Chapter 1. “Winged Words”: How We Came to Have Our Iliad

Our Iliad consists—to quote a well-known Homeric formula—of “winged words” (ἔπεα πτερόεντα). An image in a Bronze Age fresco from Pylos suggests that as early as Mycenaean times, poetry in performance was conceived of as being in flight (Plate 1). [1] As we have seen, the nature of the Iliad—as a poem that is created only in performance—has profound implications for scholars seeking to establish an authoritative text. It has been suggested that, in Homeric epic, epea pteroenta are ones that, once uttered, cannot be taken back (Latacz 1968). And yet this is precisely what editors of Homer have throughout history hoped to do; we strive to recapture the authoritative performance and make it a text.
Building on the evolutionary model outlined in the introduction, I propose in this chapter to describe how a performance tradition that was already well underway in Mycenaean Greece eventually crystallized into what we know as the Iliad, and how that poem was transmitted in various media through more than three millennia so as to reach us in the textual form that we now have. The processes of transmission that I will describe are certainly not without controversy, and in my notes I will point to important recent discussions of these complex issues. My overarching aim is to provide an overview of the research that forms the foundation of the Homer Multitext and to make plain the scholarly principles on which the discussions of the multiformity of the Iliad in the following chapters depend. I will explore above all two related questions. How did a poem that was repeatedly composed and recomposed in performance become a written text? If the Iliad is indeed ultimately derived from such performances, what are implications for the textual tradition of this poem?

Performance of the Iliad and the Odyssey in Ancient Times

Homeric poetry was known to the ancient world primarily in performance. Most scholars would agree that in their earliest incarnations the poems that came to be our Iliad and Odyssey were composed orally and in the context of performance, and that this process was occurring over hundreds of years and throughout a large geographical area. These basic facts about the creation of the Iliad and the Odyssey came to be known through two different kinds of investigation.
First, there is the evidence that can be gleaned from the poems themselves. The meter of the poetry is the dactylic hexameter, and the language of the poems is a poetic composite of several dialects that was never spoken in any one time or place (Parry 1932, Palmer 1962, Horrocks 1997 and 2010:44–49, Bakker, forthcoming). The predominant layer consists of Ionic Greek forms, with the result that a large portion of the poem might be surmised to have come into shape in archaic Ionia (Frame 2009:551–620, §4.20–4.71). But there are verses that are demonstrably much earlier, in Arcado-Cypriote and Aeolic dialects, and others much later, with a veneer of Attic Greek. Phrases, half-verses, whole verses, and even whole scenes are repeated with a regularity that indicates this poetic composite was formed within a traditional system—that is to say, it could not be the product of one person. [2] The work of Milman Parry and Albert Lord demonstrated that this traditional system of formulaic language evolved to serve the needs of poets composing in performance (Ebbott, forthcoming).
There are, moreover, several passages within the poems that depict the performance of epic poetry, such as the performances of Phemios for the suitors in the house of Odysseus and those of Demodokos for the Phaeacians in the house of King Alkinoos. These passages show a bard performing at banquets, often taking requests for various episodes involving well-known heroes. Such passages in the Homeric texts that refer to occasions of performance are fascinating windows into how ancient audiences imagined the creation of epic poetry (Ebbott, forthcoming). Certainly the process is entirely oral. References to writing in the epics are famously few and mysterious. [3] The absence of writing in the composition of the poetry is also reflected in ancient biographical traditions about Homer that conceive of him as being blind (Graziosi 2002). In the Homeric Hymn to Apollo, the narrator proclaims that he is a blind man from Chios, and in Odyssey Book 8, the blind poet Demodokos who entertains the feasting Phaeacians (and whom many readers equate with “Homer”) is said to be compensated for blindness by his talent: “Him the muse had dearly loved, but she had given to him both good and evil, for though she had endowed him with a divine gift of song, she had robbed him of his eyesight” (Odyssey 8.63–64).
From the perspective of the internal audience of these performances, such as the suitors who are entertained by Phemios in Odysseus’ house on Ithaca in book 1 of the Odyssey, or the guests who listen in rapt silence to Demodokos in Alkinoos’ house in Phaeacia in book 8, the events narrated are well known but at the same time come from the relatively recent past. The Trojan War has come to an end only ten years before the performances depicted. But for the external audience, such as Athenians at the Panathenaic festival in Athens in the Classical period, the songs of Phemios and Demodokos are the traditional material of poets working within the epic tradition. Phemios sings nostoi, songs about the homeward voyages of the various heroes from Troy; Demodokos sings about a quarrel between Odysseus and Achilles at Troy, and later, the sack of Troy. It is worth noting that despite the differences in occasion, for the external audience, the compositional process of these notionally “past” performances and that of the present, framing performance is imagined to be the same. In this way the very ancient performances represented within the poems are placed on a continuum that connects all the way to the audience’s present. [4]
The second mode of inquiry into the creation of the Iliad and the Odyssey that I wish to highlight is the one I have already discussed in the introduction to this book. As we have seen, in the 1930s Milman Parry and his assistant Albert Lord went to Yugoslavia to study the oral epic tradition that at that time still flourished there, and soon came to understand that the Homeric poems were not only traditional in content, but were in fact oral poems—that is, products of sung performance rather than composition through the technology of writing. Their fieldwork allowed Parry and Lord to discover in “Homer” the existence of a sophisticated, traditional, economical, and above all oral system that enabled great literature to be composed in performance. They showed how a singer, trained in techniques that were centuries if not millennia old, could draw upon a storehouse of traditional language, tales, and heroic figures to compose epic poetry on any given occasion. [5]
The work of Parry and Lord suggests that though both the techniques used to compose epic and the stories that were told in epic were highly traditional, in its earliest stages there would have been a great deal of multiformity in the Greek oral epic tradition, with different performance traditions competing against one another to be considered authoritative. We witness just such a dynamic in Odyssey 8, when Odysseus specifically requests that Demodokos sing the song of the sack of Troy by means of the wooden horse. Odysseus says to Demodokos:
αἴ κεν δή μοι ταῦτα κατὰ μοῖραν καταλέξῃς,
αὐτίκ᾽ ἐγὼ πᾶσιν μυθήσομαι ἀνθρώποισιν,
ὡς ἄρα τοι πρόφρων θεὸς ὤπασε θέσπιν ἀοιδήν.
Odyssey 8.496–498
If you relate these things to me in accordance with destiny
Straightaway I will speak words before all men,
saying how a god readily bestowed upon you a wondrous song.
Odysseus’ words imply that there is a correct or authoritative way to perform this song, but also that other singers might perform it differently.
Here again, as we saw in connection with the passage from Iliad 20 with which I began my introduction, we find competing epic traditions evaluated in terms of fate or destiny (moira). [6] What is “fated” is the traditional and hence authoritative version of the story. For Odysseus, however, what is “tradition” for the external audience of the epic and even for the internal audience, the Phaeacians, is in fact his own life experiences, which took place only ten years prior to the current occasion of performance. He is therefore uniquely qualified to judge the authoritativeness of the current performance. His reaction, namely his tears, reveals to us that Demodokos has succeeded. [7] The passage as a whole suggests that what makes a song authoritative is not particulars of wording—that is, the “text” of the song—but a more deeply felt adherence to tradition that results in an emotional response from its listeners, who are themselves an integral part of the dynamic of tradition. [8] As Parry observed of the South Slavic epic song tradition, “The fame of a singer comes not from quitting the tradition but from putting it to the best use” (Parry 1971:335). So too will Demodokos become famous when Odysseus “speaks words before all” about what a gifted singer he is.
Such an understanding of tradition does not preclude multiformity. Lord observed for example that songs in the South Slavic tradition could be expanded or compressed as the occasion, time available, or the audience required, and that particular episodes could be included or not. He observed variations in different performances of the same singer and also when the notionally “same” song was performed by different singers. [9] In fact, the singers that Parry and Lord interviewed claimed to sing the song the same way every time, “word for word, and line for line,” but their fieldwork shows that this was never actually the case. As Lord notes, “What is of importance here is not the fact of exactness or lack of exactness, but the constant emphasis by the singer on his role in the tradition … the role of conserver of the tradition, the role of the defender of the historic truth of what is being sung; for if the singer changes what he has heard in its essence, he falsifies truth” (1960/2000:28).
In the ensuing chapters we will find many examples from the Greek epic tradition of a different kind of multiformity, in which competing and mutually incompatible alternatives present themselves. What is the cause of such multiformity? In my book Homeric Variations on a Lament by Briseis (2002:21–36 and 49–65), I argued that variations on the story of Achilles’ captive prize woman Briseis are fundamentally connected with local as opposed to Panhellenic epic traditions. In so doing I was following Gregory Nagy, who has shown that Archaic Greek poetry refers to Panhellenic myth and poetry as “truth” while local versions of stories about gods and heroes are pseudea, or “lies.” [10] Such a conception of truth and fiction is at work in the opening lines of the Homeric Hymn to Dionysos:
οἳ μὲν γὰρ Δρακάνῳ σ’, οἳ δ’ Ἰκάρῳ ἠνεμοέσσῃ
φάσ’, οἳ δ’ ἐν Νάξῳ, δῖον γένος, εἰραφιῶτα,
οἳ δέ σ’ ἐπ’ Ἀλφειᾠ ποταμᾠ βαθυδινήεντι
κυσαμένην Σεμέλην τεκέειν Διὶ τερπικεραύνῳ·
ἄλλοι δ’ ἐν Θήβῃσιν, ἄναξ, σε λέγουσι γενέσθαι,
ψευδόμενοι· σὲ δ’ ἔτικτε πατὴρ ἀνδρῶν τε θεῶν τε
πολλὸν ἀπ’ ἀνθρώπων, κρύπτων λευκώλενον Ἥρην.

