Alcman's "Partheneion" I Reconsidered Rosenmeyer, Thomas G Greek, Roman and Byzantine Studies; Winter 1966; 7, 4; ProQuest pg. 321

Aleman's Partheneion I Reconsidered 
Thonlas G. Rosenmeyer 

I 

VERY SOON after the publication in 1863 of Aleman's Louvre 
Partheneion1 it became apparent to the scholars who tried 
to interpret it that there were elements in the composition of 

the poem which suggested that it might have been performed by a 
group divided into two half-choruses. This was the view expressed, 

1 Hereafter referred to as Partheneion I. The following abbreviated references will be 
used: AHRENS: H. L. Ahrens, 'Das alkmanische partheneion des papyrus," Philologus 27 (1868) 
241-85; 577-629. BOWRA 1936: C. M. Bowra, Greek Lyric Poetry from Aleman to Simonides 
(Oxford 1936). BOWRA 1961: C. M. Bowra, Greek Lyric Poetry from Aleman to Simomdes s 
(Oxford 1961). BOWRA 1964: C. M. Bowra, Pindar (Oxford 1964). BURNETT: A. P. Burnett, 
"The Race with the Pleiades" CP 59 (1964) 30-4. DAVISON: J. A. Davison, "Aleman's 
Partheneion," Hermes 73 (1938) 44D-58. DIELS: H. Diels, "Alkmans Partheneion," Hermes 31 
(1896) 339-74. n'ERRIcO: A. d'Errico, "n Partenio di Alemane," Annali FacLettFilos, Uni­
versitd di Napoli 7 (1957) 5-38. FARINA: A. Farina, Studi sui Partenio di Alemane (Napoli 1950). 
GARZYA: A. Garzya ed., Alemane, I Frammenti (Napoli 1954). VAN GRONINGEN: B. A. van 
Groningen, "The Enigma of Aleman's Partheneion," Mnemosyne SER. III, 3 (1935-36) 241-61. 
JANNI: P . .lanni, "Interpretazioni di Alemane," RFIS 40 (1962) 18D-5. JURENKA 1896: H. 
Jurenka, "Der aegyptische Papyrus des AIkman," SA WW 135 (1896) 1-35. JURENKA 1897: 
H . .lurenka, "Epilegomena zu Alkmans Partheneion," Philologus 56 (1897) 399-405. KUKULA: 
R. C. Kukula, "Alkmans Partheneion," Philologus 66 (1907) 202-30. MARZULLO: B. Marzullo, 
"II Primo Partenio di Alemane," Philologus 108 (1964) 174-210. PAGE 1937: D. L. Page, "The 
Chorus of Aleman's Partheneion," CQ 31 (1937) 94-101. PAGE 1951: D. L. Page ed., Aleman: 
The Partheneion (Oxford 1951). Partheneion II: Aleman fr.3 PMG (POxy. 2387). PBEK: W. 
Peek, "Das neue Alkman-Partheneion," Philologus 104 (1960) 163-80. SCHEIDWBlLBR: 
F. Scheidweiler, "Zu Alkmans Partheneion," RhM 93 (1950) 242-9. SCHWENN: F. Schwenn, 
"Zu Alkmans grossem Partheneion-Fragment," RhM 86 (1937) 289-315. SCHWYZER: 
E. Schwyzer/A. Debrunner, Griechische Grammatik II (Mi.inchen 1950). STOBSSL: F. Stoessl, 
"Leben und Dichtung im Sparta des siebenten Jahrhunderts," in Eumusia Howald (1947) 
92-114. WILAMOWITZ: U. von Wilamowitz-Moellendorff, "Der Chor der Hagesichora," 
Hermes 32 (1897) 251-63. 

For bibliographies concerning the work done on Parth. I, see van Groningen and 
Page 1951. I hasten to add that Page's work is the most authoritative available. If in the body 
of this paper I venture to disagree with Page's interpretation on more than one occasion, 
I do so because his edition is a convenient starting point, which is to say that without it, 
this paper would not have been written. I need not labor the point that anyone who tries 
his luck with Aleman is immeasurably in Professor Page's debt. 

The following items came to my attention after this essay was completed: A. Garzya, 
Studi sulla lirica greca da Alcmane al primo impero (Messina 1963) 13-46; A. F. Garvie, "A Note 
on the Deity of Aleman's Partheneion," CQ 15 (1965) 185-7; M. L. West, "Alemanica," CQ 
15 (1965) 188-202; G. Devereux, "The Kolaxaian Horse of AIkman's Partheneion," CQ 15 
(1965) 176-84; G. Devereux, "Homer's Wild She-Mules," ]HS 85 (1965) 29-32; G. Devereux, 

321 



322 ALCMAN'S PARTHENEION I RECONSIDERED 

though somewhat warily, by Hermann Diels (373-4), following up the 
hints of G. Hermann, H. L. Ahrens and W. von Christ, among others. 
Diels admitted that the poem itself offers no conclusive evidence in 
support of anyone specific division. His own proposal was, in the end, 
that very likely half-chorus A sang the first four lines of each strophe, 
half-chorus B the second four lines and the whole chorus the remaining 
six lines. 

Other scholars have proposed different divisions. 2 Blass3 thought 
that each strophe was sung by one member of the chorus; according 
to Wilamowitz4 the mythical part was sung by the whole chorus, 
whereas the latter half of the song was produced by a single soloist. 
Even van Groningen (259), who criticizes Blass and Wilamowitz for 
the implausibility of their schemes, concludes that "the first part ... 
is sung by the chorus and its leader together. Afterwards (with vs. 
36) the leader separates and the other maidens continue to sing alone." 

The fullest criticism of the attempts to analyze Partheneion I into 
anything but a monochoral ode is that of Denys Page.6 He has three 
principal objections to dividing the poem into half-choruses or any 
other sub-elements: (1) to judge from the papyrus, which is full 
oflectional signs but carries no paragraphi indicating change of singer, 
the Alexandrians had no knowledge of any such division; (2) various 
arguments from the text which have been used to demonstrate 
division, especially the argument from the supposed contrast between 
[E"JYc1v p.Ev and .?yc1[v] BE at 85-7, can be shown to be inadequate; 
and (3) the divisions proposed make for irregularities unparalleled 
elsewhere in Greek choral literature. 

Page's refutations, especially those under (2) and (3), seem to me 
eminently just. The schemes of division proposed, particularly by 
Sheppard 6 and Stoessl, are marvels of disorder; their erratic nature 

"The Enetian Horse of AIkman's Pdrtheneion," Hermes 94 (1966) 129-34. I refer to the sug­
gestions of Garvie and Devereux in the text. Garzya's chapter is largely a restatement of his 
1954 position. West's analysis differs so sharply from mine that I have thought it better 
not to allow a discussion of it to add to the polemical burden of the paper. 

2 For details, cf the summary of Page 195I.57ft". In addition to the three British scholars 
singled out for criticism by Page (Sheppard, Bury and Bowra) it may be useful to mention 
Stoessl, whose scheme is perhaps even more venturesome than those of the others. 

a F. Blass, in RhM 40 (1885) 22. 
, Wilamowitz 262; cf also Scheidweiler. 
5 Page 1937; cf also Page 195I.59ff. In the earlier study Page demolishes the evidence of 

his opponents regarding irregular divisions between speakers and singers in Attic drama. 
S J. T. Sheppard, "The Partheneion of AIkman," Essays and Studies Presented to Sir 

William Ridgeway (Cambridge [Eng.] 1913) 124-35. 



THOMAS G. ROSENMEYER 323 

completely rules out the lyrical responsion one might expect from 
singing half-choruses. Moreover, arguments from the text have 
usually been based on a mistaken understanding of the function of the 
first person singular in choral poetry. 7 It is not surprising that even 
Bowra, originally a supporter of the idea of division, has now relin­
qUished it in the second edition of his Greek Lyric Poetry. Bowra 
(1961.64) continues to think that it would be preferable if an arrange­
ment in terms of half-choruses could be worked out but admits that 
"there is no need to think that Aleman does this, and indeed it is 
difficult to arrange the semi-choirs in any way that will win accep­
tance." Other scholars who argue against half-choruses are Schwenn 
(311), Davison (445 n.3), Farina (55-64), Lesky,8 and Garzya (38). Some 
of them, as for instance Lesky, renounce the idea of half-choruses with 
patent regret. 9 Page himself (1951.48) shows his uneasiness by sug­
gesting that, though the girls sang together as one chorus, they may 
have divided into two groups for the sake of the dancing. 10 Garzya 
(38) too is willing to opt for this compromise. 

One of the reasons why some scholars continue to feel a certain 
nostalgia for the idea of half-choruses is a text which appears in Page 
1951.13 as scholion VIII. It is written against col. ii lines 14-15 (= 
Partheneion 1.48-9) and appears to make a distinction between the 
followers of Agido and the followers of Hagesichora : 

9!frrT7Jay~8o·TOVT . .... 
?Ta9!VrrT7JaY7Jar~g . 

With Page, except for the reading of iT,ll this may be read as follows: 
a, 71'( apa) T(fj~) • Ay~8o(t) Toih[ 0 ••.• ]oTa at 71'( apa) T( fj~) 'AY7Ja~x6( pa~) 

7 Cf infra n.34. That Page's opposition to the proposed schemes is, in part, based on what 
seems to me a questionable premise-viz. that" most of the lines are obviously sung by 
persons on the side of Hagesichora "-does not, therefore, affect the justice of the case 
against them. 

SA. Lesky, Geschichte der griechischen Literatur 2 (Bern 1963) 173. 
9 Ibid, "Der Inhalt legt dies an manchen Stellen nahe ... Man wird es der dialogischen 

Haltung einzelner Partien zum Trotz wahl einem Chore belassen mussen." He accepts the 
strength of Page's argument (3). 

10 For other cases of dancing half-choruses. see Ar. Eq. 242f and Av. 353, discussed below, 
p.356. It is of some interest that with regard to Parth. II Page is in considerably more doubt 
about the arrangement of the singing; at CR 9 (1959) 16 he says: "I do not understand the 
arrangements here: I do not believe that these lines" (viZ. 1-9) "can be sung by the whole 
choir ... but what is the alternative?" 

11 According to F. G. Kenyon, The Palaeography of Greek Papyri (Oxford 1899) 155, the 
evidence from the papyrus of Aristotle's AthPol and from the Anonymi Lond. Iatrica shows 
that 11'p6, (and 1T€pl) tended to be written ,T, while';' is to be read as 11'upa. 

3-G.R.B.S. 



324 ALCMAN'S PARTHENEION I RECONSIDERED 

If we assume that the note was meant to cover vv.49-50, we may 
then say that the scholion appears to register a change of performers 
between the last line of strophe 4 and the beginning of strophe 5. 
The scholion should be read in conjunction with scholion v, which 
stands opposite v.36 and appears to indicate the beginning of a 
passage in praise of Agido: 

.. ( . HaYLoovs . .... . 

On this I am inclined to follow Blass and Diels, both of whom read 
traces of apri before what appears in the text of Page. 12 Now it is true 
that the scholia are to be used with much circumspection; they can be 
wrong, as is most probably true of the explanation of II€A'1}ao€s v.60. 
I happen to think that scholia VIII and V are entirely correct; on the 
other hand scholion VII, opposite vA3, which appears to say something 
about Hagesichora, is at least out of place, if not entirely mistaken. 
But there is just enough in these comments, pitifully fragmentary 
as they are, to indicate that there may be a case for the existence of 
some kind of dramatic exchange.13 

The ancient evidence, though slight and garbled,14 encourages us 
not to surrender the notion of oLxopla without a struggle. The only 
scholar who, to my knowledge, has proposed a scheme of division 
which satisfies the implied conditions of Page's criticism (3) is d'Er­
rico. 15 His contribution does not appear to have become widely 
known. Briefly, d'Errico divides the chorus into two half-choruses of 
ten girls each, one led by Hagesichora and one led by Agido. That 

11 It must be admitted that our faith in Diels' judgement in this matter is somewhat 
shaken by his statement, p.135, that he follows the traditional explanation, "die auch der 
Scholiast zu teUen scheint," that 'Ay£80Vs in the scholion is to be taken as a subjective 
genitive. There is no telling how the scholiast felt about this; for myselfI have little doubt 
that the genitive is objective. 

18 Marzullo 197 launches a vigorous attack on Page and Bowra for disregarding the 
evidence of the scholia, which, according to M., are mostly reliable. As it rums out, M. 
himself pays little enough attention to the scholia in the matter of assessing the relative 
importance of Agido and Hagesichora. He radically minimizes Agido's role in the pro­
ceedings. 

U But cf. Pollux's reference to Tyrtaeus and Laconian choral division, immediately after 
his general statement about half-choruses (4.107, p.233 line 6 Bethe). The latter runs: 
cHrO"TaV ya.p 0 XOPM £ls Mo p.lpT] "Tp.T]8fj£, "TO p.t.v '1Tpayp.a. Ka.A£t:"Ta.£ ~)£xopla., EKa..,.lpa 8' .q p.oipa. 
TJp.£xopwv, & 8' av-ra£&vG£v, av-r£x6p£a.. Pollux's purely formal definition is safer than the 
further distinction introduced by Conradt; cf. infra n.l03. I suspect that Hephaestion De 
signis 4.74 Consb., with its curious notice of metrical changes within fourteen-strophe poems 
by Aleman, is faultily abstracted from a larger discussion of choral division. 

16 d'Errico 24ff. I did not become acquainted with this study until the bulk of my 
article was already written. 



THOMAS G. ROSENMEYER 325 

Agido is a xopayos is, he feels, stated in v.44;16 that Hagesichora is also 
a xopayos is clear from her name and has of course never been doubted 
by anyone. d'Errico assigns the singing of the odd-numbered strophes 
to the half-chorus of Hagesichora, and the even-numbered strophes 
to the half-chorus led by Agido. That is to say, strophes 4, 6 and 8 are 
Agido's, strophes 5 and 7 are Hagesichora's. d'Errico derives his proof 
of the dramatic division of the poem from cpaa€'is 73, which according 
to him conveys the general idea of: "You will not be able to say that 
you have gone to a good school, as we have"; and from ~ oux opfj's; 
50, which he regards as addressed by Hagesichora's girls to those of 
Agido. Some of d'Errico's detailed suggestions seem to me to lack 
support, nor am I sure that he has not glossed over some of the 
difficulties inherent in any position taken on the poem. But on the 
whole, as I hope to show, his scheme makes good sense. 

It appears to me that there are certain analogies, both structural and 
psychological, between Partheneion I and pastoral contests which may 
help to strengthen the case for choral division. I should like to discuss 
some of these analogies at some length before returning to the 
specific problems of the text of Aleman. 

II 
The question why Theocritus wrote his pastorals in the Doric 

dialect has no single answer. His Syracusan origins are obviously 
relevant, but they cannot provide the whole answer, since his use of 
various dialects testifies to a formal virtuosity which seems to have 
little to do with the accident of birthY Stark18 suggests that Ibycus 
had considerable influence on the creation of the Hellenistic pastoral. 
He does not give his reasons for the suggestion, other than to refer 
generally to "Zitaten und Bemerkungen bei Literarkritikern und 
Grammatikem." Similarly van Groningen (242) hints that there are 

16 d'Errico cites the support of C. del Grande in tl>OPMlrSB (Napoli 1959) 84, who 
there corrects his earlier view, TPAHJIJIA (Napoli 1952) 16 and 296, that the line refers to 
Hagesichora. 

17 It is widely assumed that Theocritus' choice of Doric has something to do with his 
dependence on Sophron. But the Anecdoton Estense (Prolegom. p.12 Wendel) makes the 
point that Theocritus' Doric is different from that of Sophron; and in fact Theocritus' debt 
to Sophron has been greatly exaggerated. His pastorals are not mimes, and even his so­
called city mimes owe less to Sophron than to other models, in spite of the scholiast's 
note, p.305 Wendel. Cj A. S. F. Gow, Theocritus n (Cambridge [Eng.] 1950) 34-5 and 265. 

