Sophocles

In dialogue Sophocles’ metre is like Aeschylus’. In song he has a manner of his own.^[1] His career is well recorded, but there is a caution. Sophocles “became canonical too easily and blandly”.^[2]

Dates. The Cambridge Companion’s chronology puts his birth in 497/6 or 496/5 BC, and his first victory in 468 with Triptolemus. It puts his death in 406/5.^[3] The sources differ slightly on the death. Scodel dates it to 405, “only a few months” before Aristophanes’ Frogs.^[2] He first produced in 468.^[2] Cropp, citing a remark of the elder Pliny, calls 468 the year of his first, “or at least first prize-winning”, production at the Dionysia.^[4]

There was no record of his birth. Greeks did not register births and deaths, and the year is worked out from the Parian Marble, an inscription from Paros that lists dated events. One entry puts his first victory in 469/8 at the age of twenty-eight. Another puts his death in 406/5 at ninety-two. Both point to 497/6, and 496 has become the accepted figure.^[5] The victory of 468 is reasonably certain, because the archon of that year made Cimon and his fellow generals the judges. Whether it was really his first production is another question.^[6]

He was still alive at the Proagon of 406, the ceremony before the Dionysia, where he mourned Euripides. He was dead by the Lenaea of 405, when the Frogs was staged.^[7]

His career of over sixty years overlapped with those of Aeschylus and Euripides, and he outlived Euripides.^[8] His grandson staged the Oedipus at Colonus in 401, some five years after the poet’s death.^[9]

Output and victories. The ancient figures disagree. Aristophanes of Byzantium referred to 130 plays, seventeen of them spurious, and another source to 123.^[2] The Suda says 123, or more according to some.^[10] Cropp reports 113 or 123, and more than 120 known titles.^[4] The biographies credit him with twenty or twenty-four victories.^[2] Gregory counts eighteen or more in about thirty attempts. Not even Aeschylus matched this “exceptional” record.^[11]

The inscribed victory list and Diodorus give eighteen wins at the City Dionysia. The Suda’s twenty-four may include wins at the Lenaea. He was never placed third.^[12] His son Iophon was also a tragedian. He won at the Dionysia in 435 and came second in 428.^[13]

Public life. He was a treasurer of the Athenian alliance in 443/2, and he was elected general. His contemporary Ion of Chios tells an anecdote about him. In it he jokes that his skill in love disproved Pericles’ low opinion of his generalship. The biography makes him pious. He was the host of the god Asclepius when the cult came to Athens.^[2]

Only two facts about him are beyond doubt. He was a treasurer of the alliance’s funds in 443/2, and his tragedies were staged at the City Dionysia.^[14] He served as general with Pericles in the Samian War of 441/0. The usual date of the Antigone, 441, rests on an ancient link between the play and this generalship, and that link may be only a guess.^[15] At about eighty-three he was one of the ten probouloi, the special officials appointed in 412/1.^[16]

He was upper-class, as Ion attests. He repeatedly put his work before the archon who chose the festival’s poets. If he was not a professional poet, Pelliccia argues, nobody was.^[17] The Frogs leaves him out of its contest in the underworld. He is too “good-tempered” to want to leave.^[2]

Man of the theatre. Aristotle credits Sophocles with the third actor.^[18] The increase of the tragic chorus from twelve to fifteen is counted a Sophoclean innovation too.^[19] He may have written a prose treatise On the Chorus, now lost.^[20] The ancient Life (23) says that he brought the Phrygian mode into tragedy.^[21]

Hephaestion preserves one of his elegiac lines. Sophocles thought that the name Archelaus would fit neither hexameter nor elegiac verse. So he wrote Ἀρχέλεως (Archeleōs), “for it was metrical to say it so”.^[22]

Among the three. In the spoken trimeter “Aeschylus and Sophocles” are paired against the later Euripides. Both resolve sparingly, on average in no more than one line in ten.^[23] Resolution means that a long position is filled by two shorts.

