Hiller seems to think that alternative (ii) offers less scope for confusion and error on the part of the later writer, though even here we must bear in mind the carefully formulated warning of  J. A. Davison (“Quotations and Allusions in Early Greek Literature,” Eranos 53 [1955]: 137 = From Archilochus to Pindar [London 1968] 81–82): “All that we can feel sure of is that Pausanias (or his authority) found in Callinus a phrase or phrases ascribed to Homer which resembled some words in the Thebais closely enough to lead him to infer that Callinus was actually quoting the Thebais and ascribing it to Homer.” How much more uncertainty would be uncaged if (i) were the truth! Hiller does well to stress (p. 326) the large number of admittedly brief and passing references to the Theban saga in the Iliad and {29|30} the Odyssey. If an allusion to one of these from Callinus was misinterpreted we should be in a sorry way. Enough said, then, to put us well on our guard against any smooth and unpremeditated deduction from this notorious passage. Wilamowitz’s cautious summing up (1916:364n1) can hardly be bettered: “Ob die bis in das dritte Jahrhundert gelesene Thebais mit der, welche Kallinos vor sich gehabt hatte, ausser dem Stoffe noch irgend etwas gemein hatte, wusste niemand.”

We may therefore conclude that, though the Argives, Argos, and Adrastus would be mentioned far more often in the Thebais and the Epigoni than in the Iliad and the Odyssey, nevertheless they are mentioned enough in the latter pair of poems (evidence marshaled by Hiller [327–328] and Scott [22–24 = 18–20]) for the story to work perfectly well if τῶν Ὁμηρείων ἐπέων as a phrase bears the meaning that a present-day reader would naturally place upon it. And (pace e.g. How and Wells or Cingano) it would surely be a little odd if Herodotus, who at IV 32 = Epigoni Τ1/F2 expresses doubt concerning the Homeric authorship of the Epigoni, were here calmly to couple it with the Thebais and label the combination τὰ Ὁμήρεια ἔπεα.

Two other passages have wrongly been thought to contain traces of this tradition of Homer as the Thebais’s author: (i) Antigonus of Carystus 25 (Rerum naturalium scriptores graeci minores p. 9 Keller) attributes to ὁ ποιητής a line and a half of gnomic advice which some have assigned to the Ἀμφιάρεω Ἐξελασία, {30|31} a work most plausibly interpreted as part of the Thebais (as we shall see pages 99–102 below). But though we may accept that ὁ ποιητής is here equivalent to Homer, the rest of the hypothesis does not follow at all (see pages 101–102 below). Nor should (ii) Horace’s Ars Poetica 146 (nec reditum Diomedis ab interitu Meleagri [orditur]) be taken even as necessarily referring to the Thebais, and still less as implying the Roman poet’s attribution of that epic to Homer: see Brink, Horace on Poetry: The Ars Poetica (Cambridge 1963–1982) 442.

By contrast, the fragment of Dionysius of Samos (ὁ κυκλογράφος) cited as Τ4 (see page 136), which claims Homer as a contemporary of the Theban as well as the Trojan War, does seem to reflect the tradition. So, for instance, among earlier scholars, Grote (History of Greece 2.129n2), and Felix Jacoby, among the more recent (in his commentary ad loc. [1^A.492]), have inferred.

We must now face another, scarcely less important, issue. When Pausanias introduces his remarks on the Thebais and its authorship with the statement ἐποιήθη δὲ ἐς τὸν πόλεμον τοῦτον καὶ ἔπη Θηβαίς, he has just been describing the siege of Thebes as the climax of the Epigoni’s campaign against that city. One might, then, expect “this war” to refer most immediately to this second, successful enterprise as well as the earlier failure. The fact has not escaped the attention of those scholars who believe the epic known as the Epigoni was part of a larger Thebais (so especially Bethe [1891:36–37 and 122]) and they triumphantly cite our passage as an evidence of this theory (see page 107 below). Even the cautious Rzach, who rejects the more extreme forms of this dogma, deduces from Pausanias’ words (1922:2374.42): “er wählt also eine Gesamtbezeichnung.” One can see why they came to this conclusion. But a closer examination of the structure of Pausanias’ whole argument here will also show the needlessness of such an inference.

The relationship between the Iliad and the Thebais has long been a matter of debate and dispute. Four older treatments of the topic still repay attention: Welcker 1865:2.320–332, Bethe 1891:174–147, Robert 1915:1.185–199, and Friedländer 1914:317–329 = 1969:34–42. Of these, the last two, which originally appeared at about the same time and do not, therefore, show awareness of each other’s conclusions, are the most rewarding. Robert’s analysis is the fuller and more detailed, but labors under the disadvantage of approaching the topic via the not particularly fruitful conviction (needlessly stressed at every turn of the argument) that the author of the relevant Iliadic passages cannot be the author of the Thebais. {32|33} Friedländer’s more concise study shows a greater awareness of the principles that underlie the whole issue. A Spanish monograph by José B. Torres-Guerra, La Tebaida Homérica como fuente de Iliade y Odisea (Madrid 1995), with English summary pp. 78–82, fails to take into account the monograph of Ø. Andersen (1978), the best treatment of the issue to date.

(i) Iliad IV 365–400

Here Tydeus’ activities are described by Agamemnon in order to supply an exhortatory paradigm for Tydeus’ son Diomedes, whom the leader of the Greeks mistakenly supposes to be skulking away from the battle, quite unlike his father. Diomedes’ alleged cowardice is in especially striking contrast to Tydeus’ behavior about the time he accompanied Polyneices on a peaceful embassy to Mycenae in the hope of gathering forces for the expedition against Troy. Mycenae would willingly have supplied such troops but παραίσια σήματα from Zeus deterred her (line 381). At a somewhat later stage, when the expedition was already under way, Tydeus was sent on another mission, this time to the enemy capital. At Thebes he went to the palace of Eteocles (line 386), challenged the {33|34} Thebans to an athletic contest, and beat them all easily, with Athena’s help. The angry Thebans set an ambush for him as he returned to the expedition, but here too Tydeus emerged victorious and killed all his assailants except for Maeon, whom he spared and sent back to Thebes, θεῶν τεράεσσι πιθήσας (line398).

(ii) Iliad V 793–813

Here again Diomedes is the recipient of a paradigmatic exhortation whose main theme is his inferiority to his father. This time the speaker is Athena, who recalls with authoritative knowledge an occasion when Tydeus went alone as a messenger to Thebes (lines 803–804) and was invited by its inhabitants to a feast. Instead, he challenged them to a contest and vanquished them all easily, thanks to Athena’s help.

Less detailed but still important are the two following passages:

(iii) Iliad X 284–289

This time Diomedes himself reminds Athena (in a prayer designed to win her support by appeal to the principle “help now, as in the past”):

(iv) Iliad V 115–120

In a very similar context (a prayer introduced by exactly the same formula of address to Athena) Diomedes requests aid εἴ ποτέ μοι καὶ πατρὶ φίλα φρονέουσα παρέστης | δηΐωι ἐν πολέμωι. Diomedes also makes general reference to his father’s career in the Theban War at Iliad VI 222–223 and XIV 110–132.

Welcker (1865:2.328–329) suo modo took (i) to be derived from the Thebais; B. Niese (Die Entwicklung der Homerischen Poesie [Berlin 1882] 129) thought it free invention. Bethe was unimpressed by the first two passages, dismissing them as “Prahlereien” (1891:175). More reasonably, Friedländer (1914:320–321 = 1969:36) notes how consistent are the premises of their story of Tydeus with the more general context of the Theban War both as revealed in the other Iliadic references and as more explicitly set out by later sources: Athena’s support is common to both accounts, for instance, and to the third version recounted in Iliad X (cf. Iliad V 115–116). Tydeus’ transference from Aetolia to Argos (Iliad XIV 119) explains his presence on the Argive side in the war against Thebes (see further page 121 below) and Iliad IV’s παραίσια σήματα recur in Pindar {34|35} Nemean IX 19–20 and Euripides Suppliant Women 155–160, for example. Andersen (1976:36) also observes that the very phrase παραίσια σήματα is a ἅπαξ in Homer, which may perhaps imply an epic source. Friedländer concludes that “ein altes Epos genau so erzählt hat,” without, of course, necessarily equating that epic with the Thebais (see page 33 above). An epic source is also presupposed by, for instance, Aly (RE 7^A [1948]: 1706.2–3) and Leaf (on Iliad IV 384).

Such a conclusion still allows considerable scope and freedom for Homer’s own innovating hand. Take, for instance, the names of the two leaders of the Theban ambush: Μαίων Aἱμονίδης … | υἱός τ’ Aὐτοφόνοιο, μενεπτόλεμος Πολυ-φόντης (Iliad IV 394–395). Robert (1915:1.192) and Willcock (145 = “Mythological Paradigms in the Iliad,” in Oxford Readings in Homer’s Iliad [Oxford 2002] 441) think (to quote the latter) that “Maeon son of Haemon has a Theban-sounding father” (Haemon is the son of Creon in F1 of the Oedipodeia: see page 18 above) “and may be authentic; noticeably he is the only survivor.” The suspiciously murderous-sounding Polyphontes and his father Autophonus have, on the contrary, often passed as Homeric inventions (for invented names in Homer cf. Willcock 144–145 = “Mythological Paradigms,” 440; L. P. Rank, Etymologiseering en verwante verschijn-selen bij Homerus [Assen 1951] 130–135; cf. E. Risch, Eumusia [Howald Festschrift (Zurich 1947)] 72–91 = Kleine Schriften 294–313; H. von Kamptz, Homerische Personennamen [Göttingen 1982] 25–28, on “redende Namen” in early epic). But if Homer {35|36} could invent appropriate names in this manner, so perhaps could other, earlier, epic poets. They could conceivably be the source for the present passage. Tydeus’ sparing of Maeon is explained by the bafflingly elliptical phrase θεῶν τεράεσσι πιθήσας (line 398), the type of abbreviated reference which is often taken to represent compression of a preexisting narrative (cf. Iliad VI 183 for precisely the same phrase, and G. S. Kirk, The Songs of Homer [Cambridge 1962] 165). Andersen, too, is prepared to take seriously the idea that Maeon’s sparing represents an earlier tradition (1978:44n11). Here also, of course, there is room for dispute, since Niese (Die Entwicklung der Homerischen Poesie [Berlin 1882] 128), who believes the whole scene to be based on Bellerophon’s adventures in Iliad VI 187–211, takes the phrase in Iliad IV to be derived from that in Iliad VI. And the later details as to Maeon which we find in other authors may be spun out of Homer rather than stretching back to the Thebais (Andersen 1978:38).

The striking detail of a Mycenae ready to act as ally in the war against Thebes but deterred by signs from Zeus (see pages 33–35 above) is often interpreted as having been introduced to explain why that great city and its Pelopid rulers were conspicuous by their absence from the roll call of cities participating in the Theban War. If this is so, who first perceived the need of such an explanation? The composer of the Iliad, or the author of the epic which some suppose to underlie this and similar passages? Robert (1915:1.191: cf. Heldensage 3.1.932) followed by Andersen (1978:35) says the Pelopids can have played no part in the Thebais. We know too little of the poem to make such a generalization with utter confidence. But the surviving traces of the tradition, as Andersen stresses, do bequeath a picture of Adrastus and Amphiaraus as the leaders of all Argos and Achaea: the Pelopids need not have featured at all. Andersen again backs up Robert, who noted (1915:1.35) how the very fruitlessness of the visit to Mycenae (together with the superfluity of Polyneices in the Homeric context) might be thought to create a “Präjudiz für Erfindung.” The motive for such an invention is well conveyed by Andersen: it is once more paradigmatic. As Agamemnon’s forebears were willing to help Tydeus, so that hero’s son should now help Agamemnon.

Like Andersen (1978:44n9; cf. 45n20), I cannot accept Robert’s view of Iliad IV 365–400 as a “stümperhaftes Autoschediasma” (1915:1.191) deriving from Iliad V 793–813. For instance, modifying his earlier remarks in Studien zur Ilias (Berlin 1901) 185, he rather perversely tries to discover minor inconsistencies between the two sections that will confirm such a relationship (1915:1.188–189). But set these two passages side by side:

and

Is it really natural to take the former as a misunderstanding of the latter, the second passage picturing a Tydeus invited to a feast and proudly challenging his hosts to a contest, and the first transforming this into a chance stumbling upon the feasting Thebans and a blatant provocation? Against Robert’s view see further Andersen 1978:44n9 (who rightly concludes that Robert “hier … legt viel zu viel in den Text hinein”) and 79–82.

The Doloneia, of course, occupies a special position, though even if it did not I doubt whether I should be impressed by Robert’s claims (1915:1.194) that X 289’s μάλα μέρμερα μήσατο ἔργα rings odd of Tydeus’ self-defense against an ambush, or that we must be disturbed by the specifying of Tydeus’ ἀγγελία as a μειλίχιος μῦθος (implying a special negotiation?) and the absence of any reference to Tydeus’ challenge. Andersen (1978:130) very sensitively explains the reason for these and other apparent divergences. We here find a concentration upon the μέρμερα ἔργα perpetrated by Tydeus on his way back from the embassy to Thebes because these and these alone are relevant to the situation in the Doloneia (where Diomedes is hardly likely to penetrate the capital of the enemy forces!). The same explanation applies to the new detail of Athena as helper to Tydeus in this encounter too (an ad hoc invention, thinks Andersen, designed to bring the situations of father and son into the closest possible similarity).

