast between divine wisdom and human ignorance,but there are other featuresof Homericreligionwhich were repudiated by Xenophanes. He rejects the conception of gods in humanform(B 11, B14)and conceivesof one god, greatestamonggods and men who is unlike men both in body and mind (B23). One insufficiently appreciatedfeatureof Xenophanes'critiqueof religionis his repudiation of religious practices,and not simply religiousconceptions.He expresses scorn for the practiceof placing pine branchesaround the house in the -divine powers belief that these branchesare somehowthemselves aXXOL (B 17),and we are told by Diogenes Laertiusthat he rebuked(x(Oa&%0fkaxL) Epimenides,a man who enjoyed a reputationas a prophet and miracle Xenophanesdenouncedthe worker.17 Accordingto Aetius18 and Cicero19 practiceof divination([xvrLxij), the attemptto acquireknowledgethrough the use of omens and portents of various sorts.20 This feature of Xenophanes'writings,the attack on divination and related superstitious practices,furnishes,I believe, a basis for a coherentaccountof his general philosophicaloutlook, and his scepticism.If I am rightabout this, then we can not only make sense of some troublesome fragments,we can also appreciatethe origins of his scepticaloutlook, and the significanceof the intellectual revolution which was effected by Xenophanes, and by the pre-Socratic philosophersgenerally. Xenophanes'rejectmnof divinationis, firstof all, not unconnectedwith other aspectsof his thought. It is a reasonableinferencefrom his conception of god as unlikemortalsin raiment,voice, or body,thatthe gods do not 7

appear in mortal form, nor do they speak directly to us. Nor, since it is unfittingfor the god to be in differentplacesat differenttimes(B 14,26), is possiblefor god to 'come or go in our midst'(,ReTXpXEOaL in B26). Yet, as
FlaceliRre states "... genuine JMVTLXi, in the original sense of the word

(mania:madness)[is]causedby possession,by the literalpresenceof thegod in the soul of the prophetor prophetess,who thus receivesthe revelation direct from heaven." Frtinkelobserves that in the attack on divination, Xenophanes "made the chasm between the here and the beyond unbridgeable"(p. 130), but the chasm had already been provided for in Xenophanes'positiveaccountof the one god. Nor is the attackon divinationunconnectedwith Xenophanes'cosmology in whchcelestialphenomenaareexplainedin termsof waterand earth (B29), or perhaps simply earth (B27). Nilsson's account of the conflict betweenreligiousand philosophicalways of thinkingwhichwas a general featureof pre-Socraticphilosophy, and which led to the diminishedinfluence of seers and oracles,servesequally well as an explanationof the connection between Xenophanes'critique of divination and his cosmology: The real clash took place between that part of religion which interferedmost in practicallife and with which everyonecame into contact everyday, namely,the artof foretellingthe future,and the attemptsof natural philosophy to give physical explanations of celestial and atmospheric phenomena,or portents,and otherevents. Such explanations underminedthe belief in the art of the seers and made it superfluous.For if these phenomenawere to be explainedin a natural way, the art of the seerscame to naught.21 That Xenophanes' cosmologyhad this anti-divinational flavoris indicated by severalisolatedremarksabout traditionalportents,as well as the subjects of his cosmologicalinterest.So far as I know, only Dodds has noticed the connection:[Xenophanesgives]"naturalistic explanationsof the rainbow (Fr. 32) and St. Elmo's fire (A39), both of which are traditional portents."'22 Rainbowsare among the most strikingand suggestiveof all naturalphenomena,and have been taken as harbingersof good fortune, but for Xenophanes,"she whom they call Iris,she too is actuallya cloud, purple,and flame red, and yellow to behold" (B32). St. Elmo's fire, the freak electrical phenomenon sometimes seen on ships' masts during storms,was considereda portentof good fortune,and was thoughtof in antiquityas two brothers(Dioscuri, later Cabiri)who were the guardian saintsof mariners in distress,23 but to Xenophanes,"thosewhichsome call the Dioscuriare little clouds glimmeringin virtue of the kind of motion 8

that they have"(A39, Guthrietrans.). presents the following brief summary of the phenomena Flaceli&re whichwere taken as omens or portents: were obviouslysigns of the will of Atmosphericphenomena,meteora, the gods; especiallyof Zeus, the god of the atmosphereand the sky. The weightiestpresageof all, the one that could negate or confirmall others,was thunder.In the Iliad, wheneverZeus wishes to encourage one of the Greekor Trojanheroeshe does so by hurlinga thunderbolt to the rightof him. Rain also comes fromZeus, and was regardedas a sign of his will, a diosemeion. But beyond the clouds and all other atmosphericphenomena were the stars: in Homer, Sirius, 'Orion's Dog', was a star of ill-omen; the Spartanswould never embarkon a campaignbefore the full moon, which appearsto be the reasonwhy they did not arriveat Marathonuntil the battle was over.24 Among other celestial phenomena thought of as omens or portentswere eclipses, shooting stars, and the phases of the moon. The industrious manticseer could find significancein dreams,sneezes,volcaniceruptions, the sounds of gongs and rustling branches, the entrails of sacrificed animals,birds,and the castingof dice or bones.25 While the explanations whichXenophanesgivesare neitherdetailednoralwaysconsistent,they do focus on many phenomenainvolvedin the practiceof divination: (l)"He says that the sun and the starscome from clouds"(A32, Plut. Strom.4, cf. A33, 38, 40). (2)"Xenophanessaid there are many suns and moons accordingto regions,sections,and zones of the earth,and that at a certaintime the disc is banishedinto some sectionof the earthnot inhabitedby us, and so treadingon nothing,as it were, producesthe phenomenon of an eclipse"(A4Ia, Aetius, Placita,II, 24, 9). (3)"Eclipsesoccur by extinctionof the sun (aIE'mv-qXMov) and the sun is born anew at each of its risings"(A41, Aet., II, 24, 4). (4)"The moon disappears each month because it is extinguished" (A43, Aet. II, 25, 4). (5)"Comets(xo,ras) are groupsor motionsof burningclouds"(A44, Aet. III, 2, II). (6)"Lightnings(&arpaznis) take place when clouds shine in motion" (A45, Aet.III,3, 6). (7)"Thephenomenaof the heavenscome from the warmthof the sun as the principlecause.For when the moistureis drawnfromthe sea, the sweet water separatedby reason of its lightnessbecomes mist and passesinto clouds, and falls as rainwhen compressed,and the
9