In song he goes his own way. He avoids Aeschylus’ favourite closing cola. In lyric iambics he admits a long anceps a little more often than Aeschylus or Euripides.^[24] He writes “several” pure aeolic strophes.^[25] He also lengthens aeolic cola by choriambic expansion, a habit West counts as especially his.^[26]

Reputation. For “Longinus”, he and Pindar at times sweep all before them like a blaze.^[27] A speaker in Xenophon puts the New Musician Melanippides on a level with Homer and Sophocles in their fields.^[28] A story of bees and honey is told of the infant Pindar, and it has a parallel in the tradition about Sophocles.^[29] His plays and their dates are on the Plays tab, his songs on the Chorus and song tab.

Seven tragedies survive complete, out of more than a hundred plays.^[1] The text quoted on this site is Francis Storr’s Loeb edition (1912–1913) in the Perseus Digital Library. It adds an eighth text, the satyr-play Ichneutae from Oxyrhynchus Papyrus 1174 (ed. A. S. Hunt, 1912). Only two of the seven are securely dated. For the others the Companions give estimates, and they do not always agree.

The Oedipus Tyrannus is generally dated to the early 420s.^[2]

Not a trilogy. Translations usually print the three “Theban plays” together, as if they were a trilogy. But they were probably composed over almost forty years. When he wrote the Antigone, Sophocles can scarcely have known that an Oedipus the King was to follow.^[3] In no case can we say which of his plays were produced together.^[4]

The lost plays. From the lost tragedies we have only some eight hundred complete lines or pairs of adjacent half-lines. Just three dozen fragments are longer than four lines.^[4] His favourite subject, to judge from the fragments, was the Trojan War, not Thebes.^[5] Well over thirty tragedies drew on it.^[4]

Papyri have given extensive fragments of the satyr-play Trackers (Ichneutae). They have also given some of a Trojan play, Eurypylus. Its hundred or so intelligible lines are the only extensive tragic papyrus fragments of Sophocles.^[6] The Trackers tells of Apollo’s search for the cattle stolen by Hermes. About two thirds of its action belongs to the satyrs of the chorus.^[7] A papyrus scrap records a production of Sophocles in the same contest as Aeschylus’ Danaid tetralogy.^[8]

His Tereus was produced before 414.^[9] The fragments of Tereus and of his two Tyro plays show that filicidal mothers and seduced heroines were not peculiar to Euripides.^[10] His lost Thamyras was about the Thracian singer who dared to compete with the Muses. It is one of the special cases in which a stringed instrument seems to have appeared on the tragic stage.^[11]

Sources and inventions. Stesichorus’ Oresteia already had Orestes’ Nurse and Clytemnestra’s dream. Both are important “in Aeschylus and in Sophocles”.^[12] Sophocles may have invented Antigone’s burial of her brother. He probably also thought of having Electra herself deceived by Orestes’ false report of his death.^[13] Rutherford has raised a possibility about the Erythraean paean to Asclepius, inscribed in the early fourth century. It may be a fifth-century work of Sophocles.^[14]

In the metrical handbook. West quotes from all seven plays, but the Oedipus at Colonus only in a footnote on prosody.^[15] Satyric drama in general may be a little less strict about Porson’s Law, though only a little. West names no play for this.^[16] Dactylo-epitrite is used several times in the earlier plays of Sophocles and Euripides, and hardly at all after 415.^[17]

Sophocles’ text is followed here from the poet’s script to the Alexandrian edition; the late-antique selection, the Byzantine manuscripts and the modern editions are on the Manuscripts tab. The general story is told on the page How the texts reached us.