18 R. Stark, "Theocritea," Maia 15 (1936) 366 n.14. 



326 ALCMAN'S PARTHENEION I RECONSIDERED 

parallels between Aleman and Theocritus and that these parallels 
have often been drawn. In actual fact, there has been very little 
detailed attention to the possible connections between Theocritus' 
pastorals and the archaic choral lyric, especially Aleman. 19 

In his description of Sparta, Pausanias (3.15) comments on the 
heroon of Cynisca, daughter of Archidamus, near the Platanistas; she 
was the first woman to raise horses and to win a victory at Olympia 
with a chariot. Nearby, Pausanias continues, is a monument to 
Aleman, who wrote songs "whose charm was not at all diminished 
by the fact that they were written in the Laconian tongue, an un­
attractive dialect": cTJ, 1TOt7JaaVTL &,ap.aTa ouoev l~ ~oov1]v aUTWV lAvp.7}-

- A' < \ - II " ",J.. Th vaTO TWV aKWVWV TJ yl\waaa, TJK,aTa 1Tap€xop.€VTJ TO €V'f'WVOV. en 
there are the precincts of Helen and Heracles; the former near the 
tomb of Aleman, the latter close to the wall. What follows is a discus­
sion of the quarrel between Heracles and the sons of Hippocoon. 
Thus in one and the same passage we have talk of the Platanistas, 
a woman who has to do with horses, Aleman, Helen and the 

1. Direct evidence that Theocritus knew and imitated Aleman is nil. We know that 
Aleman was studied in the Hellenistic age, when Sosibius the Laconian (to be distinguished 
from other Sosibii active in Alexandria during the first half of the third century B.C.) wrote 
a lIfpl 'AAKp.avo" fr.6 FGrHist III B p.595; although Jacoby (FGrHist III B 2 pp.635-7) may 
have put his dates lower than the evidence requires. In fact, knowledge of Aleman can be 
traced through the fifth and fourth centuries; cf FGrHist III B 2 p.648. Later Alexander 
Polyhistor wrote a geographical commentary on Aleman: frgg. 95 and 96, FGrHist III A 273. 
When the Alexandrians established a canon of the nine lyricists, Aleman was included (see 
Wilamowitz, Die Textgeschichte der griech. Lyriker [Berlin 1900] 16ft"); for the judgement of 
the literary historians, cf "sweet Aleman" in AP 9.571.2 and 01}Avp.£Aft, 'AAKp.avo, d1}S6v£, 
AP 9.1B4.9. There is therefore no reason to doubt that Theocritus was acquainted with 
Aleman's work, even though he did not imitate his dialect in every detail (Wilamowitz, 
op.cit. 53). The attempt of W. Schmidt (Gesch.gr.Lit. I [Munchen 1929] 466 n.B) to demon­
strate imitations of Aleman in Theocritus and Apollonius Rhodius was unsuccessful. But 
two of Theocritus' glosses point if not to Aleman at least to Doric choral poetry, They both, 
dlTT}, and fiO'1TV1}AO" occur in Id. 12, a curious little poem in which Theocritus somewhat 
approaches the learned manner of Callimachus. The schol. pp.249f Wendel reports that 
dlTT}, is dialectal and compares its use by Aleman to refer to ora, E7T£P&'C1'TOV, K6pa" fr.34 PMG 
(the reading K6pa, is superior to the alternative xopO&.<;); cf also the Sud a pp.63, 66 Adler and 
Et.Mag. 43.31 GaisE. We cannot tell whether da7TV1}Ao" which Theocritus says is from 
Amyclae. was part of the same poem of Aleman; cf Callim. fr.68 Pf. The two forms 
are discussed together in Et.Mag. 43.31 Gaisf., Eust. 732.24 on II. 9.6,1500.27 on ad. 4.362 and 
1547.1B on ad. 5.480; R. B. Onians, The Origins of European Thought 2 (Cambridge [Eng.] 1954) 
199 n.14, attempts to link them etymolOgically and semantically. Ph.-E. Legrand, Etude sur 
Theocrite (Paris 1898) 256, has a catalogue of poetic words in Theocritus worth investigating 
for their origins. Especially interesting, because they point to the Doric choral lyric as well 
as to Homer, are 8aa7TA1jTt> (2.14), cf ad. 15.234, Simon. fr.522.1 PMG, Lye. 1452; and 
oraiialw, (25.229), cf ad. 3.316, 15.13, h.Hom.Ap. 362, Aleman fr.1l2 PMG, Bacchyl. 5.81. 
Finally, according to Gustav Schlatter, Theokrit und Kallimachos (Zurich 1941) 50-I, Xa6, 
(ld. 7.5) may be a borrowing from Aleman. But the scholiast merely calls it a Spartan 
word. 



THOMAS G. ROSENMEYER 327 

Hippoc06ntids. Kaibel20 maintained that Theocritus' Epithalamium 
for Helen (Idyll 18), in which the Platanistas is prominently featured, 
is an imitation of Sappho and that it permits us to get an idea of 
Sappho's epithalamia. That was before the publication of Sappho 
fro 44 LP, the Hector and Andromache papyrus. Kaibel is hard put 
to it to explain why the poem is in Doric rather than Aeolic. His 
explanation: the subject of Helen and the singing of the girls in 
Sparta called for Doric. He does not consider the possibility that the 
model might be Aleman, in spite of his recognition that this or that 
phrase has its analogue in Aleman; for example, he compares Idyll 
18.30 with Partheneion 1.4sff. 21 The Epithalamium for Helen is not a 
pastoral, but its echoes of Spartan choral lyric are of some significance 
in this context. 

According to an old and demonstrably mistaken Peripatetic tradi­
tion, the pastoral originated from girls' performances in honor of 
Artemis Caryatis or some other Laconian or Arcadian fertility 
divinity.22 The theory may have recommended itself to begin with 
because of the recognition that Laconian partheneia in honor of 
Artemis had points of resemblance with the pastoral. It is salutary to re­
member how little we know about the partheneion as a literary type. 23 

The evidence for Pindar's Partheneia was assembled by Boeckh more 
than a century ago ;24 later discussions have had very little to add to 
his magisterial treatment. Eirber25 advances the suggestion that we 
must distinguish between 7TapO'vu.r. and 7TapO€V€La; the latter are songs 
about girls, or addressed to girls, while the former are songs sung by 
girls. In this he follows the Suda. 26 But this kind of accentual distinction 

20 C. Kaibel, "Theokrits EAENHE EIIH9AAAMION" HeTltles 27 (1892) 249-59, esp. 253. 
21 Theon, hypoth. Idyll 18 (p.331 Wendel), says that some things are taken from Stesi­

chorus' Helen. But it appears that Theon was thinking of the details of the legend rather 
than of the poetic treatment. In Theocritus' epithalamium the chorus is said to be singing 
as one group: a€LSov S' &'/La TTaaat is ~v /LlAos iYKpoTIotam I TToaalTTEpmMKTOtS (vv.7-8). The 
reading a/La is that of the pap.; but even the reading of the codd., <lpa, would not affect the 
sense, except that a/La introduces a note of emphasis, as if Theocritus were stressing that 
on this occasion the song was monochoral. That epithalamia could be amoebean is shown 
by Ar. Pax 1332ff and Catullus 62. We do not know what Aleman's practice was; all we 
know is that he was much admired for his wedding songs; cf Leonidas, AP 7.19.1. 

22 Paus. 3.10.8; 4.16.5; Theoc. Proleg. pp.2, 8 Wendel. The tradition has been thoroughly 
demolished by E. Cremonesi, "Rapporti tra Ie origini della poesia bucolica e della poesia 
comica nella tradizione peripatetica," Dioniso 21 (1958) 109-22. 

23 RE has no entry under this heading. 
DC A. Boeckh, Pindari Opera quae supersunt II.2 (Leipzig 1821) 589-95. 
25 H. Farber, Die Lyrik in der Kunsttheorie der Antike (Diss. Munchen 1936) 39. 
26 S.V. 1rap8EVE'ia; cf. schol. Ar. Ra. 918 Duebner. 



328 ALCMAN'S PARTHENEION I RECONSIDERED 

is unparalleled in the ancient discussion of poetic genres and should 
probably be chalked up to a confusion in the transmission. Proclus, 
probably following Didymus, says that partheneia mix praise of gods 
and men, and this is borne out in the extant examples in Aleman and 
Pindar.27 Another tradition, that the partheneia were accompanied 
by cdJ>"ot, does not seem to be based on firm evidence but rather 
derives from the discussion of which we have an instance in Athenaeus 
4.176f, where, among several varieties of av>"os-, a 7T'apfUv£os- av>"os-, as 
well as a 7T'a,O£KOS- and an &v0PEtOS- av>"os-, is mentioned. 28 

The most notable ancient reference to partheneia is in Dionysius 
of Halicarnassus, De Demosthene 39 (=De or.vet. 1073-4). Speaking of 
the apxa{a Kat aVu77Jpa. apJLov{a, with its lack of connectives, its swift 
changes of syntax, its anacolutha, and its want of logical coherence, 
Dionysius says that all of Aeschylus is a case in point, and all of Pin dar, 
except for the Partheneia and some other poems that require a treat­
ment similar to that of the Partheneia; but that even in them there is 
an EVy€v€la and a U€JLvo77Js; which preserve the ancient patina. Diony­
sius' remark about the comparatively less austere quality of Pindaric 
partheneia deals with apJLov{a, i.e., with syntax and logical structure, 
rather than with mood. But Plutarch, in a passage to which Ahrens 
(246) drew attention,29 clearly thought that partheneia were less 
solemn than other kinds of poems; Plato, he says, preferred the 
Dorian mode because of its UEJLV077JS;, though he must have realized 
that Aleman and Pindar and Simonides and Bacchylides wrote many 
partheneia in Doric. Ahrens comments: "Die Worte OVK ~YVOE£­
~pwnK& wollen offenbar sagen, dass den Partheneia ... das UEJLVOV 
nicht zukommt."30 And indeed one suspects that Dionysius could 
not have used the term UEJLV077JS- in his comments unless he also was 
thinking of matters which strictly speaking do not come under the 
heading of apJLovla. 

With respect to one of the two Pindaric partheneia which have come 

21 Proclus, Chr. 319B32ff (ed. Severyns 1.2 (1938) 2Uff). Cf. Ahrens 246ff. 
28 Cf. also Pollux 4.81. 
28 Plu. De mus. ch. 17, U36EF. The question of authorship does not concern us here. 
30 Plu. De mus. 1136F: O?JK ~')'V&EL S' (sc. Plato) OTL 11'0'\'\« LJdJpLa 1I'ap8lPELa ..• '~KIL&v£ Kat 

lILv&;.pw£ Kal.l:£lLwvlMJ£ Kal. BaK")(V~I.&rJ£ 1I'E1I'0l.",.«£, &.Uci: p.~v Kal. 1I'poaoSLa Kal. 7Ta£iivES, Kal. p.Evro£ 
OT£ Kal. TpayLKOI. olKTol 1I'OTE £11'1. TOU LJwplov TP07TOV £ILEAwL&r787jaav Kal T£va £PWT£Ka.. Thus 
Plutarch and Dionysius tell against the view of Mrs. Burnett, p.30, that scholars have 
been wrong to make a pretty and playful poem out of a solemn one. She attempts to 
«clear the air of 'Sapphic intimacy", while restoring to Aleman"s choral ode the public 
solemnity of a major Laconian occasion:" 



THOMAS G. ROSENMEYER 329 

down on any scale (fr.94b Snell), Bowra (1964.363) comments that 
"Pindar makes the girls use words appropriate to their age and sex 
and, as Dionysius sees, adopts a simpler and easier style. He is con­
scious that he is making concessions; for he makes the girls sing ... 

I must think maidenly thoughts 

And utter them with my tongue." 

One may wonder whether this does full justice to the complexities of 
Dionysius' use of the term apf-Lovta,31 but the point about the differ­
ence in tone must be right. Bowra continues: "Since the poem is a 
oacpV7]cpopLKIIV and belongs to a special ceremony, this may be why 
Pindar treats it in this exceptional way, since he does not follow this 
method in all his Maiden-songs. Yet this dramatic device of making 
the girls speak for and about themselves is already present in Aleman 
and comes from an ancient tradition, which Pindar is at full liberty to 
follow. Certainly this poem catches something of Aleman's lightness 
and gaiety when it speaks of the maidens' dress (5), or garlands (9), or 
the calming effect of their song (11-15), or their affection for their 
family (26-27), or their pride in its achievements (31ff)."32 

Bowra, then, takes it for granted (1) that some Pindaric partheneia, 
perhaps especially the oacpV7]cpopLKa, were more lighthearted than 
others, and that this lightheartedness was a function of their dramatic 
nature: they featured young girls singing for and about themselves; 
and (2) that this girlish gaiety is to be found also, and more typically, in 
Aleman's Partheneia. 33 Bowra's remark about the dramatic difference 
between partheneia and other poems is just, but we can go further 
than that. As Mary Lefkowitz has shown,34 we must make a fairly 
radical distinction between Pindar's Epinicians on the one hand and 
his Paeans and Partheneia on the other. In the Epinicians there are few-

31 Cf. also U. von Wilamowitz-Moellendorff, Pindaros (Berlin 1922) 435. 
31 It should be noted that the particular expressions which Bowra instances as character­

istic of the Alcmanic gaiety are very close to some of the principal features of bucolic 
cheerfulness, with the important exception of family pride. For Aleman's attention to 
(rustic?) food, cf. infra n.43. 

sa For drama and liveliness, see also the scrap of a Hellenistic .. partheneion" in Col­
lectanea Alexandrina ed. J. U. Powell (Oxford 1925) 193. 

U Mary R. Lefkowitz, "TO KAI ErQ: The First Person in Pindar," HSCP 67 (1963) 177-
253. Perhaps Miss Lefkowitz's results are a little too tidy; also, perhaps, she does not suffi­
ciently explore the question whether two statements of the same type may have entirely 
different functions in the light of the thematic context in which they appear. But the larger 
findings of her inquiry seem to me justified and important. 



330 ALCMAN'S PARTHENEION I RECONSIDERED 

according to Miss Lefkowitz, no-choral statements; that is to say, the 
poetic ego, the persona, is singular, and roughly identical with the 
poet. In the Partheneia, on the other hand, the ego is never that of the 
poet but always that of the person or persons, or perhaps better, the 
characters performing the dance-song. Thus Bowra's contention that 
in fr.94a Snell, Pindar is speaking for himself, must be amended. 
The masculines (e.g., cPLAlwv v.U) are comparable to the occasional 
masculines in the female choruses of drama,35 rather than to the 
occasions when Aleman refers to himself by name, as in fr.95 
PMG.36 

Anyone who comes afresh to the remains of Pindar's Partheneia is 
immediately struck by the frequency of references to Pan in them. We 
are reminded of Philostratus, Imagines 2.12, an ecphrasis of Pindar: 
the picture is full of bees, Pan and dancing. Partly this prominence of 
Pan is an accident of transmission; several of the notices come from 
the scholia to Theocritus. In fact Pan appears to have been less con­
spicuous in the two regular books of Partheneia than in the book of 
KEXWpLG/-dva 7TapO€VELva. 37 Pindar's most celebrated address to Pan, 
fr.95 Snell, occurred in that book; it is a hymn to Pan and the Mother 
of the Gods, to whom Pindar was indebted, we are told, as the 
result of a religious experience. 38 Whatever the truth of that 

Ii Cf. U. von Wilamowitz-Moellendorff, on Eur. Hipp. 1103ff, in Aischylos-Interpretationen 
(Berlin 1914) 195. 

