University of California Press California Studies in Classical Antiquity

Elegiac and Elegos
Author(s): Thomas G. Rosenmeyer

Source: California Studies in Classical Antiquity, Vol. 1 (1968), pp. 217-231
Published by: University of California Press
Stable URL: http://www.jstor.org/stable/25010574
Accessed: 23-04-2018 10:25 UTC
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
University of California Press is collaborating with JSTOR to digitize, preserve and extend
access to California Studies in Classical Antiquity
This content downloaded from 5.225.59.71 on Mon, 23 Apr 2018 10:25:29 UTC
All use subject to http://about.jstor.org/terms
THOMAS G. ROSENMEYER
Elegiac and Elegos

Aristotle (Poetics ch. 1) mentions the E'AEyEoTrotol, with the
writers of epic, as poets who write verse without music.1 This sets them
off against others, such as the dithyrambic, nomic, tragic, and comic
poets who use 'eAos along with jvO6zos and pe'rpov in their production.
The whole discussion is part of Aristotle's introductory section in which
he states, with an air of delighted surprise, that the Greeks have no word
for literature. He does not argue that elegiac poetry is non-musical;
he simply assumes it, as an obvious truth. The casualness of the passage
1 Aristotle's classification of elegiac as non-musical continues to be operative

later; see Dionysius Thrax ars gramm., irepi ypac. ?2: we should read (=deliver) tragedy
IpWcuKS, comedy fLwrTKws, ra 'eeyeia Atyvp3Zs, epic eTrovWs, lyric /tLLeAtCS, rovS 8 oKTroVS Vfe?t
/fevWs KCa yoepws. The position of elegiac, between comedy and epic, speaks for itself.
XAyvpws means "clear", but not necessarily (or usually) "plaintive," pace Harvey (cf. infra,
this note) 170-171. Cf. also Photius's report of Proclus's Chrestomathy sections 24-27 Severyns,
where elegiac appears between epic and iambic (A. Severyns, Recherches sur la Chrestomathie
de Proclos 1.2 [1938] 98 conjectures that this must have been the original order in section 12
also), thus implicitly casting doubt on the connection with elegos and lament which Proclus
accepts.
The following abbreviated references will be used: Bowra: C. M. Bowra,
Early Greek Elegists (Harvard 1938); Campbell: D. A. Campbell, "Flutes and Elegiac Cou
plets," JHS 84 (1964) 63-68; Dover: K.J. Dover, "The PoetryofArchilochus," EntretiensFond.
Hardt 10 (1963) 181-212; Garzya: A. Garzya, Studi sulla lirica greca (Messina 1963); Harvey:
A. E. Harvey, "The Classification of Greek Lyric Poetry," CQ (1955) 157-175; Huchzer
meyer: H. Huchzermeyer, Aulos und Kithara in der griechischen Musik (Emsdetten 1931);
Lasserre: F. Lasserre, ed. tr. comm., Plutarque: De la musique (Lausanne 1954); Page: D. L.
Page, "The Elegiacs in Euripides' Andromache," in Greek Poetry and Life (Oxford 1936)
206-230; Peek: W. Peek, Griechische Grabgedichte (Darmstadt 1960); Wilamowitz 1913:
U. von Wilamowitz-Moellendorff, Sappho und Simonides (Berlin 1913); Wilamowitz 1921:
U. von Wilamowitz-Moellendorff, Griechische Verskunst (Berlin 1921).
9--.S.C.A.
218 Thomas G. Rosenmeyer
does not lessen its importance; it is our earliest explicit testimony about
the mode of performance of elegiac.2
In the light of this, it is surprising that most moderns have
taken it for granted that the elegiac was sung, and accompanied by the
aulos, well into the fifth century and beyond. Bowra's assertion may
stand for many: "The elegiac, then, came into existence as a flute
song, and such it remained for some three or four centuries."3 Wilamo
witz and Felix Jacoby attempted to do justice to Aristotle's testimony by
suggesting that elegiac was ranked as recited rather than lyric verse
because the performer spoke rather than sang his lines to the aulos
accompaniment.4 But Rhode undertook to demolish Wilamowitz'
arguments, and Jacoby found few followers.5 The habit of thinking of
elegiac as aulodic is so strong that in a recent translation of Aristotle's
Poetics, his words on elegiac are glossed: "A plaintive song accompanied
by the aulos."6 Other scholars continue to feel the difficulty. W. Peek:
elegiac "was sung to the flute, or at least delivered by means of a dy
namic, modulated speech-song; it was and remained, for a while, a
lyric genre."7 H. Koller ventures to find his own intriguing solution:
from the start, he feels, elegiac was embellished, "umspielt," by the
aulos, that is, the reciter took turns chanting and playing his instrument.8
In 1964, D. A. Campbell went beyond the hesitations of
Wilamowitz and Jacoby and showed, in a well-argued and fully docu
mented article, that the case for an originally and essentially musical
elegiac is even weaker than they thought.9 He did this by methodically
reviewing the evidence, both from within the corpus of elegiac poetry
2 By "elegiac" I mean a poem, or collection of poetry, written in elegiac

