¹ See Young, D. C., Three Odes of Pindar: A Literary Study of Pythian 11, Pythian 3, and Olympian 7 (Leiden, 1968 [Mnemosyne Suppl. 9)], pp. 27–68Google Scholar.

² von Wilaraowitz-Moellendorff, U., Die Textgeschichte der griechischen Lyriker (Berlin, 1900 [Abhand. der Königl. Gesellschaft der Wissenschaften zu Göttingen, Phil.-Hist. Klasse. N.F. 4.3]), p. 48Google Scholar, and Pindaros (Berlin, 1922), p. 280Google Scholar. I can find no reference to the Third Pythian as a letter before Wilamowitz in 1900, but he mentions the idea so casually that it scarcely seems an innovation.

³ Young, D. C., ‘Pindar Pythians 2 and 3: Inscriptional ποτ⋯ and the “Poetic Epistle”’, HSCP 87 (1983), 31–42Google Scholar; Slater, W. J., ‘Pindar's Pythian 3: Structure and Purpose’, QUCC N.S. 29.2 (1988), 51–61Google Scholar.

⁴ So too Lefkowitz, M. R., The Victory Ode: An Introduction (Park Ridge, N.J., 1976), pp. 142–57Google Scholar.

⁵ Drachmann, A. B., Scholia vetera in Pindari carmina ii (Leipzig, 1910), p. 62Google Scholar, ad Inscr. a.

⁶ For the poetic plural of στ⋯ϕανος cf. Ol. 6.26, Pyth. 2.6, Isth. 3.11.

⁷ Drachmann (n. 5), p. 5.

⁸ Farnell, L. R., The Works of Pindar ii (London, 1932), p. 135Google Scholar. Young (n. 3), p. 42 n. 33 is also inclined to see the silence as significant.

⁹ Young (n. 3), pp. 31–42.

¹⁰ Young (n. 3), p. 38 n. 24 refers to this passage from Homer but thinks that it means that Homer was aware of inscriptional practice. This begs very large questions – the date and literacy of Homer. It is much more likely that the writers of epitaphs in the sixth and fifth centuries were familiar with hexameter poetry and with passages such as this one. The ποτ⋯ of inscriptions may be considered a legacy of epic poetry in another way as well. Those who are thought to have died ‘in the olden days’ are thereby assimilated to the Heroic Age. Their natural company is the great of the past, not the lesser men of the present (this understanding gives the ποτ⋯ of an inscription erected after a battle immediate relevance – it does not have to wait for future generations before it acquires meaning).

Hector typically thinks of what people will say ‘at some point in the future’: cf. also Il. 6.459, 479; 22.106. His mother speaks this way too at Eur. Troades 1188–91. Lefkowitz is surely right, pace Young (n. 3), p. 38 n. 24, to see the ποτ⋯ of 1188 as relevant to the ποτ⋯ of 1190 – when Hecuba imagines Astyanax's future we remember Hector's doing the same at Il. 6.479. Euripides will have known epitaphs and their language (and we are clearly in the world of writing in these lines), but in the context the passage functions as does the passage from Homer quoted above: poetry makes explicit the future point for the backward glance.

¹¹ Young (n. 3), p. 35 n. 18 makes light of them. He must do so, as they are inconvenient for his theory.

¹⁸ Vallet, G., ‘Pindare et la Sicile’, in Pindare (Geneva, 1985 [Entretiens Hardt 31]), 285–327, p. 311Google Scholar.

¹³ Another reason that has been advanced for a date of 474 for the Third Pythian is its resemblance to the Ninth Pythian, a poem that can be securely dated to that year. Both odes (they are the only two in the corpus to do so) take stories from the Ehoiai of Hesiod dealing with the loves of Apollo. On the similarities between the two poems see Bernardini, P. A., Mito e attualità nelle odi di Pindaro (Rome, 1983 [Filologia e Critica 47]), pp. 62–7Google Scholar, and Köhnken, A., ‘“Meilichos Orga”. Liebesthematik und aktueller Sieg in der neunten pythischen Ode Pindars’, in Pindare (n. 12) 71–116, p. 77Google Scholar. Köhnken, , however, without discussion simply dates the Third Pythian to 476Google Scholar, the year of the First Olympian.