For some say that you were born at Dracanum; others say on windy Icarus;
some say you were born in Naxos, divinely born, snatched from the thigh,
and others say that at the Alpheus River with deep eddies
Semele conceived and bore you to Zeus who delights in thunder.
Still others say, Lord, that you were born in Thebes,
But they lie. The father of gods and men bore you
far from men, hiding you from white-armed Hera.
As Nagy argues (1990b:43), “various legitimate local traditions are here being discounted as false in order to legitimize the one tradition that is acceptable to the poet’s audience.” The Iliad must likewise assert a version of the Achilles story that supersedes competing local variants. As I note with reference to Briseis, the Iliad does this in two ways. First, it leaves out or leaves obscure many local details about a romance between Achilles and the various girls from the many towns he captures. Second, the Iliad often includes within its own narrative allusions to other versions, thereby asserting the primacy of its own narrative at the expense of competing variations. A well-known example of this way of incorporating variation occurs at Iliad 5.634–647. In this battle exchange Tlepolemos taunts Sarpedon and claims that those who say that he is the son of Zeus are “liars”—pseudomenoi. As Miriam Carlisle has pointed out, in the Iliad Sarpedon is certainly the son of Zeus, but elsewhere there are traces of a competing versions of Sarpedon’s lineage. Tlepolemos’ use of pseudomenoi here is a way of referring to competing (and mutually exclusive) traditions, not objectively false tales. [11]
Another kind of evidence for the multiformity of Archaic Greek epic comes from Archaic and Classical Greek vase paintings, which sometimes depict “Homeric” scenes, but usually in ways that do not precisely correspond with how the story is told in the epics as they have come down to us (Snodgrass 1998). I argued in my 2002 book that while epic poets represent myth through verbal narrative, artists, drawing from the same storehouse of tradition, represent myth visually, each medium having its own rules and conventions (Dué 2002:32). The relationship between the vase paintings and poetry is therefore not linear but more like a triangle, with verbal and visual artists drawing on the same traditional narrative material and the two kinds of storytelling evolving alongside one another (and no doubt influencing one another). In this way I differ somewhat from other scholars, who have focused on the question of whether the Archaic vase painters knew “Homer” or “the Iliad.” Rather, I argue that vase paintings can be used as evidence for different ways of telling the story of the Trojan War in a time when the Iliad and the Odyssey were themselves still to a certain degree fluid. [12]
When we take into account the evidence of vase paintings, we see that the narrative traditions about the Trojan War and even the events recounted in the Iliad were indeed multiform in the seventh and sixth centuries BCE. In my 2002 book, I looked at vases that depict the taking of Briseis from Achilles (Dué 2002:28–31). In some versions of the story, Agamemnon evidently came to take Briseis himself, while in other versions (including our Iliad) he sends two heralds. Both versions survive on vase paintings: see, for example, a red-figure skyphos attributed to Macron, which depicts Agamemnon leading Briseis away by the wrist (Plate 2). The Iliad too hints at the multiformity of the tradition here. In Iliad 1, Agamemnon threatens to come himself to take Briseis away:
εἰ δέ κε μὴ δώωσιν ἐγὼ δέ κεν αὐτὸς ἕλωμαι
ἢ τεὸν ἢ Αἴαντος ἰὼν γέρας, ἢ Ὀδυσῆος
ἄξω ἑλών·
Iliad 1.137–139
But if [the Achaeans] do not give me [a prize] I myself will take one,
your prize, or the one of Ajax or Odysseus
I’ll go and take…

         ἐγὼ δέ κ’ ἄγω Βρισηΐδα καλλιπάρῃον
αὐτὸς ἰὼν κλισίηνδε τὸ σὸν γέρας
Iliad 1.184–185
          And I myself will go to the tent and take
the fair-cheeked Briseis, your prize.
In actuality (as our text presents it) Agamemnon sends two heralds to take Briseis (1.318–325). Yet elsewhere characters refer to the incident as if Agamemnon had come in person (1.356; 1.507; 2.240; 9.107; 19.89). It seems as if two versions have become conflated in the received textual tradition. Agamemnon himself suggests the possibility of an alternative version of these events when he first orders the two heralds to take Briseis:
                                     οὐδ’ Ἀγαμέμνων
λῆγ’ ἔριδος τὴν πρῶτον ἐπηπείλησ’ Ἀχιλῆϊ,
ἀλλ’ ὅ γε Ταλθύβιόν τε καὶ Εὐρυβάτην προσέειπε,
τώ οἱ ἔσαν κήρυκε καὶ ὀτρηρὼ θεράποντε·
ἔρχεσθον κλισίην Πηληϊάδεω Ἀχιλῆος·
χειρὸς ἑλόντ’ ἀγέμεν Βρισηΐδα καλλιπάρῃον·
εἰ δέ κε μὴ δώῃσιν ἐγὼ δέ κεν αὐτὸς ἕλωμαι
ἐλθὼν σὺν πλεόνεσσι· τό οἱ καὶ ῥίγιον ἔσται.
Iliad 1.318–325
                                     Nor did Agamemnon
let drop the contention with which he first threatened Achilles,
but he addressed Talthybius and Eurybates,
who were his two heralds and quick attendants:
“Go to the tent of Achilles the son of Peleus
and taking beautiful-cheeked Briseis by the hand bring her [to me].
But if he won’t give her I myself will take her,
coming with many men. And it will be a very chilling encounter.”
The Iliad, through the voice of Agamemnon, directly alludes to an alternative sequence of events that was current, I suggest, in the song culture when these verses were composed.
A more fundamental and structural variation on the story of the Iliad has been explored by Leonard Muellner (2012), who argues that surviving vase paintings represent an alternative epic tradition in which Achilles’ dominant emotion is akhos (‘sorrow, grief’), not mēnis (‘wrath’). In the akhos variation, Achilles covers his head and refuses to speak when his comrades beg him to return, which would be a substantial departure from Iliad 9 as we now know it. See for example Plate 3, showing the other side of the vase in Plate 2, where Achilles is veiled and looks down, not addressing his comrades. Muellner uncovers this variation in the vase paintings through the iconography of veiling, a visual way of representing akhos (cf. Penelope at Odyssey 1.334, Telemakhos in Odyssey 4.115, Odysseus in Odyssey 8.85). Achilles is no longer explicitly veiled in our texts, but at one time he may have been; Muellner argues that the akhos version has survived primarily in vase painting, while the mēnis version has survived in poetry. The akhos variation would seem to have very deep roots in the poetic tradition as well, however. As Muellner points out, the Catalogue of Ships in book 2 of the Iliad twice explains the reason for Achilles’ absence from battle. In the first (2.686–694), the emphasis is on Achilles’ grief for the taking of Briseis, while the second (2.768–773) emphasizes his wrath. [13] Even more fundamentally, the very name of Achilles contains the word akhos (Palmer 1963:79), and the Iliad shows “a pervasive nexus” between the word akhos and Ἀχιλ(λ)εύς, which is “integrated in the inherited formulaic system and hence deeply rooted in the epic tradition” (Nagy 1976:216).
Such glimpses of competing epic traditions point to an exceptionally creative and dynamic early performance history of the Homeric poems. Further below I will explore how such variation came to be largely screened out of the Iliad as we now have it. For now I simply want to observe that multiformity is a natural and expected phenomenon of Greek myth, and it should not surprise us that the artistic and poetic traditions that transmit these myths would likewise reflect this multiformity. The work of Parry and Lord and subsequent scholars shows that multiformity is a feature of oral epic traditions more generally, and helps us to understand by analogy the mechanisms by which the multiformity of myth entered the poetic system.

How Old Is the Iliad?