windsscatterit" (A46, Aet. III, 4, 4). Epicurus,whose naturalisticexplanationsof these phenomenaresemble states both in content and terminology those given by Xenophanes',26 of Xenophanes' conclusion an implicit me to be to what seems explicitly account: We are bound to believe that in the sky revolutions,solstices, eclipses, risings and settings, and the like, take place without the ministrationor command,either now or in the future,of any being who at the same time enjoys perfectbliss along with immortality.27 In short,Xenophanes'cosmology,as well as his conceptionof the one god, probablyserved to support his attack on divination throughsigns. Men personifynaturalphenomenaand thinkof them as visiblesignsof the will of the gods, but they are in realityonly changesdue to the motionof clouds and the kindlingand extinctionof fires.The truedivinityexistselsewhere. These inter-connections do not, taken in isolation, show that Xenophanes'epistemologicalremarkshave anythingto do with the practice of divinationbut they do suggest that his rejectionof divinationwas not devoid of philosophicalimportance.Further,since divinationis essentiallya meansfor acquiringknowledgewe oughtat leastto considerthe possibility(though this has never,to my knowledge,been attempted)that these two aspects of Xenophanes'thought are relatedin some important way to one another.The crucialquestion is whetherB34 is itself illuminated by drawingattentionto the attackon divination,and to the significance attachedto divinationin the poetic tradition. We can begin by consideringXenophanes'reason for concludingthatno man has seen or knownwhat is oaxos (and that thereneverwill be anyone who knowsabout the gods and everythingelse of whichhe speaks):
al yap
XvLTa Ru1XXLOTTaTVXOL TETEAOLEVOV tvrrv,
OXVTOSOR1

OVX MoUb.

(EIrrWV) simply as As Frankel has argued, we need not think of TvIXOL 'chanced' (to say) or 'accidentally'(said), but ratheras 'succeeded'(in as 'especially'or j&AXLOTa (saying).Thus (takingTra saying), or 'correctly' means'succeedabove othersin 'morethan others')Ta [1AXLOTO E'rwcv TvXOL

saying'.28

"for even if someone should What then can be made of TETEXE0fLE'VoV: still he would not succeed above all othersin sayingwhat is nvuXramLEvov,
has been understood as the 'complete truth' or 'what know"? TETeXEOa14vov

is completelytrue'(Kirkand Raven, Freeman,Burnet)or simplyas 'true' (Guthrie),or 'whatis completelytrue'(Kirkand Raven,Freeman,Burnet) or simply as 'true'(Guthrie),or 'what is reallypresent'(Frankel),but the

10

literal meaning of rE'r?AFpkVoV, 'what is completed, accomplished,

broughtabout' (from ?rXw) has been largely ignored.29 So far as I can timesin Homer.30 occurstwenty-three In eighteen determine,TET-EXErLEVVOS of these passages,it is linked with speaking or saying, and the followng passages illustratethis repeated formulaicexpression,"speakingof that which has been broughtabout or will be broughtabout"(Murraytrans.): (l)"for this will I speak and verily this thing shall be broughtto pass" (T-r 6 xa; TETXEake'VOV 'anTi), II. I, 212 = II, 257; VIII, 401, VIII, 454; Od. II, 187,XVII,229, VIII, 82. (2)"He arose and spoke a threateningword, that hath now been broughtto pass"(6 &i EEXEaGRvosFCTC),II. I, 388. (3)"I will declareto thee as it verily shall be broughtto pass" (s xvi II. XXIII, 410 = XVI, 440, XXIII, 672, XIX, TETXEGRACVOVEaTL),
487.

(4)"I would that this word of thine might be fulfilled"(i'?r0S TETErXwp?vov E'l), Od. XV, 536 = XVII, 163,XIX, 309. This passagefromthe Odyssey (XVII, 153ff.) gives an indicationof the sort
of context in Homer in which someone speaks of what is TETeX?wEVOV:

Then among them spoke also the godlike (OeoLo's) Theoclymenus,

saying:

'Honored wife of Odysseus,son of Laertes,he truly has no clear understanding (oviuapa o18?v); but do thou hearkento my words,for with certain knowledge will I prophesy to thee (&TpEXE'S ytTPp aoi
and will hide naught. Be my witness Zeus above all gods, pav'rrnvaoFaL),

and this hospitableboardand the hearthof noble Odysseusto whichI come, that verily Odysseusis even now in his native land, restingor moving,learningof these evil deeds, and he is sowingthe seeds of evil for all the wooers. So plain a bird of omen did I markas I sat on the benchedship, and I declaredit to Telemachus'[Od.XV, 536]. Then wise Penelope answeredhim: 'Ah, stranger,I would that this
word of thine might be fulfilled (Q'IosTETrEXEUVOVE'ti).'

