An Oracle of Apollo at Miletus Drew-Bear, Thomas;Lebek, W D Greek, Roman and Byzantine Studies; Spring 1973; 14, 1; ProQuest pg. 65

An Oracle of Apollo at Miletus 
Thomas Drew-Bear and W. D. Lebek 

}{ 

HONORIFIC INSCRIPTION at Miletus, displaying letter-forms of 
the late second or the third century, praises a lady as UpE[LO:V] 

S£a {3lov TfjC 7Tp[O 7TOAE]WC lIoAuxSo[c 'A]07]vac alpEO[ELcav] Vrro 
TOV at/s~[vSE]CTChov OEOV LJ£9[VJLEW]C 'A7TOMwvoc €v[2] CWCPPOCVV'[J Kat 
[ ] 

, ., A O[ n • A \' ,. 0 A [ tI] CEJLV OT7]'rt V7TO TWV E£JWV aVTOV I\OY£WV Ka£ EVKWJL£ac EtcaV 0 v TWC' 

after which there follows the text of the oracle, in hexameters. The 
end of the inscription informs us about the persons who arranged for 
• • liSA A' '\ tE' 'liS' t 'E ' , Its erectIon: J.f.LapKO£ VP7]I\£O£ PJLE£ac Ka£ J.f.L£VV£WV Ot PJLE£OV T7]V 
\' , yl\vKVTaT7]V JL7]TEpa. 
This document, found in 1900 or 1901, was recently published by 

P. Herrmannl with interesting observations concerning the proso­
pography of Miletus, and was subsequently discussed by R. Merkel­
bach,2 who advanced the interpretation of several passages; but a few 
difficulties and points of interest perhaps deserve further study. The 
text of the oracle is given by Herrmann as follows (we have disposed 
it here by lines, not as it is laid out on the stone): 

'0.', ' , '" , r' 'IT ' ",E p..EV, W vaETa£, <:,aKOpov 7TEp£ .n..paTOYEVE£7]C 
6JLcpfjc JLaVT£7T6,\o£O OEOKAVTEOVTEC iKEC8E, 

~ KOpVcp1]V < P >~gaca 7TEPtcWCPpovoc [ .. ] YEVETfjpOC 
, , \ [A ] "Q' , 4 E£V07TI\£OV CK£<P>T wc a 7Tap auavaTOtc£ XOPEVCEV, 

£VOEV 7TaVToI[ . ... €CP]E7TEW A&XEV aKpo7ToA7]aC, 

€COA~{V} JLEV 7TO[AlEC]C£ {307]Oooc, €V SE TEXVatcW 
• '\ \[]A' , , , EV7Tal\aJLoc K~ E£TO tcW E7TtcTanc EPY07ToVOtcW' 

• .I.. [A 't 't] A , • A'''~ , 8 aVTo",av WV. YEVE. 7Jc yap EXP7]V £EP7] LOa TE£JL7JV 

07]AVTEp7]V UgacOa£ acp' aiJLaTOc EVyEVE~PWV' 
, \ \' .,,. <\ A '.1..0 "A.I.. ' al\l\ E7T£ aKav EI\EW yEpawv '" aCEV ",POYEVE£7]C 

LJII[. ]IA SE Kwptc 7TOT~ 7TapOEVov 'ATPVTWV7]V' 
< "Q\' '{'}"'" " 12 <7] > JLEV yap ual\aJL0£O Ka£ OVK ECT aJLV7]Toc EPWTOC, 

• ~\, '0 \ ,. \'~ < , 7J OE yaJL0tc£ TE 7]I\E Ka£ EVKEl\aootc VJLEVa£oK. 
Tolyap 7TE£8oJLEVO£ Molpa£c Kat IIaAA&S, CEp..~V 
't' A\ 0' D < .I.. ' " ..,aTopVE£l\av ECUE £EP7]",OPOV ap7]TE£pav. 

1 Chiron 1 (1971) 291-98 and pl.u-m [hereafter cited as HERRMANN). The stone was copied 
and squeezed but subsequently smashed and some of the fragments are lost. 

I ZPE 8 (1971) 93-95 [hereafter. MERKELBACH). 