couplets. The term "elegy" is better restricted to the Hellenistic and Roman uses of the
verse form.
3 Bowra 7. Cf. also G. Luck, Die roemische Liebeselegie (Heidelberg 1961)
18-21.
4 For the references, see E. Curtius in RE 5 (1905) 2262-2263. See also
Wilamowitz 1921.101-102. Wilamowitz's middle position derives in part from his interest in
ancient notices concerning 7rapaKaTaAoy ; cf., e.g., his Die Textgeschichte der griechischen Lyriker
(Berlin 1900) 53 and n. 2. For a similar hesitation in antiquity, see Demetrius On Style 167; he
complains that some of Sappho's poetry is not really "poetic," and that it does not go with
dance or lyre, "unless there is such a thing as conversational lyric": el' ,s eSil Xopos
SLaAEKTLKO'S
5 E. Rohde, Der griechische Roman3 (Leipzig 1914) 149f n. 1.

6 K. A. Telford, tr. comm., Aristotle's Poetics (Chicago 1961) 3.
7 Peek 13.
8 H. Koller, Musik und Dichtung im alten Griechenland (Bern 1963) 126-131.
9 Campbell passim.
Elegiac and Elegos 219
and from ancient remarks about elegiac, and conc
sufficient to support a claim which would fly in the
We may now abandon a fiction which never was v
first place, and which has had the unfortunate effec
some that other types of spoken poetry, such as iamb
delivered to aulos accompaniment.10 If in what f
review some of Campbell's evidence once again, it i
his argument, excellent as it is, may be supplemented
may show that other internal references, if taken
references to auloi in elegiac, would lead to prepo
second, it may be possible to say more about the rela
and elegos, a putative kinship which had much to do
aulos myth.
* * *
In the pastoral lyri

lines which are in fact no
it will not do to confus
production. Let us disti
music, which we may te
circumstantial. The latter
to the accompanist(s)-us
the future, more rarely
choral poetry.12 There a
phanes, in Alcman, Pind
conclusive. Pindar's Pyth
aulos, presents a dactylo
the instrument. In spite o
of saying whether this
Similarly, when Bacchyl
the city of the victor,14
10 Huchzermeyer 29ff.
11 This is my general impre
merits a more exact investigatio
12 For the dramatists, see
Pratinas 708.6-7 PMG. In the lat
plained to everyone's satisfactio
13 The following example
PMG; Pindar 01. 5.19-20; Nem.
"Sophocles" 737b.2-4 PMG.
14 Bacchylides 9.68, with B
the gods,15 there is no assurance that the references to auloi tell us
anything about the production of the hymn in question. Nor, obviously,
does Pindar's song about the Locrian invention of music for voice and
auloi permit us to make inferences about the accompanist.16 Sappho's
words, in the Hector and Andromache poem,17 about the music in the
bridal procession in Troy should not be taken to mean that Sappho
accompanied herself with a mixture of aulos and another instrument
(castagnets?). Or is the poem choral after all?
In none of these selective examples does the narrator, or
one of his characters, talk about himself as playing the aulos.18 It is
more difficult to demonstrate that what is true of the pastoral piper may
be true also of the persona of an elegiac or an iambic poem. But once we
acknowledge that a musical instrument may be part of a poetic reality
precisely as any other object in the life of an imagined character is, the
case should not be so very difficult after all. When Theognis, or the
speaker of a Theognidean poem, tells us that he is silent (420), what are
we to make of it, except to conclude that the silence is not part of the
auditory texture of the line, but part of his poetic role ? The same with
his statement (313-314) that he is a madman surrounded by madmen.
Real madmen do not, as a rule, talk in elegiacs (nor, to avoid petitio,
do they sing, except KaraXPTrlanTKcS). His tavla, here, is referential, not
circumstantial. When Archilochus (56A D.) gets ready to lower sail,
are we to imagine him going through his nautical maneuver while
reciting, or singing, his poem to the (boatswain's?) pipe?19
Campbell has shown that none of the internal evidence
necessarily documents aulos accompaniment. I think we can go further,
in the light of the considerations just introduced, and argue that by and
large the evidence is referential, hence cannot indicate performance. As
in pastoral poetry the term a&ei'v is not uncommon; but it is generally
agreed that when Solon (2.2) uses the term o j,20 and Aristophanes Pax
15 Pindar Dithyr. 75.18-19.
16 Pindar fr. 140b.2-4.
17 Sappho 44.24-25 LP; cf. Bacchylides 10.54.
18 But see Archilochus 76 D. Can we be sure, in spite of Monum. Par. (fr. 51
I A 47 and IV A 52 D.), that Archilochus is talking about himself? At any rate, the meter
makes it less than likely that the speaker is conducting a paean, unless "Lesbian paean" is a
nickname for quite another type of composition.
19 See also Archilochus 77 D., which is obviously not from a dithyramb. For
references to aulos in elegiac, besides Campbell 64ff, cf. also Huchzermeyer 38ff.
20 Cf. Plutarch Solon 8: there is no mention of an aulete, and Pisistratus is
made to refer to Solon as tr3 Aeyowv. Cf. Campbell 65-66.
1267-1268 has aSeLv, both in a circumstantial sense, t
the reciting of epic rather than to a musical setting.
is the model passage to show the irrelevance, to the m
of the mention of musical instruments: "I rejoice dri
to the accompaniment of an aulete;22 I also rejoice ho
lyre."
xaLPC O) SE 7 rvwv KML VT5' aVXAriprpos aEo&Iov,

