Ancient Greek Elegy


[[This essay is an online version of an original printed version that appeared in The Oxford Handbook of the Elegy, ed. Karen Weisman (Oxford: Oxford University Press 2010) 13-45. In this online version, the original page-numbers of the printed version are indicated within braces (“{” and “}”). For example, “{13|14}” indicates where p. 13 of the printed version ends and p. 14 begins.]]
The word elegy comes from the ancient Greek language, which attests the word elegos (ἔλεγος) and its derivatives elegeion (ἐλεγεῖον), and elegeia (ἐλεγεία). These words were used in two senses.
One sense had to do with the singing of a sad and mournful song, to the accompaniment of a wind instrument called the aulos (αὐλός), a double reed resembling what we know as the oboe. It is relevant that the Greek word elegos and its cognates are etymologically related to the Armenian word elegn, which refers to the same kind of wind instrument (Chantraine 2009 s.v.).
The other sense had to do with the rhythm of two particular verses combined as a couplet. The two verses are the elegiac hexameter and the elegiac pentameter, while the combination of these two verses is the elegiac couplet. A poem constructed by way of elegiac couplets is an elegy.
At first sight, the two senses of the word elegos and its derivatives seem unrelated to each other, since many examples of ancient Greek songs or poems composed in the elegiac couplets of elegy seem to have nothing to do with the singing of sad and mournful songs. A closer look at the surviving evidence, however, may help undo such an initial impression. Although there is currently no consensus in the scholarly world of classical studies about the origins of elegy, an argument can be made that elegy evolved from traditions of singing songs of lament.
For the term lament, I offer this working definition: lament is an act of singing in response to the loss of someone or something near and dear, whether that loss is real or only figurative. The essential background on the ancient Greek traditions of lamentation can be found in the foundational book of Alexiou 2002 (the first edition was {13|14} published in 1974; the second edition features an important new introduction by Yatromanolakis and Roilos).
The argument here is not that the singing of lament was destined to remain the basic function of the ancient Greek poetic form of elegy. After all, there is evidence to show that elegy became vastly diversified in its functions, and such diversification was already underway at a relatively early period (Irwin 2009).
The argument, rather, is simply this: the earliest recoverable function of elegy was the singing of elegos in the sense of a mournful song or lament. Such a sense of the ancient Greek noun elegos, still active in some uses of the derivative noun “elegy” and the derivative adjective “elegiac” in English ([cross-reference to the relevant chapter in the Handbook]), is clearly attested in the tragedies of Euripides (Aloni 2009). A telling example is a passage in his Trojan Women where the lamenting figure of Hecuba says that she is singing her elegoi, her “mournful songs of tears” (119: δακρύων ἐλέγους).
To understand the elegiac function of elegy, we need to take a closer look at its actual form, the elegiac couplet, which can be defined in terms of its meter. (By meter I mean the stylization of rhythm in poetry: Nagy 1974:145; also Allen 1973:13–14, 258; and Nagy 1990a:39–42). Here is a metrical map of the elegiac couplet, showing that its meter is a combination of a hexameter consisting of six feet with a pentameter supposedly consisting of five feet:

Hexameter – ⏑ ⏑ – ⏑ ⏑ – | ⏑ ⏑ – | ⏑ ⏑ | – ⏑ ⏑ – –
A B C D
Pentameter – ⏑ ⏑ – ⏑ ⏑ – | – ⏑ ⏑ – ⏑ ⏑ 
A

The symbols used in this metrical map are as follows: “–” stands for a long syllable, while “” stands for a short syllable; both in the hexameter and in the pentameter, the last syllable of the verse is a special category, though I write it simply as a long syllable in both cases (on “the law of indifference” in the last syllables of verses, see Allen 1987:134); “|” stands for predictable word-breaks, that is, for places where words predictably begin and end.

As is immediately evident from looking at this metrical map, the basic rhythmical sequence of the elegiac couplet was –, known as a “dactyl” because it consists metaphorically of one long “joint” followed by two short “joints,” like the human finger or daktulos (δάκτυλος). In a comedy by Aristophanes (Frogs 650–651), there is talk of a rhythm described as κατὰ δάκτυλον, “dactyl by dactyl,” which is evidently a dactylic rhythm.
What is not evident from this metrical map is the existence of an alternative rhythmical sequence – – that could replace the basic rhythmical sequence –. This alternative sequence – –, known as a “spondee,” was named after a rhythm associated with the stately slowness of singing while pouring a libation or spondē (σπονδή). The spondee was a derivative of the dactyl, as we know from the fact that the second long syllable of – – was a contraction of the two short syllables of – (Allen 1987:112–114; Nagy 1996b). {14|15}
In introducing my metrical map of the elegiac couplet, I described the hexameter as consisting of six feet. Now I add more details to this metrical description:
  1. Any one of the first five feet of the hexameter can have the rhythm of a dactyl (–) or, alternatively and secondarily, of a spondee (– –).
  2. As for the sixth foot, it has the rhythm of an incomplete dactyl (–), though it looks like a spondee (– –); this interpretation is in line with what we know about the final syllable of the hexameter (see again Allen 1987:134 on “the law of indifference”; also Allen 1973:98).
  3. In the hexameter, the main word-break occurs either at point A or at point B. So the occurrences of word-breaks at points A and B are mutually exclusive. As for occurrences of word-breaks at points C and D, their occurrences are not mutually exclusive, nor are they affected by occurrences of word-breaks at either A or B.
  4. After the main word-break, whether at point A or at point B, it becomes markedly more rare for wording that contains the rhythm of a dactyl (–) to be replaced by wording that contains the rhythm of a spondee (– –).
In introducing my metrical map of the elegiac couplet, I described the pentameter as consisting of five feet. Now I add more details to this description as well:
  1. Either one of the first two feet of the pentameter can have the rhythm of a dactyl (–ᴗᴗ) or, optionally and secondarily, of a spondee (– –). Likewise, either one of the last two feet of the pentameter can have the rhythm of a dactyl (–ᴗᴗ) or, optionally and secondarily, of a spondee (– –).
  2. As for the fifth foot, we are confronted with a problem: the first two and the last two feet of the pentameter do not add up to five feet. And five is the number of feet required by the name of the meter, pentameter. So, where is the missing fifth foot? This problem can be solved by counting not the feet but the syllables of the pentameter, which has fourteen syllables altogether (after we discount the optional and secondary replacements of dactyls by spondees). The four feet we have already counted are occupied by twelve syllables. Two syllables have yet to be counted: the first of these two is the long syllable following the first two feet of the pentameter and the second of these two is the long syllable following the last two feet. These two matching long syllables combine to form the rhythmical equivalent of the missing fifth foot of the pentameter.
  3. In the pentameter, the main word-break occurs at point A. This point is positioned exactly at the rhythmical center of the pentameter.
  4. After the main word-break, at point A, it becomes markedly more rare for wording that contains the rhythm of a dactyl (–ᴗᴗ) to be replaced by wording that contains the rhythm of a spondee (– –). In this way, point A in the pentameter is similar to point A in the hexameter. In another way, however, it is radically different. It has to do with two different kinds of splitting at point A. In the hexameter, the third foot (–ᴗᴗ) splits into two rhythmically equal halves (– and ᴗᴗ) at point A. So, what comes after point A in the hexameter is the second half of the third foot. In the pentameter, by contrast, the whole pentameter (–ᴗᴗ–ᴗᴗ–|–ᴗᴗ–ᴗᴗ–) splits into two rhythmically equal halves (–ᴗᴗ–ᴗᴗ– and –ᴗᴗ–ᴗᴗ–) at point A. So, {15|16} what comes after point A of the pentameter is not the second half of a third foot, as in the hexameter, but the second half of the pentameter. The rhythm of the first half of the pentameter, after it reaches point A, goes on to repeat itself in the second half.
In this metrical analysis of the elegiac couplet, we see two kinds of split at the main word-break. In the hexameter, the main word-break at point A (or B) splits the verse into two rhythmically asymmetrical parts. In the pentameter, by contrast, the main word-break at point A splits the verse into two rhythmically symmetrical parts, two perfect metrical halves. So the hexameter of the elegiac couplet is asymmetrical while the pentameter is symmetrical.
The symmetry of the pentameter, which completes and thus defines the elegiac couplet, creates a symmetry for the whole couplet. And this symmetry in form is relevant to the argument that the elegiac couplet, as the form of elegy, is connected with an “elegiac” function, that is, with the function of singing mournful songs. As we will now see, such an “elegiac” function of the symmetrical elegiac couplet can be analyzed in terms of its opposition to the epic function of the asymmetrical hexameter.
Up to now, we have been considering only the kind of hexameter that combined with the pentameter to form an elegiac couplet. But there was another kind of hexameter that combined not with the pentameter to form an elegiac couplet but with other hexameters to form a series of consecutive hexameters. This other kind of hexameter was the basic form of the genre of epic, as most prominently represented in the Homeric Iliad and Odyssey, which contain respectively over 15,000 and over 12,000 consecutive hexameters. So, this kind of hexameter could have an epic function when it was uncombined with the pentameter of the elegiac couplet. I have described the epic function in this way here in order to make a point about the “elegiac” function. Just as we may speak of the genre of epic as marked by the hexameter uncombined with pentameter, we may also speak of the genre of elegy as marked by the hexameter combined with pentameter.
Here and hereafter, I use the word genre with reference to forms that have functional relationships with each other or with other forms, as in the case of the ancient Greek genres of epic and elegy (for more on functional definitions of genre, especially with reference to epic, see Nagy 1999b).
In Indo-European languages other than Greek, there is comparative evidence for prehistoric phases in the formal characteristics that distinguish the genres we know as epic and elegy in the historical phases of their existence. In terms of such prehistoric phases, we may speak of a basic functional opposition between epic and lament. The most decisive comparative evidence comes from the Slavic traditions of epic and lament, as explored by Jakobson (1952). In these traditions, the primary meter of epic is a decasyllable consisting of two asymmetrical parts separated by a main word-break, and the first of these two parts has wording composed of four syllables while the second has wording composed of six. By contrast, the meter of lament is a decasyllable consisting of two symmetrical parts separated by a main word-break, and each of the two parts has wording composed of five syllables (Nagy 1974:184–186). Other Slavic meters of epic and lament show comparable patterns of asymmetry and {16|17} symmetry respectively (Nagy 1974:186–188). Such patterns of opposition in the formal features of epic and lament are prototypes of the opposition we see in Greek traditions between the epic function of the hexameter, with its two asymmetrical parts, and the elegiac function of the pentameter, with its two symmetrical parts (Nagy 1974:185; relevant is the discussion of Allen 1973:120).
In the case of the attested Greek genres of epic and elegiac poetry, more needs to be said about the formal opposition between the asymmetry of the epic hexameter and the overall symmetry of the elegiac couplet, which has its own hexameter combined with its own pentameter. I focus here on the hexameter as combined with the pentameter in the elegiac couplet. This hexameter as used in the genre of elegiac poetry is formally distinct from the hexameter as used in the genre of epic poetry as well as in other genres, such as oracular poetry. While the epic or oracular hexameter simply maintains its asymmetry in the process of combining with other hexameters, the elegiac hexameter actually highlights its own asymmetry in the process of combining with the symmetrical elegiac pentameter that follows it in the elegiac couplet. An example of such idiosyncratic highlighting is the fact that the elegiac hexameter, unlike the epic hexameter of Homeric poetry, tends to avoid a word-break at point A (Barnes 1995:150). That is because a rhythm that hypothetically ends at point A of the elegiac hexameter would then replicate the rhythm that ends at point A of the following pentameter as well as the rhythm that starts at the same point A of that pentameter. So, in order to avoid such a three-way symmetry, where the first part of the hexameter would hypothetically replicate in rhythm the first as well as the second parts of the following pentameter, the elegiac hexameter tends to prefer a word-break at point B instead of point A.
Greenberg (1985:260) says it best when he concludes, on the basis of such metrical idiosyncrasies in the elegiac couplet: “[W]e should not think of elegy as derived from the hexameter.” As he points out, it is not a matter of “tailoring” the phraseology of epic hexameter to fit the phraseology of the elegiac couplet. Elegy has its own system of meter and phraseology. “It is different, and has its own rules.”
In the light of such findings, I will summarize the formal characteristics of elegy as a genre in the attested phases of Greek literature. In contrast to epic as a genre, which uses a combination of hexameters in a series of hexameters, elegy as a genre uses a combination of its own special elegiac hexameter with its own special elegiac pentameter in the elegiac couplet. Whereas the hexameter as used in the genre of epic existed independently of the pentameter, elegiac hexameter as used in the genre of elegy existed in a co-dependent relationship with the pentameter. The co-dependency is evident from three facts:
  1. The asymmetry of the elegiac hexameter had to be accentuated to highlight the symmetry of the pentameter.
  2. The pentameter had no other purpose but to follow up on the elegiac hexameter. So, although the epic hexameter could be part of a series of consecutive hexameters, a pentameter could not be part of a series of consecutive pentameters. The pentameter could only be part of an elegiac couplet. Only the elegiac couplet could be part of a series of consecutive elegiac couplets. {17|18}
  3. In such a series of consecutive elegiac couplets, the symmetry of the elegiac couplet could be further accentuated by way of rhyming the last syllable at the end of the first half of the pentameter with the last syllable at the end of the second half (Nagy 1974:100). To illustrate, I quote without translation an elegiac poem attributed to Theognis (the conventional datings of this figure vary, from the late seventh century BCE all the way to the early fifth). In this poem (Theognis Scroll I verses 173–182) the rhyming pattern that I have just described is visible in the pentameters of five consecutive elegiac couplets (at 174, 176, 178, 180, 182, the ends of the rhyming syllables are highlighted):
          Ἄνδρ’ ἀγαθὸν πενίη πάντων δάμνησι μάλιστα,
καὶ γήρως πολιοῦ | Κύρνε καὶ ἠπιάλου […ou | …ou]
175    ἣν δὴ χρὴ φεύγοντα καὶ ἐς βαθυκήτεα πόντον
ῥιπτεῖν καὶ πετρέων |Κύρνε κατ’ ἠλιβάτων. […ōn | …ōn]
καὶ γὰρ ἀνὴρ πενίηι δεδμημένος οὔτε τι εἰπεῖν
οὔθ’ ἕρξαι δύναται | γλῶσσα δέ οἱ δέδεται.[…ai | …ai]
χρὴ γὰρ ὁμῶς ἐπὶ γῆν τε καὶ εὐρέα νῶτα θαλάσσης
180    δίζησθαι χαλεπῆς | Κύρνε λύσιν πενίης.[…ēs | …ēs]
τεθνάμεναι, φίλε Κύρνε, πενιχρῶι βέλτερον ἀνδρί
ἢ ζώειν χαλεπῆι | τειρόμενον πενίηι. […ēi | …ēi]
Having considered these three facts pointing to a co-dependency between the hexameter and the pentameter, we need to reassess the functions of these meters as markers of the genre that is elegy. In an opposition of hexameter and pentameter, the hexameter is the unmarked member, since this meter may be used in genres other than elegy, such as epic and oracular poetry, whereas the pentameter is the marked member, since it is used only in the genre of elegy.
The markedness of the pentameter can be seen most clearly in the use of this meter for the purpose of expressing a meaning that supplements whatever meaning is already being expressed by way of the hexameter. In order to observe such an interaction between units of meaning as conveyed in these two meters, I outline here the overall metrical and thematic structure of a ten-line elegiac poem, Poem 4 of Tyrtaeus (whose lifespan is conventionally dated to the latter part of the seventh century). First I give the text of the poem:

