Segal attributes to me an obtuseness so enormous that I am surprised he finds it believable (even in me). At the end of the Bacchae‘the terrible suffering is there, as Seaford (finally) acknowledges in his review.’ This implies that I had somehow been unaware that realizing that you have just torn apart your son involves terrible suffering. He writes that there is ‘a word of mine that Seaford particularly dislikes’, namely ‘complexity’. But to dislike the word ‘complexity’ (as opposed to a claim to superior complexity) would in my view be almost insane. Again, it would be astonishingly obtuse to maintain (as S. claims I do) that ‘every tragedy uses aetiological myth about hero cult or about civic or ritual foundations in the same way or that this use will remain constant over several decades’.

Camouflaged by this kind of absurdity is an important methodological issue. For my reading S. still uses the words ‘rigidity’, ‘exclusionary’, ‘reductive’, ‘superimpose’, ‘monolithic’, ‘simply’, whereas his reading is ‘more open to the potential multiplicity of meaning’. But I am in this respect not as far from S. as he imagines. I have never claimed that tragedy is ‘always in a direct one-to-one correspondence’ with ‘cultural patterns’ (yet another example of S. attributing to me a view which I consider to be absurd). Of course tragedy does not in any simple way reproduce ritual or aetiological myth. My point is rather (and here is my real difference with S.) that when we interpret the complexity (yes, the infinite complexity) of the Bacchae we must (I emphasise) equip ourselves with detailed understanding (absent from S.’s book) of the Dionysiac mysteries, of aetiological myth, of the polis, and of a whole lot else. The fact is that S.’s dominant interest is in the woven quality of the text: in analogy, polarity, inversion. This he does very well, for a vast range of texts, including the Bacchae.

I respond to S.’s detailed criticisms as follows. He accuses me of a ‘slip’ in accusing him of a ‘slip’ in mistranslating Hippolytus 1425. What he wrote is that the suffering of innocent victims on stage ‘contrasts with the foundation of a cult by a distant Olympian in the remote future.’ But it is not founded ‘in the remote future’. S. exaggerated the contrast by simply misunderstanding di’ aionos makrou, which means (in effect) ‘to last for ever’. He now says that what he meant was ‘the “remote”, later time, indicated by Artemis’ future tenses, when these rites will be performed’. Note his crucial climbdown from ‘foundation’ to ‘performed’. But even this is misleading, for he is still (by confusing futurity with remoteness) trying to hold on to the implication of his earlier error, namely that there will be a significant interval (how long?) between the suffering of Hippolytus and the performance of his cult. Further, he adds the new point that the prenuptial cult founded is appropriate to the hero lamented in it ‘only in a ritual sense, but hardly so in terms of the character of Hippolytus that the play has constructed.’ Can he be unaware of the female resistance to sex and marriage (and concomitant lamentation) characteristic of prenuptial ritual? Perhaps I do him an injustice.

Of my criticisms of his misunderstanding and mistranslation(s) of Bacchae 860-1 ( … en telei theos / deinotatos, anthropoisi d’ epiotatos) S. writes ‘Seaford may dislike polarity, but I see no way to remove it from this passage’. Now to ‘dislike polarity’, or to want to ‘remove it from this passage’, would be almost insane. An advantage of my interpretation (of the transmitted text) is precisely that, unlike any other, it makes sense of the polarity between the two phrases each taken as a whole (not merely between deinotatos and epiotatos). This is particularly important, given the symmetrically emphatic positions of en telei and anthropoisi (see further my commentary).

Against my view that the polis does not ‘positively reject Dionysos’ S. had advanced 1295 ( emanhte, pasa d’ exebakcheuthe polis) as a ‘strong and clear reference to the responsibility’ of “all the city”‘. Now he specifies that he meant responsibility for the god’s vengeance. He also now argues that the polis in this line ‘seems to be identified with the raging women’. But even if we accept this, mention of being put into a Bacchic frenzy is certainly not a strong and clear reference to responsibility for the vengeance of Dionysos. Perhaps S. has in mind that the frenzy was a punishment for rejecting the story of Dionysos’ birth, and so a sign of responsibility. But it was only the royal women, i.e. precisely not the whole polis, who rejected the story (26). No less odd is S.’s new claim that lines 50-52 are further evidence for the ‘responsibility’ of the city. In these lines Dionysos says that if the polis of the Thebans tries by armed force to take the maenads from the mountain he will lead the maenads against them. But of course in the event the tyrant, despite his threats of military force, goes to the mountain alone. The earlier attack by the herdsmen on the maenads is not by the ‘polis of the Thebans’, and even if it were, the distinction between Pentheus and the polis (which is my whole point) would actually be reinforced by the episode, for the herdsman-messenger recognises the miraculous power of Dionysos whereas Pentheus persists in his hostility.

On metatheatricality S. now writes that ‘both ancient and modern audiences … like to think about how literature constructs its world of make-believe, how it convinces us of the reality of that world and what kind of “truth” that represented world can have alongside the other, more “factual” truths that guide our lives’. I very much doubt that ancient audiences (as opposed to some philosophers) pondered such issues (and how often do even modern ones?). But more importantly, a few paragraphs earlier S. wonders how I can deploy aetiological myth in interpreting the play, given that ‘we do not yet have time machines’ to enable us to know the ancient audience’s response. How on earth then can S. blithely assert, given that he too has no time machine, that the ancient audience was interested in metatheatricality? The clue is in his unthinking phrase ‘both ancient and modern audiences’. In the earlier passage he claimed that ‘if Seaford is right about Bacchae, there is an enormous and unbridgeable gap between modern audiences’ response to the play and the response Seaford constructs for the original audience’. Well, of course there is an enormous gap. It is a gap that I attempt (to the limited extent possible) to bridge. Modern audiences of the Bacchae do not revere and fear Dionysos. They do not participate en masse in his festivals. They do not achieve salvation by being initiated into the mysteries. They do not live in a polis whose crucial survival depends on the collective performance of cult (impelled by the warning given in aetiological myth). They do not on the whole live in daily fear of tyranny. They are not hostile to Thebes. And so on. S. ‘s position on the other hand is in effect that because (not having time machines) we cannot be sure that the response of the ancient audience was different from that of most modern ones, we can assume that it was the same, i.e. that they saw metatheatricality, irresolvable contradiction, irredeemable suffering, and so on.

When I say that he ‘privileges disorder’ I am, according to S., ‘completely ignoring a main point’ of his book. But then, a few words later, he quotes his own statement that the tragic element in the play ‘brings together both order-creating and order-dissolving forces of personal and social life in a balance that is always shifting and unstable’ — precisely an example of the privileging of disorder: order and disorder are brought together in a balance that is always shifting and unstable (i.e. disordered).

S.’s aim in quoting this statement is in fact to correct my remark that it was, unknown to him, anticipated by Oudemans and Lardinois, who actually refer to his earlier book. I concede my errors, and so accept the correction. If others (O. and L.) had come to the same idea independently, that would, if anything, have constituted support for the idea. But now alas it seems that S. was after all the originator of what I regard to be a vacuous manoeuvre.

Another error of mine corrected by S. is my translation of the aorist exebakcheuthe as ‘was in a bacchic frenzy’ rather than ‘was put into a bacchic frenzy’. Finally, another (frequent) claim of S. that I am happy to concede is that in some basic matters he has the majority of interpreters on his side. I am well aware that I stand outside that mainstream to which he has so creatively contributed.