1. History, myth, poetry

Patrick Kavanagh's short poem confronts the reader with a number of questions which will preoccupy us in this survey. The Homeric poems show us a world which in many respects seems primitive and remote; even if the expedition of the Greeks against Troy really happened, even if it took place on the scale which the Iliad asserts, and lasted the full ten-year span, it would still be ‘a local row’ compared with later historical conflicts, ancient or modern. Can the bad-tempered disputes of warrior chiefs, the violent revenge of a savage and undisciplined soldier, the lies and posturing of a vagabond rogue, still move or excite an audience today? It will be necessary to show here some of the ways in which Homer gives the conflict at Troy, and the homecoming of Odysseus, a timeless importance, so that these mere episodes in the vanished heroic age – long past even for the poet and his audience – become microcosmic images of human life. The vast subject of Homer's influence upon later western literature cannot be even superficially addressed here; but occasional comparisons and illustrations may help to show how much subsequent poets and artists have found in the Iliad and the Odyssey to enlighten and inspire their own work.

‘I made the Iliad from such / a local row.’ But what was the local row about, and where and when did it happen? The Greeks always assumed that the war was an authentic historical event, and modern faith in this was strengthened by the landmark excavations in mainland Greece and at Troy by the archaeologist Heinrich Schliemann and his many successors. On the one hand, a glorious civilization had existed in Greece in times far earlier than the classical age; on the other, a city with mighty walls had existed near the Hellespont, had been rebuilt many times, and at some stage had been destroyed by fire. Supporting evidence seemed to appear in Hittite documents, which mentioned a people called the Ahhijawa, a name which could be an earlier or foreign form of ‘Achaeans’; and there were other possible correspondences – most intriguingly, evidence of a monarch ruling in ‘Wilusa’ (Ilium) who bore a name transmitted in documents as Alaksandu, which strongly resembles ‘Alexandros’, the alternative name for the Trojan Paris. It was tempting to suppose that the legends were substantially true, that a great force of Greek invaders attacked and sacked the city now called ‘Troy VIIA’, and that some time later the fall of the Mycenaean civilization brought an end to this prosperous era; later mythology saw this in terms of an age of heroes followed by subsequent decline. The discovery that ‘Linear B’, the language of the Mycenaean age, was in fact an earlier form of Greek, made it possible to hypothesize a continuous tradition of poetic narrative, which preserved a record of these great days through the Dark Ages.Footnote ²

Bold and imaginative reconstruction has continued, with the advance of archaeological and linguistic research. A picture emerges of large-scale conflicts between Mediterranean communities and the societies dominating the eastern seaboard in the late second millennium.Footnote ³ Yet caution is still in order. If there was a continuous tradition from Mycenaean times, we cannot assume that the tradition was free of distortion or misunderstanding (for instance, those who do not dismiss the Wooden Horse as pure fiction sometimes suppose it to be a confused echo of an Oriental siege engine). Some features of the tradition are, in any case, implausible: the thousand ships, the ten years spent far from home are surely epic exaggeration; nor can even the most romantic of readers suppress doubts about a national campaign fought to recover one leader's wife. Overseas raids and adventures are a familiar part of the epic repertoire and probably of the real-life background, as many passages in the Odyssey suggest. In the Iliad, Nestor recounts his own exploits as a young man in a cattle-raiding expedition against the people of Elis (11.670–762): as has often been suggested, this may well give a more accurate idea of the scale of warfare in early Greece than the vast assemblage of peoples listed in the Catalogue of Ships.Footnote ⁴ At all events, even if some general account survived from Mycenaean times, it is hardly likely that any details of the course of events were preserved.Footnote ⁵

Whatever the historical facts behind the Iliad, they are separated from the epic itself by a gap of at least four hundred years (the traditional date for the fall of Troy was 1183 bc, and archaeologists currently place the destruction of Troy VIIA not far from that date).Footnote ⁶ The Odyssey, being a tale about an individual's fortunes and an island community of little prominence in Mediterranean history, is still less likely to preserve much in the way of historical reality. A great deal of the Odyssey, in any case, is set in a world of magic and monsters, remote from authentic history or geography. The poems do contain elements of an older tradition: the wealth of Agamemnon, lord of Mycenae ‘rich in gold’, Ajax's tower-like shield, and Odysseus' boar's-tusk helmet are examples. Sometimes these memories are blurred or conflated with later conventions: for the most part the warriors carry two thrusting-spears, but the single heavy spear, a Mycenaean weapon, is occasionally recalled, in some passages to arouse awe and a sense of latent power (see Il. 16.140–4). Certain verbal phrases and formulae also clearly have a long history, and it has been argued that many of these go back to Mycenaean times or even beyond: this would imply a long and continuous tradition of poetry on heroic themes.Footnote ⁷

Later ancient criticism assumed that the Greek epics preserved historical tradition, but it was always recognized that the poet had enhanced or embroidered what he inherited: Herodotus remarked that Homer in the story of the abduction of Helen had selected from various versions one which seemed more appropriate to epic story-telling (2.116.1, referring to the Cypria), and Thucydides commented that it was natural for a poet to exaggerate the importance of the events he described (1.10). Much has been made of the distinctions between different types of story, such as ‘saga’, which is conceived as more realistic, though concerned with heroic combat, and ‘folk tale’, an elusive term often applied to more fantastic or magical narratives; definitions are difficult, and the relation of both terms to ‘myth’ is unclear.Footnote ⁸ In general it would seem that both Homeric poems include folk-tale or magical elements, but the Iliad, with its military theme and firm geographical setting, is more like saga, while the Odyssey, especially in its first half, is closer to folk tale.Footnote ⁹ But any strong contrast of this kind swiftly breaks down on closer analysis: the Iliad includes a cap of invisibility, a talking horse, a warrior who fights with a river god, and a body magically preserved against decay; the Odyssey, in its own way, shows us more of social and domestic ‘realities’ than the more sombre and dignified Iliad will admit. Moreover, the presence of the Olympian gods in both poems, but more prominently in the Iliad, makes Homer a peculiarly special case: these colourful and potent figures, intervening freely in mortal affairs for good or ill, seem to belong to neither saga nor folk tale, but transcend such categories.

Comparative studies have shown that, even when narrative poetry of a heroic type deals with events closer in time than the Trojan War, and better documented, startling distortions may be introduced.Footnote ¹⁰ The Song of Roland recounts an episode in the campaigns of Charlemagne, and a Latin chronicler records the death of ‘Hruodlandus’ (i.e. Roland) in battle in 778; but the battle in question, an attack by Basques, was of no historical importance. In the heroic style, the poet turns it into a huge conflict between Christian and Infidel, and Roland is slain by a mighty Saracen; the whole poem embodies the spirit of the twelfth-century Crusades.Footnote ¹¹ Similar transformations and intermingling of history and imagination can be found in other traditions: the Nibelungenlied, for example, or the Serbian epics.Footnote ¹² The Norse Saga of the Volsungs moves from pure myth – the tale of Sigurd the Dragon-slayer, rich in magic and close to the world of the Norse gods – into a more historical era, with Atli (Attila the Hun) in conflict with the Burgundian Gunnar (Gundaharius); the legends here have connections with those in the Nibelungenlied. But even here the saga is demonstrably remote from history, and combines characters who could never have met. Obviously, the fact that modern historians can identify errors does not prove that historical accuracy is impossible in heroic epic. It has been suggested that the broader background may often be more accurate (though less precisely described) than the central incidents;Footnote ¹³ but the probable limitations of this kind of evidence need to be recognized. It may be that the Homeric epics tell us more about the attitudes and outlook of the Ionians of the eighth or seventh century bc than about the actions of the Mycenaeans of the twelfth.