Scripts and family archives. A tragedy was written as a script for the actors and chorus who performed it under the poet’s direction. By the later fifth century, though, Sophocles and Euripides may have been writing for readers as well.^[1] Lyric was written out as prose. A colon is a short unit of verse. The division of lyric into cola is commonly ascribed to Aristophanes of Byzantium (c. 255–180 BC in Sampson, c. 257–180 in Reynolds and Wilson).^[2] After him, texts divided into cola become the norm.^[3]

Easterling thinks that in the early days the preservation of scripts depended a great deal on family archives. Dramatists’ descendants competed with plays left unperformed at their deaths. Sophocles’ grandson did so with the Oedipus at Colonus in 401.^[4]

Revivals and official copies. From 386 an “old” tragedy was revived each year at the City Dionysia. On the surviving records the provision was used mostly for Euripides. There is some evidence for revivals of Sophocles, and none for Aeschylus.^[5] Under the orator Lycurgus, about 330 BC, Athens established an official copy of the three tragedians. Actors were not to deviate from it.^[6] Whether earlier official copies existed is disputed.^[7]

Alexandria. The Alexandrian edition included every play the scholars could find.^[8] Aristophanes of Byzantium referred to 130 plays of Sophocles, seventeen of them spurious.^[9] So the question of authenticity was already being asked. The text of tragedy was probably established by reference to the Athenian official copy.^[10]

That copy was probably not a critical edition, and it may already have contained errors and spurious lines.^[11] Modern scholars regularly credit Aristophanes of Byzantium with editing the whole of Greek tragedy, but there is no direct evidence for it. The scholia to Sophocles name him only once.^[11]

The Alexandrian editors divided lyric into cola. Whether they still had the music is disputed. Fleming and Kopff argue that musical and metrical knowledge was handed down to them. Parker and Prauscello argue that it was lost.^[12]

Colometry is editorial. The line-division of the lyrics in a printed text is not Sophocles’. The metrical analysis is an interpretation too. Several tragic passages, Ajax 628–634 among them, can be read either as ionic or as aeolic, and West prints both analyses side by side.^[13] The medieval manuscripts agree fairly well on the colometry of easy metres. In difficult ones they vary so widely that they may go back to different ancient colometries.^[14]

Dialect in the copying. Copyists trained in Attic tended to replace forms from other dialects with Attic ones. So in the lyrics of tragedy, many original Doric forms have been eliminated by successive generations of scribes.^[15]

The limits of the evidence. In tragedy the same strophic pattern never recurs more than twice. So period-end shows itself only now and then: by hiatus, by brevis in longo, or by anceps next to anceps. The antistrophe repeats the strophe’s pattern, and this is called responsion. Textual corruption may destroy responsion. The editor may then be unable to say whether strophe or antistrophe preserves the true scheme.^[16] One example of imperfect aeolic responsion in Sophocles is Trach. 960 against 929, where a telesillean is answered by an anaclastic form. It stands as an exception to the rule of exact responsion.^[17]

In the trimeters, there are a handful of apparent breaches of Porson’s Law in tragedy, and most are suspect on textual grounds.^[18] At the join between lines the dramatists “show some disrespect” for period-end, and Sophocles takes some licences there. He places a postpositive first in a line (OT 1084 f.), and a prepositive last (Ant. 171 f.). He elides across the line-break, once even ταῦτ’ (OT 332).^[19]

Papyri. Ancient copies on papyrus bear witness to strands of the text that are lost everywhere else.^[20] For Sophocles they have restored much of the satyr-play Ichneutae (P.Oxy. 1174)^[21] and part of Eurypylus.^[22]

The Ichneutae papyrus shows how an ancient copy laid out a play. Its lines are indented at two levels, which mark changes of metre and changes of speaker. No papyrus of Sophocles carries musical notation.^[23]

The Transmission tab ends with the Alexandrian edition. The seven plays are followed here through late antiquity and Byzantium to the printed and modern editions; the general history is on How the texts reached us.

Seven plays. Until about AD 250 writers quote from many of the roughly three hundred plays of the three tragedians. After that, they quote only seven plays each of Aeschylus and Sophocles, the very seven we still have, and ten of Euripides.^[1] Wilamowitz explained this by a school syllabus fixed by one influential schoolmaster. Reynolds and Wilson find the theory “too schematic”. They would abandon the idea that a conscious selection by one person decided what survived. Kovacs leaves open whether the selection was deliberate, or a chance result of popularity and the availability of commentaries.^[2] The point is disputed.