          Φοίβου ἀκούσαντες Πυθωνόθεν οἴκαδ’ ἔνεικαν
μαντείας τε θεοῦ καὶ τελέεντ’ ἔπεα·
ἄρχειν μὲν βουλῆς θεοτιμήτους βασιλῆας,
οἷσι μέλει Σπάρτης ἱμερόεσσα πόλις,
5        πρεσβυγεν<έα>ς τε γέροντας· ἔπειτα δὲ δημότας ἄνδρας
εὐθείαις ῥήτραις ἀνταπαμειβομένους
μυθεῖσθαί τε τὰ καλὰ καὶ ἔρδειν πάντα δίκαια,
μηδέ τι βουλεύειν τῆιδε πόλει <σκολιόν>·
δήμου τε πλήθει νίκην καὶ κάρτος ἕπεσθαι.
10      Φοῖβος γὰρ περὶ τῶν ὧδ’ ἀνέφηνε πόλει. {18|19}

They heard Phoebus Apollo at Delphi and from there brought back home
the oracular pronouncements of the god and the words [epea] that were meant to come to
pass:
“To lead in deliberations, that is what the kings should do, honored as they are by the gods.”
Their concern is the city of Sparta, the city loved by its people.
5        “And the elders, senior that they are, and then the men, who are members of the
community …”
responding as they do in accordance with the straight utterances,
“… they should speak things that are fair and they should do all things that are just …”
and they should not offer any plan for this city of ours that is crooked,
“And, as for the masses of the community, victory and superior force should keep them
company.”
10      So, you see, that is how Phoebus Apollo made his revelation about these things to the city.

Now I give the metrical and thematic outline:

  1. At line 1, a hexameter introduces the whole poem and is followed at line 2 by a pentameter that supplements the introduction.
  2. At lines 3, 5, 7, 9 of the poem, four hexameters contain oracular poetry as supposedly uttered by the god Apollo himself at the Delphic Oracle.
  3. At lines 4, 6, 8, three pentameters supplement—by way of a relative clause (line 5), a participial phrase (line 6), and a coordinate clause (line 8)—what is being expressed by the oracular hexameters at lines 3, 5, 7, with the surprising result that the hexameters of lines 3, 5, 7 are syntactically independent of the pentameters of lines 4, 6, 8. Then, at line 10, a pentameter supplements not only what is being expressed by the oracular hexameter at line 9 but also what is being expressed by the whole poem (Nagy 1974:185n37).

In this Poem 4 of Tyrtaeus, we see elegiac poetry in the act of actually quoting the oracular hexameters by way of its own elegiac hexameters and then supplementing those oracular hexameters by way of its own elegiac pentameters.

There are two important ramifications here that affect our understanding of elegiac poetry as a genre:

  1. In quoting the oracular hexameters, this elegiac poem is referring to uses of hexameters in genres other than elegy. In this way, the genre of elegy shows its capacity for performing the functions of forms that belong to other genres.
  2. In supplementing the oracular hexameters by way of pentameters, this elegiac poem is not extending the quotations of hexameters. It is extending only the specifications that are framed within these hexameters. The pentameters that follow the oracular hexameters are not oracular quotations in their own right. These pentameters only supplement the meanings of the quoted oracular hexameters. Still, even if they only supplement the oracular meanings, these pentameters are performing here the function of oracular poetry. So, in this way as well, the genre of elegy shows its capacity for performing the functions of forms that belong to other genres.

In some elegiac poems, this capacity of elegiac can be expressed by way of ostentatious references to the actual forms of other genres. We find an example of such a {19|20} reference in the elegiac Poem 4 of Tyrtaeus. In the pentameter at line 2 of this poem, the word epea is used in referring to the hexameters quoted at lines 3, 5, 7, 9. This word, which is epos in the singular and epea (epē) is the plural, is the source for the modern term epic. In ancient Greek, the singular form epos can be used to refer to a single hexameter while the plural form epea can refer to a series of hexameters (Koller 1972; Nagy 1979:236, 272; Martin 2005:13–14). In Poem 4, epea is an ostentatious way of referring to a series of oracular hexameters as quoted within a series of elegiac couplets containing the quoted hexameters combined with supplementary elegiac pentameters.

Comparable in its use of epos / epē is an elegiac poem attributed to Theognis (Scroll I lines 15-18):

Μοῦσαι καὶ Χάριτες, κοῦραι Διός, αἵ ποτε Κάδμου
ἐς γάμον ἐλθοῦσαι καλὸν ἀείσατ’ ἔπος,
“ὅττι καλόν, φίλον ἐστί· τὸ δ’ οὐ καλὸν οὐ φίλον ἐστί,”
τοῦτ’ ἔπος ἀθανάτων ἦλθε διὰ στομάτων.

Muses and Graces, daughters of Zeus! You were the ones who came once upon a time to the house of Kadmos.
You came there, to his wedding, and you sang a beautiful epos:
“What is beautiful is near and dear, what is not beautiful is not near and dear.”
That is the epos that came through their immortal mouths.

At line 17, this poem quotes a hexameter that supposedly recovers the actual wording of a song that had been sung once upon a time by the Muses and the Graces when they performed at the primal wedding of Kadmos and Harmonia; at line 16, which is the pentameter that precedes the quoted hexameter, the poem ostentatiously looks forward to this quoted hexameter by calling it an epos; then, at line 18, which is the pentameter that follows the same quoted hexameter, it ostentatiously looks back at it by calling it an epos all over again.

So far, we have considered how the genre of elegy shows its capacity for performing the functions of forms that belong to the genres of epic and oracular poetry. But there is more to it. This capacity of elegy is part of the overall function of elegy. In the case of Poem 4 of Tyrtaeus, for example, we can see how all the pentameters of this elegiac poem supplement the meanings of hexameters that precede them and, by extension, how they supplement the meaning of the whole elegy. Looking beyond this particular elegiac poem, we can see how the procedure of supplementing the meaning of hexameters by way of pentameters in an elegiac couplet is part of an overall procedure of supplementing the meaning of an entire poem. Such overall supplementation is a characteristic feature of the elegiac genre. From here on, I will refer to this feature as the supplementary function of elegy.
What I have just called the supplementary function of elegy will turn out to be vital for taking up again the argument that was started at the beginning of the chapter, which was, that elegy evolved from traditions of singing songs of lament. These traditions, as we will now see, had their own supplementary function. {20|21}
It must be noted right away, before we even begin to consider what we know about the traditions of ancient Greek lamentation, that we have precious little direct evidence about laments as they were actually performed in the ancient Greek world. What we do have for the most part are representations, in the verbal and the visual arts, of performed laments. The surviving evidence for such representations in song and poetry as well as in painting and sculpture is surveyed in the book of Alexiou, mentioned earlier (2002).
In the specific case of song and poetry, we can mine the surviving evidence to find whatever indications they may give about the form and the content of the laments sung by the figures who are being represented. We have already seen an example of such a represented figure: she is Hecuba in the Trojan Women of Euripides. Another example, coming up soon, is the figure of Andromache as she appears in another tragedy of Euripides, the Andromache.
Whatever we stand to learn indirectly about the form and the content of laments as performed in ancient times can be compared with what we can learn directly from studies of live recordings of laments as performed in the various surviving Greek song cultures of modern times. The cumulative results of such learning have been collected and integrated in the work of Alexiou (2002). Other works that integrate what we know about modern as well as ancient Greek traditions of lament include Sultan (1999) and Dué (2002, 2006).
In applying here what we know about Greek traditions of singing lament, we may start by reapplying the working definition that I offered at the beginning of this chapter: lament is an act of singing in response to the loss of someone or something near and dear, whether that loss is real or only figurative. Now I elaborate on this definition by outlining three characteristic features of lament:

  1. The prime mover in the singing of a lament is conventionally the one person who is most closely affected by whatever loss is being lamented. That person is conventionally a woman. She shows her sense of loss by expressing her sorrow in song. She sings her lament. And, in singing her lament, she can cry while she sings and sing while she cries (in traditional lament, the physiology of crying and gesturing is integrated with the art of singing: see Dué 2006:46). The loss that is being lamented is most commonly the death of a loved one, which can be linked with other misfortunes. The contexts of lament for the dead will vary. The lament may be an immediate response to a misfortune, as in the witnessing of a death or hearing the news of a death. Alternatively, it may be a delayed response, as at a funeral or commemoration.
  2. The woman who sings as the prime mover in the singing of lament becomes the lead singer. As the lead singer, she interacts with an ensemble of women representing a given community. The ensemble responds to the lament of the lead singer by continuing it. The continuation is an antiphonal performance, which may take various different forms, ranging all the way from stylized crying and gesturing to full-blown singing and dancing (Tsagalis 2004:48–50, 72–74). In using the term antiphonal here, I am emphasizing the fact that such performance is meant as a response to the initial {21|22} lament of the lead singer (Dué 2006:12; on antiphonal tsakismata in Modern Greek traditions of lamentation: Dué 2006:159).
  3. In singing her song of lament, the woman who sings as the lead singer may touch on any aspect of her own feelings or on any aspect of the projected feelings of the community as represented by those who sing and dance their antiphonal response to her lament. These feelings may be projected either by the lead singer or by the represented community. And the antiphonal response may be performed not only by the ensemble of women who represent the community but even by the community itself, consisting of men as well as women (Tsagalis 2004:61, 64–65, 69). In short, lament is a communalizing experience. It leads to a communalization of emotions, in all their diversities.