The bards of Homer's age and earlier were concerned to preserve the klea andrōn, the ‘glorious deeds of men’; it would seem that they believed these men to have existed in time long past, but recognized that they could not guarantee every detail of their accounts. Homer invokes the Muses in terms which admit his dependence on tradition and inspiration: ‘you are goddesses, you are present, you know all; we only hear the glorious tale, and have no knowledge’ (Il. 2.485–6). We may doubt whether Homer had authority in earlier poetry for every warrior whom he names in the Iliad; names came into his mind, and seemed to fit – a sign that the Muses were at his side, smiling upon his work. The borderline between adaptation of tradition and creative development, here as elsewhere, is impossible for us to draw, and may have been as hard for the original poet. The poet, then, is not primarily a chronicler. Nor does he attempt to connect the tale he tells with the present day, whether by offering certain characters as exemplary figures for his own age, or more explicitly by tracing the genealogy of royal families or particular patrons back to the heroes of old – a practice easily paralleled in other poetic traditions, and most influentially followed by Virgil in the Aeneid. Genealogy is important in the Iliad (less so in the more individualist Odyssey, where the hero is not competing among equals), but principally in self-assertive speeches, in which the heroes can declare themselves and define their status. There is little sign that Homer's audiences found special satisfaction in the exploits of their supposed ancestors. The most plausible case is the reference to the future destiny of Aeneas' descendants (picked up and developed by Roman readers), which has been read as addressed to self-styled Aeneadae in the Troad.Footnote ¹⁴ But this example is most notable for its isolation. For Homer and his predecessors, the glorious past mattered more than reflected glory in the present.Footnote ¹⁵

2. Poems behind poems

A major issue is whether the Homeric epics are representative or exceptional. The question can also be posed in chronological terms: do they represent the mature and typical form of the epic tradition, or a glorious final flowering? It is at any rate clear that the Iliad and the Odyssey presuppose, and were preceded by, other poems on a wide range of epic themes.Footnote ¹⁶ This is evident from references within the poems themselves. We may distinguish references to other episodes in the Trojan War and its aftermath – the abduction of Helen, the wound of Philoctetes, the gathering at Aulis, the first embassy to the Trojans, the death of Achilles, the suicide of Ajax, and so forth – from allusions to other heroic tales which are less closely related to the Trojan War: amongst these are the story of Bellerophon, the wrath of Meleager, the war of the Seven against Thebes, the birth of Heracles and his many labours in service to Eurystheus, and the Argonautic expedition.Footnote ¹⁷ Within the Trojan saga, it is assumed that the audience knows what the situation is, why the armies are at war (see esp. Il. 1.159), and who the characters are: thus Achilles storms away in anger ‘with Menoitios’ son and his comrades' (1.307), that is, with Patroclus, who is mentioned shortly afterwards without the identification being made explicit.Footnote ¹⁸

In a different category come stories about the gods. Again, it is obvious that the Olympians are familiar to the poet's audience, and that this familiarity derives not only from cult worship and visual representations but from previous song. The poems present the divine order as well established, with a clear hierarchy of power and division of provinces (1.533–5, 581; 15.185–217); we see not only from Hesiod's Theogony but also from a number of brief allusions in the poems that this was not an eternally fixed condition. There had been wars among the gods in earlier times: the overthrow of Cronos and the destruction of rival powers such as the Titans and Typhon are mentioned, but only in passing: the emphasis is on the security of Zeus's reign. The permanence of the gods in the Iliad (‘happy and existing forever’, 24.99) provides the essential foil to the short-lived and suffering mortals. Elsewhere, in the references to Zeus's adulteries and the complaints of Calypso, we catch hints of a freer intermingling of gods and mortals, with regular amatory encounters involving even the king of the gods (cf. Hes. Theog. 535; Cat. fr. 1.6–7). Again the Homeric epics are selective: of the leading figures only Achilles is the son of a divinity, and that privilege brings him little advantage: son of a goddess and a mortal, he stands between the two worlds, not wholly belonging to either.

Homer not only cited or alluded to other poems and stories but adapted them to play a part in his own work: for instance, a tale such as the death of Agamemnon, repeatedly cited but never retold at length in the Odyssey, is not evoked merely for the sake of digression, or even to advertise the range of the poet's repertoire; it is there to provide a contrast with the central story of Odysseus.Footnote ¹⁹ Other passages, such as the references in the Iliad to Heracles' experiences, can also be ingeniously brought into relation to the larger themes of the poem. Occasionally we may surmise that unusual or obscure versions are the inventions of the poet, to suit the argument of a particular speaker (e.g. Il. 1.396–406).Footnote ²⁰ Moreover, we can see that the epics make use of a number of themes which might be used elsewhere with other heroes: heroic wrath, the wanderer's return, the descent to the underworld, the forging of divine armour, are all paralleled elsewhere in the Greek tradition, though in some cases the other texts may be echoing Homer. One case in which the borrower seems clearly to be Homer concerns the wanderings of Odysseus, and particularly the perils of the Clashing Rocks. In Od. 12.69–70, Circe warns the hero that only one ship has ever passed through these Rocks safely, ‘the Argo, known to all men, sailing back from Aeëtes’ (70) – known, presumably, through poetry. There is good reason to think that other characters and events in these books of the Odyssey have also been taken over from the Argonautic saga: Circe herself, sister of Aeëtes and Medea's aunt, seems to belong to that world.Footnote ²¹ None of this deprives Homer of originality; rather, it shows him as an active participant in a tradition which thrives on competition and constant reworking of well-established themes.

The complexity of Homer's relation to his ‘sources’ may be illustrated in more detail from the story of Meleager, as told by Phoenix to Achilles in Book 9 of the Iliad.Footnote ²² Phoenix comments that, until now, no-one could have found fault with Achilles' anger at Agamemnon, but if he still holds out in his refusal to accept the compensation, he may suffer the fate of Meleager, who persisted too long in a similar resentful resignation from war, and lost all chance of gifts in the end. The narrative of Meleager includes a number of parallels with the sequence of events in the Iliad itself: in particular, the name of the hero's wife, who eventually persuades him to abandon his wrath, is given as Kleopatra (though reference is also made to another name), apparently an anticipation of the parallel role of Patroclus, who will succeed in persuading Achilles to relent.Footnote ²³ The motif of Meleager's anger, which is not found in other versions of the tale, appears to be Homer's invention, in order to bring the story into line with the main plot of the epic. A further point is that, in later versions, Meleager dies on the battlefield as a result of his mother's magical revenge for the slaying of her brothers: it is disputed whether Homer suppresses this finale (which would be ill-suited to Phoenix's persuasive task), or whether the version is a later addition to the story.Footnote ²⁴ It is, in any case, clear that Phoenix's narrative performs several functions. Like other ‘paradigms’, it offers an argument to show why the listener, here Achilles, should follow the speaker's advice. But Phoenix's perspective is limited, in two respects. First, he assumes that the worst thing that can happen is for Achilles to lose the chance of the gifts now being offered. In fact, in Book 19, Achilles will receive the gifts anyway, despite his persistence in anger; but they will be meaningless to him because of the far greater anger and grief aroused by the death of Patroclus. Second, in the more theological passage which precedes the Meleager tale, Phoenix advises Achilles to respect the spirits of Prayer, the Litai, or he may suffer the consequences (9.502–14). This simple moralizing has sometimes been taken as endorsed by the poet,Footnote ²⁵ but that reading seems to give undue importance to the words of a minor character; nor does Phoenix's ideal picture of divine justice receive much confirmation elsewhere in the poem. If Achilles is ‘punished’ for excess, it is in a more subtle and less overtly moralizing way. Finally, if the poet is indeed aware of some version in which Meleager's wrath ends in death, it is natural that Phoenix should suppress so grim a conclusion, but it is possible that the audience, alert to the omission, may anticipate the analogous fate of Achilles and relish the irony of the characters' efforts to evade the consequences of this parallel.Footnote ²⁶

3. Poetic language

The traditional nature of Homeric poetry is also clear on the level of language.Footnote ²⁷ The poems are composed in a version of Greek which was never spoken by a single people or in a single place: it is an amalgam of several dialects. Of these the predominant is Ionic, spoken by the Greeks of Asia Minor (Turkey), on some of the Aegean islands, such as Chios, and elsewhere. Second place goes to Aeolic, the form of Greek spoken by the people of northern Greece (Boeotia, Thessaly) and on Lesbos. Of lesser importance are the elements of diction perhaps originating in southern Greece in Mycenaean times and labelled Arcado-Cypriot (because this dialect survived in Arcadia and Cyprus in historical times).