8e Marzullo 187-8 contends that in Parth. 1.39 lywv I)' &dl)w must be the poet speaking in 
his own person, avowing that, for a man to be EVtPPwv, he had better not engage in the 
irresponsible ventures of which the myth of the Leucippids is an example, but he must 
sing of Agido. 

87 Cf. Vita Pindari Ambrosiana, ed Drachmann, vol. I p.3: 'TTapOEVlwv {3', ""pETaL I)E Kat y' ;; 
E'TTLyp&,pETaL KEXWPLUpivwv 'TTap8EVlwv. 

18 For details, see Wilamowitz, Pindaros (Berlin 1922) 270-2 and Bowra 1964.49-50. For a 
passage which presumably may tell us something about the spirit and the color of hymns 
to Pan and the Mother, see Ar. Aves 737ff. But E. Fraenkel's inferences, Beobachtungen ~u Aristo­
phanes (Roma 1962) 209-11, seem to me too precise. Rather, Phrynichus has written music 
which is very much like certain rustic songs in honor of the two divinities. For Pan as the 
'dog' of the Mother, Wilamowitz compares Call. Hymni 4.228. For the joint hymning of the 
Mother and Pan by nocturnal girls' choruses (surely not "night after night" as Lattimore 
translates), cf. Pyth. 3.77-9. Pindar pictures the girls as singing 'TTap' l,.,.dv 'TTp60vpov; one won­
ders whether this is not the source of the tradition that Pindar had a chapel in front of his 
house dedicated to the joint service of Pan and the Mother and that his own daughters were 
among the celebrants. Bowra feels that the opening line offr.95 Snell is so close to that of an 
Attic skolion, fr.887 PMG, "that Pindar must surely have picked up something from it, 
and his poem may not be very far in date from his visit to Athens c. 474 B.C." I would rather 
explain the similarity between the two passages as evidence of the fact that these partheneia 
had something about them in diction and mood which put them close to the lively, un­
solemn, amoebean skolia. The utilization of formal similarities for the purpose of compara­
tive dating is always hazardous. 



THOMAS G. ROSENMEYER 331 

tradition,39 it is surely significant that when Pindar writes about Pan, 
the pastoral divinity, it is in the form of partheneia. It is impossible at 
this distance to determine whether the partheneia of the third book 
differed substantially from those of the other two books, and how. 
Nor can we tell precisely how the 8acplffJcpoptKeX differed from regular 
partheneia. 40 If fr.104b Snell is from a (jacpV1JcpoptKoV, one cannot help 
being struck by its similarity to Aleman fr.56 PMG, which is surely 
from a partheneion also. The rustic color, the joy in food and physi­
cal satisfaction, as well as the gaiety, put us in mind of Aristophanes 
in his more gentle moments and of Theocritus. 

The world conjured up in Aleman's partheneia is one of 
intimacy and good fellowship. Colin Edmonson41 has published 
a roof tile bearing women's names, connected with the worship of 
Apollo and Hyacinthus at Amyclae. He thinks that this and other 
similar catalogues of names, of which there are a number of examples 
in Spartan inscriptions, may have had a ritual purpose. The fact that 
in this instance the names are written on a roof tile rather suggests a 
less formal occasion. It is easy to imagine a situation in which the 
girls, excited by their association in some venture, perhaps in a 
ceremony, or by the recollection of it, decided to record this associa­
tion, more or less in the manner of modern school girls who write 
their names into year books or club records. In any case, the co­
incidence of these graffiti on the one hand and the list of girls' names 
in Partheneion 1.70-7 is evident.42 The spirit of the lines is one of 
playful raillery, of teasing. To find similar examples of youthful 
exuberance, of young people closely known to one another and 
emphasizing their bond through banter as well as declarations of 
affection, we must skip several centuries until we come to the Socratic 
circle, and, later, the minuscule world of the pastoral singers. 

Some parts of the Dorian choral tradition, then, especially the 
partheneia (and perhaps also some of the paeans), point forward to the 
Hellenistic pastoral in several ways: a greater gaiety by comparison 
with the more solemn songs of cult and veneration; an interest in the 

3& It is no longer fashionable to shrug it off as unconcernedly as Gildersleeve did in his 
Pindar: The Olympian and Pythian Odes (New York 1885) xiii. 

co Proclus, Chr. 320f (Severyns [supra n.27] 211-32); L. R. Farnell, The Works of Pindar II 
(London 1932) 425-7. I have not been able to use F. Sbordone, .. Partenii pindarici e dafneforie 
tehane," Athenaeum 18 (1940) 26-50. 

n C. N. Edmonson, "A Graffito from Amyklai," Hesperia 28 (1959) 162-4. 
n For a detailed analysis of Parth. 1.70-7, see p.347 below. 



332 ALCMAN'S PARTHENEION I RECONSIDERED 

data of nature, and food; 43 an emphasis on intimacy; and, though this 
must remain a statement without proof, a certain impersonal quality 
whereby, in this type of choral lyric, the poet allows his characters to 
sing without obtruding his own ego, a procedure which is quite 
distinct from other kinds of archaic lyric poetry, choral or non-choral, 
and which in poetry is found again primarily in drama and in the 
Theocritean pastoral. 

III 
One of the many difficult problems concerning Partheneion I is 

determining the concrete situation which we are to envisage for its 
performance. It is generally held that the poem differs from, say, the 
odes of Attic drama in that it involves an act of worship, with an 
offering brought to the goddess. But there are significant differences 
of scholarly opinion concerning the identity of the goddess, the nature 
of the gift, the time of the service,44 the precise role played by Agido 
and Hagesichora, and the relevance of the legend told in the first half 
of the song. 

It may be useful to remind ourselves that the proof of the ritual 
status of the hymn is extremely slight. It hinges on v.61, cpapos cp€poluCts. 
Now whatever the meaning of the phrase and of the larger sentence 
of which it is apart, 45 it is hardly sufficient to persuade us that we are 

U For food in Aleman, ef. Bowra 1961.66-9, and frgg. 17, 19, 20, 42, 96 PMG. In fr.95 
Aleman refers to himself as a furnisher of food; for the question whether Aleman himself 
could be a participant in a partheneion, ef. p.338 below. Fr.56 remains something of a 
mystery; somebody is addressing a girl who is supposed to have made a cheese from 
lion's milk in a golden bucket on many past occasions. Apparently the festival has some­
thing to do with Hermes, but also with other gods. The offering is to be compared to the 
Kp'{1avwTolof fr.94. The suggestion of Willem den Boer. Laconian Studies (Amsterdam 1954) 
264, that the reference is to a mythic parallel with the Spartan ordeal of cheese stealing 
(if. Plu. Lye. 17.5-18.2) seems to me to rest on little evidence and to be intrinsically unlikely 
because of the feminine construction. As for a feeling for nature, the main exhibit is the 
much disputed fr.89. Whether it is by Aleman or not (negatively H. F. Fraenkel, "Bine 
Stileigenheit der friihgriechischen Literatur," GGN 1924. pp.83-4 n.4; ef. also his Diehtung 
und Philosophie des frUhen Grieehentums [APA PhilMon 13, New York 1951] 229 n.19; P. Maas. 
Greek Metre transl. Lloyd-Jones [New York 1962] 9; but see the protest of R. pfeiffer in 
Hermes 87 [1959] 5 n.l). the fact that the ancients could ascribe the lines to Aleman tells us 
much about the kind of poetry associated with his name. Cf. also fr.60, with its intriguing 
combination of two plants favored by Theocritus, iAlxpvaos and lonmpos. 

44 Wilamowitz 255, against the majority of his colleagues, insisted that the song was sung 
in broad daylight. For the details of the scholarly debate concerning the points mentioned, 
ef. Page 1951 passim. 

(5 See the discussion below, pp.343ff. The suggestion of M. West. "Alemanica," CQ 15 
(1965) 198, to read </>aFos for </>apos is unworthy of its brilliant author. 



THOMAS G. ROSENMBYBR 333 

dealing with a hymn sung as part of an official cult ritual. Even if the 
offering of the ~&pos (or ~apos) was part of a cult in Sparta, rather than 
a dedication by victors in a singing contest, the fact that the girls here 
sing of carrying such an implement does not immediately signify 
that the song is part of a ritual, primarily calculated to please or move 
or in some other way affect the disposition of the deity. When Archil­
ochus says of himself in fragments 76 and 77 D. that he is a leader of 
the paean or that he knows how to perform the dithyramb of Diony­
sus, we know from the meter that he is not doing anything of the 
sort. Similarly, when a tragic or comic chorus addresses a prayer or 
offers gifts to a god we are not invited to think that we are participating 
in an act of worship in the ordinary sense of the word. The Attic stage, 
not to mention the Callimachean hymn, teaches us that it is possible 
to have a poetic reality which has many of the earmarks of a cult 
reality but which is designed to impress as poetry, even as entertain­
ment, without practical relevance to the relation between man and 
his god. 46 

The poem itself, therefore, does not help us very far in determining 
the conditions under which it, or any partheneion by Aleman, would 
be performed. But there are three ancient passages which mention 
Spartan women's festivals in terms which may be pertinentY 
Plutarch refers to a 7Tavvvx,s €OPT~ in which the women feasted with 
the girls, with some of the chief ladies by themselves in the large 
avopwv. Pausanias speaks of a group of women called Dionysiades who 
engage in a racing contest which was first instituted by command 
from Delphi. Athenaeus has an account of the Hyacinthia, which goes 
back to the Laconica of Polycrates. All three passages contain elements 
which throw light on the possible conditions of Partheneion I, as 
scholars working on Aleman have long known. Athenaeus' report is 
particularly intriguing. During the three-day Hyacinthia, we are told, 
there are at first no paeans to the gods, nor do the celebrants wear 
garlands or eat bread. But after this initial mourning period the 
spirit changes. There are songs to the gods and performances by 
boys on Kt(}apat and aVAot. Others ride on horseback through the 
(}laTpov. The girls also have their fun; some are conveyed on 

48 P. D. Arnott, Greek Scenic Conventions in the Fifth Century B.C. (Oxford 1962) 53fI, reminds 
us that true sacrifices are not permitted on the stage and that the central altar has no 
ritual standing. 

47 Plu. MOT. 775D, Paus. 3.13.5, Ath. 4.17.139c-p. Cf Diels 369-71. 



334 ALCMAN'S PARTHENEION I RECONSIDERED 

decorated Kava8pa (wicker carts), others race with teams of horses. 
Everybody turns out and views the spectacle. Other writers con­
firm that the Hyacinthia also featured women's choruses and 
contests. 48 

The importance of these reports has often been noted. In Partheneion 
I, also, there is a reference to racing (59), and the many equine touches 
in the poem gain added significance if we assume that the horse race 
is closely connected with the performance of the poem itself. It is 
worth emphasizing further that Athenaeus' account stresses the 
spectacle and the fun rather than the solemnity of the occasion. The 
performances, or some of them, take place in the 8€a-rpov. Now it 
would be anachronistic to assume that Aleman's partheneia were 
performed in a theater, before a group of interested but detached 
spectators who had paid to witness the spectacle. Still it would be fair 
to regard Partheneion I as a performance of sorts, a poem written 
for entertainment and possibly elevation but not for the purpose of 
worship.49 It is difficult to conceive of a cult exercise that would not 
be allergic to the praise (and chaffing) of persons which fills at least 
half of the poem. If the hymn were part of a religious ceremony, the 
encomia lavished on Hagesichora and Agido might well make the 
goddess impatient. By the same token, banter along the path of a 
procession is one thing;50 but banter written into a cult hymn is a 
logical absurdity. 

I am, therefore, inclined to think that the poem constitutes a 
performance, much along the lines of the choral hymns performed in 
the theater of Athens. Bruno Snell's remarks on the metrics of 

48 See Eur. He!. 1465ff; Philostr. Vit.Soph. 2.12. For further accounts of women's races, 
see Page 1951.56. 

4t One wonders whether there are any securely attested cases of noted writers con­
tributing hymns to be sung in the public worship of an established divinity. Cf the doubts 
raised about Sophocles' paean to Asclepius by J. H. Oliver, "The Sera pion Monument and 
the Paean of Sophocles," Hesperia 5 (1936) 91-122. Oliver thinks that Philostratus (VA 3.17) 
attributed to Sophocles the famous (and anonymous) paean to Apollo of which a copy has 
been found in Cenchreae. It is my own feeling that Sophocles' paean and the Delphic 
paean which Aeschylus was asked to write (Porph. De abst. 2.18) were designed as '7Tpoolp.ta, 
analogous in status to the Homeric Hymns, dedications to the god; songs that could be and 
surely were performed in honor of the god, but were not intended to replace or rival the 
traditional hymns which formed the musical part of the service proper and served the 
community regularly to revalidate their relations with the god. 

60 For Y£<PVPtCTP.OS along the path of a cult procession, cf Hesychius s.v. y£<pvpls and 
y£<pvptCTTal; also Strabo 9.400. See also schol. Ar. Pluto 1014, with its reference to women 
chaffing each other from carts; the notice retains its value even if the scholiast is mistaken 
about the festival in question, as is held by L. A. Deubner, Attische Feste (Berlin 1932) 73-4. 



THOMAS G. ROSENMEYER 335 

Partheneion 1,51 viZ. that in some ways it anticipates some of the 
practices of Attic drama, may throw additional light on the relation­
ship. Saying that Partheneion I is likely to have been a performance 
rather than a ritual song does not obviate the need to find out about 
the formal addressee, the nature of the gift, and so forth. On the 
whole, I should think, the weight of the evidence favors the majority 
view that the goddess addressed is Artemis, rather than Helen, or 
Eileithuia, or one of the other divinities that have been suggested. 52 
I also think it likely that the performance took place at night. Fortu­
nately these matters are beyond the immediate scope of this paper. 
The essential point is that on the evidence available we are not 
required to think of Partheneion I as constituting part of a religious 
ceremony. Only with this firmly in mind can we give fair considera­
tion to the next question, which is whether there was a rival chorus. 

Let us, for the purposes of this analysis, distinguish between 
'agonistic' and 'amoebean'. The former designates the external 
rivalry of a chorus with another chorus, presumably for the acquisi­
tion of a prize or some other distinction. The latter refers to the 
internal rivalry and jockeying for position between members of the 
same chorus, especially between two half-choruses. In the first 
edition of his Lyric Poetry Bowra supposed that both situations exist in 
Partheneion I; and I hope to show that Bowra's initial instincts were 
right. His assumption of an agonistic situation continues to hold the 
field. 53 It has, for better or worse, been held by a number of scholars, 
including Diels (361), Kukula,54 Page (1951.57), and Garzya (68 and 
note), though the latter's language is sufficiently hesitant to have 
suggested to Harvey55 that he found no evidence in the poem for the 
existence of a rival chorus. Among the critics who deny the existence 
of a rival chorus are Wilamowitz (259), Marzullo (197ft) and Burnett 

1i1 B. Snell, Griechische MetrikS (G6ttingen 1962) 47; cf also infra n.l07. Snell's point is 
that, contrary to the practice of Pindar and Bacchylides, who usually favor one metrical 
type in writing a poem, Aleman and Ibycus anticipate the freer metrical flow of dramatic 
odes by using different metrical species side by side in one and the same strophe. 

52 But see now A. F. Garvie in CQ 15 (1965) 185-7, who makes Phoebe, daughter of 
Leucippus, a plausible candidate for the position of the goddess in the poem. 

53 Bowra 1961.56-8: "His song indicates ... some kind of contest, which is suggested by 
such words as p.aXOlJ7-m at 63, rXp.vva, at 65, 'lTOVWV at 88, and lp~vas Eparus at 91 ... " "There 
is then something to be said for the view that Aleman's choir competes against another 
choir called 'The Doves' and that the competition consists of appearance, singing, and 
running." 

