¹ The following are the articles on the end of the Septem to which I shall be referring (normally by author' name only): Wilamowitz-Moellendorff, U. von,‘Drei Schlussscenen griechischer Dramen’, Sitz. der Berl. Akad. 1903, 436Google Scholar ff.; Wundt, M., ‘Die Schlussscene der Sieben gegen Theben’, Philol. 65 (1906), 357Google Scholar ff.; Platt, A., ‘The Last Scene of the Seven Against Thebes’, CR 26 (1912), 141Google Scholar ff.; Kohl, R., ‘Zum Schluss von Aischylos Sieben gegen Theben’ Philol. 76 (1920), 208Google Scholar ff.; S.C. Manginas, (1953), 305 ff.; Pöitscher, W., ‘Zum Schluss der Sieben gegen Theben’, Eranos 56 (1958), 140Google Scholar ff.; Lloyd-Jones, H., ‘The End of the Seven against Thebes’, CQ N.S. 9 (1959), 80CrossRefGoogle Scholar ff.; Fraenkel, E., ‘Zum Schluss der Sieben gegen Theben’, MH 21 (1964), 58Google Scholar ff.; Dawe, R.D., ‘The End of the Seven against Thebes’, CQ N.S. 17 (1967), 16CrossRefGoogle Scholar ff.; Petersmann, H., ‘Zum Schluss von Aischylos Sieben gegen Theben’, Ziva Antika 22 (1972), 25 ff. For fuller bibliography see Lloyd-Jones and Nicolaus.Google Scholar

² Page, in the apparatus of his 1972 O.C.T., implies that 875–960 might also be spurious. But the style of these lines seems wholly Aeschylean, and on Page' own showing, since they contain nothing ‘quod sororibus unice conveniat’, they can hardly be by the same interpolator as 861–74.

³ The fact of which Manginas and Ptitscher make so much, that some expressions here do have close parallels in genuine Aeschylus, proves nothing.

⁴ The text here can be emended to give some other meaning, but this does not affect the real problem.

⁵ I take it, however, despite Wilamowitz, (app. crit. of his Aeschyli Tragoediae (Berlin, 1914)) and Taplin, that the interpolator did intend to give the ode, or parts of it, to the sisters. The reason he introduced them as early as 861 will have been so that they could lend variety and subjective pathos to the lyrics at a time when a purely choral section of such length would have been intolerable to an audience.Google Scholar

⁶ This particular argument would be met by the compromise solution of Kohl, who accepts Antigone but rejects Ismene, or that of Robert, C., Oidipus (Berlin, 1915), i. 375, and Pötscher, who accept the sisters in the lyric section but reject the final scene. For reasons which will be apparent neither of these compromises seems to me acceptable.Google Scholar

⁷ The children' songs at Eur. Alc. 393 ff., Andr. 504 ff., presumably performed by actual boy singers, are clearly not comparable. Some have supposed that Ismene' few words at Soph. O.C. 1724–1736 were sung by a supernumerary, but even if this were true the procedure would remain highly im probable for Aeschylus sixty years earlier (see Dawe, p. 22).

⁸ Mention of the sisters at 974 is no doubt the result of mere scribal corruption, as Wilamowitz says.

⁹ Several editors (and Wilamowitz in ‘Drei Schlussscenen’) follow Haupt in keeping 995 undivided, deleting 996 and retaining 997. This solution does give the required symmetry, but it does not suffice to make the lines suitable for the semi-choruses. On the other hand there is no need to delete 995 as well as 996 f., with Srebrny, S., Critica et Exegetica in Aeschylum (Toruń, 1950), 34 ff.Google Scholar

¹⁰ At 1068 has naturally been taken to mean ‘help Antigone to bury’. But it need not. The semi-chorus has just referred in the masculine to the implication being that other citizens besides the chorus are having to make a decision in the matter and that some may defy the edict; the context thus makes it natural in any case to talk of ‘taking part in the burial’.

