The Theban Epics

  Davies, Malcolm. 2015. The Theban Epics. Hellenic Studies Series 69. Washington, DC: Center for Hellenic Studies.

Chapter 1. Oedipodeia


ἡ Οἰδιποδ(ε)ία (Tabula Borgiana [T: see page 133 for text] and Σ Euripides Phoenician Women [F1: see page 133 for text]), meaning “the poem about Oedipus” (for the variation in spelling and [perhaps] the principle see Stesichorus’ Εὐρωπ(ε)ία fragment 96 with Davies and Finglass ad loc.); or τὰ Οἰδιπόδια (scil. ἔπη) by analogy with the Cypria and Naupactica as Pausanias (F2: see page 134 for text) cites the work? There is not really enough evidence to decide, although Pausanias is the likelier to be wrong because his context provides more opportunity for corruption through assimilation (τὰ ἔπη … ἃ ὀνομάζουσι). On the “Iliac tablets” exemplified by T, see M. Squire, The Iliad in a Nutshell: Visualizing Epic on the Tabulae Iliacae (Oxford 2010).


In F1 and F2 the work is cited anonymously with the formula οἱ τὴν Oἰδ. γρά-ψαντες / ὁ τὰ ἔπη ποιήσας ἃ Οἰδ. ὀνομάζουσι. There is no substantial difference between these two expressions. The once-popular notion (so, for instance, Rzach 1922:2358.7–8, etc.) that the plurals in the former (and in such analogous phrases as Epigoni F3 (see page 144 for text) oἱ τὴν Θηβαΐδα γεγραφότες, Cypria F21 οἱ τῶν Κυπρίων ποιηταί) indicate uncertainty as to authorship is illogical and unparalleled, and quite out of the question in the present instance. The use of plural for singular is idiomatic in this type of anonymous citation: see K. Alpers, Das attizistische Lexikon des Oros (Berlin 1981) 82n14. The Tabula Borgiana’s unique attribution of the work to Cinaethon must be viewed very skeptically, not because of the doubts as to its reliability expressed by W. McLeod (“The {1|2} ‘Epic Canon’ of the Borgia Table: Hellenistic Lore or Roman Fraud?,” Transactions and Proceedings of the American Philological Association 115 [1985]: 153–165; for a defense, see Squire 2011:44–47), but because of the general tendency of later writers to attach authors’ names to epics earlier writers had cited anonymously. [1] Wilamowitz suggested his supplement for the relevant passage in 1884:334 purely exempli gratia. For its problematic relation to the space required see McLeod 1985:159–160. West’s supplement is proposed in West 2013:3.

Jedes Urteil über die älteste form wird dadurch erschwert, dass wir die thebanische Epen nicht wiederherstellen können.

F. Dirlmeier, Der Mythos von König Oedipus2 (Mainz 1964) 14

We would learn a good deal about the handling of the Oedipus legend by the Attic tragedians and later writers, and other similarly inestimable advantages would accrue, if only we possessed some reliable information as to the general contents of this early epic. Unfortunately, the number of actual fragments that we have is tiny, and no one would call them particularly informative. At least, F1 on Haemon’s death at the claws of the Sphinx tells us relatively little, and F2 on the mother of Oedipus’ children has been dismissed as no more helpful by several scholars. Other critics, however, are more sanguine, and suppose that, if combined with later sources, this latter fragment can be used to open up surprisingly wide areas of the now-vanished poem. Here, then, we already meet the two incompatible attitudes that will clash again and again in these pages: the sanguine and the skeptical. Needless to say, even the more optimistic scholars disagree among themselves, and are divided, for instance, as to which of our later sources can legitimately be combined with F2 of the Oedipodeia to produce valid evidence.

Both types of dissension are fully represented even in the short list of treatments of our epic that follows: Bethe 1891:1–23, Robert 1915: passim, but esp. 1.149–168; Deubner 1942:2–27 = 1982:636–661. There is a critique of these three fundamental studies in Stephanopoulos 1980:103–110. Note also de Kock 1961 (with bibliography in 13n35 and 15n43) and 1962, and Wehrli 1957 = 1972. For a more general bibliography of treatments of Oedipus see G. Binder, Die Aussetzung des Königskindes: Kyros und Romulus (Beiträge zur Klassischen Philologie 10 [Meisen-heim 1964]), 142–143; Lowell Edmunds 1981a (with bibliography on 30–39) and 1981b:221–238; and T. Petit, Oedipe et le Chérubin: Les sphinx leventins, cypriotes et grecs comme gardiens d’immortalité (Göttingen 2011). {2|3}

Bethe’s first chapter gives a brief, exuberant, and colorful sketch of the Oedipodeia’s contents as he thought they could be recovered from later sources. Robert brought a massive weight of learning to crush this rash attempt at expanding the boundaries of knowledge. Deubner, while admitting the justice of much of Robert’s negative criticism, believed that more could be salvaged from the wreckage of Bethe’s theory and more inferred from a careful analysis of some later evidence than Robert allowed, though he himself often disagrees with Bethe over the details of reconstruction.

Under the influence of Robert’s destructive onslaught upon his predecessors, Wilhelm Schmid tried to confine his remarks on the poem in GGL 1.1.202 to what we certainly know. This proves to be circumscribed enough: the epic was 6,600 lines long; it mentioned Haemon as one of the Sphinx’s victims; it gave the name of Euryganeia to the mother of Oedipus’ children. Schmid relaxed his scruples enough to permit two inferences: Oedipus’ rescue of Thebes from the Sphinx and his marriage to his mother must also have fallen within the poem’s scope.

Even this rigorous approach may have admitted too many uncertainties: thus the cautious and skeptical Robert allowed the hypothesis that the Oedipodeia mentioned the Sphinx’s riddle, something which other scholars find quite inconceivable (see pages 9–13 below). Clearly we must carry out a careful and scientific examination of the credentials of this and all other similar suggestions before we can allow them anything approximating a serious hearing.

The best way of proceeding seems to be to fix in our minds the outlines of the Oedipus story as familiar to us from later writers and then isolate each individual detail and ask what grounds there are (if any) for supposing that detail to have featured in our epic. But before embarking on this painstaking examination, we must first inspect the notorious “Pisanderscholion” which will form a suitable prelude to our task. For, according to Bethe, it handily contains within itself a summary of the contents of the whole epic, a ὑπόθεσις, as it were, for the entire Oedipodeia. After the onslaught upon it by Robert and then Deubner, this position seemed, for nearly a century, to have been abandoned as unsustainable; but finally it found perhaps surprising adherence in one of the last of Lloyd-Jones’ contributions to classical scholarship. On the identity of Pisander see in particular Deubner 1942:5–18 = 1982:639–652; Keydell’s article in RE s.v. “Peisandros (13)” (19 [1938]: 146–147); Jacoby ad loc. (1A.493–494) and in his Nachträge thereto (1A2.544–547); de Kock 1962; Ed. Fraenkel, “Zu den Phoenissen des Euripides,” Sitzungsberichte der Philosophisch-philologischen und der Historischen Klasse des Königlich Bayerische Akademie der Wissenschaften 1 (1963): 6–7; Mastronarde’s commentary on Euripides Phoenician Women (Cambridge, 1994) 31–38; Lloyd-Jones 2002 = 2003; N. Sewell-Rutter, Guilt by Descent (Oxford {3|4} 2008) 61–65. The text to be discussed occurs in Σ Euripides Phoenician Women 1760 (1.44 Schwartz = argumentum 11 of Mastronarde’s Teubner text of this play) = Peisandros FGrHist 16 F10.