How far back does the performance tradition of the Iliad go? Possibly very far indeed. Greek is an Indo-European language, and Greek mythology and poetics are thought to have evolved out of an Indo-European tradition that predates the Greek language. The dactylic hexameter in which Homeric epic is composed is related to the lyric meters of Sanskrit poetry, and Gregory Nagy has argued that the Greek and Indic meters are cognate, stemming from Indo-European prototypes. [14] This relationship alone is suggestive of how very ancient the poetic traditions that produced our Iliad and Odyssey may be.
Linguists have shown, moreover, that it is possible, through examination of cognate formulas, to recover some of the poetics of the Indo-European tradition which Greek inherited. The first such formula ever discovered, and arguably the most telling, gave birth to the study of comparative Indo-European poetics. [15] The phrase κλέος ἄφθιτον ‘imperishable glory’ is attested both in Archaic Greek lyric poetry (Ibycus, PMG S151.47–48) and in epic (Iliad 9.413) and is cognate with the Indic śrávas … ákṣitam, attested in the Sanskrit Rig-Veda (1.9.7; see also 1.40.4, 8.103.5, and 9.66.7). The passage in which the phrase appears in the Iliad is a crucial one, in which Achilles lays out his justification for continuing to sit out of the war:
μήτηρ γάρ τέ μέ φησι θεὰ Θέτις ἀργυρόπεζα
διχθαδίας κῆρας φερέμεν θανάτοιο τέλος δέ.
εἰ μέν κ’ αὖθι μένων Τρώων πόλιν ἀμφιμάχωμαι,
ὤλετο μέν μοι νόστος, ἀτὰρ κλέος ἄφθιτον ἔσται·
εἰ δέ κεν οἴκαδ’ ἵκωμι φίλην ἐς πατρίδα γαῖαν,
ὤλετό μοι κλέος ἐσθλόν, ἐπὶ δηρὸν δέ μοι αἰὼν
ἔσσεται, οὐδέ κέ μ’ ὦκα τέλος θανάτοιο κιχείη.
Iliad 9.410–416
My mother the goddess Thetis of the shining feet tells me
that there are two ways in which I may meet my end.
If I stay here and fight around the city of Troy,
my homecoming is lost, but my glory in song [kleos] will be unwilting [aphthiton]:
whereas if I reach home and my dear fatherland,
my kleos is lost, but my life will be long,
and the outcome of death will not soon take me.
As we will see further below, the theme of the hero as a plant that blossoms beautifully and dies quickly is an important thread in Greek lament traditions (Dué 2007), and it is also a metaphor that encapsulates what glory means in the Iliad. Here Achilles reveals not only the crux of this choice of fates around which the Iliad itself is built, but also the driving principle of Greek epic song. The unwilting flower of epic poetry is contrasted with the necessarily mortal hero, whose death comes all too quickly (Nagy 1979:174–184; Nagy 2013:408–410). κλέος ἄφθιτον is just one example among many, but it reveals that an essential theme of the Iliad long predates the Iliad itself and suggests that there may be many aspects of the Greek epic poem that are equally old. [16]
Moving forward in time, we can find tantalizing archaeological and linguistic evidence to suggest that something like our Iliad was already been being performed in the Greek Bronze Age. The fresco of the lyre player from Bronze Age Pylos cited at the beginning of this chapter would seem to illustrate for us a performance tradition in the Mycenaean palaces, but given that the Linear B texts from Mycenae, Pylos, and elsewhere preserve nothing like literature, do we have any way of linking the content of our Iliad to those Mycenaean performances? I believe that we can. Let me begin by outlining some of the points of contact between our Iliad and the Bronze Age more generally, before moving on to explore the antiquity of the poetry and the story as we now have it.
The Aegean Bronze Age is the context in which our Iliad appears to be set, and the Greeks themselves believed that the Trojan War took place at the end of the Bronze Age. [17] The major heroes, such as Agamemnon and Nestor, generally come from places that were Mycenaean palatial centers. [18] Conversely, Athens, while it certainly existed in Mycenaean times (Iakovidis 1983:73–90, Hurwit 1999:67–84), does not seem to have been the major center of power and influence that it came to be in later times, and Athenian heroes do not play a very important role in the Iliad either (Dué 2006a:91–95). [19]
Likewise, the Catalogue of Ships in book 2 of the Iliad, the roster of record for the Trojan War, would seem to have its origins in the Bronze Age, although it certainly evolved to contain information from later eras, and scholars differ widely in their approach to the Catalogue’s relationship to history. [20] The Catalogue is controversial for many reasons, and we must proceed cautiously if we want to make any assertions about it all. It is by no means the only catalogue in surviving Homeric poetry, but at nearly four hundred verses in length it is by far the longest. Its placement in the narrative, at the start of a battle in the tenth year of the war, seems odd. [21] The list follows a circuitous geographical progression that begins in Boeotia, and the region of Boeotia and neighboring areas is disproportionately represented. [22] And although the Catalogue seems to reflect, for the most part, the political geography of Bronze Age Greece, there are many exceptions that are hard to explain. As Oliver Dickinson (2011: 154–155) has recently concluded:
All in all, the Catalogue is a strange compilation, and it does not seem possible to devise any rational explanation for its peculiarities. Here, as with many Homeric problems, the lack of pre-Homeric or contemporary “heroic” poetry is a major obstacle to the creation of plausible hypotheses. The most that can be safely said is that the Catalogue is likely to have been compiled from materials of different origins and dates and that care has been taken to harmonize it to other Greek traditions; but, although in some parts it does show a degree of historical consistency, on the whole it is most unlikely to bear any resemblance to the probable political configuration of those parts of Greece that it covers at any time period.
In fact, as I will explore in more detail in chapter 4, many of the controversies associated with the Catalogue of Ships can be at least partially explained if we understand it to have been composed as part of a conservative system of oral composition in performance that evolved over many centuries. Names and places that seem obscure to us would have had a prominent place in the epic tradition at one time or another. Some places that flourished in the Bronze Age no doubt were obscure already even for ancient audiences of the Archaic period, but a brief record of their local heroes was preserved and eventually crystallized as part of the Catalogue. In chapter 4 I will argue that in many ways such a catalogue can function as an index to the full diachronic expanse of the epic tradition itself.
Similarly, the work by E. S. Sherratt on the archaeological layers in the Iliad, which I cited in the introduction, indicates that the poem reflects the material culture of more than one time period. Some aspects of material culture referenced in the Iliad undoubtedly reflect the understanding of a Bronze Age world, while others are demonstrably Iron Age or later in their worldview. Two of Sherratt’s examples will demonstrate the complexities of the situation. In Iliad 23 Achilles offers a lump of iron as a prize in the funeral games for Patroklos:
αὐτὰρ Πηλεΐδης θῆκεν σόλον αὐτοχόωνον
ὃν πρὶν μὲν ῥίπτασκε μέγα σθένος Ἠετίωνος:
ἀλλ᾽ ἤτοι τὸν ἔπεφνε ποδάρκης δῖος Ἀχιλλεύς,
τὸν δ᾽ ἄγετ᾽ ἐν νήεσσι σὺν ἄλλοισι κτεάτεσσι.
στῆ δ᾽ ὀρθὸς καὶ μῦθον ἐν Ἀργείοισιν ἔειπεν:
‘ὄρνυσθ᾽ οἳ καὶ τούτου ἀέθλου πειρήσεσθε.
εἴ οἱ καὶ μάλα πολλὸν ἀπόπροθι πίονες ἀγροί,
ἕξει μιν καὶ πέντε περιπλομένους ἐνιαυτοὺς
χρεώμενος: οὐ μὲν γάρ οἱ ἀτεμβόμενός γε σιδήρου
ποιμὴν οὐδ᾽ ἀροτὴρ εἶσ᾽ ἐς πόλιν, ἀλλὰ παρέξει.
Iliad 23.826–835
Now the son of Peleus set in place a lump of pig-iron
which had once been the throwing-weight of Eetion in his great strength:
but now swift-footed brilliant Achilleus had slain him and taken
the weight away in the ships along with the other possessions.
He stood upright and spoke his word out among the Argives:
“Rise up, you who would endeavour to win this prize also.
For although the rich demesnes of him who wins it lie far off
indeed, yet for the succession of five years he will have it
to use; for his shepherd for want of iron will not have to go in
to the city for it, nor his ploughman either. This will supply them.” [23]
Sherratt assesses the passage as follows:
We have here what seems to be a rather odd situation. A lump of unworked iron … has been regarded as a prized possession for a long time, first as the favourite throwing-weight of a king and hero, then as something worth taking as a spoil of war, then as worth having as a prestigious prize. Yet it is suddenly—almost as an afterthought—recognized as having its prime desirability in a potential for utilitarian use, as a source of agricultural and pastoral tools. (Sherratt 1990:810)
In the Bronze Age, iron was an exotic luxury, minimally worked, valuable simply for being a precious metal. But after the Bronze Age, blades appear, and then daggers and swords. Sherratt concludes:
the first part of the passage in Iliad xxiii concerning the prize would seem to accord best with an attitude to iron which prevailed between the 16th and 12th centuries, while the second part belongs to a time from c. 1000 on when iron tools were regularly produced in Greece. (Sherratt 1990:811)
Shields are another item of material culture catalogued by Sherratt as evidence for the long range of time in which the Iliad must have been being composed. I quote Sherratt again:
Aias’ shield [is] an extraordinary affair which, as Aias enters the battle in Iliad vii.219, is described as tower-like (eüte purgon). It is made of seven layers of oxhide to which an eighth layer of bronze has been added, apparently as an afterthought (vii.223). As if this were not enough, this shield, a few lines further on in the thick of the fight, suddenly acquires a boss (messon epomphalion: vii.267) which has no part in the original description. Hector too has a very odd shield, at one point described as extending from his neck to his ankles (Iliad vi.117) and at another as completely circular (vii.250). That is a shield worth trying to imagine! (Sherratt 1990:810)
Sherratt points out that tower-like rectangular or figure-of-eight shields are known from the early Mycenaean period, but they became redundant once bronze body armor was introduced in the fourteenth century. It is only at the very end of the thirteenth century, in the period Sherratt terms “post-palatial,” that smaller hand-held shields (which could include a boss) are found in archaeological contexts (Sherratt 1990:811–812).
Sherratt’s analysis of the conflation of several historical eras in the Iliad’s representation of items of material culture demonstrates that in most cases it is not possible to isolate Bronze Age artifacts and conceptions of how to use them from those of later eras. The realia of multiple time periods have entered the formulaic diction of the system, with the result that they are all inextricably bound up in and interdependent on one another. Achilles’ long “Pelian ash” thrusting spear may well be, like the boar’s tusk helmet of Iliad 10, an heirloom handed down from an earlier generation, as the archaeological evidence would suggest (see Plate 4), but it can also be thrown at Aeneas like the throwing spears found only in post-palatial contexts (Sherratt 1990:811). Such differing conceptions of how items of material culture function reflect the poem’s very long history, with composition in performance taking place over many centuries, in a poetic tradition that would seem to stretch back at least as far as the sixteenth century BCE.

Narrative Layers: How Old Is the Story of the Iliad?