To illustrate these three characteristic features of lament, I start by quoting the narration of a lament in epic. The epic is the Homeric Iliad. And the lead singer of the lament is Briseis, a beautiful aristocratic woman who had been captured as a war prize by Achilles. Briseis is shown here in the act of lamenting the death of Patroklos, the best friend of Achilles (Iliad XIX 282–302):

          Βρισηῒς δ’ ἄρ’ ἔπειτ’ ἰκέλη χρυσέῃ Ἀφροδίτῃ
ὡς ἴδε Πάτροκλον δεδαϊγμένον ὀξέϊ χαλκῷ,
ἀμφ’ αὐτῷ χυμένη λίγ’ ἐκώκυε, χερσὶ δ’ ἄμυσσε
285    στήθεά τ’ ἠδ’ ἁπαλὴν δειρὴν ἰδὲ καλὰ πρόσωπα.
εἶπε δ’ ἄρα κλαίουσα γυνὴ ἐϊκυῖα θεῇσι·
Πάτροκλέ μοι δειλῇ πλεῖστον κεχαρισμένε θυμῷ
ζωὸν μέν σε ἔλειπον ἐγὼ κλισίηθεν ἰοῦσα,
νῦν δέ σε τεθνηῶτα κιχάνομαι ὄρχαμε λαῶν
290    ἂψ ἀνιοῦσ’· ὥς μοι δέχεται κακὸν ἐκ κακοῦ αἰεί.
ἄνδρα μὲν ᾧ ἔδοσάν με πατὴρ καὶ πότνια μήτηρ
εἶδον πρὸ πτόλιος δεδαϊγμένον ὀξέϊ χαλκῷ,
τρεῖς τε κασιγνήτους, τούς μοι μία γείνατο μήτηρ,
κηδείους, οἳ πάντες ὀλέθριον ἦμαρ ἐπέσπον.
295    οὐδὲ μὲν οὐδέ μ’ ἔασκες, ὅτ’ ἄνδρ’ ἐμὸν ὠκὺς Ἀχιλλεὺς
ἔκτεινεν, πέρσεν δὲ πόλιν θείοιο Μύνητος,
κλαίειν, ἀλλά μ’ ἔφασκες Ἀχιλλῆος θείοιο
κουριδίην ἄλοχον θήσειν, ἄξειν τ’ ἐνὶ νηυσὶν
ἐς Φθίην, δαίσειν δὲ γάμον μετὰ Μυρμιδόνεσσι.
300    τώ σ’ ἄμοτον κλαίω τεθνηότα μείλιχον αἰεί.
Ὣς ἔφατο κλαίουσ’, ἐπὶ δὲ στενάχοντο γυναῖκες
Πάτροκλον πρόφασιν, σφῶν δ’ αὐτῶν κήδε’ ἑκάστη.

Then Briseis, looking like golden Aphrodite,
saw Patroklos all cut apart by the sharp bronze, and, when she saw him,
she poured herself all over him in tears and wailed with a voice most shrill, and with her
hands she tore at
285    her breasts and her tender neck and her beautiful face.
And then she spoke, weeping, this woman who looked like the goddesses:
“O Patroklos, you have been most gracious to me in my terrible state and most gratifying to
my heart.
You were alive when I last saw you on my way out from the shelter {22|23}
—and now I come back to find you dead, you, the protector of your people
290    —that is what I come back to find. Oh, how I have one misfortune after the next
to welcome me.
The man to whom I was given away by my father and by my mother the queen
—I saw that man lying there in front of the city, all cut apart by the sharp bronze,
and lying near him were my three brothers—all of us were born of one mother –
they are all a cause for my sorrow, since they have all met up with their time of destruction.
295     No, you did not let me—back when my husband was killed by swift-footed Achilles,
killed by him, and when the city of my godlike Mynes [= my husband] was destroyed by him
– you did not let me weep, back then, but you told me that godlike Achilles
would have me as a properly courted wife, that you would make that happen,
and that you would take me on board the ships,
taking me all the way to Phthia, and that you would arrange for a wedding feast
among the Myrmidons.
300    So now I cannot stop crying for you, now that you are dead, you who were always
so sweet and gentle.”
So she [= Briseis] spoke, weeping, and the women kept on mourning in response.
They mourned for Patroklos, that was their pretext, but they were all mourning, each and
every one of them, for what they really cared for in their sorrow.

In the logic of the epic narrative here, Briseis is not just weeping, not just speaking words of sorrow. She is represented as singing a lament (Dué 2002:70–71, 81; 2006:43–44). And the words of her lament are quoted inside the epic narrative. Following this quotation in Iliad XIX (287–300) is the quotation of another lament for Patroklos, this one performed by the hero Achilles himself (315–337); here too, as in the case of Briseis, Achilles is represented as singing a lament (Tsagalis 2004:86, 139–140).

Such quotations do not represent the actual meter of lament as sung in real laments, since the genre of epic regularly uses its own meter, which is the hexameter, in representing other genres that it quotes, including unmetrical genres (Martin 1989:12–42; also pp. 87–88, specifically on lament). In other ways, however, as we see clearly from the words of Briseis, such quotations do represent the three characteristic features of lament as I outlined them earlier:

  1. Briseis, even though she is not the wife of Achilles but merely his war prize and potential war bride, is the woman most closely affected by the death of Patroklos, the best friend of Achilles, and so it is she who assumes the role of prime mover in the performing of lament for Patroklos.
  2. As the prime mover of lament, Briseis becomes the lead singer in the performing of lament for Patroklos. As the lead singer, she interacts with an ensemble of other women who had also been captured as war prizes. The ensemble is shown in the act of responding to her song of lament by continuing it with their own lament, in antiphonal performance: Ὣς ἔφατο κλαίουσ’, ἐπὶ δὲ στενάχοντο γυναῖκες “So she [= Briseis] spoke, and the women kept on mourning in response” (XIX 301). The verb epi –stenakhizesthai, which I translate here as “keep on mourning in response,” is the conventional way for epic to refer to any form of antiphonal performance. {23|24}
  3. In singing her song of lament, Briseis as lead singer touches on her feelings as a captive woman who has become the war prize of Achilles—and who hopes to become his war bride. She also touches on the projected feelings of the ensemble of captive women who respond to her lament in antiphonal song. These women too are war prizes, and they must therefore share in some ways the sorrows felt by the lead singer as she sings her lament. But the lead singer laments primarily the death of Patroklos and only secondarily her own misfortunes, while the ensemble of women who respond in antiphonal song are lamenting primarily their own misfortunes and only secondarily the death of Patroklos. Sorrow over the death of Patroklos seems to be the primary concern of Briseis—to the extent that her lament projects the sorrow of Achilles, which is a driving theme in the plot of the epic. By contrast, the sorrow expressed by the ensemble of captive women over their own misfortunes seems to be a primary concern only for them. Or is it? In the lament of Briseis, the sorrow of the captive women is projected as the primary sorrow of Briseis herself over her own misfortunes, which had been caused by the deaths of her former husband and her kinfolk at the hands of Achilles. Briseis shows that she remembers that old sorrow, since her wording indicates that she had wanted to lament her dead husband in the same way that she now laments the dead Patroklos. That death in her past is relevant to the death of Patroklos in the present. And the love of Briseis for her former husband and her kinfolk is relevant to her love for Patroklos as a stand-in for Achilles. There is a diversity of emotions here. And the antiphonal exchange of laments between the captive women and their lead singer leads to a communalization of these emotions. The example of Briseis, then, supports the argument that lament is a communalizing experience. It leads here to a communalization of diverse emotions.
Such communalization in antiphonal exchanges of laments between lead singer and ensemble can be more straightforward. A salient example is the story of another lead singer in the Homeric Iliad, Andromache. On three distinct occasions, the epic narration shows this woman in the act of singing a lament as a lead singer, and, each time, her words of lamentation are quoted. On the first occasion, she is lamenting the future death of her husband, Hector (VI 407–439); on the second occasion, she laments him when she first sees him dead, witnessing the horrific sight of Hector’s corpse being dragged behind the chariot of Achilles (XXII 477–514); then, on the third occasion, she laments him on the formal occasion of his funeral in Troy (XXIV 725–745). On each of these three occasions, the lament sung by Andromache as lead singer is followed by antiphonal lament performed by an ensemble of women attending her (VI 499–500, XXII 515, XXIV 746). After two of the three laments performed by Andromache, the antiphonal lament of the women is described in these words: Ὣς ἔφατο κλαίουσ’, ἐπὶ δὲ στενάχοντο γυναῖκες “So she [= Andromache] spoke, weeping, and the women kept on mourning in response” (XXII 515, XXIV 746). We have already seen exactly the same words describing the antiphonal lament performed by the ensemble of women responding to the lament of Briseis as lead singer: Ὣς ἔφατο κλαίουσ’, ἐπὶ δὲ στενάχοντο γυναῖκες “So she [= Briseis] spoke, and the women kept on mourning in response” (XIX 301). {24|25}
In these examples of lamentation, we have seen two different kinds of community being represented by the ensemble of women who respond antiphonally to the song of lament performed by the lead singer. In the case of the three different ensembles of women responding to Andromache on three different occasions of lament, the community they represent is Troy. In the case of the women responding to Briseis, on the other hand, the community they represent is mixed: just like Briseis, who is a captive woman and a war prize of the enemies of Troy, they too are captive women and war prizes, but their feelings of solidarity with each other are mixed with their feelings of solidarity with Briseis, whose love for Patroklos and Achilles, the greatest enemies of Troy, is now driving her to cross over to the other side. Mixed feelings, mixed community. And these mixed feelings of community find their self-expression in the antiphonal exchange of lament, which communalizes these feelings into a larger sense of community.
In all the examples of lament as narrated within the genre of epic, the narration never shows the actual words performed by any ensemble in antiphonal response to the feelings of the lead singer. And so the feelings of the ensemble are never expressed directly. They are only projected in the words performed by the lead singer in antiphonal exchange with the ensemble.
By contrast, representations of lament within the genre of tragedy do show the actual words being performed by the ensemble who sing their antiphonal response to the lament sung by the lead singer—and who thus express their own feelings directly. We see a striking example at the beginning of the Trojan Women of Euripides, where a lengthy outburst of lead singing in the lament by Hecuba (98–152) is followed by briefer outbursts of her lead singing, each of which is followed in turn by antiphonal responses from an ensemble of Trojan women (153–234). As the drama proceeds, there are further lead songs of lament performed by Hecuba, followed by further antiphonal responses performed by the Trojan women (as at 278–291 / 292–293, 1240–1250 / 1251–1259). Not only does tragedy show in such cases the actual wording of the antiphonal responses performed by the ensemble of Trojan women: it also represents the actual performance of such responses. The performers of the representation are the male chorus of tragedy, who sing and dance what a female chorus would be expected to sing and dance in real laments. The underlying reality in such a representation is this: women who sing and dance antiphonal responses in real lament are a chorus in their own right (Dué 2006:12; also p. 22 with reference to the work of Calame 2001 on female choruses).
There is a parallel underlying reality in the representation of lead songs of lament performed by Hecuba in the Trojan Women of Euripides: these performances represent what a female lead singer would sing in a real lament. Whereas the antiphonal laments that respond to the laments of Hecuba are performed by a male chorus that sings and dances what the represented chorus of Trojan women sings and dances, the laments of Hecuba as the represented lead singer are performed by a male actor who sings or recites the represented songs of lament. He sings the laments composed in choral meters, as in the case of the first lament of Hecuba (98-152), while he recites {25|26} the laments composed in iambic trimeter, which is the meter that represents spoken rather than sung words, as in the case of a later lament by Hecuba (1240–1250).
Such an actor of tragedy needs to be a virtuoso singer as well as an actor. That is because the artistic complexity we find in most of the laments sung by Hecuba as lead singer, especially in her very first lament, requires such virtuosity. Here I return to this first lament as sung by Hecuba in the Trojan Women (98-152). This lament is performed as a monody. The term monody can be defined in general as a medium of song intended for solo performance. In the case of tragedy, however, monody needs a more narrow definition: monody in tragedy is a medium of song intended for solo performance by a professional actor who is also a virtuoso singer. To be contrasted with monodic songs composed for virtuoso monodic performance in tragedy are choral songs, that is, songs composed for singing and dancing by the tragic chorus. Such choral songs, as distinct from monodic songs, require no special virtuosity. That is because the members of the male chorus of tragedy were non-professionals, unlike the actors of tragedy, who were professionals. In the fifth century BCE, which we know as the classical era of Athenian tragedy, the non-professionalism of the tragic chorus was an essential part of the civic ideology of Athenian State Theater (Nagy 1990a:382–413).
A master of classical tragedy like Euripides, no matter how ambitious he may have been as a composer, had to work under severe constraints in composing the music to be sung and danced by the male chorus assigned to him by the State. There are broad hints in ancient sources (as in the Aristotelian Problems, 19.15) about the frustrations experienced by masters of tragedy in having to compromise their own virtuosity as composers in order to accommodate various limitations in the singing and dancing abilities of their various non-professional performers (Nagy 1990a:343). One way out for a creative composer like Euripides was to concentrate his musical ambitions on the composition of monody, to be performed by a professional actor who doubled as a virtuoso singer. After all, the essence of monody in the tragedies of Euripides was the virtuosity it showed in its performance. It is in this light that we need to consider the monody sung by the professional actor who represented Hecuba as the lead singer of the chorus of Trojan women in the original production of the Trojan Women of Euripides.
From what we have seen so far, we already know that monodic performance, as it exists in tragedy, is not restricted to any single metrical structure. This fact needs to be connected with another fact: in ancient Greek literature, many genres exist only in the form of monodic performance, not choral performance. One such genre, it is relevant to note, is elegy. Here I return to my earlier working definition: monody is a medium of song intended for solo performance. In terms of this definition, elegy is a monodic medium. By contrast, lament is not an exclusively monodic medium. It is also a choral medium.
This formulation is relevant to what we have just been considering, the monodic laments performed by Hecuba in the Trojan Women. And it is even more relevant to what we are about to consider, a monodic lament performed by Andromache in another tragedy. This lament, as we will see, is an elegy. To say it another way, it is a composition consisting of elegiac couplets. {26|27}
The elegy occurs near the beginning of a tragedy of Euripides known as the Andromache, which was first produced sometime in the last third of the fifth century BCE. In this elegy, we are about to see the figure of Andromache in the act of singing a song of lament as an expression of her sorrow over her misfortunes. Now that her husband Hector has been killed and Troy has been destroyed, this beautiful and aristocratic princess has been reduced to the degrading status of a captive woman, a war prize taken from Asiatic Troy to the Helladic city of Phthia, where she is fated to become the slave of the evil princess Hermione. Andromache sings her lament in the form of a monody (Euripides, Andromache 91–117):