The first item concerns the reason for the Sphinx’s sudden appearance within Theban territory: she was sent by Hera (presumably in her role as γαμοστόλος: cf. Lloyd–Jones 2002:9 = 2003:28) to punish the Thebans for their toleration of Laius’ indulgence in homosexuality and his consequent abduction of Chrysippus. Before considering this section in full we must note (with Robert [1915: 151–155]) the presence within it of a digression on the Sphinx and her victims which one would not readily attribute to the same source as the surrounding context:

ἦν δὲ ἡ Σφίγξ, ὥσπερ γράφεται, τὴν οὐρὰν ἔχουσα δρακαίνης. ἀναρ-πάζουσα δὲ μικροὺς καὶ μεγάλους κατήσθιεν, ἐν oἷς καὶ Αἵμονα. τὸν Κρέοντος παῖδα καὶ Ἵππιον τὸν Εὐρυνόμου τοῦ τοῖς Κενταύροις μαχεσα-μένου. ἦσαν δὲ Εὐρύνομος καὶ Ἡιονεὺς υἱοὶ Mάγνητος τοῦ Αἱολίδου καὶ Φυλοδίκης. ὃ μὲν οὖν Ἵππιος καὶ ξένος ὢν ὑπὸ τῆc Σφιγγὸς ἀνηιρέθη, ὁ δὲ Ἡιονεὺς ὑπὸ τοῦ Οἰνομάου, ὃν τρόπον καὶ οἱ ἄλλοι μνηστῆρες.

At first sight, the agreement with F1 of the Oedipodeia over Haemon as one of the Sphinx’s victims might seem to support Bethe’s case. But in fact, it is the remainder of the account of the victims which undermines it. The strangely disproportionate attention (not taken into account by Lloyd-Jones [2002:4 = 2002:23] when he disputes Robert’s claim) here paid to the strictly irrelevant Eurynomus and Eioneus indicates, as Robert (1915:154) saw, a source in the form of a “mythologische Traktat” which summarized the legend of the Lapiths’ battle along much the same lines as Diodorus Siculus IV 99. This part of the scholion, then, can safely be segregated from any reconstruction of the Oedipodeia.

Nevertheless, even Bethe himself (1891:10–15) was obliged to argue that the pristine outline of the epic story had been blurred to obscurity by the interpolation of material from tragedies by Euripides and Sophocles. The razor-sharp intellect of Robert (1915:1.163–167) objected that, even if allowances were made for these alleged intrusions, the scholion remains as incoherent as before. We shall consider in a moment whether this is really so. Let us first register approval of Robert’s point (1915:1.152; misunderstood by Lloyd-Jones [2002:9 = 2003:29] when he writes “even Robert allows that the marriage with Euryganeia comes from the Oedipodeia”) that the mere mention of the name is no guarantee of epic origin (see page 22 below for other apparently independent testimonies which give this name to Oedipus’ wife). From the assumption that epic origin was so guaranteed sprang Bethe’s initial interpretation of the “Pisanderscholion” as a handy resumé of the Oedipodeia. The inadequacy of this approach must by now be plain.

But, of course, to exclude a given scholar’s theory about epic sources is not definitively to rule out any hypothesis concerning an epic source. This could be done if there were grounds for confidence in a totally incompatible hypothesis, such as Deubner’s notion that two tragedies by Euripides underlie Pisander’s narrative. A brief examination of this influential idea will not, therefore, be totally irrelevant to a study of the epic Oedipodeia. Deubner argued (1942:6 = 1982:640) that in order to reject the picture of an epic source for this and most portions of the scholion, one need not resort to Robert’s extreme interpretation of the whole as a confused mélange (attacked also by Lesky [1712.39–40], Jacoby [495], etc.). Robert was right to point to the inconcinnity involved in the clumsy change of subject at the climax of the narrative’s first half:

ἀπελθὼν τοίνυν ἐφονεύθη ἐν τῆι σχιστῆι ὁδῶι αὐτὸς (scil. Λαίος) καὶ ὁ ἡνίοχος αὐτοῦ, ἐπειδὴ ἔτυψε τῆι μάστιγι τὸν Οἰδίποδα. κτείνας δὲ αὐτοὺς [scil. Οἰδίπους] ἔθαψε παραυτίκα σὺv τοῖς ἱματίοις ἀποσπάσας τὸν ζωστῆρα καὶ τὸ ξίφος τοῦ Λαίου καὶ φορῶν κτλ.

But Deubner’s economic hypothesis (1942:7–9 = 1982:641–643) was that between the two sentences Pisander has changed his source, and that the two different sources are to be equated with Euripides’ Chrysippus and Oedipus. The former will have supplied all the information about Laius as πρῶτος εὑρετής {5|6} of homosexuality and Hera’s punishment, indeed everything down to ἔτυψε τῆι μάστιγι τὸν Οἰδίποδα, with the obvious exception of the digression on the Sphinx considered on page 4 above.

The second question posed above is rendered all the more difficult by the near impossibility of deciding on reliable sources for the reconstruction of these works. On Euripides’ Chrysippus see the bibliography offered by de Kock (1962:31n97), and now Kannicht (TrGF 5.2.877–879). De Kock himself (1962:31–36) has no great difficulty in arriving at a reconstruction of the plot which is remarkable for its almost total lack of any common ground with the “Pisanderscholion.” See further Mastronarde 33–34.

As for the Oedipus, here too scholars have disagreed over the details it will have contained. For a bibliography of recent attempts to reconstruct the play see D. Bain, “A Misunderstood Scene in Sophokles,” Greece and Rome 26 (1979): 145 = Greek Tragedy (Greece and Rome Studies 11 [1993]), 93n17; and now Kannicht, TrGF 5.1.569–570. The play’s most famous fragment (F 541 Kannicht), spoken by a θεράπων of Laius (ἡμεῖς δὲ Πολύβου παῖδ’ ἐρείσαντες πέδωι | ἐξομματοῦμεν καὶ διόλλυμεν κόρας), has no counterpart in anything Pisander tells us of Oedipus’ varied career, and Deubner does his case no good at all by seeking (1942:19 = {6|7} 1982:653) to declare this precious piece of evidence spurious. We must therefore ask ourselves whether Delcourt (Oedipe et la légende du conquérant 2 [Paris 1981], xviii) and de Kock (1962:22–23; cf. Lloyd-Jones 2002:8 = 2003:27) are not right to posit a deliberate anacoluthon or a minor lacuna as a simple, unmomentous solution for that awkward change of subject.

If we reject Bethe’s picture of the source as epic, and Deubner’s picture of the source as dramatic, and if we believe that Robert’s contemptuous dismissal of the scholion as a hopelessly confused mélange goes too far, then we have precious little room for maneuver. Perhaps de Kock’s hypothesis (1962:23–24) of a learned Hellenistic mythographer ingeniously and idiosyncratically combining older and newer motifs, some of them from drama, is not so far from the truth. We may, at any rate, heartily agree with his final conclusion on the passage of Pisander (1962:37): “the important deduction … is that, because we cannot determine all its sources with absolute certainty, we have no right to rely on it alone in our reconstruction of the Oedipodeia.

Let us now turn to the main features of the Oedipus story as familiar to us from later authors and see whether there is any chance of gauging the likelihood that they featured in our epic.

Components of the Story

The Rape of Chrysippus

Lloyd-Jones states the facts with memorable precision: “Robert showed that Bethe had not proved that the Oedipodeia used the Chrysippus story, but he did not show that it cannot have used it” (Justice of Zeus, 120; cf. his later, fuller, but less cautious treatment 2002:5–6 = 2003:24–25). Given our present state of knowledge, we have no hope of deciding either way. It is, however, striking that so many scholars (especially Robert 1915:1.157) should have so vehemently denied the very possibility of the motif’s occurrence in epic, and one cannot help concluding that prejudice rather than probabilities swayed their minds. The lack of any specific testimony concerning the legend before the time of Euripides’ Chrysippus is no very impressive argument. And the tendency of some myths to gain a homosexual coloring in later authors (especially the Alexandrian poets: cf. Kroll, RE s.v. “Knabenliebe,” 11 [1922]: 903; Dover, Greek Homosexuality, 199) does not entail that every legend containing such features must be late. Homosexuality is certainly absent from the world of the Homeric epics (cf. Dover 194 and 196–197; Griffin 1977:45 = 2001:378 and Homer on Life and Death [Oxford 1980] 104n4), and the absence is most easily recognized by the manner in which Hebe usually ousts Ganymedes as cupbearer to the gods {7|8} (on Iliad XX 231–235 cf. Dover 196). But the likeliest explanation of this state of affairs lies in the sphere of deliberate omission on aesthetic grounds rather than mere ignorance, and we are by now perfectly familiar with the process whereby features specifically excluded from Homer’s works reappear in later epics. It is therefore no surprise to find that the Homeric Hymn to Aphrodite 202–206 makes reference to Zeus’ passion for Ganymedes (see Faulkner ad loc. and cf. J. Th. Kakridis, “Die Pelopssage bei Pindar,” Philologus 85 [1930]: 463–474 = Μελέτες καὶ Ἄρθρα 55–63 = Pindaros und Bakchylides [Wege der Forschung 134 (1970): 175–190]), as does Ibycus fr. 289 ΡMGF; cf. Richardson’s commentary on the Homeric Hymn to Demeter (Oxford 1974) 279–280, and Wilkinson’s on the Ibycus fr. (pp. 253–258). The same lyric poet represents Rhadamanthus and Talos (Daedalus’ nephew) as homosexual lovers (fr. 309 PMGF). The adjectives applied to Haemon in F1 of the Oedipodeia are very suggestive in this context (Thebes punished for the abduction and death of one handsome youth by the abductions and deaths of countless handsome youths?). Given the likely motive (see page 6 above) for Chrysippus’ suicide, Lloyd-Jones (2002:5–6 = 2003:24) was probably right to argue that it is not unthinkable in early epic.