64 Kukula 216. He points out that Pindar and Bacchylides use p.a)(TJ and vlKrJ of contests. 
55 A. E. Harvey, in his review of Garzya in Gnomon 28 (1956) 91. 



336 ALCMAN'S PARTHENEION I RECONSIDERED 

(31). The issue has been reopened with the publication of the new 
papyrus, Partheneion II, which contains the word aydw. Marzullo 
contends that the term need not refer to a contest but may mean an 
assembly place, as it does in Homer. Peek (167) also decides in favor 
of the meaning 'place of assembly' and cites Pindar, Paean 6.60, in 
evidence. However, even though the spatial sense of aywv is attested 
earlier than the abstract sense of 'contest', it usually, even in Homer, 
refers to a place of contention rather than mere assembly. 56 Hence 
Barrett 57 is probably right in suggesting that the reference in Par­
theneion II is to a contest of some kind: "The surviving lines remind us 
strongly of the Louvre partheneion ... It seems from fr.1 that there 
will be others singing besides our chorus, in what was perhaps a 
contest of some kind ... " On balance, then, what with the supportive 
evidence of Partheneion II and the agonistic language of Partheneion I 
itself, the assumption of a rival chorus is not implausible, and causes 
fewer difficulties than the opposite hypothesis. 58 

The issue of 'amoebean' structure, specifically of the operation of 
half-choruses, or S,xopLa, has also been raised anew by the publication of 
Partheneion II. According to Peek (169ft) the new poem presents 
alternating half-choruses, although he admits that this interpretation 
is largely speculative. In actual fact, it seems to me, Peek does not 
exploit the evidence as fully as he might. His view that three successive 
strophes concern themselves with Astymeloisa makes him overlook 
the possibility that in one or the other of them the singers turn 

56 H. J. Mette, in B. SnellJH. Erbse, Lexikon des fruhgriechischen Epos (Gottingen 1955) colI. 
134-6, on &y~v: "Oer ay~v ist urspriinglich das Ergebnis eines aytw, so der &ywv 8EWV das 
Ergebnis eines 'Zuges', der die Gotterbilder auf die Akropolis usw. geleitet, der ciywv VEWV 
das Ergebnis eines 'Heereszuges'." In Homer and Hesiod the noun &y~v occurs only in the 
singular. Mette cites from schol. BI II. 18.376 to the effect that ay~v has five different 
meanings: aywv 8~ C1T}p.atvEL 1TEVTE· TOV T01TOV WS TO (8260), TO a8poLup.a (Y 33), TO 1TAfj8os 
WS TO (Q I), TO <1t1AOV ••• TOV vaov WS malitia Kal <H 298). Similar and derived statements 
are to be found in Et.Gen., Apollon. Lex. and Hesychius. But in spite of Mette's etymology 
and the variety of interpretations suggested by the scholiast, the passages from Homer 
assembled by Mette show clearly that in the majority of the cases ay~v signifies an 'area­
for-contention' or a 'body-in-contention'. Even some passages which Mette puts under 
the heading 'Ensemble, Versammlung' should be crossfiled under the heading 'body­
in-contention': e.g., when the golden automata of Hephaestus are said to move on their 
own from their house to the tlEwS &y~v and back, the expression may be due to the fact that 
, ... hen the gods assemble, there is always a stir and a contest of wills, at least in Homer. As 
for the &ywv VEWV (Mette's Bib), one is reminded of the racing ships that started out 
for the Sicilian campaign, Thuc. 6.32.2. &y~v, then, is a body swarming with activity and 
tension, or the activity itself. 

57 w. S. Barrett, in his review of POxy. XXIV in Gnomon 33 (1961) 683, 
58 Compare the same conclusion of Page 1951.57. 



THOMAS G. ROSENMEYER 337 

against Astymeloisa. If, e.g., in fr.3 PMG col. ii v.65 we read eX oE 'ToJv 
7TV'\€WV' €xoW'cx and imagine Astymeloisa's co-leader as the subject, we 
may set up the following amoebean situation. Vv.64ff: Astymeloisa 
does not answer me, but she with the 7TV'\€ci)Jl, like a star, steps out 
with slim legs, and her hair is beautiful. Vv.73ff: Astymeloisa is the 
darling of the people; if she were to take my hand, I would be her 
worshipper. Or again, the sequence might be the other way round; 
aJMyw v.76 (cf Partheneion 1.2) may be negative: though Astymeloisa 
is the darling of the people, I do not care for her, and turn to X. A 
number of interpretations along these lines could plausibly be 
suggested; certainty is obviously impossible. In any case there is no 
compelling reason to assume that Astymeloisa is the only girl praised 
in the parts of the poem that we have; and Peek's assumption that 
Partheneion II employs half-choruses is more easily defended if we 
are ready to assume the contrary. 

If we now turn to Partheneion I we find that there has been violent 
disagreement over the comparative status of Hagesichora and Agido. 
Page (1951.45-6) is of the opinion that "Agido is a kind of Chief 
Assistant to Hagesichora, a Second-in-Command." The terms, with 
their evocation of law and order in colonial Africa, are witty enough; 
but one wonders what the parallels for such a position might be, 
either in choral poetry or in ritual practice. Page's conclusion is based 
on a purely quantitative procedure, the counting of occurrences of 
names, which for want of other criteria must remain an unsatisfactory 
method of inquiring into dramatic subtleties. Furthermore, Page 
seems to me to be taking too much for granted in the last three 
strophes. If, for instance, he says that Hthe girl 'with the lovely yellow 
hair' in v.IOI is surely the same person as the girl with the 'hair like 
purest gold' in VV.5lff," this means that in Page's opinion either there 
was only one eligible blonde in Sparta, or, if we must assume rivalry 
between two choral leaders, the contest is between a blonde and a 
brunette. In actual fact, all girls, in archaic poetry, are blondes. 

Other critics who uphold the superiority of one girl over the other 
encounter similar difficulties. Davison (446), though recognizing that 
Agido is just slightly more beautiful than Hagesichora (she "wins 
only 'by a short head' "), says that since the KE'\7J~ at 50 is Hagesichora, 
aV'Ta at 45 must be she too, since in both cases the talk is of a horse. 
Van Groningen (348-9), while conceding that the fourth strophe is in 
praise of Agido and the fifth in praise of Hagesichora, nevertheless 



338 ALCMAN'S PARTHENEION I RECONSIDERED 

concludes that there is only one full chorus, of which Hagesichora is 
the (non-singing) leader, while Agido (also non-singing), though the 
superior of Hagesichora, is given very short shrift in the poem itself. 
Schwenn (306) similarly suggests that Agido is the priestess, Hagesi­
chora the choral leader. Bowra (1961.62-3) concedes to Page that 
Hagesichora is more important than Agido, "but of course it does not 
stop the choir from continuing to praise Agido." 59 

In view of the contradictory opinions held on this score, the simplest 
solution, viZ. that there are two half-choruses each with its non­
singing leader, alternately praising the leaders, may have something 
to recommend it after all. A detailed demonstration of this from the 
text will be given below. Here we may anticipate by stating that 
apparently Aleman wrote at least two different kinds of songs for 
'TTapfNvoL: (1) songs in which two half-choruses contend with one 
another (Partheneion I and perhaps II); and (2) songs in which he and 
the girls address one another. The conclusion that Aleman appeared 
in some of the partheneia in his own person is inescapable. In fr.26 
PMG, consisting of four lyric hexameters, the poet sings to the girls 
that he is too old to dance at their pace, and wishes he were a K7JPVAOS, 
flying along with his d:AKV6vES. 60 In fr.38 the girls sing of themselves 
as commending the (male) KL8apLUTd.s who is, if not Aleman himself, a 
stand-in for the artist. In fr.39 and possibly also in frAO we seem to 
have a acppay{s, which must mean that Aleman reserved solo passages 
for himself; cf Pollux 4.66, where we learn about the role and the 
position of the ucppay{s in citharoedic poetry, especially in Terpander's 
nomos. Fragments 39 and 40 are too brief to furnish assurance that 
they come either from a choral poem or a monody. But the fact that 
the poet is referred to in the third person in fr.39 and the parallel 
of the bird comparisons in Partheneion 1.87 and 101 argue for the 

6' For a survey of scholarly opinion concerning the identity and role of Agido. see 
Garzya 35ff. 

60 For a song in which lyrical utterances by an individual alternate with singing by the 
chorus. cf Bacch. Dith. 18 Snell. The first and third strophes are sung by the chorus. the 
second and fourth by Aegeus. There are no amoebean patterns of responsion. however. 
Whether the avowal of feebleness in Aleman fr.26 PMG is to be regarded as applying to the 
chorus master in his own person or to a character acted out, we cannot say. The gossip 
reported in Athenaeus 13.600f, fr.59 PMG. appears to suggest that Aleman managed to 
express his personal feelings and desires in his choral poetry. Page's criticism of Antigonus, 
p.41 PMG, seems excessive. It is true that Aleman's bird flies along with, rather than with 
the support of, the haleyon birds. But Antigonus recognized correctly that the popular 
notion of the haleyons' supporting the K7JP';;>\Os with their wings lends a distinct color to 
Aleman's picture, in spite of the difference in detail. See, however, G. L. Huxley, "Studies 
in Early Greek Poets, II: Aleman's KoAvfL{JwaaL," GRBS 5 (1964) 26-8. 



THOMAS G. ROSENMEYER 339 

choral alternative. Ibycus fragments 286 and 287 PMG are, on the 
basis of their metrics, generally thought to come from choral odes; 
and they allow for the expression of the innermost feelings of in­
dividuals. It is difficult to imagine how this could have been managed 
in any other way than through exchanges between singers. 61 

IV 
I may tum now to a selective commentary on Partheneion I, re­

stricting myself to discussion of those lines whose readings have a 
bearing, positive or negative, on the suggestion that the poem 
exhibits dichoria. 

COMMENTARY 

Lines 2 and 12: OVK] ... &Myw, and 7TlXpTJao/LES. It is tempting to suppose that 
these expressions of disinterest, which occur in two successive strophes, are a 
function of amoebean responsion; cf. Theoc. 5.112-5. If so, it would be interest­
ing if the heroes of the ancient war could be divided into two groups. Half­
chorus A would then be expressing its lack of appreciation for the heroes 
acclaimed by half-chorus B, and vice versa. But our text is too fragmentary to 
be sure of anything. For the suggestion that the Deritids and the Hippo­
coontids were praised by two different groups, see Stoessl95. 

Lines 40-3: &pw F' c.:Yr' aALOV, OV7TEP &/LW 'AYL8c1 /LaPWPETlXL 4>lXlVTJv. Page 
translates: HI see her (sc. Agido) like the sun, which Agido summons to shine, 
as our witness." Page is undoubtedly right in his view that 4>alV7]v is intransi­
tive. He is less convincing when he opposes the old view (Bergk, Diels, 
Jurenka; now also Treu62) that /LaPWpo/LlXL here equals /LlXPTVPW, i.e. <to testify 
in a court of law'. If, as Page himself assumes, the poem is performed in a 
night ceremony, what sense is there in saying that Agido summons the sun 
to shine? Further, if /LlXPWPETlXL meant <summons', we should expect the 
aorist of 4>alvw; cf. Ii. 9.240: apaTaL 8~ TaXLaTlX 4>lXvT]/LEVlXL 'Hw 8iav. Rather, 
Agido is our evidence that there is a sun; in the absence of the heavenly body 
we need somebody like Agido to remind ourselves of the sun's existence. 
Translate: HI see Agido as the sun; she is our witness of the sun's shining." 
The phrase is analogous to a lyric comparison; cf. Sappho's E/L7Tpl7TETlXL •.. 
£.OS' ••• /L~VlX, 96.7-9 LP. It is true that /LarrrVpO/LfXL in the sense of /LapTVPW is 

61 It is difficult to guess what kind of a song fr.16 comes from and whether the lines refer 
to Aleman himself. 

U M. Treu, Von Homer Zur Lyrik (Zetemata 12, Munchen 1955) 263-4. Cf now also P. 
lanni, "Agido ed Agesichora," RFIC 92 (1964) 59-65, who emphasizes the agonistic tenor of 
f«XprtJpop.a,. 

4-G·R.B.S. 



340 ALCMAN'S PARTHENEION 1 RECONSIDERED 

attested only once, in Plato, but even weak attestation takes precedence over 
faulty sense, particularly in the case of a poet like Aleman whose vocabulary 
is known to us only very imperfectly. 

Line 44: Page 50 n.l comments: «xopayos in v.44 cannot be Agido; for VtV 
would then have to refer to Agido, and VLV is never used thus reflexively." 
Strictly speaking the use here is not reflexive but 'anaphorical'. Chantraine63 

defines ''l'anaphorique JLtV" as the JLLV in a subordinate clause which refers to 
the subject of the main clause and cites a number of such cases in Homer. 
Schwyzer (190-1) remarks: «Die anaphorischen Pronomina der 3. Person 
stehen statt des Reflexivs im allgemeinen nur in Infinitiv- and Partizipial­
konstruktionen und Nebensatzen." He cites cases from the Iliad, the Odyssey 
and Herodotus. He even makes some allowance for JLLV in cases of reflexive 
reference. Thus there is no reason why vtV should not refer to Agido.64 The 
sense is "Agido is like the sun; but our respected leader does not permit us to 
praise her or to find fault with her." That is, her appearance is so self-evident 
that there is no need for us to enlarge on it, either approvingly or disapprov­
ingly (the latter is interesting, but not surprising if the context is amoebean). 
Some critics wish to make a distinction between XOPOUT(x:ns and xopayos. 
Since the former, as we shall see and as is generally agreed, refers to Hage­
sichora, the champions of Hagesichora's sole eminence in the poem are 
compelled to argue either that xopayos also refers to Hagesichora or that the 
meaning of xopayos is not 'choral leader'. Pluto Apophth.Lac. 219E, is our best 
evidence that the two terms are interchangeable: LlaJLwvlSas TaX8€tS EUXaTos 

TOV xopov lmO TOV TOV xopov iUTWVTOS, €ry€, €l1T€V, clJ xopayl, ~'€VP€S 1TWS Kat, 
• r , II """ I aVTTJ 7J xwpa CXTLJLos ovacx €VTLJLOS Y€V7JTCXL. 

Lines 45-9: cxv-ra is emphatic: by herself, without help from others in the 
form of a eulogy, Agido impresses everyone as outstanding. To compare the 
girl with a well-built horse is good Spartan language.65 Even more important, 
however, especially for the use of the masculines, is the Homeric model, 
Il. 9.123-4.66 As for tm01T€TpL8lwv ov€lpwv, it is at present impossible to say 
whether the meaning is 'winged dreams' or, as the scholiast suggests, 'dreams 
that live under the Rock.' Since the epithet seems to be largely ornamental, 
the point is not a crucial one. 

Summing up: strophe 4 is in praise of Agido. It is surprising that this should 
ever have been doubted, seeing that there is no mention of Hagesichora 
whatever, 

6S P. Chant raine, Grammaire homtrique II (Paris 1953) §228 p,54. 
U The view of d'Errico 26 that v,v refers to the sun need hardly be considered. 
85 Cf. Diels 355. 
81 "Part of the inherited stock-in-trade of the professional poet's vocabulary" (J. A. 

Davison, "Quotations and Allusions in Early Greek Literature," Branos 53 [1955] 139), or 
direct quotation? Because of the masculines and also for general reasons affecting the 
relationship between the Homeric poems and the archaic lyric, I would assume the latter. 



