Skip to main content Accessibility help

The invention of Homer

  • M. L. West (a1)


I shall argue for two complementary theses: firstly that ‘Homer’ was not the name of a historical poet, but a fictitious or constructed name, and secondly that for a century or more after the composition of the Iliad and Odyssey there was little interest in the identity or the person of their author or authors. This interest only arose in the last decades of the sixth century; but once it did, ‘Homer’ very quickly became an object of admiration, criticism, and biographical construction.



Hide All

1 It is not the purpose of the present paper to establish these points, which are not essential to my argument, and this is not the place for a bibliography of separatist and unitarian statements. Those who cling to the belief that one man was responsible for both poems seem to me to be hindered from a just assessment of the contrary evidence by a romantic attachment to the traditional idea of the one supreme poet.

2 Alcm. PMGF 39; Sol. 20.3; Thgn. 19; 771; Stes. PMGF 212.

3 Phemius’ song about the ’AΧαιŵν νóστος is called ‘new’ (Od. 1.352), with regard to the fact that its subject matter was very recent, but there is no suggestion that he personally composed it or that it belonged to him more than to other poets.

4 Cf. Durante, M., Sulla preistoria della tradizione poetica greca, vol. 2 (Rome, 1976), 185–7. He notes that in the South Slavic epic tradition no guslar's name is remembered from the centuries before Vuk Stefanović Karadžić made his famous collections (from 1813 on.)

5 Some Old Norse texts were similarly identified by their place of origin, like the Greenland Lay of Atli and the Flateyjarbók (Flat Island Book). According to one view, the name Edda comes from the place-name Oddi in south-west Iceland, the home of Snorri Sturluson. For the Iliad poet's connection with Ilios, cf. Schadewaldt, W., Iliasstudien (Leipzig, 1938), 125, n. 0; West, M. L., MH 52 (1995), 217, n. 43.

6 Burkert, W., MH 29 (1972), 75.

7 Durante (n. 4), 189. He refers also to the Euboean Hομεριος (IG 12[9].56. 135; lead tablet from Styra, fifth century), ‘che peraltro è omonimo di Zεɉς ‘Oμριος, e quindi ha ragione teoforica.’ The Aetolian Oμαρος ‘può ben essere una Rückbildung del nome precedente, qual è ad esempio παναíτωλος rispetto a παναιτώλιος (così Bechtel, Pers[ onennamen], pp. 525, 532).’

8 A. Körte, RE iiA, 1321–2.

9 Durante (n. 4), 190–1.

10 Ephorus, FGrHist 70 F 1. Lycophron's use of ὂμηρος for ‘blind’ (422) is of course inspired by this theory.

11 Pind. Nem. 2.1–2; Pl. Phaedr. 252b; Ion 530d; Rep. 599e; Isoc. Helen 65.

12 Harpocr. s.v. ‘Oμηρíδαι; Acusilaus, FGrHist 2 F 2; Hellan. 4 F 20; the Crates in question is identified by Jacoby (FGrHist 362 F 5) as the Athenian writer on rituals. Besides Jacoby's commentaries ad locc. see his important additional note in vol. IIIb (Noten), 407–10. The Seleucus fragment (76 Müller) is overlooked by Nilsson, M. P. in his brief treatment of the Chian Dionysia, Griechische Feste von religiöser Bedeutung (Leipzig, 1906), 306.

13 G. W. Nitzsch, De historia Homeri, vol. 1 (Hanover, 1830), 130; id., Die Sagenpoesie der Griechen (Braunschweig, 1852), 317; A. Kirchhoff, SPAW 1893.904; W. Schmid, Gesch. d. gr. Lit. I.1 (with O. Stählin, Munich, 1929), 158; Burkert, W. in Bowersock, G. W. et al. (edd.), Arktouros. Hellenic Studies presented to Bernard M. W. Knox (Berlin-New York, 1979), 55. The conversion into an Olympiadic dating would be due to Hippostratus himself after the example of Timaeus; cf. Jacoby (n. 12), 595. A dating by Olympiads also appears in F 4.

