^{page 49 note 1} The Theodore Spencer Memorial Lecture of 21 November 1950 (London, Faber & Faber, 1951), 32. In the first act of Tom Stoppard's play Rosencrantz and Guildenstern are Dead Guildenstern puts the point more ironically: ‘What a fine persecution—to be kept intrigued without ever quite being enlightened.’

^{page 49 note 2} ii (1969), 142.

^{page 50 note 1} This limitation of scope, together with the organization of my material that it involves, has certain inevitable consequences, which have given rise to demurrers on the part of some Euripidean scholars with whom I have had the privilege of discussing this paper: particularly Professor Richard Kannicht, to whose generously detailed comment my final draft owes a great deal. The central objection is this: that by presenting my ideas in the way that I do, I deliberately turn my back on a question of central importance: the question, to whom are the developments in each particular scene that I discuss a surprise? To the audience primarily, or only to the characters on stage at the time, or to characters and audience together? That this distinction is always relevant and occasionally of crucial importance, I make no attempt to deny; but if my paper had been focused on that question, it would have been a different paper, and several points of a different kind would have been at least blurred, if not entirely removed from the field of vision.

^{page 50 note 2} Op. cit. 131 f.

^{page 51 note 1} A trifle cynically, and yet at the same time (as Kannicht does well to remind me) with a precisely calculated dramatic intention. Electra's suicide threat in the brief preamble to the messenger scene (755) picks up the similar threats that she expressed at the end of the previous episode (686 ff., 695 ff.), and thus emphasizes the depressive side of her hysterical character at a moment of nervously tense uncertainty.

^{page 53 note 1} Op. cit. 131. Cf. also Méridier's and Barrett's annotations ad loc., and Arnott, Peter (not to be confused with the author of this paper), Greek Scenic Conventions (Oxford, 1962), 36 f.Google Scholar

^{page 53 note 2} In Sophocles' Philoctetes, which was produced some years after Euripides' Electra, the older dramatist plays with dramatic convention in a way curiously parallel to that of Euripides here. Philoctetes asks the chorus to withdraw (1177), the chorus agree to do so (1180), but then Philoctetes changes his mind (1182) and the chorus remain on stage. Here clearly Sophocles plays his game with the conventions of tragedy in a situation of high dramatic quality, in order to illuminate one effect of pain on a heroic character.

^{page 54 note 1} There is some doubt over the precise moment of the chorus's exit; cf. the commentaries of Dale on 374 and Kannicht on 385.

^{page 54 note 2} It would have been an attractive bonus to the argument if I had been able to claim that this was the first time that a tragic chorus had made so dramatic an exit on to the stage and off by the palace's central door in the middle of a play, but such a suggestion would be injudicious: too many fifth-century tragedies have passed into total oblivion. Admittedly, there are only four or five other known instances of a tragic chorus leaving its post in the orchestra in the middle of a play. At Aesch. Eum. 231, Soph. Aj. 814, Eur. Alc. 743, and Rhes. 565, the chorus leaves the orchestra and then later return through the parodoi, so never stepping on the stage. Eur. Phaethon, fr. 781 Nauck² = 245 ff, Diggle, is more puzzling. It is often alleged that here either the main chorus itself or a secondary chorus is ordered into the palace during the play by Merops (cf. Decharme, P., Euripide et l'esprit de son théâtre, 424 ff.Google Scholar; Ritchie, W., The Authenticity of the Rhesus of Euripides, 118 ff.)Google Scholar, but recently powerful voices have been raised against that interpretation (cf. Diggle, 's edition of Phaethon, 150, n. 2).Google Scholar

^{page 54 note 3} Cf. Kannicht's edition of the play, ii. 10 f.

^{page 55 note 1} This translation is basically Edmonds's.

^{page 55 note 2} pp. 50 f.

^{page 56 note 1} The Greek runs λαβὼν δ⋯ κ⋯πτει … When the actor paused on this verb at the penthemimeral caesura, the audience could not immediately have known whether an object such as ν⋯τον Αἰγ⋯σθου was likely to follow it or not.

^{page 56 note 2} Cf. Gredley, B., GRBS ix (1968), 415 ff.Google Scholar, especially his comment: ‘It would seem that Euripides has deliberately created a false impression of Helen's murder, and suspended its refutation for over a hundred verses in order to make the news of her disappearance the more unexpected and effective.’

^{page 57 note 1} The author of this paper must acknowledge responsibility for this and other pieces of translation from the Orestes. Elsewhere the generally excellent versions of the Chicago Complete Greek Tragedies are used, except where a more literal rendering of a phrase or sentence is required in order to convey Euripides' purpose more exactly.

^{page 58 note 1} Unfortunately, Euripides' deliberate effect has been marred by later interpolators. After the Phrygian slave has finally and clearly announced the mysterious disappearance of Helen, he is confronted in the text as vie now have it by Orestes, and a dialogue of muddled opacity follows in which Helen is referred to as dead (1512 f., 1534, 1536), disappeared (1555 f.: ‘an idle rumour’), dead again (1566, 1579), and disappeared again (1580 ff.), without any apparent dramatic point for the twitching and switching. Muddle is removed and point restored by the deletion of 1503–36 as a later interpolation, as Gredley's paper (loc. cit. p. 56, n. 2 above) convincingly argues. M. D. Reeve informs me that in a paper to be published in GRBS he has put forward supplementary arguments in support of Gredley.

^{page 59 note 1} CPh lvii (1962), 170–2.

^{page 59 note 2} Cf. Reinhardt, K., Tradition und Geist (Göttingen, 1960), 237.Google Scholar

^{page 60 note 1} Cf. especially Strohm, H., Euripides: Interpretations zur dramatischen Form (Zetemata, xv: Munich, 1957), 126 f.Google Scholar; Hourmouziades, N. C., Production and Imagination in Euripides (Athens, 1965), 30 and 168Google Scholar; and Webster, T. B. L., ‘Euripides: Traditionalist and Innovator’, in The Poetic Tradition (edited by Allan, D. C. and Rowell, H. T.: Baltimore, 1968), 29 and 32.Google Scholar

^{page 61 note 1} In his note on 894, where Page wrongly assumes an extra exit (1053) and an extra return (1069) of the children. Contrast his note on 1053.

^{page 62 note 1} He is not a ‘messenger’, despite what the manuscripts say; see Kannicht's commentary, ad loc., and (more briefly) Dale, ad loc., and Gredley, 410 n. 2. Kannicht's commentary discusses this scene in lively and illuminating detail, but his general interpretation of it differs substantially from mine.

^{page 62 note 2} Dale's commentary, on 711 ff.

^{page 64 note 1} The Bostonians, ch. 19.

^{page 64 note 2} This paper has benefited greatly from the positive criticism of several scholars: particularly A. D. Fitton-Brown, R. Kannicht, and M. D. Reeve. My debt to them is gratefully acknowledged.