But if we have no evidence that the rape of Chrysippus was missing from the Oedipodeia, we have no evidence either of its presence. Wilamowitz (Hermes 59 [1924]: 270 = Kleine Schriften 4.363–364) thought he could produce the latter. In one of the letters of Julian the Apostate (80 [p. 97.19 Bidez-Cumont]) the paradosis runs ὥσπερ ἐξ ἁμάξης εἰπεῖν οἷα ψευδῶς ἐπὶ τοῦ? Λαυδακίδου Ἀρχίλοχος. For the corruption in the penultimate word Weil conjectured Λυκάμβου and this correction has been widely accepted (e.g., by M. L. West, who prints it in his edition of Archilochus [p. 64] among the testimonia for the epode dealing with the eagle and the fox with no indication that it is a conjecture). Wilamowitz, however, suggested a different remedy: Λαβδακίδου (i.e., Laius), with an insinuation of pederasty against Julian’s acquaintance Lauricius. If Archilochus did indeed bring such a charge against his victim, he must have derived it from an earlier “Theban epic” (unknown to Julian, who therefore calls Archilochus’ charge “false”). This conjecture hardly involves an alteration (for the interchangability of β and υ in manuscripts see Aristotle Rhetoric 3.14.1415B38 [αὐτοκάβδαλα Α, -καυδαλα β]; Theocritus Idyll 5.109 and Gow ad loc.), but the hypothesis of derivation from an earlier Theban epic does not necessarily follow, and the question of an epic origin for the rape of Chrysippus must remain open. [5] {8|9}

The Oracle to Laius on the Consequences of Begetting a Son

On the sources and likeliest origin of this see Fontenrose, The Delphic Oracle (Berkeley 1978) 96–98 and 362–363. It might at first be thought that our views as to its presence or absence in the epic must depend on what we think of Chrysippus’ presence or absence. Wehrli (1957:110 = 1972:63), however, though refusing that figure to our epic, is prepared to countenance an oracle which warned a guiltless Laius of the consequences of begetting a child. He compares the utterance to the blameless Croesus concerning his projected war on Persia (see Fontenrose 302) and the famous “son of Thetis” prophecy.

The Exposure of Oedipus

As Rzach stresses (1922:2360.50–62), our ignorance of the epic’s contents is so comprehensive that we cannot tell whether the version familiar from tragedy was used (exposure on Mount Cithaeron) or the variant preserved in Σ Euripides Phoenician Women 28 (1.252 Schwartz): τινὲς δὲ ἐν λάρνακι βληθέντα καὶ εἰς θάλασσαν ῥιφέντα τòν παῖδα προσπελασθῆναι τῆι Kορίνθωι φασίν. Both motifs are primitive and popular: on the former see Stith Thomson, Motif-Index R 131, S 301, on the latter S 141, S 331; Bethe 1892:72–73; A. B. Cook, Zeus (Cambridge 1925) 2.671–673; N. M. Holley, “The Floating Chest,” Journal of Hellenic Studies 69 (1949): 39–47; G. Binder, EM s.v. “Danae” (3.264), and more fully in Die Aussetzung des Königskindes, 142–144 and index s.v. “Ödipusmythos.”

The Parricide

On the general question of this episode’s connection with the Oedipus legend see Edmunds 1981a:47–48. Mastronarde (34–35 and n2) observes that one feature of the Pisander scholion which looks relatively early is its location of Oedipus’ killing of Laius on Mount Cithaeron, in contrast to the later almost-universal placing of it at the crossroads in Phocis. Lloyd-Jones (2002:7 = 2003:26) thinks Oedipus’ exposure on Cithaeron may be from the Oedipodeia, and notes (9 = 28) the existence of a cult of Hera (see page 4 above) on the self-same mountain. See further Fowler 2013:403.

The Sphinx’s Riddle

On the connection with the Oedipus legend see Edmunds 1981a, esp. 18–21; on the possibility of its appearance in epic, Lesky, RE 3A (1928), s.v. “Sphinx,” {9|10} 1711–1712, Mitteilungen des Vereines Klassischer Philologen im Wien 5 (1928): 3–12 = Gesammelte Schriften 318–326; Lloyd-Jones, Dionysiaca (Page Festschrift [Cambridge 1978]) 60–61 = Academic Papers [I] 332–334; Petit, Oedipe et le Chérubin, passim). On the scene in art see Simon 1981:12–70; K. Schauenberg in Praestant Interna (Hausmann Festschrift [Tübingen 1982]) 230–235 (bibliography in 230nn1–2); LIMC VII.1 V 3–9.

The riddle is reported in differing forms by different authors. I record here the text printed by Lloyd-Jones, to whom the reader is referred for details as to its sources and the significant verses (see too Edmunds 1981a:32n16):

ἔστι δίπουν ἐπὶ γῆς καὶ τετράπον, οὗ μία μορϕή,
καὶ τρίπον, ἀλλάσσει δὲ φυὴν μόνον ὅσσ’ ἐπὶ γαῖαν
ἑρπετὰ κινεῖται καὶ ἀν’ αἰθέρα καὶ κατὰ πόντον.
ἀλλ’ ὁπόταν τρισσοῖσιν ἐπειγόμενον ποσὶ βαίνηι,
ἔνθα τάχος γυίοισιν ἀϕαυρότατον πέλει αὐτοῦ.

The particular question that concerns us is whether Robert was right to suggest (1915:1.56–57 and 168) that the lines (or something like them) emanate from an early epic such as the Oedipodeia or the Thebais. The original publication of a papyrus fragment belonging to Euripides’ Oedipus (TrGF 5.1.573; cf. Edmunds 1981a:33n22) led Lloyd-Jones (Gnomon 35 [1963]: 447) to suppose that Robert’s thesis had been strengthened: the fragment showed that Euripides used a different version of the riddle from that cited above: this latter must then have possessed considerably more authority than the tragedian’s for it to survive so long and to be quoted by so many later authors. In the later treatment (from 1978), just cited, Lloyd-Jones displayed considerably more skepticism, largely due to his acquaintance with the second of Lesky’s articles.

In his second article, after decisively establishing that the Oedipodeia is the only early epic in which the riddle could possibly claim to be both totally relevant and appropriate, Lesky turned to the disproving of Robert’s thesis as applied to this poem. His counterarguments fall under the headings of the general and the specific, and within the former category belong his attempts to demonstrate that in the Oedipodeia the Sphinx does not yet seem to have featured as a poser of riddles: rather she was a mere brutal murderer. Intrinsic to this whole argument, of course, are the assumptions that this is exactly how the Sphinx operated in the earliest form of the legend of Oedipus’ encounter with her, and that the familiar riddle version is a later intellectualizing refinement. Such a reconstruction has found favor with several scholars (especially Edmunds [1981a:18]), although few have expressed the view with the force and clarity that Lesky devoted to it in his RE article (1716–1717). His argumentation there rests on two types of reasoning. {10|11} First is the a priori intuition that in legends involving heroes, brute force must logically precede the more sophisticated employment of cunning and guile, so that a more straightforward version in which a normal unintellectual monster is crushed by strength of arm must be presupposed by the extant story of the riddle and a battle of wits to solve it. Second is the evidence of an often-cited lekythos now in Boston (97.374: LIMC VII.1 B2.78, s.v. “Oidipous”), which shows a naked man labeled as Oedipus wielding a club against the Sphinx, for all the world as if he were Heracles. The image on this vase can now be supplemented by several other artifacts (useful list and bibliography: Edmunds 1981a:35n37) [6] similarly suggestive of a tradition wherein Oedipus killed the Sphinx in straightforward manner, with sword or spear. [7]

The a priori arguments seem to me altogether too crude and simplistic. It would be one thing to insist that stories where heroes win through by strength and might are earlier in kind than stories where the hero relies on his wits; it is a quite different proposition to claim that all examples of the latter originally ran along the lines of the former, as if Odysseus’ cunning escape from the Cyclops’ den was originally effected by sheer brute strength! And indeed tales of cunning and stratagem seem basic to mankind in general (note in particular M. Detienne and J.-P. Vernant, Les ruses de l’intelligence: La metis des grecs [Paris 1974] = Cunning Intelligence in Greek Culture and Society [Sussex 1978] passim). The riddle in particular as a motif in folk-literature is both primitive and widespread: see Stith Thompson, Motif-Index 6 s.v. “Riddles: Guessing with life as wager,” and especially the entries s.v. Η 512, 541.1, and 541.1.1; H. Fischer’s article s.v. “Rätsel” in EM 11.267–275; Edmunds 1981a:5–12; and M. L. West, Indo-European Poetry and Myth (Oxford 2007) 281.

There is more than a little to be said in favor of the hypothesis that the riddle was ab initio and always thereafter connected with the Sphinx (so, for instance, O. Crusius, Literarische Zeitungsblätte (1892) 1699; Walter 69–70). That does not entail, of course, that the version of the riddle now under discussion appeared in the Oedipodeia. Lesky’s specific arguments against this possibility are more convincing than the general ones just examined. At least his linguistic observations seem irrefutable: verse 2’s ἀλλάσσει is alien to epic (the word first occurs in Theognis 21; on ἐπαλλάξαντες in Iliad XIII 359 see LfgrE 1 col. 535), and τάχος … ἀφαυρότατον in verse 5 presupposes for the adjective a sense unexampled in Homer (whence, presumably, the variant μένος … ἀφ. offered by some authors). The use of hexameters is no necessary index of epic origin: see the fragment of Euripides’ Oedipus mentioned on page 10 above and Radt, TrGF 4.237 (on F190). If Lesky is right, we need not seek (with Rzach [1922:2358–2358] and several other scholars: cf. U. Hausmann, Jahrbuch der Staatliche Kunstsammlungen in Baden-Würtemberg 9 [1972]: 20) to establish a connection between the Oedipodeia and the famous Vatican cup (Vatican Η 569: ARV 2 451 = LIMC VII.1 Vb.19, s.v. “Oidipous”; cf. Simon 1981:28–31 and plate 15) redated ca. 470 by Beazley ap. Fraenkel on Aeschylus Agamemnon 1258 (and in ARV 2 as cited), which shows the Sphinx addressing Oedipus from her column and beside her the words (κ)αὶ τρί(πουν) which begin verse 2 of the riddle. Simon (1981:30–31) revives the old idea that the vases may reflect Aeschylus’ Sphinx. Other attempts to decide the issue either way are unsatisfactory. West’s observation (on Hesiod Works and Days 533) that if the riddle “had come in the epic Oedipodeia, Athenaeus might have been expected to quote it from there instead of from Asclepiades FGrHist 12 F7” is not particularly convincing. Scholars have shown some skepticism about accepting the one fragment Athenaeus cites from the Thebais as evidence of direct knowledge of the original poem. In the case of the Oedipodeia there are no grounds whatsoever for thinking that Athenaeus had read the epic. (West has since changed his mind and prints the riddle as a fragment of our poem as preserved by Asclepiades on p. 40 of the Loeb FGE [2003].) Although the vase Rzach cited is no evidence for the supposition, he may still have been right (1922:2358.52–53) to suppose that our epic mentioned the Sphinx’s riddle.

In spite of the above uncertainties as to the Sphinx’s role, it seems safe enough to follow the vast majority of scholars in deducing from F1 (see page 18 below) that the hand of the queen (Oedipus’ own mother) was the reward for the Sphinx’s conqueror in our epic, together with royal rule. For this popular {12|13} folk-tale motif see in particular Stith Thompson, Motif-Index 6 Τ 68, Wehrli 1957:113 = 1972:66n28.

The Dénouement and Its Consequences

μητέρα τ᾽ Οἰδιπόδαo ἴδον, καλὴν Ἐπικάστην,
ἣ μέγα ἔργον ἔρεξεν ἀϊδρείηισι νόοιo,
γημαμένη ὧι υἷϊ, ὁ δ’ ὃν πατέρ’ ἐξεναρίξας
γῆμεν· ἄφαρ δ’ ἀνάπυστα θεοὶ θέσαν ἀνθρώποισιν.
ἀλλ’ ὁ μὲν ἐν Θήβηι πολυηράτωι ἄλγεα πάσχων
Καδμείων ἤνασσε θεῶν ὀλοὰς διὰ βουλάς.
ἡ δ’ ἔβη εἰς Ἀΐδαο πυλάρταο κρατεροῖο,
ἁψαμένη βρόχον αἰπὺν ἀφ’ ὑψηλοῖο μελάθρου,
ὧι ἄχεϊ σχομένη. τῶι δ’ ἄλγεα κάλλιπ’ ὀπίσσω
πολλὰ μάλ’, ὅσσα τε μητρὸς Ἐρινύες ἐκτελέουσι.

Odyssey xi 271–280

Ever since Welcker (1865:2.313–314), many scholars have believed that the above lines provide, in effect, a handy summary of the latter part of the Oedipodeia. For a bibliography see Deubner (1942:34 = 1982:668n2), who himself advances further arguments in favor of the hypothesis. He has convinced many sober scholars even in recent times. Thus we find Griffin writing (1977:44n32 = 2001:375n38): “It was pointed out in antiquity (Pausanias IX 5.10) that the word ἄφαρ seems to rule out the production of children. This is the more striking as it has been shown by Deubner … that this passage of the Odyssey is based on the version of the cyclic Oedipodeia, in which Oedipus had by her [scil. his mother] two sons, Phrastor and Laonytus.” But it is my contention that Deubner has shown nothing of the sort, and that much of Griffin’s article merely underlines the implausibility of the hypothesis he here accepts. Griffin himself (1977:44 = 2002:375) has established that the Odyssean episode takes its place within a series of passages where Homer has sought to eliminate grisly details of family murder and strife. He instances as analogous the omission of the tale of Iphigenia’s sacrifice by her father and “the silence in the Odyssey about the way in which Clytemnestra died.” [

In the present case the parricide and union with the mother are so basic to the story that they must be accorded a mention. But this mention is of the briefest, and nothing is said of Oedipus’ self-blinding or of the unhappy children {13|14} born of the incestuous relationship. With the latter omission we might compare Homer’s refusal to bless the guilty liaison of Paris and Helen with children (see Griffin 1977:43 = 2002:373). Scholars are becoming gradually more willing to accept that idiosyncratic mythological details in Homer are much likelier to be the product of the poet’s innovation in or reworking of myth than to represent an accurate and painstaking summary of some now lost epic for which they are a valuable source of information. This is true of the story of Meleager as it appears in Iliad 9 (see Davies and Finglass on Stesichorus fr. 183). I believe it is also true of the story of Oedipus in Odyssey xi.