¹¹ The different words for lamentation are discussed by Alexiou, M., The Ritual Lament in the Greek Tradition (Cambridge, 1974), pp. 102 f., 225 f. None of the distinctions of meaning that she draws can be relevant in the present context.Google Scholar

¹² A plausible guess would be that was followed by a term meaning ‘unburied’ in a paraphrase of the Homeric

¹³ The point has been made before by Wilamowitz, , Aischylos: Interpretationen (Berlin, 1914), pp. 89 f., and others, but the natural conclusion - that Antigone's speech and the anapaests are by different authors - has not been drawn.Google Scholar

¹⁴ Once again the inconsistency has been noted (see e.g. Dawe, p. 21) and the point has then been left hanging.

¹⁵ I am here assuming, with all recent editors, that 1054–1065 belong to the whole chorus. Attempts to give anything before 1066 to semi-choruses run into difficulties, since 1062–1065, which favour Polynices, will presumably have to be delivered by the same semi-chorus as 1066–1071. Lloyd-Jones's attempt (pp. 111 f.) to distinguish between whole semi-choruses and their leaders is highly conjectural.

¹⁶ Sometimes in tragedy a choral ode commenting on the action will take its point of departure from somewhere earlier than the end of the preceding episode. It would be more surprising if an anapaestic system which itself contributes to the action could do the same.

¹⁷ I do not suppose there will be much support for Petersmann's view that the interpolator's references to the city in general, in place of Creon, can be explained by anti-Theban sentiment after the Battle of Delium in 424. For one thing one would hardly expect anti-Theban sentiment at Athens to be directed against the .

¹⁸ Of course if the interpolated scene replaced an extended ending by Aeschylus, as some have thought (Wilamowitz, , ‘Drei Schlussscenen’, D.L. Page, Actor' Interpolations in Greek Tragedy (Oxford, 1934), p. 32), then the latter could also have taken account of the burials. But then it could have looked remarkably like the ending which my theory postulates.Google Scholar

¹⁹ I am aware that arguments of this kind are subjective and can be made to cut both ways; it is no doubt possible to give al account of the scheme and purpose of the Septem that will leave no place for the fina scene (see e.g. Nicolaus, pp. 85 ff.). I woulc maintain, however, that my own account is at least equally good.

²⁰ Schadewaldt's theory would in fact meet some of my arguments; but the anapaestic passage seems to me perfectly coherent and internally consistent (apart from 1064), and I can therefore see little methodological justification for regarding it as a patchwork by more than one author.

²¹ See Garvie, A.F., Aeschylus' Supplices: Play and Trilogy (Cambridge, 1969), pp. 194 ff.Google Scholar

²² Another possibility, which would also favour my case, is that a subsidiary chorus sings 1034–1051 but the main chorus divides at 1052.

²³ The parallel with the Prometheus is noted by Snell, B., Aischylos and das Handel, im Drama, Philol. Suppl. 20.1 (1928), 94. I am assuming, more for the sake of argument than from conviction, that this play is by Aeschylus; if it is not, my argument must be adjusted accordingly.Google Scholar

²⁴ Of the various discussions of the nautical imagery in this play the best is that of Nes, D. van, Die maritime Bildersprache bei Aischylos (Groningen, 1963), pp. 75 ff. and passim.Google Scholar

²⁵ e.g. Müller, G., Sophokles: Antigone (Heidelberg, 1967), pp. 21 ff.Google Scholar

²⁶ These prophecies would indeed be incompatible with casting Polynices out unburied; but that is not what happens. Presumably we are not to exclude from Greek tragedy every development that threatens to prevent the fulfilment of a prophecy?

²⁷ Page is probably right to imply that in 912 is the best that can be done; and I sympathise with his inability to swallow the frigid and misplaced rhetoric of (as 913 is usually punctuated), though I am not sure what alternative there is. But 914 looks sound enough in itself. , though paralleled in a third century Boeotian inscription (SEG 3. 357. 2), is here doubtless an Aeschylean coinage formed from on the analogy of the word is not connected with as the schol. and some commentators think.

²⁸ In reply to Lloyd-Jones Dawe claims that ‘in Suppliants all the power resides with the king Pelasgus’; but even if that were true (and it is far from obvious) it would not alter the crucial fact that anachronistic procedures are alluded to. Nor can I agree with Dawe that the last scene of the Septem shows a ‘preoccupation with the as a political entity’ any greater than is demanded by the new turn in the plot; and indeed we may remember that the were already credited with minds of their own at 6–8, as was the at 198 f.