The work of Sherratt suggests that epic poetry featuring warriors and weapons was being composed at least as early as the sixteenth century BCE, but were the songs of the Mycenaeans anything like our Iliad? I will adduce two different types of evidence, visual and linguistic, that support the idea that the basic narrative structure of not only the Trojan War myth but even of the Iliad itself came into existence well before the end of the Bronze Age. [24]
The Iliad is an epic poem about a sequence of events that take places in the tenth year of the Trojan War, a war in which Greeks under the leadership of a Mycenaean leader attack and besiege a well-walled city in the Troad. Just such a tale may well be represented in Bronze Age art. Sarah Morris (1989) argued that the frescoes that survive in the so-called West House at the site of Akrotiri on the island of Thera represent a Trojan War–like narrative tradition that very likely would have found expression in verbal art of the time as well. The frescoes, which date to the seventeenth century BCE, [25] form a border along the upper walls of room 5 of the West House. The east and west segments feature Nilotic landscape scenes with fantastical creatures, including a griffin. The precise subject of the south segment is much debated, but it can generally be described as depicting ships crossing the sea between two towns, one larger than the other. The ships are festively decorated and seemingly peaceful (though they certainly convey warriors, with their boar’s tusk helmets hung up on pegs), and people watch them come and go from the walls of each town (see Plate 5). The north wall segment (see Plate 6), however, clearly depicts Mycenaean warriors with oxhide shields and boar’s tusk helmets attacking a walled city on a rocky coastline not unlike that of the nearby Anatolian peninsula. An attack from the sea seems to be taking place at the same time, with dead or drowning men depicted in the water near some arriving ships.
Morris interprets the miniature frescoes in the light of other surviving artifacts from the Bronze Age that juxtapose peace and war. She argues that the juxtaposition is an ancient theme that found expression in both the visual and the verbal art of the Bonze Age: “Its images are cognates of epic motifs, without being illustrations of particular episodes. … The rich repertoire of visual formulae and motifs mirror their proliferation not only in other visual arts but in poetic narratives now believed to be in circulation in the early Mycenaean age” (Morris 1989:515). The city at peace and the city at war featured on the shield of Achilles forged by Hephaistos in Iliad 18 may be a verbal attestation of this ancient visual theme, but so too are the many Homeric similes that invoke peaceful activities in the midst of battle and the former peace of Troy (τὸ πρὶν ἐπ’ εἰρήνης, Iliad 9.403), recalled at several key moments in the Iliad (Morris 1989:527–529).
Although Morris does not argue that particular Homeric episodes are being represented in the Theran frescoes (the walled city being attacked need not be Troy, for example), she nevertheless finds remarkable correspondences between the surviving Homeric epics and the stories seemingly depicted in the frescoes. The cities at peace on the south wall show, in Morris’s reading, a successful return home from a military expedition. Could the fresco be a visual counterpart to a nostos-style song of return like the Odyssey? As Morris notes, the harbor towns in this fresco “belong to the world of Aegean topography but also recall the poetic description of that world in Homer,” such as the harbors of Ithaca, the land of the Laestrygonians, Aiaia, and Scheria (Morris 1989:518–519). At the very least, the helmets hung up in every boat suggest a successful return from a military expedition. Of course, in our Odyssey, Odysseus is the only one of his men to return home, after he has lost every one of his ships. But the Theran fresco may tell a tale of homecoming in which many soldiers make it back safely. Morris points to Herakles’ expedition against Troy with 6 ships (alluded to in Iliad 7.451–453, 20.145–148, 21.442–457) as being of a scale on par with this fresco. [26]
The north wall features the “complementary theme of war” (Morris 1989:522), and parallels with the Iliadic tradition may be found here in virtually all surviving scenes of the combined land and sea attack depicted. [27] A scene that has been previously interpreted to be a ritual gathering at a peak sanctuary may instead be a council of elders and allies who debate what to do about the invading army (or alternatively, an embassy between the attackers and the city leaders). Several mounds and hills are featured as gathering places for the Trojans and their allies in the Iliad (e.g. the tomb of Ilos at Iliad 10.414–416). People watch the battle from the city walls, as in, for example, Iliad 3. The epic theme of the cattle raid (such as that narrated by Nestor in Iliad 11.669–684) may be featured in the top right of the surviving portion of the fresco. [28] Warriors with oxhide-covered tower shields, boar’s tusk helmets, and long thrusting spears march toward seemingly unsuspecting shepherds, while nearby women carry water in jugs on their heads. Similar episodes of ambush on unsuspecting shepherds are alluded to repeatedly in the Iliad; examples include Achilles’ ambushes of Andromache’s brothers (6.423–424), of Antiphos and Isos (two sons of Priam, 11.101–106), and of Aeneas (20.90–92). The wall depicted in the lower right may be just such a wall as the Achaeans are said to have built upon landing at Troy to protect their ships. [29] Similar collocations of land and sea attack (together with a cattle raid) can be found elsewhere in Bronze Age art, including on a stone rhyton from Epidaurus and a silver rhyton from Mycenae. [30] Close parallels with the fresco can also be seen in the depiction of the city under siege (i.e., the city at war) on the shield of Achilles (Iliad 18.509–540). As Morris concludes: “We can identify half a dozen thematic groups—arrival and departure by sea, council and battle, ambush and defense, the peaceful outskirts of a city, warriors landing and soldiers drowning—that correspond to epic traditions” (Morris 1989:530).
I would add to Morris’s analysis that the thematic pairing of going out on a military expedition (as featured on the north wall; see Plate 6)—be it an attack on a foreign city, an ambush, or a cattle raid—and the successful return (as featured on the south wall) is deeply embedded in the poetics of Homeric epic. [31] Mary Ebbott and I have explored in depth the ancient theme of ambush in Homeric epic, and we have argued (see especially Dué and Ebbott 2010:76–79 and 276–278) that the success or failure of an ambush is expressed in terms of whether the participants return home. Ambush episodes share thematic language and details with the theme of the journey, especially the nostos, or journey of homecoming. Spying missions or ambushes were conceived of as having the overall structure of a journey, and the two themes share the particular spatial movements of going out and, more importantly, coming home. If spies do not return, they cannot share the crucial information they were sent to obtain. Moreover, because of the stealth involved, if the warriors on an ambush die, their loved ones may not know where they died or be able to recover their bodies. We see this concern expressed similarly in connection with deaths that happen during a journey. For example, Telemakhos says that if his father had died at Troy, he would have had a proper burial and the kleos of a warrior, but since he apparently died before reaching home, he is without kleos (Odyssey 1.234–243). In Iliad 10, the other Achaeans anxiously await the homecoming of Diomedes and Odysseus, reflecting both the necessity of the return and the danger involved in such missions (Iliad 10.536–539). When they do arrive, they are greeted in the same language used elsewhere in the epics for welcoming those who have just completed a journey (Iliad 10.542). In contrast, when Dolon leaves the Trojan camp, the narrator says that he will not bring the information back to Hektor that he is being sent to gather; since he will not come back, we know immediately that his mission will not be a success (Iliad 10.336–337).
The similarities between homecoming journeys and ambush-themed missions are revealed in several places. When Priam is about to set out on a nighttime infiltration into the enemy camp in Iliad 24, Hecuba asks him to pray to Zeus for his arrival back home (τῆ σπεῖσον Διὶ πατρί, καὶ εὔχεο οἴκαδ’ ἱκέσθαι / ἂψ ἐκ δυσμενέων ἀνδρῶν, Iliad 24.287–288). The failure of an ambush, even in very compressed versions of ambush narratives, is expressed in terms of a failed return home. For example, in the ambush of Tydeus by fifty Thebans, we hear that “he killed them all, and released only one to return home” (πάντας ἔπεφν’, ἕνα δ’ οἶον ἵει οἶκον δὲ νέεσθαι, Iliad 4.397). Similarly, in the failed attempt to ambush Bellerophon, the ambushers never return home (κρίνας ἐκ Λυκίης εὐρείης φῶτας ἀρίστους / εἷσε λόχον· τοὶ δ’ οὔ τι πάλιν οἶκον δὲ νέοντο· / πάντας γὰρ κατέπεφνεν ἀμύμων Βελλεροφόντης, Iliad 6.188–190). So, when Diomedes is looking for a partner for the nocturnal spying mission in Iliad 10, he chooses Odysseus for the qualities that make him a good ambusher, especially his ability to get home:
τοῖς δ’ αὖτις μετέειπε βοὴν ἀγαθὸς Διομήδης·
εἰ μὲν δὴ ἕταρόν γε κελεύετέ μ’ αὐτὸν ἑλέσθαι,
πῶς ἂν ἔπειτ’ Ὀδυσῆος ἐγὼ θείοιο λαθοίμην,
οὗ περὶ μὲν πρόφρων κραδίη καὶ θυμὸς ἀγήνωρ
ἐν πάντεσσι πόνοισι, φιλεί δέ ἑ Παλλὰς Ἀθήνη.
τούτου γ’ ἑσπομένοιο καὶ ἐκ πυρὸς αἰθομένοιο
ἄμφω νοστήσαιμεν, ἐπεὶ περίοιδε νοῆσαι.
Iliad 10.241–247
Among them in turn Diomedes well-known for his battle-cry spoke,
“If you are ordering me to choose a companion myself,
how could I overlook god-like Odysseus,
whose heart and audacious spirit are especially ready
for every kind of labor, and Pallas Athena loves him?
With him accompanying me even from burning fire
we could return home [nostos], since he is an expert at devising [noos].”
Athena, the goddess who loves both Odysseus and Diomedes, tells Diomedes during the ambush to remember his homecoming (νόστου δὴ μνῆσαι μεγαθύμου Τυδέος υἱὲ, Iliad 10.509). Only that way can the mission be a success. [32]
The mission may also be something other than an ambush or raid, however. When Patroklos commences his fatal impersonation of Achilles in Iliad 16, Achilles prays to Zeus that Patroklos will be able to push the Trojans back from the Greek ships in face-to-face combat and return unscathed (ἀσκηθής, 16.247). Zeus grants only part of the prayer, however:
τῷ δ᾽ ἕτερον μὲν ἔδωκε πατήρ, ἕτερον δ᾽ ἀνένευσε:
νηῶν μέν οἱ ἀπώσασθαι πόλεμόν τε μάχην τε
δῶκε, σόον δ᾽ ἀνένευσε μάχης ἐξαπονέεσθαι.
Iliad 16.250–252
The father granted the one [part of his prayer] to him, but denied the other:
To push away war and battle from the ships
he granted, but he denied his safe return [nostos] from battle.
Patroklos’ failure to return home is emphasized in the lamentation that arises following his death:
                                     αὐτὰρ Ἀχαιοὶ
ἀσπασίως Πάτροκλον ὑπ᾽ ἐκ βελέων ἐρύσαντες
κάτθεσαν ἐν λεχέεσσι: φίλοι δ᾽ ἀμφέσταν ἑταῖροι
μυρόμενοι: μετὰ δέ σφι ποδώκης εἵπετ᾽ Ἀχιλλεὺς
δάκρυα θερμὰ χέων, ἐπεὶ εἴσιδε πιστὸν ἑταῖρον