No unbiased reader of Odyssey xi 275–280 would for one moment conclude that they were in any way compatible with (let alone suggestive of) a version in which, after all the dreadful revelations, Oedipus calmly proceeded to take a second wife and father four children upon her. But Deubner has not yet finished with the Odyssean passage: far from being content with reading into these lines the birth of four infants after the great dénouement, he goes on to extrapolate the birth of two infants before it!

Deubner, in common with many scholars, assumes that Pausanias can be trusted in his remarks upon the Oedipodeia’s presentation of the facts pertaining to its hero’s married life. I myself prefer to follow Robert and others (see page 21n17 below) in supposing Pausanias to be guilty of a fairly elementary blunder in the matter of Euryganeia’s identity (see page 21 below). That there is considerable danger in attributing to Pausanias an error of this kind in connection with a poem that has now vanished almost without trace I do not deny. But those who accept Deubner’s views here are in no position to throw stones. Or are they unable to detect the inconsistency inherent in trusting Pausanias without demur when he is talking of a lost epic, while faulting him twice over in connection with an extant one? For if Deubner is right to suppose that Jocasta/Epicaste bore Phrastor and Laonytus to her son in the Oedipodeia, and if he is further right in conjecturing that the Odyssean lines are based on the Oedipodeia, then Pausanias must be doubly wrong, both in his particular interpretation of the word ἄφαρ, and in his general deduction that the Odyssey knew of no children begotten by Oedipus upon his mother.

Deubner (1942:36 = 1982:670) pleads for an “elastic” interpretation of ἄφαρ here and in the allegedly analogous instances at Odyssey ii 95 and 169 and in Homeric Hymn to Demeter 454. His claim will not survive a reading of the excellent article on this word that Führer has contributed to LfrE. ἄφαρ can indeed have a nontemporal signification and it is under this heading (“2 modal: in der Tat, wirklich, schon”) that Odyssey ii 169 appears (1697.70–72). The other Odyssean passage is ranked under subsection Id (“sogleich”) at 1697.35–37 and Homeric Hymn to Demeter 454 under Ia 1696.15–16 (“erstaunlich, schnell”). These are all {14|15} accepted meanings for the word and so is that assigned to the passage under discussion (Ib: “schnell, gar bald” [1696.53–57]), with a reference to the implied absence of incestuous offspring in the version here followed. Deubner’s claim—that the word need not exclude an interval of a year between marriage and the emergence of the truth—is most decidedly to be rejected. See further Fowler 2013:404n25 against other ancient and modern misreadings of the relevant Greek word.

And who, after all, are the two sons Phrastor and Laonytus whom Deubner wishes to have brought into the world during this year’s interval? Why, they figure in Pherecydes FGrHist 3 F95:

Oἰδίποδι (φησί) Κρέων δίδωσι τὴν βασιλείαν καὶ τὴν γυναῖκα Λαίου, μητέρα δ’ αὐτοῦ Ἰοκάστην, ἐξ ἧς γίνονται αὐτῶι Φράστωρ καὶ Λαόνυτος, οἳ θνήισκουσιν ὑπὸ Μινυῶν καὶ Ἐργίνου< lac. stat. Jacoby>. ἐπεὶ δὲ ἐνιαυτὸς παρῆλθε, γαμεῖ ὁ Οἰδίπους Eὐρυγάνειαν τὴν Περίφαντος, ἐξ ἧς γίνονται αὐτῶι Ἀντιγόνη καὶ Ἰσμήνη, ἣν ἀναιρεῖ Tυδεὺς ἐπὶ κρήνης, καὶ ἀπ’ αὐτῆς ἡ κρήνη Ἰσμήνη καλεῖται. υἱοὶ δὲ αὐτῶι ἐξ αὐτῆς Ἐτεοκλῆς καὶ Πολυνείκης. ἐπεὶ δὲ Eὐρυγάνεια ἐτελεύτησε, γαμεῖ ὁ Οἰδίπους Ἀστυμέδουσαν τὴν Σθενέλου.

These issues are vitally important, for several scholars have drawn conclusions crucial for the character of our epic (and for the “evolution” of the Oedipus legend) from the identification of the traditions supposedly represented by Odyssey xi, Pherecydes, and Σ A Iliad IV 376 with what once stood in the Oedipodeia. Thus Deubner (1942:38 = 1982:672) infers an “epic” tradition in which Oedipus figured as a much more robust character than his counterpart in Attic tragedy: he does not blind himself, he can bring himself to marry again, he continues to rule over the Thebans, and he finally dies in battle. This is an archaic, and therefore the oldest and original, presentation of the hero. Likewise de Kock, who on other topics arrives at conclusions that drastically disagree with Deubner’s, claims (1961:16–17) that in the Oedipodeia “we find ourselves in a world completely different from that of the tragedy,” and that this epic presents us with “a hero who is clearly not deeply affected by the effects of patricide and mother marriage.” If this last remark were true it would be a remarkable epic indeed, and a remarkable mental attitude to incest and murder within the family, very important for studies of the development of Greek morality. Similarly now Fowler 2013:404–405: “the incestuous offspring [were] germane in tragedy, but absent from epic … the tragedians … raised the level of horror.” {16|17} And so it is worth stressing that this picture is largely based upon the very economic summary in the Nekyia from which it would be unreasonable to expect a detailed account. In such a context, “Homeric decorum” (Mastronarde p. 21) will have found it easy (and congenial) to skirt the awkward and fearful questions of incestuous children and Oedipus’ guilt. It is hard to see how an epic whose very title implies a detailed account of the career and suffering of the hero could ever have similarly avoided these basic issues.


F1 (see page 133 for text)

On the plurals in οἱ … γράφοντες see page 1 above. The phrase which in our MSS immediately precedes the quotation (oὐδεὶς οὕτω φησὶ περὶ τῆς Σφιγγός) has been variously emended or deleted. Vian’s suggestion οἱ τὴν Oἰδιποδείαν γράφοντες, οἵτινές εἰσιν, made in 1963:207n5, takes its inspiration from the idioms used in quoting such epics of uncertain authorship as the Titanomachy (T2) and the Cypria (F7). If the epic’s allusion to the Sphinx has dropped out, then the disputed phrase requires no remedy, since for all we know the sequel to the two extant lines presented the Sphinx in a unique way. The unparalleled nature of this epic’s treatment of Haemon as victim of the Sphinx is stressed by Vian (1963:207–208), and this may be what our scholion originally intended to convey.

It would perhaps be misleading to suggest that there is anything strictly and literally unique about the epic’s presentation of Haemon as a victim of the Sphinx. This is a tradition which recurs in at least three other authors. We have already encountered it in “Pisander” (see page 4 above), where it is said of the Sphinx that she ἀναρπάζουσα μικροὺς καὶ μεγάλους κατήσθιεν ἐν οἷς καὶ Αἵμονα τὸν Κρέοντος παῖδα. A more detailed description of the exact circumstances of Haemon’s destruction is to be found in Apollodorus III 5.8:

χρησμοῦ δὲ Θηβαίοις ὑπάρχοντος τηνικαῦτα ἀπαλλαγήσεσθαι τῆς Σφιγγὸς ἡνίκα ἂν τὸ αἴνιγμα λύσωσι, συνιόντες εἰς ταὐτὸ πολλάκις ἐζήτουν τί τὸ λεγόμενόν ἐστιν, ἐπεὶ δὲ μὴ εὕρισκον, ἁρπάσασα ἕνα {18|19} κατεβίβρωσκε. πολλῶν δὲ ἀπολομένων, καὶ τὸ τελευταῖον Αἵμονος τοῦ Κρέοντος, κηρύσσει Κρέων τῶι τὸ αἴνιγμα λύσoντι καὶ τὴν βασιλείαν καὶ τὴν Λαΐου δώσειν γυναῖκα.

A similar picture of Theban deliberations, but without the detail of Haemon’s death in Σ Euripides Phoenician Women 45 (1.255 Schwartz) = Asclepiades FGrHist 12 F7B. These passages may all, as Vian (1963:207–208) suggests, derive from the Oedipodeia.