Αν.     … ἡμεῖς δ’ οἷσπερ ἐγκείμεσθ’ ἀεὶ
θρήνοισι καὶ γόοισι καὶ δακρύμασιν
πρὸς αἰθέρ’ ἐκτενοῦμεν· ἐμπέφυκε γὰρ
γυναιξὶ τέρψις τῶν παρεστώτων κακῶν
95      ἀνὰ στόμ’ αἰεὶ καὶ διὰ γλώσσης ἔχειν.
πάρεστι δ’ οὐχ ἓν ἀλλὰ πολλά μοι στένειν,
πόλιν πατρώιαν τὸν θανόντα θ’ Ἕκτορα
στερρόν τε τὸν ἐμὸν δαίμον’ ὧι συνεζύγην
δούλειον ἦμαρ ἐσπεσοῦσ’ ἀναξίως.
100    χρὴ δ’ οὔποτ’ εἰπεῖν οὐδέν’ ὄλβιον βροτῶν,
πρὶν ἂν θανόντος τὴν τελευταίαν ἴδηις
ὅπως περάσας ἡμέραν ἥξει κάτω.
᾿Ιλίωι αἰπεινᾶι Πάρις οὐ γάμον ἀλλά τιν’ ἄταν
ἀγάγετ’ εὐναίαν ἐς θαλάμους Ἑλέναν.
105    ἇς ἕνεκ’, ὦ Τροία, δορὶ καὶ πυρὶ δηϊάλωτον
εἷλέ σ’ ὁ χιλιόναυς Ἑλλάδος ὠκὺς Ἄρης
καὶ τὸν ἐμὸν μελέας πόσιν Ἕκτορα, τὸν περὶ τείχη
εἵλκυσε διφρεύων παῖς ἁλίας Θέτιδος·
αὐτὰ δ’ ἐκ θαλάμων ἀγόμαν ἐπὶ θῖνα θαλάσσας,
110    δουλοσύναν στυγερὰν ἀμφιβαλοῦσα κάραι.
πολλὰ δὲ δάκρυά μοι κατέβα χροός, ἁνίκ’ ἔλειπον
ἄστυ τε καὶ θαλάμους καὶ πόσιν ἐν κονίαις.
ὤμοι ἐγὼ μελέα, τί μ’ ἐχρῆν ἔτι φέγγος ὁρᾶσθαι
Ἑρμιόνας δούλαν; ἇς ὕπο τειρομένα
115    πρὸς τόδ’ ἄγαλμα θεᾶς ἱκέτις περὶ χεῖρε βαλοῦσα
τάκομαι ὡς πετρίνα πιδακόεσσα λιβάς.
ΧΟΡ.  ὦ γύναι, ἃ Θέτιδος δάπεδον καὶ ἀνάκτορα θάσσεις
δαρὸν οὐδὲ λείπεις,
Φθιὰς ὅμως ἔμολον ποτὶ σὰν Ἀσιήτιδα γένναν, …

But I, involved as I am all the time in laments [thrēnoi] and wailings [góoi]
and outbursts of tears,
will make them reach far away, as far as the aether. For it is natural
for women, when misfortunes attend them, to take pleasure [terpsis]
95      in giving voice to it, voicing it again and again, maintaining the voice from one mouth to the
next, from one tongue to the next.
I have here not one but many things to mourn:
I mourn the city of my fathers. I mourn Hector, dead. {27|28}
And I mourn the rigid fate allotted to me by an unnamed force [daimōn], a fate
to which I am yoked,
having fallen captive to a life of slavery—so undeserved!
100    You must never call any mortal blessed [olbios]
before he dies and you see him on his last day alive,
and you see how he lives through that day before he finally goes down below.
To Ilios [= Troy] with its steep walls did Paris bring not a wedding to be celebrated
but some kind of aberration [atē]
when he brought to the wedding chamber, as his partner in bed, Helen herself.
105    Because of her, O Troy, by spear and fire were you captured by the enemy.
Seized you were by the thousand ships of Hellas sent by swift Ares,
and so also was my husband Hector taken from me, wretched that I am.
Around the walls [of Troy]
was he dragged from the chariot driven by the son of the sea-dwelling Thetis.
And then I myself was taken out of my chamber and brought to the shore of the sea.
110    Hateful slavery did I place as headwear upon my head.
And many a tear came falling, all over the complexion of my face as I left behind
my city and my chamber and my husband lying in the dust.
I cry O for me, wretched that I am! Why did I have to see the light of day
as a slave of Hermione? Worn down by her domination,
115    to this statue of the goddess do I come as a suppliant, embracing it with both hands,
and I dissolve [tēkesthai] (into tears) like a stream that flows from a spring in the rocky
heights.
chorus
My lady, you who have been sitting there on the sacred ground and precinct of Thetis
for some time now, unwilling to leave,
I, a woman from Phthia, have come, approaching you, a woman born in Asia …

We come to a decisive moment in the ongoing argumentation about the relationship of elegy and lament as we contemplate the monodic singing of this elegy here in the Andromache of Euripides. Here it all comes together. The elegiac couplets at lines 103–116, as sung by the actor who represents Andromache in the act of lamenting her misfortunes, show elegy in perfect convergence with lament.