THOMAS C. ROSENMEYER 341 

L · 50 4· l' , < ~ <, "'E ,< ~ \ , ~, ~ , .t. ~ Ines -. 7] OUX Op7]ts; 0 JLEV KEI\7]S V7JTtKOS' a OE xatTa Tas EJLaS aVE'f'tas 
'AY7Jatx6pas £7rav8EL xpuaos [w]s &Xy]paTos. "Do you not see? The horse is 
Enetic; but the hair of my cousin Hagesichora blooms like unalloyed gold." 
Page, in spite of a touch of adversativeness in his translation,67 thinks that 
these lines are all complimentary to Hagesichora. But 'Enetic' is not necessarily 
complimentary, in spite of Devereux's recent argument68 that Homer's 
reference to Enetic range mules, Il. 2.851, implies a compliment to the mares 
that breed them. Our earliest direct evidence for Enetic horses being regarded 
as choice racing horses comes from the fifth century; schol. Eur. Hipp. 231, 
tells us that Leon of Sparta in 440 B.C. was the first to win an Olympic victory 
with an Enetic team. Now Eust. 361.10-40 in his comments on II. 2.852 complains 
that Euripides took his cue from Leon's victory to talk, anachronistically, about 
Enetic horses in the heroic past. The complaint is paralleled in the Euripides 
scholion, which also reports that the Eneti had originally lived in Paphlagonia 
before they settled on the Adriatic. Hence, Devereux argues, we must dis­
tinguish between two varieties of Eneti, those of Paphlagonia, who breed 
fine mares for the production of mules; and those of the north Adriatic. 
Since the horses of the latter, as we are told by several sources,69 are harness 
horses rather than riding horses, and since they are more distinguished for 
their speed than for their looks, Aleman must (Devereux concludes) be 
thinking of the Paphlagonian Eneti. But in spite of Devereux's careful sifting 
of the sources, his case for a well-known breed of Paphlagonian Enetic mares 
is circumstantial at best.70 The case for Adriatic Enetic harness horses is much 
better attested. Once Devereux's premise, that Aleman is paying a compli­
ment, is challenged, the difficulties ofinterpretation disappear. What is wanted 
is a reference to a horse that is neither overly attractive nor primarily a riding 
horse. Thus the Adriatic species fits excellently; that there are no other pre­
classical references to the breed is an argument from silence which this very 
passage should be allowed to demolish. In effect, the girls are saying "Don't 
you have eyes to see? The noble steed (of whom you sing)71 is-an Enetic 
pony. But our cousin Hagesichora ... " The strophe, then, starts with a piece 
of chaffing: Agido, by comparison with Hagesichora, whom the chorus is 

67 Page 1951.22. Marzullo 195 n.3 apparently misunderstands Page's meaning. 
68 G. Devereux, "The Enetian Horse of AIkman's Partheneion," Hermes 94 (1966) 129-34. 
69 See especially John K. Anderson, Ancient Greek Horsemanship (Berkeley 1961) 37, who 

cites Strabo 5.1.9. Strabo is not explicit in relating his tale of the wolf-brand horses to the 
Enetic breed, but Anderson is probably correct in making the connexion. 

70 W. S. Barrett, in his edition of Eur. Hipp. (Oxford 1964) 204, appears to favor the 
Adriatic over the Paphlagonian identification, though he does refer to Page's doubts on 
this head, Page 1951.88: "It is easy to presume familiarity with Paphlagonia, difficult to 

find traces of acquaintance with the far northern Adriatic." 
71 The warning of Jurenka 1896.14 that the article in v.50 is demonstrative is worth 

heeding. His further comment that K'ATJS is part of the predicate is not equally binding. 



342 ALCMAN'S PARTHENEION I RECONSIDERED 

about to praise, is a Connemara pony, fast perhaps, but small, half-wild and 
not entirely attractive.72 

L" 55 7 "" , ~ .I..'~ , \ , <A ' , mes -: TO T apyvpLov 7TPOUClJ7TOV OLa.,.aoav TL TOL /\EyW; rrJULxopa fLEV 
av-ra. HHagesichora's face is silver. But why talk about it explicitly? Here 
she is before you!" This sequence is the pendant to the praise of Agido at 
40-5. Like Agido, Hagesichora does not need to be praised in detail since her 
beauty is openly in evidence. Further, Kukula (208) may well be right that 
&pyr5PLOV implies a reference to the moon, which would thus answer to the 
mention of the sun at 41. But even without this, the correspondence between 
the two eulogies is clear. Page's translation "So much for Hagesichora," is mis­
leading; his own citations (1951.89) of Pi. Ol. 4.24 and Dem. 18.282 show that 
his understanding of av-ra is not exactly what is indicated by his translation. 
The pronoun is in fact used to emphasize the presence of persons near the 
speaker; cf the examples given by Kuehner-Gerth.73 

Lines 58-9: a o£ oEVTEpa 7TEO' 'AYLOc1 TO FE'iOO~ 'l1T1TO~ , If3TJvwL KOAaea,o~ opa­
fL~TaL. Page continues to feel that 'AYLow must be accusative,74 in spite ofthe 
apparent view of the scholiast B (POxy. 2389 fr.6 col. i lines 2-4) that it is 
nominative. The reading of the papyrus is such that it favors the scholiast's un­
derstanding of the passage:7 5 "The second one after her (viZ. after Hagesichora) 
in looks, Agido will run, a Colaxaean horse with an Ibenian." The scholiast 
says that according to Aristarchus an Ibenian horse is better than a Colaxaean.76 
There are here two possible points of comparison: (1) beauty, (2) speed, We are 
told that Agido is less beautiful than Hagesichora. There is to be a race ;77 

will Agido also be slower than Hagesichora? The answer to this depends in part 
on the interpretation of 7TEO' and of the dative 'If3TJvwL. The suggestion of 
Garzya (52) that 7TEO' derives from 7TEoopafL~TaL does not strike me as plausible; 
he himself notes that fLETaTpEXw in the sense of 'run with' is not attested. The 
position of 7TEO' would seem to indicate that it is to be taken adverbially with 
oEVTEpa, and Page acknowledges that the scholiast so understands it: 'second 
after'. This leaves 'If3TJvwL as a dative associated with opafL~TaL. On the analogy 
of fLaXEufJaL with the dative it has been suggested that the reference here is to a 

71 Aleman fr.l72 PMG is too uncertain textually and in its ascription to be useful. 
73 R. Kiihner/B. Gerth, Ausfohrliche Grammatik der griechischen Sprache' I (Leverkusen 

1955) 645. 
H D. L. Page, in his review of POxy. XXIV in CR 9 (1959) 18; cf also PMG p. 7. 
76 See also Garzya 51 and Marzullo 196-7. 
76 G. Devereux, "The Kolaxaian Horse of Aikman's Partheneion," CQ 15 (1965) 176 n.4. 

suggests that AIkman derives the Kolaxaian horse-and also the dream horse of lines 
46-9-from Aristeas' Arimaspea. Cf J. D. P. Bolton, Aristeas of Pro co nne sus (Oxford 1962) 40f. 

77 Cf the references to racing contests given above, pp. 333f. Davison 446 asks whether the 
two girls are really thought of as running in a race, or whether the idea of the horse 
simile makes for this further image. He decides for the latter. But it is worth remembering 
that at many Greek festivals the musical agon is followed by a gymnic and an equestrian 
agon; for the Great Panathenaea the evidence was collected admirably by Davison him­
self in]HS 78 (1958) 25ff. 



THOMAS G. ROSENMEYER 343 

contest between two girls compared to horses. But there is no other instance 
of Spal-'Etv or its derivatives with the dative. Hence we are on our own. The 
interpretation will be affected by the understanding of the subsequent 
vv.60-3 about the Peleiades. For my own purposes it does not make much 
difference whether Agido is thought of as running with Hagesichora or against 
her (though my discussion of 60-3 will show that I favor the former).78 We 
must, however, insist that the passage is concerned entirely with these two 
girls and no others. The notion that Sr:.VTEpa may refer to a third girl (Bergk, 
Kukula, Scheidweiler [244-5], Page [1951.89-90], Barrett79) is not warranted by 
anything in the text and is unlikely ex hypothesi. It is in fact an idea that was 
prompted only by an uncertainty about the roles of Hagesichora and Agido. 

Lines 60-3: 'Tal. llr:.A'Y}aSr:.s yap al-'Lv opOplaL rpapos rpr:.poluaLs VVK'Ta SL' a/LjJpo­
ulav aTr:. mlPLOV aUTpov aF'Y}pO/LEvaL ~x6V'TaL. Page80 examines the various 
possibilities of interpreting llr:.A'Y}&.Sr:.s: 

(1) Hagesichora and Agido (Wilamowitz, Schwenn; cf also Farina [39] and 
Garzya [52-5]). This seems to be the opinion expressed in scholion XII and 
perhaps also in scholion B (POxy. 2389 fr.6 col. ii lines 16ff), although the 
scholia are too fragmentary to be very helpful. Garzya's position is inter­
esting. For aF'Y}pOI-'EVaL he cites Il. 6.289ff and translates: HAnd in fact these two 
doves struggle (with the rival chorus) in the ambrosial night, lifting up for us, 
who offer it to Orthria, the cloak (shining) like the star Sirius." Garzya is, 
I believe, alone in withholding the term llr:.A'Y}&.Sr:.s from the rival chorus 
without at the same time giving up the notion that there was a rival chorus. 
Further, he argues that aF'Y}pol-'EvaL cannot be intransitive and hence cannot 
refer to the rising of a constellation or star; and that Sirius is compared to the 
cloak rather than to the llr:.A'Y}&.Sr:.s. His interpretation is vulnerable on two 
points: (a) his interdiction of the intransitive use of ar:.{pO/LaL is excessive, 
especially in the light of Herodotus' frequent use of ar:.tpol-'aL in the sense of 
'depart'; (b) Garzya fails to note that in the poetic tradition Sirius usually 
has ominous or uncomplimentary connotations ;81 hence the image is less 
likely in connexion with the gift to the goddess than as a characterization of a 
rival chorus. 

(2) The constellation (Weir Smyth, van Groningen [250-1], Davison [449J; 

cf d'Errico [26--7] and Burnett passim). Davison argues that there is no evidence 

78 Cf now also G. Devereux (supra n.76) 177: "The Ibenian and the Kolaxaian horses are 
not racing against each other, but are running jointly ... " 

79 W. S. Barrett in Gnomon 33 (1961) 686. Barrett is very hesitant about this solution. 
80 Page 1951.52fI. For the Pleiades generally consult Gundel's article in RE 21.2 (1952) 

2486-2523. 
81 Hesiod connects the star with deadly drought. Homer does not refer to Sirius by name, 

but his ovA,os riaT'!/P (II. 11.62) seems to have been identified as Sirius, if we can trust the 
instincts of Ap. Rhod. Arg. 957-9. For a recent survey of the evidence, see K. Gantar, 
"Kaiser Justinian 'jenem Herbststern gleich': Bemerkung zu Prokop Aed. I 2, 10," MH 19 
(1962) 194-6, who in turn refers to Gundel in RE ZW.R. 3A.I (1927) coll. 314fI. 



344 ALCMAN'S PARTHENEION I RECONSIDERED 

of a rival chorus and that p.aXEu8txL can be used only of competitions which 
involve actual fighting, such as boxing and other contact sports. Neither point 
is very strong. Though there is no explicit evidence for the rival chorus in the 
poem, there are hints of its existence; and as Page has said (1951.57), "If 
< direct evidence' must be nothing less than the explicit statement that a Rival 
Choir exists, it is neither possible to find it nor rational to look for it." Davison's 
point about p.aXEu8txL is too rigorous and has convinced few. But the main 
objection to (2) is that it would be absurd to compare the constellation 
Pleiades to Sirius. The point of a comparison is that it connects members of 
different classes.82 Hence varying assessments of the brightness of Sirius will 
not suffice to vindicate (2); nor will the theory, derived from ProcIus on Hes. 
Erga 417, that EE{pLOS could be used as a name for the sun. 

(3) The rival chorus (Bowra, Page, Scheidweiler and others). Schol. Theoc. 
13.25 p.262 Wendel (cf Call. fr.693 pf.) reports that the <Peleiades' were 
the daughters of the queen of the Amazons and that these were the first, as 
girls, to establish dancing and a 7TtxWVX{S : 7TPW'TOV S' txVTaL XOpE{txV Ktx' 7TtxWVx{Stx 

avVEG'7'1}UavTO 7Ttxp8EVEVOVUtxL. The scholion goes on to connect the rising of the 
Pleiades with the quickening of spring, which is further proof of the impossi­
bility of associating Sirius poetically with the constellation. Finally, the scholion 
names the seven Pleiades; the reminder that the Pleiades constitute a sizable 
group is further proof that the two leaders are not likely to be so designated. 

On balance, then, identification (3) remains the most plausible. The 
reference is to a group of dancers who are engaged in a contest with the girls 
who sing this hymn. They rise like Sirius: an uncomplimentary remark 
which confirms the agonistic setting. Whether Agido and Hagesichora will 
do their racing together or in competition with one another is something 
else again. If their racing is at the same time an offering of the 4>apos, which is 
by no means more certain here than in the description of the Heraea by 
Paus. 5.16.2-5, they will have to be imagined as running together. And in 
fact yap v.60 suggests a united effort on the part of the two leaders. I am not 
sure how to explain the change of tense between SptxP.-r}'TtxL at 59 and fLtxxovrtxL 

at 63. Perhaps it would be proper to say that SptxfL-r}'TtxL refers to a specific 
action which will take place directly, while fLtxxovrtxL refers to the general 
reality of the conflict between the two rival choruses; for the latter, if. Il. 
5.875.83 Thus vv.58--63 will have the following meaning: "When Agido runs 

81 I do not understand why Page 1951.53-4 finds this argument of little weight. For the 
rest, Page's discussion of the identity of the Peleiades, pp.52-7, is magisterial. Cf also his 
reference to Paus. 5.16.2-5 for a group of sixteen women racing one another in honor of 
Hera. They also offer a robe to the goddess and split into two (half?) choruses. 

83 At the Panathenaea, as at other such festivals (e.g. the Theseia) the agonistic order was: 
(1) music, (2) athletics, (3) horse-racing. The evidence comes from Pollux 8.93, Arist. 
AthPol ch. 60, and several incriptions. J. A. Davison (supra n.77) pp.26ff gives the details; he 
seems inclined to believe that many of the contests formed part of the annual Panathenaea 
from very early times. 



THOMAS G. ROSENMEYER 345 

with Hagesichora, as she will later today, she will be second to her in beauty, 
as a Colaxaean is to an Ibenian horse. (They will run) for the Peleiades are 
contending with us, rising like Sirius (or: like a searing star) through the 
ambrosial night, as we take the 1>apos to Orthria." 

Summing up: strophe 5 is in praise of Hagesichora, with some allowance 
for cooperation between Hagesichora and Agido in the face of a rival chorus. 

L " £A ·76·'" ../.. ' , ,."" " !\ mes u--r- • OUTE yap n 7Top'f'upas Toaaos KOpas WU'T afLuvaL, OUT€ 7TOLKIJI.OS 

~ , "<:' " A <:" ,~, \ 71.T ~ , ,\ \' '[<:"] opaKWV 7TayxpvaLos, OVOE fLtTpa VOLa ... OVOE TaL .LVaVVWS KOfLaL, a"" OV a 

, ApETa OLEL8~s ... Bowra's characterization (1961.63) is typical of the usual in­
terpretation: "The choir proclaims its deficiencies ... They have no purple, 
golden armlets, or Lydian coifs, nor can they summon to their help girls 
renowned for their beauty." But it is strange that a chorus of girls, instructed 
by Aleman, should complain of not having enough in the way of ornaments 
or good looks to compete with their rivals. Kukula84 protested against this 
notion, but his own suggestion, viZ. "there is nowhere enough purple, etc. 
to be a match for us," is hardly more convincing. Before the passage can be 
understood, we have to deal with three specific problems: (a) the meaning 
of KOpas; (b) the meaning of eXfLvVat; and (c) the function of WaTE. 