The Laurentian manuscript. In the ninth century the plays were recopied in minuscule. Minuscule was the new lower-case script. The commentaries were cut down to fit the margins, and became the scholia.^[1] The most important single witness is Laur. 32.9 in Florence, of the tenth century. It is the only medieval manuscript of all seven plays of Aeschylus, and it is “of fundamental importance” for the text of Sophocles as well.^[3]

A palimpsest is a manuscript whose first writing was scraped off so that the parchment could be reused. A palimpsest in Leiden (B.P.G. 60 A) preserves parts of some Sophoclean plays. It is the “twin brother” of the Laurentian manuscript, apparently written by the same scribe.^[4] Both omit Oedipus Tyrannus 800. A stemma is a family tree of the manuscripts, and a shared error like this helps editors to draw one up.^[5] Editors call Laur. 32.9 “L”. L is so much more important than any other witness that even its minor errors are reported in the apparatus.^[6]

A line lost everywhere. All the manuscripts omit one line of Antigone 1165–1168. Eustathius, in the twelfth century, quotes the passage complete from “good copies”. But the complete text is also in Athenaeus. Close study of the variants makes it likely that Eustathius took it from him.^[7]

The triad. In the thirteenth century schools narrowed their reading to three plays of each tragedian. This is the Byzantine “triad”. For Sophocles the three were the Ajax, Electra and Oedipus Tyrannus.^[8] Most later manuscripts contain only these. The other plays survived just long enough to be taken to Italy by Renaissance collectors.^[9] Demetrius Triclinius, about 1305–1320, worked on the plays of the school curriculum with the help of Hephaestion’s handbook. He knew that strophe and antistrophe should respond.^[10]

About seven per cent of Sophocles’ work survives.^[11] After the Byzantines recovered Constantinople in 1261, many new copies were made: a few from the late thirteenth century, some two hundred from the fourteenth and fifteenth. Most contain three plays at most, and a good number only the Ajax and Electra. Triclinius edited the seven plays with scholia. His grasp of metre, and of strophic responsion in particular, let him restore Sophocles’ words in dozens of places. Modern scholarship on the text of Sophocles begins with him.^[12]

Print. The first printed Sophocles came from the press of Aldus Manutius in Venice. In 1502–1504 that press issued the first editions of Sophocles, Euripides, Herodotus, Thucydides and Demosthenes.^[13]

The Aldine and the other sixteenth-century editions rested on a poor source. L was known, but three hundred years passed before its value for Sophocles was seen. Triclinius’ text was the basis of the printed editions until the late eighteenth century. One of them, Willem Canter’s of 1579, was the first to set out the lyrics in responding units.^[14]

Modern editions. The Cambridge Companion lists these texts:^[15]

• R. D. Dawe’s Teubner (3rd edn 1996), with Dawe’s Studies on the Text of Sophocles (1973–1978);
• H. Lloyd-Jones and N. G. Wilson’s Oxford text (1990, corrected 1992), with their Sophoclea (1990);
• for the fragments, S. Radt, Tragicorum Graecorum Fragmenta vol. 4 (1977), and Lloyd-Jones’s Loeb (1996);
• Jebb’s commentary on all seven tragedies (1883–1900).

Metrical analyses of the whole of Sophocles are in Dawe’s Teubner. Analyses of all tragedy are in A. M. Dale’s Metrical Analyses of Tragic Choruses (1971–1983). West warns that those in Jebb “are antiquated”.^[16] The authenticity of the trochaic tetrameters that close the Oedipus Tyrannus (1515–1530) is debated.^[17] This site quotes Storr’s Loeb text (1912–1913), because the Perseus Digital Library makes it available under an open licence.^[18]

Drama used three modes of delivery: speech, recitative and song. Sophocles’ spoken trimeter and his recited metres are treated here; his lyric cola are on the next tab.