65 



66 AN ORACLE OF APOLLO AT MILETUS 

One of the main themes of this oracle is set forth already by the 
first word, 6tfo€. The theme 6tfo~ ... rKEcfh: occurs also in a well-known3 

oracle of Apollo at Delphi preserved by Pausanias 9.37.4: when King 
Erginos as an old man came to Delphi ~PWP"lIctJ 7TEpt TT(xt8wlI xpfj. T&8E 
~ llveta' 

'EpyLIIE IDvp.€lIo,O 7T&' llPECf3wvux8ao 
01/ ~A(JEC yEVE~V 8'~1JP.EIIOC KTA. 

In line 3 Herrmann established the correct reading <!»~gaca,4 and 
Merkelbach5 (p.94) restored the meter by writing {7T}~p'CWcpPOIlOC,6 
but there remains a lacuna of two letters before the word YEIIETfjpOC. 
Herrmann commented (p.292): "Vor yEIIETfjpOC ein runder Buchstabe, 
danach einer mit senkrechtem Strich: OJ? in der Schede von Fredrich,7 
~K auf dem Blatt mit der Umschrift."8 Merkelbach (p.94) restored 
here [EV]YEIIET'fjpOC, explaining: "Der Gedanke ist: Wie Pallas Athena 
aus dem Haupt eines edlen Vaters geboren ist, so solI auch ihre 
Priesterin von edlen Vorfahren abstammen (Vers 9)." It is true that 
EVyEIIET~PWV used of the priestess' family in line 9 reflects YEIIETfjpOC 
used of Athena's father in line 3, but it is not likely that in this oracle, 
which displays much care in the selection of unusual vocabulary, 
such a rare word should be employed twice. Moreover, in the measure 
that this word is appropriately applied to <noble' human families it is 
quite inadequate as an epithet of Zeus. We therefore suggest the 

3 This oracle is eVidently cited by Plut. An seni 784B, on a man who begins to deal with 
politics in old age: tcwc civ alTtWfLlvcp T£V~ '1Tapa.cxo£ Tel rijc IIv8lac fil'1Tfiiv "&.p' -?.\8fiC" aPrlv Kal 
8T]fLaywylav 8£~~fLfiVOC. Editors of Plutarch compare Corpus Paroemiographorum Graecorum I 
p.444: o.p' ~.\8fiC, aM' EC TelV Ko.\wvelv 'fiCO. 

'He commented (p.292) that the stone has "nach KOpV.p~V vor -T]gaca ein kleines 0: da 
fhlgaca nicht moglich ist, vermute ich Verschreibung fur p~eaca, wobei aber auf jeden Fall 
die Partizipialform nicht in das Metrum passt." 

6 This correction was also made independently by B. Snell (cf Merkelbach 95). 
6 The mason's error is due not merely to careless copying of his original (as earlier in this 

line or in line 6), for it is also a substitution according to sense (as in line 12, where W. 
Buhler remarked to Herrmann on the erroneously inserted word OUK that "sein Eindringen 
mag mit einem Versehen im Hinblick auf die Negation, die in afLV1JTOC ja ohnedies schon 
enthalten ist, zu erklaren sein"): in place of the new word Ep£C6J.ppovoc required by the 
meter, the mason substituted the equally rare '1TfiP£C6J.ppovoc, which has the same meaning 
but a different metrical value (influenced perhaps by the existence of the word '1TfiP£CCO.ppwv). 
Also possible metrically, but less appropriate to the style of this oracle with its many new 
words, would have been '1Tfipl.ppovoc (used in Homer only of women). 

7 This copy had been made before the stone was broken (cf Herrmann 291). 
8 Among the material at Herrmann's disposition was "ein unter den milesischen Papieren 

erhaltener Versuch der Textwiedergabe in Umschrift von einer mir unbekannten Hand." 



THOMAS DREW-BEAR AND W. D. LEBEK 67 

restoration [00] yEvETfjpOC. The combination of YEVET77P immediately 
after the possessive pronoun is found fairly often, for instance (with 
the same word-order: adjective, possessive pronoun, substantive) in 
Nonn. Dion. 4.303 (of Cadmus): 

, \ \ ' 'f} T: I ~ ... , I aN\a 7TO ov VptoLO T€OV y€VeT'I}poc Eaccac 

The same sequence is probably found again at 30.178: 

\ , \ ~ '\ r I \ I 
V€KPOV E/WV YEVET'T'Jpa 7Tal\LV o"WOV7'a TEI\ECCW 

In these phrases the possessive pronoun does not necessarily lay stress 
on the kinship expressed, though this may be the case here: at her 
birth the mighty goddess broke open the head of her own father. 