14 Jacoby (n. 12), 596.

15 Orlandini, P., Kokalos 3 (1957), 94–6, fig. 22/3; Guarducci, M., Annuario (1959/1960), 397; Bull, épigr. (1962), no. 397; Burkert (n. 13), 54–5. Fick, A., Die homerische Odyssee (Göttingen, 1883), 280, sought to find another reference to Cynaethus in Pliny, N. H. 4. 66, hanc (Delum) Aristoteles ita appellatam prodidit quoniam repente apparuerit enata; Aglaosthenes Cynthiam, alii Orlygiam, Asteriam, Lagiam, Chlamydiam, Cynethum Pyrpylen (Cynaethus Pyrpolen Fick) igne ibi primum reperto.

16 West, M. L., CQ 25 (1975), 161–70. On p. 168 I suggested that the Delian hymn might be dated between 570 and 547. However, Burkert (n. 13), 62, has a better argument for a later date: the hymn presupposes that Apollo has a temple on Delos (52, 56, 80), which he does not seem to have done until about 540/30.

17 Burkert (n. 13), 59–60; Janko, R., Homer, Hesiod, and the Hymns (Cambridge, 1982), 112–13. Burkert dates the event to (spring) 522, arguing that Polycrates’ death followed μετ’ ⋯λíγον Χρóνον (Phot. s.v. πὐθια καì Δ⋯λια; cf. Zen. Ath. 1.62), and that that was ‘late in 522’. But this is incorrect. The account of Polycrates’ death in Hdt. 3.122–5, which is no doubt based on Samian tradition, presupposes that Cambyses was still the King of Persia (122.3, 126.1). Cambyses died in the spring or summer of 522. It remains theoretically possible that Polycrates could have celebrated his Delian festival in the spring of 522 and still died before Cambyses, but 523 seems much likelier. So Parke, H. W., CQ 40 (1946), 105–8.

18 Kirchhoff (n. 13); Burkert (n. 13), 57–8.

19 Cf. Burkert, W. in Papers on the Amasis Painter and his World (Malibu, 1987), 55.

20 RhM 122 (1979), 197. Fehling argues bizarrely that the Homeridai did not exist at all, but were a fantasy developed from Pindar's poetic use of the word for rhapsodes.

21 The various genealogies of Homer are conveniently tabulated by Allen, T. W., Homer. The Origins and the Transmission (Oxford, 1924), facing p. 32. A late genealogy of Terpander that makes him Homer's great-great-grandson (Suda τ 354 s.v. ⋯Éρπανδρος = Terpander test. 24 Gostoli) is not relevant to the Homeridai's claims to be descended from the poet.

22 Jacoby (n. 12), 408–9, demolishing the amateurish romancings of Wade-Gery, H. T. in The Poet of the Iliad (Cambridge, 1952).

23 It was no doubt the Panathenaic canonization of these two poems that led over time to their being regarded as the only epics that were truly by Homer. ‘Um 500 sind alle gedichte von Homer; um 350 sind von Homer im wesentlichen nur noch Ilias und Odyssee’ ( Wilamowitz, , Homerische Untersuchungen [Berlin, 1884], 353, after a survey of the evidence).

24 West, M. L., The Orphic Poems (Oxford, 1983), 720, 108–11. The first extant mention of Orpheus in literature comes from Ibycus, another poet at Polycrates’ court.

25 Hdt. 1.62.4; sch. Ar. Pac. 1071.

26 Hdt. 7.6.3. In the Homeric scholia he is also accused of having interpolated Od. 11.602–4, and elsewhere he is regarded as a forger of Orphic poems (West [n. 24], 40).

27 Soranus, Vita Hippocratis 1; cf. Jacoby on Pherecydes 3 F 59; L. Edelstein, RE Supp. 6. 1295.

28 Ath. 638b; Hsch. s.v. ảμητορíδας Et. Magn. 83.15; O. Crusius, RE 1. 1828–9.

29 I will pass over a possible explanation from West Semitic which I have put forward elsewhere (The East Face of Helicon [Oxford, 1997], 622–3) and which is too adventurous to justify a second airing.

30 Od. 16.468. Cf. σϒμβάλλομαι which can also mean ‘meet, encounter’, ὂμηρα ‘hostage’ is similarly analogous to σὐμβολον a token exchanged by way of a compact. For the semantic development, cf. Durante (n. 4), 190–1.

31 Hes. Th. 39. In my commentary I compared Hymn. Ap. 164 οὒτω σøιν καλ⋯ σϒνάρηρεν ⋯οιδ⋯, and referred to the musical sense of ⋯ρμονíα, ‘tuning, attunement’. Gregory Nagy, who also regards ‘Oμηρος as a mythical, prototypical author, interprets the name as ‘he who fits [the Song] together’: The Best of the Achaeans (Baltimore, 1979), 296–300; Pindar's Homer (Baltimore, 1990), 373; Homeric Questions (Austin, 1996), 89–91; Poetry as Performance (Cambridge, 1996), 74–5. It is not clear to me whether he regards the Homeridai as prior.