XCp co) aE ?SOOyyOV XEpaM Avprv oXeW v.
Which is it to be, aulodic, or lyric? At 1065 we learn that it is proper,

while engaged in a KCCJLOs, to sing to aulos accompaniment, vr'
avAXr]rr-pos cE1$Scv. This is one of the pleasures that men and women
may have; but it is not what the speaker is doing at the moment;
KWcOacELv is not conducted in elegiacs. The typical Theognidean utter
ance is not descriptive, but imaginative, or expressive of his desires, and
often prescriptive; and when the "thou shalt" includes mention of an
aulos, we learn nothing about the delivery of the poem.23
21 At Aristophanes Pax 1279, a Homeric hexameter like the preceding ones,
singing is out of the question. Stanford's vauable reminder (W. B. Stanford, The Sound of
Greek [Berkeley, Los Angeles 1967] 28ff) that Aristoxenus and Aristides Quintilianus allowed
for types of delivery intermediate between conversational speech and lyric song (cf. also
similar warnings by Th. Zielinski, Die Gliederung der altattischen Komoedie [Leipzig 1885] part
2, ch. 2) should not cause us to drop the distinction between unaccompanied recitation,
however elevated and resonant, and accompanied chanting or singing; cf. the remarks of
Aristoxenus Elements of Harmony 1.8ff, cited by Stanford. Another question related to our set of
problems is whether skolia were sung to the accompaniment of an aulos or a lyre. Huchzer
meyer 57 decides in favor of the lyre, on such evidence as Aristophanes Nub. 1354ff and
Cicero Tusc. 1.4. The decision, though based on little evidence, is not inherently unlikely,
seeing that many Attic skolia are composed in Aeolic meters. But when H. adds that "der
Aulos begleitete beim Mahle wohl nur ... die Lieder im elegischen Masse," and refers to
Aristophanes Ves. 1217ff, he fails to convince: (1) the meal is over; (2) the auletris has played
a prelude, prior to the beginning of the skolia; and (3) the skolia themselves are, at this point
at least, iambic.
22 &Eojwv is Pierson's correction for the ms acovtwv which Young accepts
into the text; cf. however, most recently B. A. van Groningen, Theognis (Amsterdam 1966)
211-212. In any case, the disjunctive Xaipw 8e--Xalpw $8 shows that we are to think of separate
occasions, neither of which needs to be identified with the present one.
23 For passages in which the speaker expresses a desire for music, see
Theognis 1041; 1055-1058; 761; 791. Desire implies want; hence the passages cannot tell us
anything about performance. Theognis 239ff poses an interesting question. Campbell 64
thinks that the atvAcbi6a which, according to the speaker, will celebrate Cyrnus in the future,
must be Theognis's own elegiacs, but that Theognis has in mind, not his own performance,
but a more formal or elaborate occasion. To this the answer must be that we have no legiti
mate reason for the assumption that a poem can be performed in two different ways. Hence,
The external evidence is equally inadequate. Most of it
derives from a tradition which crops up most alarmingly in the ps.
Plutarchean De musica, especially ch. 8, which is a mine of ill-considered
and jumbled information.24 It records a notice, also given by Strabo,25
that Mimnermus played a tune called Kradias on the aulos. This piece
of information is supposed to go back to Hipponax (96 B.). De musica 8
continues: " In the early period the AcpSol sang eAEyeya eLeAo/oEroTLe7va,
elegiacs put to music; this is clear from the Panathenaic inscription
concerning the musical contest."26 If Hipponax really said that Mimner
mus played the Kradias on the aulos (we do not have his own words),
then the piece was not aulodic but auletic, hence the further comment
of De musica that early elegiac was aulodic is not pertinent. The reason
ing seems to be something like this: (1) Mimnermus performed a nomos
on the aulos; (2) (but wasn't Mimnermus an elegist?); (3) (Yes, but) in
the old days elegiac involved aulos playing; compare the inscription.27
It is clear from the discussion in De musica ch. 8 (cf. also
chs. 4, 9, 19, 28) that EAEyetov and 'AEyoS are, in the author's mind,
associated with the nomos and choral poetry. Actually there is some
doubt that nomoi were usually choral. W. Vetter, while stressing that
nomos is not a technical term but a name that can be applied to many
either Theognis's elegiacs are sung, or Theognis is not referring to his own lines. I think the
latter is correct; Theognis envisages a time when his own verses about Cyrnus will have made
the youth a celebrity, to be further eulogized in musical encomia. Finally, Theognis 825-830
and 939-942 (or 939-944, as Young arranges the lines) are two rather mysterious passages
which have not yet been sufficiently explicated by the commentators. But on any interpreta
tion they cannot be pressed to support the theory of sung elegiacs; one refers to the past, with
disapproval, the other to the future. The latter, in particular, appears to me to be a griphos.