The spoken trimeter

The opening of Oedipus Tyrannus is West’s own first specimen of the tragic trimeter:^[1]

A caesura is a word-end at a regular place inside the line. Lines 1 and 2 have the penthemimeral caesura, which falls after the fifth position. Line 3 has the hephthemimeral caesura, which falls after the seventh. Line 2 ends with a short syllable in the final long position (-τε, -te). This is called brevis in longo.^[1]

In tragedy a stop + liquid or nasal inside a word may be treated either way. So τέκνα (tekna, “children”) can be scanned tek-na (— ⏑) or te-kna (⏑ ⏑).^[2] West’s syllable-markers show τέκ'να (tek’na) and ἕδ'ρας (hed’rās) long here. He marks the α (a) of νέα (neā) and of ἕδρας (hedrās) long.^[1]

Resolution. Resolution means that a long position is filled by two shorts. In Sophocles, as in Aeschylus and the early iambographers, it is rare. It is found on average in at most one line in ten.^[3] The first four princeps positions may be resolved, and in drama the fifth too. But in tragedy the fifth is resolved only when the anceps next to it is short.^[1]

An anapaestic foot ⏑ ⏑ — is admitted in tragedy outside the first foot only for proper names. In Euripides, and once in Sophocles (Phil. 795), its first syllable or two may be formed by a prepositive.^[4] A prepositive is a short word that leans on the word after it. Porson’s Law is observed: when the ninth position, the last anceps, is long, no word may end after it.^[1] In OT 1 above that anceps is the short νέ- (ne-); a worked case of the law is on the Euripides page (Metres: spoken).

Across the line-join. Sophocles twice puts a postpositive at the beginning of a line (OT 1084 f. οὐκ ἂν ἐξέλθοιμ’ ἔτι ‖ ποτ’ ἄλλος, ouk an exelthoim’ eti ‖ pot’ allos, “I could never turn out to be another”). He places a prepositive in final position (Ant. 171 f. παίσαντές τε καί ‖ πληγέντες, paisantes te kai ‖ plēgentes, “striking and struck”). Like the comedians, he elides over the line-break when both the last position of one line and the first of the next are long. He usually does so with δ’, τ’ or μ’ (d’, t’, m’), but once with ταῦτ’ (taut’, OT 332).^[5] In stylistic terms, no other Greek poet runs one line into the next so forcefully. This is the so-called Sophoclean enjambment.^[6]

The middle of the line. In a small percentage of tragic trimeters there is a caesura at the end of the third foot. It almost always comes with an elision at that point, but occasionally without.^[1] One example without elision shows Sophocles’ harsh parallelism. It is Antigone’s answer to Ismene:

The line has its regular penthemimeral caesura after εἵλου (heilou). But the near-rhyme of the infinitives ζῆν (zēn) and κατθανεῖν (katthanein) stresses a pause at mid-line, after the sixth position. That is the very place where a Greek iambic trimeter should not break.^[7]

Plutarch reports that Sophocles himself described three periods of his style. The first imitated Aeschylus’ grandeur. The second was a harsh and contrived manner. The last was the style “most expressive of character, and the best” (our translation of Plutarch). Probably all the surviving plays belong to the last.^[6]

Dialogue forms. In stichomythia the speakers have one line each. Sophocles shares it with the other tragedians. A single line can also be divided between speakers, and this is called antilabe. It is used to great effect in the exchange in which Oedipus forces the truth from the herdsman (OT 1173–1176).^[8]

Recited metres

Between speech and song there seems to have been a mode of delivery somewhat like recitative. Actors and chorus used it, especially to introduce a choral song. Its anapaests often mark entrances. The chorus of the Ajax enters with a preamble of thirty-eight anapaestic lines before its lyrics. It is not known whether such passages were delivered by the whole chorus or by its leader.^[9]

A scene of dialogue may modulate into anapaestic recitative. That is often a signal that the play is about to end, as at Ajax 1402 ff. and Philoctetes 1409 ff.^[10] One of the two tragic trochaic tetrameters without the normal caesura is Phil. 1402.^[11]

The lyric cola of Sophocles are described here, mostly in West’s terms. How much of each play is sung, and by whom, is on the Chorus and song tab.

