¹ On the linguistic density and allusiveness of Heraclitus’ style, see Kahn, C.H., ‘A new look at Heraclitus’, American Philosophical Quarterly 1 (1964), 189–203Google Scholar and his commentary on the fragments, The Art and Thought of Heraclitus (Cambridge, 1979), especially 87–95.

² Following Graham, D.W., ‘Heraclitus and Parmenides’, in Caston, V. and Graham, D.W. (edd.), Presocratic Philosophy. Essays in Honour of Alexander Mourelatos (Aldershot and Burlington, 2002), 27–44Google Scholar (relevant discussion at 31 n. 22), I prefer this reading to παλίντονος, the alternative. παλίντροπος is the reading found in the best source, Hippolytus (Haer. 9.9.2), who in fact is the only source who gives the full citation and who gives it in a context where it is clear that he has a copy of the book before him. Moreover, since παλίντονος is a more common epithet for a bow, παλίντροπος is the lectio difficilior. παλίντροπος is also found at Plut. 473F (all manuscripts apart from D), 1026B; παλίντονος at Plut. 369B, 473F (MS D) and Porph. De antro 29. As the anonymous reader points out to me, Heraclitus would not have been alone among archaic authors in playing with the -τροπος suffix, given Sappho, fr. 71.4 Voigt, κα[κό]τροπος, occurring in a passage which, like the Heraclitus fragment, ostensibly concerns invective and musicality.

³ On this doctrine, see D.W. Graham, Explaining the Cosmos: The Ionian Tradition of Scientific Philosophy (Princeton, 2006), 122–9 and his entry on Heraclitus in the Stanford Encyclopedia of Philosophy, online at https://plato.stanford.edu/.

⁴ M.L. West, Ancient Greek Music (Oxford, 1992), 56–7.

⁵ Kirk, G.S., Heraclitus: The Cosmic Fragments (Cambridge, 1954), 203–21Google Scholar ruled out the possibility that the musical sense of the word is operative here, ‘for this would be totally unsuitable to … the bow’ (208), but this is unduly restrictive, especially given the characteristically polysemous nature of Heraclitus’ writing. Cf. Arist. Eth. Eud. 7.1.1235a25 = A22, where Heraclitus is reported as seeing the coexistence of high and low notes as necessary for ἁρμονίη.

⁶ Cf. B48, τῶι οὖν τόξωι ὄνομα βίος, ἔργον δὲ θάνατος, ‘And so the name of the bow is life (βίος) but its function is death’, punning upon two different senses of the same word (albeit differently accentuated). For these interpretations of B51, see C.H. Kahn (n. 1 [1979]), 195–200.

⁷ On this fragment, see Tor, S., ‘Heraclitus on Apollo's signs and his own’, in Eidinow, E., Kindt, J. and Osborne, R. (edd.), Theologies of Ancient Greek Religion (Cambridge, 2016), 89–116CrossRefGoogle Scholar.

⁸ Bakker, E.J., The Meaning of Meat and the Structure of the Odyssey (Cambridge, 2013), 98–100CrossRefGoogle Scholar. The D scholia at 20.155 (Dindorf 2.690, who refers to them as ‘V’ scholia) identify this as the festival of Apollo νουμήνιος.

⁹ Thalmann, W.G., Conventions of Form and Thought in Early Greek Epic Poetry (Baltimore, 1984), 175–6Google Scholar notes the resemblance to the Homeric Hymn to Apollo at this point and also comments on how this is the point at which Odysseus’ words and deeds, after his disguise on Ithaca, finally unite.

¹⁰ As Kahn (n. 1 [1979]), 199 argues.

¹¹ Plato's Socrates sees this as the climax of Odysseus’ narrative (Ion 535b–c).

¹² For this latter point I am indebted to the anonymous reader, who draws my attention to the discussion of the simile by Graziosi, B., Homer (Oxford, 2016), 54–6Google Scholar.

¹³ For the fixity and reputation of the Homeric epics by this stage, see Cassio, A.C., ‘Early editions of the Greek epics and Homeric textual criticism in the sixth and fifth centuries b.c.’, in Montanari, F. (ed.), Omero tremila anni dopo (Rome, 2002), 105–36Google Scholar.