The dance of the newborn goddess in line 4 is an allusion to one of 
the current explanations of her epithet IIaAAcXc. Because Athena was 
born from the KOpVCP~ of Zeus she obtained as her lot the aKpo7T6A'I}ac.9 

The thought expressed in such etymological allusions may be illus­
trated by passages in the discussion of Athena by the Stoic Cornutus, 
De N at.Deor. 20: YEV€COat OE Kat €K Tfjc TOU LJ tOC KEcpaAfjc A€YETaL • . • €7T€t 

... \ '() I \ , I I ... I t.J.. \ I , (187) TOU !-'EV av PW7TOU TO aVWTaTOV !-,EpOC TOU cw!-'aTOC "I K6t,al\'I} ECTL • • • 

\ ') , \" '''"'' 0' , I " Kat 7TOI\EWC yap Kat OtKOV Kat TOV fJLOV 7TaVTOC 7TpOCTaTLV 7TOt'l}TEOV TYJV 

rPp6V'T'JCLV (cf lines 6-7)10 ... IIaAAac OE MYETat Ota T~V !-'E!-'VOEV!-'€V'T'JV 
, " , , , \ \, , \'''~ 

7T€pt aVT7JV VEOT7JTa • • • CKtpT7JTLKOV yap Kat 7TaN\O!-'EVOV TO VEOV. wpVVTa£ 

OE aVT~V €V Ta'ic aKp07T6AEC£ jLcXAtCTa. The enumeration of the OVVcXjLEtC of 
the goddess in the long'relative clause which begins with 7] in line 3 

and extends over five lines is a traditional stylistic element in such 
contexts,u to which is added here an acquaintance with philosophical 
literature perhaps indicated also by the epithet KpaToYEvE{7J given to 
Athena at the end of line 1: this is a new word, but KpaToYEv~c,12 also 

9 See Merkelbach 94; in line 5 he suggested 1Tavrol[wv] rather than 1TavTol[acJ. Here the 
restoration [St]'!1T€W is a possible alternative. Sanctuaries of Athena were usually located on 
an acropolis: cf. D. Knoepfler, BCH 96 (1972) 287. The use of the verb '\ayxavw to designate 
the functions 'allotted' to a divinity is traditional: see K. Keyssner. Gottesvorstellung und 
Lebensauffassung im griech. Hymnus (Stuttgart 1932) 64-65, with the example cited from Pro­
clus 7.21 concerning Athena: ~ '\ax€c aKpo1T(5'\TJa KaO' vl{lt'\oq,OtO KO'\WVT}C. For the thought cf 
Aelius Aristides' discourse on Athena (2. 10. 17ff) : aTE': Si lv KoprxPii T€ TOV 'O,\VP.1TOV Kal lK 
KOPVcpijC TOV ~t(lc y€voP..!VT} 1TO,\€WV .•. 1Tacwv Tac Kopvcpac EX€t KaTa KpaToc dJc &,\TJ8wc 
VPTJKv'ia • •• TTJpovca TO cVp.{30'\ov TfjC a&ijc y€v.!awc (cf also 12.3-4). 

10 Cf. ibid. 188: Tove S' ao,\ovc €VP€"iv p.Ev MY€Tat, KaOa1T€p Kal Tllia lv Ta'ie T.!xvatC y'\acpvpa· 
acp' oJ Kal E1TteTane (cf line 7) Tije TaAaewvpy{ae lCTt. 