These passages, and others in Homer31provide ample justification for

reading 'TETEXeuvov rliuWv as "speaking or saying what is brought to pass".

This fitswell in the contextof line 3 of B 34, since it is obviousthatone kind of thing that one might succeed in saying is a true predictionabout future events.One can also connectthis withr&RLA&LUTa sincesome personsmight be thoughtto 'succeedabove all othersin sayingwhatcomes to pass.'32 The full message of lines 3-4 is that even if one succeeded above others in speakingof what is brought to pass, still he himself does not know, but belief or opinion is alloted to all things.
11

It is not yet clear whether B 34, in its entirety, espouses a general scepticismor simplyscepticismaboutthe claimsto knowledgof thosewho predictingevents,but thislatterscepticismis presentin succeedin correctly lines 3-4 and does serve as the basis for Xenophanes'claim in lines 1-2. Xenophanes'scepticismthen involves,at least in part,an implicitrepudiation of a central figure in Greek religion, the oracle or prophet,and a repudiationas well of the statureenjoyedby these figuresin the Homeric epic. Yet doubtsabout the infallibilityof prophetshad alreadyappearedin theme in of divinationwas a re-current Homer, and the trustworthiness So while Xenophanes'scepticismabout divinationis classicalliterature.33 revolutionaryas a repudiationof the entire enterprise,it is not wholly withoutprecedent. It shouldalso be noted that the passagequotedat lengthfromOd.XVII, 153 ff. not only links up prophecywith saying what is -rvreXeavov,but also displaysa connection,found elsewherein Homer,betweenthe gift of prophecy possessed by the mantic seer and 'sure' or 'clear' knowledge oAv, but that he himself (Theoclymenusclaims that Telemachusov oa&Ka
will give an exact or certain (&TpExtws)prophecy). The form aaxps which

appearsin line 1of B34 does not appearin Homer,but a64a knowingand aa&4speakingdo occur. To say or know in a way which is aa'a is, at the very least, to say or know what is true(cf. II. IV, 404: "Sonof Atreusutter et6?') when thou knowesthow to speak truly(oa&a lEI-v)", not lies (i but it commonlycarriesa special emphasison knowingor sayingthe full, of the returnof clear,and detailed truth(cf. Od.XVII, 106:"tell me a&4pa not what is known but thy father").On occasion it serves to characterize ratherthe mannerin which somethingis known,and designatesa knowing that is sure,certainor expert(cf. II. XV, 632: "unskilled(ovia6xx Ei&es) to olta fight a wild beast";II. XX 201 = 432: "I know well of myself(oo&4a thou II. VII, 226: "Hector,now verilyshalt XaL aVTos) how to uttertaunts"; man to man whatmannerof chieftansthere know of a surety(a&aac'CaGE) be among the Danaans").It is such certainknowledgethat is claimedby
Pulydamas (II. XII, 228 ff.: "on this wise would a soothsayer interpret, one d&1BE) of omens, and to that in his mind had clear knowledge(aa&4o OvRuj

whom the folk gave ear"), Athene, disguised as Mentes, speaks to Telemachus(Od. I, 200 if.): I will now prophesyto thee as the immortals put it in my heart, and as I think it shall be broughtto pass (TEXMEoaL), nor one versedin the signsof though I am in no wise a soothsayer(tmivnLs), as claimingto AlthoughAtheneis not portrayed birds(olwv&va&'a e18Ws). be (a&a E8W's in such matters(indeed any claim of propheticwisdom would be at odds with her disguisein purelymortalform),the factthatshe 12

claimsto be neithera prophetnorone skilledin signsof birdsindicatesthat Thus when Xenophanes such skill is typicallyclaimed by the prophet.34 assertsin B34 that no man has seen or known Tor aaoes it is quite possible thathe had in mind a sureor certainknowledgeof this sort.Since lines 3-4 concede that someone might succeed in saying what comes to pass, it is unlikelythat Xenophaneswishes to deny that men sometimesattaintruth. Since these lines serve as Xenophanes'reason (yap) for denying that men apprehend whatis aac+s,or in a mannerwhichis aa4is,35it is likelythathe intendsto deny that men ever enjoysureor certainknowledgeof the truth, even if they do sometimessucceedin sayingit (and can, as in line 4, believe or supposethat it is true). It is now possible,I believe, to see Xenophanes'remarksas a reflection of, and in partas a reactionagainst,some basic waysof thinkingthat were embeddedin archaicpoetry and religion.While adopting the traditional contrastbetween human and divine capacities,especiallythe capacity to know, he rejectedthe belief that this gulf is bridgedby the interventionof divine beings in mortal form, or that the gods somehow speak to men throughsignsor inspiredprophets.Xenophanes'repudiationof divination was probably based on his own positive theology and his de-anthropomorphizedcosmology,and is one facet of his attack on the religion of Homer and Hesiod. He denied that men who correctly predict events therebypossess knowledge,and this repudiationof knowledgeby divination seems to have led him to adopt a general scepticism about the capacityof meremortalsto attainsureor certainknowledgeaboutthe gods and everythingelse of which Xenophanesspeaks. But there s a remainingproblem.The inference is monumentallyfallacious:divinerswho claim to know the futurereallydo not know, thereever be fore no man has had certainknowledgeof the truthnor will there. anyone who has knowledgeabout the gods and everythingelse of which I speak. How could Xenophaneshave reached a general scepticalposition
simply from the failure of some men to know some things?
IV - THE GROUNDS FOR XENOPHANES' SCEPTICISM