κείμενον ἐν φέρτρῳ δεδαϊγμένον ὀξέϊ χαλκῷ,
τόν ῥ᾽ ἤτοι μὲν ἔπεμπε σὺν ἵπποισιν καὶ ὄχεσφιν
ἐς πόλεμον, οὐδ᾽ αὖτις ἐδέξατο νοστήσαντα.
Iliad 18.231–238
                                     But the Achaeans
gladly drew Patroklos out from under the arrows
and laid him on a litter. His dear comrades stood around him,
mourning. And among them swift-footed Achilles spoke
shedding hot tears when he saw his trusted comrade
lying on a bier torn by sharp bronze,
(his comrade) whom he sent with horses and chariot
to war, but he never received him back home again.
This same language is used of Achilles himself, who is lamented by his mother immediately upon Patroklos’ death, so closely are the two deaths intertwined:
τὸν μὲν ἐγὼ θρέψασα φυτὸν ὣς γουνῷ ἀλωῆς
νηυσὶν ἐπιπροέηκα κορωνίσιν Ἴλιον εἴσω
Τρωσὶ μαχησόμενον: τὸν δ᾽ οὐχ ὑποδέξομαι αὖτις
οἴκαδε νοστήσαντα δόμον Πηλήϊον εἴσω.
Iliad 18.57–60
After nourishing him like a plant on the hill of an orchard
I sent him forth in the hollow ships to Ilion
to fight with the Trojans. But I will not receive him again
returning back home to the house of Peleus.
These words are echoed by Achilles a few lines later, when he tells Thetis that he will re-enter battle:
νῦν δ᾽ ἵνα καὶ σοὶ πένθος ἐνὶ φρεσὶ μυρίον εἴη
παιδὸς ἀποφθιμένοιο, τὸν οὐχ ὑποδέξεαι αὖτις
οἴκαδε νοστήσαντ᾽, ἐπεὶ οὐδ᾽ ἐμὲ θυμὸς ἄνωγε
ζώειν οὐδ᾽ ἄνδρεσσι μετέμμεναι, αἴ κε μὴ Ἕκτωρ
πρῶτος ἐμῷ ὑπὸ δουρὶ τυπεὶς ἀπὸ θυμὸν ὀλέσσῃ,
Πατρόκλοιο δ᾽ ἕλωρα Μενοιτιάδεω ἀποτίσῃ.
Iliad 18.88–93
But as it is [it seems that it was all done] in order that there be infinite grief in your heart
for your child who has perished, the one who you may not receive again
returning back home, since my spirit won’t allow me
to live nor go among men unless Hektor
first struck by my spear loses his life,
and pays me for despoiling Patroklos, the son of Menoitios.
Like Patroklos, when Achilles re-enters battle in Iliad 20 it is on a mission from which he will never return home.
Coming back unscathed, ἀσκηθής, is used elsewhere in Homeric epic in connection with both a return from spying/ambush (Iliad 10.212) and a return home upon completion of a journey. It is used prominently in Odyssey 5 of the need to have both Telemakhos (Odyssey 5.26) and Odysseus (Odyssey 5.144, 5.168) return to their fatherland (πατρίδα γαῖαν). At the very beginning of the story of his wanderings, Odysseus says that he would have returned home unscathed if the sea currents had not prevented him from making his way to Ithaka (Odyssey 9.79–81). As Mary Ebbott and I have argued, arriving unscathed seems especially associated with sailing journeys (Dué and Ebbott 2010:278), as Odysseus also uses it in one of his Cretan lies, in this case for an easy sailing from Crete to Egypt (Odyssey 14.255). [33] In one additional case in the Odyssey, Odysseus tells Achilles in the underworld that Neoptolemos was very successful in battle and ambush, and later adds that he was never hurt in the fighting—he was ἀσκηθής (11.535), never touched by a weapon. Although the context has referred to both battle and ambush, the word is used to describe Neoptolemos as he boards his ship for the journey home, again displaying its deeply embedded connection to both sailing journeys and nostos in the formulaic language of epic.
Do the frescoes of the West House on Thera tell a story featuring a surprise military attack on a walled town together with the all-important safe return of the warriors? If Morris’s thesis is correct and the miniature frescoes of the West House really do depict an epic-like narrative, they would be among the oldest-known attestations of narrative in visual art. Moreover, the frescoes would seem to be the creation of Minoan (or Minoan-influenced) craftsmen on the Cycladic island of Thera telling a Mycenaean Greek story (or at least a story featuring Mycenaean Greeks). [34] How are we to understand the relationship between the epic story-telling traditions of the Mycenaean Greeks on the mainland and the artists of Thera at this early date? Morris is perhaps then rightly cautious about asserting specific correspondences to the story of the Iliad or the Odyssey in the Theran frescoes. Instead she posits that they are evidence for the existence at the time of Thera’s destruction of a poetic tradition with formulaic and thematic parallels to surviving Greek epic, a precursor tradition to that in which the Iliad and the Odyssey were ultimately formed (Morris 1989:530–531). [35] She follows Gregory Nagy (1979/1999:140), Emily Vermeule (1983:142), and others in suggesting that such a poetic tradition (and indeed that of the Trojan War/Iliad) may have been inspired ultimately by the activities of the Mycenaeans on the west coast of Anatolia over multiple centuries, as attested in Hittite archives and the material record of the Bronze Age (Morris 1989:531–535).
But there is other evidence to suggest that the deep structure of the Iliad is as old or nearly as old as the Theran frescoes. The main story of the Iliad as it has been handed down to us is not the taking of Helen, the Trojan War as a whole, or the sack of Troy, but rather Achilles’ cosmic wrath in the tenth year of the war, his withdrawal from battle, the devastating consequences, and his ultimate return to battle with knowledge of his own imminent death. The devastating consequences of Achilles’ wrath and withdrawal include something he never expected (17.401–411, 19.328–30), the death of his closest comrade, Patroklos, at the hands of Hektor. Building on the work of Whitman (1958), Van Brock (1959), and others, Gregory Nagy has shown that Patroklos is Achilles’ therapōn, a word which conveys a relationship of ritual substitution (see especially Nagy 1979/1999:94–117 and 289–295). This relationship becomes fulfilled when Patro-klos leads the Myrmidons into battle in place of Achilles, wearing Achilles’ armor. Patroklos’ subsequent death, the lamentation for him, and his funeral preview those of Achilles. [36] Achilles’ death does not take place within the narrative confines of the Iliad itself, but it is nonetheless enacted in the sacrificial death of Patroklos.
The deaths of Patroklos at the hands of Hektor, of Hektor at the hands of Achilles, and of Achilles at the hands of Paris are fundamentally interconnected in the Iliad. Achilles’ death is prophesied by various characters and foreshadowed repeatedly after the death of Patroklos, including by his mother, Thetis, who tells Achilles at Iliad 18.96: “Your own death awaits you straightaway after that of Hektor” (αὐτίκα γάρ τοι ἔπειτα μεθ’ ῞Εκτορα πότμος ἑτοῖμος). With his dying breath Patroklos foretells the death of Hektor, who in turn with his dying breath foretells the death of Achilles. This interconnectedness is made manifest in the exchange of armor that takes place among these three characters (Whitman 1958:199–203). Patroklos goes into battle wearing Achilles’ famous divinely crafted armor in order that he be mistaken for Achilles, Hektor strips Patroklos of this armor upon killing him and puts it on himself, whereupon Achilles must get a new set of divinely made armor, in which he confronts Hektor, who is still wearing his old set. Each of these characters then becomes an embodiment of Achilles; in killing Patroklos, Hektor arguably kills Achilles as well, and in killing Hektor, Achilles kills himself.
It was at one time common in Homeric scholarship to view this sophisticated structure as the finishing touch of the final poet in the tradition, that is, of “Homer,” and indeed the character of Patroklos has been interpreted by scholars who conceive of Homer in this way as being the creation of the master poet. [37] It was a central tenet of my 2002 book that such arguments are fundamentally flawed; they misunderstand the way composition in performance within a traditional system operates, and wrongly equate oral methods of composition with simplicity and a lack of both creativity and planning (see especially Dué 2002:83–89). But even setting aside such conceptual arguments about the nature of Homeric composition and the way that tradition operates, there is evidence to suggest that Patroklos’ story, especially as it intertwines with Hektor’s, is—rather than being the final touch of a master poet at the end of the tradition—in fact incredibly ancient.
The interconnectedness of the deaths of Patroklos and Hektor is reflected in the formulaic language used to tell their story. The same set of three verses marks the passing of each warrior:
Ὣς ἄρα μιν εἰπόντα τέλος θανάτοιο κάλυψε·
ψυχὴ δ’ ἐκ ῥεθέων πταμένη Ἄϊδος δὲ βεβήκει
ὃν πότμον γοόωσα λιποῦσ’ ἀνδροτῆτα καὶ ἥβην.
Iliad 16.855–857 and 22.361–363
When he had thus spoken the finality of death covered him;
His life-breath left his body and flitted down to the house of Hades,
lamenting its sad fate and leaving behind the youth and vigor of its manhood. [38]
The meter of lines 16.857/22.363 provides an indication of the great antiquity of the verse:
— — | — |— | — | — | — —
ὃν πότμον γοόωσα λιποῦσ’ ἀνδροτῆτα καὶ ἥβην.
The first syllable of the word ἀνδροτῆτα (‘manhood’) must be scanned short, something which is possible only if we assume the verse to have been composed before the linguistic changes that resulted in *anr̥tāta becoming ἀνδροτῆτα (cf. ἀνδρειπότες at Iliad 2.651, where the ἀνδρ- must similarly be scanned short). [39] I quote Calvert Watkins’s succinct explanation of the significance here: “Since we know that the change > or/ro (other dialects ar/ra) had taken place in Greek by the time of the Linear B tablets … the lines with *anr̥tāta could not have been composed any later than 1400 BC or so. They furnish us with a terminus ante quem for the fixation of the formulaic vehicle of a key feature of the thematic structure of the Iliad: these two deaths in equipoise.” [40]
In sum, two very different kinds of evidence, the frescoes of ancient Akrotiri and linguistic evidence embedded within the formulaic diction of the Iliad, suggest that a song tradition similar to our Iliad may have existed before the end of the Bronze Age, quite possibly even hundreds of years before. It is not possible to fully reconstruct such an Iliad, nor should we expect that any song performed at this time was fixed and unchanging. Surviving evidence indicates that not only were there numerous competing epics in circulation during the Bronze Age (the epic traditions that eventually resulted in the poems of the Epic Cycle), but very likely there were competing variations on the Iliad being composed and performed in different geographical locations. [41] In the introduction I noted the evolutionary models of Nagy and Sherratt, both of whom view the early phases of the Iliad and Odyssey traditions as being creative and generative, which is to say multiform. The Late Bronze Age would have been an extremely generative period of time for the creation of epic poetry, to use Sherratt’s terminology, [42] and, as we will see, it falls fully within period 1 of Nagy’s evolutionary model, the period he deems “relatively most fluid.”
I propose now to turn to the textual transmission of the poems. As we will see, the earliest textual witnesses of the Iliad and the Odyssey that have survived, the fragmentary papyri from Egypt (discussed further below), postdate this fluid tradition I have been discussing by nearly a thousand years, but nevertheless contain a great deal of variation that points to a creative and dynamic early history of the poems.