How, when, and where this amalgam evolved is obscure to us. Its advantages to the poets are somewhat clearer. Ionic and Aeolic offer alternative words and morphological forms which coexist in the Homeric text, and these are usually of differing metrical shape. Thus we find five metrically differing forms of the verb ‘to be’ (εἶναι, ἔμεν, ἔμμεν, ἔμεναι, ἔμμεναι). The poet can employ these different forms to suit the needs of the metre. The common word for ‘ship’, ναῦς, has dative plural ναυσί in Attic, but the epic also uses νηυσί (Ionic), ναῦφι (an archaic form scarcely found after Homer), and even νήεσσι, an artificial combination involving Ionic stem and Aeolic termination. The poems are also inconsistent in using forms which derive from different periods: for instance, the old genitive in –οιο is found alongside the later and more familiar –ου. A key discovery in terms of the ‘layering’ of Homeric language was Richard Bentley's identification (1732) of the effects of the consonant ‘w’, represented in some inscriptions by the letter ϝ, digamma. Digammas do not occur in our manuscripts, because the sound was lost in Ionic at an early date (though it survived as a numeral). But many Homeric verses will only scan if a digamma is assumed at the start of a word that in the established Homeric text begins with a vowel (e.g. ἄναξ, οἴκος). Hence lines or formulae that need a digamma are likely to be older than those which are composed without respecting its presence. Bentley's detection of this principle, derided by his contemporaries, is now seen as one of the most brilliant insights in modern philology.

Metrical needs also explain variation between contracted and uncontracted forms of verbs, addition or omission of augment, and similar licences. Words can be modified or used where they are linguistically wrong but poetically convenient (e.g. εὐρύοπα, properly an accusative, is sometimes treated as nominative or vocative). But metre, however important, does not explain everything. Sometimes strange words and exotic forms seem to be used because they have poetic distinction. At least some of these will be archaisms, and their meaning was unclear to later Greeks, perhaps even to the Homeric poets themselves; misunderstandings sometimes generated fresh usage.Footnote ²⁸ The dual form Αἴαντε, ‘the two Ajaxes’, seems originally to have meant the brothers Ajax and Teucer, but in some passages is reinterpreted to apply to the greater and the lesser Ajax. Cult titles could become fossils (did Argeiphontes really mean ‘slayer of Argos’, or is this the guesswork of poets and scholars?). In Aristophanes' time the meaning of the expression ἀμενηνὰ κάρηνα was disputed in the schoolroom (Banqueters fr. 233 K-A), and Aristotle refers to the so-called ‘glosses’, words which could not be understood (διάκτορος, μέροψ, ἐριούνιος are examples) (Poet. 1459a9–10).

In short, Homer's language was not the Greek spoken by any of his audience. Nor was it unique to Homer, or even to a single region of the Greek world. The linguistic usage of Hesiod and other early hexameter poets closely resembles that of the Homeric poems. Generations of bards had collectively developed a special poetic language which could be modified for different subject matter but which was characterized by archaism and artificiality. Innovations undoubtedly occurred, but they are hard to plot when so much is uncertain about chronology, absolute or relative.Footnote ²⁹ What matters is that the epic style maintained a level of dignity and decorum appropriate to the grandeur of its typical subject matter.

4. Light from the East?

The origins of Greek epic may lie further back, and further east, than dark-age Ionia. Near Eastern texts, not least the famous ‘Epic of Gilgamesh’, yield remarkable parallels or precedents for Homeric conceptions, episodes, and even phraseology.Footnote ³⁰ The cosmogony implied in the account of Oceanus and Tethys (Il. 14.201, 302) recalls the Babylonian Enuma Elis, the epic of creation.Footnote ³¹ The conception of the gods as gathered in assembly is frequent in Akkadian and Hittite. The idea of the fatal letter found in the Bellerophon story is first attested in the story of David and Uriah (Il. 6.168–70, 2 Samuel 11); the Old Testament also offers a parallel to the failed seduction and revenge of Sthenoboiia (Il. 6.160–66; cf. Potiphar's wife, Genesis 39).Footnote ³² Even the supremely Homeric statement of the opposition between god and man has precedent in the heroic realism of Gilgamesh:

Above all, Gilgamesh loses his beloved comrade Enkidu, and grieves for him with Achillean devotion, raging ‘like a lioness deprived of her whelps’; elsewhere he journeys to the Underworld, like Heracles and Odysseus in Greek mythology: Gilgamesh, like the Greek heroes, is forced to recognize the limits of mortality.Footnote ³⁴ It has long been recognized that prominent features of Hesiod's theology (especially the myth of succession among the gods, and the sequence of the ages of men) are indebted to eastern myth;Footnote ³⁵ similarly, the Homeric poems need to be set against a wider background, and recent scholarship has charted some of the channels of communication – migrant craftsmen, diviners, healers, and not least wandering bards.Footnote ³⁶

Yet, for all the gratitude we may feel for this interdisciplinary research, questions and doubts still persist. Though the links with Gilgamesh are generally accepted, many other similarities between the poetry and mythology of the Near East and those of archaic Greece may strike the reader as inconclusive – parallels rather than sources. If warriors in different cultures are compared with wolves or birds of prey, if a king is described as ‘like a father to his people’, why should we seek a direct connection? Many such resemblances could easily arise independently. Are we dealing with diffusion or polygenetic development? More fundamental is the question of the effect of those ideas and details which have been imported. How are they adapted and used in their new context? Do motifs or concepts derived from the Near East delight the audience as exotic novelties, or is the effect one of mystery and awe? Some of the chief advocates of this approach have been content to accumulate material, but the critic needs to look harder at the nature of the resulting compound.Footnote ³⁷ Again, what actually happens when the mythic structures and motifs of the Near East are combined with those from the Indo-European tradition? Martin West has produced two extraordinarily rich books on these two streams of influence,Footnote ³⁸ but the naïve reader asks how we are to understand the relationship between them, and whether Homer and his fellow poets simply drew on one or other stream unconsciously or according to taste.

One episode in which the connections seem particularly clear and persuasive is the Deception of Zeus (though even here sceptical voices are audible).Footnote ³⁹ A cluster of suggestive motifs have been found – the references to the division of the universe among the gods and to the love-making of Oceanus and Tethys, and the magical kestos (an ornate and seductive feminine garment) which Aphrodite lends to Hera. If the scene involving Hera and Zeus does indeed owe something to the scene in which Ishtar (the Semitic goddess who corresponds to Aphrodite) attempts to seduce Gilgamesh, then the Homeric version has transformed an episode of antagonism and violent rejection into something more delicate and light-hearted: Gilgamesh's jeering list of Ishtar's former lovers becomes Zeus's fond reminiscences of his own amorous career.Footnote ⁴⁰ The direct connections can be questioned, but the comparison is still a revealing one for the understanding of Homer's divine comedy.

Other questions of influence and development arise with the stories that, like the Cyclops tale, are found frequently in other cultures but only in texts of a later date than Homer: do they all depend ultimately on the story as told in the Odyssey, do they emerge independently, or are Homer and some or all of the later versions dependent on a common source?Footnote ⁴¹ It is most unlikely that Homer invented the Cyclopes, who are mentioned elsewhere in the Odyssey in more incidental contexts; moreover, he fails to make explicit the chief feature of these monsters, and one which is crucial to the plot: that a Cyclops has only one eye (Hes. Theog. 143).Footnote ⁴² There are other ways in which the story seems to have been modified or adjusted to increase tension or to suit the ethos of the Odyssey. Here and elsewhere, the comparison with other versions, whether older or younger than Homer, serves to bring out the particular qualities and thematic concerns of the Homeric epics.Footnote ⁴³ Whether the perennial fascination of these themes depends on their intrinsic merits as material for exciting stories or on some deeper psychological or anthropological roots is a question which goes beyond the province of the literary critic.Footnote ⁴⁴

5. Early Greek hexameter poetry

However, the Iliad and the Odyssey did not exist in a vacuum. There were other important poets composing in the epic hexameter around the time of the great epics, and it is time to consider how far we can put the Homeric poems in a specifically Greek context.