24 I assume that in many cases the confusions of De musica go back to its
sources, among them Heraclides Ponticus; cf. the corrections recommended by H. Weil and
Th. Reinach in their treatment of chs. 8 and 9, in their Plutarque: De la musique (Paris 1900)
23ff. F. Wehrli, Die Schule des Aristoteles 7 (Basel 1953) 112-115 has nothing on the relative
merits of this section of Heraclides's work on music. Cf. also U. von Wilamowitz-Moellen
dorff, Timotheos: Die Perser (Leipzig 1903) 89.
25 Strabo 14.1.28.
26 For lejLEXA'qroLrLE'zva cf. infra, p. 224. I do not understand Lasserre's
comment, p. 158, that the Panathenaic inscription is cited to explain why the elegiac writer
Mimnermus appears in Hipponax in the capacity of aulete.
27 The logic is faulty; the inscription we do not know. As regards the un
reliability of much that is said about music and poetry in the later accounts, see Campbell's
showing, 67-68, that Athenaeus 14.620C (the poetry of Homer, Hesiod, Archilochus,
Mimnermus, and Phocylides was composed to be sung) and 14.632D (Xenophanes, Solon,
Theognis, Phocylides, and Periander did not compose their verses to be sung) cancel each
other out.
different things, finds that songs called nomoi are
"durchkomponiert," and often monodic, or purel
When De Musica 4 and 5 informs us that Clonas o
nomoi for aulos in the wake of Terpander's nomoi fo
clear whether his nomoi were aulodic or auletic, a d
occasionally obscured in the tradition of which D
Strabo, and Pausanias are representatives.29 For s
whose invention is ascribed to Clonas, including
is featured in the series, Sacadas is claimed as the
account of the tradition about Sacadas is still t
although Hiller himself, perhaps unavoidably, fe
harmonizing tendencies of which he accuses his p
showed that Sacadas was mainly known for his wo
most famous composition, the vdLOS IIvOiKdos with w
Pythian games in 582, 578, and 574 B.C.,31 appears to
of Programmmusik for solo aulos, representing th
with the serpent. Hiller allows for the production of
pieces by Sacadas; though some of the evidence on wh
especially De musica 8 and 9, is unreliable, there is
should deny Sacadas some measure of versatility.
was built on his accomplishments as an aulete.32
It is against this background that we must ta
information (De musica 3; 5; 8) that Clonas, Polym
28 W. Vetter, s.v. vo'dos in RE 33. Halbband (1936) 84
mowitz-Moellendorff, Timotheos: Die Perser (Leipzig 1903) 93-94 was t
of vo'tos on a sound footing; cf. H. Grieser, Nomos (Heidelberg 193
the subject is summarized by R. P. Winnington-Ingram in Lustrum
suggests, probably correctly, that after Sacadas aulodic vof4os bec
votlos was usually citharodic. But see Telestes 810 PMG who disting
voduos (ev a3vAos) and a Lydian v Lvos (dovct:voIs rr'jKTrSWv iaApoZs)
29 See the apparatus and commentary of Weil and Rei
20-21. Even the great August Boeckh allowed the ancient confu
writing a brief paragraph defining vodos in which just about eve
de metris Pindaricis iii. 6.201.8: Nomi quidem, qui aut avtA,&tKol au
simplicis erant metri, citharoedici ex hexametris heroicis, quam
laudatur, auloedici ex distichis elegiacis; paulatim vero priscae simp
catior structura, adeo ut ne antistrophas quidem haberent.
30 E. Hiller, "Sakadas der Aulet", RhM 31 (1876) 76
Frazer, ed., Pausanias' Description of Greece V (1913) 245-246.
31 Paus. 2.22.9; 4.27.7; Strabo 9.421; Pollux 4.74; 78.
32 Cf. also Pollux 4.81: Pythian auloi played ro aXopov atVl
ol Ue XoptKOli t&vpda'flotLs rpoarv'ovv. May we infer that typically
honor of Apollo, was solo work?
were composers of eAeyeZa and pdA-q. On the one hand, Clonas and
Sacadas were responsible for a kind of music for (voice and?) aulos
called elegos; on the other we learn (Pausanias 10.7.4) that avAwSOLa had
an extremely short life at the Pythian games, being in fact presented only
once, in 582 B.C. Thereafter, according to Pausanias, the Amphictyons
decided that avAwSlia was not appropriate to the sanctity of the occasion;
q yap ,vA8,&'a ueA) re -rEv avAov roa aKV0pWTOTaTa Kal A UEyeax [Operot]
rTpoaaSodeva Tros aVAoZs.33 If there is anything we can make of this
garbled text, it is that avAwSlia consists of dismal (or: disfiguring?)34
tunes on the aulos and dirges sung along with the aulos. Note the un
usual order; the aulos appears to be the lead, and the voice the accom
paniment. And as if this was not bad enough, Pausanias proceeds to cite
a dedication in which an Echembrotus of Arcadia was hailed for win
ning a victory at Delphi
'"EAArqcT ' aeSCwv ICsAEa Kal eAEyovs.
What are we to make of the distinction between LEAvr and EAeyot, which
also underlies the association of FpA?) and Aeyetea wEp/EAo7rop77yLva in
De musica 8 ?35 It may well be that pLAX, as in Pausanias's report of the
decision of the Amphictyons concerning av'Ac8ia, refers to the playing
33 ldAXq re is Dindorf's plausible emendation for codd. fcr-qr). Wilamowitz