Other discrepancies between what the Thebais and what the Iliad have to say about the career of Tydeus allow of an easy explanation along lines that are by now very familiar (see page 14 above): the wish to avoid grim and grisly stories and (a point particularly stressed by Andersen [1978:17 and 141]), the need to preserve Tydeus as a suitable paradigm for his son. Both considerations will make clear at once why the cannibalistic propensities revealed in Thebais F5 (see page 140 for text) get no mention in the Iliad. Likewise this poem says nothing of the tradition that Tydeus killed one of his uncles or the son of one {37|38} of his uncles (on which see page 121 below), a tale we know to have appeared in the Alcmaeonis and may guess to have featured in the Thebais. The Iliad does indeed name the uncles (XIV 115–118) and Σ AB Iliad XIV 120 notes that the verb πλαγχθείς may be a covert allusion to the exile that resulted from this killing. Again, the Iliad has nothing to say of the role of the Delphic oracle and its prophecy about the boar and the lion (see page 63 below). But even if we were not aware that Homer generally shies away from excessive dependence on the oracular and the prophetic (see Griffin 1977:48 = 2001:383) we might observe (with Robert himself 1915:1.196) that Tydeus’ story is after all told allusively, and that the Delphic origin of this oracle is by no means guaranteed (see page 64 below).

The basic presuppositions of the Iliad and the Thebais, then, are similar. As to their relationship, our initial antithesis between Welcker’s derivation of Iliadic details from Thebais and Niese’s free invention may not, after all, be so absolute. As Andersen puts it (1978:36–37), the picture of Tydeus’ single-handed expedition, if not a total invention, has at least been reshaped to give that hero the prominence required by the paradigmatic context.

Mecisteus

Friedländer rightly observes (1914:318–320 = 1969:34–37) that here, as with the Iliadic references to Tydeus, a few lines imply and conjure up a rich hinterland of mythical presuppositions which are fully consistent with the traditions of the Theban War as we recover them from later writers. As one of the seven Argive chieftains and a hero who fell before Thebes (see page 70 below), Mecisteus can only have participated in the funeral games of Oedipus if they were celebrated prior to the outbreak of hostilities. In other words, Oedipus’ death is here conceived as occurring in Thebes and before the expedition of the Seven against that city, precisely the same conception that is entertained in Sophocles’ Antigone and several other later works of literature. And there seems to be at least one further parallel for the friendly relations envisaged as existing between Thebans and Argives at this stage: Σ Iliad XXIII 679 (5.472 Erbse) saw the relevance of the version whereby Ἡσίοδός (fr. 192 MW) φησιν ἐν Θήβαις αὐτοῦ [scil. Οἰδίποδος] ἀποθανόντος Ἀργείαν τὴν Ἀδράστου σὺν ἄλλοις ἐλθεῖν ἐπὶ τὴν κηδείαν τοῦ Οἰδίποδος. One would naturally suppose that {38|39} the marriage between Argeia and Polyneices (on which see page 63 below) is to be connected in some way with this visit, though precisely how one need not venture to speculate. Mimnermus fr. 21 W may also belong here: Μ. δέ φησι τὴν μὲν Ἰσμήνην προσομιλοῦσαν Περικλυμένωι (see page 96 below) ὑπὸ Tυδέως κατὰ Ἀθηνᾶς ἐγκέλευσιν τελευτῆσαι. As Friedländer observes (1914:319 = 1969:35n48), these events seem unlikely in wartime, and if the exiled Tydeus had proceeded first to Thebes and only afterwards to Argos, he could have parti-cipated in Oedipus’ funeral games, and killed Ismene before encountering Polyneices at the gates of Adrastus’ palace.

Conclusion

Homer’s numerous references to the Theban War do indeed presuppose a tradition very similar to what we would independently guess to have stood in the Thebais. The Theban and Trojan wars “dominated epic tradition” (West on Hesiod Works and Days 162), and it is almost unthinkable that the composer of the Iliadic passages considered above was ignorant of some poetic work on the earlier war. The relationship of this work to the cyclic Thebais must remain obscure, but it cannot have been very different in content. On the other hand, Homer’s tendency to invent mythical details for his own purposes must not be underestimated, and several features which earlier scholars derived from the Thebais may rather be explained, with Andersen, as ad hoc creations, or at the very least careful adaptations to fit the new context.

The relationship between the Iliad and the Thebais is very much a special one. Other works, of course, have been thought to reflect the now-lost epic. It will be convenient to examine below under the relevant headings such writers as mention the Seven against Thebes, the striking of Capaneus by Zeus’ thunderbolt, and so on. In what follows I merely list, with a few appropriate comments, some studies not covered on pages 2–3 above in connection with the Oedipodeia, involving those authors that are most frequently supposed to have drawn upon the Thebais. On the origins of the story as a whole see Ernst Howald 1939—hard to get hold of, but stimulating. Also Dirlmeier 1954:151–158 = 1970:48–54, which, by accumulating potentially relevant material from the ancient Near East, interestingly anticipates the thesis of Burkert (1981:29–48 = 2001:150–165) on the eastern origins of the story. On Pindar’s indebtedness to the Thebais there is a useful article by Stoneman (1981:44–63), whose main fault is an occasional {39|40} uncritical acceptance of some of the reconstructions produced by Bethe, Friedländer, and the rest. For further bibliography see Kühr 2006. Finally, we should bear in mind that Statius, the one poet to have composed an epic Thebaid that is still extant, “die Thebais notorisch nicht gelesen hat” (Robert 1915:1.228–229; cf. 1.172, 202, etc.). See also R. Helm, RE 18.3 (1949): 996.49–55; D. Vessey, Statius and the Thebaid (Cambridge 1973) 60, etc.

Robert (1915:1.181–182) claimed that the artistic evidence afforded little help in reconstructing our lost epic, and that continues to be true, by and large, in spite of the accretions to our knowledge since he wrote. Artifacts will be mentioned as and when they seem likely to be relevant.

On the correct quantity of the middle vowel (Θηβᾰις) see Housman, “Notes on the Thebais of Statius (Continued),” Classical Quarterly 27 (1933): 72–73 = Classical Papers 3.1221–1222. The adjective κυκλικός is appended to the title, presumably to distinguish it from Antimachus’ epic, by the quoters of F2, F3, and F7 (see ad locc. for the ultimate sources of these passages).

The Thebais, like the Epigoni, is said to have contained seven thousand lines, a fact which excited Roscher (Die Sieben- und Neunzahl im Kultus und Mythos der Griechen [Leipzig 1904) 47–48) to the conclusion that each epic was divided into seven books of a thousand lines each, “ein deutlicher Beweis, wie weit in diesem Falle die Zahlensymbolik gegangen ist.” The allusion is to the seven gates of Thebes, but even if we are as impressed by the coincidence as Roscher himself was, we will only have obtained a small insight into the perverse mentality of some anonymous scholar. On book-division in general see S. West, The Ptolemaic Papyri of Homer (Cologne 1967) 18–24; S. West and M. L. West, “Comment,” Symbolae Osloenses 74 (1999): 68–73 = M. L. West, Hellenica 1.182–187.

In fact we have no evidence at all as to the number of books into which the Thebais was divided.

Scholars once assumed that the legend of the Seven against Thebes originated in the Mycenaean age of Greece (discussion and bibliography in Dirlmeier 1954:154 = 1970:53; cf. J. T. Hooker, Studies in Honour of T. B. L. Webster [Bristol 1988] 2.61 = Scripta Minora p. 279). More recently, Burkert has ingeniously and persuasively argued that “the tale of the ‘Seven against Thebes’ is the epic transposition of a purification ritual of ultimately Babylonian origin” (1981:42 = 2001:160). He notes several potential parallels between the Greek story and an Assyrian magical text involving “Seven Demons with formidable wings,” or rather figurines thereof, which are opposed by figurines of seven protective deities, and also “twins fighting each other in the gate” (1981:41–42 = 2001:159). But examination of the question as to when such Oriental influence can have made its influence felt in Greece leads Burkert to a conclusion he himself finds disturbing: “If any connection between the Babylonian and the Theban ‘Seven’ is accepted, the tale cannot have been created in Greece before 750 B.C.” (1981:44–45 = 2001:161).

Both sets of findings are at odds with the stress on “late” linguistic features in those fragments initiated by Wilamowitz and Wackernagel, and taken over by scholars like Bethe and Rzach or (more recently) Kirk and Griffin (“Die erhaltenen Verse sehen nicht danach aus, als hätte sie Kallinos gelesen” [Wilamowitz 1914:104]). They also raise important questions of principle, especially regarding the relationship between formulaic style and oral composition. Notopoulos’ simplistic assumption that the former is in itself sufficient guarantee of the latter receives specific refutation in Kirk’s important study of “Formular Language and Oral Quality” (Yale Classical Studies 20 [1966]: 153–174, esp. 169–174 = Homer and the Oral Tradition [Cambridge 1976] 183–200, esp. 195–200). Several studies have called into doubt the once-popular assumption that formulaic composition automatically entails orality: for a useful summary of their conclusions see Lloyd-Jones, “Remarks on the Homeric Question” (in History and Imagination [Trevor-Roper Festschrift (1981)] 7–10 = Academic Papers [I] 18–21; to whose bibliography add M. L. West’s “Is the ‘Works and Days’ an Oral Poem?” [in I poemi epici rapsodici non omerici e la tradizione orale (Padua 1981) 53–67 = Hellenica 1.146–158]), with its stress on the possibility that oral and literary modes of composition need not represent absolute alternatives. See also R. Janko, Homer, Hesiod, and the Hymns (Cambridge 1982) general index s.v. “formula, definition of ’’ and “orality, criteria for”; West, in Der Übergang von die Mündlichkeit zur Literatur (Tübingen 1990) 33–50 = Hellenica 1.159–175. {42|43}

Doubtless Burkert is right to draw the distinction he does (37 = 156) between the formulaic character of the Thebais’s fragments and, on the one hand, the allusive and playful adaptations of Homeric phraseology practiced by Panyassis and, on the other hand, the Meropis’s totally un-Homeric style. But no one has ever suggested that the Thebais was as late a composition as Panyassis’ Heracleia. It does not follow that it was contemporary with the Iliad or the Odyssey. On the question of the relationship between the Iliad and the Thebais (or an earlier form thereof) see pages 32–40 above. On the more general issue of relative dating of early Greek epic see Relative Chronology in Early Greek Epic Poetry, ed. Andersen and Haug (Cambridge 2012), esp. the final chapter by West (pp. 224–241).

Ἄργος: for “the very first word” of a poem as indicating “the singer’s subject” see West on Hesiod Theogony 1. Within the sphere of epic as narrowly defined we think at once of μῆνιν ἄειδε θεά (Iliad I 1), ἄνδρα μοι ἔννεπε Mοῦσα (Odyssey i 1), Ἴλιον ἄειδω (Ilias Parva fr. 1.1). See further B. A. van Groningen, The Proems of the Iliad and the Odyssey (Mededeelingen der Koninklijke Nederlandsche Akademie van Wettenschappen, afdeeling Letterkunde 9 [1946]) 6–7, W. H. Race, “How Greek Poems Begin,” Yale Classical Studies 29 (1992): 20. Willcock on Iliad I 1 observes “the first word … shows that the plot of the Iliad is to be primarily psychological, and that at any rate we do not have here a simple chronicle of the fighting at Troy,” and compares the first word of Odyssey i 1. By contrast, the cyclic epics, as Aristotle and Horace saw (cf. Brink on Horace Ars Poetica 143–144) give what Griffin (Homer on Life and Death [Oxford 1980] 1; cf. “Critical Appreciations VI: Homer, Iliad 1.1–52,” Greece and Rome 29 [1982]: 129 = Homer [Oxford 1998] 69) calls the “straightforward narrative of an obviously significant event—the war of the gods and the Titans, the whole Theban War, the capture of Troy.” But given this, it is odd that an epic called the Thebais should begin with a reference to Argos: see pages 45–46 below. ἄειδε: in early epic the verb is equally applicable to the activity of the Muse and that of the poet inspired by the Muse: see W. Kranz, “Sphragis: Ichform und Namensiegel als Eingangs- und Schlussmotiv antiker Dichtung,” Rheinisches Museum 104 (1961): 6 = Studien zur antike Literatur und ihrem Fortwirken (Heidelberg 1967) 29n5.