²⁹ In fact I believe the antiphonal pattern of the lyric stichomythia should here be sustained with a text on the lines of Weil's Teubner Edition. I am firmly convinced that 999 is spurious (an intrusive gloss unconnected with the other interpolations).

³⁰ Many scholars have indeed thought that Eteocles does sacrifice himself for the good of the city; recent revivers of this ‘Opfertod theory’ are Dawe, PCPS 189 (1963), 37Google Scholar ff., and Kirkwood, G.M., Phoenix 23 (1969), 9Google Scholar ff. I cannot accept it, however; for some good counter-arguments to Dawe's view see Ferrari, F., Annali della Scuola Norm. Sup. di Pisa Ser. 3. 2. 1 (1972), 141 ff.Google Scholar

³¹ That Eteocles foresees his death in the duel has been denied by Patzer, H., HSCPh 63 (1958), 109 ff., and others. They argue on the basis of 672 that he expects to win and that when he talks of his death (684, 689–91, 697, 702–4) he must be referring to death from the pollution of fratricide. But that is not the most natural way of taking these lines, and it is better that we should attribute Eteocles' apparent confidence at 672 to the rhetorical context and not seek to pin him down to a single consistent set of expectations.Google Scholar

³² Though Headlam, W., On Editing Aeschylus (Cambridge, 1891), p. 95, quotes some late parallels: Polyb. 3. 53. 2, 18. 25. 4, Diod. 11. 32. 4, A.P. 9. 304 (Parmenion).Google Scholar

³³ Aelian openly quotes from Aeschylus's work elsewhere: Hist. An. 9. 42, 12. 8; fr. 22.

³⁴ For a phrase in enjambment giving an unexpected twist to the sense cf. Sept. 427 f., 531 f. (as compared with 47), Cho. 984 f., 988 f.

³⁵ See Dawe, , The Collation and Investigation of Manuscripts of Aeschylus (Cambridge, 1964), pp. 182 f.Google Scholar

³⁶ I feel, however, that it would have more point if there were a double meaning in not just ‘to pay the penalty’ but also sarcastically ‘to receive the honour (of burial)’ (as some of the older commentators in fact took the phrase). is used of honours paid to the dead according to the manuscripts (usually emended) at Soph. El. 915.

³⁷ On catachresis see Schuursma, J.A., De poetica uocabulorum abusione apud Aeschylum (Amsterdam, 1932), Fraenkel on Ag. 149.Google Scholar

³⁸ Corruption is certain in Pers. 664 f., probable in P. V. 582.

³⁹ Cho. 606 is admittedly in a corrupt ode, but I cannot believe in the ‘incendiary woman’ we meet in Page's text; Althaea is being charged with murder, not arson. , as a sinister riddling reference to her plan of burning a brand, is much more appropriate. And needs an epithet.

At Eum. 378 Page accepts the lengthening, presumably as an epic licence in a dactylic context; but we cannot reasonably assume that Aesch. made a distinction in this matter between dactyls and other metres if Sophocles and Euripides did not.

⁴⁰ At Cho. 854 it is probable, though not certain, that the right reading is

⁴¹ which Page prints, will hardly do. Until late Greek is found only at II. 15. 104 - and then as a participle, which, as can be seen from Fraenkel's note on Ag. 1174, need not presuppose the existence of the finite verb.

I am extremely grateful to Mrs. P.E. Easterling and Dr. R.D. Dawe for generous help. For the views expressed here I am, of course, solely responsible.

Addendum

It is unfortunately only at the last minute that I have been able to see Professor Hartmut Erbse's interesting article ‘Zur Exodos der Sieben (Aisch. Sept. 10051078)’, Serta Turyniana (Urbana, 1974), pp. 169 ff.

The first of the article's two parts consists of replies to linguistic and stylistic charges brought by Nicolaus and Fraenkel against the iambic part of the exodos. On lines 1005–25 Erbse's arguments and mine frequently support or supplement each other, but at some points we are in sharp disagreement; I would suggest that the last part of my article should be read in conjunction with the first part of Erbse's. Of the stylistic merits of 1026–53 I remain sceptical.

Erbse's Part II seeks to show that the themes of the exodos are not alien to the play as a whole. I am hardly an impartial judge, perhaps, but I find in Erbse's arguments confirmation of my own feeling that the burials have indeed a place in the Septem but that it takes special pleading to say the same of Antigone.