Performance and the Earliest Texts of the Iliad and the Odyssey

Lack of evidence prevents us from knowing a great deal about the transmission of epic poetry from the end of the Bronze Age through the so-called Dark Ages and into the Archaic period, but once again the poems themselves provide some clues. The song traditions of the Bronze Age continued—which is to say, the system in which epic poetry was created continued to evolve—even as the occasions and circumstances of performance changed. Sherratt views this time period as being one of “active generation and recreation of inherited tradition in several regions, leaving original remnants” (1990:817). And indeed the material and cultural world of the Iron Age is featured throughout both epics. Iron Age elements identified by Sherratt and others in the poems include the use of cremation for burial, armor (including slashing swords, horned helmets, double throwing spears, and bossed shields), fighting tactics, the use of iron for everyday objects, and architecture (Sherratt 1990:819). The maritime activities described in the stories told by Odysseus and others (which feature raiding, sacking of cities for plunder, slave-taking, and overseas trade) have also been seen to be reflections of an era of colonization and an Iron Age world. [43]
The adaptation of the Phoenician alphabet for the writing down of Greek and the advent of literacy in the Archaic period probably did not have a large impact on the oral poetic tradition and people’s reception of it for several centuries, although this is of course a subject of debate among scholars. [44] In Archaic and Classical Greece, the primary access to the Iliad and the Odyssey for most people would have been in the performances of professional rhapsodes (González 2013 and forthcoming [c]). What “texts” were these rhapsodes performing? And how do their performances relate to the texts we now have? Both Sherratt and Nagy attribute the fixation of the Iliad and the Odyssey and the loss of multiformity to performance, not the existence of written versions. When the primary venue for the transmission of Homeric poetry came to be the highly regulated competitive performances of the epics, first in regional festivals such as the Panionia and then at Panhellenic festivals, the Iliad and the Odyssey became “possessions,” which is to say, static and unchanging: “The emphasis had shifted from statement to possession. From now on the creative function of the bard (aoidos) gave way to the relaying rôle of the rhapsode.” [45]
This process took many centuries, however, and multiformity persisted long after the time most scholars generally begin to think of the Iliad and the Odyssey as fixed texts. [46] In its most basic form (as outlined in Nagy 2009:4–5) the stages of Nagy’s evolutionary model for the text fixation of the Iliad and the Odyssey are as follows: [47]
Period 1 of Homer was the relatively most fluid period, with no written texts, extending from the early second millennium BCE to the middle of the eighth century in the first millennium BCE. When I say that this is a most fluid period, I mean that epic was most susceptible to change in this period of its evolution.
Period 2 of Homer was a more formative, or Panhellenic, period, still without written texts, extending from the middle of the eighth century BCE to the middle of the sixth.
Period 3 of Homer was a definitive period, centralized in Athens, with potential texts in the sense of transcripts, extending from the middle of the sixth century BCE to the later part of the fourth. Somewhere near the start of this period, there was a reform of Homeric performance traditions in Athens.
Period 4 of Homer was a standardizing period, with texts in the sense of transcripts or even scripts, extending from the later part of the fourth century BCE to the middle of the second. Somewhere near the start of this period, there was another reform of Homeric performance traditions in Athens.
Period 5 of Homer was the relatively most rigid period, with texts as scripture, from the middle of the second century BCE onward. This period starts with the completion of the editorial work of Aristarchus of Samothrace on the Homeric texts, not long after 150 BCE or so.
Nagy posits that the Panathenaic festival in Athens, where strictly regulated contests in the performance of Homeric poetry were taking place as early as the sixth century BCE, was the context within which the Iliad and the Odyssey became crystallized into a relatively fixed form. [48] The resulting “Panathenaic” texts may have remained in flux for some time, influenced by a variety of factors, including political pressure from those in power (Frame 2009). The tyrant Peisistratos, for example, who is credited with the reorganization of the Panathenaia in 566 BCE and possibly the institution of rhapsodic contests, [49] is cited by several ancient sources as the organizer of a so-called Peisistratean recension, which produced the first written and authoritative text of the Homeric poems. [50] The story has a close affinity with tales in other cultures about how an oral tradition came to be authoritatively fixed in writing (Nagy 1996b:70–75). Nevertheless, there may be a clue here as to how the first written texts of the Iliad and the Odyssey were commissioned. The other epics that are known to have circulated in antiquity, often referred to collectively as the Epic Cycle, were not performed at the Panathenaia and have not survived in written form. [51]
In any case, at some point during the Archaic or Early Classical period in Athens, texts of the Iliad and the Odyssey began to be produced. [52] However this was done, it must have required great cost and Herculean effort, since writing would have been a new technology, the materials needed would have been expensive and difficult to acquire, literacy would have been restricted to an elite minority, and the performers would have been unaccustomed to slowing down their composition to meet the needs of a scribe. Milman Parry and Albert Lord faced many of these same obstacles when they attempted to capture in writing the performances of bards in the South Slavic epic song tradition during their fieldwork in Yugoslavia in the 1930s. Whatever texts were produced at this time were copied, and copied again and again for centuries. I have suggested above that a “Panathenaic text” may be the exemplar upon which the written texts of Classical antiquity were based. But our evidence suggests strongly that no single exemplar has reached us from Classical Athens, and there may have been many Panathenaic texts, given that each festival performance was a new composition, even if a highly regulated one.
With the exception of a few ancient quotations that survive in other texts (discussed in chapter 2 below), Homeric papyri are the oldest surviving witnesses to the text of Homer. These papyrus documents are all fragmentary, and range in date from as early as the third century BCE to the seventh century CE. The vast majority of the fragments were discovered in Egypt and now reside in collections located all over the world. The papyrus fragments of Homeric poetry reveal that the texts of the Iliad and the Odyssey were still somewhat fluid even after the Classical period in Athens. It is only starting around 150 BCE that the texts seem to become standardized, closely resembling the much later manuscripts of the medieval period (Haslam 1997). Because this date coincides with the height of the scholarly activity centered around the great Ptolemaic library in Egyptian Alexandria, it has been theorized that scholars such as Zenodotus, Aristophanes of Byzantium, and Aristarchus played an important role in establishing the relatively standardized text of Homer that is found in the medieval manuscripts of the Iliad and the Odyssey (Reynolds and Wilson 2014:8). Others have suggested alternate explanations, such as the rise of the book trade around this time, which must have resulted in greater diffusion of a common text (S. West 1967:11–17, Nagy 1996b:96–99, Schironi, forthcoming [a]).
Papyrus fragments are extremely significant for Homeric studies (Dué and Ebbott 2017). First, as already noted, they are ancient witnesses to the text of Homer. The medieval manuscript tradition of Homer begins with the tenth-century CE manuscripts of the Iliad known commonly as the Venetus A (Marcianus Graecus Z.454 [=822]) and D (Laurentianus 32.15). The Venetus A is the one upon which modern editions of the Iliad are primarily based, but some papyrus fragments predate the medieval tradition by as many as twelve hundred years. The papyrus fragments give us an otherwise irrecoverable picture of the Iliad and the Odyssey as they were performed and recorded in ancient times. As we will see in chapter 3, Homeric papyri reveal a state of the Homeric texts in antiquity that can be quite surprising. There are numerous verses in the papyri that are seemingly intrusive from the standpoint of the medieval transmission. These additional verses, the “plus verses,” are not present in the majority of the medieval manuscripts of the Iliad. Other verses that are canonical in the medieval manuscripts are absent from the papyri—these may be termed “minus verses.” Also prevalent is variation in the formulaic phrasing within lines. In other words, it seems from this most ancient evidence that the poems were performed and recorded in antiquity with a considerable amount of fluidity.
Because the papyri that predate 150 BCE present such surprising variations from the medieval texts of the Iliad and the Odyssey, they are often termed “wild.” This term, however, is very misleading from a historical point of view. As we will see, the quotations of the Iliad and the Odyssey that we find in such fourth-century BCE authors as Plato, Demosthenes, and Aeschines likewise present numerous plus verses, minus verses, and other significant variations from the medieval texts of Homer. In other words, the multiformity of the so-called wild texts of the oldest papyri is confirmed by the quotations, which present a similar picture of the Iliad and the Odyssey in antiquity (Dué 2001a). To put it still another way, the further back in time we go, the more multiform—the more “wild”—our text of Homer becomes. This is the exact opposite of what we should find for an author who composed in writing, where we would expect to see more uniformity in the textual witnesses the closer we came to the author’s lifetime.
The so-called wildness of the earliest papyri becomes more understandable and even expected when we take into account the evidence that the Homeric scholia provide from the work of the Alexandrian scholars. Scholia are notes in the margins of medieval manuscripts. [53] They are therefore part of the medieval transmission of Homeric poetry, but the notes derive ultimately from the work of the Alexandrian scholars and especially Aristarchus, the great second-century BCE scholar and editor whose critical work on the text of Homer is referred to throughout the scholia (Nagy 2004:3–24; Schironi 2018 and forthcoming [a and b]). These scholars had available to them at the library of Ptolemaic Alexandria a large number of Homeric texts, and it is clear that there were differences among them—that is to say, they contained multiforms. As Nagy reconstructs it, the editorial process of the great editors and heads of the library, Zenodotus, Aristophanes of Byzantium, and especially Aristarchus, was to comment upon these texts in scrolls of commentary (hupomnēmata), which were keyed to a standard text. [54] Such an understanding of Aristarchus’ methodology is supported by the following comment in the scholia of the Venetus A at 9.222 (the main text of the Iliad here reads αὐτὰρ ἐπεὶ πόσιος καὶ ἐδητύος ἐξ ἔρον ἕντο), which I quote with Nagy’s translation (2004:49):
ἄμεινον οὖν εἶχεν ἄν, φησὶν ὁ Ἀρίσταρχος, [εἰ] ἐγέγραπτο “ἂψ ἐπάσαντο” ἢ “αἶψ’ ἐπάσαντο,” ... ἀλλ’ ὅμως ὑπὸ περιττῆς εὐλαβείας οὐδὲν μετέθηκεν, ἐν πολλαῖς οὕτως εὑρὼν φερομένην τὴν γραφήν.
It would have been better, says Aristarchus, if it had been written “ἂψ ἐπάσαντο” or “αἶψ’ ἐπάσαντο”; nevertheless, because of his extreme caution, he changed nothing, having found in many of the texts this attested way of writing it.
It is only in subsequent eras that Aristarchus’ preferred readings from the hupomnēmata were inserted into editions put together by later scholars (Nagy 1996a:107–152 and forthcoming). Aristarchus himself does not seem to have ever published his own text of Homer with his own preferred readings. But even if he had, we would know from his commentaries about readings in the many other texts that were available to him, and so once again we are forced to confront the multiformity of the Homeric tradition. It is not my intention here to reargue Nagy’s theories about Aristarchus, but rather to make clear my own assumption, following Nagy, that the multiforms attributed in the scholia to Aristarchus and other scholars associated with the library of Alexandria are not editorial conjectures of the sort made by nineteenth-century philologists, but observations of multiformity culled from the wide array of texts available to them. [55]
Under Nagy’s model, the text of the Iliad had become largely fixed by the middle of the second century BCE. This fixation took place over several centuries and occurred primarily within the context of Athenian performance traditions and regulations, especially in connection with the Panathenaic festival (Nagy 2004:25–39), although as Nagy cautions, “the textual evidence allows us to reconstruct a Panathenaic tradition, relatively less multiform than other epic traditions, but this evidence cannot be reduced to a single ‘uniform’ Panathenaic text” (2004:38). It has been argued that just such an Athenian text is the one upon which Aristarchus commented (Schironi, forthcoming [a]). What we recognize from the scholia of our medieval manuscripts is that the scholars at the library of Alexandria who produced editions of and commentaries on the Iliad had access not only to something like a Panathenaic version but also to a wide range of editions from a wide range of people and places, including the so-called city editions (which are attributed to particular cities) and a number of editions attributed to particular individuals. At the same time, there were texts circulating that are referred to in the scholia as the koinai, the standard or common texts, and these seem to derive ultimately from Athens (Nagy, forthcoming). Some groups of texts are referred to in the scholia as being “more refined” (χαριέστεραι) while others are “more common/standard” (κοινότεραι), and a wide variety of other adjectives are employed as well to describe the various kinds of editions to which the Alexandrian scholars had access (see West 2001:37, Nagy 2004:88, and Schironi, forthcoming [a]). For now I only emphasize that even in periods 4 and 5 of Nagy’s model, the “standardizing” and “relatively most rigid” periods respectively, there is still considerable evidence for multiformity in the textual tradition, and indeed the bulk of the attested multiforms of the Iliad are found in sources deriving from these two periods.
The fact is that even if we accept (as I do) Nagy’s theory that by the second century BCE a relatively standard Panathenaic text of the Iliad and the Odyssey was in circulation, there are still multiple potential performance-derived sources for the multiforms attested in the scholarship of the great Alexandrian editors (such as Zenodotus and Aristarchus) and subsequent ancient scholars. Non-Panathenaic texts from places other than Athens, and possibly even Panathenaic texts from earlier time periods, were in circulation. We do not know the ultimate sources of editions attributed in the scholia to named individuals such as Rhianus of Crete or Antimachus. How were these editions created? Are they perhaps a product of specially commissioned performances? Or could they be the result of collation by an editor of a number of different texts? Because I and my collaborators on the Homer Multitext are not attempting to reconstruct a single text of the Iliad, I am interested not in demonstrating that such multiforms should be disregarded or dismissed but rather in what we can learn from them about the performance of the Iliad in various places and at various times. It will be my assumption throughout this work that every attested multiform is at least potentially derived from the performance traditions of Nagy’s period 4, in which transcripts of epic performances were likely being made and circulated, forming the raw material for the editions that were created by scholars at this time.
This multiformity, as I have argued, can be best understood if we conceive of our earliest texts as the products of a traditional system of composition in performance. The variations recorded in the early Homeric papyri and the Homeric scholia are the vestiges of a once-vibrant performance tradition of the Iliad and the Odyssey, in which no poem was ever composed, performed, or recorded in exactly the same way twice. In the earliest stages of the tradition that produced the Iliad, each performance would have resulted in an entirely new, and entirely traditional, composition. By the time of the earliest surviving papyrus fragments, the oral composition in performance tradition of Homeric epic poetry seems to have died out. But variations in the ancient textual tradition, which are, as I say, reflexes of this once oral and performative tradition, persisted for several more centuries. Moreover, performances of this poetry continued even as written texts were created, sold, and acquired as prestige objects. The variations preserved for us in the Homeric papyri and the scholia are a unique window into the performance tradition that generated them.