ἄειδε θεά: of course the same pair of words occurs in the same metrical sedes in the first line of the Iliad. L. E. Rossi (“Estensione e valori del colon nell’esametro omerico,” Studi urbinati di storia, filosofia e letteratura 39 [1965]: 250n33 and “La fine alessandrina dell’Odissea e lo ζῆλος Ὁμηρικός di Apollonio Rodio,” Rivista di filologia e di istruzione classica 96 [1968]: 160) has interpreted this as a direct allusion to that poem by the composer of the Thebais. Kranz (6–7 = 29–30) had already expressed a more cautious attitude, preferring to think in terms of a general stylistic feature of ancient epic as opposed to the obvious imitation of Homer with which we are presented by Orphic fr. 48 Kern: μῆνιν ἄειδε θεὰ Δημήτερος ἀγλαοκάρπου. See further J. Redfield, “The Proem of the Iliad: Homer’s Art,” Classical Philology 74 (1979): 98–99 = Oxford Readings in Homer’s Iliad 460. {43|44}

θεά: for this way of referring to the Muse at the start of a poem see Davies and Fin-glass on Stesichorus fr. 90.9.

ἔνθεν: for “the expansion by means of a relative clause of the subject of song initially named” as “a regular feature of epic proems” see West on Hesiod Theogony 2. Compare in particular Iliad I 2 (μῆνιν) … | … ἣ μυρί’ Ἀχαιοῖς ἄλγε’ ἔθηκε; Odyssey i 1 (ἄνδρα) … ὃς μάλα πολλά κτλ.; Ilias Parva fr. 1.2 (Δαρδανίην) … | ἧς περί. For Latin examples see Coleman on Statius Silvae 4.22.

1 – 2. ἄνακτες ‖: even in the absence of the next line we can see that there was enjambement between it and the first verse of the poem, and this feature is the third of Rossi’s reasons (for the other two see above on ἄειδε θεά and πολυδίψιον) for supposing that the very opening of the Thebais specifically imitated the openings of the Iliad and the Odyssey. The latter’s initial verse and its sequel are indeed enjambed in an equally striking fashion: ὃς μάλα πολλά ‖ πλάγχθη. But even here I prefer to talk in terms of a general stylistic feature common in early epic rather than specific copying. A line from a later poem such as Apollonius of Rhodes IV 2–3 ἦ γὰρ ἔμοι γε ‖ yields more readily to an interpretation as allusive imitation of the feature. Our instance and the Odyssey’s provide a case of “necessary enjambement”: see Milman Parry, “The Distinctive Character of Enjambement in Homeric Verse,” Transactions of the American Philological Association 60 (1929): 200–220 = The Making of Homeric Verse (Oxford 1971) 251–265; G. S. Kirk, “Have We Homer’s Iliad?,” Yale Classical Studies 20 (1966): 105–152 = Homer and the Oral Tradition (Cambridge 1976) 146–182; G. P. Edwards, The Language of Hesiod in Its Traditional Context (Cambridge 1971) 85–100; Richardson, Homeric Hymn to Demeter pp. 331–338; Janko, Homer, Hesiod, and the Hymns (Cambridge 1982) general index s.v., etc. For Argos as the point of departure for another great expedition celebrated in epic cf. Euripides Electra 1–3 (ὦ γῆς παλαιὸν ἄρδμος [Herwerden, Haslam: ἄργος codd.] Ἰνάχου ῥοαί, | ὅθεν ποτ’ ἄρας ναυσὶ χιλίαις Ἄρη | ἐς γῆν ἔπλευσε Tρωιάδ’ Ἀγαμἑμνων ἄναξ. There, of course, the situation is slightly different, since the farmer is apostrophizing the locale in which the play is set.

Howald (1939:7) convincingly argues that the stories of Oedipus and of the Seven against Thebes were originally independent entities: each is too complex and elaborate to be prologue or sequel to the other, and each has a quite distinct character. The two have been artificially united by the device of Oedipus’ curse on his sons, and this expedient may well have been the invention of the Thebais. On curses in early Greek literature see Watson 1991, esp. 12–18. On ancestral curses in particular see West in Sophocles Revisited (Lloyd-Jones Festschrift 1999) 31–45 = Hellenica 2.287–301; R. Gagné, “The Sins of the Fathers … ,” Kernos 24 (2008): 109–124; and N. Sewell-Rutter, Guilt by Descent (Oxford 2008). On Oedipus’ curses in general W. Bühler, Zenobii Athoi Proverbia 5 (Göttingen 1999) 452–458.

In some respects the two fragments are very similar: in both, Oedipus becomes angry; in both the result of that anger is that he curses his two sons. But whereas in F3 the action that angers him is perpetrated by both sons so that {46|47} the joint curse is instantly explicable, in F2 only Polyneices seems responsible; nevertheless, Eteocles too falls under his father’s curse. Furthermore, the action that evokes the curse in F2 is on the face of it designed to honor Oedipus, so that Oedipus’ response seems at first paradoxical. And the contents of the curse are different, though closely connected: in F2, despite the abrupt termination of the extract, war between the two brothers is clearly prophesied; F3 more specifically mentions their deaths at each others’ hands. This climax in clarity and grimness (cf. my note on Sophocles Trachiniae [Oxford 1991] lines 43–48) suggests that in the original epic the two episodes did indeed stand in this order (cf. Welcker 1865:2.334–335). For the principle that a similar event’s happening twice constitutes decisive proof of an underlying tendency see C. W. Müller, “Der ‘zweite Beweis’ als Wahrheitskriterium,” Hermes 127 (1999): 493–495 = Kleine Schriften 182–185, esp. Erbse’s citation ap. n12 of the present two fragments. Both filial misdemeanors concern τροφή, which is the reason given for anger and curse at Aeschylus Seven Against Thebes 786 and Sophocles Oedipus at Colonus 1265–1266, 1362–1369.

Let us now confine ourselves for the moment to F2. We shall try to obtain a clearer picture of events: first and foremost, why is Oedipus so angry? Polyneices sets before him a fine silver table that had belonged to Cadmus, and also (presumably on it) a fine golden goblet full of wine. The possessions are then described as the precious γέρα of Oedipus’ own father, Laius, so one presumes that the goblet too once belonged to Cadmus and that it and the table were handed down within the family from father to son. Athenaeus adds to this picture by claiming that Oedipus had previously forbidden the goblet to be brought before him. Whether {47|48} he derives this information from elsewhere in the epic or is merely making an obvious inference, one would suppose the ban to apply to the table also. The only conceivable motive for Oedipus’ extreme vexation at the presence before him of these family heirlooms and kingly symbols must be that devised by Welcker (1865:2.334) in the wake of Eustathius and accepted by Bethe (1914:102–103), Robert (1915:1.175), Rzach (1922:2364.21–22), and practically all scholars: Oedipus does not want to be reminded by these objects of the father he had unwittingly killed and supplanted. Perhaps, too, as Robert added, he does not wish to be reminded by these royal tokens of his former prosperity and happiness. Erika Simon objected (1981:10 and n13) that the tokens had earlier belonged to Labdacus and Cadmus, and advanced the novel hypothesis that Oedipus was vexed because in setting before him the utensils used in the hero-cult of Cadmus, his sons were treating him as if he were already dead. But our fragment says nothing of Cadmus’ hero-cult or the practice of “Totenmahl” (on which see page 119 below). It does, however, stress (lines 5–6) that the objects belonged to Laius. Perhaps this is another un-Homeric feature: the dining table is not the symbol of social harmony (see my remarks in “Feasting and Food in Homer: Realism and Stylisation,” Prometheus 23 [1997]: 97–107), but a source of discord.

Finally, here are a few minor comments on the context of the present quotation in Athenaeus. Kaibel wished to delete the words δι᾽ ἐκπώματα. The plurality need not disturb us (cf. A. C. Moorhouse, The Syntax of Sophocles [Mnemosyne suppl. 75 (1982)] 4–5 on plurals for instruments and tools) but the phrase is unnecessary and, worse, inadequate as a motivation for the curse, since (as we have seen [page 47 above]), the silver table plays its part too in angering Oedipus. However, as Robert pointed out (1915:2.66n28), Athenaeus is citing the whole passage for this one detail of the cup, and the double mention is appropriately emphatic. Athenaeus also seems to err in stating that both sons placed the cup before Oedipus (παρέθηκαν) but the mistake is venial, especially when it occurs in a sentence that began with the perfectly accurate statement that the incident led to the cursing of both sons. Eustathius has taken over from Athenaeus both of these small errors, and also the failure to refer to the silver table (irrelevant, as we have just seen, to the context in which Athenaeus cites the epic lines).

1. αὐτὰρ ὁ διογενὴς ἥρως: cf. Iliad XXI 17 αὐτὰρ ὁ διογενὴς δόρυ μὲν λίπεν κτλ.; Odyssey xxiii 306 αὐτὰρ ὁ διογενὴς Ὀδυσεύς κτλ.; Iliad V 308 αὐτὰρ ὅ γ’ ἥρως |, IV 489 | Aἴας διογενής. In a list of adjectives ending in -ης which can also operate as proper names, Σ A Iliad XVI 57 (4.173 Erbse) happens to juxtapose αὐτὰρ ὁ διογενής from the above passages with Πολυνείκης, but (pace e.g. Welcker; Schneidewin, “Zu den bruchstücken der homerischen dichter,” Philologus 4 [1849]: 747; Nauck, Mélanges gréco-romains, 374; Allen, Oxford Text of Homer vol. 5, p. 113) this is fortuitous and has nothing to do with our line. ξανθὸς Πολυνείκης | : cf. Iliad III 284 = XVII 18 ξανθὸς Μενέλαος |; Hesiod Theogony 947 ξανθὴν Ἀριάδνην |. For other instances of the epithet in this position within the hexameter see W. D. Meier, Die epische Formel im pseudohesiodeischen Frauenkatalog (diss. Zurich 1976) 157.

2. | πρῶτα µέν: cf. Odyssey xxii 448 | πρῶτα μὲν οὖν; xxiii 131 | πρῶτα μὲν ἄρ; Iliad VI 179 πρῶτον μὲν ῥα. Οἰδιπόδηι: for the various forms of this name in epic and tragedy see A. Sideras, Aeschylus Homericus (Hypomnemata 51 [1971]) 101. καλὴν παρέθηκε τράπεζαν: cf. Odyssey v 92 θεὰ παρέθηκε τράπεζαν |; i 138–139 ξεστὴν ἐτάνυσσε τράπεζαν | … ταμίη παρέθηκε φέρουσα.

2 – 3. παρέθηκε τράπεζαν | ἀργυρέην: for the enjambement cf. Odyssey i 441–442 θύρην δ’ ἐπέρυσσε κορώνηι | ἀργυρέηι; xv 103–104 υἱὸν δὲ κρητῆρα φέρειν Μεγαπένθε’ ἄνωγεν | ἀργυρέον. It is of the type that Milman Parry (as page 45 above) termed “unperiodic enjambement … the addition of an adjectival idea … describing a noun found in the foregoing verse” (“The Distinctive Character of Enjambement in Homeric Verse,” 206 = The Making of Homeric Verse 255), and Kirk (“Have We Homer’s Iliad?” [as on page 45 above] 107 = Homer and the Oral Tradition 149) “progressive enjambement.”

3. Kάδµοιο θεόφρονος: see Kirk, “Formular Language and Oral Quality” [as on page 42 above], 169 = Homer and the Oral Tradition 195 for the epithet (“a compound unique in the epic tradition”) as “a clear departure from the thrift of the oral epic. The standard laudatory epithet for this position in the verse is δαΐφρονος (28 times in Homer).” θεόφρων again only in Pindar Olympian VI 41. αὐτάρ ἔπειτα |: the notorious cyclic formula: cf. Pollianus Palatine Anthology XI 130.1–2 τοὺς κυκλί<κ>ους {49|50} τούτους, τοὺς ‘αὐτὰρ ἔπειτα’ λεγόντας | μισῶ, λωποδύτας ἀλλοτρίων ἐπέων (on which see Cameron, Callimachus and His Critics [Princeton 1995] 396–398; Griffin 1977:49; Campbell on Quintus Smyrnaeus XII 139). The phrase at line-end is not unknown to Homer (see Campbell) but in the present case one is reminded of the naϊve repetition of ἔπειτα in our earliest examples of Greek prose: cf. Ed. Fraenkel, “Additional Note on the Prose of Ennius,” Eranos 49 (1951): 50–56 = Kleine Beiträge zur klassischen Philologie 2.53–58, Dover in Classical Contributions (McGregor Festschrift 1981) 24–25 and n47 = Greek and the Greeks 29 and n46.

4. δέπας ἡδεὸς οἴνου |: the same phrase ends a line at Odyssey iii 51. The double absence of digamma (ϝηδεός ϝοίνου) is striking in both passages and a probable index of “lateness” (Wilamowitz 1884:366n45; Bethe 1891:40n20). On the line-end phrase μελιηδέος οἴνου see Chantraine, Gramm. hom. 1.123; on other Homeric instances of ἡδύς and οἶνος sans ϝ see 151 and 145 respectively. Most are easily removed. For the adjective’s application to wine see Arnott on Alexis fr. 46.9 KA.