Since aa4n'VELaV was thought to be possessed by the gods alone (as in AlcmaeonB 1),it mightbe arguedthatB34 does not rejectthe possibilityof all humanknowledge,but simplyassertsthat no man everenjoys the clear and certainknowledge,possessedby the gods, especiallythe one god who is greatestabovegods and men. If so, then the failureof seersand oraclesto gain knowledgeeven when they speak truly,could be viewed as a reason13

able basis on which to doubt that any man ever sharesin the synopticand certain knowledgeof the gods. So the argumentis not: since some men those don't know some things, no one knows anything,but rather:since to do fail gods the of who could be mostexpectedto sharein the knowledge so, then no man ever does so (even if they can acquireknowledgeby their own inquiry). This proposalhas some merit, but it falls short of being a convincing account,for reasonsthat becameclearin our discussionof earlierattempts to restrict the scope of Xenophanes' scepticism: while line I denies knowledgeof what is ax4os,line 2 denies knowledge(withoutspecification of type) of the gods and everythingelse of which he speaks;line 4 asserts only that belief or seeming is allotted to all things, and B 18 does not say that men acquireknowledgethroughtheir own seeking. It is simply unreasonableto think that a man who says no man knew or will know with and that belief or seeming is assignedto everything respectto everything, means to say implicitlythat some men do know something.The proposal Whatis usefulhoweverin this firstattemptto mitigatethe mustbe rejected. fallaciousnessof Xenophanes' inference is its recognitionof seers and oraclesas paradigmsof a sort: they, if anyone,could be expectedto share To completethe account,we mustexplain in the knowledgeof the gods.36 how, in two differentrespects,Xenophanesarguesfor a generalsceptical thesison the groundsthatsince the conditionsnecessaryfor knowledgeare they are not met even in the most promisingor favorablecircumstances, never satisfied. The first paradigmis referredto in line 3 of B34 - even if someone shouldsucceedabove othersin sayingwhat is broughtto pass,still he does not know. What must be rememberedis the ratherobvious point that the most favorableor promisingcase that could be made for the art of divination is its trackrecord,i.e. a citation of instanceswhere the predictions made by seersand oraclesturnedout to be right.This was in fact the kind of 'proof supplied on occasion by those who claim to possess prophetic powers.37 Thus, although wavrLxi is not explicitly mentioned in line 3, Xenophanesis challengingwhat is in fact the most favorablecase to be made for knowledge through divination, and claiming that even when someone succeeds in saying truly what comes to pass, he still does not know.His reasonfor this claimis not statedbut as I have alreadysuggested, it is likely tied to his own conception of the gods and his alternative naturalisticexplanation of omens and portents of various sorts. Since divination does not supply knowledge, given even the most favorable outcome, it can be reasonablyconcluded that we cannot acquireknowl14

edge by meansof its techniques. was perhapsthe most It should also be noted that while foreknowledge characteristic claim of the diviner, and successfulpredictionits strongest support, divination was by no means confined to the future. The most famous seer in Homer, Calchas,is describedas "the best of divinerswho knows things that were, and were to be, and that had been before"(II. I, 70). We do not generallyknow the sorts of questions put to the famous oracles at Delphi and elsewhere, but the leaden tablets excavated at Dodona display a variety of topics on which the oracle was consulted. These includequestionsabout the past and present,as well as the future.38 Epimenides,who received Xenophanes'rebuke,is describedby Aristotle as one "who did not practisedivinationabout the future,only about the obscurities of the past"(Rhet., 1418a 21 ff.). Thus,a repudiationof IiaVTLX1l would resultnot only in a scepticismabout the diviner'scapacityto know the future,but an equal scepticismabout their capacity to penetratethe obscuritiesof the presentand past. Butwhy would it follow from the failureof these men to know anything, The answerlies in the status thatno man has knownor will knowanything? of seersand oraclesas paradigmcases,and in the background assumptions about knowledgewhich were well establishedin the poetic traditionboth precedingand enduringafter Xenophanes'time. Partof this background has been alreadystated: 'men could have no certainknowledge,that was reservedfor God'. What needs to be added is only, as Guthrie puts it, "fa commonplace of poetry, expressed in invocastions to the muses and elsewhere,that mankindhad no sure knowledgeunless the gods chose to reveal it."39This conception of knowledge through divine revelationor to the Catalogueof inspiration is explicitin Homer(e.g. in the introduction Shipsat Il. I1,484-493:the godsknoweverythingand mortalsknownothing unless the gods choose to reveal it) and Hesiod (Theogony,26 if.), and it occursin the writingsof laterphilosophers.Parmenides presentshis way of truth as a revelationfrom "the goddess who leads the man who knows throughevery town" and Parmenides"will.learn all things"even though "thereis no truth in the beliefs of mortals"(D-K B 1). Empedocles also dismissesthe claimsof mere mortalsto have comprehendedthe truth(D-K B2),but he invokesthe muse to lead him on to the heightsof wisdom(D-K B3). Not uncharacteristically, Empedocleslinked his special insight with the attainment of semi-divinestatus(D-K B 112).In short,given the 'poetic epistemology'of Xenophanes'time, the attainmentof certain knowledge requireseitheran ascentof mortalsto the level of the gods or a descent of the gods into human affairs,and Xenophanes denies that either of these
15

everoccurs.The firstof these is madeclearby Xenophanes' refusalto think thatsomeonecould be both mortaland immortal(cf. A 13:if theyare gods, do not lament for them, if they are men, do not sacrificeto them),and the second possibility is ruled out as 'unfitting'for the true divinity. If one views certain knowledge as the prerogativeof the gods, and makes, as Frankelputs it, "the chasm between the here and the beyond unbridgeable" (ibid.,p. 130),a scepticismconcerninghuman knowledgebecomes logicallyinescapable. has B34becomesclearand coherent:mankind Given these assumptions, no certainknowledgeunless the gods impartit to us, or some men succeed in attainingthe statusof the gods. But the gods do not come amongus and they do not speak to us either in their own voices or throughsigns and oracles.Those who mightbe most thoughtto enjoy revealedknowledgeof the truthdo not do so, for even if they succeedabove othersin sayingwhat comes to pass,still they do not know, and belief is allottedto all things.So the certaintruth no man has seen nor will there ever be anyone who has knowledgeabout the gods and everythingelse of which I speak. of I have argued that a clear, consistent, and coherent interpretation Xenophanes'scepticismcan be providedby attendingto the religiousand poetic tradition in which he stood, and we can now also gain a more realisticappreciationof his achievements.His scepticismis not likely to appeal to contemporary philosophers; it rests on assumptions about knowledgeand divine revelationwhich are no longerwidelybelieved,and it is closely tied to aspects of Homer'sreligion which are now mainly of historicalinterest.Nor can he, without exaggeration,enjoy the status of being an earlyproponentof the theoriesof laterGreekscepticsor modern philosophical views of the nature and growth of scientific knowledge. There are similaritiesbetween his scepticalthesis and the conclusionsof the latersceptics,but the groundsfor his scepticismareverydifferentfrom theirs, and there is no good reason to think that he espoused Popper's 'principleof rationalknowledge'. What is noteworthyin Xenophanes'thought is his articulationof the contrastbetween belief and knowledge,and his contentionthat whatever truthis to be gainedmustcome as a resultof humaninitiativeand inquiry. to thinkthat this could result While he remainedtoo much a traditionalist in certainknowledge,for thatwas reservedfor the gods, he did believethat men could discoverwhat resembledthe truth,or whatwas at leastlikelyto be true.None of this, it seems to me, constitutesthe emergenceof a 'robust empiricism'.But Xenophanes' call for investigation,his repudiationof divination, and his related demythologized cosmology, constitute a 16

departurefrom earlierways of thinking that is justly thoughtof as revolutionary.40 University of Maryland

Exceptwhere noted to the contrary,the Greek text of the fragmentsis taken from Diels, 6th edition rev. W. Kranz, 3 vols (Berlin 1952). Die Fragmente der Vorsokratiker, Hereaftercited as D-K. Selections from Xenophanes are cited by numberand letter (A: Leben und Lehre, B: Fragmente). 2 It is quoted by Sextus Empiricus in this form on three occasions (Adv. Math. VII, 49, 1TnOL 'rvwXTtl is quoted at Pyrrh.Hyp. II, 18). Plutarchhas 110;VIII, 326, and 86xos8' E7'n in the first line, but this has been rejected in D-K, following Frankel's yivE.r'for 'LbiEv argumentin "Xenophanesstudien",Hermes60 (1925) since yEVFT' requiresa separation of TO xao?s from Ei&.s that is impossible in genuine Archaicstyle. Furthertextualsources are listed in Guthrie,A Historyof GreekPhilosophy(CambridgeU. P., 1967),Vol. 1, 395. Frankel's paper has been translated by M. R. Cosgrove and A. P. D. Mourelatosand included in the latter's The Pre-Socratics:A Collection of Critical Essays (New York, 1974),pp. 118-131. Subsequent referencesto Frankelare to this translationof his paper. 3 According to Diogenes Laertius (Lives IX, 20), "Sotion says that he was the first to maintain that all things are incognizable, but Sotn is in error"(Hicks trans.). 4 See for example, Frankel: "otois unites the notion of completeness ... with that of reliable, faithful, and unadulteratedapprehending"(ibid, p. 127). A fuller discussion appearslater in this paper (Section III). 5 D-K A24 (Arius Didymus in Stobaeus, Ecl. II, 1, 17) 's pO'6pa o rV XOELnOFV, 0ps.iv 018E 86xos8' ?'r 'ar&OLTrTvxTaL, and Varroin Augustine,De Civ.Dei 7, 17:homnisest enimhaec
6

opinari, Dei scire. D-K 24B 1: InEpi Tf;WV&'fXv

wv, -nEp! 'rCv 611TC4V o.54n'vaEMv

1V OEEoi EXovTL, (,S 8i &6pS&nots

The text is not certain. While a contrast between divine knowledge and TEex1LaCpe6axL. mortalconjectureis clear, it might be readwith equal sense, "concerningthe non-evident, the gods have a clear understanding,concerning things mortal, men merely conjecture from signs",or perhaps "concerningthings mortal,the gods have a clear understanding, concerningthings non-evident, men merely conjecturefrom signs." 7 Bruno Snell, The Discoveryof the Mind, trans. T. G. Rosenmeyer (Oxford & Boston, 1953),esp. Ch. 7 "Human Knowledge and Divine Knowledge Among the EarlyGreeks". See, for example, the prelude to the 'catalogue of ships' in the Iliad&"for you are goddesses, you are at hand and know all things, but we hear only a rumor and know nothing" (485-486); Theognis, Elegiac Poems (141-2): "we mernpractise vain things, knowing nought, while the gods accomplishall to their mind." For furtherexamples, see Guthrie, pp. 398-399. This contrast, coupled wth a conception of X6yos as human contrivance, forms the basis for Untersteiner'sview of B34 (cf. Mario Untersteiner, Senofane(Firenze, 1967),esp. pp. ccix-ccxxvi. 8 Studiesin PresocraticPhilosophy,ed. D. J. Furley and R. E. Allen (London, 1970), 152. Popper'spaper appeared originally in P.A.S. (N. S.) 59 (1958-59); a similar account is and Refutations(London, 1963). given in his Conjectures 9 To make Popper's thesis even remotely plausible, we must read rrTrXveaJVoV E?XUV in B34, as 'sayng the final or complete truth', but (as will be developed in detail later) has an ordinarysense of that which is completed, made actual, or brought TE-Te)XwaRvov

17

about, and perhapsalso, following Frgnkel,that which is real or present.In neithercase can TETEXEOFLVOVEtilrv be taken as 'sayingthat which is the complete or final explanation or theory'. 10Even so, Aristotle's view is that knowledge is gained not through the refutationof Justpriorto the discussion previousconjectures,but, so far as possible, theirconfirmation. of incontinence Aristotle states, "Here as in other cases we must set down the phainomena and begin by considering the difficulties, and so go on to vindicate if possible all the common conceptions about these states of mind, or at any rate most of them and the most important"(NE VII 1 1145b 2-6). We are indebted to G. E. L.Owen's de la methode,Louvain, 1961)for "Tithenaita Phainomena"(in Aristoteet les probWmes an apppreciationof the extent to whch phainomenameans not 'the observed facts', but
"common conceptions" (iEi'vo). 11Ibid., p. 152. Snell (ibid.) also assumes that what men discover is knowledge, and,

of Xenophanes'doctrine thoughhe seems not to realizeit, this makes his characterization dangerously close to an explicit inconsistency. He attributes the following views to Xenophanes "human knowledge is in its essence deceptive" (p. 139), "only apparent knowledge" (p. 140), "fallacious"(p. 141), and "men acquire knowledge through their own striving(p. 140),"man'sown initiative,his industryand zeal, become crucialfor the acquisitionof knowledge"(p. 140),"knowledgeconsists of the data gained from inquiry and search"(p. 140). I find his one attemptto reconcilethese two positionsexceptionally opaque "knowledgeas such is obscure, but it is illumined by searching"(p. 140). 12 See for example, Lehrer's "Why Not Scepticism?"Philosophical Forum 2 (1971), 283-98. 13 I do not think we can exclude the possibility of 'itv beng a kind of 'mental seeing'. (clear, kv 4peaLv)and what is acx4is Homer had alreadyspoken of 'mentalseeing' (W&aOOO certain, true) is more naturallythought of as propositions,accounts, stories, ratherthan the objectsof sense perception. 14 It is at least worth noting that Ea'vmL,although etymologicallyconnected with verbsof seeing, had already by the time of Homer acquireda bnroadersense in which one could consistentlysay, "I know (ol8a) even though I have not seen". The following passage is from Bk. XX of the Iliad, 203 ff. each other's parentsand lineage, for we have heard tales told in ... we know ('i8[uev) olden days by mortal men, but with sight of eyes hast thou never seen [known]my parents nor I thine (064eLb' ou'r'Lres) (Murray trans.). fur Philologie 109 Cf. also, Heitsch,"Das Wisssendes Xenophanes",RheinischesMuseum (1966) 193-235,and Iliad VI, 150; Hesiod, Theogony53-62, 915-17. 15 Frankel's later paraphraseindicates that he does not adhere to his early translation ("and what is precise no man has seen") and adopts the expansion I have suggested:'to
rwv rapL: a reliable knowledge with ovU&is yvoiLT&v rvrwr YE ff p1i?v oa4?savOp6pwos pIrpxy~L&

respect to all of the objects spoken of here, particularlyconcerning the gods, is not possible for men" (p. 128). 16 CharlesH. Kahn, Anaximander (New York, 1960), and the Originsof GreekCosmology p. 134. 17 Plato tells the story of Epimenides'visit to Athens and his prophecyabout the Persian invasion(Laws,642d-e, see also the 0. C. D., p. 33 1). Aristotlealso refersto him as a seer at Rhet. 1418a 21 ff. 18 i-Evo46vils xovi'E'Tixovpos &daLpOvaO rv PaVTLX)V,"Aetius, Placita, V, 1, I (D-K A52). 19"Of these - to mention the most ancient - Xenophanes of Colophon, while asserting