11 See the examples discussed by E. Norden, Agnostos Theos (LeipzigJBerlin 1913) 169fT. 
11 The Thesaurus s.v. KpaToy€V1}C has also a cross-reference to the word Kopvcpay€V1}e, which 

occurs in Pluto De Is. et Os. 381E, a discussion of the Pythagorean name for the equilateral 



68 AN ORACLE OF APOLLO AT MILETUS 

a hapax legomenon, occurs in Porphyry's work of allegorical interpreta­
tion De Antr.Nymph. 32, where the word was clearly invented to 
explain Od. 13.346.13 

In line 7 the restoration K~[Ewo]iCLV is a possible alternative.14 The 
first editor's supplement in the following linelli was replaced by 
Merkelbach with the convincing restoration aVrocpav[ oik KOVP ]77c: the 
verb governing [EP77tSa TELf.L~V is Sleac8aL in the following line, and the 
subject of Sleac8aL is clearly 877Av-rlP77, here used as a noun16 as in an 
anonymous epigram on Agamemnon, Anth.Pal. 9.495: 

Line 10, however, is still not in order. Neither syntax nor sense is 
clear, but Herrmann offers no translation or comment. To elucidate 
this line we must first examine its context. Now the three following 
hexameters1 ? describe an opposition said to exist between Athena and 
Aphrodite, and this opposition is based on the two goddesses' different 
attitudes toward love and marriage: Athena has nothing to do with 
such activities, which on the contrary form the central interest of 

triangle. For KPcrr0'Ywrlc beside (now) KpaTO')'£VtdT}, if. #fJO')'wrlc beside at/Jpo'YlVEta and 
KV1TPO'Ywrlc beside KV1TpoylvEta. 

13 The verse reads a?n-/zp l'/l'~ KpaTdc ,up.lvoc TavU.pv>.Aoc lAatT}. and Porphyry comments 
'AOT}viic p.& 'Yap Td t/JVT6v, t/Jp6J1T}Ctc ~E 'r7 'AOT}vii. KpaTO')'EJIOVc ~' oVC1Jc rijc OEOV, OlKE'ioV T6'/1'oV 0 
6 ' "... ,,\ , .... \_, ,,\ 6' • EoAoyoc E'EVP£V E'/I" KpaTOC TOV IU.P.EJIOC aVTT]v Ka 'Epwcac. 

U The adjective KAEtTCk (which is the Homeric form) is not elsewhere attested in the 
inscriptions of Didyma. 

15 To support his restoration Herrmann cited (p.293) an honorific inscription referring to 
another oracle of Apollo (I. Didyma 243 line 7): ylvovc a?n-04>art: wv '/I'fJO')'ov]wv WC 0 Lh~vP.Etk 
lp.«prVPT}c[£v]. and explained (p.294) that a?n-ot/Jav[wv] is perhaps equivalent in meaning to 
a?n-oxOwv. The parallel cited is itself a restoration, however, and therefore of no probative 
value, and on the other hand there exists no evidence to support the proposed explanation 
of a?n-ot/Javt7c: on the contrary, the word is used elsewhere not of men but of divinities (cf 
L. Robert, eRA! 1971 p.611 n.1; Merkelbach comments that "Hier bezeichnet es die 
Gouin, die aus eigener Kraft in Epiphanie erschien"). 

11 This usage, not registered in LSJ or its Supplement, is mentioned in the Thesaurus (with 
the example from the Anth.Pal.). 

17 On line 11 Herrmann remarked (p.Z9Z) "sollte ~"1P'~ gemeint sein, statt des zerdehnten 
3"1p,,~q.?" Although the latter form is apparently not elsewhere attested (Herrmann cites 
no instances), examples of analogous lengthening from the poetry of Imperial times are 
cited by K. Meister, Die homerische Kunstsprache (Leipzig 19Z1) 68".1; nevertheless the in­
terpretation offered by Merkelbach (p.95): ~[P],~ "mit Dehnung des ," is preferable, be­
cause it avoids assuming an error by the mason, who inscribed one alpha and not two 
(lengthening of the iota might have been facilitated by the long iota of the future and aorist 
in forms such as ~T}p,c6p.EOa and l&r]p'c£v). 