¹⁴ Young has an excellent discussion of these themes in his literary essay on the poem (n. 1).

¹⁵ I do not claim that the ποτ⋯ is significant enough for there to be an echo of line 5 in line 74. ποτ⋯ is a word Pindar regularly uses in introducing mythical narrative. But it appears here in the very phrase that associates Cheiron with Asclepius. When they are next brought together (63ff.) Pindar has himself understood the undesirability of desire for the absent.

¹⁶ E.g. Slater (n. 3), p. 59.

¹⁷ Burton, R. W. B., Pindar's Pythian Odes: Essays in Interpretation (Oxford, 1962), p. 146Google Scholar, observed that the first-person singular pronoun in the epinician odes of Pindar never excluded the poet. About the same time Lefkowitz, M. R., ‘ΤΩ ΚΑΙ ΕΓΩ: The First Person in Pindar’, HSCP 67 (1963), 177–253Google Scholar, investigated the matter in detail and concluded that the first person could refer only to the poet and that there was no choral ‘I’ at all in the epinicia. She returned to this question in ‘Pindar's Pythian V’, in Pindare (n. 12), 33–69, pp. 47–9, and in an article, ‘Who Sang Pindar's Victory Odes?’, AJP 109 (1988), 1–11Google Scholar: her conclusion here is that not only does the first person not include the chorus but that the odes were for the most part not performed chorally. Two further articles that appeared at almost the same time as Lefkowitz's and that share her conclusion – that there is little reason to believe in choral performance of the epinician odes – are Heath, M., ‘Receiving the κ⋯μος The Context and Performance of Epinician’, AJP 109 (1988), 180–95Google Scholar, and Davies, M., ‘Monody, Choral Lyric, and the Tyranny of the Hand-book’, CQ 38 (1988), 52–64CrossRefGoogle Scholar. Most recently, Carey, C., ‘The Performance of the Victory Ode’, AJP 110 (1989), 545–65Google Scholar, and Burnett, A., ‘Performing Pindar's Odes’, CP 84 (1989), 283–93Google Scholar, have defended the traditional belief that the poems were sung chorally. But even if this is the case, the first person is not necessarily the voice of the group. Carey, for instance, believes that lines 63–79 of the Third Pythian refer to the poet himself and not to the chorus (p. 561 n. 41).

¹⁸ E.g. Young (n. 1), pp. 45–6.

¹⁹ Of both the foregoing interpretations it is tempting to say with Gow, A. S. F., Theocritus ii (Cambridge, 1965), p. 130Google Scholar (re the Seventh Idyll): ‘A theory based upon the assumption that he means the reverse of what he says starts at some initial disadvantage.’

²⁰ E.g. Herington, C. J., Poetry into Drama: Early Tragedy and the Greek Poetic Tradition (Berkeley, Los Angeles and London, 1985), pp. 190–1Google Scholar.

²¹ At Ol. 4.9–10 κ⋯μος appears to mean ‘celebratory song’ and to refer to the poem that is being sung since, unlike festal dance-κ⋯μοι, this one is a χρομιώτατον ϕ⋯ος. Gerber, D. E., ‘Pindar's Olympian Four: A Commentary’, QUCC N.S. 25.1 (1987), 7–24, p. 16Google Scholar, thinks that κ⋯μος at Pyth. 3.73 is similar and means simply ‘song’. But Ol. 4 is quite clearly a short processional (whether at Olympia or Camarina is disputed). The κ⋯μος, if it is a song, is also a procession, perhaps a dance: it is mentioned with the deictic pronoun (and with the receptionand arrival-motifs discussed by Heath [n. 17], pp. 185–90). It is a κ⋯μος very like that of Ol. 14, another short processional ode which refers to τ⋯νδε κ⋯μον (16).

²² Ol. 10. Nem. 3, Nem. 9, and Isth. 2 are poems which seem to celebrate a victory at a subsequent date. If it is felt that Pyth. 3 is simply another such instance, the question of the dating will be immaterial – it will be a commemorative poem at a date after the victory, but exactly when it was composed will be unimportant. My argument is, basically, that the rhetorical stance of Pyth. 3 is internally consistent and that its mood is unique. Special explanation is, consequently, necessary.