At lines 91–102, before the elegiac couplets get started at lines 103-116, the actor who represents Andromache is speaking in iambic trimeter, which is the meter that represents spoken rather than sung words (as we saw already in the lament of Hecuba at lines 1240–1250 of the Trojan Women of Euripides). Then, at line 103, we see an abrupt switch (as marked by the absence of an expected word of syntactical connection) from speaking in iambic trimeter to singing in elegiac couplets. The singing of the elegy gets started at line 103 and continues through line 116.
It is particularly relevant to elegy that the lamenting words of Andromache highlight what she calls her fate, as allotted by some unknown force, at line 98. In Modern Greek traditions of women’s lament, the actual word for “lament” is μοιρολόγι, which means “speech of fate” (Nagy 1999a:80n; decisive for this interpretation of the meaning is the dialectal variant form μορολόγι). This word, in and of itself, is a most eloquent indication of the extraordinary capacity of lament in making relevant to a given moment of sorrow all the other sorrowful moments of life, and, far more than that, in making relevant the sum total of life’s lived experiences. We see {28|29} in this capacity another vital point of contact between lament and elegy, as we pursue the ongoing argument that elegy evolved out of lament.
The elegy of Andromache in the Andromache of Euripides actually shows the evolutionary relationship between lament and elegy. We see it in the simple fact that this elegy is complete as an elegy, but it is incomplete as a lament. The elegiac singing of Andromache is not a complete performance of lamentation. It is only part of a performance, as we see from what happens after Andromache finishes her song of lament. This lament, which is a monody sung by the tragic actor representing Andromache, leads into an antiphonal lament, sung and danced by the chorus of tragedy. At line 117, the hexameter of what was expected to be the next elegiac couplet is picked up as the first line of a choral song—a song that is meant to be sung and danced by the chorus. This choral song is decidedly not elegiac in form. From line 117 onward, there are no more elegiac couplets to be heard, since the hexameter at this line will not be followed by a pentameter. Rather, this hexameter at line 117 is the beginning of strophe α of a choral song consisting of trochaic as well as dactylic rhythms; then, the hexameter at line 126 is the beginning of antistrophe α. The choral rhythms are maintained by the hexameters at lines 135, and 141, which are respectively the beginnings of strophe β and antistrophe β of the choral song.
So, the elegy of Andromache, as a monody, is only a part of the lament here, since the choral song of the women of Phthia continues where Andromache left off (Allan 2000:55–57; Faraone 2008:127–137). From the plot of the tragedy, we can see how the continuation comes about. An ensemble of local women from Phthia has just arrived at the scene of action, finding there a foreign woman from Troy, Andromache. She is in a state of abject isolation, lamenting her misfortunes as she sings her monody of sorrow. The local women react to the sorrow by singing and dancing an antiphonal lament in response to the monodic lament sung by Andromache. This sharing of sorrow between the two sides, by way of the monody sung by Andromache on one side and the choral song sung and danced by the local women on the other side, leads to a communalization of emotions, in all their diversities.
By way of this communalization, the local women from Phthia have in effect accepted the foreign woman from Troy as their lead singer. For the male chorus who represents this ensemble of women, the actor and virtuoso singer who represents Andromache in singing her lament is in fact the lead singer, since the monody he sings leads into the singing and dancing of the chorus.
Such communalization of emotions is anticipated in the initial lamenting words of Andromache herself, since she speaks about lamentation not only in terms of her own monodic lament but also in terms of antiphonal lament—which is the kind of lament that the tragic chorus will be performing when it arrives on the scene. In her state of abject isolation, the figure of Andromache speaks openly about expressing her sorrow in lament, saying that the voicing of lament is a woman’s way of sharing her sorrow with other women, who will then continue the lament by voicing their own sorrows, “maintaining the voice from one mouth to the next, from one tongue to the next” (95). These lamenting words of Andromache, which merely imagine the antiphonal lament of a chorus of women, anticipate what will actually happen when {29|30} the monody of Andromache leaves off: the lead song of lament performed by Andromache (101–116) will now be taken up by an antiphonal song of lament that is no longer imagined but actually performed (117–146). And the performers of this antiphonal lament are no longer an imaginary chorus but the tragic chorus that sings and dances in the tragedy of Euripides.
When Andromache in her song of lament actually speaks of women’s songs of lament, there are two words used in speaking of such songs, thrēnoi and góoi (the singular forms are thrēnos and góos). This usage is telling us something about the special ways in which the genre of tragedy refers to lament. To understand these special ways, we must first consider how these same two words are used in the genre of epic.
In epic, we learn that thrēnos and góos refer to two distinct genres of lament: thrēnos refers to lament as performed by men who are professional singers, while góos refers to lament as performed by non-professional singers, whether they are men or women (Nagy 1999:112). On the occasion of Hector’s funeral in the Iliad (XXIV 720–723), the word thrēnoi (721) marks the songs of men described as aoidoi or “singers” (720), while góos (723) marks the songs of women led by Andromache as the first of three lead singers set to perform a lament (again, 723); the second lead singer is Hecuba (747), and the third is Helen (761). The laments of the lead singers in each case lead into antiphonal singing and dancing, as indicated by the verb epi–stenakhizesthai / epi–stenein “keep on mourning in response” with reference to those who perform laments in response to the lead singing; those performers are women (722, 746) or, as we will see in more detail later, a combination of men and women (776). In the case of the thrēnoi, the lead singing is indicated by the noun exarkhoi or “leaders” with reference to the professional male lead singers (721). In the case of the góos, the lead singing is indicated by the verb exarkhein / arkhein or “lead” with reference to the non-professional female lead singers Andromache (723: arkhein), Hecuba (747: exarkhein), and Helen (761: exarkhein). Besides the funeral of Hector, there is another occasion where the word góos is used with reference to both the lead singing and the antiphonal performance of lament: this time, it is at the funeral of Patroklos, where the performers of the góos are not women but men (XXIII 10, 12), whose mourning is led by Achilles himself (XXIII 12, 17: arkhein). In epic narrative, antiphonal lamenting may be performed not only by women representing a community but even by the community itself, in its entirety. Toward the very end of the Iliad, at the funeral of Hector, when Helen finishes performing her lead song of lament, the antiphonal response to her lament is performed not by an ensemble of women but by the entire community or dēmos of the Trojans (XXIV 776). Evidently, the antiphonal response is being performed here by men as well as women, and thus the lament of women is being transformed into “a collective civic sorrow” (Dué 2006:44; also Tsagalis 2004:61 on Iliad XXII 429, where antiphonal lament is performed by the politai or “city-dwellers” of Troy in response to the lament of Priam).
By contrast with epic, which shows a distinction between thrēnos and góos as genres, we see no such distinction in the monody of Andromache as represented in the Andromache of Euripides. Both words thrēnos and góos refer in this monody to laments performed by women. Nor can we find a clear distinction between thrēnos {30|31} and góos anywhere else in tragedy. For example, in the tragedy of Euripides known as Iphigeneia in Tauris (143–147), the male actor who represents Iphigeneia sings in a monody: ἰὼ δμωαί, | δυσθρηνήτοις ὡς θρήνοις | ἔγκειμαι, τᾶς οὐκ εὐμούσου | μολπᾶς ἀλύροις ἐλέγοις, αἰαῖ, | ἐν κηδείοις οἴκτοισιν “O my handmaidens! How involved I am in thrēnoi so sad for singing as thrēnoi, coming as they do from the kind of song that has no kindly Muse, with its elegoi having no lyre to accompany them, aiai, in their pitiful expressions of anxieties.” In the wording of this monody by Iphigeneia, thrēnos refers to non-professional female lament, not to professional male lament. Moreover, the word elegoi here means simply “mournful songs,” which cannot refer to the genre of elegy as we know it, since the monody of Iphigeneia is composed in choral meters and not in elegiac couplets; we have already seen another such use of the word elegoi in the stylized sense of “mournful songs” in the monody of Hecuba as sung in the Trojan Women of Euripides (119: δακρύων ἐλέγους).
An explanation for the ignoring of distinctions between the separate genres of thrēnos and góos in tragedy can be found in tragedy itself as a genre in its own right. This genre, like epic, is capable of representing other genres within itself. Unlike epic, however, the genre of tragedy represents other genres directly and in some ways more realistically. For example, the female form of lament known as góos in epic is represented directly in tragedy. The singing and dancing of the góos is shown in its entirety. To be seen and heard are not only the singing of the lead song performed by the lead singer but also the singing and dancing of the antiphonal song performed by the chorus. Also, tragedy represents these songs realistically by retaining their multiform meters. By contrast with what we see in the genre of tragedy, the genre of epic represents only indirectly and less realistically any genre other than itself. In the case of the góos as performed by women, for example, epic quotes within its narration only the words of the lead song performed by the lead singer. Also, epic represents these quoted words unrealistically by transforming their potentially multiform meters into the uniform meter of epic, the hexameter. As for the antiphonal song of the góos as performed by a chorus of singers and dancers, the genre of epic does not represent the words of the chorus, even indirectly, and it narrates the action of its antiphonal performance in a most reductive and thus unrealistic way by compressing an entire sequence of choral responses within the frame of a single word, epi–stenakhizesthai (alternatively, epi–stenein), in the stylized sense of “keep on mourning in response.”
Having seen how the genre of tragedy is more realistic than the genre of epic in its representation of lament, we can also see why it ignores the distinction between the separate genres of thrēnos and góos. Lead singers such as Andromache and Iphigeneia can refer to their laments as thrēnoi because their singing is actually performed by professional male singers who are comparable to the professional male singers of thrēnoi in the epic, not to the female singers of góoi, who are of course non-professionals. In brief, the conventions of Athenian State Theater, where males play female as well as male roles, makes the overlapping of these two genres of lament inevitable.
It is not enough to say, however, that tragedy ignores the distinction between professional male lead singing of thrēnoi and non-professional female lead singing of {31|32} góoi simply because professional male singers are used for such performances. More than that, tragedy ignores this and other such distinctions because it is understood to be the most definitive imaginable form of performance. Tragedy is the ultimate genre, from the standpoint of the civic ideology that shapes the festivals of Athens, especially its main dramatic festival, the City Dionysia, which is the main venue for Athenian State Theater. From the standpoint of this ideology, there is no need to make distinctions in referring to male or female lament, since the definitive form of lament is male by virtue of being performed in the all-male genre of tragedy (Nagy 1994/1995:21). Nor is there a need for Athenian State Theater to make distinctions in referring to Greek or non-Greek lament, as we see in the Persians of Aeschylus, since the definitive language of lament is Greek, not Persian (Ebbott 2000).
In short, the references to female lament in tragedy are in reality references to female lament only as represented by male lament. Such a representation of female lament by tragedy can be considered an act of male appropriation, to which I will refer hereafter as a masculinization of women’s lament.
With its power of direct representation, which is called mimēsis in Greek, tragedy is not only more realistic than epic in the ways it shows lament in action. It is also more deceptive, since this realism makes it all the more difficult to distinguish between genres of male and female lament embedded within tragedy as a genre of all-male performance. The masculinization of women’s lament in tragedy is so realistic that it becomes barely noticeable. That is what makes tragedy in Athenian State Theater seem so threatening in Plato’s Republic (III 395d–e). As we see from the analysis of Nicole Loraux (1998:10–11), the perceived threat is that men who represent the laments of women in tragedy will start to talk and think and even feel like women, not like the men they really are. In the masculinization of women’s lament, from Plato’s point of view, the danger is that men’s lament can in turn be feminized.
Now that we can appreciate more fully the power of tragedy in representing women’s laments realistically, I am ready to make three points about the elegy sung by Andromache at lines 103–116 in the Andromache of Euripides:
  1. The elegy of Andromache, sung within the context of a lament, can be considered a realistic representation of a lead song of a real lament as really sung by women. Such a lead song is monodic, leading into an antiphonal song of lament by a chorus. This conclusion supports the theory of Page (1936), who argued that the elegiac couplets of the elegy sung in the Andromache of Euripides represent elegiac couplets as actually performed by women. In terms of this argument, “elegiac couplets sung in stanzas were used by women for a number of ritual purposes, including lament and hymns” (Faraone 2008:137, citing Alexiou 2002:131–134 on “common features of lament and hymn”). In the Hellenistic period, which extends from the fourth into the first century BCE, such a form of women’s singing in elegiac couplets seems to be imitated in learned poetry; a case in point is Hymn 5 of Callimachus, who flourished in the third century BCE (Bulloch 1985).
  2. The elegy of Andromache, if it is taken out of its context as a lament sung by a woman, can be considered a realistic representation of real elegy as really sung by men. By {32|33} real elegy I mean the existing form of elegy as we see it attested in the historical period of Greek literature starting from the seventh century BCE. As we are about to see in the analysis that follows, real elegy was strictly monodic, without a choral follow-up. And real elegy, in its monodic form, could be performed only by men.
  3. The elegy of Andromache corresponds to a pre-existing form of elegy as represented in epic. This form is the thrēnos, a monodic lead song performed by men, which leads into choral song performed by either men or women. This pre-existing form can be considered a pre-genre, to which I will refer as threnodic elegy in order to distinguish it from the existing genre of elegy. Unlike the monody of elegy, which is cut off and disconnected from any follow-up in choral performance, the threnodic elegy still retains its choral connectivity, as we can see from the representations of this genre in epic.
In brief, then, the elegy of Andromache composed by Euripides for his Andromache is exceptional in ancient Greek literature because its form represents three different but related genres: (1) lament as performed by women, (2) elegy as performed by men, and (3) threnodic elegy as performed by men who are professional singers. The representation of all three of these genres must have seemed most realistic to the audiences of Euripides, since he is ridiculed in the Frogs of Aristophanes (1301–1303) for heavy-handedly inserting what is called the thrēnos into his tragedies (Nagy 1990a:403).
The differences we see in these three genres stem from a basic tendency in the evolution of poetic forms in the prehistory of Greek literature. That tendency can be described as an ongoing differentiation of specialized forms of solo performance evolving out of unspecialized forms of group performance. What results is the emergence of specialized forms that exist only in solo performance, cut off from the corresponding unspecialized forms that continue to exist in group performance. To put it in terms of attested forms of performance in the fifth century BCE, a period of time that coincides with the classical phase of ancient Greek literature, solo performance becomes monodic while group performance remains choral (Nagy 1990a:85, 340–341). Almost all of the poetry we find surviving from the literature of this period has been composed exclusively either for monodic or for choral performance.
An exception is tragedy, along with other dramatic forms as attested primarily in the context of Athenian State Theater in the fifth century. Within the framework of tragedy and other drama, the monodic and the choral dimensions of performance are still connected, as we saw for example in the elegy of the Andromache of Euripides.
Outside the framework of tragedy and other drama, however, in all other forms of attested poetry and song making, the medium of monodic performance had already become disconnected from the medium of choral performance. A primary cause was the ongoing professionalization of monodic poetry and song making in the classical phase of Greek literature in the fifth century BCE. By this time, a man needed to be a professional singer in order to have the skills required for singing monody; and, if he happened to be an elite amateur who aspired to be able to perform monodic song in front of his fellow elites at special private events like symposia, he needed to be properly educated by professionals in order to develop the skills of monodic singing. {33|34} Only the medium of performing choral poetry remained unprofessionalized in the classical phase of Athenian State Theater. Amateurs recruited for performing in the civic event of a drama would be expected to have the minimal skills needed for the choral singing and dancing that the composer-director of the given drama would teach them. (The relevant evidence for the formulations in this paragraph is surveyed in Nagy 1990a:378–379, 406–410).
Even choral performance became professionalized in the post-classical phase of Athenian State Theater in the fourth century BCE (Nagy 1996a:173–174n74). And the trend of professionalization must have been already underway in the earlier classical phase, since any choral song featuring a lead singer who sings a monody in concert with the chorus would have required a professional actor for the lead singing. This formulation can be extended further to the professional actor of drama in general, since his professionalized function as actor had evolved from the non-professional function of lead singer in the chorus. The actor who performs as a virtuoso singer of monody can be considered a revitalized offshoot of such a lead singer.
There is one genre that remains exempt from the professionalization of monodic performance. It is the genre of women’s lament. And the same can be said about all other forms of women’s song besides songs meant for lament, such as love songs meant for courtship rituals and weddings, hymnic songs for worshipping the gods, and so on (Tsagalis 2004:82–85; Dué 2006:20). The performance of such songs by women could always remain unprofessionalized. But there is an essential qualification to be noted here: the performance of such songs could remain unprofessionalized if the performers were women. These same forms of performance could become professionalized, however, if the performers were women as represented by men.
A salient example of such professionalization by way of masculinization is the poetry and song making of Sappho, whose life and times are conventionally dated to the late seventh and early sixth centuries BCE. Songs attributed to Sappho, stemming from the song making traditions of the local women of her native island of Lesbos, could be sung not only by women but also by men, and there were two basic social contexts for such singing by men, namely, the symposium and the public festival (Nagy 2007:217–226; see in general the fundamental work of Yatromanolakis 2007 on the reception of Sappho). The tradition of singing Sappho’s songs, which required virtuoso lead singing that leads into choral singing and dancing, became professionalized as a tradition of monodic singing as performed by men at symposia, first already in Lesbos and then later on in places like Samos and still later on even in Athens, as of the early sixth century and continuing all the way into the fifth; in the case of Athens, we have reason to believe that Sappho’s monodic songs were performed there not only at symposia but also, eventually, at the major public festival of the Athenians, the Panathenaia (Nagy 2007:243–246).
The professionalization—and masculinization—of Sappho’s monodic songs in the two social contexts of the symposium and the public festival is comparable to what we find in the history of elegy. There are two historical facts about elegy that establish the comparison: {34|35}