5. αὐτὰρ ὅ γ’ ὡς: the same collocation of words at the start of a line in Iliad XII 40, XXI 550; cf. Iliad V 308, cited on line 1. On the stylistic implications of this third instance see Griffin 1977:49. φράσθη: the meaning is that given by LSJ s.v. II 4 (“perceive, observe”). Compare ἐνόησε in fr. 3.1 (Oedipus is again the subject). In other words, φράσατο would supply the same sense. For the form cf. (ἐπ)ἐφράσθης in Odyssey v 183, xix 485, xxiii 260. On its relatively recent development see Chantraine (as cited on line 4), 1.405–406 πατρὸς ἑοῖο|: same phrase, same position in Iliad XIV 11, XXIII 360; Hesiod Theogony 472.

6. τιµήεντα γέρα: cf. Odyssey i 312 (|τιμῆεν), xiii 129 (| τιμήεις). µέγα οἱ κακὸν ἕµπεσε θυµῶι: for a full and excellent analysis of the oddity of this phrase see Kirk, “Formular Language and Oral Quality” (as cited on page 42 above) 169–171 = Homer and the Oral Tradition 195–197. The poet seems to have conflated “two distinct formular applications of ἔμπεσε: an emotion ‘falls upon’ the spirit, an evil ‘falls upon’ a house.” But in the present case, muddle-headedly, “an evil” (κακόν rather than, e.g., ἄχος) falls upon Oedipus’ spirit. Did the poet mean κακόν to be equivalent to ἄτη? For ἔμπεσε θυμῶι | of emotion see Iliad IX 436, XIV 207, 306 (ἐπεὶ χόλος ἐ. θ.), XVI 206 (κακὸς χόλος ἐ. θ.), XVII 625 (δέος ἐ. θ.). For ⌣–κακὸν ἔμπεσεν οἴκωι | cf. Odyssey ii 45, xv 375. µέγα οἱ: ϝ is hardly ever neglected before third-person sing. oἱ: see West, Hesiod’s Theogony p. 100, his note on Hesiod Works and Days 526 (p. 291), Edwards, Language of Hesiod, 138n48.

7. |αἴψα δέ: these two words begin a line in Iliad II 664, Odyssey xvi 359. µεταµ-φoτέρoiσiν: It is difficult to know what to make of this. If we are supposed (i) to detect here an example of the verb μετ’ … ἠρᾶτο in tmesis, we will look in vain for an entry s.v. μεταράομαι in our lexica. And the search s.v. ἀράομαι for the construction ἀρὰς ἀ. μετὰ τινι in place of the normal ἀρὰς ἀ. τινι will be equally futile. But emendations do not convince: ἑοῖσιν ἐπ’ ἀμφοτέροισιν coni. H. van Herwerden, “Notulae ad Athenaeum,” Mnemos yne 4 (1876): 313, prob. Nauck, Mélanges gréco–romains, 374–375, and seriously considered by Wackernagel (1916:181n2); κατ᾽ ἀμφοτέροισι coni. R. Peppmüller, “Zu den Fragmenten der griechischen Epiker,” Neue Jahrbücher für Philologie und Pädogogik 133 (1886): 465, comparing κατηράτο in Iliad IX 454. The two likeliest solutions for the problems raised by the paradosis are (ii) to associate it with the Homeric construction (Schwyzer, Gr. Gr. 2.483) whereby we find a verb with μετά + dative plural as prepositional object rather than with the simple dative that later {50|51} suffices. Especially enlightening are instances where tmesis would be ruled out by the resultant form (e.g. μεταπολεμίζω) or by other considerations (Iliad I 525–526 τοῦτο γὰρ ἐξ ἐμέθεν γε μετ’ ἀθανάτοισι μέγιστον| τέκμωρ). Alternatively, if not repelled by its remarkable equation of σύν with μετά, one may suppose that we have here (iii) the form μεταμφοτέροισι as an alternative to συναμφοτέροισι, though it must be admitted that (iii) is no less a stranger to LSJ than (i) and (ii). This third interpretation is adopted by scholars of the calibre of Wilamowitz (1884:366n45) and Wackernagel (1916:181n2), who take it as a further index of the relative recentness of the poem, the latter observing that (a) the equivalence between μετά and σύν thus implied is unknown to early epic, where the former means “amid” (see further Schwyzer, Gr. Gr. 2.481–482); (b) the word to which our form is an alternative, συναμφοτέροι, is itself not found until the fifth century (though cf. Theognis 820 συναμφοτέρους, which there seems no reason to date so late). M. Leumann, Homerischer Wörter (Basel 1950) 94n56 takes this explanation a stage further: starting with the tmesis in Homeric phrases like μετὰ Tρωιῆισιν ἔειπεν (Iliad XXII 476), μετὰ δ’ Ἀργείοισιν ἔειπεν (XXIII 781), Tρώεσσι μεθ’ ἱπποδάμοις ἀγορεύσω (VIII 525), ταῦτα μετ’ Ἀργείοις ἀγορεύεις (X 250), he suggests that, in a manner constantly presupposed by his book, such extensions of the tmesis as μετ’ ἀμφοτέροισιν ἔειπεν (Iliad III 85 = VII 66; cf. too the similar examples ἄριστα / ἔρισμα μετ’ ἀμφοτέροισι γένηται at Iliad III 110, IV 38, ἔργα μ. α. ἔθηκε at Iliad III 321, φιλότητα μ. α. βάλωμεν /τίθησι at Iliad IV 16/83, ἔριν Ἀτρεΐδηισι μ. α. ἔθηκε at Odyssey iii 136) have been misunderstood and μεταμφότεροι created by misinterpretation. ἐπαράς: Iliad IX 456.

7 – 8. ἐπαράς | ἀργαλέας ἦρατο: comparable enjambement in Odyssey xi 291 (289: βόας) …|… ἀργαλέας (cf. Hesiod fr. 37.2 MW), Iliad XI 3–4 ἔριδα … | ἀργαλέην. With ἐπαράς … ἦρατο cf. Iliad II 788 ἀγορὰς ἀγόρευον and in general Fehling, Die Wiederholungsfiguren und ihr Gebrauch bei den Griechen vor Gorgias (Berlin 1969) 156–157. For the adjective ἐπάρατος in real-life curses see Watson 1991:37.

9. οὔ ο†: μή is tentatively suggested by Hutchinson in his commentary on Aeschylus Seven Against Thebes (Oxford 1985) xxix, but for the digamma see above on line 6. For a list of early attempts to solve the crux see W. Ribbeck, “Zu den Fragmenten der griechischen Epiker,” Rh einisches Museum 33 (1878): 457. Best, perhaps, was Hermann’s πατρώϊ’ ἐνηείηι φιλότητος, which Ribbeck himself adapted to π. ἐνηέι <ἐν> φιλότητι (so too, independently, Peppmüller, “Zu den Fragmenten” [as in 7n above], 465, comparing for the hiatus in the bucolic caesura ἔγχεϊ ὀξυόεντι in Iliad V 50, etc.). Kaibel’s edition of Athenaeus placed Ribbeck’s conjecture in the actual text of the fragment. Robert, however (1915:2.67), argued that both lines of approach are vitiated by the fact that Homer only employs ἐνηής of an ἑταῖρος or person (contrast Hesiod Theogony 651 μνησάμενοι φιλότητος ἐνηέος, cited by Ribbeck). The same charge may be leveled against O. Rossbach’s ἐνηῆι φιλότητι (“Epica,” Neue Jahrbücher für Philologie und Pädogogik 143 [1891]: 82) and a similar one (as seen by Robert and Peppmüller) against Meineke’s ἐν ἠθείηι φιλότητι (Analecta Critica, 212), since Homer never bestows this (or any other) adjective upon this noun when it bears the nonsexual signification required here. However, Robert’s criteria so drastically exclude most corrections from consideration that a reminder is necessary of the numerous other un-Homeric features that this fragment and others of our epic contain. Perhaps Ribbeck’s suggestion is the least unsatisfactory after all. φιλότητι | in Odyssey viii 313, x 43 (in the last passage meaning “friendship”).

9 – 10. ( πατρώια …) | δάσσοντ ’: apart from πατρώιων χρημάτων δατήροι at Aeschylus Seven Against Thebes 711 (two lines after a reference to Οἰδίπου κατεύγματα), compare the terms of Oedipus’ curse at 788–789 of the same play: καί σφε σιδαρονόμωι | διὰ χερί ποτε λαχεῖν κτήματα (see too the references to κτέανα and κτήματα in 729 and 817 and cf. Lloyd-Jones, Classical Review 28 [1978]: 214), at Euripides Phoenician Women 67–68 (ἀράς) … | θηκτῶι σιδήρωι δῶμα διαλαχεῖν τόδε and at [Plato] Alcibiades. (2) 138^c ὥσπερ τὸν Οἰδίπουν αὐτίκα φασὶν εὔξασθαι χαλκῶι διελέσθαι τὰ πατρῶια τοὺς υἱεῖς. Hermann’s conjecture δάσσαιντ’ (made in the note on Oedipus at Colonus 1377 in his 1827 revision of Erfurdt’s commentary [2.435]) held the stage until Wackernagel {52|53} (1916:254–255) objected that Homer only uses ὡς plus optative in indirectly quoted prayers (Odyssey xvii 243, xxi 201), and proposed δάσσοντ’ on the basis of Odyssey v 23–24 = xxiv 479–480 (ἐβουλεύσας| … ὡς … ἀποτείσεται). The comparison is apt; only (unlike Wackernagel) we must take ἀποτείσεται and δάσσοντ’ plus ὡς as exemplifying “the transition from modal to final use” (Schwyzer, Gr. Gr. 2.665 on Odyssey v 23–24, with which he compares Iliad VIII 36–37 βουλὴν … ὑποθησόμεθ’ … |ὡς μὴ πάντες ὄλωνται). W. Headlam (“Emendations and Explanations,” Journal of Phil ology 30 [1907]: 307), citing Sophocles Oedipus Rex 1270–1274 (Oedipus’ curse on his sons) αὐδῶν τοιαῦθ’ ὁθοῦνεκ᾽ οὐκ ὄψοιντό νιν, conjectured δάσσοιντ’, approved by Pearson (Euripides’ Phoenissae, p. xxn2), but for future indicative rather than imprecatory optative in curses see Watson 1991:23–24 (cf. 40) and on Horace Epode V 89.

10. ἀεί: cf. Iliad XII 211 Ἔκτορ, ἀεὶ μέν πώς μοι ἐπιπλήσσεις ἀγορῆισιν, XXIII 648 ὥς μευ ἀεὶ μέμνησαι ἐνηέος, οὐδέ σε λήθω, Odyssey xv 379 οἷά τε θυμὸν ἀεὶ δμώεσσιν ἰαίνει. The short α in these passages is regarded as an Atticism by Wackernagel (1916:146), who rejects the notion of an East Ionic origin. Chantraine demurs (Gramm. hom. 1.167). But as Shipp, restating Wackernagel’s case, observes (Studies in the Language of Homer ² [Cambridge 1972] 49), “if Ionic it is late, as αἰεί persists into the inscriptions … and is usual in MSS of Herodotus.” Hermann’s ἔοι (suggested in the note on Oedipus at Colonus 1377 cited above on lines 9–10), resembles Schneidewin’s αἰεὶ δ’ ἀμφοτέροισιν ἔοι π. τ. μ. τ. ( “Zu den bruchstücken der homerischen dichter,” 747) and Köchly’s εἴη δ’ ἀμφοτέροισιν ἀεὶ π. τ. μ. τ. (Coniectaneorum Epicorum fasc. I [1851] p. 10 = Opusc ula Philol ogica 1.230) and other emendations in seeking to introduce an imprecatory optative that would be idiomatic in a curse: see e.g. the funerary inscription from Asia Minor cited by J. H. M.Strubbe in Faraone and Obbink, eds., Magika Hiera (Oxford 1991) 39: ἐξωλεία καὶ πανωλεία εἴη αὐτῶι πάντων. But a corresponding verb with πολεμοί τε μαχαί τε as its subject can well have stood in the verse that originally followed line 10 (since Athenaeus is usually a careful and conscientious quoter [cf. K. Zepernich, “Die Exzerpte des Athenaeus in den Dipnosophisten und ihre Glaubwürdigkeit,” Philologus 77 (1921): 324–363 (esp. 362–363)], the omission of the line will be a transmissional error). Without emending we still have an irrevocable prayer for hateful things which supplies a positive equivalent of e.g. μηδέποτε in curses (see Strubbe 56n106) used of benefits not to be enjoyed. For “always” in curses see, e.g., Tibullus I 5.51–52 hanc volitent animae circum sua fata querentes | semper, Propertius IV 5.39 semper habe morsus circa tua colla recentes, Genesis 3:14 and 17 (God’s curse on the serpent and Adam) “all the days of thy life.” πόλεµοί τε µάχαι τε |: same phrase at line-end in Iliad I 177, V 891, Hesiod Theogony 926. But there may be a special point to the phrase here. The curse from Asia Minor cited above provides a parallel for the idiomatic (and “strengthening”) juxtaposition of nearly synonymous evils prayed for in a curse.