²³ See Wilamowitz, , Pindaros (n. 1), pp. 282–3Google Scholar. So too Most, G. W., The Measures of Praise: Structure and Function in Pindar's Second Pythian and Seventh Nemean Odes (Göttingen, 1985 [Hypomnemata 83]), p. 67, with n. 42CrossRefGoogle Scholar.

²⁴ Pausanias (8.42.9) gives a dedicatory epigram from Olympia that accompanied a monument erected by Hieron's son, Deinomenes. This epigram records Hieron's two Olympic victories with the single horse and one with the chariot. The victor-list from Oxyrhynchus (P. Oxy. II 222 Col. I 19) gives the dates 476 and 472 for the κ⋯λης-victories. This corroborates a scholium on the First Olympian (Drachmann, [n. 5] i [1903], pp. 15–16)Google Scholar, where the chariot-victory of 468 is also mentioned (this chariot victory was the occasion of Bacchylides 3).

²⁵ If there was a victory at Delphi in 482 the pattern is not so neat. But, as I have argued, the case for victories in both 482 and 478 is not strong, and 478 seems certain. 482 is in any case not during Hieron's reign in Syracuse.

On the length of the career of a racehorse in antiquity (and in modern times), see Maehler, H., Die Lieder des Bakchylides I.ii (Leiden, 1982 [Mnemosyne Suppl. 62]), p. 79 n. 6Google Scholar. If we accept that Pherenikos was not Hieron's horse at Delphi in 482 and that he was in 474, we have certain victories in 478 and 476, when he was at his peak, and no victory in his final competition. There is no reason for thinking that the victory of 472 was a victory of Pherenikos.

²⁶ I prefer this explanation to the idea that the poem is an unsolicited offering.

²⁷ If the first person of the poem is the voice of the poet and not that of a chorus and if the poet has not come to Syracuse, παρ' ⋯μ⋯ν πρ⋯θυρον would appear to refer to Thebes. But the primary implication of the phrase is that the poet is making a public prayer: cf. the prayer of the Locrian maiden πρ⋯ δ⋯μων at Pyth. 2.18 and the presence of the Cyrenean kings πρ⋯ δωμ⋯των during the festal procession at Pyth. 5.96. Pindar has already drawn attention to the public nature of his prayer in line 2 with the word κοιν⋯ν (proleptic, ‘so daϐ es alle hören können’; cf. Schroeder, O., Pindars Pythien [Leipzig and Berlin, 1922], p. 27)Google Scholar.

Victory and peril belong to the world of men, ignominy and ease to the world of women: a defeated athlete returns home to his mother, Pyth. 8.85. The prayer to the mother goddess is, thus, especially appropriate if there has been an actual defeat. Cf. my ‘Nereids with Golden Distaffs: Pindar, Nem. 5’, QUCC N.S. 25.1 (1987), pp. 25–33, esp. p. 32Google Scholar.

²⁸ Pelliccia, H., in a subtle and sensitive analysis of the rhetorical articulation of the first part of the poem, ‘Pindarus Homericus: Pythian 3.1–80’, HSCP 91 (1987), 39–64Google Scholar, has shown that the ⋯λλ' of line 77 is not the alternative to the preceding impossible conditions but that this function is fulfilled by the new condition of line 80.

²⁹ The importance of conditional statements in the Third Pythian is noted by Greengard, C., The Structure of Pindar's Epinician Odes (Amsterdam, 1980), p. 107 n. 66Google Scholar.

³⁰ The practice of quoting a poetic text and then commenting upon it first becomes common in Greek poetry about this time: see West, M. L., Studies in Greek Elegy and Iambus (Berlin and New York, 1974 [Untersuchungen zur antiken Literatur und Geschichte 14]), p. 180CrossRefGoogle Scholar.

³¹ Drachmann (n. 5), p. 82. Cf. Macleod, C. W., Homer: Iliad Book XXIV (Cambridge, 1982), p. 133Google Scholar. Among editors of the Iliad, Monro, Leaf, Ameis-Hentze, and Willcock, e.g., maintain that the proper understanding of the passage is that there are but two urns.