32 Above, p. 367 with n. 12.

33 Wilamowitz, , Die Ilias und Homer (Berlin, 1916), 366 (cf. Jacoby [n. 12], 410), thought that they were called ὂμηροι, not in the sense of ‘hostages’ but of ‘followers, attendants’; Theopompus (FGrHist 115 F 300) said that ⋯μηρεîν was an old word for ‘follow’, and derived the ‘hostage’ sense from this. (Cf. Hsch. o 714 ⋯μηρεî ༐γγϒâται, ⋯κολοϒθεî, and 717 ⋯μηρητῆρῆρες ⋯κóλοϒθοι, σϒν⋯γοροι.) Aristotle (fr. 76), in a complicated story about Homer's birth and childhood, related that he had at first been called Melesigenes, but his name was changed to Homer when the Lydians were abandoning Smyrna to the Greeks, the inhabitants were invited to follow them out of the city, and he said he wanted to ⋯μηρεîν, using this word for ‘follow’.

34 ‘Il nome di Omero’, Rendic. morali dell’ Accad. dei Lincei 1957, ser. 8 vol. 12 fasc. 1–2, 94–111, repeated with slight modifications in Sulla preistoria … (n. 4), 185–203.

35 Polyb. 5.93.10; Strab. 8.7.3 εἲκοσι μέν ἒτη διετÉλεσαν γραμματÉα κοινɂν ἒΧοντες καì σταατηγοɉς δὐο κατ’ ༐νιαϒτɂν οdz ’AΧαιοí, καì κοινοβοὐλιον εỉς ἒνα τóπον σϒν⋯γετο αὐτοîς, ༐καλεîτο δέ ‘Aάριον, ༐ν ὦι τà κοινà ༐Χρημάτιζον καì οὖ ’ἒωνες πρóτερον, and 8.7.5. The ‘Oμάριον established in Calabria in the fifth century by the Achaean colonies Croton, Sybaris, and Caulonia had the same role: κοινɂν dzερɂν καì τóπον ༐ν ι τάς τε σϒνóδοϒς καì τà διαβοὐλια σϒνετέλοϒν (Polyb. 2.39.6).

36 Paus. 7.24.2, who relates the legend that Agamemnon convenced the Greek leaders there before they set out against Troy.

37 Hsch. θ 90 θάμϒρις παν⋯γϒις, σὐνοδος …91 θαμϒρíζει ảθροíζει, σϒνάγει In a fourth century inscription from Thespiae two men are named as θαμϒρíδδοντες, evidently some kind of official role (SEG 32.503; cf. P. Roesch, Études béotiennes [Paris, 1982], 138–42, who interprets as ‘célébrer le culte de Thamyris’).

38 Crusius, O., Philologus 54 (1895), 717, ‘mit τóτε πρŵτον soll wahrscheinlich der Agon in Chalkis übertrumpft werden’.

39 Janko (n. 17), 113–14, 259–61, suggests that Cynacthus himself produced the verses to validate the performance of a ‘Hesiodic’ together with the ‘Homeric’ hymn, sc. the Pythian beside the Delian.

40 When Herodotus (5.67.1) says that Cleisthenes of Sicyon stopped the rhapsodes from reciting ‘Homeric’ poetry, or when Aristotle (Rhet. 1375b30) says that the Athenians used ‘Homer’ as evidence in support of their claim to Salamis about 600 B.C., it is unsafe to infer that the name Homer was actually used at the time of those events. Cf. Burkert (n. 19), 44.

41 Tatian, Ad Graecos 31 (= DK 8.1), names Theagenes with others under the heading of those who have enquired about Homer's poetry, his ancestry, and his date, but this need not mean that each writer in the list treated all those topics. Theagenes’ main concern was apparently to justify Homer's theology by means of allegorical interpretation.

42 Xenoph. DK 21 B 11; 10 ༐ζ ảρΧῆς καθ’ ‘Oμηρν(,) ༐πεì μεμθ⋯κασι πάντες (cf. Burkert [n. 19], 45). This might, of course, have been written as late as the 470s.