A few introductory remarks on the text of the note that is our source for this fragment: the legion inadequacies of Papageorgiou’s edition of the Sophoclean scholia can from time to time be remedied by consulting V. de Marco’s work De Scholiis in Sophoclis Tragoedias Veteribus (Reale Accademia Nazionale dei Lincei 6 [Rome 1937]), which handily corrects and amplifies Papageorgiou’s information, {53|54} and nowhere to better effect than on p. 111, which deals with the scholion that is the source for our fragment. I have incorporated the Italian scholar’s findings in this fragment’s text and app. crit. They reappear in his full-scale edition of the scholia on the Oedipus at Colonus (Rome 1952). Nauck’s small but palmary corrections of the scholion’s comment on Oedipus’ anger were made in his review of Papageorgiou (Mélanges gréco-romains, 50).

οἱ περὶ Ἐτεοκλέα καὶ Πολυνείκην = Ἐτεοκλῆς καὶ Πολυνείκης: on the idiom see S. L. Radt, “Noch einmal Aischylos, Niobe Fr. 162N (278M),” Zeitschrift für Papyro-logie und Epigraphik 38 (1980): 47–56 and “οἱ (αἱ etc.) περὶ + acc. nominis proprii bei Strabon,” Zeitschrift für Papyrologie und Epigraphik 71 (1988): 35–40 = Kl eine Schriften 236–246 and 362–374. The criticism implied by the scholion’s use of the adverbs μικροψύχως and τελέως ἀγεννῶς was attributed to Didymus by Robert (1.170); cf. Pfeiffer, History of Classical Scholarship 1.276–277. As Griffin suggests (Homer on Life and Death 14), it may reflect the inability of a later Greek mentality to understand the importance attached to food as a symbol of honor in earlier literature. This brings us to the fragment itself.

Almost immediately after citing our present fragment, the Sophoclean scholion proceeds to quote a further fragment in the form of fifteen iambic trimeters (TrGF 2 F458) which appear to presuppose the same state of affairs. Controversy has long raged over the origin, authorship, and genre of this floscule of drama: see especially Robert 1915:2.67–69. Without becoming unnecessarily embroiled in this matter, we may safely make the following comments on this other fragment’s version of events. The speaker of the fifteen verses would seem to be either Eteocles or Polyneices, for he describes how he and at least one other had been accustomed to send to the blind Oedipus a portion of the sacrifice (lines 2–3 θυσίας {γὰρ} ἀπαρχὴν γέρας ἐπέμπομεν πατρί | περισσὸν ἀρνῶν ὦμον, ἔκκριτον γέρας [κρέας coni. Methner]). But on one occasion a lapse of memory led to their sending something else (lines 5–6: ἀντὶ τοῦ κεκομμένου | ἐπέμψαμεν βόειον) and the irate old man, interpreting the change as a deliberate insult intended to escape his attention, invokes a curse upon his sons that is remarkably similar to what we find at the close of the epic fragment (lines 14–15: χαλκῶι δὲ μαρμαίροντες ἀλλήλων χρόα | σφάζοιεν ἀμφὶ κτήμασι βασιλικοῖς).

Now even if this dramatic excerpt had suffered less corruption than it has and we felt far more confident as to its source and genre, there would still be danger in resorting to it automatically in order to supplement or clarify our own particular fragment. Let us start then by approaching the epic lines in isolation to see what they will yield us independently.

Our fragment’s mention of an ἰσχίον seems to imply (see ad loc.) a sacrifice as background to the insult (so e.g. Bethe 1891:102–103). Oedipus does not participate directly as a king normally would (cf. Aristotle Politics 1285^B10), perhaps {54|55} because his hands are polluted by his crime. Clearly his sons have, in practice, taken over the duties of kingship (see Wolff in Roscher 3.2663.39–49). Teiresias in Euripides Phoenician Women 875–876 describes how Oedipus’ two sons ἄνδρα δυστυχῆ | ἐξηγρίωσαν. In the Thebais was the insult intentional or deliberate? The actual fragment represents Oedipus as exclaiming παῖδες μέγ’ ὀνείδειον τόδ’ ἔπεμψαν, but the words of a proud and angry old man are not perhaps the most reliable testimony or the most objective. The quoter of the epic ascribes the offense to forgetfulness (ἐκλαθόμενοί ποτε) and this corresponds with the explanation given in line 4 of the iambic trimeters treating of the same subject (οὐ μεμνημένοι). Here, certainly, Oedipus’ complaint about filial malice seems at odds with the reality. The same picture is implied by Plato’s remarks on the malicious nature of Oedipus’ curse (Alcibiades [2] 138^C and 141^A, Laws 931^B). See too Aeschylus Seven Against Thebes 780–781 (ἐπ’ ἄλγει δυσφορῶν | μαινομέναι κραδίαι), Euripides Phoenician Women 66 and 877 (νοσῶν). The idea that an unintentional insult, one occasioned by oversight, is as deserving of punishment as a deliberate crime accords perfectly with archaic Greek morality: see Davies and Finglass on Stesichorus fr. 85.2.

| –⌣⌣ ὡς ἐνόησε: the same phrase in the same metrical position at Iliad XV 422, Odyssey x 375, etc. χαµαὶ βάλε εἶπέ τε µῦθον: cf. Iliad VII 190 (Ajax recognizes his κληρός and as a sign of his joy) [τὸν μὲν παρ πόδ’ ἑόν] χαμάδις βάλε φώνησέν τε. Similar phrasing, very dissimilar content. χαµαὶ βάλε: the same phrase in the same metrical position at Iliad XXI 51, Odyssey xvii 490, Homeric Hymn to Hermes 118 and 298; and (with βάλον for βάλε) Iliad V 588 and Odyssey xxii 188. But in these instances the phrase has a different meaning from the present occurrence. A closer parallel {56|57} for anger expressed by the flinging to ground of an object is Iliad I 245–246 ὣς φάτο Πηλεΐδης, ποτὶ δὲ σκῆπτρον βάλε γαίηι | χρυσείοις ἥλοισι πεπαρμένον, though that action has a symbolic dimension (see Griffin, Homer on Life and Death 11–12) lacking here. βάλε εἶπε: the evidence of so corrupt a fragment is hardly sufficient to allow us to decide whether the poet gave εἶπε a digamma or intended βάλεν. εἶπέ τε µῦθον: the same formula ends a hexameter at Iliad VII 277, XI 647, XVIII 391, XXIII 204, Odyssey viii 302, xiv 494, Homeric Hymn to Aphrodite 256, 286, Homeric Hymn to Hermes 154, 218, 306, HH 7.54 (cf. Odyssey v 338: εἶπέ τε μῦθον|: ἔειπε | U⁸ : πρὸς μῦθον ἔειπε | rell.). As here, it is directly followed by a speech in oratio recta in all these instances except Iliad VII 277 (where a line supplying the subject of the verb intervenes) and Odyssey viii 302 (where no speech follows). See further R Führer, Formproblem-Untersuchungen zu den Reden in der frühgr. Lyrik (Zetemata 44 [1967]) 17–19.

2 – 3: a large number of scholars have wished to posit a lacuna between these two verses (e.g. Ribbeck [“Zu den Fragmenten” (as in F2.9n), 457], who thoughtfully appends his version of the missing line). The reasons for agreeing with them may be grouped and listed as follows:

• Aa) The eccentric μέν solitarium in line 2
• b) The lack of any object for ἔπεμψαν in the same line
• c) The extremely abrupt nature of the asyndeton at the start of line 3
• Bd) The excessive brevity of Oedipus’ speech as transmitted
• e) The presence of a marginal sign opposite line 2

As regards the oddities collected under heading A, they are all removable by simple emendations (see ad loc. for details), the majority of which also recommend themselves on grounds quite independent of the presence or absence of a lacuna. Three emendations within two lines: this is not excessive for a quotation fragment, given the extreme susceptibility to corruption of such texts. Of B we may observe that we have no right to demand a Homeric plenitude from the speeches in later epic; on the contrary, Griffin (1977:49–50), who accepts the notion of a lacuna, nevertheless refers to the “dry manner of indirect reporting” here exhibited and “the indirect and summary manner” in which the curse is reported. Certainly, the presence of the word μῦθος in the introduction to Oedipus’ direct speech implies nothing about its length: the self-same formula εἶπέ τε μῦθον heralds a one-line speech at Iliad XVIII 391–392.

δὲ Δί: on the various cases of Zeus’ name see Schwyzer, Gr. Gr. 1.576–577. As W. Schulze observed (Quaestiones Epicae [Gütersloh 1892] 241n1), no adequate parallel for εὖκτο Διῖ βασιλῆι is provided by Iliad II 169 (Διὶ μμῆτιν), X 16 (Διὶ μμέγα), or II 781 (Διὶ ϝϝώς): see further P. Maas, Gr eek Metr e [Oxford 1962] §131. Schulze himself was reduced to considering the possibility of Διεῖ (cf. Quaestiones Epicae 239–241; Schwyzer as cited; Burkert 1981:36 = 2001:155, comparing the epithet διίφιλος) or attributing the oddity to the error of “imitatoris contra Homeri usum parum intellectum inviti peccantis.” De Marco’s discovery that εὖκτο δὲ Διί stands in R clinches the case for Buttmann’s palmary εὖκτο δὲ Δί (proposed in Gr. Gr. 2¹ [Berlin 1825] 405 = Gr. Gr. 1² [Berlin 1830] 225). The corruption of Δί to Διί can be paralleled time and again from Pindar’s manuscripts (e.g. Olympian XIII 106), and the omission of ΔE before ΔII in LM through haplography was practically inevitable. It is inconceivable that even the most incompetent of epic poets could ever have commenced a hexameter with εὖκτο Διί, thereby introducing at one and the same time an unbearably harsh asyndeton and an unprecedented lengthening of the final vowel. Δῑ: for the contracted form see (apart from the Pindaric examples indicated above) the two Etruscan helmets dedicated to Zeus by Hieron in commemoration of his victory over the Carthaginians at Cyme in 474 (cf. Meiggs–Lewis, A Selection of Greek Historical Inscriptions [Oxford 1969] p. 62). This is another un-Homeric feature. Δὶ βασιλῆι: the application of βασιλεύς to this or, indeed, any god is un-Homeric and another sign of lateness: see Fraenkel on Aeschylus Agamemnon 355, Richardson on Homeric Hymn to Demeter 358 for the history and frequency of this and similar designations. The earliest parallels are Hesiod Theogony 886 Ζεὺς δὲ θεῶν βασιλεύς (where, however, it possesses a strongly predicative sense, as West ad loc. observes), Works and Days 668 Zεὺς ἀθανάτων βασιλεύς, Theogony 923 μιχθεῖς’ ἐν φιλότητι θεῶν βασιλῆι καὶ ἀνδρῶν, fr. 308.1 MW αὐτὸς γὰρ πάντων βασιλεὺς καὶ κοίρανóς ἐστιν, Cypr ia F7.3 Ζηνὶ θεων βασιλῆι. {59|60} Compare Κρονίδαις βασιλεύς in Alcaeus (frr. 38^A9, 387). καὶ ἄλλοις ἀθανάτοισι|: the same phrase at line-end in Iliad II 49. For its use as a comprehensive prayer formula see Davies and Finglass on Stesichorus fr. 85.2. On the role of the gods in fulfilling curses see on F2.8 above.

4 . χερσὶν ὑπ’ ἀλλήλων: the motif of fraternal ἀλληλοφονία recurs at Stesichorus fr. 97.21 ὑπ’ ἀλλάλοισι δαμέντας; Pindar Olympian II 41–42 ἰδοῖσα δ’ ὀξεῖ’ Ἐρινύς | ἐπεφνέ οἱ σὺν ἀλλαλοφονίαι γένος ἀρήϊον; TrGF 2 F458.14–15 χαλκῶι δὲ μαρμαίροντες ἀλλήλων χρόα | σφάζοιεν. Cf. Oedipus’ remarks to Polyneices at Sophocles Oedipus at Colonus 1373–1374 (αἵματι | πεσῆι μιανθεὶς χὠ ξύναιμος ἐξ ἴσoυ) and at 1387–1388 (συγγενεῖ χερί | θανεῖν κτανεῖν θ’ ὑφ’ οὗπερ ἐξελήλασαι). | χερσὶν ὑπ’ ἀλλήλων: cf. | χερσὶν ὑπ’ Ἀργείων (Iliad XIII 763, XXIV 168). καταβήµεναι –⌣⌣– — |: cf. Iliad XII 65, Odyssey x 432. Aἴδος εἴσω |: cf. Iliad III 322, Odyssey ix 524.

Let us now see what the two foregoing fragments tell us about the Oedipus of the Thebais. In the first place, had he blinded himself before he cursed his sons? In spite of 2.5’s φράσθη and 3.1’s ἐνόησε, Welcker supposed he had (1865:2.337, followed by, for example, Bethe [1891:104–106 and 165]). The self-blinding certainly seems basic to the story and occurs in every version (except, by implication, Homer’s, which characteristically tones down the story’s horrors [see pages 14–15 above]). Blinding and curse seem linked in the corrupt Aeschylus Seven Against Thebes 783–791.

 

The only author cited above who definitely portrays Oedipus as continuing to rule over the Thebans is Homer, and it seems safe to conclude that he was in fact the only author who ever presented this version of events. This is a feature of his normalization of the story (see pages 14–15 above) and follows inevitably upon his elimination of offspring and their father’s curse on them. In our two fragments, by contrast, he no longer participates in sacrifices, and even lacks control over the disposition of his family’s ancestral property. Still more suggestive is his use of a curse (the last resort of the weak and helpless: cf. Watson 1991: 38 and 95) to punish his sons. He presumably lacked more direct means.

That Oedipus survived long enough in Thebes to witness the fulfilment of his curse in the mutual slaughter of his sons is first explicitly suggested by Euripides Phoenician Women 66–76, etc. Bethe (1891:105, 165n7) assumed that this, like several other features of the play, derived from the Thebais. It is more plausibly attributed to the inventive mind of Euripides himself by Robert (1915:1.415), Stephanopoulos (1980:125), Mastronarde (ad loc.), etc. For artists’ depictions of Oedipus as present at his sons’ ἀλληλοφονία see Krauskopf LIMC 54ff. Whether such artists really conceived of Oedipus as literally present or merely a brooding symbol of his curse’s fulfilment is a moot point.

Let us examine the ipsissima verba of one testimony to this sequence of events:

Scholars have been surprisingly slow to detect the major incoherence here. And yet if Polyneices quitted Thebes because he (and his brother) feared the fulfilment of their father’s curse, why on earth should he return (with his brother’s active encouragement) merely because Oedipus had died? Would that death make the father’s curse one jot the less likely of fulfilment? Would it not (if anything) bring it closer?

A further difficulty resides in the phrase καὶ οὕτω τὸ δεύτερον ἔφυγε, as if Polyneices’ initial departure had likewise been enforced! Jacoby, in his commentary on Hellanicus FGrHist 4 F98 (1^A.460), notes that the words are due to contamination with the alternative tradition whereby Polyneices is extruded forcibly and once and for all. But surely the entire passage of Pausanias is a late and incoherent conflation of two originally separate and logical traditions, the forcible and permanent exclusion just mentioned, and the voluntary departure of Polyneices κατὰ συνθήκην, as we find it described by Hellanicus. For here and here alone do we find an uncontaminated explanation of how the brothers’ {62|63} peaceful attempt to avoid the curse ended in strife. According to this version, Polyneices chooses to leave the kingdom to Eteocles and departs, taking with him τὸ μέρος τῶν χρημάτων (including the tunic and necklace of Harmonia), to live in another city (Argos). In other words, Polyneices takes with him all the wherewithal for making trouble against his native land. The curse has already started to take effect by determining Polyneices’ choice.

This is not to say that Hellanicus preserves the Thebais’s version. But he seems likely to preserve Stesichorus’ version (see Davies and Finglass on fr. 97). Bethe (1891:106–107) was particularly anxious to know how Polyneices obtained the ὅρμος in the Thebais. The answer may lie here.

Howald (1939:10) convincingly argues that these two heroes originally stood outside the list of the Seven against Thebes, and belonged together as “ein altes Abenteurerpaar” (in the manner of, e.g., Theseus and Pirithous; see further his Der Mythos als Dichtung [Zurich 1937] 74–79): Tydeus a brutal bully (see page 71 below), and Polyneices a cunning rogue (see page 48 above). He further suggests (p. 12) that Polyneices originally had no father and belongs to the folktale type of individuals who are “Bastarde, aus niederem Milieu entstammend.”

Largely because his own interpretation of the above artifact was essentially incompatible with them, Robert was cautious (1915:1.204) as to the claims of Bethe and others that the more familiar version outlined above derived from epic: he thought it unlikely that the Delphic oracle would have played so significant a role in Ionian epic. But Fontenrose has pointed out (Delphic Oracle, p. 95) the general lack of evidence among the ancient sources for the Delphic origin of the oracle concerning Adrastus’ daughters. As he observes, Adrastus might have received it from a μάντις (so Apollodorus III 6.1) or, more directly, from Apollo.

We have seen (page 33 above) that the gods’ hostility to the expedition against Thebes is posited as early as the Iliad, and is a constant feature of tradition thereafter. μάντεις δ’ ἐπῆλθες ἐμπύρων τ’ εἶδες φλόγα;, Theseus asks Adrastus at Euripides Suppliant Women 155, and when this question elicits a groan, οὐκ ἦλθες, ὡς ἔοικεν, εὐνοίαι θεῶν, he rightly infers (line 157). Compare Aeschylus Seven Against Thebes 378–379, where Amphiaraus forbids Tydeus to cross the Ismenus (οὐ γὰρ σφάγια γίγνεται καλά).

Bethe (1891:26) saw a rationalized reworking of the Thebais’s catalogue of forces in Pausanias IX 9.2 (3.17 Rocha-Pereira):

and IX 9.4:

He could only identify the Arcadian force with that of Parthenopaeus (see page 72 below) and the Messenian with the Biantids.

This detail was attributed to the Thebais by Welcker (1865:2.375). Wilamowitz too, was of the opinion “dass die Stiftung der panhellenischen Nemeen auf das damals allbekannte Epos [viz. the Thebais] zurückgriff” (Glaube der Hellenen 1.392). Stoneman (1981:52–54) reassessed the grounds for this supposition and pronounced them generally good. That the victor list in Apollodorus III 6.4 coincides with the identities of the Seven against Thebes as inferred for the Thebais (pages 68–69 below) strengthens the hypothesis, as does the overall similarity between the events mentioned by Apollodorus and those in Iliad XXIII’s ἆθλα ἐπὶ Πατρόκλωι. Stoneman’s further suggestion (1981:53–54) that the death of Opheltes also fell within the Thebais gains some color from its position within a general framework of gloomy omens: Apollodorus III 6.4 has Amphiaraus rename the child Archemoros as a token of impending doom. So too Bacchylides IX 14, where the dead infant is a σᾶμα μέλλοντος φόνου. We have already seen (pages 33–35 above) that the gods are likely to have expressed their disapproval of the expedition in the Thebais by some such παραίσια σήματα.

ἑπτάπυλος Θήβη: were there seven leaders against Thebes because the city had seven gates, or did the tradition as to the number of leaders determine the tradition as to the number of gates? Wilamowitz’s pungently framed question is best answered by his own fundamental investigation 1891:191–241 = 1971:26–76 (cf. H. W. Singor, “The Achaean Wall and the Seven Gates of Thebes,” Hermes 120 [1992]: 401–411, associating the number with the seven gates in the Achaean wall of Iliad VII 336–359; Kühr 2006:211n63). Thebes was traditionally pictured {66|67} as seven-gated from earliest literature onwards: she is thus in epic (see Hesiod Works and Days 162 with West ad loc.; W. D. Meier, Die Epische Formel im pseudohesiodeischen Frauenkatalog [diss. Zurich 1976) 176] and in poets influenced by epic, even Pindar (see Slater’s Lexicon . s.v.), whose own experience of his native city could have given the epithet the lie. For, as Wilamowitz observed (1891:224–225 = 1971:59), seven gates constitute a paradoxically large number of points of attack for the aspiring enemy, and the historical Thebes at the relevant time can never have possessed more than three or four (cf. Wilamowitz 1891:193–196 = 1971:28–30; Howald 1939:3n2; Burkert 1981:39–40 = 2001:157 on the unhistorical nature of “seven-gated Thebes”). Robert (1915:1.120–121) assumed an Ionian author ignorant of the real city.

Seven operates as “eine alte Märchenzahl” on other levels as well (Seven Thieves, Seven Brides for Seven Brothers; see, in general, Stith Thompson’s Motif-Index 6 s.v. “Seven” [pp. 658–659]; Dirlmeier 1954:154–156 = 1970:51–54; Burkert 1981:44 = 2001:161; Kühr 2006:211–212; M. Davies, “From Rags to Riches: {67|68} Democedes of Croton and the Credibility of Herodotus,” Bulletin of the Institute of Classical Studies 53.2 [2010]: 35n55, etc.). It was this realization that inspired Ernst Howald’s ingenious hypothesis that the Seven under the leadership of Adrastus were originally conceived as seven demons of the Netherworld commanded by their king, the ruler of the Dead (1939:16–17). In a sort of reversal of the motif of the hero’s κατάβασις, or descent to the Underworld, these seven demons broke loose and assaulted a city in the upper world of the living, before being despatched, together with their lord, back to their usual abode. Such a theory would explain the unusual brutality that characterizes most of the Seven (see page 70 below) as well as the story’s radical transformation of other familiar motifs (see Howald 1939:14), which substitutes villains defeated in an enterprise involving a real city in the midst of the known world for the more orthodox picture of heroes victorious in some remote and otherworldly locale (be it Troy, Colchis, or the like). Howald’s case is strengthened by Burkert (1981:40–41 = 2001:158–159), who cites the parallel Babylonian epic of Erra (ninth–seventh century), wherein the god of war and pestilence and seven “matchless warriors” set out to destroy mankind.

Authors variously name the seven gates of Thebes (see Wilamowitz’s list and discussion: 1891:210–218 = 45–53; cf. Kühr 2006:212). We have no evidence as to which, if any, of the nomenclature derives from the Thebais, though, as Wilamowitz (1891:224 = 59n2) saw, the tragedians’ repeated use of Ὁ̆μŏλω̄ῐδε̆ς, in spite of the metrical difficulties it posed them, is suggestive.

Robert concludes that the Thebais’s Adrastus and Mecisteus were replaced in the Aeschylean catalogue by the colorless Eteoclus (cf. Fraenkel 1957:25 = 1964:294) and Hippomedon. The reason for the elimination of Adrastus is obvious (sec. Robert): since Aeschylus’ seven champions are each listed to be killed by a corresponding Theban hero, and since Adrastus survives the battle, he cannot be fitted into the schema. Why Mecisteus should have been ousted is less obvious, and even Eteoclus may have been pre-Aeschylean (see page 69n49 above).

What is meant by “the numerous traces which the Thebais has left in” Attic tragedy may be seen by glancing at Bethe’s list (1891:83–84) of details about the Seven which are common to Aeschylus Seven Against Thebes, Sophocles Oedipus at Colonus, and Euripides Suppliant Women and Phoenician Women. Their consistency in characterization and other respects is such that influence by the Thebais becomes the only satisfactory explanation. In the case of Amphiaraus we may regard the hypothesis as proved, since the stress on his double rôles of seer and warrior which we find in Seven Against Thebes 569 and Oedipus at Colonus 1314–1315 certainly did occur in the Thebais (F7: see page 92 below). Elsewhere we lack this kind of confirmation. But when all three tragedians agree in as many as eight places on the nature of Capaneus’ death (see page 72 below), it is hard not to divine an epic, and specifically the Thebais, as their common source.

On this hero in general see Robert, Heldensage 3.1.924–926; Nilsson, The Mycenaean Origin of Greek Mythology (Berkeley 1932) 116–117 (“his character shows traces of a high and crude antiquity which was detested by the Homeric age”: for the most extreme instance see page 81 below); and Dirlmeier 1954:157 = 1970:53 on this and other names of the Seven ending in –eus. Aeschylus Seven Against Thebes 424–425 hints at the tradition as to Tydeus’ small stature which we find in the Iliad (V 801 μικρὸς μὲν ἔην δέμας, ἀλλὰ μαχητής), and Fraenkel (1957:16 = 1964:285) suggests this may have occured in the Thebais. The Seven Against Thebes also reveals a state of hostility between this hero and Amphiaraus: Tydeus upbraids the seer for cowardice (lines 382–383); Amphiaraus denounces the blood-lust and bad counsel of Tydeus (lines 570–575). Bethe (1891:82–83) supposed this enmity to derive from the Thebais, a suggestion approved by Wecklein (1901:663–664). For ἔρις as an important motif in epic see e.g. D. L. Cairns, “Affronts and Quarrels in the Iliad,” Papers of the Leeds International Latin Seminar 7 (1983): 155–167 = Oxford Readings in Homer’s Iliad (2001) 203–219, and “Ethics, Ethnicity, Terminology: Iliadic Anger and the Cross-Cultural Study of Emotion,” Yale Classical Studies 31 (2003): 11–49.

On this figure see in general Hutchinson on Aeschylus Seven Against Thebes 526–567 and Fowler 2013:411. “P’s Arcadian birth and early metoecism to Argos is a constant detail, whether original to Seven Against Thebes 547–548 or not” (Collard on Euripides Suppliant Women 888–891 [2.331]). See Fraenkel 1957:37–38 = 1964:306. The hero’s status as Arcadian and a son of Atalanta was confidently attributed to the Thebais by Bethe (1891:86n11). More cautious are Jacoby on Hellanicus FGrHist 1 F32 (p. 328) and Fraenkel as cited. Howald (1939:12–13) sees the association of this hero (and, indeed, the Seven at large) with Argos as the product of a late tendency to locate saga in specific contexts (cf. Kühr {72|73} 2006:138n26), and attributes it to epic. On the likely original form of the name see Beazley, “Some Inscriptions on Vases: V” (as in 69n49), 313–314.

Wilamowitz (1914:99n1) and Fraenkel (1957:32 = 1964:301) suppose he featured in epic, while Bethe (1891:87n13) specifically assigned him to the Thebais’s list of the Seven. Against this latter supposition see page 70 above.