1913.298 notes: "Die Inschrift ist so seltsam, dass es schwer faellt, ihr in allem zu trauen."
We cannot be sure that it was not Oprjvot rather than eAeyela which originally stood in the
text, but it is unlikely. The association of /deAi with eAeyeCa or LAeyot is common enough; cf:
the Suda's note on Olympus. Cf. De musica 3, from Heraclides, where Terpander, and after
him Clonas, is said to have taken his own and Homer's 4Tr) and to have added tunes, t&qr/
7rep&rO'evra. Here rbM stands in the position elsewhere occupied by eAeyeta or ~Aeyot. I am
inclined to interpret this to mean that both poets used dactylic rhythms. Cf. also Alcman
39.1 PMG.
34 aKvOpwo'rorara can hardly refer to facial expression, or only to facial
expression, since auletic performance, which continued after the vetoing of aulodic, has an
even larger share of facial distortion. It is true that in classical writers, from Aeschylus to
Plato and Xenophon, aKcvBpwros always means "scowling" or "frowning." But in some of the
later writers, particularly in Plutarch, a figurative use of the word is found: De curios. 518B,
De def. orac. 417C, etc. In such cases the item to which oacvOpwnro is applied is treated as if it
had a face, or the ascription involves the idea of consequences which will be a frowning matter.
On balance I suggest that in the Pausanias passage the adjective is used figuratively, and goes
with both ldA,/ and eAeyela.
35 The verbal participle suggests that Plutarch thinks of eAEyeia as non
musical. 1AeAoiroleo originally simply meant "to make music" = "to sing"; see Aristoph.
Thesm. 42, 67; Ran. 1328; like seAosioda and fseAoroto's it was used without an object. But in
Longinus ch. 28.2 and Athenaeus 14.632D, as in our Plutarch passage, it takes an object, and
means: to set to music a piece which is originally non-musical. This is particularly clear from
the Longinus passage: Plato used tropes and periphrasis to add music to his speech: LAX7)v
Aaftd.v rVv A tfL ?eAoiowiroae.
of the aulos, while EAhyos points to a vocal perf
saw in connection with Sacadas, this is by no m
Euripides's later use of the term EAEyos to refer to O
lyre (Hypsipyle 62 Page) serves to confuse the pictur
The really significant issue in this snarled
and claims is the nature of the relation between e)AE
An explicit statement that elegoi were sung and c
distichs, goes back to Didymus, who takes it f
At the same time Strabo (14.64.7) and Horace (A
even in antiquity the beginnings of elegiac were n
understood.38 The combination of these two posi
know nothing about the origins of elegiac, but th
composed of elegiac couplets, has haunted scholarsh
The first scholar poet, Euripides, acted upon the c
the Andromache wrote a piece in elegiac couplets whi
invoke the shades of the early elegos of whose form
inkling.40 The passage is in the nature of an exper
may regard the sung couplets as in some way analo
meters; cf. the difficulties of, say, Sophocles Tra
Philoct. 839-842; in both passages the distribution of
and speaking is not at all clear. The Alexandrian p
36 Dover 188.ff has a good appraisal of the evidence. Th
his reconstruction is its success. Any reconstruction that manages
by accommodating all the data is likely to arouse both admiration a
37 See Severyns (supra, n. 1) 101 who refers to Et. Gu
38 Cf. also L. Alfonsi, "Sul irept IHoITrcv di Aristotel
193-200. Alfonsi, and Rostagni whose writings he cites, are perhaps
they know what was in the second book of Hepti rotrqTWv.
39 Cf. Bowra 5: "Hellenistic and Roman writers regar
mournful measure-Ovid's 'flebilis Elegeia'." Bowra notes that ou
are military and convivial; he sees the contradiction, but does
"mournful measure" is not quite correct; when the Augustans, and
the term elegi (always in the plural), the reference is either to th
Ovid Fasti 2.3), or to complaints, especially lovers' complaints (C
ment and frustration, rather than death, provide the occasion f
"mournful," then, is yet another symptom of the muddle resulting
the data concerning elegos and elegiac.
40 For Euripides's assumption that elegos = lament, cf
analyzes the metrics of Andromache's lament; the scarcity of di
regularity of the caesura after the third longum, seem to him to
model, as does the "Doric" dialect. He further speculates, less convi
ness of the lament may be due to the performance of the Andromache
445 and 734); Page guesses Argos, ca. 421 B.C. Garzya 175-179 finds,
that the sequence is not threnodic, but an attempt to recapture the
Callimachus (fr. 7.13 Pf.) and Apollonidas (A.P. 10.19.5) drew on
Euripides's experiment and put the equation on its head; the very
obscurity of the term EAeyos made it an attractive substitute for the
more humdrum (in their days, not earlier, as we shall see) EAEyedov.
This comes to be imitated by the Roman elegists. Hence there is noth
ing surprising about Pausanias (10.7.5) and De musica (8) confusing
the terms in reverse and writing EAEyela where we should have expected
9'AEyoL.41
One thing on which most sources seem to agree is that the
old elegos was a mournful tune.42 This, in turn, may have encouraged
the close association of elegos with flute playing, because according to
one tradition the aulos was at first used exclusively for songs of mourn
ing.43 The consequences of this combination, along with the muddling
of the history of elegos and elegiac, have been disastrous. "Elegos meant
lament, ... and the elegy was recited to the accompaniment of a flute
.. .".44 Scholars have been hard put to it to square this dogma with the
fact that so much of early elegy is not lament, and so much aulos playing
is connected with occasions of a joyous or playful nature.45 To cope
with this difficulty, some speak of "threnodic elegy" whenever they
wish to talk about that elegiac tradition which, they think, sprang
directly from the old elegos-lament.46 But the qualification merely
41 See Campbell 67.
42 Cf. the evidence collected by Page. He also shows that the sources indicate
a readiness to think of elegos as accompanied by a string instrument. The assumption that
elegos was a kind of lament was held to be corroborated by various etymologies; cf. schol.
Aristoph. Aves 217 (p. 215 Duebner) who, apparently following Didymus, derives gAeyos from
e e AEyeav; and the modem discussions cited by H. Frisk, Griechisches Etymologisches Woerterbuch
fasc. 6 (1957) 486, s.v. ZAeyos. For the ancient etymologies, see also Severyns (supra, n. 1)
99-101.
43 Plutarch, De E apud Delph. ch. 21. Cf. also the passages cited by E. Reiner,
Die Rituelle Totenklage der Griechen (1938) 67-70, with notes. Reiner's distinction between
yoos (private, informal) and Op/vos (public, stylized) does not concern us here.
44 P. Friedlaender and H. B. Hoffleit, Epigrammata (1948) 66.
45 See the evidence compiled by Huchzermeyer, and his conclusion p. 27.
For dvavAos in the sense of "sad," cf. Euripides Phoen. 791, and Campbell 66, who refers to
Miss Dale's note on Euripides Alcestis 447. "Unmusical" or "unfestive" are other possible
translations; but whatever the exact meaning of the negative adjective, it argues against a
necessary tie-up between the aulos and grief.
46 Friedlaender (supra, n. 44) 68-69 speculates that Anacreon 100 and 102
D., and perhaps 101 D., were "specimens of that very genus of poetry which we sense in the
background of the elegiac epitaphs," i.e., poems "sung before the tomb." But surely eyosae
in 100 D. makes that quite impossible. (Their authenticity is not here at issue). Cf. also Page
214, n. 2, on Archilochus 10 D.; he refers to Plutarch, quom. adulesc. poet. aud. 23B, who quotes
the poem.
serves to make the case for an original identity o
even more precarious. Add to this that it is now
recognized that the earliest writers of elegiac cou
and that the genre itself, and its language, bear t
origin.47 If Tyrtaeus was a Spartan who compo
(digamma-less) Ionic because elegiac was at that t
erty, one may well wonder what is left of the connec
and the Peloponnesian elegos of Clonas, Sacadas
their peers.
Nor does the use of elegiac couplets as funerary verse help
to associate elegiac with the elegos-lament. There were funerary hexa
meters as well as elegiacs, and even some funerary iambics.48 It is true
that it was the elegiac couplet rather than hexameters which soon
achieved a virtual monopoly on tomb markers and votive tablets. The
reason for this is that the couplet is a closed system, better suited to the
needs of a self-contained avowal than the open-ended hexameter or
other stichic patterns. There is very little doubt, however, that
funerary elegiac is a specialized adaptation of the heroic hexameter for
lapidary use. Note the occasional difficulties encountered by epi
graphers in their decipherment of early dedicatory inscriptions whose
fragmentary state makes it difficult to determine whether they are
written in hexameters or in elegiacs.49
Kenneth Dover has recently reminded us that the term
E'AEyEov for an elegiac couplet does not crop up before the fifth cen
tury.50 Earlier writers usually prefer to use the same term, c'rr-,, to desig
47 See Dover 193; also Bowra 43.