To do so means giving more attention to several works which have already been mentioned in passing: Hesiod, the Homeric Hymns, and the Epic Cycle. Traditionally these were all seen as chronologically and generically later than the Homeric poems, but this view, while it may be broadly correct, has been challenged and in any case needs refining. We must distinguish between Hesiod's authentic surviving poems (the Theogony and the Works and Days) and the many other poems attributed to him in antiquity. Of the so-called Homeric Hymns, none can be ascribed to the same poet as the Iliad and the Odyssey, and many of the shorter items in the collection are much later. As for the Epic Cycle, none of the poems survive and they are doubtless of very varied dates. Finally, just as the Homeric epics spring from a tradition, so we must acknowledge that these other works probably had their predecessors, some of which may have predated the Iliad. The world of early hexameter poetry was not restricted to heroic epic: there was evidently also a tradition of theogonic poetry, of wisdom poetry, of hymnic narratives, of catalogue poetry. The example of Hesiod shows that one poet could compose in different hexameter genres. The best approach will be to discuss each of the above categories in turn, relating them to Homer.

6. Dating the poems

In discussion of all these texts, chronology presents severe problems. Modern readers do not always realize just how fragile and approximate the arguments for dating ancient texts usually are. Ancient testimony seldom helps, partly because of the lack of any universal dating system. In the late fifth century the Athenians recorded the victories at the dramatic festivals year by year in inscriptions, and these lists were used by later scholars and editors; hence we have reliable dates for many of the surviving plays. This is altogether exceptional, and nothing like this existed in earlier centuries. For the most part, modern scholarship has constructed a chronology based either on internal evidence – for instance, references to events, practices, or institutions which can be given an approximate date – or on the assumption that one work is definitely later than another (so that even if an absolute chronology cannot be established, some clarity is achieved in relative chronology).

Neither of these approaches is free of difficulties. Hesiod is traditionally dated by his participation (mentioned in Works and Days 650–9) in the funeral games for Amphidamas, a warrior of Chalcis in Euboea, which most interpreters associate with the Lelantine War of the late eighth or early seventh century, but we know so little about this war that some scholars have even questioned whether it really took place.Footnote ⁵⁸ It is essentially this datum that leads to Hesiod being given a date in most reference works around 700, and Homer is placed c.750–700 because he is generally thought of as being earlier than Hesiod. Our total ignorance of Homer's identity and origins may seem to justify an earlier date, although this largely arises from the anonymity of epic narrative style: how much would we know of Hesiod if he did not talk about himself? Current scholarship divides between those who hold to a date in the eighth century for the Iliad (whether c.750 or elsewhere) and those who see no obstacle to down-dating that poem to the first quarter of the seventh century or even later.

Archaeological evidence is also invoked, though this is treacherous ground.Footnote ⁵⁹ Agamemnon wields a Gorgon-shield, and the figure of the Gorgon is not found in Greek art before 700, and not used on shields until c.680; therefore, it is argued, the Iliad must be later than 680. Clearly this is an argument from silence; if another, earlier Gorgon-shield were to be found, its value would be nullified. Some of the other arguments for dating sections of the poem are unsoundFootnote ⁶⁰ or more speculative, but we should attach some weight to the judgement of archaeologists that Achilles' reference to the wealth of Delphi would be improbable much before c.700.Footnote ⁶¹

An important test case in the argument is the famous cup discovered on Pithecusae (Ischia), off the west coast of Italy, an early Greek colonial site. This cup is inscribed with verses which use formulae from epic: ‘I am the cup of Nestor, good to drink from; whoever drinks from this cup, straight away he will be filled with longing for Aphrodite of the lovely garland.’ The cup is currently dated c.740 bc.Footnote ⁶² We naturally think of the scene in Iliad 11.632–7, where Nestor's bond-slave prepares a cup of refreshment, and both the drink and the cup are described in some detail. If the inscription does indeed allude to this scene, the prior date of the Iliad is settled; but it can of course be argued that Nestor and his cup may have had a place in much earlier poetry, and that the scene in Book 11 draws on those earlier works.

The arguments for relative dating depend on our ability to detect the influence of one poem on another. Older discussions tended to assume that Homer was the earliest surviving Greek poet, and therefore all later resemblances were imitations of passages in Homer. But ‘echoes’ of Homeric scenes and lines are often illusory, since the imitation may be of epic material in general, and not of what we think of as ‘Homer’. Imitation may indeed be the wrong term, since we are often dealing with ‘typical’ scenes or sequences, part of the repertoire of bards in general. The very idea of ‘quotation’ or ‘allusion’ has been judged invalid by those who prefer to speak in terms of ‘traditional referentiality’ or ‘resonance’.Footnote ⁶³ On this model the poet does not refer to other poetic texts but to the poetic tradition in general: no poem has privileged status as a source. There are problems with this position, which seems to assume a steady state, a wonderfully huge and coherent body of tradition, while leaving little room for innovation and diversification. Some parallels and correspondences between extant texts do seem more specific and significant. The Odyssey surely does follow and imitate or echo the Iliad; the Hymn to Demeter seems influenced by both the Odyssey and the Hymn to Aphrodite.Footnote ⁶⁴ Similarities may cluster or cohere in such a way that it is hyper-sceptical to question a direct relationship; but this is a matter of scholarly judgement, and in an area where so little is known for certain, different scholars may legitimately differ in their view of the probabilities.

Richard Janko has scrutinized the language of Homer, Hesiod, and the Homeric Hymns from the standpoint of philology: applying a number of criteria of linguistic change (observance of digamma is one of the more readily understood), he finds confirmation of the ‘traditional’ sequence of relative dates, with Iliad first, then Odyssey, then Hesiod's Theogony and Works and Days, and later still the Hymns.Footnote ⁶⁵ This is reassuringly close to what might be considered likely on other grounds: Hesiod himself in the Works and Days refers back to a passage in the Theogony, and we shall see that there is good reason to believe the Odyssey a later poem than the Iliad.

Janko also offered some hypothetical ‘absolute’ dates, though strictly as illustrative possibilities. That is, if the Iliad were dated to 750–725, he believed that his findings would establish, within a margin for error, secure ‘absolute’ datings for each later work. Unfortunately this argument has been misunderstood as a claim that these dates were indeed established as the definite dates of composition.Footnote ⁶⁶ But even relative datings are valuable, if these statistical arguments are accepted as decisive. Not all have agreed that they are. It is objected that statistics take insufficient account of differences between poets and between different types of passage; that poets of different age or working in different local traditions may have different preferences in the epic linguistic repertoire, regardless of their chronological position.Footnote ⁶⁷ The field divides between those who treat Janko's linguistic arguments with particular respect and those who prefer to give greater weight to other criteria of dating.

7. The debate on the oral epic tradition: Milman Parry and his followers

The Homeric epics, whatever their actual date of composition, were at some stage committed to the written form in which they have come down to us. But did the original poets make use of writing, or are we dealing with a purely oral tradition, oral not only in terms of practice and performance but also in the process of transmission? This has been the predominant theme of Homeric scholarly debate through much of the last century.