THOMAS DREW-BEAR AND W. D. LEBEK 69 

Aphrodite.ls It seems inevitable that three lines of this oracle were 
devoted to underlining this opposition only because it was in some 
way relevant to the conditions that governed the choice of a priestess 
of Ath~na.19 

Merkelbach (p.95) made this suggestion: "lch schreibe cP8aCE ~ 
(=rpOac' ~) statt rpOacEv. Der Vers ist schwierig ... Ich nehme als Sinn 
an: Man soIl eine weibliche Person aus gutem Haus erwahlen, jedoch 
(&,u') ausserdem (i1T~) eine solche, welche (m ihren Ameil an den 
Caben der Aphrodite (atcav YEp&WV 'ACPpOYEVEtT}C) schon (cpB&CE) 
erhalten hat (i..\E LV) , ij( TLc) lcP8acE atcav i..\ELV YEp&WV • Acppo8tT'Y/C. Ein 
Relativum ist jedenfalls n6tig urn einen symaktischen Zusammen­
hang herzustellen." But Merkelbach himself drew attention to certain 
weaknesses: "Die Kiirzung des langen Vokals (~) im Hiat ist zwar 
korrekt, aber doch ziemlich hart. Die Wortstellung ist kiinstlich 
verschrankt." Such objections do in our opinion render this suggestion 
unappealing.20 

18 The locus classicus for the antagonism between Athena and Aphrodite is of course the 
description of the battle between the gods in II. 21.420ff; more relevant here is the beginning 
of the first Homeric hymn to Aphrodite, lines 9ff: 

ou yap ot ~va3~v €pya 1TOAVXpvcov 'AtjJp03lTT}c, 
&X\' apa OL1TOAEfLol TE aaov Kat €pyov ~ Ap1Joc, 
vC/Lival T~ fLaxat T~, Kai &yAaa €py' aAEY&VEtV. 
JIPWT1J TIKTovac av3pac t1TtX8ovlovc J3t3agE ..• 

Artemis is described as ya/Lwv afLVqToc in Opp. eyn. 1.34, where she states her aversion to 
the a8&p/LaTa JIoVToy~vd1JC. 

19 Herrmann (p.293) attempted a different explanation of the "Frage nach der Bedeutung 
der Gegeniiberstellung von Athena und Aphrodite in den spateren Versen, wo auf einen 
schon alteren Anspruch Athenas hingewiesen zu sein scheint. Wenn das eine Anspielung 
auf lokale Gegebenheiten ist, konnte man es so deuten, dass durch die vom Orakel 
geforderte Neubesetzung des Priestertums der Athena Polias ein in Milet bestehender 
alter Kult der Athena wiederbelebt und gegenliber dem jlingeren Aphrodite-Kult, der 
vielleicht grossere Bedeutung erlangt hatte, aufgewertet werden sollte." But there is no 
indication in this text of an "alteren Anspruch Athenas," and in general this oracle concerns 
the personal affairs of Satorneila rather than questions of public policy. The same editor 
offered also an alternative suggestion (p.294), that Satorneila "bereits Priesterin der Aphro­
dite war und nun zusatzlich auch das Priester tum fUr Athena Polias iibernehmen sollte"; 
but he remarks himself that "freilich wird auf ein solches doppeltes Priesteramt in dem 
erhaltenen Teil der Ehreninschrift nicht hingewiesen," and in any case the spectacle of two 
goddesses squabbling over the services of a priestess was surely not intended here. 

20 To support his restoration with its artificial word order Merkelbach cited Callim. 
fr.43.5D-53 (Pfeiffer): 

otaa A~OVTlvovc - - - - - - - - - - - - -­
Kai MEyap~ic £TEPOt TOUC a1Tlvaccav JKEt: 

N"awt MEyapijEc, €xW 5' Evf30tav €VtC1TEiv 
4>Dt.aTo Ka, KECTOV 3EcmSnc ~v ~ EpvKa. 



70 AN ORACLE OF APOLLO AT MILETUS 

The sense thus expressed, however, is just what is required in this 
line, for it is stated here that the priestess of Athena must assume the 
obligation of sexual abstinence.21 On the other hand this obligation 
need not seem too harsh, because she has already enjoyed her full 
share of love. In order to express this meaning it is not necessary to 

alter what was inscribed upon the stone, for another solution makes it 
possible to obtain satisfactory meter and word order. Only the stop 
inserted by the first editor at the end of line 9 must be removed, and 
the second word in line 10 must be understood as a variant orthog­
raphy, due to iotacism, of ~7TEl: "it was necessary that a woman with 
the blood of noble ancestors should receive the priestly honor, but 
after she had already obtained her full share of the gifts of 
Aphrodite." 