  1. There were two basic social contexts for the singing of elegy by men, namely, the symposium and the public festival (Bowie 1986:14–21, 34)
  2. As we have already seen, elegy was a monodic medium, restricted to singing by men, but the actual form of the elegiac couplet had a broader range, since it was not restricted to singing by men. Within the broader context of threnodic elegy, the choral lead singer could sing in elegiac couplets, to be followed up by the singing and dancing of the chorus, and such lead singing could be performed by women as well as by men.

Using the analogy of Sappho’s monodic songs, then, we can say that the formation of elegy as a genre stems from traditions of monodic performance by men at symposia. Such monodic performance, in the context of the symposium, can be described as a medium that has been disconnected from choral performance, that is, cut off from its roots in the chorus. The medium of choral lead singing, which leads into the singing and dancing of the chorus, loses its original choral context in the process of becoming transformed into the newer medium of monodic singing at symposia. Such a loss of choral context in the newer medium affects what I described earlier as a communalization of emotions between the choral lead singer and the chorus in the older medium. In the newer medium of monodic singing at symposia, such communalization is taken out of context, decontextualized.

Such decontextualization of shared emotions in monodic singing at the symposium is only partial, however, since the monody itself can represent a communalization of emotions even without the presence of a real chorus. In the context of monodic singing at a symposium, the choral context can be partially recontextualized by way of representing the shared emotions of a chorus that is no longer there. The monody can still pretend to have a choral follow-up, a choral response. But now there is simply one voice that expresses and represents what is being felt, both in the singing of an imagined lead singer and in the antiphonal response of an imagined chorus. And the chain of recontextualization in monody can extend even further, from the symposium into the public festival. The choral lead singing that is recontextualized by being represented in monodic singing at symposia can be further recontextualized by being represented in monodic singing at public festivals, where the singing takes the form of competitive performances by professional singers called kitharōidoi or “citharodes,” performers who sing lyric songs to the accompaniment of the seven-string lyre or kithara (Nagy 2004:37–38).
On the island of Lesbos, in the context of public festivals, citharodes would sing the songs of Sappho as a model of women’s lead singing, in the context of an imagined chorus, and the songs of Alcaeus as a model of men’s monodic singing, in the context of an imagined symposium. Similarly, in the context of a real symposium, men would sing the songs of either Sappho or Alcaeus or other models of monodic singing. And, beyond their original setting in Lesbos, these songs were appropriated and maintained in Athens, where they kept on being performed both in the less formal context of the symposium and in the more formal context of the public festival of the Panathenaia (Nagy 2007:243–246). {35|36}
Corresponding to the citharodes, whose monodic singing of lyric songs was accompanied by the kithara, there were professional singers called aulōidoi or “aulodes,” whose monodic singing of corresponding songs was accompanied by the aulos, and we know for sure that these two forms of monodic singing, citharody and aulody, were symmetrically institutionalized at public festivals like the Panathenaia (Nagy 1990a:104). And, though we do not know for sure, we have reason to think that aulodes who sang to the accompaniment of the aulos in aulodic competitions at the Panathenaia would be singing songs composed in elegiac couplets. In other words, they would be singing in the genre of elegy. To back up this line of thinking about the singing of elegy at public festivals, I point to the cumulative evidence showing that the singing of elegy at symposia was conventionally accompanied by the aulos (Bowie 1986:27).
In terms of the ongoing argument, then, the traditions of aulody stemmed from the singing of men’s and women’s elegy (1) by amateur men performing for each other at symposia and (2) by professional aulodes performing for civic assemblies at public festivals. And the repertoire of men singing elegy at symposia and at public festivals would include songs composed in elegiac couplets as sung by women who were lead singers of lament. Such songs composed in elegiac couplets could have included elegiac songs of Sappho. There is an attested reference (Oxyrhynchus Papyri 1800 fr. 1) to an ancient edition of Sappho’s songs in which one of the scrolls contained a whole collection of her elegeia or “elegies”; there is also a reference to her elegeia in the Byzantine encyclopedia known as the Suda (Σ 107).
We have noted that the sharing of emotions in the monody of elegy has been taken out of its earlier context, which was a mournful exchange between lead singer and chorus. The occasion for such an earlier context, as we have also noted, was lamentation over personal misfortunes, and the most common example of such lamentation was the mourning of the death of a loved one. For elegy, however, both the context and the occasion have changed. The context is no longer an exchange between lead singer and chorus, and the immediate occasion is no longer the coming together of mourners at a funeral. Instead, the context is now monodic singing, and the immediate occasion is the coming together of revelers at a symposium, where men drink wine together as they recline on sympotic couches and enjoy each other’s company while conversing or taking turns in singing monodic songs.
In general, monodic singing at symposia would feature either virtuoso citharodic singing, where the singer accompanies himself on the kithara, or, more commonly, aulodic singing, where the singer is accompanied by a girl or boy playing on the aulos—or even by a fellow symposiast playing as a spontaneous accompanist. In particular, one special kind of aulodic singing at symposia would be the monody of elegy, that is, of singing in elegiac couplets to the accompaniment of the aulos.
All this is not to say that a man could not lament a death in a monody sung in elegiac couplets at a symposium. He could, but such a lament would not be the lament he would sing on the immediate occasion of, say, mourning a death at a funeral. His lament would now be sung on the immediate occasion of a symposium, where he would be sharing his emotions not with fellow mourners but with his fellow {36|37} symposiasts. Yes, he could still be singing the kind of lament he would sing at a funeral, or even the kind of lament that his women kinfolk would sing at that funeral. But his lament would be decontextualized. His lament would not be real lament but elegy.
In real lament, as still attested in recordings of songs sung by women who have kept up the traditions of singing lament in the surviving song cultures of the Greek-speaking world today, we find a stunningly wide diversity of ways in which to express emotions. Besides the countless different ways of expressing the most relevant primary emotion of sorrow over the death of a loved one and over all the sufferings occasioned by that death, there are also countless different ways of expressing another relevant primary emotion, and that is love—love not only for the beloved dead but also for any of the living who are mourning both for the dead and for each other. Such expressions of love in Greek lament can even be transferred to expressions of love in love songs sung on occasions such as courtship rituals and weddings (Dué 2006:20). And the diversity of other relevant emotions extends to expressions of the mourner’s diverse feelings of fear, anger, or hatred as well. All these diverse ways of expressing emotions are tied to the singing woman’s full range of lived experiences. In short, there is a remarkably wide range of possibilities for the self-expression of a woman who is singing a lament.
We may expect a correspondingly wide range of possibilities for the self-expression of a man who is singing an elegy. For a man, in fact, the range of emotional self-expression is even wider than it is for a woman, since the singer of elegy at a symposium, even if he is lamenting the dead, must take note of the immediate occasion for his expression of lament, which is a sympotic occasion. Such is the dramatized occasion of Poem 13 of Archilochus, whose lifespan is traditionally dated to the first half of the seventh century:
          κήδεα μὲν στονόεντα Περίκλεες οὔτέ τις ἀστῶν
μεμφόμενος θαλίηις τέρψεται οὐδὲ πόλις·
τοίους γὰρ κατὰ κῦμα πολυφλοίσβοιο θαλάσσης
ἔκλυσεν, οἰδαλέους δ’ ἀμφ’ ὀδύνηις ἔχομεν
5        πνεύμονας. ἀλλὰ θεοὶ γὰρ ἀνηκέστοισι κακοῖσιν
ὦ φίλ’ ἐπὶ κρατερὴν τλημοσύνην ἔθεσαν
φάρμακον. ἄλλοτε ἄλλος ἔχει τόδε· νῦν μὲν ἐς ἡμ<έα>ς
ἐτράπεθ’, αἱματόεν δ’ ἕλκος ἀναστένομεν,
ἐξαῦτις δ’ ἑτέρους ἐπαμείψεται. ἀλλὰ τάχιστα
10      τλῆτε, γυναικεῖον πένθος ἀπωσάμενοι.