(a) Page comments (1951.90): "KOpas: rather 'abundance' than (as usual) 
'over-abundance'; cf Hes. Th. 593; Heraclitus fr.ll1 D." Page does not 
consider a third possibility, viZ. 'satisfaction', i.e., the pleasurable gratification 
of a physical desire. But this is in fact the common meaning of KOpOS and 
KopEvvvfLat in Homer. Usually the source of KOpOS is food (11.19.167; Od. 14.28), 

but there are many possible grounds of satisfaction, and the three uses of the 
noun (as distinguished from the verb, which is more common) are associated 
with grief (Od. 4.103), fighting (Il. 19.221), and with sleep, love, singing, and 
dancing (ll. 13.636): 7T(XV'TWV fLEV KOpOS eaTl, Kat V7TVOV Kat c/>L>"077JTOS I fLo>"7rijs'T€ 

y>"VKEpfjS Kal, eXfLvfLovoS 0PX7JOfL0'iO. It is true that in later lyric poetry KOpas 

comes to have a meaning close to vfJptS; cf alav~s KOPOS, a persistent or 
obsessive state of being puffed up, at Pi. Ol. 2.95 and 1. 3.2. But the Hippo­
cratics continue the use of the term in the sense of pleasurable satisfaction of 
the moment: most foodstuffs are compounded and unhealthful, ;gw TWV 7TPOS 

-,]8oV7}v T€ Ka~ KOPOV ~PTVfL'VWV TE Ka~ eaKEvaapivwv (De vet. med. ch. 14 line 50 
Jones). Here KOpas goes closely with ~SOV7}, and designates 'pleasure in and 
beyond satisfaction'.85 The most plausible interpretation of 64-5, therefore, 
is: "I do not take as much pleasure in (or: get as much satisfaction from) 

at Kukula 211. Marzullo's explanation of the passage, pp.202-3, is colored by his thesis 
of an erotic connexion between Agido and Hagesichora. Cf also Diels. 

8& The sense is anticipated in Hes. Theog. 592-3: he says of women that they 
1ffjp.a p.lra 6V1JTO raL P.ET' av/)paaL vaLETaovaLV 
ouAop.I,,?]. 1f€VI?]. ou aup.cpopoL, aAAa K6pOLO· 

they are a help, not in poverty, but in pleasure. The comment is bitterly ironical, since 
sexual pleasure is not one of the commodities which Hesiod counts among the things that 



346 ALCMAN'S PARTHENEION I RECONSIDERED 

crimson ... " On this interpretation, crimson material turns out to be not a 
desired stage prop but an imagined source of pleasure, comparable to other 
sources of delight such as pins, hats, and-girls. 

(b) al'QvCtt. Page 1951.52 n.1 refers to schol. Lips. Hom. It 5.266 (=Eust.ll. 
546.29): r/JTJui. ya.p 0 YPCtI'f'Ct'TtKOS .' AptCT'Tor/JaV7JS (cf. Nauck Ixi p.213) 'TO 
al'Vv€u8Ctt o~ I'0VOV C17JI'Ctlv€tv 'TO KCtKWS '1TCt8oV'TCt aV'Tt8tCt'Tt8£vCtt, llia. 'Tt8€iu8Ctt 
KCti. aV'Ti. .ptAoO 'ToO al'€l.pCtu8Ctt O'Ttovv. Aristophanes cites as examples of this 
meaning the present passage from Aleman (reading al'QvCtu8Ctt rather than 
al'QvCtt) and Thuc. 1.42.1. Page professes not to understand Aristophanes' 
comments. The Thucydidean passage, however, seems to supply a possible 

I I , i: I ~. I • ~ , , 8 "I h cue: V€W'T€POS 'TtS ... CtstOV'TW 'TOtS 0I'Otots TJI'CtS CtI'VV€U Ctt: et t e young 
men consider it proper to treat us as we have treated you." The force of 
al'Vv€u8Ctt here is <to respond', to react to a stimulus. So when Aristophanes 
compares al'Qv€u8Ctt to al'€t{J€u8Ctt, he appears to think of the simple Homeric 
usage <to answer', or <to take one's tum'.86 Hence the deprecatory .ptAov and 
the generalizing O'Ttovv. It is true that Aleman has the active rather than the 
middle; but both occur in Homer, and I see no difficulty in applying what 
Aristophanes says about al'Qv€u8Ctt to al'QvCtt. In Homer, the response is 
always a military response, i.e., a defense. But it may well be that in its 
origin al'Qvw is not restricted to the field of battle. The variety of con­
structions (with acc., with dat., or with a preposition) indicates that al'Qvw is 
not, in essence, a verb that requires an object.87 Availing myself of Aristo­
phanes' views, I propose to translate al'QvCtt as <answer or <respond'. The 
chorus talks of itself not as defending itself against the rival chorus but re­
sponding to the sentiments expressed in the previous strophe, hence re­
sponding to the other half-chorus. The use of al'QvCtt, then, is not agonistic 

benefit society. I prefer this interpretation to Page's translation of KOPOS by 'plenty'. Hera­
clitus fr.65 D. presents an interesting problem: KaAt"' 8~ av-rd XfYTlCTp.ouVVTjV KalKopov. XfYTlCT­
p.oCTVvq 8~ E(Fr£V 1] 8taKOCTp-T/CTtS KaT' av-rov, 1] 8~ EIC1TVPWCTtS KOpOS. In frgg. 67 and 11 KOPOS is 
contrasted with }o.,p-os. Both }o.tp-OS and XPT/CTP-0uVVT/ signify 'hunger' or <need'. G. S. Kirk. 
Heraclitus, the Cosmic Fragments (Cambridge [Eng.] 1954) 357-61, seems to imply that KOpoS 
was used variously by Heraclitus to mean (1) fulness and (2) overfulness (his own transla­
tion 'satiety' does not really fit the situation as he describes it). But he admits that any­
thing we might say on this score is speculative. In view of its contrast with Atp-OS, however. 
I consider it more likely that Heraclitus thought of KOpOS as denoting a phYSiolOgical or 
psychological experience rather than a condition or a state. To demonstrate this, a de­
tailed analysis of Heraclitus' series of opposites would be required. 

86 Cf. also Simonides' use of ap-wt"CT8a" fr.6U PMG, which the Suda 1.148 Adler glosses: 
avrl TOU X&p'Tas a'ITo8,80vat. 'to return favors'. See also the entries in Hesych. 1.3856 and 
3860, p.134 Latte (gloss. Cyrill.). according to which ap-wt"w and ap-wt"CT8a, equal cmo8,86va, 
or aVTa'lTo8,80va,. 

87 The old view that a"ww and &.p-t"t{3w are etymolOgically related. i.e., that both of 
them derive from IE *meu (cf. Latin movere), has now been abandoned. For references. 
see H. Erbse, in LexfrgrEpos (supra n.56) colI. 619 and 650. 



THOMAS C. ROSENMEYER 347 

but amoebean. But what is the connexion between ToaaoS" KOPOS" and &/Lt5vaL? 
This depends on the translation of WO'TE. 

(c) WO'TE. It is important to realize that WO'TE cannot be the consequent 
of Toaaos, because of what follows. If for the moment we dismiss warE &/Lt5vaL 
from consideration, the rest reduces to a simple though grammatically harsh 
praeteritio: "We do not get as much satisfaction from scarlet material ... 
nor does a golden dragon pin (please us as much), or a Lydian toque ... or 
Nanno's hair, or the divine Areta, or ... " It is true that beginning with 
7TOLKl>..OS" BpaKwv we have nominatives rather than genitives, so that the initial 
construction based on KOPOS" is side-tracked. But it is generally agreed that the 
sequence is a praeteritio and that its items parallel one another. Beginning with 
v.73 Aleman relaxes the pattern of the sequence even further by introducing 
the house of Ainesimbrota, with its charming inmates: ovB' ES AlV7'}aL/Lf3p[ 0]­

TaS" ev80iaa c/>aaEts- 'Aarac/>lS" [T]lf /LOL ylVOLTO Ka~ 7TOTLy'M7TOL tPD..v>J...a ... 
Thus, instead of continuing the catalogue of ladies who do not measure up 
by comparison with someone else not yet mentioned, Aleman introduces a 
touch of concreteness and refracts the line: "Nor will you go to the house of 
Ainesimbrota and say: Let Astaphis be mine, and let Philylla look at 
me ... "88 In spite of these resolutions, the underlying thought remains the 
same; everything in the extended praeteritio points to the final acclamation 
to which all such series move. The initial 'Too-aoS' points toward an ultimate 
($uuov ('we do not get as much satisfaction from Y as from X); the sub­
sequent independent statement points toward an adversative statement 
('you will not say Y should look at you, but you will hope for a glance from 
the beautiful X'). It goes without saying that the girls mentioned in vv.70-6 
are all beautiful. They are the acknowledged beauties of the Spartan jeunesse 
doree, analogous to the various handsome youths eulOgized in some of Socrates' 
introductory gambits. They are mentioned because along with crimson 
garments and gold brooches they share the quality of desirability.89 At the 
same time, like the wardrobe and the jewelry, they are supposed to pale into 
insignificance by comparison with the X toward whom the praeteritio is 
pointed. It does not much matter precisely what kind of a house Ainesimbrota 
ran; Page (1951.65) may well be right that she was the keeper of a training 
school for chorus girls, though there is no evidence for this whatever. But we 
can be sure that the girls mentioned are not in the chorus; it would be 
awkward in the extreme if the members of the chorus (or half-chorus) were 
to refer to themselves for the mixture of praise and deprecation required in 

88 The second person .paa,(is has generalizing force; cf Schwyzerp. 244 (B iv 3 am. Through­
out the lines preceding 73 the singers avoid the first person, as if the praeteritio were headed 
for expressing a universal rather than a personal value. 

F~ Cf. also Plato, Hp.Ma. 287Eff. Hippias instances young girls and gold as prime examples 
of beauty. 



348 ALCMAN'S PARTHENEION I RECONSIDERED 

a praeteritio. Granted that this is the drift of the passage, that is to say, that 
TOCTCTOS' anticipates a OCTCTOV (which, to be sure, is never realized), then ~CT'T' 
ap.vv(u can be nothing more than a parenthetical phrase, analogous to such 
expressions as wS' £l1T£LV, and meaning 'to respond'. True, WCT'T£ introducing 
an absolute or limitative infinitive is not paralleled, but there is some suggestion 
that the limitative infinitive is in its origins consecutive.90 This is shaky ground. 
My argument requires either that WCT'T£ was at one time equivalent to wS' with 
the limitative infinitive, or that a T was added in the text of Aleman when it 
was felt that TOCTCTOS' needed a consequent. An additional difficulty is that 
limitative infinitives are more common in prose than in poetry and that 
the first examples occur considerably later than Aleman. But it is probably 
fair to assume that at a time prior to the writing of literary prose, poetry was 
less self-conscious about excluding the patterns of living speech. What is 
more, this would not be the only colloquialism in Aleman's verse; cf. the 
parenthetical F£{1TOtp,l K' in v.85. In any case, the interpretation here suggested 
seems to me less awkward than the various uncertainties which result 
from the notion of a defense and a victory being read into the thought of 
the passage. 

Line 77: aM' <AY11CTtxopa p.£ T£{p£t. Ever since Lobel warned Page that the 
papyrus reads not n7P£L but T£{p£L scholars have been in a quandary about the 
sense of the line. Page (1951.91) simply confesses that he does not under­
stand; his reference to Hes. fr.105 Rz. does not help since in Hesiod the subject 
of the verb is not a person but €pwS'. d'Errico (31) proposes the translation, 
«We are worried about Hagesichora." He understands the whole strophe as 
follows: "You, the girls of Hagesichora, will have a difficult stand against us, 
since you do not have a Nanno, or an Areta ... ; only Hagesichora, your 
leader, has us worried." But like the interpretation of Kukula (supra n.84), 
d'Errico's does not convince; the strophe is not about running, but about 
pleasure and beauty. Harvey91 decides that emendation is the only recourse 
and proposes p.' i(}£{p£t. Garzya (1954.62) ventures to disregard Page's warning 
and to read rrJP£L after all. The only scholar who accepts and translates T£{p£t 

without blinking is Bowra :92 "The word means 'wears out' ... " But then, 
because of his initial assumption that the strophe is in praise ofHagesichora, he 
concludes that the phrase means: "We are worn out with praising her; there 
obviously is nothing more to say." I suggest that we should take T£tp£t quite 
literally. In Homer, Tdpw can be predicated of a wound (II. 16.510) or a pain 
(Il. 16.61) or old age (Il. 4.315) or sweat or hunger or bad smells; or even of a 

.0 Cf Schwyzer 379; also Gottfried Herman, Opuscula I (Leipzig 1827) 227, cited by A. 
Roseth, "Die Entstehung des absoluten Infinitivs," AJP 43 (1922) 220-7. Roseth himself 
has a different view of the matter; he holds that the absolute infinitive comes about by 
ellipsis, when an auxiliary verb of possibility drops out. 

11 A. E. Harvey, in his review of Garzya in Gnomon 28 (1956) 89-90. 
til Bowra 1961.61. But cf also M. Treu, in Gnomon 26 (1954) 170. 



THOMAS G. ROSENMEYER 349 

depressing song (Od. 1.342), i.e., of anything that does not please and positively 
displeases or even disgusts. T€tPO/L(U, then, is in some respects the opposite of 
KOpEVVV/LCt.L. Homer does not give us a person as the subject of T€lpw, and in 
fact Aleman's usage may be the only one of its kind. But it is clear that the 
word was available to express the notion of violent displeasure. Ar. Lys. 960 

has the word in a comic context; the chorus of old men sympathize with 
Cinesias after Myrrhina has disappointed his expectations: €v 8€tvwt y' cL 8VO"T"f}v€ 
KCt.KWt T€lpEL tf;vx~v €gCt.7TCt.'T"f}(1Els. I suggest that Aleman uses the word, extrava­
gant and usually impersonal as it is, in order to emphasize what turns out to 
be an egregious anticlimax. "But Hagesichora grates on me" comes to be 
substituted for the eulogy of X toward which the praeteritio has ostensibly been 
moving. Hence the eulogy would have been H Agido is the loveliest." Hence, 
the strophe is sung by the half-chorus of Agido. The substitution of a taunt of 
Y for the praise of X is an amoebean technique. Such cX7Tpou86K"f}TCt. are not 
uncommon in comedy and the pastora1.93 It makes particularly good sense 
in an atmosphere of intimate banter, in which insults are not taken amiss and 
in which the allegiances of the participants are evident without the need for 
the kind of information which modern theater audiences must have. The 
girls led by Agido can depend on their commitment to Agido being under­
stood by all; taunting Hagesichora is immediately recognized as a direct 
outcropping of this commitment. 

Summing up: strophe 6 is, by indirection, in praise of Agido. This contention 
is further strengthened by the opening of Strophe 7, which is unambiguously 
and argumentatively in praise of Hagesichora. 