The mention of her two sons at the end of this inscription makes it 
clear that Satorneila22 had been married; and if she is assumed to be 
of a certain age at the time of her selection as priestess, it could fairly 
be said that she had already received her share of love.23 Did Apollo 
in this oracle require her to leave her husband? A different solution 
is preferable, for the true situation may be deduced from the phrase 
7TE,(J6p.EVO' Molpatc in line 14. This is an oracle of Apollo and it concerns 
a priestess of Athena: why then does it state that the citizens of 
Miletus must obey the Mo'ipa,? The injunction is no mere rhetorical 
embellishment, but rather an allusion to the most important factor 
in Satorneila's personal situation. The Fates were not beneficent 
deities but rather the bringers of inescapable doom, and that is why 

But this is not an apposite parallel, for the basic structure of these lines is dear: o18cx or 
lxw Jv"'T£iv with an object in the accusative. Understanding of the unusual word order is 
therefore not really as difficult as it would be in line 10 here; furthermore Callimachus 
evidently desired to avoid monotonous enumeration, which is not a problem in the 
orade. 

11 Note in this context the adjective C£JLvrlV at the end of line 14, which is applied to 

Icx'T0pv£'O..cxv at the beginning of the following line rather than (as would also have been 
possible) to the preceding word naMaS,; the section in prose at the beginning of the in­
scription states that Satorneila was alp£8[£iccxv] tTT[t] cw4>pocJvn KCXt [c£JLv]On]'T'. 

IS On the name (the Latin Saturnina) cf Herrmann Z95 n.ll. 
la A similar meaning would result from modification of Herrmann's word divisions in 

line 10: aA.\' ITT, alc' avt:'\t:iv 'Y'pcx <Lv .p8acw • A.ppoywt:lT/c, "but it is her fate (ITT, for lTT£c'T') to 
renounce the gifts of Aphrodite which she had enjoyed previously" or "to which (by her 
appointment) she had become superior"; but .p8av£,v 'T'VOC llas no real parallel in either of 
these senses, and this oracle has no hint elsewhere that Athena is in any way superior to 
Aphrodite. 



THOMAS DREW-BEAR AND W. D. LEBEK 71 

they are mentioned here: Satorneila is a widow.24 Only after the death 
of her husband did the Milesians consult Apollo about the choice of a 
priestess; thus the selection of this lady, with these special obliga­
tions,2s may now be seen as peculiarly appropriate. 

It is apparent that the difficulties encountered by previous editors 
in interpreting this oracle derive from the fact that it is constructed 
largely of delicate allusions to particular circumstances in the honor­
and's private life. A literal translation will make clear how we under­
stand this text: 

"Late, 0 townsmen, concerning a priestess of Athena 
have you corne to hear the divine inspired voice­
who split the head of her very wise father 

4 and leaping among the Immortals danced a martial rhythm 
whence she obtained as her lot to preside over ... citadels, 
a noble helper of cities, and in the arts 
a skilled instructor for famed craftsmen: 

8 for it was necessary that the honor of the priesthood of the 
self-appearing Maiden 

be received by a woman with the blood of noble ancestors, 
but after she had previously obtained her share of the gifts 

of Aphrodite, 
for the Cyprian goddess vies with virgin Athena, 

12 since the one is uninitiated in love and the bride-chamber 
but the other rejoices in marriages and melodious bridal 

songs. 
Accordingly in obedience to the Fates and to Pallas 
appoint chaste Satorneila as holy priestess." 

The carefully chosen vocabulary and elegant structure of this poem 
display an equal elaboration. In his dialogue De Pyth.Or. 396c Plutarch 

II On the employment of older widows as priestesses in cults which required chastity 
see E. Fehrle, Die kultische Keuschheit im Altertum (Religionsgeschichtliche Versuche und 
Vorarbeiten VI, Giessen 1910) 95, with the examples there cited. 