Vases often depict the Sphinx carrying off a youthful male victim which clings beneath her belly (see LIMC VIII.1 IVB, s.v. “Sphinx,” esp. 3–4 [p. 1161], Simon 1981:16, K. Schauenberg, in Praestant Interna [U. Hausmann Festschrift (1982)] 232, and Petit, Oedipe et le Chérubin, 123–124 on “le sphinx ravisseur”). The late black-figure lekythos painter who repeats this subject four times was therefore awarded the title of the “Haemon Painter” by E. Haspels, Attic Black-Figured Lekythoi (Paris 1936) 130–141. The name is convenient and probably harmless, so long as we remember the warning delivered by its inventor (130n3): “I do not wish to imply … that the victim of the Sphinx on our vases is necessarily” Haemon or, indeed, any other definite and specific person. Most scholars have assumed that in the Oedipodeia, as in Apollodorus, Haemon was the last of the Sphinx’s victims. The ἔτι and the superlatives in the first line of our fragment are consistent with, though not in themselves indicative of, such a hypothesis. However, as Lesky stresses (see page 10 above), the very logic of the story also points in this direction: the death of the Theban regent’s son and heir finally creates an overwhelming crisis concerning the city’s future and Creon is forced to adopt the expedient (so familiar in folk tale: see page 13 above) of offering the kingdom to whatever stranger shall rescue the city. Haemon’s death, Creon’s proclamation, and Oedipus’ success must follow closely on each other’s heels for the story to work. So too concludes Vian (1963:206).

Whether the Oedipodeia placed the death of Haemon in the same circumstances as those reported by Apollodorus we have no means of knowing. I do not accept, however, that Lesky has decisively excluded any link between the riddle and this epic in general or our fragment in particular. As part of his attempt to establish that the riddle cannot have featured in the Oedipodeia (see page 11 above), this scholar argued that Asclepiades’ version of events can have nothing to do with the epic and must be a relatively late attempt to reconcile the familiar picture of the riddling Sphinx with the alleged earlier picture of the Sphinx as a normal ravening monster. Inevitably the same objections must apply to Apollodorus’ version, which inserts Haemon’s death into the background of events established by Asclepiades. {19|20}

I have already explained my reluctance to accept the popular reconstruction of the “original” form of Oedipus’ encounter with the Sphinx (page 12 above). In the present case, one must add that Lesky’s approach leaves no room for any early version linking the riddle with the death of Haemon (or, indeed, any other Theban). Why such a version should be excluded on principle I cannot see: granted the possibility of the coexistence of a riddling Sphinx and Haemon’s death at her hands, Apollodorus’ is the only conceivable way of combining the two motifs, and there seems to me to be nothing obviously absurd or redolent of a late compromise in his version of events. The famous Hermonax pelike (Vienna 3728: ARV 2 1.485.24 = LIMC VII.1 VA2.490, s.v. “Oidipous”), which shows eleven Theban elders deliberating over the riddle while the Sphinx looks on ominously, perched upon a pillar, has therefore been taken to derive from our epic (Robert 1915:1.168, Rzach 1922:2358.22–28, etc.; cf. Petit, Oedipe et le Chérubin, 127 on “le sphinx à la colonne”). Although unconfirmable, the possibility should not be excluded as categorically as it is by Lesky.

1 . κάλλιστόν τε καὶ ἱµεροέστατον ἄλλων: the language seems surprisingly erotic for epic: cf. Theogn. 1117 = 1365: (Πλοῦτε, θεῶν | ὦ παίδων) κάλλιστε καὶ ἱμεροέστατε πάντων (cf. Ibyc. S 173.7 PMGF ]ιστε παιδῶν [(where Page suggested κάλλ]ιστε). This may be connected with the original conception behind the Sphinx’s addiction to the snatching up of young men: see Vian 1963:206–207. There is no call to suppose (with e.g. Küllenberg, de imitatione Theognidea (1877) 23, van Groningen on Theogn. 1117) that our epic is the inspiration for the verses of Theognis cited above. As Wilamowitz observed (Sappho und Simonides 120n1, followed by J. Kroll, Theognisinterpretationen (1936) 7n18 and Vetta on Theogn. 1365) the Oedipodeia’s use of accusatives “ist ganz schlecht aus dem Vocativ … gemacht: hier ist das Epos jünger, nicht notwendig als diese Verse, aber wohl als diese Wendung in einem erotischen Trinkverse.” ἱµεροέστατον: the LSJ entry s.v. ἱμερόεις is deficient and erroneous: the word is only used of things by Homer (ἀοιδή, ἔπεα, ἔργα, etc.). Hesiod applies it to females: Th. 359 (of Calypso as in HHDem 422) and fr. 291.3 MW. LSJ should not have cited Pind. fr. 33c (= 87) 2 Sn. as an instance of its application to persons (thereby misleading e.g. Page, Sappho & Alcaeus 59 and Gow, Theocr. 2. Addenda [p. 592]): the reference there is to the island of Delos. The only other early example besides the present where it is attached to a young man has a decidedly homoerotic tinge: Theogn. 1365 (cited above). Note, however, Wilamowitz’s popular supplement at Sappho fr. 17.10: Θυώνας ἰμε[ρόεντα παῖδα: see Page as cited. For its use (often homosexual) of persons in Alexandrian authors see Theocr. Id. VII 118 and Gow ad loc., Kost on Musaeus Hero and Leander 20. [16] On vase-paintings which depict the Sphinx snatching up desirable young men see page 19 above. The superlative plus genitive construction is guaranteed by the two Theognidean verses cited above. It is fairly common in Homer: see Iliad I 505–506 ὠκυμορώτατος ἄλλων | ἔπλετ’. For further examples and discussion see Schwyzer, Gr. Gr. 2.116n1 and Chantraine, Gramm. hom. 2.60.

2 . Κρείοντος ἀµύµονος: on the meaning of the epithet here see A. A. Parry, {20|21} Blameless Aegisthus (Mnemos. Suppl. 26 [1973] 78) who argues that “in view of the emphasis on [Creon’s] son’s beauty, the reference is surely to Creon’s looks.” Non sequitur.

On Creon’s role in Theban myth see Vian 1963:183–193; on his son’s, 206–208.

F2 (see page 134 for text)

On ancient traditions regarding Oedipus’ wife/wives see in general Fowler 2013:403–408. On attempts to amplify our fragment by reference to Pherecydes FGrHist 3 F95 and Σ A Iliad IV 376 see pages 15–16 above. It is important to reconstruct the stages in Pausanias’ argument here, leading up to the mention of the Oedipodeia’s version: (1) Pausanias takes Odyssey xi 271–280 to entail that Epicaste/Jocasta bore Oedipus no children. (2) Pausanias’ reason for this inference is that in these lines Oedipus’ murder of his father and marriage to his mother are followed shortly (ἄφαρ) by the gods’ disclosure of those deeds. (3) Therefore Epicaste/Jocasta can have had no time before the disclosure to bear to Oedipus as many as four children. (4) Rather, the four children were begotten by Oedipus upon an entirely different woman, to wit Euryganeia. (5) And this latter personage is alluded to in the Oedipodeia. (For the use of δέ here to answer a question [πῶς οὖν … τῶι Oἰδίποδι; ἐξ Eὐρυγανείας δὲ τῆς Ὑπέρφαντος ἐγεγόνεσαν] see Denniston, GP 2 171 [ii b].)

This is just as well, for if the idea of Euryganeia as the sole mother and wife of Oedipus has the support of those ancient authors quoted above, the alternative interpretation of her as a second spouse whose union with Oedipus is quite free from incest was obviously widespread in antiquity. We have already seen the testimony offered at the end of ‘Pisander’ ’s narrative:

φασὶ δὲ ὅτι μετὰ τὸν θάνατον τῆς Ἰοκάστης καὶ τὴν αὑτοῦ τύφλωσιν ἔγημεν Εὐρυγάνην παρθένον, ἐξ ἧς αὐτῶι γεγόνασιν οἱ τέσσαρες παῖδες.

