¹ I follow Page's text and punctuation in this stanza. (codd.) cannot be excised, since does not fit and does not fit Hermann's could be right, but is uneconomical. is necessary, impossible: Blomfield's seems certain, and, if is to have its normal position, must be preceded by sense-pause (cf. Fraenkel, E., ‘Kolon und Satz, II’, NGG (1933), 319 ff.Google Scholar = Kleine Beiträge i. 93Google Scholar ff.; JHS 97 (1977), 128 n. 11),Google Scholar being understood as subject to (see Fraenkel on Ag. 71). Then need to be governed by a transitive verb, and Butler's is best.

² The intensive repetition of is not brought out in the translation here (cf. n. 22 below).

³ Reading Enger's governed by with active sense (cf. PV 860 ). = ‘wins past’, ‘outstrips (all) in victory’ (so Blass). See Appendix. In here rendered ‘wicked’, the privative is pejorative rather than negative (see Fehling, D., Hermes 96 (1968), 150–5,Google Scholar esp. 152, Die Wiederholungs-figuren vor Gorgias (1969), p. 288,Google Scholar cf. E. Or. 163 For Scylla this means illicit love; for the Lemnian women, with whom the correspondence here is closest (see pp. 253 ff. below), it means a love which frustration and jealousy have turned to hatred.

⁴ Reading for in 624, in 629, and assuming a verb meaning ‘I mention’ or the like to have been displaced by the second in 628. See Appendix.

⁵ For this curious story see Burkert, W., CQ N.S. 20(1970), 1–16.CrossRefGoogle Scholar

⁶ Das Opferam Grabe (1896), p. 213;Google ScholarHerakles ² (1895), ii. 20 ff.Google Scholar

⁷ As is remarked by Oehler, R., Mythologischi Exempla in der älteren griechischen Dichtung (1925), p. 81, though he does not do justice to Wilamowitz's interpretation.Google Scholar

⁸ The romantic motivation; first found in Ovid, but not necessarily Hellenistic.

⁹ Cf. Oehler, , op. cit. (n. 7), p. 79.Google Scholar

¹⁰ Hom. Il. 24.535 ff.; E. Ba. 337–40.

¹¹ There is no such example to be found in Oehler's careful survey.

¹² Hence ‘awake’ (Erfurdt), ‘take up’ (West, M. L., BICS 24 (1977), 99Google Scholar): but neither is particularly appropriate.

¹³ Preuss, followed by Wecklein, Blass, Groeneboom, and Rose. Cf. Pers. 93–100, where a comparable transposition is required in lyrics, and ibid. 552–61, where a whole stanza (or its equivalent) is omitted in the text of M and added in the margin, showing how the error might have occurred. The text of the Choephori, which depends on M alone is still more vulnerable to such errors, as is that of the Supplices, where 88–90/93–5 are commonly transposed. There may also be some dislocation in Cho. 429–55 (cf. Dawe, R. D., The collation and investigation of manuscripts of Aeschylus, [1964] 161–4,Google Scholar and Eranos 64 (1966] 1–21, esp. 6–13).Google Scholar

¹⁴ See Groeneboom's note. Cf. Ag. 785–9 (Groeneboom), PV 30, 507. On the systematic connection between and see Palmer, L. R., ‘The Indo-European origins of Greek justice’, Transactions of the Philological Society 1950 (1951), pp. 153–65.Google Scholar

¹⁵ Holtsmark, E. R. (CW 59 (1966), 251Google Scholar) argues that the destruction of the Lemnian women implied by 636 leads directly to the idea of retribution in the last strophic pair. But the retribution awaiting Clytemnestra follows more naturally on her crime.

¹⁶ CP 62 (1967), 182–4.Google Scholar

¹⁷ On the tendency of Greek poets to invent new features in mythical exempla to suit the case they are illustrating, see Oehler, , op. cit., p. 7;Google ScholarKakridis, J. T., Homeric Researches (1949), pp. 86 ff.Google Scholar; Willcock, M. M., CQ N.S. 14 (1964), 141 (nn. 1,2).CrossRefGoogle Scholar

¹⁸ Lebeck, Anne, The Oresteia: a study in language and structure (1971).Google Scholar

¹⁹ In particular, the primary reference of 623–4 is toClytemnestra, not Aegisthus, as Verrall and others have held (see below, p.259 and n. 29); though Aegisthus' partnership is implied by and perhaps by

²⁰ Oehler, , op. cit., p. 78;Google ScholarHoltsmark, (op. cit. (n. 15), pp. 215–16)Google Scholar, who also points out that the exempla exhibit the same pattern.

²¹ Holtsmark, (loc. cit. (n. 15)) makes the focus of the whole ode, and thereby seeks to justify the traditional order. do not think he succeeds, though I gladly tccept much of what he says.Google Scholar

²² Bundy, E. L., Studia Pindarica [University of California Publications 18.2 (1962), 38. ‘What do I boast beside the mark?’, i.e. ‘My boast is relevant’, is a possible rendering, but it fits the context less well.Google Scholar

²³ The structure is similar to that of the opening of our ode: the climax of the initial priamel is itself a priamel.

²⁴ The point holds good for any text giving the sense here, which as Page says is the sense required (app. crit. ad loc.).

²⁵ The intensive repetition can be brought out only at the cost of some freedom in translation. ‘Ruthlessness’ conveys the idea of better than the usual equivalent ‘daring’

²⁶ On ring-form as a frame for mythical exempla see Oehler, , op. cit., p. 7;Google ScholarWillcock, , op. cit. (n. 13), p. 142,Google Scholar with n. 2; Köhnken, A., Die Funktion des Mythus bei Pindar (1971), pp. 66–8, 123–50.Google Scholar Mr. Peter Pickering suggests that may simply be a further case of intensive repetition. We might compare the repetition of and cognates in the first stasimon of the Prometheus Vinctus (cf. Euripides and the Judgement of Paris, JHS Suppl. 11 (1965), 18 n. 2Google Scholar). That occurs at the beginning and end of the ode, however, is hardly accidental there either. (We might also regard the verbal echo in Cho. 630 as marking a cor-responsion between the climax of the exempla and that of the opening priamel, cf. n. 20.).

²⁷ Cited by Wecklein in his Greek edition (1910). Wecklein himself took to govern the following genitives, a usage not justified by the comparative genitive with (see Blass's note, and cf. JHS 96 (1976), 134 n. 53).Google Scholar

²⁸ This is one of the meanings given by Phrynichus ap. Bekker, Anecdota i. 8.8. The word evidently existed, since it is cited in more than one inflection, but it is not clear that the lexicographers or the scholiast knew what it meant, nor why a negative adjective derived, as it presumably is, from should mean ‘shameless’. Possibly ‘incautious’, ‘rash’ would be nearer the mark, cf. ibid. and the normal use of

²⁹ Aeschylus, , Choephori (1894),Google Scholar Appendix, § 17. Blass shares his view, despite Ag. 1435, which he prays in aid, citing in Sept. 603, Eum. 560 (cf. also S. Trach. 1046); for the positive force of he cites in Sept. 603, Eum. 560 (cf. also S. Trach. 1046). Lloyd-Jones, reading (with of Clytemnestra: see below), renders ‘honoring the hearth of the house that lacked warmth’ (The Libation Bearers, by Aeschylus (1970), p. 45Google Scholar). He also compares Ag. 1435–6 and explains: ‘a hearth that had only a man like Aegisthus to light it lacked proper warmth’. So Fraenkel, on Ag. loc. cit.:Google Scholar the hearth is normally kindled by the legitimate lord of the house; Clytemnestra is there accepting Aegisthus as such, the chorus here mean that he is a usurper. We might compare Ag. 968–9 But that is part of an elaborate complimentary metaphor from which Aegisthus is naturally excluded, while the obvious erotic overtones of 1435–6, with in Cho. 625, make it very difficult to take in this way.

³⁰ Lucian, , Nigrinus 57Google Scholar cited by Groeneboom to support the contrary view, merely suggests that Lucian made the same error. Fraenkel, in his note on Ag. 483–4 insists that means ‘authority’, ‘rule’, and so also here and at PV 405. This could be the sense with my interpretation: the chorus honour a queen who does not yield to or kill the rightful lord on his return from war. But the word could mean ‘will’ or ‘spirit’ in any or all of these places: the extension from the sphere of war is easy-enough (cf. ), and the analogy of meaning ‘authority’ in E. Hipp. 975 etc. is not conclusive.

³¹ It has also been suggested that has intruded from a scholion. This is possible, but since gives the right sense, it is more likely that some derivative of stood in the text.

³² Strictly the logic is: ‘You too, Deianeira, should have good hope, since it is no painless lot Zeus has laid on mortals’ (see Denniston, , GP, p. 195); and so in Cho. 623 ‘since I have recalled cruel pains, not untimely do I mention also …’. For the postponement of with this sense, cf. e.g. Trach. 280, 1275, O. T. 1409. (Denniston's interpretation of Trach. 127–8 is rightly followed by subsequent editors; the sense ‘not even Zeus’ suits neither the logic of the passage nor the classical concept of Zeus.)Google Scholar

³³ Cf. Dale, A. M., The Lyric Metres of Greek Drama ² (1968), p. 168;Google Scholar though her examples are of the type whereas as prosodiac-enoplian clausula is usually interpreted as a catalectic form of ithyphallic, which one would not expect to correspond with Schroeder conjectured in 839, giving in both places; Maas, in his copy of Schroeder's Cantica now in my possession, accepted the licence in responsion.

³⁴ Cf. Conomis, N. C., Hermes 92 (1964), 31.Google ScholarAndr. loc. cit. is also usually taken to be a hypodochmiac. Pace Conomis, (op. cit., p. 32Google Scholar), there are certainly dochmiacs in the vicinity, and this might justify the licence in a prosodiac clausula.

³⁵ Some follow the scholia in taking to mean , ‘vain of thought’, or ‘volatile’, ‘unstable’ in mind, cf. Hesychius The thought would then be very like that of S. Aj. 416 This could be right, but the evidence for this meaning is not very convincing: we should expect to be positive rather than negative. is also recognized in the scholia, but the meaning of the epithet is still a difficulty, and the simple dative could not stand by itself: must be altered to give it a construction. (Emperius' introduces an ellipse of the copula with a participle, which is extremely rare except with participles regularly found in the accusative absolute, e.g. (cf. K-G. ii. 40 f.).

³⁶ has three terminations elsewhere in Aeschylus, but is found with two in Sophocles (Ichneutae 239) and Euripides. It is normally trisyllabic in tragedy, but (1) occurs at Pers. 985, and is generally read ibid. 271; (2) Sept. 278 ff. is corrupt, but the phrase seems sound, and cannot be trisyllabic without drastic surgery such as Murray's; (3) it is pretty certainly disyllabic at S. Ichneutae 239: this is a satyr-play, but as is not a comic word, it is fair evidence; (4) the disyllabic form is implied by Sept. 72 S. O. C. 1533

³⁷ So Frisk, , s. v., with references to other viewsGoogle Scholar (see esp. Björck, G., Das Alpha impurum (1950), pp. 340–2).Google Scholar