The Homeric poems themselves portray heroes singing, and poets performing at feasts: Phemius at the court of Odysseus, Demodocus in the land of the Phaeacians. These bards seem not to need written texts: they perform with instrument in hand, a kitharis or a phorminx, conventionally translated ‘lyre’, and are probably imagined as chanting rather than singing throughout. They follow the directions of their aristocratic patrons: Penelope asks Phemius to choose another song, and the disguised Odysseus calls for a laudatory narrative of his own exploits in the sack of Troy. Telemachus mentions the attraction of a new song, and Demodocus is described as singing ‘the lay whose renown reached as far as broad heaven at that time’ (Od. 8.74). Fluency and sound organization are admired, in poetry as in rhetoric (Od. 8.489; contrast Il. 2.213, 246); still higher praise is given for the vividness of the narrative.Footnote ⁶⁸ Odysseus tells Demodocus that he has given a marvellous account of the sufferings of the Achaeans at Troy, ‘as though you yourself had been present, or heard it from another’ (Od. 8.491). Poetry gives delight and enchants: the audience sit spellbound (Od. 1.325–6; 17.518–20). It can also arouse grief and pain, particularly when it calls to mind past suffering (1.337–42, 8.521–31); at the same time, the listener should be strong enough to endure this sadness and reflect on the meaning of that suffering (1.353–5). The scenes involving bards and song can be combined to yield a coherent and subtle conception of poetry and its function.Footnote ⁶⁹ These representations of singers at work, honoured in their society and rewarded with a good cut of meat, may well bear some similarity to the situation of Homer himself. Ancient readers already made this assumption, suggesting that the blind Demodocus was a mirror image of Homer.Footnote ⁷⁰ Without embracing the biographical approach, modern scholars naturally scrutinize these scenes in the hope of recovering some picture of the poet's original setting. By combining the slender indications of the poems with later Greek texts on music, imaginative scholarship has even offered a reconstruction of the original ‘tune’ with which the poet accompanied his song.Footnote ⁷¹

All of this may be justifiable, but we should remember that Homer's bards are performing in a heroic world, and that the Odyssey in particular seems deliberately to elevate the status of the poet (e.g. Od. 22.344–9; cf. 17.383–5). These are idealized images of the poet and his audience, particularly the picture of Demodocus in the fairyland world of Scheria. We should not draw too close a connection between the poet himself and the singers in his poem. To take only the most prominent difference, the songs of the bards in the Odyssey are very short; Demodocus sings three in the course of a crowded day and still leaves plenty of time for Odysseus' own tale. By contrast, it is hard, perhaps impossible, to find an occasion on which the Iliad or the Odyssey could actually have been performed: even in an age which was accustomed to long poetic recitations and appreciated the singer's art, how could such lengthy works ever have been heard in their complete form? A major festival event, perhaps the Panionia at Mount Mycale near Ephesus, could have provided a suitably grand occasion,Footnote ⁷² but even this context hardly explains the sheer size of the epics. Yet the cumulative evidence for close integration and careful planning of both epics as monumental unified works compels us to accept that the ideal performance context would be one which allowed for total recitation of the entire poem.

We do not know who Homer was, and ancient biographies, largely fanciful in the manner of the genre, are no help; the fact that there was even speculation as to the origin of the name proves that antiquity was no wiser than we are about the man behind it.Footnote ⁷³ In the early period, as we saw, many other poems were attributed to Homer (even Thucydides assumes his authorship of the Hymn to Apollo), and it may be that the name was simply applied to all or most of the songs preserved by the group known as the Homeridae (‘sons of Homer’), who claimed descent from the great bard.Footnote ⁷⁴ Later readers became more discriminating; by the fourth century it could at least be asserted that he had composed only the Iliad and the Odyssey (Xen. Symp. 3.5). This became the standard view, although the so-called ‘separatists’ had their doubts.Footnote ⁷⁵ It remains convenient to use the name ‘Homer’ as shorthand for the creator of both epics, even if one supposes that the Odyssey is in fact by a later imitator.

Equal darkness covers the place and date of composition of the epics. Notoriously, different cities and islands of the eastern Mediterranean competed for the honour of being Homer's birthplace (this dispute is the subject of various epigrams in Anth. Pal. 16.293–99). The Ionian coastline or the island of Chios, with its later association with the ‘Homeridae’, have strong claims; by the time of Simonides (early fifth century bc), Homer could be cited as ‘the Chian’ (fr. 19 West²). Some of the descriptive passages in the Iliad concerning places near Troy probably rest on first-hand experience: in particular, the poet knows that Poseidon could have seen Troy from the topmost peak of Samothrace, though the island of Tenedos might have been thought to obstruct his view.Footnote ⁷⁶ That the Odyssey poet had similar knowledge of Odysseus' kingdom, the island of Ithaca, is less likely, given the geographical difficulties; and his acquaintance with remoter regions such as Egypt, important though these are in the Odyssey, was extremely hazy.Footnote ⁷⁷

The question of the date of composition of the Iliad and Odyssey is bound up with the problem of the process of creation and transmission.Footnote ⁷⁸ In antiquity the Homeric poems were seen as literary creations: that is, readers supposed that they had been created with the aid of writing, as was the work of later poets. Although the Jew Josephus declared that Homer was illiterate, this was for no good critical reason but formed part of a polemic in which he denigrated the classic texts of paganism in contrast with the authority and dignity of the scriptures (Against Apion 1.12). In modern times it was F. A. Wolf in his Prolegomena (1795)Footnote ⁷⁹ who set forth the fundamental objections to the hypothesis that Homer wrote his poems down, in the way in which we normally imagine a poet at work in a study. The arguments partly concerned the history of writing in early Greece – how early was the technique available, and how readily would the necessary materials come to hand? – and partly exploited the inconsistencies of language, style, and construction which Wolf and his successors believed that they could detect in both epics. Hence arose the long-standing conflict between the ‘analysts’ (those who cut up or separated the poems into layers or independent lays) and the ‘unitarians’ (those critics who opposed this procedure and insisted on the fundamental unity and quality of both epics).Footnote ⁸⁰

Although the concept of oral composition was current from the time of Wolf,Footnote ⁸¹ it was the work of Milman Parry which gave this notion a central place in the interpretation and study of Homer.Footnote ⁸² He addressed the question of what made Homer's poems so different from later epic, and his answer was that these poems were not simply the creations of an individual but the poems of a people, in that they were oral creations, drawing on a poetic tradition which went back over several lifetimes, even centuries. The evidence for this was above all linguistic, and Parry's most important work focused on the repetitions and variations in linguistic phrases or repeated ‘formulae’ which recur in the Homeric texts. ‘Much-enduring godlike Odysseus’ is an obvious example, tied to a particular character; another, which can be used in contexts involving many characters, is ‘He, wishing them well, addressed them and declared…’ (Il. 1.253, etc.). These recurring formulae most commonly consist of whole hexameter lines, or of half-lines beginning or ending at the ‘caesura’ or pause point in mid-line. In other words, they are phrases created to ease the task of a poet composing in hexameter verse.Footnote ⁸³ Homer has often been described as an improviser, a misleading term. On the level of metrical technique and formulaic vocabulary he does indeed work with tools which were devised to ease the process of oral composition; but the Iliad and the Odyssey could not have been produced without long preparation and premeditation. A corollary of this is that the formulae, besides being useful as building blocks, can also have an aesthetic contribution to make: ‘what is technically convenient can also be poetically effective’.Footnote ⁸⁴

The systems of formulaic phrases which Parry analysed were so numerous and showed such economy (in the sense that superfluous or duplicated phrases were kept to a minimum) that it became impossible to suppose that this was merely one poet's linguistic repertoire, all devised for his own use: it could only have been built up gradually by generations of bards. This conclusion was confirmed by philology. The language of Homer is an artificial tongue, never spoken by a single race or state at a given period; it is an amalgam of elements from different dialects and periods. Such a linguistic mixture could not have evolved in a single poet's lifetime; the poetic language was developed by poets over a period of generations. Rather than being the first and greatest of poets, Homer was shown to have composed with a long line of predecessors behind him, all of whom had contributed to the subject matter and diction of his own work. Some have rebelled against Parry's conclusions, feeling that they deny Homer any claim to originality or invention; but this is to misconceive the debate. Any artist imitates or makes use of his predecessors; the relation between tradition and innovation is a subtle and fruitful one, for the literate and for the oral poet.Footnote ⁸⁵ Nevertheless, the Homeric epics, though created in a tradition, stood out even in antiquity as exceptional in quality; this was, after all, the verdict of Aristotle and of Horace.Footnote ⁸⁶

Parry's researches were published gradually rather than being brought together in a book, and his untimely death in 1935 meant that some time elapsed before his ideas had their full impact. Nor have they ever influenced continental scholarship as much as they have Anglo-American work on Homer, though there are signs that things are changing. Much has been done to extend and develop his ideas, particularly in the analysis of formulaic systems. Parry was mainly concerned with the noun-epithet formulae (‘much-enduring godlike Odysseus’ and the like), but made some moves towards the examination of larger sequences in terms of recurrent patterns: motifs, themes, typical scenes, and so forth.Footnote ⁸⁷ Subsequent work in this area, notably by Fenik, has been especially valuable.Footnote ⁸⁸ By comparing a set of arming scenes or scenes in which a stranger arrives and is given welcome, episodes which occur in comparable form in different parts of Homer's work, we can see some of the poet's repertoire and observe how he plays variations on a ‘standard’ sequence; the overall similarity highlights differences which are often significant.Footnote ⁸⁹ Sometimes we may suspect that the poet is adopting a standard sequence in order to introduce a new or unexpected element, surprising his audience. In general, Parry's methods help to make us more alert to Homer's craft as a storyteller.

In the initial excitement arising from Parry's work large claims were sometimes made about the consequences for criticism: for instance, that the Homeric poems were entirely composed of formulae (a claim that can only be sustained by unacceptably expanding the concept of a formula); that Homer, composing orally and improvising, was incapable of advance preparation or subtlety of design; that oral poetry in any case did not seek the same effects as literary poetry, and could not be judged by the same criteria; that, indeed, a whole new form of ‘oral poetics’ is required.Footnote ⁹⁰

There has also inevitably been a reaction.Footnote ⁹¹ Some of Parry's formulations were clearly vulnerable, and some of his followers went too far in ruling out any kind of traditional ‘literary’ criticism. On the other hand, it is evident that the formulaic style, and other features of Homeric poetry which distinguish it from the epics of Apollonius or Virgil or Milton, do need to be taken into account in criticism. At the very least, Parryism can explain the frequency with which lines or phrases are repeated in the poems. It is surely necessary to go further, and to see Homer's narratives as composed for performance. Sometimes inconsistencies and oddities can be explained or excused by this assumption, though perfectionist critics may still be irked when ‘Homer nods’.Footnote ⁹² ‘Schedios is killed by Hector not once but twice; we do not weep for him. Chromios is killed three times, and innumerable eyes are dry.’Footnote ⁹³ Again, if the poets are using traditional plot lines and motifs, the problems or obscurities may sometimes be explained by the blending or juxtaposition of slightly different motifs (the difficulties in Iliad 2 concerning Agamemnon's testing of the troops may be an example of this, though characterization also plays a part). Both analysts and oralists should bear in mind that there are many anomalies and inconsistencies in works certainly composed with the aid of writing.Footnote ⁹⁴

On a verbal level, the analysts had complained that Homer often used language inappropriately, for instance in applying ‘stock’ epithets where they did not fit (Penelope's ‘fat’ hand in Od. 21.6 was a notorious example), and here the oral theory offered an answer, but an unsatisfying one for many readers. Parry went so far as to claim that the poet used these phrases so automatically that they no longer possessed a real meaning for him. What needs saying here is that the cases of inept use are remarkably rare, and need to be balanced against examples of Homer's conscious and tactful adjustment of formulae to suit a particular situation. A small but telling instance occurs at Iliad 16.298, when Zeus, who is often given the epithet ‘gatherer of the clouds’ (nephelēgereta), is described instead as ‘gatherer of the lightning’ (steropēgereta), a variant which is found only here. The reason is clear: the poet is describing how Zeus disperses the clouds, and the normal epithet would naturally jar.Footnote ⁹⁵

Parry stressed the element of tradition in the Homeric epics, while others have sought to isolate particularly creative uses or modifications of traditional material, or have even maintained that Homer is working against or remoulding the tradition.Footnote ⁹⁶ Another fruitful line of research is to explore the complexity of Homer's diction and styles: whereas Parryism seemed to imply a monolithic formulaic style, recent work has made clear that there are important distinctions between the style of the speeches and that of the narrative; moreover, different speakers may be characterized by different styles.Footnote ⁹⁷ In all these areas, what matters is to recognize that tradition and originality are combined, even if there are limits to how far we can define the relationship between the two.

A different approach, again seeking to vindicate the aesthetic as well as the functional qualities of the formulaic language, makes the traditional nature of that language a major resource for the poet. By using phraseology and typical episodes which he and others have used elsewhere, he evokes these other uses, recalling for the audience a world that they intuitively recognize. ‘Traditional referentiality’, to quote its chief exponent, ‘entails the invoking of a context that is enormously larger and more echoic than the text or work itself, that brings the lifeblood of generations of poems and performances to the individual performance or text.’Footnote ⁹⁸ Formulae and themes metonymically draw on a larger world: ‘pars pro toto’ is an expression that Foley has used repeatedly.Footnote ⁹⁹ The approach allows for variation and distortion of the norms; audiences are experienced enough to detect such abnormal usage. This is a very suggestive approach, though problems remain (such as how far different bards might differ in their styles and subjects, and how to allow for the evolution and invention within the tradition).

The other aspect of Parry's work which particularly caught the imagination of his followers was the comparative approach, pursued through fieldwork on existing oral poetic traditions, above all in Yugoslavia in the 1930s.Footnote ¹⁰⁰ There it proved possible to hear, record, and question bards composing without literacy, and to test the reliability of their memories of previous songs. Through this fieldwork, so Parry believed, a poetic tradition comparable with that of Homer could be observed and preserved. Fascinating though the comparison has proved, subsequent scholarship has been cautious in making deductions about the consequences for Homer. In quality and metrical subtlety the Yugoslav poems seem to fall short of Homer's; nor do these bards generally compose on such a large scale, except on special request.Footnote ¹⁰¹ The differences between ancient Ionia and modern Europe mean that special caution is needed in tracing the development or degeneration of an oral poetic tradition.Footnote ¹⁰² These analogies can only be suggestive, not probative. In the end, Parry's work on modern traditions may be found less valuable for Homeric studies than his analysis of Homer's own style and formulae.

8. But could Homer write?

Milman Parry tended to say or imply that his researches had ‘proved’ that Homer was an oral poet, in the sense of one who not only performed orally but made no use of writing in the composition of his works or at any stage. Modern readers of Homer – ‘readers’ is the key word – find it almost impossible to imagine the effort of memory involved in preserving long poetic texts such as these without any aide-mémoire. True, actors and operatic singers achieve feats scarcely less miraculous even in our own time, but they have a written text to memorize.

It can be argued, however, that this is the wrong model. According to Parry, and to many who have studied oral poetic traditions more widely than he, the singer is not memorizing and reproducing, but composing afresh, re-creating; each new performance is in a sense a new song, and tape recorders will show differences of some significance even when a singer does claim to be repeating ‘the same’ song. According to this view, the poet is in large part ‘improvising’, a word with disturbing implications. Did the Iliad, or parts thereof, go through this process in Homer's lifetime, being re-created many times? In that case, we may well suspect that the poet refined his work and added fresh beauties, rather than simply starting afresh each time; but how can we be sure that we have the ‘best’ version? The problem becomes greater when a different poet takes over the song, and remakes it anew. If a period of several generations intervenes before the poem is committed to writing, it may undergo extensive changes. In a way, it is not ‘Homer's’ poem. We may think of the paradox of the axe hanging on the wall for generations: if the blade has been replaced twice, the handle once, is it still ‘grandfather's axe’? Hence the attractions of assuming a literate Homer; a written text may be subject to the normal mishaps of miscopying but at least imposes some fixity. Lord and others have supposed that Homer himself was illiterate but dictated his works to willing disciples or scribes, so making it possible for the master's works to live on.Footnote ¹⁰³ Others have wished to go a step further and cut out the middleman: could not Homer himself have learned to write and used this opportunity to preserve, perhaps to enhance and enlarge, his greatest poem? It is an intriguing proposal that the discovery of writing itself could have been the stimulus which prompted a poet of exceptional gifts to create or develop a poem of extraordinary scope and length, in the knowledge that the instrument now existed to ensure his work's survival.Footnote ¹⁰⁴ More far-fetched is the hypothesis that the alphabet was itself devised specifically as a means of preserving epic poetry.Footnote ¹⁰⁵

Current orthodoxy places the introduction of the alphabet from the Phoenician into the Greek world in the late ninth or early eighth century: the earliest known Greek inscriptions come from c.770–750 bc.Footnote ¹⁰⁶ Whether we suppose that an alphabet, if available, would have been utilized to put Homer's poetry in a permanent form, or whether suitable writing materials could be obtained at this date, will partly depend on the importance which we suppose was attached to poetry in this period. Papyrus was an expensive import even in fifth-century Athens; leather has also been suggested.Footnote ¹⁰⁷ The difficulties are formidable, but the notion that the Iliad was consigned to writing from the start, at a date in the late eighth or early seventh century, is not a physical impossibility. The special attraction of this theory is that it explains the composition of so massive an epic, which would hardly have been composed for performance alone. Alternatively, some look hopefully to the so-called Homeridae, who, if they recognized the stature of their poet, may perhaps have sought to memorize rather than re-create.

