¹ The lemma ναῦς occurs 589 times in the Iliad, and is the third most common substantive in the poem (after Ἀχαιός and ἀνήρ).

² The total count differs depending on the definition of ‘epithet’ applied. O'Sullivan lists 34 epithets of ships to be found throughout early Greek epic: LfgrE, Band 3, s.v. νηῦς, ναῦς, 381–400. By his criteria, Dee finds a total of 48 different epithets for ships in the Iliad and the Odyssey (the second-most diverse system in the poems, after that for ἵππος), of which 31 are present in the Iliad: Dee, J.H., Epitheta rerum et locorum apud Homerum (Hildesheim, Zurich and New York, 2002), 373–90Google Scholar. Cf. Grey, D., ‘Seewesen’, in Buchholz, H.-G. (edd.), Archaeologia Homerica. Band II, Kapitel G (Göttingen, 1990), 93Google Scholar, who finds 22 in the Iliad.

³ Metrically see Alexanderson, B., ‘Homeric formulae for ships’, Eranos 68 (1970), 1–46Google Scholar and, less detailed, Parry, M., The Making of Homeric Verse (Oxford, 1971), 109–13Google Scholar. For discussions of ship construction, see Morrison, J.S. and Williams, R.T., Greek Oared Ships (Cambridge, 1968)Google Scholar; Kurt, C., Seemännische Fachausdrücke bei Homer (Göttingen, 1979)Google Scholar; Casson, L., Ships and Seamanship in the Ancient World (Baltimore, 1995)Google Scholar; and Mark, S., Homeric Seafaring (College Station, TX, 2005)Google Scholar. On poetics, see Vivante, who does consider epithets for ship: Vivante, P., The Epithets in Homer (New Haven and London, 1982), 65–71Google Scholar and 193–6. He discusses ‘hollowness’ in passing at 13 and 118 but with emphasis on the representational nature of the epithet ‘irrelevant to the narrative occasion’ (i.e. divorced from contextual application). This view will be challenged.

⁴ Constraints of space restrict my analysis to the Iliad and exclude a similarly detailed analysis of koilos in the Odyssey and the complex ways in which that poem is in dialogue with the Iliad. A cursory survey suggests that this would be a fruitful avenue for further analysis.

⁵ And see Grethlein on the danger of circular arguments when correlating the text of epic and material remains: Grethlein, J., ‘From “imperishable glory” to history: the Iliad and the Trojan War’, in Konstan, D. and Raaflaub, K.A. (edd.), Epic and History (Chichester and Malden, MA, 2010), 122–44Google Scholar.

⁶ For the first conclusion, see Casson (n. 3), 44: ‘The ships were “hollow”, i.e., undecked.’ For the second, see Mark (n. 3), 97; Morrison and Williams (n. 3), 45; and Kurt (n. 3), 36–75, whose discussion of koilos leads him to speculate whether Homer knew of ‘cargo’ ships.

⁷ ‘Einen Hohlraum enthaltend, mit Fassungsvermögen ausgestattet, geräumig’, LfgrE, Band 2, 1470–1. Janko, R., The Iliad: A Commentary. Volume IV: Books 13–16 (Cambridge, 1992), 57Google Scholar, discussing κοίλῃς ἐπὶ νηυσί (on 13.107), presents both options: ‘The epithet may stress the ships’ capacity or their lack of a deck.’ Cf. Jong, I.J.F. de, Homer Iliad Book XXII (Cambridge, 2012), 182Google Scholar (on Il. 22.465, to which we shall return).

⁸ Cf. Parry, A. Amory, Blameless Aegisthus: A Study of ΑΜΥΜΩΝ and other Homeric Epithets (Mnemosyne Supplementa 26) (Leiden, 1973), 165–6CrossRefGoogle Scholar.

⁹ Foley, J.M., Immanent Art: From Structure to Meaning in Traditional Oral Epic (Bloomington, 1991)Google Scholar, developed in id., Homer's Traditional Art (Philadelphia, 1999). Cf. Kelly, A., A Referential Commentary and Lexicon to Homer Iliad VIII (Oxford, 2007)Google Scholar, especially 5–17, for a useful summary of traditional referentiality and a defence of its necessarily subjective application.

¹⁰ As Kelly (n. 9), 6 notes: ‘The challenge, therefore, is to detect the traditional quality of the “element” through the semantic significance of its context, not as something which the author or singer must combat in order to make sense, but as an informative source of associative meaning.’

¹¹ All references are to the Iliad unless otherwise indicated. See also 2.113 and 2.288, 4.239, 5.716, 9.20, 12.15–16 (a prolepsis) and 18.326–7.