48 See Peek 12; Friedlaender (supra, n. 44) 157-158. It is difficult to see a
difference in kind between public elegiacs and public iambics; the iambs from Ptoion, IG 12
472 (Friedlaender no. 167), could equally well have been composed in elegiacs. I am doubtful,
however, about Dover's suggestion, p. 189, that Archilochus did not regard his (personal)
poems in elegiacs and in iambs as belonging to different genres, and that perhaps he used
the word iamboi "with reference to all the forms of poem which he composed, their common
characteristic being not their metre or language but the type of occasion for which they were
composed-their social context, in fact."
49 Perhaps the most interesting case is that of the spit-holders from Pera
chora with their fragments of writing which Wade-Gery attempted to reconstruct as an elegiac
(H. Payne, Perachora I (1940) 262, 266; cf. also T. B. L. Webster, "Notes on the Writing of
Early Greek Poetry," Glotta 38 (1960) 263, n. 1; while Miss L. H. Jeffery, The Local Scripts of
Archaic Greece (1961) 124 arranges the remains so that they turn out to be two hexameters.
There is little doubt that Miss Jeffery's reading is the more likely one.
50 Dover 187ff.
nate either epic or elegiac verse,51 and this usage continues to be found
in Hellenistic and Graeco-Roman discussions.52 In later treatments of
material from the sixth and fifth centuries B.C., especially in Hephaestion
and the metrical scholia, we sometimes come across the term CA)yela in
the place of the more usual EAEyelov. Whether Anacreon's 'AEyeAla53
were any different from the elegiacs he is known to have written may be
doubted. Aristotle is the first to use the feminine form, but we cannot be
sure that this form of the word in the text of the papyrus is not due to the
fact that by the second century A.D. EAEyEtov and EAXyeda had come to be
used interchangeably.54 One notice in Hephaestion deserves special
attention (4.1-3 Consbr.): he remarks that Sophocles in his EAEyEyaL
could not accommodate the name "Archelaus" either in the ECros or in
the eAEyelov, and so he wrote 'ApXCAEws. There are two points here:
(1) Sophocles wrote EAEyecaL; and (2) EAeyEZov may mean the last line of
an elegiac couplet.55 It is, I think, a reasonable supposition that the
term E'Ayeia was introduced when an effort was made to distinguish
between an elegiac poem as a whole, and the second line of an elegiac
couplet, with its characteristic rhythm. For all we know this happened
in the time of Aristotle, or perhaps even earlier. In any case we may
assume that Sophocles's EAEyedac were, simply, elegiac couplets, whether
of the ordinary kind, or akin to those in the Andromache.56
The foregoing discussion will, I hope, have established the
51 See Solon 2.2; Theognis 20; 22; Herodotus 5.113.