18

the existence of gods, was the only one who repudiateddivination in its entirety (divinationemfunditussustulit)",Cicero, De Divinatione,Falconer trans.(Loeb), I, iii, 5. 20The most detailed and comprehensive study of Greek divination is still Bouch& Leclercq,Histoirede la Divinationdansl'Antiquitt,3 Vols. (Paris, 1879-82).His discussion of Xenophanes (Vol. I, 33-34) treats the attack on divination as a consequence of Xenophanes' conception of god's majesty, and influential primarilyin the Sicilian commedians' (Aristoxenes, Epicharmus)attacks on les devins de carrefour.Other valuable accountsof the extraordinary techniquesemployed by professionaland amateurseersare provided by W. R. Halliday, Greek Divination (Chicago, 1913); R. Flaceli&re,Greek Oracles, D. Garman trans. (New York, 1965); and M. P. Nilsson, GreekFolk Religion (New York, 1948),esp. "Seers and Oracles",pp. 121-139.
21

Nilsson, ibid., 136. Sophocles expresses a scepticism about divination (O. T, 499-512) which closely parallels what I think is Xenophanes' thesis: the gods have perfect knowledge, but there is no sure test (xpiaLs a&Xiq&s) that a mortalseer (Rx&vTMS) attainsknowledge even if he excels above others in his skill of interpretingomens (Gooca). Cf. the summary by J. C. Kamerbeek, "The contrast between divine and human knowledge ?v -6] explains their scepticism as to the truth of Teiresias' words, based on their faith in Oedipus." (The Plays of Sophocles (Leiden, 1967), p. 120.). The summarystatement of 499-512 given above is based on the translationby RichardJebb, Sophocles, The Plays and Fragments(Amsterdam, 1966). 22@ E. R. Dodds, The Greeksand the Irrational(Berkeley, 1951),p. 196, n. 7. 23 Nilsson, ibid, p. 92, 121. 24 Flaceli6re,ibid., p. 18. 25 Cf. FlaceliUre,ibid., "Divination by Signs". Xenophanes is said to have visited Etna and commented on the periodic frequency of volcanic eruptions(Aristotle, De Mirab., 833 a 15). We are also told, though the reason is not given, that Xenophanesdisapproved of dice (A16). These fragments are however not obviously related to his attack on divination. 26 Epicurus' naturalisticexplanations of celestial phenomena consistently follow those given-byXenophanes, though he is not mentioned by name: "the risingand setting of the sun, moon, and stars may be due to kindling and quenching (vcva4iv xaxi aiEmLv)"; or it may be due to "theircoming forwardabove the earth or by its intervention";eclipse of the sun may be due to the quenching of its light (xaT& Ji3EaLv); lightningmay be due to the motion of atoms in the clouds; comets are due to fires in the heavens, etc. (Diogenes Laertius,Lives, X, 91-93, 96-98, 101, 111). While Epicurusconcedes that the facts allow for a plurality of explanations, he insists that the exclusion of myth is a necessary condition (IL6vov o ,uvHos &7iGTW) for understandingand peace of mind (103). 27 Letterto Herodotus in Diogenes Laertius,Lives, X, 76. Following Epicurus,Lucretius attacks rewigious superstition on the basis of alternative physical explanations of the motions of heavenly bodies, eclipses, lightning, clouds, rain, volcanic eruptions, the seasons, plagues, rainbows,etc. (De RerumNatura,V, VI). 28 Frinkel, ibid., p. 126. 29 At one point Frankel translatesTreTrEXUaivov Eimcv as "sayingsomething which turns out to be true", but he later discards this in favor of 'articulatingwhat is really present' (ibid., 126-127). Guthrie (ibid., 395, n. 4) characterizes TETEX&G[Livov as 'a typically Homericword' but does not indicate what it typically means in Homer. 30 Cf. R. J. Cunliffe, A Lexicon of the Homeric Dialect (1924), 337. A great many more