¹² As Finley, M.I., The World of Odysseus (Edinburgh, 1954), 199Google Scholar writes: ‘In the final analysis, how can prepotence be determined except by repeated demonstrations of success? And the one indisputable demonstration of success is a trophy … there could be no honour without public proclamation, and there could be no publicity without the evidence of a trophy.’ Sarpedon's justly famous speech is the finest exposition of this system in the Iliad (12.310–18). Schein, S., The Mortal Hero (Berkeley, 1984), 67–72Google Scholar provides a useful summary. See further Redfield, J.M., Nature and Culture in the Iliad (Chicago, 1975), 30–9Google Scholar; Van Wees, H., Status Warriors: War, Violence and Society in Homer and History (Amsterdam, 1992)Google Scholar, especially ch. 3; and Zanker, G., The Heart of Achilles: Characterisation and Personal Ethics in the Iliad (Ann Arbor, 1994), 11–13CrossRefGoogle Scholar for a much fuller discussion than is possible or necessary here.

¹³ The point is in fact made about Odysseus’ lack of ships at Od. 4.558–60 and 5.15–16.

¹⁴ Many more examples will occur. See below under ‘type C’ for explicit uses of koilos in this context.

¹⁵ As Nestor at Od. 3.130–1 and 3.153–4. Menelaus explicitly states that he has brought his wealth home in his ships (Od. 4.78–82).

¹⁶ Cf. the words of the herald at Aesch. Ag. 574–9, especially 574: νικᾷ τὸ κέρδος, πῆμα δ᾽ οὐκ ἀντιρρέπει. Note also that we find an iteration of koilos at Od. 13.216: Odysseus checks that the Phaeacians have not stolen (rather than won) any of his prizes and taken them away in their ‘hollow’ ship.

¹⁷ Cf. an identical usage at 22.114–16, both examples together form type ‘C’ of the formulaic pattern.

¹⁸ See Russo's survey: Russo, J., ‘The formula’, in Morris, I. and Powell, B. (edd.), A New Companion to Homer (Leiden, 1997), 238–60Google Scholar. Cf. Finkelberg, M., ‘Oral theory and the limits of formulaic diction’, Oral Tradition 19 (2004), 236–52CrossRefGoogle Scholar.

¹⁹ The term was originally used in Hainsworth, J.B., ‘Structure and content in epic formulae: the question of the unique expression’, CQ 14 (1964), 155–64CrossRefGoogle Scholar, then expanded in id., The Flexibility of the Homeric Formula (Oxford, 1968). Cf. Hoekstra, A., Homeric Modifications of Formulaic Prototypes (Amsterdam, 1965)Google Scholar. Subsequent scholarship has suggested a modification of this relationship by emphasizing that the formulaic constituents are not equal but rather form a ‘nucleus’ and a ‘periphery’. See Visser, E., ‘Formulae or single words? Towards a new theory on Homeric verse-making’, WJA 14 (1988), 21–37Google Scholar and Bakker, E.J. and Fabbricotti, F., ‘Peripheral and nuclear semantics in Homeric diction: the case of dative expressions for “spear”’, Mnemosyne 44 (1991), 63–84CrossRefGoogle Scholar.

²⁰ This is somewhat analogous to Nagler's concept of a pre-verbal Gestalt, but does not require the Chomskyan deep structure that underpins his model. See Nagler, M.N., ‘Towards a generative view of the Homeric formula’, TAPhA 98 (1967), 269–311Google Scholar and id., Spontaneity and Tradition: A Study in the Oral Art of Homer (Berkeley, 1974).

²¹ At Od. 24.50 and Ilias Parua (F20 Davies). Though we are focussed on the Iliad, both examples have a bearing on the argument, and have been considered where appropriate.

²² See—among others—Redfield (n. 12), 204–10 for the view that the funeral games are analogous to battle as an opportunity for the hero to gain fame.

²³ Cf. Agamemnon's prize for spear-throwing at 23.892–3.

²⁴ We might also note a parallel usage in a fragment from the Ilias Parua: αὐτὰρ Ἀχιλλῆος μεγαθύμου φαίδιμος υἱὸς | Ἑκτορέην ἄλοχον κάταγε<ν> κοΐλας ἐπὶ νῆας, ‘now greathearted Achilles’ shining son | led Hector's wife down to the hollow ships’ (F20 Davies), which fits well with the contextual usage of pattern A, and relies on the same nexus of associations engendered by koilos.

²⁵ Alexanderson (n. 3), 29.

²⁶ Friedrich, R., Formular Economy in Homer: The Poetics of the Breaches (Hermes Einzelschriften 100) (Stuttgart, 2007), 18–19Google Scholar.

²⁷ θοάς: 1.12; 1.371; 2.8; 2.17; 2.168; 6.52; 10.450; 10.514; 11.3; 24.564. ἰόντ᾽: 15.116; 24.118; 24.146; 24.195. Alexanderson does not consider this, since his investigation is a strict analysis of metrical equivalence, and as such he pays no attention to developments in ‘formulaic extension’. Again, it may be more productive to think about a possible variety of ‘dictional patterns’.

²⁸ 14.354, 15.305, 17.691, 22.417, 24.203 and 24.519.

²⁹ On breaches of economy, see Friedrich (n. 26), especially 78–83 on the avoidance of contextual unsuitability. Cf. Parry (n. 3), 155.

³⁰ Cf. Finkelberg, M., ‘Oral formulaic theory and the individual poet’, in Montanari, F., Rengakos, A. and Tsagalis, C. (edd.), Homeric Contexts: Neoanalysis and the Interpretation of Oral Poetry (Trends in Classics – Supplementary Volume 12) (Berlin, 2012), 73–81Google Scholar.