¹⁴ Currie, B., Homer's Allusive Art (Oxford, 2016), 9–11Google Scholar notes this methodological principle.

¹⁵ Thalmann (n. 9), 176.

¹⁶ Following here the standard interpretation of Heraclitean War, as represented by Graham (n. 3), 144.

¹⁷ On (possible) engagements with Heraclitus in the play, see M.J. Arp, ‘Pre-Socratic thought in Sophoclean tragedy’ (Diss., University of Pennsylvania, 2006), 14–50. Sophocles at Phil. 931 makes the same pun on βίος as occurs at Heraclitus B48 (see n. 6 above).

¹⁸ Goldhill, S., Sophocles and the Language of Tragedy (Oxford, 2012), 34CrossRefGoogle Scholar.

¹⁹ Goldhill (n. 18), 34 n. 45, reading παλίντονος.

²⁰ Graziosi, B., Inventing Homer (Cambridge, 2002), 29Google Scholar.

²¹ For meaningfulness supporting the hypothesis of allusion, see Currie (n. 14), 31–3, with further references.

²² On this issue, see Cassio (n. 13) contra (e.g.) the development of the Homeric epics proposed by Nagy, G., Homeric Questions (Austin, 1996)CrossRefGoogle Scholar.

²³ First proposed by Bernays, J., ‘Heraklitische Studien’, RhM 7 (1851), 90–116Google Scholar, at 114–15 n. 2 = Gesammelte Abhandlungen (Berlin, 1885), 1.37–63, at 1.62–3 n. 1. Bernays has been followed most recently by Graham (n. 2) and (n. 3), 148–85, whose arguments I find convincing contra (e.g.) Stokes, M.C., One and Many in Presocratic Philosophy (Washington, DC, 1971), 111–27Google Scholar; Nehamas, A., ‘Parmenidean being/Heraclitean fire’, in Caston, V. and Graham, D.W. (edd.), Presocratic Philosophy. Essays in Honour of Alexander Mourelatos (Aldershot and Burlington, 2002), 45–64Google Scholar; Palmer, J., Parmenides and Presocratic Philosophy (Oxford, 2009), 341–5CrossRefGoogle Scholar.

²⁴ The key study is Havelock, E.A., ‘Parmenides and Odysseus’, HSPh 63 (1958), 133–43Google Scholar. Mourelatos, A.P.D., The Route of Parmenides: A Study of Word, Image, and Argument in the Fragments (rev. edn, Las Vegas, 2008), 31–4Google Scholar is more cautious, but the resemblances to the Odyssey are more specific than he allows. Parmenides’ youthful narrator is taken on the path that ‘bears the knowing mortal to every city’ (B1.3), an expression that recalls the man who ‘saw the cities and got to know the mind of many men’ (Od. 1.3). He goes to a place where ‘there are the gates of the paths of Night and Day’ (B1.11 ἔνθα πύλαι Νυκτός τε καὶ Ἤματός εἰσι κελεύθων), a near-identical expression to Od. 10.86, ‘near are the paths of Night and Day’ (ἐγγὺς γὰρ νυκτός τε καὶ ἤματός εἰσι κέλευθοι), a description of the location of Laestrygonia. Parmenides’ road is πολύφημον (B1.2), a word that recalls the name of the Cyclops, Odysseus’ most famous adversary, who is also hinted at later on in the goddess’ reference to ‘the wandering works of the round-eyed (κύκλωπος) moon’ (B10.4): κύκλωψ, before Parmenides, is only ever used in reference to one or more of the mythical monsters.

²⁵ For this sort of multiple reference, see Thomas, R.F., ‘Virgil's Georgics and the art of reference’, HSPh 90 (1986), 171–98Google Scholar, at 188–9 = Reading Virgil and his Texts (Ann Arbor, 1999), 114–41, at 130–2; Fowler, D.P., ‘On the shoulders of giants: intertextuality and classical studies’, MD 39 (1997), 13–34Google Scholar, at 16 = Roman Constructions (Oxford, 2000), 115–37, at 118–19; Nelis, D.P., Vergil's Aeneid and the Argonautica of Apollonius Rhodius (Leeds, 2001), 5Google Scholar.

²⁶ Hinds, S., Allusion and Intertext (Cambridge, 1998), 47Google Scholar.