52 Theocritus epigr. 21 calls Archilochus's elegiacs frrea; I assume that he
does so on the basis of an old terminology. In Hephaestion, who uses the term ero? to desig
nate the hexameter, irr7 may also refer to the combination of hexameter and pentameter;
example: 65.11 Consbr. Hermesianax and Callimachus used the term rrevradtEepov to distin
guish elegiacs from straight hexameters; cf. Wilamowitz 1913.298 note. Perhaps it was an
awareness of the confusion between EAEyEZov and ,AEyos which prompted some to avoid using
the term EAeyelov to refer to elegiacs. Conversely 'AeAcyEov may be applied to a hexametric
couplet; see Vita Homeri 36 (As'); and Pherecrates Cheiron fr. 3.7-9 Meineke (153.7-9 Kock,
Edmonds) features eAEyeZov in the sense of "hexameter line."
53 Hephaestion 5.2 Consbr. What he quotes is a hexameter, fr. 95 Bergk (55
Gentili). Hephaestion 9.10ff Consbr. also speaks of an EXAyela ets 'AAKaiLdSqv, and cites two
couplets from it.
54 Aristotle, 'AO.7roA. 5.2. Severyns (supra, n. 1) 1.1 (1938) 209-211 pro
vides specimens of variation in Photius, between EAEye'a and EAEyeZov or E'Ayela. Clearly the
scribes did not know any difference between the words, but tended to replace the feminine
with the neuter.
55 Cf. Hephaestion 51.21 Consbr., and Appendix Dionysiaca ch. 4 (=
Hephaestion pp. 315-316 Consbr.). On the other hand, at 63.4-6 Consbr. EAEyeiov clearly
means elegiac couplet. XEyos does not occur in Hephaestion.
56 Cf. my remarks supra, p. 225.
following points, or at least made them probable. T
dence for a family tie between elegos and elegiac. E
epic verse, as early references to it prove.57 There is
was sung, and some evidence that it was not. The
assume any necessary connection between elegiac a
That leaves us with one last question: how
of poetry introduced by Tyrtaeus and Callinus and
in a subsequent age, to be called by the name EAE
could not but remind people of the old elegos perform
Sacadas and their fellow producers of nomoi and sung
of reliable information, we can only guess. Wilamowi
it is not always warranted to think of the unit -
hemiepes, i.e., as half of an epic period; he points
Aeschylus Suppl. 843-846 = 853-856,59 where the un
elsewhere it appears along with other metrical units.
Wilamowitz comments60 that Pindar and Bacchyl
add, the playwrights) avoid the doubling of the unit
while Stesichorus (219 PMG 2) and Simonidies (581
Actually Wilamowitz's distinction is hard to maint
each in Stesichorus and Simonides is no better tha
Pindar.62 Wilamowitz, again, is inclined to be
ancient hymns employed refrains of the metrica
Page speculates that formulas of that type in And
such as Cot (yc; 7F/zEAEa and KaC TO v 4uvv LV eAEcas m
57 See Wilamowitz 1913.291: "Von einem elegischen S