In Aeschylus, of course, each of the Seven Argive commanders is allotted a Theban warrior as opponent. Prima facie we would expect this to be Aeschylus’ own invention, one demanded, as Robert (1915:1.246) puts it, by the play’s dramatic framework. Besides, as Wilamowitz (1891:225 = 60) and Robert saw, the Thebais seems to have made Periclymenus the opponent of both Amphiaraus and Parthenopaeus (see pages 78–79 and 90 below). Aeschylus fails to mention him at all. Polyneices must have been matched with Eteocles, of course, from the very first; and Aeschylus’ ranking of Melanippus against Tydeus also probably derives from the Thebais, where the enmity of the two reached a grisly climax (see pages 80–81 below), though in fact it was Amphiaraus who finally despatched Melanippus. On Melanippus’ presumed importance in the Thebais see Fraenkel 1957:14 and n1 = 1964:283 and n4.

For handy surveys of vases that possibly depict the Seven commanders see Small, pp. 135–138 and M. Tiverios, “Sieben gegen Theben,” Mitteilungen des Deutschen Archäologischen Instituts, Athenische Abteilung 96 (1981): 145–161. The likeliest candidates (LIMC VII.1 s.v. “Sieben” 730ff.) are the cup by Macron (Louvre G271: ARV ² 461.33), ca. 490/80; a hydria in Basel (the Borowski Collection), ca. 470/60; a lekythos by the Terpaulus Painter (Agrigentum Mus. Civ. 23: ARV ² 308.5, Paralip. 357), first decade of the fifth century; and the cup by the Kleophrades Painter (Aths. Nat. Mus. Acrop. 336 [B 87]: ARV ² 192.105), ca. 480 (for illustrations see Tiverios, plates 43–45). Each vase presents a similar scene. {74|75}

Thanks to the relative inexplicitness of this type of representation, we could not be sure, even if we knew for certain that the figures involved were the Seven, how precisely to interpret their activities. Tiverios supposes they are preparing for battle by arming themselves; but other scholars (see Small as cited on page 40 above) have taken them to be leaving home for the war.

Tiverios’ hypothesis of a literary source (our Thebais) is perfectly arbitrary and leads him into unconvincing stratagems like his attempt (p. 147) to produce a single unitary figure from the warrior holding a helmet in his hand (Terpaulus Painter and Kleophrades Painter), the warrior holding a greave (Macron), and the warrior holding a sword (Basel hydria). Parthenopaeus and Adrastus seem safely identified on these vases. As for the rest of the figures, one may more safely generalize Small’s conclusion vis-à-vis the Basel hydria (p. 176): “the other warriors take genre poses and must remain nameless.”

Those who are reluctant to accept that Onasias’ painting derives (see page 24 above) from the Oedipodeia have sometimes wondered whether the Thebais may not in fact be the inspiration behind this artifact (cf. Robert 1915:1.180). The obstacle to this is roughly the same as before: why does Pausanias fail to add the Thebais’s name? Why, indeed, does he mention Onasias at all?

For a useful survey of the numerous artifacts that have been thought to display the duel of the brothers see Small (104–108), who stresses the frequency with which the cautious scholar must abandon the combatants as unidentifiable or anonymous participators in a generic duel, and shows that numerous Etruscan urns are likelier interpreted as revealing the duel of Aeneas and Turnus, or Arruns and Brutus, or Romulus and Remus. {75|76}

I am sympathetic to the idea that this latter version of events stems from the Thebais (Bethe 1891:123–126; cf. Wilamowitz 1891:225 = 1971:60; Stoneman 1981:49 etc.), not so much for the reasons alleged by Bethe as because it is hard to conceive of any other source for an account that survives into such late authors and yet is at odds with the tragic vulgate. The motif of the victorious battle followed by the unsuccessful assault on the walls is certainly both Homeric and epic, as Bethe observed. Likewise, the striking down of the triumphant hero from the walls he seems set to scale and the routing of his side, with the sallying forth of the besieged in consequence, also have numerous analogies in epic, the Aethiopis’ Achilles in particular (cf. West 2013:149).

In Euripides’ account, Parthenopaeus is thrown from the wall like Capaneus. In Apollodorus, by implication, he is killed in the hand-to-hand fighting that accompanies the Argive retreat caused by Capaneus’ death. Bethe (1891:125) thinks this latter context more appropriate both for the kind of battle we expect in an epic and for the mighty son of Poseidon. The Thebais certainly seems to have had Periclymenus attack Amphiaraus at this stage of the conflict: see page 90 below.

Why Aeschylus should have chosen to omit Periclymenus is, as Bethe says (1891:88), a mystery: see page 68 above.

In saying that Amphiaraus slew Melanippus and cut off his head and brought it to Tydeus, Σ Gen. is in agreement with all our sources save Apollodorus III 6.3 (see Davies and Finglass on Stesichorus fr. 92); Statius (Thebaid VIII 739–750), who, having already despatched Amphiaraus to subterranean gloom, appropriately transfers the action to the hateful Capaneus (see page 76 above); and Libanius (Progymnasmata R 4.1100 [8.40 Förster]), who assigns the task to a nameless companion of Tydeus. On Melanippus’ function in the Thebais see further {81|82} page 78 above. The fight between him and Tydeus is traced back to the widespread motif of the Combat Myth (the Champion against the dark Antagonist) by Fontenrose as cited, 124–126 and n42.

We have here, then, a nice example of the different approaches of art and literature. For I agree with Beazley (p. 7) when he suggests that “the complete personification of Athanasia may be due to a painter, who from the nature of his {82|83} art, had to choose, and could not sit on the fence between person and thing.” He is clearly right to observe that none of the literary allusions to the incident need imply personification, not even Σ Pindar Nemean X 12, cited above, which tells how ἡ Ἀθηνᾶ τὴν ἀθανασίαν παρήγαγε but proceeds to describe immortality as a “gift” (δωρεά, δῶρον).

We may close with a few points of detail. The word Ἀθανασία cannot have occurred in the Thebais, since, as Beazley reminds us (7; cf. R. Renehan, “A New Lexicon of Classical Greek,” Greek, Roman, and Byzantine Studies 24 [1983]: 13), it will not fit into a hexameter and is not securely attested until the fourth century. Apollodorus’ picture of Athena παρὰ Δίoς αἰτησαμένη … φάρμακον … δι’ οὗ ποιεῖν ἔμελλεν ἀθάνατον αὐτόν [scil. Tύδεα] is strongly reminiscent of a scene in the Aethiopis as summarized by Proclus where we have Ἠὼς παρὰ Διός αἰτησαμένη ἀθανασίαν and presenting it to her son Memnon (cf. West 2013:148–149).

That Tydeus died and was buried at Thebes is implied by Iliad XIV 114 (Tυδέος, ὃν Θήβηισι χυτὴ κατὰ γαῖα κάλυψεν): cf. Σ Gen. ad loc. (2.135 Nicole): ὃς ἐν Θήβαις ἐτελεύτησεν, ἐν Θηβαικῶι πολέμωι. Zenodotus athetized and Aristophanes deleted the line (see Σ A ad loc. [3.583 Erbse] and Erbse’s note ad loc. for a bibliography of attempts to understand this odd attitude). Σ T ad loc. (3.584 Erbse) explains the Homeric verse ὅτι οὐ κατὰ τοὺς τραγικοὺς ἐν Ἐλευσῖνι μετηνέχθησαν οἱ περὶ τὸν Καπανέα, which looks, as Severyns saw (1928:224), to be a relic of the usual Aristarchean contrast between Homer’s version and that of οἱ νεώτεροι. On the Thebais ’s version of the fates of the defeated Argive corpses see further page 93 below.

Severyns (1928:222) thought (a) an important document for Pausanias’ ignorance of the Epic Cycle: “visiblement, il n’a pas connu, par une lecture personelle, le passage … auquel il fait allusion,” for he automatically reproduces a verse which is not very explicit on Poseidon’s alleged parentage. In fact the alleged oral tradition of the priests is very dubious (Severyns takes it too seriously) and may be a supposititiously circumstantial and “historical” way of conveying (more Herodoteo) an Alexandrian dispute over the significance of κυανοχαίτης. At any rate, Severyns’ skepticism will neatly dispose of Van der Valk’s: the latter says that if the Thebais had really related Arion’s filial relationship to Poseidon, Pausanias would have adduced it. But if Pausanias’ knowledge was indirect this line of argument must fall.

Van der Valk’s attempts to discredit the testimony of (c) are no more fortunate. It is arbitrary of him to suppose that because (b) is shorter than (c) it represents a “correction” of (c)’s error in attributing too much to our epic. And his customary explanation of the contents of the relevant scholion as a “guess” (this time based on the common knowledge that the Thebais named Adrastus as Arion’s owner) is as unconvincing as usual. Further arguments in favor of (c) as an integral part of our fragment are to be found in Janko 51–55.

A description of Arion’s passage through the hands of successive owners would be quite in keeping with early poetry (see, for instance, the account of Agamemnon’s scepter in Iliad II 101–105), as Severyns (1928:221) saw. For the significance of the mention of Cycnus see Janko 51–55. On Arion and his role in saving Adrastus’ life see in general L. Malten, “Das Pferd im Totenglauben,” Jahrbuch des Kaiserlich Deutschen Archäologischen Inst ituts 29 (1914): 201–208. Building on foundations laid by Wilamowitz (“Lesefrüchte,” Hermes 35 [1900]: 563–564 = Kl eine Schriften 4.140), Malten ingeniously inferred from passages such as Propertius II 34.37 (vocalis Arion), Statius Thebaid VI 424 (praesagus Arion), and ibid. XI 442–443 (fata monentem | … Ariona) that the horse could speak in a {86|87} human voice, not only in the context indicated by these passages (the Nemean funeral games for Archemorus), but as a warning to his master Adrastus of the impending catastrophe at Thebes. This warning he attributed (p. 205) to our epic, comparing the prophecy of Achilles’ horse at Iliad XIX 400–418. A similar picture, with Arion and Adrastus as the models for the Homeric Achilles and Xanthus, was drawn by E. Heden, Homerische Götterstudien (diss. Upsala 1912) 136–138; cf. B. C. Dietrich, Death, Fate and the Gods (London 1965) 237. For artifacts that depict Adrastus’ escape with his horse (including the famous relief from Gjölbaschi-Trysa) see Krauskopf in LIMC I.1 H1 (pp. 235–236).

Arion’s begetting was sometimes located in Arcadia and sometimes in Boeotia. For the details see Pfeiffer on Callimachus fr. 652 and Fontenrose, Python 367–371; for discussion see Wilamowitz, Dietrich, and Fontenrose as cited. Bethe (1891:92–93) infers that the Boeotian version is the earlier (because that land, unlike Arcadia, is suited for the breeding of horses and because Copreus in [b] can be linked to Boeotia [cf. Σ T Iliad XV 639 (4.133 Erbse): Κοπρεύς … ἄλλος Bοιώτιος, Ἁλιάρτου παῖς]), and equates this older form with the Thebais’. Wilamowitz {87|88} (1891:225n1 = 60n1) took the reverse view on account of the Arcadian spelling Ἐρίων (see page 87 above). For further bibliography and discussion see Dietrich as cited (126n2).

εἵµατα λυγρά: Welcker (1865:2.369) ludicrously supposed these to be garments of mourning (“Trauergewand”: the rendering was still taken seriously by Wecklein [1901:685]), an idea rightly scotched by Bethe (1891:93n25): as if a defeated general fleeing for his life from the battlefield will have time to slip into something appropriately gloomy! We should follow Bethe in turning first to those Odyssean passages where similar phrases are used (λυγρὰ δὲ εἵματα ἕσσε περὶ χροΐ [xvi 457] and τὰ δὲ λυγρὰ περὶ χροΐ εἵματα ἕστο [xvii 203, 338; xxiv 158]) of Odysseus’ disguise as a decrepit old {88|89} beggar. “Dirty, tattered” is the meaning suggested by the context of all these passages, and a very similar meaning seems required in our own line. Iliad XVIII 538, from the battle scene on the Shield of Achilles (εἷμα δ’ ἔχ’ ἀμφ’ ὤμοισι δαφοινεὸν αἵματι φωτῶν), conveniently reminds us of the way in which εἷμα in the singular can be used of a warrior’s armor and suggests the nature of the grime in the present passage. For a comparable exploitation of tattered clothing in elevated poetry we may cite Xerxes’ rags in Aeschylus’ Persae. On the meaning of λυγρός see further I. Anastassiou, Zum Wortfeld “Trauer” in der Sprache Homers (diss. Hamburg 1973), esp. 154–155. For the possibility that the noun is corrupt see the following note.