My own view, then, would be that the seventh-century date, while possible, is not proven, and I am content to see the Iliad as reaching something very like its final form towards the end of the eighth century, and quite possibly being committed to writing within the poet's own lifetime. The Odyssey, whoever its author may be, follows in its wake, after an interval of at least a decade. Only gradually would the two poems become known beyond the region (wherever that was) in which they were created and recorded, and it would be a long time before they became ‘classic’ texts to which other poets and artists might consciously allude.Footnote ¹⁰⁸ The sheer length of these poems may have meant that they were at first actually less popular and would certainly have meant that most audiences would only have heard parts; geographical distance must also be taken into account (wherever they were composed, the process of diffusion would be different from modern ‘publication’ and far slower). Advocates of a seventh-century date point to the startling fact that Alcman, Stesichorus, and the Lesbian poets (c.600) are the earliest certain instances of allusion,Footnote ¹⁰⁹ but the yawning absence of early Greek poetry outside the hexameter tradition deprives that argument of a great deal of its force.

9. Sixth-century Athens

According to later sources it was in the sixth century, during the reign of the Athenian tyrant Pisistratus (d. 527) and his sons, that an authoritative written text of Homer was prepared, for use by rhapsodes who recited the epics at the great festival of the Panathenaea.Footnote ¹¹⁰ These sources are not only much later but mutually contradictory, and there have been attempts to dismiss the whole ‘Pisistratean recension’ as a fiction; but, although we should not envisage anything like a scholarly publication, it seems likely that some such centralized text was established, and that this stage in the tradition accounts for some of the ‘Atticisms’ or traces of Attic dialect in our texts of Homer. The idea that before this date no coherent epic songs existed, and that it was Pisistratus who brought various songs together and created ‘Homer’, was brilliantly asserted by Wolf and has been re-argued since,Footnote ¹¹¹ but seems fundamentally implausible: the Greeks thought of Homer as an older figure, and did not associate him with Athens; the Athenians themselves make a poor showing in the Iliad, which makes it unlikely that the poem was radically modified in that city; the assertions that the Athenians did in fact interpolate certain passages presupposes a pre-existing text; and the theory contradicts the chronological arguments of Janko. It is much more likely that Pisistratus or his sons ‘procured the first complete set of rolls to cross the Aegean’,Footnote ¹¹² and that their contribution was later exaggerated.

The earliest testimony to this activity by the Pisistratid family ascribes it to Hipparchus, his son: ‘he first conveyed the poems of Homer to this country, and made the rhapsodes go through them in order at the Panathenaia, taking turns at the task, as they still do today’ (Pl. Hipparch. 228b). These rhapsodes (‘weavers of song’), professional reciters of poetry which is not their own, are often seen as a new phenomenon, distinct from the genuine oral poet; the dividing line may not be quite so sharp.Footnote ¹¹³ Our view of them is biased by Plato, who satirizes the pretensions of one such figure, the complacent Ion, who can perform Homer to perfection but cannot explain him. The poets and their kinsmen no longer have a special authority over their poetry; by becoming a classic, Homer has entered the public domain.Footnote ¹¹⁴

A more specific allegation in antiquity, preserved in the Homeric scholia, was that Book 10 of the Iliad, often called the Doloneia, was not part of the original poem, but was separately composed and incorporated in the epic by Pisistratus (schol T. 10.1). The association of the change with Pisistratus may be merely the result of the earlier stories about his involvement with the Homeric text, but the suggestion that Book 10 is a later addition gains some support from the linguistic, structural, and ethical peculiarities of that book, as well as from the lack of cross-references to it in the rest of the Iliad. Although the tenth book presupposes the Iliad, the Iliad does not need the Doloneia.Footnote ¹¹⁵

This leads on to the general question of addition to and subtraction from the Homeric text during the course of its transmission.Footnote ¹¹⁶ If a whole book has been interpolated, what other changes may have been made at different times? Other cases were indeed cited, or at least problems were identified, by ancient critics: most alarmingly, the conclusion of the Odyssey (see below, p. 98).Footnote ¹¹⁷

10. Text and scholars

It hardly needs saying that commitment of the epics to writing, and even the possible establishment of an ‘authorized’ text for the Panathenaic recitations, did not ensure a calm and straightforward transmission of the text from that point to the present day. In virtually all textual traditions, the work comes down to us by various routes, and is recopied many times; even setting aside deliberate interference, it is an established fact that new errors will occur at every stage.Footnote ¹¹⁸ Although there is much that we cannot reconstruct, it seems clear that Homer circulated in many versions which differed significantly from one another during the centuries after Pisistratus: the evidence comes partly from quotations in other literary sources, partly from the early papyri, which often deviate from the text that later became ‘standard’, in particular including additional lines. The divergences are sometimes so marked that scholars refer to early extracts as the ‘wild’ papyri.Footnote ¹¹⁹ Alexandrian scholars did much work identifying possibly spurious lines and deciding between variant readings; after the third century bc the text seems to have settled into more or less the form which we have inherited, and it is usual to associate this development with the name of Aristarchus, one of the most distinguished scholars of Alexandria.Footnote ¹²⁰

In modern times, the analysts claimed that they had identified many later additions, and sought to reduce the poem to its earliest and purest state by purging it of such intrusions. This practice is misguided and in any case fruitless: even supposing that all such cases were objectively identified, we cannot assume that they have been simply added to a core which remains uncontaminated. If that assumption is not made, then the supposed original is irrecoverable. But in an oral tradition the very concept of an ‘original’ version may in any case be a chimaera. Nevertheless, many critics have continued to build up counter-arguments to the more drastic version of analytic criticism, showing how well integrated and intricately structured both epics in fact are. Scenes are demonstrably connected, later events are foreshadowed, explicit anticipation is common, and the poet has a firm control over an extended but carefully organized plot. Against those who maintained that oral composition must be loose and ‘paratactic’, this new wave of unitarians insists on the subtlety and complexity of Homer's art. This is a version of the argument from design: random combination or cumulative addition could not have resulted in this satisfyingly ordered microcosm. Schadewaldt and Reinhardt, Bannert and Reichel in German, Macleod and Taplin in English, have charted an impressive range of connections and cross-references which surely imply the master hand of a ‘monumental poet’ who planned the Iliad as a whole. Similar arguments have been applied to the Odyssey by Garvie and myself, among others. It remains true that some of these intricate structures are more visible and significant than others; some, it may be said, are perceptible only to the eye of faith. More disquieting is the argument that many of the thematic echoes and links which the unitarians detect can equally be explained by the conscious development of the poem by a succession of bards; to seek a single poet of surpassing genius is merely romantic misunderstanding of the nature of an oral tradition.Footnote ¹²¹ Such a hypothesis offers one possible explanation for the problems at the conclusion of the Odyssey.