³² Fera, M. Cannatà, ‘Pindaro interprete di Omero in “Pyth.” 3, 81–2’, GIF 38 (1986), 85–9Google Scholar, maintains that controversy over the correct understanding of the lines from Homer already existed in Pindar's time.

Plutarch is fond of the passage and quotes or alludes to it five times in the Moralia. Only once (105c) does he give the version we find in our texts of Homer. He quotes Plato's text at 24a–b and at 600d provides a commentary obviously based on Plato's version. It is unclear which version he has in mind at 369c and 473b, but quite clear that here too he thinks that there are but two jars in all.

³³ As Johansen, H. Friis and Whittle, E. W. believe, Aeschylus: The Suppliants iii (Copenhagen, 1980), p. 343ad 1070Google Scholar. One of the scholia to Ol. 1.60, μετ⋯ τρι⋯ν τ⋯ταρτον π⋯νον (Drachmann, [n. 5], i [1903], pp. 40–1)Google Scholar, tries, not very convincingly, to bring the expression into line with the passage in Pyth. 3. For a discussion of this phrase, ‘the most controversial in O. l’, see Gerber, D. E., Pindar's Olympian One: A Commentary (Toronto, Buffalo and London, 1982 [Phoenix Suppl. 15]), pp. 99–103Google Scholar.

³⁴ Gildersleeve, B. L., Pindar: The Olympian and Pythian Odes ² (New York, 1890), p. 269Google Scholar.

³⁵ Though the ⋯τ⋯ρ is adversative it is not strongly so: for a similarly weak contrast between ⋯τ⋯ρ and μ⋯ν cf. Pyth. 4.168–9. The δ⋯ of line 100 also provides continuity rather than contrast.

³⁶ See Janko, R., ‘P. Oxy. 2509: Hesiod's Catalogue on the Death of Actaeon’, Phoenix 38 (1984), pp. 299–307CrossRefGoogle Scholar.

³⁷ Achilles was, like Asclepius, taught by Cheiron (cf. Nem. 3.43–55). Both mythical sections of the poem thus end with the fiery death of a ward of Cheiron, and both deaths are followed by a gnomic reflection about what mortals must (χρ⋯: 59, 103) expect from the gods.

There are two actual pyres in the poem, that of Coronis and that of Achilles. Coronis, is killed τ⋯ξοισιν ὕπ' Ἀρτ⋯μιδy03BF;ς (10)Google Scholar, Achilles, τ⋯ξοις (101)Google Scholar of an unnamed assassin. The parallelism is surely deliberate, and another connection is thus established between the two mythical sections of the poem. Pindar suppresses direct mention of the agent of Achilles' death. That agent was Apollo, who also had Coronis slain. Apollo is mentioned neither as the source of the arrows that killed Achilles nor as having sung at the wedding of Peleus and Thetis (cf. Nem. 5.22–5 and Il. 24.63), perhaps because the one would suggest the other and Pindar's presentation of the wedding as a moment of unalloyed felicity between ills would thereby be damaged: the wedding redounds to the discredit of Apollo in Homer (Il. 24.63) and in Aeschylus (cf. TrGF ii.350, where Apollo speaks dishonestly). On the wedding of Peleus and Thetis in Pindar and earlier, see March, J. R., The Creative Poet: Studies on the Treatment of Myths in Greek Poetry (London, 1987 [BICS Suppl. 49]), pp. 3–26, esp. pp. 20–3Google Scholar.

³⁸ Dissen, L., Pindari carmina quae supersunt ii (Gotha and Erfurt, 1830), p. 211Google Scholar.

³⁹ For a selection, see Mezger, F., Pindars Siegeslieder (Leipzig, 1880), pp. 65–6Google Scholar.

⁴⁰ Boeckh, A., Pindari opera quae supersunt II.ii (Leipzig, 1821), p. 255Google Scholar: ‘De ipsa filii vel filiae morte ne dubites.’ The second malum for Boeckh is, of course, Hieron's illness.