Lyric iambics. Sophocles admits a long anceps more often than Aeschylus or Euripides. A clausula is the closing cadence of a period. He does not use Aeschylus’ aristophanean and pherecratean clausulae. He allows himself slightly more freedom in exact syllabic responsion, as at OT 194 against 207.^[1]

Aeolic. Sophocles has several pure aeolic strophes.^[2] Longer cola are sometimes created by dactylic or choriambic expansion. West counts the choriambic type as distinctively Sophoclean. He analyses Phil. 706 ff. as glᶜ | glᶜ (both with dragged close) | phᶜ | ∧ia∧ tl ia∧ | dod (dragged) ‖ gl²ᶜ | ia | tl | r |||.^[3]

Dale had set out the same song. She used it to show how glyconics grow into asclepiads by added choriambs, and how the cola overlap word-end. She reads 710 as meant to be heard as aeolic, a catalectic version of the two asclepiads before it.^[4] A colon has a “dragged” close when its end — ⏑ — becomes — — —. Sophocles and Euripides occasionally drag the end of a colon in this way.^[3] Exact responsion is the rule with anaclasis. But at Trach. 960 ~ 929 a telesillean (tl) answers its anaclastic form (tl¨).^[3]

Dochmiacs are found in every extant tragedy.^[5] They are the quintessential metre of dramatic lyric, unknown in lyric outside drama.^[6]

Ionics are “occasional” in Sophocles. OT 483 opens a period with a “hypermetric” long syllable before the ionic run. Ajax 628–634 admits either an ionic or an aeolic reading.^[7] Dale discusses the same lines. The repeated choriamb, with a long anceps throughout, lets the opening be heard as a contracted ionic metron. She concludes that Sophocles developed a special “aeolo-ionic”, built only of longs and double shorts. It is a problem almost entirely his.^[8]

Dactyls. Sophocles and Euripides sometimes write longer dactylic periods than Aeschylus. But they mark them off into tetrameters by word-division, as in El. 129 ff.^[9]

Two habits are his own. First, he avoids the Aeschylean practice of emphasising contractions by word-end.^[9] Second, he follows an acatalectic tetrameter with a colon that does not begin on a princeps, and so forces a change of step (El. 134–135, OT 171 f., El. 125 f.). Normally successive princeps positions are never separated by more than two shorts or one long. West counts this device as typical of Sophocles, though not confined to him.^[9] Dactylo-epitrite appears in his earlier plays, and scarcely after 415.^[10]

Polymetry. The long astrophic monodies and lyric dialogues of the last plays of Sophocles and Euripides show the greatest variety of metre within a single composition.^[11] Phil. 1169–1217 begins in iambics, and turns to ionics in mid-sentence at 1175. Then come anapaestic, aeolic and dactylic cola, and dactylic tetrameters from 1196 to 1207. The close runs through iambic, aeolic and dactylo-epitrite cola (E², e ⏑ D —) to a final penthemimer. So every lyric type is here except cretics and dochmiacs.^[12] West supposes that its singers also had to manage changes of mode, as the “new music” did.^[13]

Responsion at a distance. A strophe and its antistrophe need not stand together. In the Philoctetes 391–402 is answered by 507–518, after a long scene in trimeters.^[14]

Language. The lyric metres of tragedy follow the tradition of archaic choral song, the tradition of Pindar and Bacchylides. Their language has a Doric patina.^[15] Whether a line was sung is often decided by the metre, and by key linguistic features such as Doric alpha. The manuscripts tend to confuse these.^[16]

The Metres tabs describe Sophocles’ cola. How song is placed and used in his plays is described here.