zu reden, ist ein Unding." "Euphorion, der Nachtreter des Kallim
meter." It would take a bold man to show that the sentiments ex
Theocritus 8.33-60, a sequence of elegiacs, differ appreciably from
ble hexameter passages.
58 Wilamowitz 1921.101; his remarks assume his own
which posits brevis in longo, hence disqualifies the sequence from
hemiepes.
59 Cf. also Euripides Troad. 1094-1098 = 1112-1116; Alcestis 590-591
599-600 where, however, the rhythm of the strophe obscures the pattern.
60 Wilamowitz 1921.432.
61 Cf. also Alcman 39.2-3 PMG.
62 Note such prominent exceptions as 01. 13.17 = 40 = 63 etc.; also Paean
5.41-42 = 47-48; cf. Snell6 II 166. It should be noted that Simonides does not use DD in
531 PMG, the encomium on the victims of Thermopylae, where we might have expected it to
occur. In fact none of the texts cited by Reiner (supra, n. 43), i.e., laments attributed to lyric
writers, exhibit the pattern.
refrains in traditional dirges.63 Could it be that fifth-century choral
composers disdained the doubling of this unit because they felt that the
ascendancy of non-musical elegiac had disqualified the sequence for
use in song ?64 Be that as it may, the process of displacement and label
ing may be pictured as follows. To begin with, there was the elegos, at
home in the Peloponnesus, probably a species ofnomos, that is, astrophic
and either for solo instrument or for voice and accompaniment. At the
time when the elegos was receding in prominence, that is, in the course
of the sixth century, elegiac verse from Ionia, at first merely an exten
tion of epic and called 'E'y, happened to emerge into prominence; and
when it had become entrenched as one of the principal varieties of
spoken verse, somebody applied to it the name EAEye'a or AE'yeLov,
perhaps because the clash of longa in the middle of the second line of
the couplet reminded the nomenclator of similar things in lyrics with
which he was familiar, and which he took to represent the ancient
elegos. On this admittedly hypothetical assumption-though we should
not forget that the coining of metrical terms such as Asclepiadean and
Adonic and lecythion is a notoriously willful business-the connection
between elegiac and elegos would be one of nomenclature, of lexical
transfer, rather than of natural affinity or even filiation. There are
parallels in the history of modern poetry. The humanist sonnet, to take
one instance, has little to do with the folk tune whose name was put
under obligation to label the highly sophisticated, and recited, poems of
the Italian masters.65 The precise details of the appropriation of the
Provencal term are no longer discoverable. But no one claims that the
sonnets of Giacomo de Lentino, or Guittone di Arezzo, were written
63 Wilamowitz 1921.95; Page 219.

64 E. Fraenkel, "Lyrische Daktylen" (Kleine Beitraege zur klassischen Philo
logie 1 [1964] 165-233; reprint of his 1918 article in RhM) has nothing on double hemiepes.
Heliodorus 3.2 has a hymn to Thetis which consists of fifteen consecutive pentameters. I
assume that this is merely a prose writer's attempt to write, and rival, heroic lyric.
65 Modern scholars have generally returned to Du Bellay's view that the
sonnet is a learned Italian invention, rather than of Provencal, Northern French, or even
Arabic ancestry. See W. Moench, Das Sonett: Gestalt und Geschichte (1955) 267-268; also E. H.
Wilkins, "The Invention of the Sonnet," Modern Philology 13 (1915) 463-494, reprinted in a
revised form in W.'s The Invention of the Sonnet and other Studies in Italian Literature (1959) 11-39.
The section disproving the nineteenth-century notion of a popular origin of the sonnet is
omitted from the revised version, for the reason that nobody believes it any more. "It is now
agreed that the sonnet was an artistic invention, not a popular growth." The theory of a
popular growth had been based almost exclusively on the circumstance that the term 'sonet'
(diminutive of 'so') derives from Provencal folk culture; it means "sound," "lay," "song."
to be sung, to the accompaniment of rustic fiddles or
to that of more courtly instruments.66 And we may
peare's "Why is my verse so barren of new pride" is n
a continuous tradition which has its roots in count
have a way of being stolen, or misapplied, as The
sorrow.
University of California
Berkeley
66 Moench (supra, n. 65) 41 considers it a paradox that sonnets came to be

set to music, particularly by the Dutch and Italian polyphonists. In the Renaissance we
distinguish between songs composed to be sung (ballads, madrigals, etc.), and poems which
were set to music after first having been published without music, i.e., poems to be recited.
In antiquity, as far as we can tell, poems were composed for singing or for recitation. The
idea of singing a poem that was not originally written for singing voice seems to be foreign
to the ancient tradition.

University of California Press California Studies in Classical Antiquity

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

University of California Press California Studies in Classical Antiquity

Uploaded by

Copyright:

Available Formats

Elegiac and Elegos

Author(s): Thomas G. Rosenmeyer

Elegiac and Elegos

1 Aristotle's classification of elegiac as non-musical continues to be operative

2 By "elegiac" I mean a poem, or collection of poetry, written in elegiac

5 E. Rohde, Der griechische Roman3 (Leipzig 1914) 149f n. 1.

In the pastoral lyri

xaLPC O) SE 7 rvwv KML VT5' aVXAriprpos aEo&Iov,

Which is it to be, aulodic, or lyric? At 1065 we learn that it is proper,

33 ldAXq re is Dindorf's plausible emendation for codd. fcr-qr). Wilamowitz

47 See Dover 193; also Bowra 43.

51 See Solon 2.2; Theognis 20; 22; Herodotus 5.113.

57 See Wilamowitz 1913.291: "Von einem elegischen S

63 Wilamowitz 1921.95; Page 219.

66 Moench (supra, n. 65) 41 considers it a paradox that sonnets came to be

You might also like