2~ Perhaps these obligations may have some connection with the reproach uttered at the 
beginning of the oracle, &pE tKEdh, because if applied regularly they might have rendered 
this priesthood less attractive and hence difficult to fill (note that &pI in line 1 is explained 
by y&p in line 8); but it is also possible that this was not an established custom (for then the 
explanation in lines 11-13 might have been unnecessary), but rather a special ordinance 
imposed here because it was appropriate for Satorneila. Herrmann (p.293) poses the ques­
tion "ob dahinter eine Hingere Vakanz eines schon bestehenden Priestertums zu sehen ist 
oder ob es iiberhaupt um die Neueinrichtung eines nach Aussage des Orakels schon lange 
faIligen Kultes geht." 



72 AN ORACLE OF APOLLO AT MILETUS 

k \ \ ' ",I.,.., () , ........ A ,. f" \ remar s 7TON\(UCLC E'f"f aV/LacaL TWV E7TWV 0 LOyEVLavOC, EV OtC Ot XP1]C/LOt 

>.iyov-raL, ti]v cpaVA6rt}Ta Kat ti]v E~TlAELav.26 To such complaints this 
oracle of Didyma presents a striking contrast. Repetition of words is 
carefully avoided (note that no goddess is mentioned twice by the 
same name). In the antithesis of lines 12 and 13 are four different 
expressions for sexual relations, and the central idea of the priesthood 
is expressed by 'aK6pov in line 1, lEp1]t8a TEL/L~V in line 8, and lEP1]CP6pOV 
&:p~TELpav in line 15. The poem abounds in Homeric words and forms, 
and its tendency toward elevated language is responsible for the 
numerous words which occur here for the first time: KpaToYEVEt1]C, 

£PLCWCPpovoc, Elv67TAtov beside the well-attested £V67TALOC, lEp1]t8a used 
as an adjective,27 EryEVET~PWV, which takes its place next to the already 
known EryEVlT1]c and EryEvlTELpa, and finally lEP1]CP6pOV, which is not 
cited by the dictionaries beside lEpOCP6pOC and lEpiicp6poc :28 all in a 
poem of fifteen lines. 

Such effective use of words, combined with the knowledge demon­
strated of Stoic theology, betray a poet of considerable erudition and 
skill who has also given his work a very sophisticated structure. We 
have already noted the three different expressions for <priestess' 
which occur in the first and last verses and again in the middle of the 
poem, emphasizing what is after all the central theme of this oracle; 
nor is this the only tectonic device. The first two lines consist of an 
address to the inhabitants of Miletus, revealing the topic concerning 
which they have decided to consult Apollo,29 and the last two hexa­
meters address the Milesians once more in order to give them the 
god's response. Enclosed between this beginning and end is a des­
cription of the birth, power and functions of Athena, whose cult forms 
the basis of the whole matter, to which there corresponds the elegant 
depiction of the rivalry between Athena and Aphrodite. These two 

211 In Lucian, lupp.Trag. 6, Hermes, told by Zeus to make a proclamation in verse, replies 
that he is afraid of being laughed at like Apollo for his metrical mistakes (cf. Pluto MOT. 
396D: TOQC 13~ 'lToAAoQC TWV Xf'T/cpiiiv opwp.VJ Ka~ Tote P.lTPOK Ka~ Tote ovOp.ac, 'lTATJp.p.€A€lac Ka~ 
t/Jav>'orrrroc &.va'ITf'ITATJcp.lvovc). 

17 Although this usage of the word is not attested elsewhere, the formula in which it 
occurs (always at the end of a line) is characteristic of Didyma: see L. Robert, CRAl 1968 
pp.580-81, and Herrmann 293 n.4. 

28 LSJ cites no example of any of these three forms used as an adjective: but see J. and L. 
Robert, BullEpigr (REG 63) 1950 p.l71 no.134 (now IG X.l.l 58). 

29 Note the emphatic position of the words o.p~ ... rK€c8f forming a strong hyperbaton 
at the beginning and end of this two-line section; the words within this frame supply the 
details. 



THOMAS DREW-BEAR AND W. D. LEBEK 73 

sections concerning the goddess surround the statement of the 
qualifications necessary for the new priestess. The whole oracle is thus 
arranged into five nearly symmetrical and mutually interlocking 
sections. Not only does it honor Satorneila, it also has considerable 
value as a poem; her sons did well to have it cut on stone. 

THE CENTER FOR HELLENIC STUDIES 

December, 1972