Medieval Transmission and Beyond

After papyrus ceased to be used, the Iliad and the Odyssey were copied onto parchment codices, like the Venetus A. [56] The Venetus A is the earliest-extant complete medieval manuscript of Homer, hand copied and assembled by one or more Byzantine Greek scribes in the tenth century CE. (The few medieval manuscripts that predate it contain commentary on and paraphrases of the poem but not a complete text.) The nearly two hundred Homeric manuscripts that succeed it are remarkable for the relative uniformity of their texts, and in this respect they differ considerably from the ancient witnesses. But although they do not vary in radical ways from one another, it is important to understand that the medieval manuscripts of Homer do not descend from a single exemplar, nor is there a medieval vulgate for the Iliad or the Odyssey. It is clear that a substantial number of texts survived the transfer from papyrus scrolls to parchment codices and that there were therefore multiple channels of transmission. What is not entirely clear is why the versions that survived resemble each other so closely. As I noted above, it has been postulated that the editorial activities of the scholars associated with the library at Alexandria played a role in the standardization of the Homeric text. But this theory does not entirely account for the continued multiformity of the text in the medieval period.
The Venetus A and several other deluxe manuscripts that survive from the tenth to the thirteenth centuries CE are invaluable to us for much more than their texts of the Iliad. As I have noted, these manuscripts contain not only the text of the poem but also excerpts from the scholarly commentaries of these same Alexandrian scholars, which are copied into its margins and between lines of the text. (Of the more than two hundred medieval manuscripts of the Iliad, only a small number are deluxe editions complete with scholia like the Venetus A.) These writings contain notes on the text that explain points of grammar and usage, the meaning of words, interpretation, and disputes about the authenticity of verses and the correct text. As we will see in chapter 4, the scholia often preserve examples of the multiformity of the Homeric tradition that are attested nowhere else. The material contained in these marginal notes derives from scholarly works that predate the manuscript’s construction by a thousand years or more. And like the ancient papyri, which give us their surprising picture of the state of the Homeric text in antiquity, the scholia give us a fascinating historical window into the evolution of the Iliad, and will be the source of many of the examples discussed in this book.

Establishing the Text

The first printed edition of the Greek text of the Iliad was made in Florence in 1488–1489 (without scholia), and this printing was the first crucial step toward making the Iliad widely available to a modern audience. It was edited by Demetrius Chalcondylas, who no doubt consulted several manuscripts available to him at that time (Proctor 1900:66, Sandys 1908:104, and Geanakoplos 1962:57–58). But the application to the Homeric texts of the techniques of textual criticism, in which scholars seek, using a variety of interpretive and deductive methods, to establish the correct text of an ancient author, would not begin in full force for three more centuries. Moreover, the medieval texts on which the early printed editions were based are not the same as the ones that we use now. The Venetus A manuscript of the Iliad was not published until 1788, when Jean-Baptiste d’Ansse de Villoison rediscovered it in the Marciana Library in Venice. The manuscript had been there for more than two centuries. It had belonged to the collection of the Greek scholar Cardinal Bassilios Bessarion (1403–1472), whose private library eventually became the core of the Marciana’s collection after his death. [57]
Between the early Renaissance and the late eighteenth century, when the Venetus A was published by Villoison, the figure of Homer was consistently the focal point of scholarly controversy. The so-called Homeric Question was in the process of being formulated. The “question” (which was, in reality, many questions) was concerned above all with the authorship and creation of the poems. Did the Iliad and the Odyssey have the same author? If so, when did he live? Could he write? Did he compose the poems in their entirety, or are parts of them interpolated by later authors? How did the poems come to be in the form that we now have them? Because of the wealth of scholia contained in the Venetus A, with the publication of that manuscript Homeric scholars of the late eighteenth century suddenly found themselves blessed with a treasure trove of information about what scholars of the second century BCE knew about Homer. It seemed that it would now be possible to reconstruct Homer and Homer’s original text, and that all of the Homeric questions could be solved (Nagy 2004:3–24).
This was the belief of Villoison, the editor of the editio princeps of the Venetus A, who viewed the scholia as an authoritative witness to an authoritative edition of Homer, constructed by the premier textual critic of Homer in antiquity, Aristarchus. But the views of another scholar, Friedrich August Wolf, proved to be more influential. In his 1795 work, Prolegomena ad Homerum, Wolf questioned the authoritativeness of the scholia and the work of the Alexandrian critics. Wolf argued, moreover, that the Homeric poems had been transmitted by rhapsodes in an oral tradition that had corrupted the texts irreparably over time. For these reasons, Wolf asserted, the true and genuine text of Homer could never be recovered. Wolf produced his own editions of the Greek texts of the Iliad and the Odyssey (1804–1807), relying heavily on the medieval transmission of the two poems and disregarding much of the textual work of the Alexandrians of antiquity. Wolf’s editions established a kind of notional medieval vulgate that continues to be followed to this day in modern printed editions: the line numbers in use by all modern editors are those of the highly conservative “Wolfian vulgate.”
The debates associated with the Homeric Question only intensified after the first publication of the Venetus A and its scholia in 1788 and have continued to dominate scholarly discussions of Homer ever since. Inevitably, how each modern editor of the Homeric poems answers this question has to a large extent determined the text that is printed. [58] When we seek to understand how we have come to have the text of the Iliad that we know today, it is important to understand that the debates between Villoison and Wolf and their scholarly successors about the figure of Homer and nature of the Homeric texts still guide the choices of modern editors.
The most recent critical printed edition of the Iliad is Martin West’s 1998–2000 Teubner edition. It is based on his conception of Homer, the poet and the man, as explained on the first page of his introduction: Ilias materiam continet iamdiu per ora cantorum diffusam, formam autem contextumque qualem nos novimus tum primum attinuit, cum conscripta est; quod ut fieret, unius munus fuit maximi poetae (“The Iliad contains material diffused through the mouths of singers for a long time, but the form and construction that we now know was first attained when it was written down. In order for this to happen, it was the work of one, very great poet”). West acknowledges here the oral tradition that furnished material on which the Iliad is based, but then says that our Iliad took its form when it was first written down. This was the work of a maximus poeta who could write. That the poet was also the writer is made clear as West continues: per multos annos, credo, elaboravit et, quae primum strictius composuit, deinceps novis episodiis insertis mirifice auxit ac dilatavit (“Throughout many years, I believe, he labored over it, and what he had at first put together concisely, he later wonderfully expanded and extended by inserting new episodes”). As I have argued already elsewhere, the insertion of credo here is telling (Dué 2006b). West is forced to admit, already on the first page, that his conception of Homer is a matter of faith. And because West believes in a maximus poeta, his goal is to restore the transmitted text as closely as possible to the composition of that poet. For him, the superfluity of other possible readings that survive from antiquity are of little interest.
Indeed, West’s editorial approach is in keeping with what all modern editors over the past three hundred years or so have done. But this methodology is problematic when it is applied to the Homeric texts (Dué and Ebbott 2009, 2010:153–165, and 2017). The practice of textual criticism, as applied to classical Greek texts, has historically had the goal of recovering the original composition of the author. To create a critical edition, a modern editor assembles a text by collating the various written witnesses to an ancient Greek text, understanding their relationship with each other, knowing the kinds and likelihoods of mistakes that can occur when texts are copied by hand, and, in the case of poetry, applying the rules and exceptions of the meter as well as grammar. The final published work will then represent what she or he thinks are the author’s own words (or as close to this as possible). An editor may follow one manuscript almost exclusively, or pick and choose between different manuscripts to compile what seems truest to the original. The editor also places in the apparatus criticus what she judges to be significant variants recorded in the witnesses. The reader must rely on the editor for the completeness of the apparatus in reporting variants. For a text that was composed and originally published in writing, this goal of recovering the original text may be valuable and productive (though the value has been challenged by recent scholarship), even if it may never be fully achieved because of the state of the evidence. [59]
But if, as I have argued, the Iliad was not composed in writing, this editorial system cannot be applied in the same way. The evidence outlined at the start of this chapter supports the thesis that the Homeric epics come from a long oral tradition in which they were created, performed, and re-performed, all without the technology of writing. This fundamental difference in the composition and history of this poetry means that we must adjust our assumptions in our understanding of the variations in the written record. What does it mean when we see variations, which still fit the meter and language of the poetry, in the witnesses to the texts? Instead of “mistakes” to be corrected, as an editor would treat them in the case of a text composed just once in writing, these variations are testaments to the system of language that underlies the composition in performance of the oral tradition.
For these reasons the editors of the Homer Multitext project ( do not attempt to answer questions of authorship, nor do they seek a single authoritative text. Rather, the Homer Multitext takes a historical point of departure and has as its central goal to make available an accurate picture of the transmission in all its complexity. The editors of the Multitext assert that poems that were part of a fluid and dynamic performance tradition should not be frozen in a single snapshot view, and instead intend to publish the Iliad at many different historical points of transmission. The implications of this approach for the present study are that it allows us to consider the Iliad diachronically—that is, as a system evolving over time and space. It allows us to approach each attested ancient “variant” from the medieval sources in the textual transmission of the Iliad as at the very least potential multiforms, which is to say, genuine products of the Homeric system of composition in performance. The multiforms explored in ensuing chapters can be analyzed within the contexts that produced them, and understood as historical witnesses to the thousand-year-long performance tradition from which they emerged. Rather than hide these witnesses in highly abbreviated Latin in small type at the bottom of the page, I seek in this book to magnify them and investigate what they can tell us about the poetics of the system we call “Homer.”
To sum up this wide-ranging historical overview of the history of the Iliad, I offer the following schema, which synthesizes the work of myself and the other scholars I have cited in this chapter on the need for a multitextual approach to the Iliad:
Multiformity of oral tradition → bottleneck 1 (Panionia) → bottleneck 2 (Panathenaia) → koinai (“standard” or “common” texts) → Alexandrian editorial work (esp. Aristarchus) → multiformity and discussion of multiformity in the surviving textual record → need for a multitext edition
Yes, the Homeric textual tradition is complicated and multiform, and it would be easier for editors if we didn’t have to acknowledge it. But as I hope to show in the following chapters, the complexity and multiformity offer us a wealth of riches for more fully and accurately appreciating the poetics of a poem that was over a thousand years in the making.