To care about mourning, Pericles, is not something that any one of the citizens
would find fault with, on the occasion of festivities [thaliai = symposia].
Nor would any city find fault.
You see, that’s the kind of men the waves of the loudly roaring sea
have swept under, and now we have our lungs all swollen with painful sorrows,
5        yes, our lungs. But you see what the gods have done for our incurable misfortunes,
my friend. They have placed as a cover over them a strongly resistant endurance
as an antidote. Different people have this thing happen to them at different times.
This time, {37|38}
it was our turn. And we mourn the bleeding wound inflicted on us.
Next time, it will happen to those who are next in line. So, come on, it’s time to get going,
as fast as you can.
10      It’s time for you to get over it and endure, pushing aside the kind of grief [penthos]
that women have.

The sympotic singer here is lamenting the death of beloved companions who drowned at sea. Just as the lungs of the drowned men are swollen with the salt water of the sea that has drowned them, so also the lungs of the men who lament them are swollen with the salt water of the tears they shed for their dear companions. This reference here to men’s lamenting for the dead, however, is contextualized in the setting of a symposium, where a man is singing a monodic song of elegy. And this sympotic setting of elegy makes a big difference. At a symposium, men must be men in contemplating death in particular and mortality in general. Their civic identity is foregrounded: it is essential for the symposiasts to be conscious of what the citizens think (line 1), what the city thinks (line 2). Men must sing elegy not the way women sing and dance lament. That is the thinking behind the ostentatious rejection, in this sympotic context, of a typical woman’s way of expressing her penthos (line 10). As we see most clearly in epic, this word penthos along with its synonym akhos means “grief,” and both words refer to lament as performed either by women or by men (Nagy 1999:94–117).

The point of the sympotic singer’s ostentatious rejection of women’s lament is this: whereas men can express their grief not only by way of singing lament on occasions like funerals but also by way of singing elegy on the occasion of a symposium, women cannot have such a sympotic occasion to sing elegy, restricted as they are to expressing their corresponding grief by way of singing lament. The lament of men in their sympotic singing of elegy may be a stylized and representational form of lament. Still, by differentiating itself from women’s lament, elegy can still claim to be a form of men’s lament.
Elegy as a form of men’s lament can extend from the narrower context of the symposium into the wider context of a public festival. A shining example is a lengthy monodic elegy composed by Simonides, whose lifespan is dated to 556–468 BCE. The elegy is Poem 22 of Simonides, preserved only in fragmentary form (ed. West 1998). The monodic singing of this elegy was evidently a public act of festive lamentation for the citizen warriors who died in the battle of Plataea in 479 BCE (Boedeker and Sider 2001; Nagy 2005). Though the actual venue for this elegiac performance has yet to be ascertained, one possibility I highlight here is that Simonides entered his Plataea Elegy for performance in the aulodic competitions at the festival of the Panathenaia in Athens.
The public festival is not the only other context for elegy as a form of men’s lament. Yet another context is the epigram, which is a poetic text inscribed on a monument meant to be seen by the public at large. A telling example is an epigram dated to the mid-sixth century BCE and deriving from the region of Ambracia (Supplementum Epigraphicum Graecum XLI 540). This epigram, inscribed on a monument enshrining the remains of a group of men who had been ambushed and killed while performing a diplomatic service for their community, is an elegy that laments the men while telling their mournful story within the poetic space of five elegiac couplets (Cassio 1994 has produced an improved edition of the text, with commentary). The {38|39} speaking voice of the elegy starts off by declaring monodically in the first person: andras tousd’ … olophuromai, “I lament these men” (line 1), while addressing the community in the second person as politai or “citizens” (line 9). The death of these men, as the elegy declares, was a penthos or “grief” for the community of fellow citizens, and this civic grief ethalle, that is, “it reached full bloom” at the time of their death (line 6). The verb thallein, “reach full bloom,” is a metaphorical reference here to the festive atmosphere of a symposium, as we see from contexts where the derivative noun thalia or “festivity” refers to a generic symposium (as at line 2 of Poem 13 of Archilochus, quoted above; also in Homeric Hymn to Hermes 56, 454). So the epigram pretends to maintain and even perpetuate that sympotic atmosphere in its elegiac words, permanently inscribed in stone.
Simonides, who was a celebrated composer of such elegy as performed at public festivals, became an equally celebrated composer of epigrams featuring such elegiac poetry. The composition of such elegies inscribed as epigrams could be commissioned by a community just as readily as the composition of elegies performed at their public festivals. In the case of Simonides, there survives a sizeable number of such elegies inscribed on public monuments and attributed to this master poet (Petrović 2007). The actual tradition of composing such elegies for epigrams became widespread already in the sixth century BCE (Aloni 2009). By the time of Simonides, whose era extends well into the fifth century, the epigrammatic uses of elegy became so versatile that they extended far beyond the originally sympotic contexts of this genre (Elmer 2005:10 on an elegiac epigram in Carmina Epigraphica Graeca, no. 429, dated around 475 BCE; p. 15 on an elegiac epigram attributed by Pausanias 10.27.4 to Simonides, whose composition was inscribed on the monumental painting of Polygnotus adorning the Leskhe of the Cnidians at Delphi). This epigrammatic function of elegy is vigorously continued in the Hellenistic period (Fantuzzi 2004).
Despite its claim to be a form of men’s lament, elegy can simultaneously make a counter-claim: that it is also an antidote for the grief and sufferings of lament. We find such a counter-claim in Poem 13 of Archilochus, quoted earlier, where the sympotic setting of elegy is seen as a pharmakon or “antidote” for the grief of mourning (line 7; Martin 2001:67–68, also with reference to Theognis 1133–1134; Lentini 2006:225n24). In such a convivial sympotic setting, as I noted earlier, men drink wine together as they recline on sympotic couches and enjoy each other’s company while conversing or taking turns in singing monodic songs. In such a setting, as in the setting of women’s laments, there is a wide range of possibilities for expressing emotions—not only the primary emotion of sorrow but also love, especially love, as also fear, anger, hatred, and so forth. And all these emotions can be tied to lived experiences—or even to imagined ones. These experiences, especially when it comes to affairs of the heart, can reflect negative as well as positive values.
It can be said in general about sympotic singing that the singer is free to represent negative as well as positive forms of behavior. In any given sympotic song, the “I” of the singer can be represented as roguish or decorous, antisocial or social, uncivil or civil, so long as the overall mentality of representation privileges the positive side of such oppositions (Nagy 1996a:216–223). In terms of this overall positive mentality, {39|40} men who come together at a symposium should be decorous, social, and civil, since the symposium as it represents itself in sympotic poetry is a community of companions who have social ties with one another, and these ties are communal, even civic (Nagy 2004:44–48).
Such a civic self-representation of the symposium by sympotic poetry is a key to explaining how elegy, as a primary medium of sympotic poetry, became also the medium of self-expression for the speaking voice of a civic model like Solon of Athens, whose lifespan is dated to 630–560 BCE, who was figured as the lawmaker of the Athenians (Nagy 1985:36–41; on the civic function of Solonian elegy in general, I cite the foundational work of Irwin 2005). In the elegiac poetry attributed to Solon, which teaches by example how to achieve order in society, the opposite forces of social order and disorder are made parallel to the opposite forces of orderly and disorderly behavior at a symposium (Poem 4.10; commentary by Nagy 1985:59–60; 1990a:430). Like Solon, another such civic model is Theognis of Megara, already mentioned, whose speaking voice claims for him the authority of a lawmaker (lines 543–546, 805-810; Nagy 1985:36–38). Still another such civic model is Tyrtaeus of Sparta, whose voice of authority articulates the Spartan social order (Nagy 1990b:269–275).
The civic function of elegy led to a wide variety of applications in the context of public festivals celebrated in cities large and small; at such festive occasions in any given city, elegiac poetry served the purpose of glorifying the cultural heritage of that city, and such poetry tended to be prodigiously detailed and lengthy (Bowie 1986). We have reports from the ancient world about such monumental elegies celebrating various city-states: examples include a lengthy elegiac poem about the city of Smyrna, attributed to Mimnermus of Smyrna and Colophon, dated to the seventh century BCE, and another lengthy elegiac poem about the city of Colophon and its daughter city Elea, attributed to Xenophanes of Colophon, dated to the mid sixth and early fifth centuries BCE (Bowie 1986, Irwin 2009). Such monumental poetry, which is sometimes called “catalogue elegy,” persisted into the Hellenistic period, as exemplified by the Aetia of Callimachus (Faraone 2008:139, Cameron 1995:277–289).
The civic seriousness of such elegy as sung at public festivals must be contrasted with the mixture of seriousness and lightheartedness we find in elegy as sung at symposia. In the good company of fellow symposiasts, the spirit of conviviality can induce even the voice of elegy to abandon from time to time its seriousness in tone and to forget about the cares and worries of civic existence. For example, affairs of the heart can at times overrule the affairs of state in the poetic agenda of elegy (examples abound: one among many is a poem embedded in Theognis 1063–1068). Comparable is the ostentatious abandonment of civic virtue in the atmosphere of hedonistic abandon that pervades Roman love elegy, as analyzed by Paul Allen Miller in The Oxford Handbook of the Elegy.
Despite so many such moments of celebrating the pleasures of life in ancient Greek elegy, however, death and mortality are never far away. Sorrow over mortality presses against the joy of being alive, and that is why life’s transient pleasures must be seized and enjoyed all the more. That is the pervasive sentiment of elegy. And I close by showing a most exquisite declaration of this elegiac sentiment in Poem 1 of Mimnermus: {40|41}

          τίς δὲ βίος τί δὲ τερπνὸν ἄτερ χρυσῆς Ἀφροδίτης;
τεθναίην ὅτε μοι μηκέτι ταῦτα μέλοι,
κρυπταδίη φιλότης καὶ μείλιχα δῶρα καὶ εὐνή,
οἷ’ ἥβης ἄνθεα γίνεται ἁρπαλέα
5        ἀνδράσιν ἠδὲ γυναιξίν· ἐπεὶ δ’ ὀδυνηρὸν ἐπέλθηι
γῆρας, ὅ τ’ αἰσχρὸν ὁμῶς καὶ καλὸν ἄνδρα τιθεῖ,
αἰεί μιν φρένας ἀμφὶ κακαὶ τείρουσι μέριμναι,
οὐδ’ αὐγὰς προσορῶν τέρπεται ἠελίου,
ἀλλ’ ἐχθρὸς μὲν παισίν ἀτίμαστος δὲ γυναιξίν·
10      οὕτως ἀργαλέον γῆρας ἔθηκε θεός.

What is life, what is pleasurable, without golden Aphrodite?
I want to be dead the moment I reach the time these things no longer matter to me –
I mean, secret lovemaking, sweet love gifts, bed of love.
Oh, how the blossoms of youth are ready for picking
5        for men and women alike! But then, once the time comes for the painful arrival
of old age, which turns even a good-looking man into something repulsive,
that is when a man’s thinking, over and over again, gets worn down by bad anxieties.
No, he gets no pleasure even from looking at the rays of the sun, no.
So there he is, someone boys will have nothing to do with, and gone is any respect
from women.
10      This is how painful it is, old age is, and it was made to be this way by the god.

In this sad and mournful song lamenting old age and its sure signs of a death yet to come, we can hear the sounds of real lament as really performed in lament. The formal cries of sorrow in real performances of lament, most commonly oimoi and aiai, are echoed by the arrangements of sounds produced in this song of elegy. In the second verse, the lamenting cry of oimoi is echoed in the rhyming vowels positioned at the end of the first half of the pentameter, …(m)oi, and at the end of the second half, …oi. In the seventh verse, the lamenting cry of aiai is echoed in the rhyming words positioned at the beginning of the first half of the hexameter, ai–, and at the end of the second half, –ai. There is further echoing at the end of the word kakai or “bad,” which ends at the main word break of this hexameter and which describes the merimnai or “anxieties” of the mournful man. And the sound of this word for “anxieties,” merimnai, refracts the sound of the name of this mournful man of elegy, pronounced in the original Greek as Mimnermos. (On the poetics of representing an echo as a refraction of light: Nagy 1996a:24).