11. Reception and interpretation

These and many other passages attest the special prestige of Homer in the classical period. Simonides could refer to him as ‘the Chian’ (fr. 19), but in time he became simply ‘the poet’ – no ambiguity was possible. Aristophanes' Aeschylus calls him ‘divine’ (Ran. 1034); the real Aeschylus was said to have declared his tragedies ‘fish-slices from the banquet of Homer’ (Ath. 8.347c). The highest praise for a later poet was to call him ‘Homeric’ (D. L. 4.20, on Sophocles). Prose writers were praised in the same terms (‘Longinus’ 13.3 on Herodotus; SEG 48.1330). Homer was not the pathfinder of tragedy alone; he can also be seen as the inspiration for historiography, which shares his concern to preserve the ‘glorious deeds of men’, for rhetoric, even for New Comedy. The ancient novel, with its parted lovers wandering in strange foreign lands but eventually reunited, is another obvious descendant.Footnote ¹²³ Quintilian and others compared Homer with the Ocean, which encircled the world and flowed into every lesser river.Footnote ¹²⁴

Homer has indeed been called ‘the Bible of the Greeks’, but this description requires some important qualifications. At no stage were the epics sacred texts: they were not kept in temples or interpreted by priests; although Homer was gaining pre-eminence by the end of the sixth century, other early texts were quoted and cited to illustrate mythical ‘facts’ or to back up an argument; as we have seen, Herodotus paired Homer with Hesiod as a formative influence on the Greeks' conception of the gods. Above all, Homer was not venerated without question. ‘Poets tell many a lie’, said the Athenian Solon, echoing Hesiod.Footnote ¹²⁵ The Pre-Socratic thinker Xenophanes was more specific: ‘Homer and Hesiod attributed to the gods all the things which bring reproach and disgrace among men – theft, adultery, deceit’ (B11). Heraclitus was more forthright: ‘Homer deserves to be thrown out of the public contests and given a beating’ (B42). Pindar expressed reservations as to the reliability of Homer's testimony about Odysseus: ‘I hold that the renown of Odysseus is more than his sufferings, because of Homer's sweet singing… Art beguiles and cheats with its tales’ (Nem. 7.20–3). Plato refers in the Republic to ‘an ancient quarrel between poetry and philosophy’ (10.607bc),Footnote ¹²⁶ and excises offending passages, particularly those which show gods or heroes in an ignoble light. The controversy over the immorality of Homer's gods continued throughout antiquity.Footnote ¹²⁷ But the morality of the heroes, the impetuous Achilles and the devious Odysseus, also aroused strong feelings. The tragic dramas of fifth-century Athens constantly rework Homeric themes and scrutinize epic values from the standpoint of a later age: enthusiastic admiration is mingled with questioning and even protest.Footnote ¹²⁸

Interpretation arises out of competition and polemic, for later poets sought to compete with or surpass the epic master. Hesiod, elaborating on the function of Strife in the community, speaks of competition between potters, carpenters, beggars, and bards (Op. 25–6).Footnote ¹²⁹ Hesiod's own reference to the Muses' lies has sometimes been taken as a jibe at epic narrative, probably wrongly; but Stesichorus revised Homer's mythology and his own, rejecting the tale that Helen went to Troy with Paris (PMG 192). In the sixth century Theagenes allegorized Homer; in the fifth Protagoras criticized his grammar. We recognize the beginnings of scholarly commentary.Footnote ¹³⁰ Many passages of Homer are quoted and discussed, not always in negative vein, by speakers in Plato's dialogues or the works of Xenophon. Although interpretation was often whimsical – selecting passages to prove a point without considering context, applying anachronistic criteria, or finding fault in order to show off – the eventual effect was to generate a rich and varied body of critical discussion.

The special importance of Homer meant that his poems, and especially the Iliad, were more closely studied than any other works of pagan antiquity. Historians debated the date of the fall of Troy, parodists spoofed epic phraseology, rhetoricians analysed the forms of argument employed in the speeches, students of mythology sought to reconcile or relate different legends, and grammarians puzzled over linguistic forms, tried to explain difficult words, and assessed the merits of variant readings. One work which would have shed much light on the state of interpretation in classical times was Aristotle's Homeric Problems, now lost except for a few extremely suggestive fragments; the much more sketchy treatment of some Homeric topics in a chapter of the Poetics gives an idea of the acumen with which Aristotle must have cut through much arid or ill-conceived argument.Footnote ¹³¹ After the lengthy complaints in Plato and others about the unworthy behaviour of Homeric gods and men, there is something refreshing about the curt response of Aristotle:

Above all, the ancient commentaries on Homer, which digest and pass on many of the observations of the great Hellenistic editors, provide not only precious mythological data and testimony to the textual tradition of the poems, but literary comment which may sometimes appear naïve, but retains interest and value.Footnote ¹³³ It is mere arrogance to suppose that only the most recent critical methods and concepts can offer any insight to the modern reader.Footnote ¹³⁴ Similarly, reception studies have brought home to many readers how much can be learned from the responses of earlier generations, whether amateurs or professionals and whether they came to Homer in the original or through the countless translations.Footnote ¹³⁵

Many of the topics considered by scholars and students of Homer today were pioneered by the ancient critics: the study of plot construction, characterization, rhetorical techniques, decorum and generic convention, preparation, foreshadowing, retardation, ironic double meanings – the list could be extended. The ancients were also more alert to matters of metre and audible effects, an area much neglected in our post-oral culture.Footnote ¹³⁶ Thucydides and Aristotle formulated the principle of poetic licence (Thuc. 1.10.3; Arist. Poet. 25); Longinus, in a brilliant paragraph, summed up unforgettably the differences between the Iliad and the Odyssey (On the Sublime, 9.11–15). We need not be too modest: some advances in critical understanding have become possible because we know more about philology and linguistic development than ancient scholars did; scientific archaeology has given us a far clearer insight into the dark ages and the earlier era of Mycenae than was available to Thucydides. Other improvements are technical. We have more readable printed texts, supplemented by systematic grammars, Homeric lexica, and computerized concordances: the study of Homeric vocabulary and formulaic diction is far more straightforward for us today than it was for Aristarchus or even for Wolf and his followers.

Some modern approaches, although perhaps anticipated in earlier times in certain specific observations or insights, do seem to represent fresh steps towards a deeper understanding. One which has already been mentioned is the study of creative or imaginative development of ‘stock’ formulae and themes. Another is the application of a sociological or anthropological approach to Homeric society, scrutinizing the institutions, the structures of the community, and above all the values and morality which motivate the characters in the epics.Footnote ¹³⁷ Closely related to this is the study of Homeric religion in relation to worship and cult in historical times: in one way this is the hardest of all tasks for modern readers, brought up in a completely different religious tradition (or in none), but there is another sense in which we have an advantage over the Greeks, since we are viewing the subject from outside, with a quite different perspective. After much ground-breaking and original research we may now hope that we have a better understanding of the relation between Homer's poetic presentation of the gods and the historical realities of ritual and cult.Footnote ¹³⁸ Valuable work has been done on the use of imagery, metaphor, and even symbolism (in the sense of poetic representation of physical objects which also conveys a deeper meaning), matters which at one time seemed too fanciful or sophisticated for Homer.Footnote ¹³⁹ The extended simile is not the only way in which Homer adorns and obliquely sheds light on his narrative.

Modern Homeric scholarship has also welcomed the advent of narratology – again, the systematic development of an insight already explicitly discussed in ancient texts. Narratology examines the sequence and emphases of a narrative: not just ‘what happens?’ but ‘how is it described?’.Footnote ¹⁴⁰ It also stresses the distinction between the poet's authorial viewpoint and that of his characters, and pays special attention to the passages in which the distinction is blurred or ambiguous. Instead of taking the story for granted, we are encouraged to look more closely at the way in which it is told. This has already had beneficial effect, not least in demonstrating the futility of the old debate as to whether Homer is ‘objective’ or not; the terminology of former discussion is seen to be hopelessly over-simplified.Footnote ¹⁴¹ A related enquiry is to examine the defining structures of Homer's plots, seminal for so many later works. A bold and intellectually demanding attempt to do just this has been made by Nick Lowe, drawing not only on narratology but on cognitive science and game theory: the contrasts he draws between the Iliad and the Odyssey as models for later plots in tragedy, comedy, and the novel are illuminating for the original poems and for later literature as well.Footnote ¹⁴²

These and other critical techniques cannot be adequately explored in the abstract; they need to be illustrated through examples drawn from the epics themselves. In the pages which follow only a few instances can be provided from the abundance of Homer's store. As Dryden wrote of Chaucer, ‘here is God's plenty’.