³¹ Alexanderson (n. 3), 29. He also does not fully join the dots between koilos and material gain, making the connection for only ‘six out of the eleven occurrences’.

³² On the allusion to alternative potential narrative paths, see de Jong, I.J.F., Narrators and Focalizers: The Presentation of the Story in the Iliad (London, 2004 ²), 68–90Google Scholar; Morrison, J.V., Homeric Misdirection (Ann Arbor, 1992)Google Scholar; id., ‘Alternatives to the epic tradition: Homer's challenges in the Iliad’, TAPhA 122 (1992), 61–71; Louden, B., ‘Pivotal counterfactuals in Homeric epic’, ClAnt 12 (1993), 181–98Google Scholar; and Lang, M., ‘Unreal conditions in Homeric narrative’, GRBS 30 (1989), 5–26Google Scholar. See also Reinhardt, K., Die Ilias und ihr Dichter (Göttingen, 1961), 107–20Google Scholar on the thematic potential for an early ‘Heimkehr unverrichteter’.

³³ We might also note how these parallel offers also serve to confirm Achilles’ pre-eminence; the offer in Book 7 is made to the two Atreidae, but by Book 24 Priam supplicates Achilles directly, who feels confident enough in his own position to grant the appeal and to promise a ‘break’ from the war for Hector's funeral (24.669–70).

³⁴ We should bear in mind, however, that this passage was athetized in antiquity. Commentators noted that the motivation given by Nestor for burning the bodies of the dead is inconsistent with other passages in the poem, where it is suggested that the dead will be buried in tombs at Troy (Σ A on 7.334). West, M.L., Homerus Ilias (Stuttgart, 1998), 219Google Scholar brackets the passage in his edition on the grounds that ‘utique mos Atheniensium insinuatur’, citing Jacoby, F., ‘Patrios nomos: state burial in Athens and the public cemetery in the Kerameikos’, JHS 64 (1944), 37–66CrossRefGoogle Scholar.

³⁵ To these two uses we might add the final iteration of pattern ‘A’ outside of the Iliad, used in relation to Achilles’ death and the common soldiers’ resulting desire to sail home empty-handed at Od. 24.50. This passage contains numerous verbal references to the Iliad (not only the deployment of the formulaic pattern but, for instance, μέγας μεγαλωστί, λελασμένος ἱπποσυνάων at line 40). Cf. Usener, K., Beobachtungen zum Verhältnis der Odyssee zur Ilias (Tübingen, 1990), 104–8Google Scholar, who argues that this passage in the Odyssey is directly influenced (‘beeinflußt’) by the Iliadic use. Alternatively, see Currie, B.G.F., ‘Homer and the early epic tradition’, in Clarke, M.J., Currie, B.G.F. and Lyne, R.O.A.M. (edd.), Epic Interactions: Perspectives on Homer, Virgil and the Epic Tradition (Oxford, 2006), 1–45Google Scholar, at 40 for the suggestion that this passage is a quotation (we might prefer referential usage) of a lost Memnonis that predated both the Iliad and the Odyssey.

³⁶ Morrison (n. 32 [1992]), 75–6 provides a good discussion of the danger that firing the ships poses to Achaean nostos.

³⁷ cf. Σ T on 13.107: ἐπιφέρει <δὲ> τὰς ναῦς, ἐν αἷς ἐστιν αὐτοῖς ἡ σωτηρία, ‘he mentions the ships, in which lies their [the Greeks’] salvation’ (Erbse, III 421).

³⁸ There is, of course, Hera's use at 5.791, but this does not upset the overall scheme, as it proleptically marks and establishes the coming narrative theme.

³⁹ On ‘phrase clustering’, see Hainsworth, [J.]B., ‘Phrase clusters in Homer’, in Davies, A. Morpurgo and Meid, W. (edd.), Studies in Greek, Italic, and Indo-European Linguistics (Innsbruck, 1976), 83–6Google Scholar.

⁴⁰ I am indebted to CQ’s anonymous referee for this point.

⁴¹ LfgrE, Band 2, 162–3.

⁴² Michael Clarke's recent book-chapter rightly draws attention to the dangers of accepting handy linguistic equivalents from our lexica without considering how the meaning of a word is the result of its usage in context: Clarke, M., ‘Semantics and vocabulary’, in Bakker, E.J. (ed.), A Companion to the Ancient Greek Language (Chichester and Malden, MA, 2010), 120–33CrossRefGoogle Scholar.

⁴³ In Gottlob Frege's example: the ‘morning star’ and the ‘evening star’ are both terms that denote the same thing (i.e. Venus), but which differ in their connotational associations. Frege, G., ‘Über Sinn und Bedeutung’, Zeitschrift für Philosophie und philosophische Kritik 100 (1892), 25–50Google Scholar.

⁴⁴ (1): 2.516; 2.602; 2.680; 2.733. (2): 8.334; 10.510; 10.531; 11.274; 11.281; 11.400; 11.520; 12.38; 13.423; 16.296; 17.453; 17.625.