φέρων: previously rendered as “wearing” (Evelyn-White in his Loeb text of Hesiod, the Homeric Hymns and Homerica [London 1914] 485, Huxley 1969:44 etc.; Krauskopf, LIMC 1.1.232, etc.). But the verb does not mean this anywhere else in Greek: φορῶν is what this translation requires, and the two verbs are often interchanged in MSS (see Barrett on Euripides Hippolytus [Oxford 1964] line 316). Perhaps the poet intended this here: for the Attic contraction in epic see φοροῦσ’ (Cypria F 4.3). Synezesis (φορέ͜ων) is another possibility. εἵματα λυγρὰ φέρων could hardly mean “enduring filthy garments” (Odyssey xviii 134–135 ἀλλ’ ὅτε δὴ καὶ λυγρὰ θεοὶ μάκαρες τελέσωσι, | καὶ τὰ φέρει ἀεκαζόμενος τετληότι θυμῶι is a merely formal parallel), or “bringing” them “with him” (Grote, History of Greece 1.268–269). But we cannot absolutely exclude the possibility of the “weak” meaning “with” here, which φέρων, like ἄγων, ἔχων, and λαβών, can exhibit in tragedy (see Stinton, “Agamemnon 1127 and the Limits of Hyperbaton,” Proceedings of the Cambridge Philological Society 21 [1975]: 85 = Collected Papers on Greek Tragedy 101). Compare West’s translation 2003:55. Alternatively, we may follow W. Beck, “Thebais Fr. 6^A Davies (Pausanias 8.25.8),” Museum Helveticum 38 (2001): 137–139 in emending the first word of our fragment to σήματα (an easy change), with reference to the tokens attached by the Seven to Adrastus’ chariot at the start of their expedition, as keepsakes for their heirs should they fail to return (cf. Aeschylus Seven Against Thebes 49–51). σὺν Ἀρίονι: “with the help of” A.: cf. Iliad V 219–220 σὺν ἵπποισιν καὶ ὄχεσφιν | ἀντιβίην ἐλθόντε. Other epic instances of σύν with this nuance in Chantraine, Gramm. hom. 2.135. Ἀρίονι: this (not -είονι) is the right spelling: see LfrgE S.V. (1. col. 1304), Pfeiffer on Callimachus fr. 223. Ἀρίονι κυανοχαίτηι |: cf. [Hesiod] Shield of Heracles 120 ὣς καὶ νῦν μέγαν ἵππον Ἀρίονα κυανοχαίτην |. The epithet is also associated with horses in Iliad XX 224 (Boreas) ἵππωι δ’ εἰσάμενος παρελέξατο κυανοχαίτηι |, Hesiod Theogony 278 τῆι δὲ μιῆι παρελέξατο Κυανοχαίτης | (of Poseidon’s intercourse with Medea, from whose severed head Pegasus later springs), Antimachus fr. 50 Matthews πατρί τε κυανοχαῖτα Ποσειδάωνι πεποιθώς (probably referring to Arion: see Matthews ad loc.). But of course it is most often employed in Homer of Poseidon (Iliad XIII 563, Odyssey ix 528, etc.), to whom it is so often applied that it can be used as a substitute for his actual name (Iliad XX 144, Odyssey ix 536, Hesiod Theogony 278). μελαγχαίτης is used of centaurs in [Hesiod] Shield of Heracles 186 and on the François Vase.

Adrastus had a son called Κυάνιππος (Apollodorus I 9.13). If Adrastus was originally the leader of a host of underworld demons, as Howald supposed (see above page 68), the epithet becomes even more significant (cf. Homeric Hymn to Demeter 347 Ἄιδη κυανοχαῖτα, Euripides Alcestis 439 ἁίδας ὁ μελαγχαίτας θεός). On Tydeus’ opponent Melanippus see page 73 above. {89|90}

On Amphiaraus in myth see Robert 1915:1.205–214; Howald 1939:13–14; Usener, “Der Stoff des griechischen Epos,” 37–39 = Kl eine Schriften 4.234–239; F. Bener, Die Amphiaraossage in der griechischen Dichtung (diss. Zurich 1945); P. Vicaire, “Images d’Amphiaraos dans la Grèce archaïque et classique,” Bulletin de l ’ Association Guillaume Budé 1 (1979): 2–45; Braswell’s commentary on Pindar Nemean XI (Berlin 1998); and Matthews’ on Antimachus, general index s.v. ; LIMC I.1 s.v. “Amphiaraos” E (pp. 694–697). On the connection between horse and chariot, underworld and death which the manner of Amphiaraus’ end implies see (apart from Usener) Malten (as cited page 86 above); Dietrich, Death, Fate, and the Gods, 131–132; Richardson on Homeric Hymn to Demeter 18. For Baton and Amphiaraus in the Underworld in art see LIMC I.1 E 20 (p. 85); S. I. Johnston and T. J. McNiven, “Dionysus and the Underworld in Toledo,” Museum Helveticum 83 (1996): 29. Howald (1939:14) assumes that the hero’s relatively merciful fate was a later modification induced by reluctance to accept a collective retribution suffered by every one of the seven commanders. The fullest surviving literary treatment of his disappearance is Pindar Nemean IX 24–27:

That this description derives from the Thebais was suggested by Welcker (1865:2.366) and approved by such scholars as Wilamowitz (1891:225 = 1971:60 and n3), Robert (1915:1.246), Stoneman (1981:49), and Braswell ad loc.

Apollodorus III 6.8 continues the narrative quoted above with the following: ὁ δὲ σὺν τῶι ἅρματι καὶ τῶι ἡνιόχωι Βάτωνι, ὡς δὲ ἔνιοι Ἐλάτωνι, ἐκρύφθη. Amphiaraus’ charioteer appears on the above vases and on two Volterran urns (Small, cat. 13 and 14, pp. 20–21 and 150–151). He is also given some prominence in depictions of Amphiaraus’ departure from home and family (see page 104 below); scholars have generally inferred that he was mentioned in the Thebais. Amphiaraus’ chariot is specified as a quadriga by Hyginus Fabulae 250 and Propertius II 34.39 (compare the two vases mentioned above): see page 90 above.

“Die übrigenzitate eines A[sclepiades] in den Pindarscholien gehören sicher einem kommentator, d.h. doch wohl dem Myrleane” (Jacoby, FGrHist 1^A.487.39–40, citing A. Adler, “Die Commentare des Asklepiades von Myrlea,” Hermes 49 [1914]: 39–46). Thanks to the Pindaric scholion ad loc. we know that the relevant portion of the sixth Olympian exemplifies a common practice in early Greek poetry (see R. Kassel in that part of his article “Dichterspiele” which deals with Metaphrasis [Zeitschrift für Papyrologie und Epigraphik 42 (1981): 11–17 = Kleine Schriften 121–128]), whereby phrases or lines from epic were re-cast in similar metres by later poets. But how far does Pindar’s indebtedness to the author of the Thebais extend? The question must be answered by examining first of all the individual line and then the general context. {91|92}

(iii) That the allusion to the Thebais is confined to verse 17 of Pindar’s sixth Olympian was argued by Wilamowitz (Isyllos von Epidauros 163n4) and Robert (1915:1.248). For more generous interpretations of ταῦτα εἰληφέναι see Rzach 1922:2371.30–31. If we wish to decide whether the Thebais’s influence extends further than the words just considered, we would do well to start by examining the first part of Adrastus’ speech as reported by Pindar: ποθέω στρατιᾶς ὀφθαλμὸν | ἐμᾶς. Many scholars have automatically assumed that this too derives from the Theban epic (Ribbeck [“Zu den Fragmenten der griechischen Epiker,” Rheinisches Museum 33 (1878): 458] restored the noble hexameter ὀφθαλμὸν ποθέω στρατιῆς {92|93} ἐὺν Ἀμφιάρηον and alii aliter finxerunt). As Bethe observes (1891:58–59), no very staggering implications for the history of the myth would follow if this assumption could be proved correct. “This means that in the Thebais too, after the battle was over, Amphiaraus was not to be found among either the fallen or the survivors—was in fact translated.” So writes Rohde (1.114n2 = Engl. trans. 103n2), one of the most enthusiastic supporters of a generous interpretation of the Greek phrase ταῦτα εἴληφεν. But since the mysterious disappearance of Amphiaraus must be basic to any version of the story, Rohde’s enthusiasm is here perhaps misplaced.

Is there any good reason why this part of the speech too should not ema-nate from the Thebais? The only objection with which I am acquainted is Robert’s (1915:1.248): that ὀφθαλμός is unlikely to have possessed the required metaphorical sense in early epic. Certainly, whether we take the metaphor to be one where “ὀφθαλμός is … used metaphorically of anything precious” (Stevens on Euripides Andromache 406; cf. W. Schadewaldt, “Experimentelle Philologie,” Wiener Studien 79 [1966]: 75–76 = Hellas und Hesperia, 2nd ed., 1.491–492) or (with D. Bremer, Licht und Dunkel in der frühgriechischen Dichtung [Bonn 1976] 239) detect an allusion to the motif “des leitenden Blickes,” the closest analogies are provided from tragedy and elsewhere in Pindar, and early epic has nothing comparable. However, in view of the numerous un-Homeric features of the few surviving fragments of the latter, this argument cannot be pressed.

But what of the area over which Bethe and Rohde do disagree—as violently as possible? I mean the latter’s intuition that “Pindar must have taken over not merely the words of the lament of Adrastus but the whole situation that led up to these words, as he described it, from the Thebais.” Welcker (1865:2.324, 367) had certainly taken for granted that the Thebais’s framework for the mention of Amphiaraus’ prowess as seer and warrior was, as with Pindar, a speech by Adrastus at the funeral of the Greeks who perished at Thebes. See Bethe 1891:94n27 for a bibliography of those who take this view. Bethe himself (58–59, 94–96) thought otherwise. His primary objection was to the whole idea of the cremation of the dead, which he believed alien to early epic. Homeric analogies, he argued, would lead us to expect the abandonment of the defeated army’s corpses to the open air and the tender mercies of birds and beasts of prey. And such a fate is precisely what befalls the body of Capaneus in Aeschylus fr. 17 Radt (from the Ἀργεῖοι), the body of Polyneices in Sophocles’ Antigone, and the bodies of the invading army in general at verses 1080–1083 of the same play. Only when the burial of the dead becomes widely regarded as a sacred duty does literature turn its attention to the fate of the corpses of the Argive expedition. Then it is that plays such as Aeschylus’ Ἐλευσίνιοι (TrGF 3 pp. 175–176 Radt) and Euripides Suppliant Women are written, and Pindar (on Bethe’s interpretation) revises the {93|94} earlier myth at verses 15–18 of Olympian VI to bring it into conformity with contemporary religious beliefs concerning the dead.

How then did the Thebais’s Adrastus praise Amphiaraus (for Bethe rather eccentrically retains from the theory he is criticizing this particular feature) and lament his loss? Bethe, who on page 95 of his book sneers at Welcker over “einer neuen Bethätigung seines erfinderischen Geistes,” proceeds, on page 96, to a display of his own inventive spirit by conjuring up the following vivid context for the encomium of Amphiaraus: speeding over the battlefield in wild flight from his pursuers, Adrastus suddenly sees the noble seer sink from sight and with him the last vestige of hope. An immeasurably more exciting and logical framework for Adrastus’ speech of praise (Bethe finds) than the comparatively feeble and banal adaptation of the motif by Pindar, whose wording first leads us to expect precisely the same picture (lines 12–14 αἶνος … ὃν ἐνδίκας | ἀπὸ γλώσσας Ἄδραστος μάντιν Οἰκλείδαν ποτ’ ἐς Ἀμφιάρηον | φθέγξατ’ ἐπεὶ κατὰ γαῖ’ αὐτόν τέ νιν καὶ φαιδίμας ἵππους ἔμαρψεν) and then oddly and awkwardly postpones the speech (and with it Adrastus’ realization of the significance of his loss) until some unspecifically later occasion when the dead were buried at Thebes (lines 15–16: ἑπτὰ δ’ ἔπειτα πυρᾶν κτλ.).

As Rzach observes (1922:2367–2368), this fragment may derive from an account of Tydeus’ genealogy. {95|96}

At the start of his somber catalogue of things we do not know about the Thebais, Robert (1915:1.180–182) gave pride of place to Oedipus’ wife, what her name was, and whether she was alive or dead by the time her sons clashed for the last and fatal time. His suggestion that, as in the Oedipodeia, she went by the name of Euryganeia (1.180–181) is a rather misleading guess, which is best ignored.

However, Robert’s own interpretation of the available data (an angry Athena—Athena Onca: cf. Aeschylus Seven Against Thebes 486–487 and 501–502—demands the punishment of her votaress, who has offended by intercourse with Periclymenus, son of Poseidon and therefore [cf. Robert, Heldensage 3.1.924–925] the goddess’s enemy) is perfectly compatible with a peacetime visit to Thebes by Tydeus, and this is precisely what Friedländer postulated [see page 34 above]). {96|97}

Rzach maintained (1922:2372.44–47) that if the corpses of the Argive chieftains were cremated in the Thebais on seven pyres, Polyneices’ corpse must have been among them. This is not necessarily the case (see page 94 above).

Did Creon feature in the Thebais? We know him to have been named as the father of a victim of the Sphinx in the Oedipodeia F1, which only makes sense if (see ad loc.) he was also conceived as regent of Thebes. But that does not necessarily guarantee his appearance in our epic. {98|}