Less choral song, more regular odes. Choral lyric takes up on average 17% of the lines of a Sophoclean play. In Aeschylus it takes 34–55%, and in Euripides 15%.^[1] In the Electra it is 11%.^[2] Counting all sung lines, the Oedipus Tyrannus has the fewest of his plays, just under a fifth. His plays have fewer sung lines than those of Aeschylus and Euripides.^[3] Aristotle says that Sophocles treats the chorus as “one of the actors”, and that Euripides does so less successfully (Poetics 1456a25–27).^[4]

A stasimon is a choral song sung between episodes. Sophocles’ stasima are shorter than Aeschylus’, most commonly two strophic pairs. They are also more standardised in plan. They tend to begin with something removed from the play’s world, a maxim or a myth, and make the connection with the action only later. Each strophe is answered once, by its antistrophe. The next pair takes a new metrical form, often reusing the same rhythms in another arrangement. This lets the poet change mood within an ode.^[5] In the second stasimon of the Oedipus Tyrannus (863–910) the relevance to Oedipus appears only in the last lines, on the fading oracles given to Laius.^[6]

Who sings. Where their identity is certain, Sophocles has twenty-four male and fifteen female choruses. That is the reverse of the preference in Aeschylus and Euripides. In the Ajax and Philoctetes the chorus is of adult citizen soldiers.^[7]

In Aeschylus and Euripides, song mostly goes to characters who are not adult Greek men. Sophocles is the important exception. He gives his heroes extended lyric at moments of extreme pain: Heracles consumed by the poisoned robe (Trach. 983–1043), and Ajax on realising what he has done (Ajax 348–427).^[8] Cicero later criticised the portrayal of the suffering Heracles, some of it in lyric metres.^[9] Song can also mark joy, as in Electra’s recognition of her brother (El. 1232–1287). And actor and chorus need not both sing. At the end of the Oedipus Tyrannus the chorus-leader’s spoken iambics answer Oedipus’ lament (1313–1366).^[10]

Forms. The parodos is the chorus’s entrance-song. In a dialogic parodos it becomes a lyric exchange between the chorus and a character already on stage. This appears only in Sophocles’ late plays, the Electra, Philoctetes and Oedipus at Colonus. In the Oedipus at Colonus the chorus’s promise of safety occupies the symmetrical strophic section. Its horror at Oedipus’ name fills the astrophic epode (208–253).^[11]

A kommos is a lyric exchange between actors and chorus. The song over the sleeping hero in the Philoctetes (828–864) is a kommos, in which Neoptolemus chants four hexameters between strophe and antistrophe. The chorus begins as if singing a paean, a prayer to Sleep for healing. Then it urges Neoptolemus to steal the bow. So the paean, a song of healing, is turned towards killing.^[12]

Genres of song. The entrance-song of the old men in the Antigone (100–161) has marked paeanic elements.^[13] In the Women of Trachis the female chorus mixes the male paean (210) and the dithyramb (219) in one song (205–224). Plato condemned such mixing of lyric genres, and it is usually associated with the New Music. But it is found in tragedy from Aeschylus on.^[14] The women anticipate a wedding-song for Deianira and Heracles. The audience knows that it would better suit his new marriage with Iole.^[13]

This song has no antistrophe. Its iambic metres change quickly as it moves from wedding-song to paean to dithyramb, and it probably changed mode as well. The aulos-player had a showpiece here. The women even sing that they will not reject the aulos, “tyrant of my mind” (216–217).^[15]

Modes and instruments. Early tragedy used mainly the Dorian and Mixolydian modes. Aristoxenus says that Sophocles was the first Athenian poet to bring the Phrygian mode into his songs and to mix in the manner of dithyramb.^[16] He is the only tragedian known to have played the kithara on stage, in the title role of his Thamyras.^[17]

Endings. Some of Bacchylides’ myths end abruptly. Pelliccia contrasts them with “the appalling disquiet” caused by two of Sophocles’ endings. The Electra ends without mention of the Furies, and the Women of Trachis without a hint of Heracles’ apotheosis.^[18]

The new music. Actors’ song grew, and the chorus’s share shrank. This reflected the musical fashion of the later fifth century, which gave an outlet for virtuoso solo performance.^[19] West’s showpiece of late polymetry, Philoctetes 1169–1217, is analysed on the Metres: sung tab.