Lines 78-81: OV yap a K[Ct.]N\lu4>vpos 'AY7Jutx[o]p[Ct.] 7TCXp' Ct.VT€L, 'Ayt8oL •.. 
OwunJp[ux T'] a/L' €7TCt.tV€L; The yap makes it likely that this rhetorical question 
is a rejoinder to the implications of the preceding lines. The sense is: "What 
you say is nonsense; does not Hagesichora, she of the fine ankles, stand 

93 Cf Theoc. Id. 7.125; "May Molon choke himself with this tussle" has some of the 
same force, Molon being introduced in order, by negative implication, to wish Aratus 
a better fate. Partheneion 1.77 is unusual in that it combines two related forms of humor: 
(1) the cX7rpoa80KTJTov, and (2) the terminal punchline, which often operates with the element 
of surprise and equally often introduces a note of vulgarity. In pastoral poetry, the punch­
line terminates either an individual song or the whole poem. It is usually the winner in 
the contest or the dominant character who achieves this break-through. See Theoc. Id. 
1.151-2; 4.58, 63; 5.149-50; 7.126-7; 9.35-6; 10.67-70, and the comments of E. Fraenkel, 
Horace (Oxford 1957) 60-1. For surprise endings in modern pastoral poetry, cf Ronsard's 
Adonis (Venus forgets Adonis and takes up with Anchises), and Robert Frost's New Hamp­
shire: «At present I am living in Vermont." As for (1), the Tractatus Coisl. mentions the 
€K TOU 1Tapa 1TpoaooKlav as one of the sources of laughter. Lane Cooper, An Aristotelian Theory 
of Comedy (New York 1922) 249-50 gives examples and refers to Arist. Rhet. 3.11.1412al9ff. 
See also W. J. M. Starkie, The Acharnians of Aristophanes (London 1909) lxvii f, who criticizes 
Aristotle for defining the technique as pertaining to Y'AWS a1Tl~ T<VV 1Tpayph.TWv, which 
explains why Aristotle does not analyze the stylistic phenomenon involved. Starkiegives a 
list of examples from Aristophanes. The book on the rhetoric of humor in Aristophanes 
remains to be written. 



350 ALCMAN'S PARTHENEION I RECONSIDERED 

right here to disprove your position?" What follows is fragmentary, but it 
is generally conjectured that there is some talk about Hagesichora standing 
next to Agido and commending the 8wcrrljpux. As I have indicated above, there 
is no need to assume that the two leaders are acting as priestesses in a 
religious service. The emphasis on the close association of the two leaders 
is probably to be seen as a parallel to the joint (or competitive) running 
projected in vv.58-9. Here, as there, their very closeness will make the 
superiority of Hagesichora even more obvious. In the sequel, vv.82-4, there 
is some allowance for joint action, very much as in vv.60-3; whatever the 
differences between the two leaders, their stand vis-a-vis the gods is united, 
as is that of the whole chorus. 

L o 8A '91 [] , , ,. [!l' , " 8' , •• lnes .,.- : XO poa-ru:ns, F€L7TOtP.t 1(, fi JYWV P.fiV aV'Ta 7Tap €VOS p.a'Tav a7TO 
8 ' \ '\ \, i: ' '[] "" - \ 'A' -' \ '" , ,-, , pavw I\fil\aKa yl\avs' fiyW v O€ 'Tat f.l,€V wn p.Wlta-ra FavocxVTJv €pw' 7TOVWV yap 

.,. , , " 'i: rA '[] '" • '''' ['] , '[ -] • 'fJ ap.tv La'TWp fiyfiV'TO' E~ yrJUtXOp as O€ VfiavtOES tp 7lvas fipa'T a S €7Tfi avo 
For xopoa-ra'TLS, see above on v.44. In what follows, the girls, speaking in the 
choral singular, compare their own singing to the hooting of an owl from 
the rafter. The emphasis is on ath-a, here in the sense of 'I alone', without 
the choral leader who is addressed. That this leader is Hagesichora is clear 
from 90. For a fuller understanding of the image of the owl we must wait 
for an examination of vv .100-1 below. The praise of Hagesichora is interrupted 
by a further mention of the goddess who is the object of devotion, Aotis, who 
has in the past been a saviour divinity.94 We do not know whether the 7T()VOL 

of v.88 are connected with the filp-r]VTJ of v.90, whether, that is, the divine 
salvation from troubles and the peace upon which Hagesichora has launched 
the girls are one and the same. For a discussion of the difficulties and of 
past attempts to cure them, see tke comments of Page (1951.93-5). Page 
himself declares for a contrast between the (general) peace owed to the 
goddess and the (particular) victory owed to Hagesichora. Garzya (66-9) 
prefers to think that Hagesichora is thought of as an instrument of the 
goddess and that both 7T6vot and €lp-r]VTJ are without reference to a specific 
past incident but designate effort and achievement in agonistic endeavour. 
Certainty in this is impossible; at any rate the singers ascribe some function 
in achieving peace to Hagesichora, and I consider it likely that the reference 
is to a past victory in a festival such as the one in which they are now engaged.95 

Summing up: strophe 7 is in praise of Hagesichora, with some allowance 
for joint action between Hagesichora and Agido. The pious remarks about 

94 Garzya 67 is probably right when he protests against the notion of Bowra 1936.58 
(tacitly corrected in Bowra 1961) and others that the troubles referred to are the second 
Messenian War. But Garzya's own proposal (68) to interpret both £y£VTO v.89 and brl{Jav 
v.91 as gnomic aorists breaks down for want of parallels. 

96 Cf Davison 453. Garzya 68 rightly compares Pi. 01. 1.98: fL"At.TO"aaav w8lav. I 
wonder whether Ar. Eq. 579 is not at least ambiguous, that is, whether the reference is not 
as much to the dramatic competition as to any military promises. Cf also Thesm. 1147. 



THOMAS G. ROSENMEYER 351 

the gods and Aotis carry us back to the respectable sentiments that had, from 
all we can gather, served as the keynote of strophes 1-3 and the beginning 
of 4. Whether this is connected with the fact that the poem is drawing to its 
close must remain a matter for speculation. If the papyrus did not show the 
coronis against the fourth line of col. iv, it would hardly have occurred to 
anyone that the poem should terminate so abruptly. For, as I will try to show 
next, the last strophe is no less amoebean in spirit and execution than the 
preceding ones. 

Lines 92-8: It is generally agreed that the very defective text contains refer­
ences to a trace-horse, to a helmsman on a ship and to someone whose 
singing is compared to that of the Sirens. Both Page and Garzya have their 
difficulties because they are unwilling to grant that more than one leader 
may be involved. Page (1951.97ff) argues that a 8196 must refer to the chorus, 
since a 8l101 surely refers to Hagesichora. In support of this he elaborates a 
difference between 96 (the whole chorus) and 85 (each girl in the chorus) 
which is not borne out by the text. Garzya (70) with some considerable hesi­
tation proposes to retain Hagesichora as subject in 96. It is evident that a OE 
96 is the climax of a priamel which Otterl096 has clarified as follows: 

The other horses follow the trace-horse. (92-3) 
The sailors follow the helmsman. (94-5) 
We are effective only because of the ability of our leader. 

(96-101). 
The details of the priamel are not completely verifiable, but the broad out­
lines are clear. Since the representative of authority appears to be mentioned 
in the first colon of each element of the priamel, a Sl 96 must refer to the 
leader. Her ability is, temporarily and parenthetically, compared with that 
of the Sirens, to the advantage of the latrer.97 But it looks as if the leader's 
prowess comes to be underlined once more in what follows, vv.98ff. The 
mention of the Sirens, therefore, is an interruption of a larger sequence in 
praise of the leader. It thus answers to the mention of Aotis in the previous 
strophe, vv.87-9. Just as the praise of Hagesichora was then tempered by the 
manoeuver of listing her side by side with the goddess and thus getting her 
in the goddess's shade, so the leader presently extolled is said to be inferior 
to the Sirens, and yet a virtuoso in her own right. 

Lines 98-9: aVTp S' €VSEKa] 7Tal8wv SEK . ....... EL' Scholiast XIX bases his 

96 W. A. A. van Otterlo, "Beitrag zur Kenntnis der griechischen Priamel," Mnemosyne 8 
(1940) 148-53. 

97 Bowra 1964.26 cites Pindar fr.94b, 13-20 Snell, from a partheneion, and adds: "Here 
Pindar no more thinks of the danger lurking in the Sirens' song than does Aleman when he 
uses it as a type for the most beautiful song ... " The Pindaric passage shows that the 
Sirens were regarded as spirits that calm the sea and rough winds. Is Aleman familiar with 
this notion? If so, the sentiment would connect with the elp1/11TJ of line 91. But cf Georg 
Weicker, Der Seelenvogel in der alten Literatur und Kunst (Leipzig 1902) 42. 



352 ALCMAN'S PARTHENEION I RECONSIDERED 

explanation of 'ten in the place of eleven' on Pausanias' discussion (5.16.2) 
of the team groupings at the Heraea, with its organization by three age levels. 
Page (1951.99) is inclined to accept the comparison; others who associate SlKa 
(or OEKas-) with the chorus are Bowra, van Groningen, Wilamowitz and 
Jurenka. But it is hard to see what team divisions by age have to do with this 
song, which is sung before the racing starts. Also Pausanias' remark that the 
sixteen women formed two choruses (rather than three, as might have been 
expected) indicates that for singing purposes age was not the chief criterion. 
Hence 7Tatowv in our text should not be pressed to mean 'very young girls' 
in contrast with older girls. In any case, since a O€ 96 refers to the leader, it is 
natural to assume that this same leader is the subject of the verb whose ending 
is preserved at the end of 99. This is actually the explanation given in the 
second half of the scholion: ~TJui.v OOV rT]v XOpTJYov . .• aVTi. EvOEKa aLOELV O€Ka, 
which may be translated: "he says that the choral leader ... sings as ten in 
the place of eleven." Page98 rightly warns that aVT{ cannot mean 'against'; 
'against', in an agonistic or amoebean context, is 7TPOS- or 7TO'T't, as is clearly 
shown in Theocritus. Precisely what the sense of ' as ten in the place of eleven' 
is I cannot say; perhaps Stoessl (101) is right in suspecting that a proverbial 
expression lurks behind it. But I take it that the expression is complimentary. 
For the use of aVT{ to designate worth, cf Il. 9.116. 

Lines 100-1: ~8€YYE'T'aL o ... w . .. '. Eav8w poalm KVKVOS-· a 0' E7TL}L€PWL 
gav8aL KOI'..lUKCU [ •.. The compliment is continued with the passage about 
the swan. Since ~8€yyE'T'aL, once more, is most plausibly assigned to the sub­
ject of the priamel, i.e., the choral leader, the passage apparently is a simile: 
"she sings as sweetly as the swan ... " There is no reason to assume that the 
O€ which connects the new sentence with the previous statement is adversative. 
The simile of the swan is thus in direct responsion to the image of the owl in 
the same position in the preceding strophe. But whereas then the half-chorus 
had, for the sake of extolling the virtue of Hagesichora, referred to them­
selves as hooting like an owl, the supporters of the leader in the present 
stanza sing not about themselves but about their leader and top their 
opponents by praising the voice of their leader in terms of a bird simile, 
which in effect turns the technique of their opponents against them. It is clear 
from the compliments that the leaders are regarded as exceptionally fine 
singers. They do not participate in the singing of this song. One assumes that 
Partheneion I was part of a program which included other selections, in which 
the leaders could demonstrate their qualities as singers.99 What follows, again 
connected by a continuative OE, is a reference to the attractive blond hair of 
the leader. 

98 Page 1951.98 n.2. Cf the flouting of this warning in d'Errico 34-5. 
99 Each half-chorus has praised its leader on three counts: (1) her beauty, (2) her ability 

as a racer on foot or on horseback, (3) her quality as a singer or dancer or both. Employing 



THOMAS G. ROSENMEYER 353 

Summing up: what remains of this strophe can be analyzed as referring to 
one, and only one, unnamed leader. She could be either Hagesichora or Agido; 
there is nothing in the words themselves to sway us in one direction or the 
other. But especially the responsion between vv.85-7 in honor of Hagesichora 
and vV.lOO-l seems to me to make for the conclusion that the eighth and final 
strophe is in support of Agido. 

v 
I hope that the foregoing analysis has put sufficient flesh on the 

skeletal notions offered in sections I-III and that the case for an 
amoebean structure of Partheneion I is now at least as strong as the case 
for monochoral delivery. More than probability is impossible in a 
field as open to hypothesis and reconstruction as the early choral 
lyric and Aleman in particular. It has not been my intention to say 
anything about the possible relationships between Aleman and later 
poetry or to contribute another chapter to the study of the origins of 
the pastoral. My citations from other choral writers and from 
Theocritus are to be regarded as heuristic only. I felt that these 
pointers might help to widen the range of speculation and provide 
thereby some new insights into the scope of the choral lyric. Happily 
this generation no longer subscribes to the principle that the simple 
precedes the more complex and that in reconstructing a song of 
Aleman we therefore must look for something rudimentary. At the 

the traditional praise by comparison, particularly the simile, for (1) and (2) the poet has made 
use of the archetypal epic horse: cf II. 6.506-11, combining the attributes of speed, elegance, 
beauty and vitality. But for (3), especially for the leader as singer, there is little epic pre­
cedent. Surely one may assume the bird simile as an ancient formula. Aleman must take 
his cue from folk patterns when he says, as he appears to do in fr.39 PMG, that he learned 
his music from KaKKa/3l8Es, which according to Athenaeus is a species of partridge: cf also 
frAO PMG and Ath. 9.374D. In vv.85-7 and 96-101 the singing is compared once to the 
hooting of an owl (see the protest of Page 1951.93 against the imputation by some com­
mentators that the owl's cry is ill-omened), and again to the song of a swan. The bizarre 
matching of swan and owl is found in Vergil, Eel. 8.55: certent et cycnis ululae. In place of the 
owl, Theocritus introduces hoopoes (5.137), wasps (5.29) and frogs (7041). Theideaof a singing 
contest between ill-matched rivals is a fixed motif of the pastoral tradition (see Ernest 
Dutoit, Le theme de l'adynaton dans la poesie antique [Paris 1936] 33), often as part of an 
&8ovaTov, as in the Vergil passage; cf also Eel. 9.29: sed argutos inter strepere anser olores. 
Aleman's is thus the first of many passages which later come to characterize singing 
contests, particularly the pastoral amoebean. No other examples have reached us from the 
time of Aleman or over three centuries thereafter, for Pin dar's image of crows vs. the bird 
of Zeus (01. 2.86fT) is meant in a different sense. I suspect, however, that there were other 
instances in archaic choral lyric, and that this was one of the features which led Theocritus 
and his successors to look to early Dorian lyric for part of their inspiration. Cf now also the 
remarks of G. Devereux in CQ 15 (1965) 176 nA. 



354 ALCMAN'S PARTHENEION I RECONSIDERED 

same time we expect, in our reconstructions of fragmentary material, 
to find patterns which are paralleled elsewhere in the history of Greek 
literature. If the present analysis of Partheneion I is plausible, the poem 
should please us on several scores. In its details it is often close to 
Homer; its mood alternates between lyricism, gnomic seriousness 
and mischievous banter; and in structure it exhibits a surprising 
amount of responsion, conceptual as well as formal, between the 
strophes. The responsion is never obvious or monotonous; like 
Theocritus, Aleman knows how to deflect answering patterns and to 
achieve subtle asymmetries. All this, and more, makes the poem 
hard to classify, but easy to enjoy. 