43 Heraclitus, DK 22 B 42, 56, 105.

44 Simon. eleg. 11. 15–18, 19.1–2, 20.13–15; PMG 564. Note also the apophthegm about Hesiod and Homer attributed to Simonides in Gnom. Vat. 1144 (FGrHist 8 F 6; D. A. Campbell, Greek Lyric, 3.366).

45 Aeschin. Ctes. 183; Plut. Cimon 7.6; Page, D. L., Further Greek Epigrams (Cambridge, 1981), 257, lines 841–2 (cf. Hdt. 7.161.3).

46 Pind. fr. 264, 265; Pyth. 4.277, 3.112–15; Nem. 7.20–3; Isth. 3/4.55–9; cf. fr. (anon.) 347. Homeric tradition as the τριπɂς ảμαζιτóς Pae. 7b. 11.

47 Bacchyl. fr. 48.

48 Hdt. 2.23, 53, 116–17; 4.32.

49 On this development, see Burkert (n. 19), 56–7.

50 Xenophanes, Heraclitus, see above; Simonides, PMG 542, 579, 581; Pratinas, PMG 713; Epicharmus fr. 88 (IEG 2.45); Pindar, , see references collected in my Ancient Greek Music (Oxford, 1992), 345, n. 73. If we can trust the Peripatetic Megaclides, Stesichorus had already mentioned older poets by name: Xanthus (PMGF 229); Hesiod (PMGF 269). Cf. Janko, R., CQ 36 (1986), 41–2.

51 West, M. L., The Hesiodic Catalogue of Women (Oxford, 1985), 136–7.

52 Cf. Wilamowitz (n. 33), 370–1, 375–6, 439; Marx, F., RhM 74 (1925), 406–8, 417.

53 West (n. 50), 348. Anacreon may have introduced the barbitos to Athens (ibid., 58). I suspect he may also have brought knowledge of the Lesbian poets, with whom acquaintance is shown from early in the fifth century (e.g. PMG 891).

54 Ath. 3a; cf. Aloni, A., L'aedo e i tiranni (Rome, 1989), 121–2. We hear of a γραμματισγτ⋯ς called Maeandrius who held a position of trust with Polycrates, Hdt. 3.123.1.

55 Heraclitus, DK 22 B 129; cf. West (n. 24), 8–9.

56 Neanthes, FGrHist 84 F 29; cf. Burkert (n. 6), 77–8.

57 A risible name in Plato's opinion (Rep. 600b), through attested as that of an Ephesian historian (FGrHist 417). Burkert (n. 6), 78, observes that the termination -ειοι was typical of hetairiai, clubs, political parties, philosophical schools, and the like, whereas as a patronymic it would be archaic or Aeolic.

58 Plato, loc. cit., Call. Epigr. 6 Pf., Certamen 18, Procl. Vit. Hom. p. 100. 11 Allen = 26.16 Wil., etc.; cf. Burkert (n. 6), 76, n. 10.

59 Another area of rivalry between the two parties appears in the legend about Lycurgus’ reception of the Homeric poems in Sparta. Ephorus (70 F 149 § 19) had Lycurgus meet Homer in Chios, whereas Heraclides Lembus (after Aristotle, fr. 611.10) said that he got the poems from Creophylus’ descendants. Cf. Burkert (n. 6), 77.

60 West, M. L. in La Fable (Fondation Hardt, Entretiens sur l'antiquité classique 30, Vandæuvres-Genève, 1984), 116–19, 123–6.

61 Crusius (n. 38), 720–1, noting the part played by Apollo at the start of the Iliad, suggested that the hymn served as an introduction to that epic.

62 Janko (n. 17), 256–8, thinks that the (Delian) hymn was composed and posted up much earlier, at a time when Apollo had no temple of his own and was sharing Artemis’. He then has difficulty in explaining the references to a temple of Apollo in the hymn.

63 Burkert (n. 13), 60; cf. id. (n. 19), 53; id. (n. 6), 78, n. 19, ‘die Ausgestaltung der Panathenäen stellt sich neben die fast gleichzeitige Ausgestaltung der Dionysien durch die Tragödie’. On the uncertainty of the conventional date for the establishment of tragedy at the Dionysia cf. West, M. L., CQ 39 (1989), 251–4.

64 Johansen, K. Früs, The Iliad in Early Greek Art (Copenhagen, 1967), 223–7, 236–40.


Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed