¹ Excision was first proposed by Bergk, T., Griechische Literaturgeschichte, iii (1884), 512 n. 14Google Scholar, who regarded 1021–55 and 1056–80 as alternative versions. The next major assault was Müller, G., ‘Interpolationen in der Medea des Euripides’, SIFC 25 (1951), 65–82Google Scholar. A full bibliography of the question is found in Reeve, M. D., ‘Euripides, Medea 1021–1080’, CQ n.s. 22 (1972), 51–61CrossRefGoogle Scholar, who argues strongly for deletion. Three more recent discussions are Zwierlein, O., ‘Die Tragik in den Medea–Dramen’, Literaturwissenschaftliches Jahrbuch 19 (1978), 27–63Google Scholar; Lloyd–Jones, H., ‘Euripides, Medea 1056–80’, WüJbb N.F.6a (1980), 51–9Google Scholar; and Erbse, H., ‘Zum Abschiedsmonolog der euripideischen Medeia’, Archaiognosia 2 (1981), 67–82Google Scholar, a reference I owe to Professor Lloyd–Jones.

² Christmann, E., Bemerkungen zum Text der Medea des Euripides (diss. Heidelberg, 1962), pp. 125–45Google Scholar.

³ Mastronarde, D. J., Contact and Discontinuity: Some Conventions of Speech and Action on the Greek Tragic Stage (Berkeley, Los Angeles, London, 1979), p. 110Google Scholar. Incidentally, I do not see why he posits an attendant in addition to the Paedagogus. K. H. Lee, per litteras, asks ‘Why get rid of the Paedagogus if not to leave Medea alone with the children?’.

⁴ Bain, D., Masters, Servants and Orders in Greek Tragedy: Some Aspects of Dramatic Technique and Convention (Manchester, 1981), p. 33Google Scholar. See also his earlier Actors and Audience: A Study of Asides and Related Conventions in Greek Drama (Oxford, 1977), pp. 26–7Google Scholar.

⁵ See the trenchant criticisms of both the geistesgeschichtliche Schule and the Seelendrama approach in Schlesinger, E., ‘Zu Euripides' Medea’, Hermes 94 (1966), 26–53Google Scholar and Lloyd-Jones's summary (p. 51) of the fin-de-siècle currents of thought that gave rise to them.

⁶ Erbse (p. 73) writes ‘Tα⋯τα kann nur [!] das Attentat bezeichnen, und als Subjekt zu ⋯κφε⋯ξεται versteht man mühelos [!] die Person, für die es kein Entrinnen gibt’. One can only gasp at Erbse's audacious dogmatism. The possibility many scholars regard as by far the less likely of two is described as the only possibility, and the suppletion many consider impossible is ‘effortless’.

Kvičala proposed transposing 1064 to follow 1066, as one of my anonymous readers pointed out. This makes it easier to supply Creon's daughter, though one would still welcome a parallel to the change of subject between two third-person verbs in the same line connected by ‘and’. This remedy would also commend itself more if there were not so many other problems in the passage.

⁷ Steidle, p. 160 n. 51, suggests that, grammatically difficult as it may be, τα⋯τα must be the subject of ⋯κφε⋯ξεται and the verb must be taken either as a passive or — better — as active, with suppletion of τ⋯ μ⋯ γεν⋯сθαι. His clear-sighted rejection of the other alternatives is to be commended, but his last suggestion — supported by no parallels — strikes me as desperate. It may be that at some point a Greek wrote τα⋯τα οủκ ⋯κφε⋯ξεται. and meant by it ‘these things shall not escape fulfilment’, but I doubt whether it was a fifth-century dramatist who did so.

⁸ There are not many passages in Greek tragedy which deserve to be called ‘high fustian’ more richly than 1056–8:

Nay, nay, my pride and anger, do not so!

Let be, hardhearted wretch, spare thou my sons!

Living with me shall they make glad thy heart.

Translation is more than usually adequate. Note that the author either makes Medea apostrophise her anger, and the pronouns in 1058 (‘me’ and ‘thy’ above) distinguish her thumos from Medea herself, which is a frigid conceit; or the thumos is Medea herself, in which case the pronouns in 1058 are just bad writing.

⁹ Reeve says (p. 56n. 4), ‘As regards the rhetorical structure of the two [Neophron and Euripides] Jachmann is right, but there is more poetry in Med. 1056–80 (1069–75).’ I note that he seems to wish to exclude from this praise the earlier lines in the passage, which shows, I think, that his Stilgefühl agrees roughly with mine.

¹⁰ See Denniston, , GP 250–2Google Scholar for the use of non–connective κα⋯ δ⋯ to mark ‘vivid perception by mind, ear, or eye’, ‘the provision or completion of something required by the circumstances’, and as an approximation to ἤδη. Fortunately we need not assign our passage exclusively to one of these categories.