19

examples could be provided by broadening the criterion to include variant verbs for 'speaking'(e.g. &-yop?w- to speak publicly)and other formsof TExi. (e.g. the prediction of Calchasat IL 11,330). 31 Cf. the predictions of Melantheus(Od. XVII, 299), Antinous (Od. XVIII,82), Odysseus (Od. XIX, 547). 32 Seers and oracles are of course the paradigmcases, but the descriptioncould refer to anyone who succeeds in correctlypredictingthe future. Accordingto Diogenes Laertius (Lives, 1, 23) Xenophanes admired Thales for his ability to predict eclipses and set the solstices. B. L. van der Waerden (following an explanation given by M. Schramm) accountsfor Thales' predictionby pointingout that Thales predictedonly that an eclipse would occur in a certainyear (Herodotus, 1, 74) and, given enough backgroundinformation about precedinglunar and solar eclipses, it was possible to discoverthat in some years solar eclipses were likely to occur (Science AwakeningII: The Birth of Astronomy (New York, 1974), pp. 120-122). Xenophanes' admiration for Thales (assuming the accuracyof the story) need not be at odds with his scepticismabout divination,since he may have creditedThales with 'skill in conjecture',not knowledge. Euripidesadopts this position when he says that "the good prophet is the man skilled in conjecture"(Hel., 744-757), Platocreditsthe oraclesand prophetswith 'well-aimedconjecture'(E8okia) but insists that this is still only true opinion, not knowledge (Meno, 99c). 33 In Book II of the Odyssey,Eurymachastells the prophet Halithersesto go home and prophesyto his children,for 'many are the birdswho under the sun's rayswander;not all which of them mean anything'(181-2), "Nor do we care for any prophecy(?oumpo&ni's), you, old sir, may tell us, which will not happen, and will make you even more hated" 202). See also Hector's scorn for the &'FTEX0&vEcL 8' eTL R&XXov, (FIuOIs &xxp&c0vTov, prophecyfrom birdsgiven by Pulydamas(II. XII, 228ff.: "one birdonly is best, one omen - to fight for our country").Aristophaneswas later to ridicule the soothsayersin The Knights and The Birds. Sophocles' OedipusRex contains occasional sceptical remarks about the reliability and legitimacy of prophecy (lines 500-515) but since Teiresias' predictionis ultimatelyconfirmed, one cannot suppose that Sophocles' intention was to undermineconfidence in divination. Divination was not repudiated by either Plato or Aristotle, at least not in all its forms, and it was defended by the Stoics. The most extensive criticismamong later philosopherswas provided by the Epicureans,as can be seen in the remarksof the Epicurean Boethos in Plutarch'sdialogue, On the Pythian Oracles:".... the Sibylsand Bakishave foretoldevery sortof event and misfortune:if it so happens that a number of them have come to pass, nonetheless at the time they were uttered their prophecies were lies, even if fortuitous circumstancesshould eventually appear to make them true."Quoted in Flaceliere,ibid., p. 81. 34 So characteristic in fact that it becomes naturalto speak of the prophetsthemselvesas acxvis - sure or unerring(cf, Liddell and Scott: aaoAs; for example, the descriptionof Teiresiasin Sophocles' Oed. Rex, 286: a0w-r'TaTr). 35 The syntax of aa4es (direct object of 'MEv, accusative of respect, or adverbial accusative) is unclear, but I do not see that a reasonable interpretationof the fragment presupposesa definitive answer. What is oevos may be what is not known, or it may be the respectin which one does not know, or the mannerin which one does not know. The in lines 1-2are set in clear contrast(kv- -e) and Ei'&s ';&Ev importantpoint is that aot4os, oiv with 86xosin line 4. Denniston cites this fragmentas an example of an ovv (in To ,uEV ooois) emphasizing a prospective p.?v (The Greek Particles (Oxford at the Clarendon Press, 1954), p. 473). This contrast makes it unlikely that To Gacxi' functions as an

20

independent sentence as, for example in Cleve's: "and this is sure by all means" (The Giantsof Pre- SophisticPhilosophy,p. 28). 36 The great importanceof seers and oracles in both public and privateaffairs is perhaps sufficiently well known to need no furtherargument. One need only point to the great popularityof the traditionaloracles at Delphi, Dodona, and Claros(near Colophon), and accordingto Herodotus,the incessantuse made of seers in militarymattersas the clearest evidence of the exalted position whichhheyenjoyed. The majorrole played by divinersin Greek religion is explained in detail by Nilsson, ibid., 123-139. 37 In Book 11 of the Odyssey,Halitersespredictsthe fate that is to befall Penelope'ssuitors, and then argues: "I who foretell this am not untried, I know what I am saying. Concerninghim, I say that everything was accomplished in the way I said it would be at the time the Argives took ship for Ilion, and with them went resourcefulOdysseus. I said that after much suffering, with all his companions lost, in the twentiethyear, not recognized by any, he would come home. And now all this is being accomplished(vvvrrTavr TEXrt'Tr)." (170-176, Lattimoretrans.). A similar defense is offered by Euthyphro, the self-proclaimed theological expert in Plato'sEuthyphro (3c), when he complains of his receptionin the assembly:"When I tell them in advance what will occur they laugh at me, and yet I have never made a prediction that did not come true." 38 Included among those questions recordedwere these: whether a man's wife will bear him a child, whetherthe child which his wife is carryingis actuallyhis, whethera man will do well by breedingsheep, and, my favorite,"AgisasksZeus Naios and Dione whetherhe lost the blanketsand pillows himself or whetherthey were stolen by someone outside the household." For other examples see C. Caraponos,Dodone et ses ruines(Paris, 1878)68 ff., and the Bulletinde Corr.Hell., 83 (1959) 669-73. 39 Ibid, p. 398. It has been thought that Xenophanes rejected the view that mortals derive theirknowledge from the gods, and held instead that men gain knowledgethrough their own inquiry (Snell, ibid., 139-144;John Robinson, An Inlroductionto Early Greek Philosophy New York, 1968, pp. 55-56). But this is not implied by the fragments. Xenophanesdenies that the gods revealedall thingsto mortalsfrom the beginningand he repudiatesdivination, but he nowhere rejectsthe assumption that if knowledgeis to be attained at all by mortals, it must come by divine revelation. What man can discover throughseeking is 'the better'(B 18)which can easily be the 86xosof B34, and which may resemble or be similar to what is true (B35). One relevant fragment on this issue is the
81 0VTrTo;L tantalizingly brief B36: 6.6rr6oaa
T4E1ivAOLV

ErWop&aOaL,"as many as they have

revealed to mortals to look upon." But it does not say anything about knowledge.The majorobstacle in the path of thinking that Xenophanes espouses knowledge gained ugh inqury is one of consistency. Like Snell, Robinson seems untroubled by atributing to Xenophanes the following: "only through patient inquiry does the truth come to be known" and "the truthitself is known only to god" (ibid., p. 56). 40 I am indebted to William Fortenbaugh, David Glidden, G. B. Kerferd, David Konstan, MarthaNussbaum, and Gregory Vlastos for their criticismof earlier drafts of this paper. I would also like to thank A. P. D. Mourelatos,Carl Brumbach,William Sewell, and Ronald Swigger for their assistance during the early stages of my researchon this topic.

21