Boeckh's interpretation at least has the merit of trying to bring the myths of the two sections of the poem together. For a recent discussion of the problem of the unity of the two sections of the poem, which remains broken-backed in most interpretations, and for a proposed solution, see Buongiovanni, A. M., ‘Sulla composizione della III Pitica’, Athenaeum N.S. 73 (1985), pp. 327–36, esp. 33 IffGoogle Scholar.

⁴¹ Mullen, W., Choreia: Pindar and Dance (Princeton, 1982), pp. 100–9Google Scholar, has studied the interaction of χ⋯ριῂ and π⋯μα in the odes. His statistics show (not altogether surprisingly) that Pyth. 3 is exceptional in its emphasis on π⋯μα.

⁴² A correct etymology, as it happens: see Frisk, H., Griechisches etymologisches Wörterbuch i (Heidelberg, 1973), p. 341Google Scholar. But for purposes of the poem it is the word-play that is important. The correctness of the etymology is not.

If we are tempted to look for a Grundgedanke in the poem, the quotation from Homer, with its strategic position in the structure and its resonance in the language and the myths of the poem as well as in the personal situation of the addressee, would be an excellent candidate. Young, though he is critical of the theory of Grundgedanken, finds the Third Pythian more amenable to this theory than the other odes of Pindar. But the Grundgedanke that has traditionally been found in this poem is something rather different, i.e. γν⋯θι σεαυτ⋯ν (see Young [n. 1], p. 65). The fundamental idea of both good and bad fortune's being of divine origin suits the myths of both parts of the poem, however, whereas the idea of self-knowledge or self-restraint is less applicable to the stories of Cadmus and Peleus than to the stories of Coronis and Asclepius (for a discussion of the inadequacy of most analyses of the poem to account for the myths of the second section, see Buongiovanni [n. 40]). There is a more sympathetic treatment of the idea of Grundgedanken now by Heath, M., ‘The Origins of Modern Pindaric Criticism’, JHS 106 [1986], 85–98CrossRefGoogle Scholar.

⁴³ Lines 103–4 say much the same thing: prosperity comes πρ⋯ς μακ⋯ρων.

⁴⁴ ‘…en délice il change son absence.…’ The phrase is from Paul Valéry's Le Cimetière marin, a poem that takes its epigraph from Pythian 3. Another fine poem that shows the direct influence of the Third Pythian is Hölderlin's hymn, Der Rhein: see the study by Seifert, A., ‘Die Rheinhymne und ihr Pindarisches Modell: Struktur und Konzeption von Pythien 3 in Hölderlins Aneignung’, Hölderlin-Jahrbuch 23 (1982–1983), 79–133Google Scholar.

⁴⁶ Drachmann (n. 5), i [1903], p. 68, ad Ol. 2.29c. Diodorus and the scholiast agree on the essential point, that Hieron wished to get rid of Polyzelus by sending him to fight in a mainland war, but disagree on other points (e.g. whether Polyzelus actually went).

Polyzelus and Gelon are never mentioned in Pindar's poetry – the situation at Syracuse was strikingly different from that at Acragas, where Theron and Xenocrates were on the best of terms and both patrons of Pindar. At Pyth. 1.48 the plural εὑρ⋯σκοντο is surprising immediately after the singular παπ⋯μειν' (referring to Hieron). No subject is expressed, but the reference is to the leaders at the battle of Himera where it is likely that both Polyzelus and Hieron were with Gelon. Line 79 is likewise vague, with the plural able to accommodate a Polyzelus and a Hieron who are not mentioned by name.

⁴⁶ Chamoux, F., L'Aurige de Delphes (Paris, 1955 [Fouilles de Delphes 4.5]), pp. 26–31Google Scholar.

⁴⁷ See, e.g., Woodford, S., An Introduction to Greek Art (Ithaca, 1986), p. 86Google Scholar; Mattusch, C. C., Greek Bronze Statuary: From the Beginnings through the Fifth Century B.C. (Ithaca and London, 1988), pp. 127ffGoogle Scholar.

⁴⁸ Pindaros (n. 2), p. 237.

⁴⁹ E.g. Mattusch, loc. cit. (n. 48). Jeffery, L. H., The Local Scripts of Archaic Greece (Oxford, 1961), p. 266Google Scholar, thinks that the change was made by the Geloans about fifteen years after the date of the original inscription.