Pherecydes too (FGrHist 3 F95) has a similar tale:

Οἰδίποδι (φησί) Κρέων δίδωσι τὴν βασιλείαν καὶ τὴν γυναῖκα Λαίου, μητέρα δ’ αὐτοῦ Ἰοκάστην, ἐξ ἧς γίνονται αὐτῶι Φράστωρ καὶ Λαόνυτος, οἱ θνήισκουσιν ὑπὸ Μινυῶν καὶ Ἐργίνου <lac. stat. Jacoby>. ἐπεὶ δὲ ἐνιαυτὸς παρῆλθε, γαμεῖ ὁ Οἰδίπους Eὐρυγάνειαν τὴν Περίφαντος, ἐξ ἧς γίνονται αὐτῶι Ἀντιγόνη καὶ Ἰσμήνη, ἣν ἀναιρεῖ Τυδεύς ἐπὶ κρήνης … υἱοὶ δὲ αὐτῶι ἐξ αὐτῆς Ἐτεοκλῆς καὶ Πολυνείκης. ἐπεὶ δὲ Εὐρυγάνεια ἐτελεύτησε, γαμεῖ ὁ Οἰδίπους Ἀστυμέδουσαν τὴν Σθενέλου.

However, the value of the former is highly dubious (see pages 4–7 above). As for the latter, even Jacoby (ad loc., p. 416), who refuses to accept Robert’s approach to the present epic fragment, is obliged to admit of the Pherecydean version “kontaminiert hat Ph. sehr naiv, indem er aus den verschiedenen namen für die muttergattin eine reihe von ehen machte.” (Fowler ad loc. [2013:406] is a little more reluctant to accuse Pherecydes of “such an elementary lack of understanding.”) Stephanopoulos (1980:105–106) has tried to use Pherecydes’ fragment as a means to refute Robert’s interpretation of Pausanias’ words. But his first argument merely terminates in the conclusion that Oedipodeia and Thebais {22|23} are unlikely both to have called Oedipus’ wife and mother Euryganeia. Since this is no essential part of Robert’s reading of the present fragment, little is achieved by the denial. The second argument asks why, on Robert’s reckoning, Pherecydes should make Euryganeia, rather than Jocasta, bear Oedipus the four famous children. But nothing is achieved either by wondering why Pherecydes chose this rather than that form of contamination. The proliferation of extra children and wives for Oedipus is reminiscent of the way in which late authors devise increasingly numerous husbands for Helen, and increasingly numerous offspring for her and for Menelaus (cf. Griffin 1977:43 = 2001:373). Robert’s argument is at its strongest when it concerns itself with the basic significance of the Oedipus story as a whole. It is instructive to pose the question “Can we imagine the Sophoclean Oedipus marrying again?,” even though whether the epic hero begot children on his mother like his counterpart in tragedy is precisely what we are disputing, and it is equally unproven, in fact, that epic’s Oedipus blinded himself. Nevertheless, the dread and terror of the original myth surely derive from the fact that the hero marries his own mother and has children by her. The grimness of his and his offspring’s dilemma is absurdly diluted if their mother is not his too, and if Oedipus proceeds to behave like a Tacitean Claudius caelibis vitae intolerans. The introduction of additional and normal wives, for Oedipus to have normal children by, looks very like a later attempt to purge the story of some of its horror. [

An analogous effect is aimed at in the brief narrative of Odyssey xi 275–280, where Oedipus is implicitly denied any offspring by Epicaste. But there the legend is only referred to elliptically and in passing and the elimination of children is quite in the manner of the Homeric epics, with their notorious aversion for grim tales of strife within the family. On the whole, the other early epics seem to have differed from Homer’s in this respect (see Griffin’s article passim). They showed little inclination to omit such horrors and had little opportunity to do so since these horrors were often basic to the plot. As soon as we have any specific information about epic’s presentation of Oedipus (witness the two curses he delivers in the Thebais) we are able to recognize a figure not so very different from the character familiar to us from Attic tragedy (see page 60 below).

One must not dogmatize as to the contents of epics which have vanished with such approximation to totality as the Oedipodeia. Nevertheless, I find it next to impossible to believe that an epic ever existed in which Polyneices and Eteocles lived quietly unexceptionable lives and the Theban War did not take place. One would require more direct and convincing testimony for so remarkable a scheme of events than is afforded by Pausanias. It may be objected that the fortunes of Polyneices and Eteocles lay outside the scope of the Oedipodeia. The objection is inept: we do not know at what point in the saga the epic closed, and even if it did exclude this particular area, its composer must have drawn upon some tradition which will have supplied him with ideas about the fate of Oedipus’ sons.

Suppose, however, that the Oedipodeia did lack the motif of incestuous offspring, but still envisaged the father’s curse upon his sons and their death at Thebes. Difficulties still arise. As we shall see, the vast majority of scholars suppose that the brothers featured as incestuously begotten in the Thebais. Is it really plausible that so infinitely more compact and logical a schema should have occurred to one epic poet but not the other?

Onasias’ Painting: LIMC VII.1s.v. “Septem” I.3 (p. 710)

I have so far postponed consideration of this important (but no longer extant) artifact which Pausanias mentions immediately after his reference to F2 of the Oedipodeia. “A grief-stricken Euryganeia at the battle of her sons” is compatible with either interpretation of the afore-mentioned fragment. It fits perfectly, of course, with the idea that Euryganeia was Oedipus’ second wife, not his mother, and bore him four children. But it is equally consistent with the notion that Euryganeia is merely another name for the mother of Oedipus upon whom he begets incestuous offspring. In this latter case we must suppose that Onasias was following a tradition whereby this woman survives the catastrophic {24|25} revelations like Stesichorus’ queen (see Davies and Finglass on fr. 97 of that poet) or Euripides’ Jocasta (in the Phoenician Women); unlike Homer’s Epicaste (Odyssey xi 271–280) or Sophocles’ Jocasta (in the Oedipus Rex). How a painter could have signified which of the two (second wife or surviving first wife) he was actually depicting is not very clear.

Should we therefore agree with, for example, Bethe and Stephanopoulos (see the latter 1980:107 and n12) that Onasias’ painting reflected a scene described in the Oedipodeia? P. Corssen (Die Antigone des Sophokles [Berlin 1898] 22 [followed by Rzach (1922:2359.30–53)]) had already argued that, if the Oedipodeia did indeed contain such a scene, Pausanias (or his source) would have continued to cite it, rather than turning to Onasias as if he supplied a detail missing from the epic. Besides, we should not be in too much of a hurry to assume that the Oedipodeia closed at so relatively late a stage in the story and included so much of matter that must also have occurred in the Thebais.

Εὐρυ-γάνεια as daughter of Ὑπέρ-φας is discussed by M. Sulzberger in “ΟΝΟΜΑ et ΠΡΑΓΜΑ: Les noms propres chez Homère et dans la mythologie grecque,” Révue des Études Grecques 39 (1926): 395, as an example of the widespread tendency, in early myth and epic in particular, for minor characters to have parents with similar names. He compares Καλητώρ son of Κλυτίος in Iliad XV 419. In fact, it is the variation in the form of both daughter’s and father’s name that provides most of the problems from this stage onwards. Let us put behind us the controversy over the number of Oedipus’ wives and concentrate on nomenclature. Apollodorus III 5.8 clearly entails precisely the same tradition as our fragment: εἰσι δὲ οἳ γεννηθῆναι τὰ τέκνα φασὶν ἐξ Εὐρυγανείας αὐτῶι τῆς Ὑπέρφαντος (Aegius: Tεύθραντος). Cf. Σ Euripides Phoenician Women 13 (1.249 Schwartz): καὶ τὸν Οἰδίποδα δέ φασιν Ἐπικάστην τε τὴν μητέρα γεγαμηκέναι καὶ Εὐρυγάνην. This version of the name recurs in “Pisander”: ἔγημεν Εὐρυγάνην παρθένον. For the alternative forms compare Εὐρώπεια / Εὐρώπη and see page 1 above. {25|26} Hyperphas’ name too seems capable of metamorphosis. At least, Ὑπέρφας is presumably to be equated with the Περίφας whom Pherecydes FGrHist 3 F95 makes father of Εὐρυγάνεια. (For the phenomenon of “minor variation of the same name” see Henrichs in Interpretations in Greek Mythology, ed. J. Bremmer [London 1987] 251). And Epimenides fr. 15 DK is represented by Σ Euripides Phoenician Women 13 (1.249 Schwartz) as stating that Laius married Εὐρύκλειαν τὴν Ἔκφαντος … ἐξ ἧς εἶναι τὸν Οἰδίποδα (rightly denounced by Fowler [2013:408] as “a mishmash … from which it is difficult to extract a coherent account”).