There are even further echoes of lament. In the eighth verse, where the mournful man of elegy laments that he gets no pleasure even from looking at the rays of the rising or setting sun, I have put into my translation an added “no” of denial to convey the sound of the final syllable –ou of this verse, which is the ending of the word ēeliou meaning “of the sun.” The added “no” conveyed by the sound of this final –ou is a plaintive echo of the sound of the ou– that began this verse, where the ou– of oude really means the “no” of denying the pleasure to be had from looking at the rays of the sun. The whole song begins on a note of denying pleasure, since the word terpnon in the first verse, meaning “pleasurable,” is acoustically denied by the word that follows it, ater, meaning “without.” The denial is made by the sound of ater, not only by its meaning, “without,” since the sound ter of ater echoes the sound ter of terpnon, {41|42} meaning “pleasurable.” What is “not pleasurable” would be aterpnon, which is how the sound of the word ater begins, to be followed by the wording for “golden Aphrodite,” which completes the verse and completes the idea. And the idea is this: to deny pleasure is to deny golden Aphrodite, the goddess of sensual pleasure and love. More than that: to deny Aphrodite is to deny life.
The refusal to deny life is to lament, lament sadly and mournfully but sensually as well, the way that women lament as well, sensually. There is a pleasure to be had in the sensuality of lament. We have already heard Andromache saying that, as she started singing her elegiac song in the Andromache of Euripides (87–90):

          But I, involved as I am all the time in laments and wailings and outbursts of tears,
will make them reach far away, as far as the aether. For it is natural
for women, when misfortunes attend them, to take pleasure [terpsis]
90      in giving voice to it all, voicing it again and again, maintaining the voice from one mouth
to the next, from one tongue to the next.

That sensuality of lament is passed on, from one lamenting woman to the next. Men hear their song of lament, and they too pass it on, singing elegy. That sensuality gives pleasure, and such pleasure is an elegiac pleasure, derived from the pleasure that women take in passing on their own sorrows. Such are the delights of elegy.

Works Cited

Alexiou, M. 2002. The Ritual Lament in Greek Tradition, ed. 2. revised by D. Yatromanolakis and P. Roilos. Lanham MD.
Allan, W. 2000. The Andromache and Euripidean Tragedy. Oxford.
Allen, W. S. 1973. Accent and Rhythm. Prosodic Features of Latin and Greek: A Study in Theory and Reconstruction. Cambridge.
Allen, W. S. 1987. Vox Graeca: The Pronunciation of Classical Greek, ed. 3. Cambridge.
Aloni, A. 2009. “Elegy: Forms, Functions and Ways of Communicating.” In Budelmann (2009) 161–188.
Bailey, C., ed. 1936. Greek Poetry and Life: Essays Presented to Gilbert Murray on his Seventieth Birthday. Oxford.
Barnes, H. R. 1995. “The Structure of the Elegiac Hexameter: A Comparison of the Structure of Elegiac and Stichic Hexameter Verse,” in Fantuzzi and Pretagostini (1995) I 135–162.
Bartol, K. 1987. Greek Elegy and Iambus: Studies in Literary Sources. Poznan.
Beissinger, M., Tylus, J., and Wofford, S., eds. 1999. Epic Traditions in the Contemporary World: The Poetics of Community. Berkeley.
Bierl, A., Lämmle, R., and Wesselmann, K., eds. 1997. Literatur und Religion I. Wege zu einer mythisch–rituellen Poetik bei den Griechen (Basiliensia – MythosEikonPoiesis, vol. 1.1). Berlin / New York.
Blok, J. H., and Lardinois, A., eds. 2006. Solon of Athens: New Historical and Philological Approaches. Leiden.
Boedeker, D., and Sider, D., eds. 2001. The New Simonides: Contexts of Praise and Desire. Oxford.
Bowie, E. L. 1986. “Early Greek Elegy, Symposium and Public Festival,” Journal of Hellenic Studies 106:13–35.
Budelmann, F., ed. 2009. Cambridge Companion to Greek Lyric. Cambridge.
Bulloch, A. 1985. Callimachus: The Fifth Hymn. Cambridge.
Calame, C. 2001. Choruses of Young Women in Ancient Greece: Their Morphology, Religious Role, and Social Function (tr. D. Collins and J. Orion). ed. 2 of Calame 1977 vol. I. Lanham MD. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Calame.Choruses_of_Young_Women_in_Ancient_Greece.2001
Cameron, A. 1995. Callimachus and His Critics. Princeton.
Cassio, A. C. 1994. “I distici del polyandrion di Ambracia e l’ ‘Io Anonimo’ nell’ epigramma Greco.” SMEA 33:101–6.
Chantraine, P. 2009. Dictionnaire étymologique de la langue grecque: histoire des mots (ed. J. Taillardat, O. Masson, and J.-L. Perpillou), with a supplement “Chroniques d’étymologie grecque” 1–10 (ed. A. Blanc, Ch. de Lamberterie, and J.-L. Perpillou). Paris.
Dué, C. 2002. Homeric Variations on a Lament by Briseis. Lanham MD. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Due.Homeric_Variations_on_a_Lament_by_Briseis.2002
Dué, C. 2006. The Captive Woman’s Lament in Greek Tragedy. Austin, TX. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Due.The_Captive_Womans_Lament_in_Greek_Tragedy.2006
Ebbott, M. 2000. “The List of the War Dead in Aeschylus’ Persians,” Harvard Studies in Classical Philology 100:83–96.
Elmer, D. F. 2005. “Helen Epigrammatopoios,” Classical Antiquity 24.1:1–39.
Fantuzzi, M. 2004. “The Epigram.” In Fantuzzi and Hunter (2004) 283–349.
Fantuzzi, M., and Hunter, R. 2004. Tradition and Innovation in Hellenistic Poetry. Cambridge.
Fantuzzi, M., and Pretagostini, R., eds. 1995 / 1996. Struttura e storia dell’ esametro greco. I / II Rome.
Faraone, C. A. 2008. The Stanzaic Architecture of Early Greek Elegy. Oxford.
Figueira, T. and Nagy, G., eds. 1985. Theognis of Megara: Poetry and the Polis. Baltimore.
Foley, J. M., ed. 2005. A Companion to Epic. Oxford.
Gagarin, M., ed. 2009. The Oxford Encyclopedia of Ancient Greece and Rome. Oxford.
Gentili, B., and Prato, C., eds. 1988. Poetarum Elegiacorum Testimonia et Fragmenta. Leipzig.
Greenberg, N. 1985. “Appendix: Language, Meter, and Sense in Theognis,” in Figueira and Nagy (1985) 245–260.
Householder, F. W., and Nagy, G. 1972. Greek: A Survey of Recent Work. The Hague.
Irwin, E. 2005. Solon and Early Greek Poetry: The Poetics of Exhortation. Cambridge.
Irwin, E. 2009. “Elegy,” in Gagarin (2009) forthcoming.
Jakobson, R. 1952. “Studies in Comparative Slavic Metrics.” Oxford Slavonic Papers 3:21–66, reprinted in Jakobson (1966) 414–463.
Jakobson, R. 1966. Selected Papers IV. The Hague.
Koller, H. 1972. “Epos,” Glotta 15:16–24.
Lardinois, A. P. M. H. 2006. “Have we Solon’s Verses?” in Blok and Lardinois (2006) 15–37.
Lardinois, A., and McClure, L., eds. 2001. Making Silence Speak: Women’s Voices in Greek Literature and Society. Princeton.
Lentini, G. 2006. “Un lamento da donna: il fr. 10 V. di Alceo (ΕΜΕ ΔΕΙΛΑΝ) alla luce dei frr. 6 e 130b V.,” Seminari Romani di Cultura Greca 9:219–242.
Loraux, N. 1998. Mothers in Mourning, with the essay “Of Amnesty and its Opposite.” Tr. C. Pache. Ithaca.
Martin, R. P. 1989. The Language of Heroes: Speech and Performance in the Iliad. Ithaca. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Martin.The_Language_of_Heroes.1989
Martin, R. P. 2001. “Just Like a Woman: Enigmas of the Lyric Voice,” in Lardinois and McClure (2001) 55–74.
Martin, R. P. 2005. “Epic as Genre,” in Foley (2005) 9–19.
Nagy, G. 1972. Introduction, Parts I and II, and Conclusions. Greek: A Survey of Recent Work (F. W. Householder and G. Nagy) 15–72. Janua Linguarum Series Practica 211. The Hague.
Nagy, G. 1974. Comparative Studies in Greek and Indic Meter. Harvard Studies in Comparative Literature 33. Cambridge MA.
Nagy, G. 1985. “Theognis and Megara: A Poet’s Vision of His City.” Theognis of Megara: Poetry and the Polis (ed. T. J. Figueira and G. Nagy) 22–81. Baltimore. http://nrs.harvard.edu/urn-3:hlnc.essay:Nagy.Theognis_and_Megara.1985
Nagy, G. 1994/1995. “Genre and Occasion.” ΜΗΤΙΣ: Revue d’Anthropologie du Monde Grec Ancien 9/10:11–25. In Nagy 2012 v1. http://nrs.harvard.edu/urn-3:hlnc.essay:Nagy.Genre_and_Occasion.1994
Nagy, G. 1996a. Poetry as Performance: Homer and Beyond. Cambridge. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Nagy.Poetry_as_Performance.1996
Nagy, G. 1996b. “Metrical Convergences and Divergences in Early Greek Poetry and Song.” In Fantuzzi and Pretagostini (1996) II 63–110.
Nagy, G. 1999a. The Best of the Achaeans: Concepts of the Hero in Archaic Greek Poetry, ed. 2 Baltimore. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_Nagy.Best_of_the_Achaeans.1999
Nagy, G. 1999b. “Epic as Genre.” In Beissinger, Tylus and Wofford (1999) 21–32.
Nagy, G. 2004. “Transmission of Archaic Greek Sympotic Songs: From Lesbos to Alexandria.” Critical Inquiry 31:26–48. In Nagy 2012 v2. http://nrs.harvard.edu/urn-3:hlnc.essay:Nagy.Transmission_of_Archaic_Greek_Sympotic_Songs.2004
Nagy, G. 2005. Review of Boedeker and Sider 2001, Classical Review 55:407–409.
Nagy, G. 2007. “Did Sappho and Alcaeus Ever Meet?” Literatur und Religion: Wege zu einer mythisch–rituellen Poetik bei den Griechen I (ed. A. Bierl, R. Lämmle, and K. Wesselmann) 211–269. MythosEikonPoiesis 1.1. Berlin and New York. In Nagy 2012 v2. http://nrs.harvard.edu/urn-3:hlnc.essay:Nagy.Did_Sappho_and_Alcaeus_Ever_Meet.2007
Obbink, D. 2001. “The Genre of Plataea. Generic Unity in the New Simonides.” In Boedeker and Sider (2001) 65–85.
Page, D. L. 1936. “The Elegiacs in Euripides’ Andromache.” In Bailey (1936) 206–230.
Petrović, A. 2007. Kommentar zu den Simonideischen Versinschriften. Leiden.
Sultan, N. 1999. Exile and the Poetics of Loss in Greek Tradition. Lanham MD.
Tsagalis, C. 2004. Epic Grief: Personal Lament in Homer’s Iliad. Berlin.
West, M. L. 1974. Studies in Greek Elegy and Iambus. Berlin.
West, M. L., ed. 1998. Iambi et Elegi Graeci I / II. Oxford.
Yatromanolakis, D. 2001. “To Sing or to Mourn? A Reappraisal of Simonides 22W2”, in Boedeker and Sider (2001) 208–225.
Yatromanolakis, D. 2007. Sappho in the Making: An Anthropology of Reception. Cambridge MA. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_YatromanolakisD.Sappho_in_the_Making.2008