¹¹ Strictly speaking, we cannot tell from Galen, our source, just how Chrysippus interpreted these lines. He seems to have used them in an argument about the nature of the conflict between reason and passion, maintaining that the passions are really a form of judgement: see SVF iii, p. 124 and compare ii, pp. 255–6. Galen objected that the lines contradict Chrysippus’ thesis.But whether Chrysippus thought that Medea's lines showed passion overcoming moral judgement or prudent self-regard is not recoverable. Other authors, however, make it clear that at some point Medea's lines became a tag to express the defeat of one's sense of moral obligation.especially toward others. See Galen, , CMG v.306 KGoogle Scholar. (=111 3.14–16 in P. De Lacy's edition [Berlin, 1978–80]), Plutarch, , De vit. pud. 533Google Scholard, who names perjury and other similar crimes as the bad things one knows well one is doing when giving in to improper requests, and Lucian, , Apol. 10Google Scholar. This interpretation is taken over by Snell, B., ‘Das früheste Zeugnis über Sokrates’. Philotogus 97 (1948), 125–35Google Scholar and Scenes from Greek Drama (Berkeley and Los Angeles. 1964).47–69Google Scholar. It subtly influences much subsequent discussion of the play. See for example Kitto, H. D. F., Greek Tragedy (3rd ed., rpt. 1970), p. 196Google Scholar.

Readers interested in the discussion of this speech by Stoics and others will profit from Gill, C., ‘Did Chrysippus understand Medea?’. Phronesis 28 (1983), 136–49CrossRefGoogle Scholar. a reference I owe to Professor Reeve.

¹² That βουλευμ⋯των in 1079 cannot mean ‘moral insight, knowledge of right and wrong’ is clear both from the word itself and from its use elsewhere in the play. Emendations have been proposed, e.g. μαθημ⋯των (Koechly) or сωφρόνων for τ⋯ν ⋯μ⋯ν (Stadtmueller), but they lack any probability. Diller's attempt to make κρε⋯ссων mean ‘in control of’ and Dihle's attempt to interpret θυμόϲ as Medea's tenderer feelings and to read the whole passage as a renunciation of her plans have been justly rejected. See Diller, H., Hermes 94 (1966), 267–75Google Scholar; Dihle, A., ‘Euripides' Medea’, SB Heidelberg, Phil.-hist. K.1. 5 (Heidelberg, 1977)Google Scholar; Diller is answered by Reeve, p. 59 n. 2, Dihle by Zwierlein, pp. 35–7.

¹³ Phaedra's whole speech has been badly misunderstood, largely because of its imagined relevance to the history of ideas. On the argument of the speech, see Claus, D., ‘Phaedra and the Socratic Paradox’, YCS 22 (1972), 223–38Google Scholar and Kovacs, D., ‘Shame, Pleasure, and Honor in Phaedra's Great Speech’, AJP 101 (1980), 287–303Google Scholar. Another good parallel is Alc. 1080, tellingly cited by Steidle (n. 2 above), p. 148 n. 82 against Snell's Geistesgeschichte.

¹⁴ Another example may be Med. 1051, where we should perhaps read τολμητ⋯ον τ⋯δ’. This makes the reference of τ⋯δε easier to understand besides making ⋯λλ⋯ function more naturally.

¹⁵ It is not only a heroic Medea we see here but also, as I will argue elsewhere, a Medea who is the unwitting agent of higher powers, which use her for their ends without concerning themselves with the ruin of her happiness. Line 1080 suggests the Achilles theme of a wrath that brings ruin on its object (a ruin that is both deserved by him and intended by the agent, as Achilles' wrath brought harm to Agamemnon and the Greeks) but that also causes misery to the wrathful man himself, both effects belonging to the inscrutable ordinance of Zeus.

¹⁶ I have elsewhere recommended deletion of lengthy passages on the ground that they stand in varying degrees of contradiction with their context: Andr. 333–51 (HSCP 81 [1977], 148–56Google Scholar), Ion 595–606 and 621–32 (TAPA 109 [1979], 116–24Google Scholar), Sup. 442–55 and Pho. 549–67 (GRBS 23 [1982], 31–50Google Scholar). In all instances there is a strong linguistic case to be made.

¹⁷ Since I am not the first to propose minor surgery, I ought to make clear why I am not satisfied with other proposals. Seeck, G. A., ‘Euripides Medea 1059–68: A Problem of Interpretation’, GRBS 9 (1968), 291ffGoogle Scholar., suggests deleting 1060–3. This eliminates the motif of external necessity for the murder but leaves the desperate problem of 1064 exactly where it was. Lloyd-Jones (p. 56) suggests deleting 1059–63, but the absence of any adversative in 1064 to show that Medea has changed her mind once more is a serious drawback. In addition, 1064 remains a problem.