Pausanias’ argument clearly implies that in our epic Oedipus begot upon Euryganeia not only Eteocles and Polyneices but also Antigone and Ismene. (Robert [1915:1.181] goes so far as to state that the pair of sisters are explicitly [“ausdrücklich”] attested for the Oedipodeia.) This is rather more remarkable than scholars have generally recognized, since Antigone and Ismene are conspicuous by their absence from Homer, Hesiod, Bacchylides, Pindar, and almost all of the remains of early Greek lyric and elegiac poetry. The earliest secure reference to both is Pherecydes FGrHist 3 F95. One wonders how much area the plot of the Oedipodeia would have had to cover before either sister could have had any significant part to play. Still, although we should not forget that, in the words of Wilamowitz (1914:93), “eine Person, die nichts zu tun bekommt, kann die Sage nicht brauchen,” we should also not forget post-Homeric epic’s fondness for minor characters and superfluous children (see Griffin 1977:43 = 2001:373), or Attic tragedy’s tendency to bring into sudden prominence such previously subordinate figures (Chrysothemis is a case in point). {26|}


[ back ] 1. See my “Prolegomena and Paralegomena to a New Edition (with Commentary) of the Fragments of Early Greek Epic,” Nachrichte der Akademie der Wissenschaften in Göttingen, 1 philosophische-historische Klasse (1986): 99–100.

[ back ] 2. “So kann kein halbwegs verständiger Dichter erzählt haben” is Robert’s triumphant conclusion (1915:1.156) to this part of his argument. “So darf man aber in epischer Dichtung und in Dich-tung überhaupt nicht rechnen,” Lesky (1712.47–48) reasonably retorts.

[ back ] 3. But perhaps de Kock’s citation (1962:20n31) of Aristotle’s characterization of the Odyssey as ἀναγνώρισις … διόλου (Poetics 1459B15) does not really meet Deubner’s stress on the essentially dramatic nature of ἀναγνωρίσματα. One should not confuse the concrete objects that are the latter with the abstract process that constitutes the former.

[ back ] 4. For Deubner’s “basic misreading of the chronological sequence implied by the Greek” here see Mastronarde 32–33.

[ back ] 5. For a list of scholars who supposed the Chrysippus to have introduced the story into Greek litera-ture see de Kock 1962:27nn61 and 62. An origin in early epic (not necessarily the Oedipodeia) is preferred by e.g. Lamer (RE 12.1 [1925]: 477.32–35), Daly (RE 17.2 [1937]: 2110–2111), and K. Schefold (in Classica et Provincalia [Erna Diez Festschrift (1978)] 178–179), who deems the story “eine grossartige Konzeption, das Unheil des thebanischen und des mykenischen Königshauses auf einen gemeinsamen Ursprung, die Frevel an Chrysippos zurückzuführen,” a characterization he takes to favor a late archaic source. For a bibliography of scholars who suppose it to have featured in Aeschylus’ Theban trilogy see Mastronarde 35n1 (adding now, e.g., Lloyd-Jones 2002:11 = 2003:31–32). Mastronarde himself (35–36) is skeptical.

[ back ] 6. See too Krauskopf, Der thebanische Sagenkreis und andere griechische Sagen in der etruskischen Kunst (Mainz 1974) 89n334 and n8 below.

[ back ] 7. Brief literary resumés which tell us Oedipus “killed” the Sphinx (cf. e.g. Wolff in Roscher 3.716.3–12) are, of course, nothing to the point. This is a perfectly natural condensation of “caused her to commit suicide.”

[ back ] 8. This seems especially probable in the case of Etruscan gems that show Oedipus stabbing with sword a sometimes unresisting Sphinx (instances discussed and illustrated by Krauskopf [as cited in n6 above] 52 and plates 19.8 and 9). Etruscan art often depicts Theban legends in a lurid and bloodthirsty manner: see pages 40n11 and 83 below on depictions of Tydeus’ cannibalism.

[ back ] 9. Just as failure to understand Homer’s elimination of the gruesome led Bethe (Homer 2.2.268) to suppose that Homer’s Clytemnestra committed suicide (refuted by Griffin [1977:44n32 = 2001:375n38]), so Wecklein (1901:683 and 688) inferred from Odyssey xi that Homer was unacquainted with the tradition of Polyneices and Eteocles as sons of Oedipus.

[ back ] 10. His narrative is also derived from the Oedipodeia by C. Kirchoff, Der Kampf der Sieben vor Theben und König Oidipus (Berlin 1917) 65; Wecklein 1901:676 and 681.

[ back ] 11. On the general “Potiphar’s wife” motif see the article s.v. “Joseph, der keutsche” by Reents and Köhler-Zülch in EM 7.640–648 and W. Hansen, Ariadne’s Thread: A Guide to International Tales Found in Classical Literature (Ithaca 2002) 332–352. Cf. Davies, “ ‘The Man Who Surpassed All Men in Virtue’: Euripides’ Hippolytus,Wiener Studien 113 (2000): 55n8.

[ back ] 12. I do not intend to delve into the textual problems raised by the phrase ἐπεὶ δὲ ἐνιαυτὸς παρῆλθε in Pherecydes’ fragment. Jacoby ad loc. posits a lacuna before the phrase, since he takes it that the year elapsed after the death of Jocasta. Deubner (1942:29 = 1982:663) prefers to suppose that the year in question is to be dated after the death of the two sons.

[ back ] 13. J. Bremmer in Interpretations in Greek Mythology (London 1987) 52 has tried to accept and make sense of the idea of an early Oedipus who remarries, but I find his arguments (“the wedding may well have been a poet’s solution to the question ‘what happened next?’ In a way, the myth was finished … but an audience always wants more … to be a widower was not a permanent male status”) singularly unconvincing. J. March, The Creative Poet: Studies on the Treatment of Myth in Greek Poetry (BICS Suppl. 49 [London 1987]), 122, is another relatively recent adherent of the notion of an early epic tradition featuring a non-incestuous family, with the incest motif an Aeschylean invention.

[ back ] 14. E.g. Hermogenes on Stesichorus (Tb 28 Ercoles): σφόδρα ἡδὺς εἶναι δοκεῖ διὰ τὸ πολλοῖς χρῆσθαι τοῖς ἐπιθέτοις.

[ back ] 15. Edmunds’ position here is assailed by Bremmer (as cited page 17n13 above) 46, who to my mind merely succeeds in showing (what we all knew) that the Sphinx had been integrated into the story by the time of the earliest evidence of literature and art.

[ back ] 16. Kost omits our passage and says “von einer Person zuerst” (Homeric Hymn to Demeter 422).

[ back ] 17. Followed by several scholars (bibliography in de Kock 1961:15n45: add J. T. Sheppard, The Oedipus Tyrannus of Sophocles [Cambridge 1920] xviin3).

[ back ] 18. For a list of opponents of Robert’s approach see de Kock 1961:16n46. Add de Kock himself, Ste-phanopoulos 1980:105, Simon 1981:9n10, Mastronarde 21n3, etc.

[ back ] 19. Accepted by most scholars, even those (e.g. Stephanopoulos) who wish to accept Pausanias’ testimony.

[ back ] 20. Compare, perhaps, the tradition of Meleager’s heroic death in battle in Hesiod fr. 25.9–13 MW, which seems to follow the Iliad’s playing-down of the horrific elements in the legend. On Bremmer’s attempt to interpret the remarrying Oedipus as an early and explicable feature see page 17n13 above.

[ back ] 21. In particular Wehrli (1957:112 = 1972:65) followed by de Kock (1961:16).

[ back ] 22. For a survey of attempts to identify extant artifacts as dependent upon Onasias’ painting, see J. P. Small, Studies Related to the Theban Cycle on Late Etruscan Urns (Rome 1981) 142–145.