ApPENDIX 

Dichoria in Attic Drama 

According to schol. Ar. Equites 589, comic half-choruses, if consisting 
of dissimilar groups-men and women, women and children, young 
and old-tended to be numerically unequal, thirteen as against 
eleven.1oo This has been brought into connexion with scholion XIX, 

opposite Partheneion I v.99, which reads in part: ... aLa TO TOY XOPOV 
(lTt /Ltv tl[fl £a 1/"O:pO[ E1vWII (lTt ae tlK ". c/YrJ( 0"211 ) O~II rryll XOfYY}'YOy • ••••••. 
aVTt «i tXL8ELII t. Cf pp.351-2 above. Page (1951.99 n.4) rightly comments 
that there is nothing to connect the two scholia. aVTl, in the Alcman 
scholion, does not mean 'against' any more than in the poem itself; 
and partheneia are not likely to pit young against old, or men against 
women. It may be useful, nonetheless, to present a brief synopsis of 
what is known as strophic 8LXOplo: in drama. 101 If, as I have argued, 

100 EO"n 8' OT£ Kal ~fUx6p,a rlJ'Tavro if TO' l~ &v8pwv Ka~ yvva'Kwv .•• b BE TOWu,.O'S xopo", 
Ell'£V l~ av8pwv EL"1 Ka, yvva'Kwv .; xopo" l1TAEOVEKTE'TO TWV &v8pwv I'EPOS Ka~ ~ac:w ,y'. m 8€ 
yvvaiKES lv8EKa, El 8€ 1Tal8wv EL'Tl Ka, yvva'Kwv, ml'€v yvvaiKES ,y' ~ac:w. ol 8€ naiBEs ,a'. £l8€ 
1TpEafJu,.wv Kat VEWV, TOUS npEafJu,.as nAEOVEKTEiv 8Eiv q,aaw. 

101 Evidence from vase painting is not likely to be helpful. Women's choruses apparently 
were extremely rare in Spartan vase decoration; if. E. A. Lane, "Laconian Vase Painting," 
BSA 34 (1933/4) 99-189. Cf. also the warning of R. M. Cook. Greek Painted Pottery (Chicago 
1960) 279: "Mythical and divine occasions provided the favourite themes of Greek painting. 
Most are orthodox, and of the rest some deviate through negligence. Scholars of the 
eighteenth and nineteenth centuries were much concerned with the interpretation of 
rare myths and cults, often to fit the religious theories then fashionable; their spirit is kept 
alive by the zealots who look for illustrations of Greek drama." The closest we come to a 
partheneion is in such a painting as Villa Giulia 909 0. D. Beazley, AR VI [Oxford 1963] 



THOMAS G. ROSENMEYER 355 

Partheneion I is to be sung by two half-choruses in competition with 
one or more groups (each consisting of half-choruses or not), the 
conditions of Attic drama would seem to be relevant. 102 We are here 
mainly concerned with those passages in which each half-chorus is 
assigned a whole strophe or antistrophe or its equivalent. Other kinds 
of choral distribution are common enough. 

Attic comedy is the most obvious candidate for inspection. 
According to an old tradition, transmitted via the annotations in 
MSS R and V, the Aristophanic chorus consists essentially of two half­
choruses which operate independently in strophe and antistrophe of 
epirrhematic odes, especially in the parabasis. Unfortunately the 
tradition cannot be verified, though it is by no means implausible. 
However, even if demonstrable, this type of dichoria is not exactly 
relevant to our issue, because the epirrhematic structure with its 
interposition of sustained non-lyrical passages stands in the way of 
any amoebean immediacy. Hence I shall disregard the passages 
often cited in the handbooks (e.g., A. E. Haigh, The Attic Theatre 
[Oxford 1889] 282 n.1) and pass on to non-epirrhematic examples or to 
passages in which only a few lines of dialogue interrupt the 
pericope. 

Ar. Ranae 324ff ,..., 340ff. According to the scholiast, Aristarchus said 
that the chorus first sings as a unit and then divides into two groups. 
That is to say, vv.316-7 are sung by the whole chorus, 324ff by the 
chorus of women and 340ff by the chorus of men. There is little in the 
text itself to compel such a conclusion; but some kind of distribution 
between the women, who leave the stage at 447, and the men who 
remain, appears plausible enough. It is probably better, however, in 
this instance to talk of an initial double chorus rather than of half-

618, 1662) by a follower of Douris. The picture (Furtwangler-Reichhold plates 17 and 18) 
is that of a line of eleven girls, consisting of a flute player followed by ten singers, all of 
them moving towards the right, some in Ionic, some in Doric, some in mixed costume, 
with varying hair styles. There is no element of dichoria. 

102 W. Kranz, "Die Urform der attischen Tragodie und Komodie," NJbb 44 (1919) 145f, 
argues that the pre-Aeschylean aaTVp£K6vcontained much dichoria, especially in epirrhematic 
form, and that it was Thespis who substituted an €~cfpxwv for one of the choral groups. 
Cf however Aurelio Peretti, Epirrema e Tragedia (Firenze 1939) 91ff and passim. In a dis­
cussion of dramatic dichoria, Pollux 4.107 tells us that Tyrtaeus set up a Tp£xopla, a counter­
singing of boys, men and old men. Unless we reject Pollux's note out of hand, it must be 
regarded as evidence that in Aleman's own time, or roughly his time, choral division was 
not unknown. The reference must be to an amoebean rather than an agonistic division, 
since competition between different age groups is unheard of. 

5-G·R.B.S, 



356 ALCMAN'S PARTHENEION I RECONSIDERED 

choruses. We may compare the 1TP01TO/L1TO[ of Eumenides 1032 or the 
slave women, if indeed they are the singers, at Aeschylus, Supplices 
1034. 

Conradtl03 discusses the case of Acharnenses 557ff. The passage is 
short, and the iambic trimeters are spoken by the two coryphaei. 
The question is whether the dochmiacs starting with 566ff are sung 
by one of the half-choruses or by one of the coryphaei. The question 
cannot be resolved with the evidence available to us; in any case the 
passage is too brief and too specialized to be significant. The same is 
true of Equites 242f and Aves 353, where the chorus divides into two 
action groups, that is, where a danced dichoria takes place. I04 

This leaves, as our most important evidence for dichoria in Attic 
comedy, the Lysistrata, though here also it would be more correct 
to speak of a double chorus rather than the division of one. But the 
exchanges between the chorus of old men and the chorus of old 
women (614-35 - 636-57; 659-81- 683-705; 781-804 - 805-28) exhibit 
many of the verbal and conceptual symmetries and near-symmetries 
which we associate with amoebean. Wilamowitzl05 points out that the 
whole section is in place of a parabasis, which may help to explain the 
correspondences. But all dichoria, I venture to suggest, presses for 
amoebean patterns. 

The transmission of the exodus of the Lysistrata is too faulty to 
allow us to assign lines with any degree of certainty.lOO But most 
modern editors follow Wilamowitz in assigning 1247ff to the Spartan, 
1279ff to the chorus of the Athenians, and 1296 once more to the 

loa K. Conradt, s.v. 'HILtXOPlOV, RE 8 (1912) 236ff. Conradt's distinction (col. 235) between 
two types of division, (1) the confrontation of two groups by way of contrast, and (2) 
dichoria proper, the halving of a chorus for the sake of alternating in the singing of strophe 
and antistrophe, seems to me hazardous. All choral division tends towards the amoebean; 
it is of course true that the elements of response and feedback, such as &vcx8l7l'.\wuts, capping 
and the like, are more prominent in some cases than in others. Cf also n.14 supra. 

10& Cf w. Schmid/O. Stahlin, Geschichte der griechischen Literatur IV (Mlinchen 1946) 
50 n.: "Die starkere Besetzung des Komodienchors gegenliber dem der Tragodie erklart 
sich wohl daraus, dass streitende Doppelchore in der altesten attischen Komodie iiblich 
waren (unter den vollstandig erhaltenen hat solche nur noch Ar. Lys.)." 

106 U. von Wilamowitz-Moellendorff, Aristophanes Lysistrate (Berlin 1927) 159: "Aber die 
Teilung des Chores verbot das 71'cxpcx/3cxlv"w 71'pds 'Tdv 8fifLOV. Da hilft sich der Dichter so, dass 
er die Form der Parabase einhalt, eine 6n8..j mit J7I'lpPTIILcx von zehn Versen. Strophe und 
Antistrophe verteilen sich auf die zwei Chore. Dabei wiirde jeder einzelne zu kurz 
kommen, also folgt noch ein solches System." 

101 For a recent attempt to downgrade the r6le of the prytanis and to reserve a leading 
function to Lysistrata, see P. Handel, Formen und Darstellungen in der aristoph. Komlidie 
{Heidelberg 1963) 164-7. He suggests that the aa-rpot/>ov 1279iT is sung by Lysistrata. 



THOMAS G. ROSENMEYER 357 

Spartan. Wilamowitz107 repeatedly refers to the likelihood that the 
combination of trochees and dactyls which characterizes both the 
first song of the Spartan and the song of the Athenians, reminded the 
audience of Aleman. With regard to the second solo performance 
of the Spartan, Wilamowitz is more inclined to think that it is the 
Heingelegte Glieder," the units interrupting the largely iambic 
sequence, which establish the Alemanic ring. The Spartan's songs are 
astrophic; if we are to believe Wilamowitz, the song of the Athenians 
was matched by an antistrophe which is now lost. The dramaturgy 
as such, therefore, has no trace of dichoria. But I find it suggestive that 
it is in this play, full of dramatic confrontations and with a developed 
dichoria marking its half-way point, that we have songs strongly 
reminiscent of Aleman. It is as if choral amoebean went naturally 
together with the mention of girls and horses (1308) and Tyndarids 
gambolling along the banks of the Eurotas (1301-2). 

Dichoria in drama is not restricted to comedy. The Hyporchema (?) 
of Pratinas (fr.l D) appears to feature a chorus of satyrs in violent 
altercation with a rival chorus. Most scholars believe that there were 
two half-choruses.108 W. Schmid thinks109 that the double title of 
Pratinas' tragedy Kapva:rl8€s ~ LMuf.La£va£ must refer to a double 
chorus, since the two terms designate two different geographically 
separated entities. Other double titles might well be tested for 
the same implication. Further cases of dichoria in tragedy have been 
collected by LammersllO and others. The passages he cites are mostly 
restricted to parodi and exodi. The only one which features not con­
versation but parallel structuring, not strophes broken up into smaller 

107 Wilamowitz (supra n.l05) 195-7; also his Textgeschichte der gnech. Lyriker (Berlin 1900) 
88ff. See also the earlier reflections of R. Westphal in his Allgemeine gnechische Metrik (Leip­
zig 1865)=A. Rossbach/R. Westphal, Metrik der griech. Dramatiker und Lyriker 11.2 p.224. 

108 But cf. A. M. Dale, as reported in A. Lesky, Geschichte der griech. Literatur2 (Bern 1963) 
259 n.3. 

108 W. Schmid (supra n.104) II (1934) 178-9. Schmid's thesis cannot be more than a guess, 
since we know nothing about the plot of the play. 

110 Joseph Lammers, Die Doppel- und HalbchOre in der anti ken Tragodie (Paderborn 1931) 
143ff. He examines Aesch. Sept. 874ff; Soph. Eurypylus fr.91, Ajax 866ff, Ichn. 94ff; Bur. Ale. 
77ff, Suppl. 42 and 598, Tro. 153ff, Ion 184ff, Rh. 527ff. T. Zielinski, Die Gliederung der altat­
tischen Komoedie (Leipzig 1885) 277ff, rebuffs those who attempt to divide tragic choruses 
into half-choruses. Zielinski's conclusion, supported in principle by most scholars today, 
is that in tragedy, as against comedy, dichoria is very rare. Cf. also V. de Falco, Studi sui 
teatro greco 2 (Napoli 1958) 72ff, who lists some of the past champions of dichoria in tragedy, 
such as Muff, Hense and von Christ. They were influenced by R. Westphal, who was much 
inclined to make hasty inferences from the occurrence of certain formal elements com­
monly found in amoebean passages. And yet there is much in the analyses of Westphal, 
especially of tragic laments, which may still be usefully pondered by modem critics 



358 ALCMAN'S PARTHENEION I RECONSIDERED 

segments but apparently strophe and antistrophe sung by alternate 
groups, is Septem 875ff. The paragraphi of the transmission, as well 
as the Myns of v.895, leave little doubt that the chorus divides, in 
response to the presence of the two biers on the stage. III 

Two extant plays, then, are our principal evidence for strophic 
dichoria in Attic drama. But this is sufficient to support the notion 
that dichoria may have existed also in other types of dramatic per­
formance. The usual argument is that if Aleman innovated by 
introducing dichoria, it is surprising that (1) Pindar and Bacchylides 
exhibit no trace of it, and (2) the ancient critics and grammarians do 
not comment on it more fully than they do; see, e.g., Farina 57. The 
answer to (2) is that dichoria may, at least in the classical age, always 
have been fairly rare; but that, conversely, there are traces in the 
scholia of remarks which point in the direction of dichoria in Aleman. 
As for (1), we should remember that except for the Epinida of Pin dar 
and the slim remains of Bacchylides, what we have of archaic choral 
poetry is not adequate to serve as the basis for the formulation of 
general rules. For all we know, partheneia favored dichoria. The two 

interested in formal symmetries and reciprocal patterns in tragic lyric. One should, of 
course, endeavour to distinguish between the hieratic responses of a dirge, on the one 
hand, and the witty games of jockeying for position which we associate with amoebean 
contests. One and the same figure, anaphora, e.g., may be used to achieve different effects. 

111 Sept. 875ff, down to the end, 1009, contains many amoebean features. On the assump­
tion that Aeschylus meant all the singing to be done by the chorus or by sections of it, we 
may discover dichoria in plenty. Unfortunately it is more difficult to know which sections 
are sung by whom. Ideally, 875-80 are sung by half-chorus A and 881-7 by half-chorus B; 
again, 888-99 by A, 900-10 by B, etc., etc. But with the introduction of Antigone and 
Ismene the situation became muddled. Murray notes in his apparatus that the paragraphi 
in the MSS are «confuse appicti," Still, in some cases the wording and syntax of the anti­
strophe are closely modelled on the strophe, for example 892-4",904-6. Beginning at 961 
the two choral groups may be said to engage in a stretta, along the lines of division between 
Antigone and Ismene as proposed by the editors. The refrains at 975-7 and 98~ would be 
sung by the complete chorus. II. 24.718ff, our earliest reference to an amoebean lament 
(in this case an alternation between soloists and chorus), shows the typical patterns of 
threnodic symmetry, even though the choral statements are perforce omitted. Whether 
the common designation of the Muses as a!-,£,po!-,£va, 01T~ KaAfj, (II. 1.604, Od. 24.60, etc. ; cf 
also Vergil, Eel. 359: amant alterna Camenae, though the relation has been questioned) is an 
allusion to amoebean technique or not, must be left open. Cf also Plato, Lg. 12.947B. In 
any case I should repeat that we expect to find a difference between the formal means of 
threnodic dichoria on the one hand and y£.pvp,u!-'o, and allied types on the other. Eric T. 
Owen, The Harmony of Aeschylus (Toronto 1952) 17, reverts to the older view that in Suppl. 
1018-51 the chorus divides. This may well be a more dramatic as well as a more economical 
solution than Kirchhoff's attribution of 1034-51 to the servants. However, whatever di­
choria there may be is between 1T£p'Ko1Taland not between strophe and antistrophe, and that 
puts the passage out of contention for our purposes; it means that the symmetries are 
between statements of the same group rather than between groups. 



THOMAS G. ROSENMEYER 359 

Pindaric partheneia happen to be monochoral in the extant sections; 
but that tells us little about partheneia in general or Aleman's in 
particular. It is quite conceivable and indeed likely that if Aleman's 
partheneia, or some of them, involved dichoria, he did not innovate 
but followed previously existing models. It is instructive to recall that 
among Sappho's poems, amoebean patterns are usually felt to docu­
ment her dependence on certain forms of folk poetry, for example 
fragments 140a, 114, 115 LP. 

In sum, we cannot interpret Partheneion I on the basic of precon­
ceived notions concerning the presence or absence of dichoria in early 
choral lyric. It is precisely because dichoria is so rare in extant drama 
that we must be prepared for the chance that there were some few 
examples of dichoria at a much earlier time. 

THE UNIVERSITY OF CALIFORNIA, BERKELEY 

August, 1966