[ back ] 1. This chapter is a revised and expanded version of Dué 2009b. For an overview of the various meanings proposed for this famous phrase see Létoublon 1999. See also Martin 1989:30–37, Mackie 1996:56–58, and Beck 2005:41–43. A scholion to Euripides’ Orestes 1176 explains “winged words” this way: “because in performance/reading they leave the mouth like birds” (πτηνοὺς ἐνταῦθα λέγει τοὺς λόγους διὰ τὸ δίκην πτηνῶν ἀπέρχεσθαι τῇ προφορᾷ). (For προφορά as both “reading out loud” and “performance,” see Parsons 2012:19, whose translation I have adapted.) On the Pylos fresco see Lang 1969:79–80 (with additional bibliography ad loc.) and Immerwahr 1990:133–134. A Late Minoan IIIA ceramic pyxis from Aptera on Crete similarly depicts a lyre player surrounded by numerous large birds. (On the pyxis see also Betancourt 2007:190, who interprets the imagery differently than I do here, arguing that it is connected with cult ceremonies.)
[ back ] 2. The traditionality of the language of the Homeric poems was established by Milman Parry in his 1928 doctoral thesis, L’épithète traditionnelle dans Homère: Essai sur un problème de style homérique (= “The Traditional Epithet in Homer” in Parry 1971). See the introduction, p. 6.
[ back ] 3. On the “baneful signs” (σήματα λυγρὰ) carried by Bellerophon to the king of Lycia (related in Iliad 6.168ff.), see Nagy 1996b:14 and Shear 1998 with additional bibliography ad loc.
[ back ] 4. On this point see also Burgess 2005:127 and 2009:96.
[ back ] 5. The collected works of Milman Parry are published in The Making of Homeric Verse (edited by his son Adam Parry, Oxford, 1971). After Parry’s early death Lord went on to publish numerous articles and monographs over the course of six decades. See especially Lord 1948b, 1960/2000, and 1991 and the introduction by Mitchell and Nagy to the 2000 edition of Lord’s The Singer of Tales.
[ back ] 6. See also Nagy 1979:40–41 with further discussion of the semantic range of the word moira.
[ back ] 7. For Odysseus’ reaction, cf. Penelope at Odyssey 1.328–344 and Telemakhos at Odyssey 4.113–116. In Odyssey 4.219–233, Helen drugs the drinks with n e ¯penthes so that they can tell tales about Odysseus at Troy without weeping.
[ back ] 8. We may compare the words of the rhapsode Ion in Plato’s Ion (535e): δεῖ γάρ με καὶ σφόδρ᾽ αὐτοῖς τὸν νοῦν προσέχειν: ὡς ἐὰν μὲν κλάοντας αὐτοὺς καθίσω, αὐτὸς γελάσομαι ἀργύριον λαμβάνων, ἐὰν δὲ γελῶντας, αὐτὸς κλαύσομαι ἀργύριον ἀπολλύς (“I have to give [the audience] very close attention, for if I set them weeping, I myself will laugh when I get my money, but if they laugh, it is I who will weep at losing it” [translation after that of Cooper 1938, as reprinted in Hamilton and Cairns 1961]). On the traditional audience see Dué and Ebbott 2010:27–28 and 99–100.
[ back ] 9. See especially Lord 1960/2000:20–28 and 99–123, as well as Parry 1932:15 (= Parry 1971:336).
[ back ] 10. Nagy 1990a:60: “The alētheia of Greek poetry tends to contrast with the divergence of local poetic versions in the overarching process of achieving a convergent version acceptable to all Hellenes.” On pseudea as variant versions and not necessarily “lies”—that is, deliberate falsehoods—see Carlisle 1999.
[ back ] 11. See Carlisle 1999:62–64 as well as Pratt 1993:29–30 (with n29).
[ back ] 12. See Dué 2002:27–36 with references ad loc. On this point see also Lowenstam 2008:5 as well as González 2013:41–68. My approach differs from Lowenstam’s slightly in that he still prefers to see the vase painters as being directly inspired by poetry (as opposed to mythological traditions independent of poetry and the inherited conventions of vase painting), but he acknowledges that the poetic tradition was still multiform at this time and that there was not yet a canonical Iliad or Odyssey. This is a subtle distinction, however, and for the most part my own take and that of Lowenstam are in agreement.
[ back ] 13. Interestingly, Zenodotus athetized the first passage but not the second (Muellner 2012:213).
[ back ] 14. See Nagy 1974 and 1990a:459–464 and Katz 2005:25.
[ back ] 15. As Katz notes (2005:25), it was an observation by Adalbert Kuhn in 1853—that śrávas ... ákṣitam “fame ... unwithering” and kléos áphthiton are cognates—that led to the birth of comparative poetics as a scholarly discipline. On the subsequent scholarship see e.g. Haubold 2002, Volk 2002, and Katz 2010:361.
[ back ] 16. Scholarship since 1853 has gone on to demonstrate many additional ways in which the Iliad exhibits parallels with Sanskrit epic and other Indo-European poetic traditions. Overviews of the Indo-European poetic context out of which Greek epic may have arisen include Katz 2005 and 2010, West 2007, Cook (forthcoming), and Levaniouk (forthcoming [a]). Important early studies are those of Schmitt (1967 and 1968) and Durante (1976). See also Watkins 1995 and the work of Nagy, especially Householder and Nagy 1972 and Nagy 1974, 1990b, 2004, 2008a, and 2008b, Frame 1978 and 2009, Jamison 1991, 1994, 1996, 1999, and 2001, and Muellner 1996.
[ back ] 17. See, e.g., Herodotus 2.145 and 7.171, Thucydides 1.8–11, Eratosthenes (FGrHist 241 F 1d), with proposed dates ranging in ancient authors from 1334 (Douris, FGrHist 76 F 41) to 1135 BCE (Ephorus, FGrHist 70 F 223).
[ back ] 18. The newly discovered Bronze Age palace at Sparta solves an old mystery as to why a place so integral to the narrative tradition of the Trojan War did not appear to be of significance in Mycenaean times.
[ back ] 19. I do not mean to imply that Athens was completely insignificant in the Bronze Age or that it played no role in the song traditions then developing. Nagy (2015b) has argued that myths involving contact between Athens and the Minoans were already in circulation during the Mycenaean period on Crete.
[ back ] 20. On the Catalogue’s connections to the Bronze Age see Allen 1921, Page 1959, Simpson and Lazenby 1970, and Latacz 2004:219–249. Some scholars date the Catalogue considerably later, however; see Kirk 1985:129–170 with further references ad loc. and Sammons 2010:138–139. On the poetics of the catalogue and its relationship to the Iliad as a whole see Visser 1997, Sammons 2010:135–196, Tsagalis 2010b:329–330, and below, chapter 4. West (2007:70–71) includes catalogues of fighting groups and their leaders among the inherited elements of Indo-European poetic tradition in Greek epic.
[ back ] 21. See also Sammons 2010:137 with additional citations ad loc.
[ back ] 22. On the geographical progression, see Kirk 1985:183–186, Visser 1997, and Brügger, Stoevesandt, and Visser 2003:153–154.
[ back ] 23. Translation is that of Lattimore (1951), as quoted by Sherratt (1990).
[ back ] 24. For some other examples beyond what I discuss here of (Hittite) Bronze Age material seemingly preserved within the Greek epic poetry and myth associated with Troy, see Bachvarova 2016:351–356. She highlights in particular the names Paris and Priam and the close relationship between Apollo (~Apallu) and Troy. Bachvarova, however, sees this material as having entered into the Greek tradition via Near Eastern contact in the early Iron Age, and not as evidence for the antiquity of the Iliad or Greek epic poetry about Troy.
[ back ] 25. Bronze Age chronology and the dating of the volcanic eruption that destroyed the island of Thera have long been controversial, but scientific evidence suggests that the eruption took place between 1627 and 1600 BCE (Manning, Ramsey, et al. 2006).
[ back ] 26. See also Morris’ follow-up article (2000) as well as the work of Shaw (2000), who also interprets the miniature frescoes as a narrative involving departure, voyage, and homecoming, as I do here. For other explanations of what is depicted on the south wall, see Morgan 1988:143–165, Immerwahr 1990:74, Doumas 1992:45–97, and Nagy 2013:633–635 and 2015a.
[ back ] 27. For still further parallels with the epic tradition besides those adduced here, see Thomas and Conant 2007:44–51.
[ back ] 28. On the cattle raid as a traditional epic theme see (in addition to Morris 1989:527–528) Martin 2000 and Dué and Ebbott 2010:80–84.
[ back ] 29. This wall (the ones the Achaeans are said to have built upon landing at Troy) is mentioned in Thucydides (1.11.1) and Herodotus (2.118), and has been the source of much confusion and scholarly debate, especially as it relates to the Achaean wall built in Iliad 7. See, eg., Leaf 1900:297, Scodel 1982, Boyd 1995, Porter 2006, and Pache 2014, together with Morris 1989:524–526.
[ back ] 30. See Morris 1989:523–529 with further references ad loc.
[ back ] 31. Cf. Shaw 2000:272: “In interpreting the arrival of the ships as a return to their base—as I do here—it is important to recall the major role nostos played in early Greek literature involving war and naval expeditions.”
[ back ] 32. Conversely, in the Odyssey, Odysseus’ men nearly forget about their nostos in the land of the Lotus Eaters (νόστου τε λαθέσθαι, 9.97), and it is only Odysseus’ intervention by force that allows the men to make it back from that particular mission.
[ back ] 33. Cf. Solon fr. 19 (West), who says that Kypris sent him home unscathed (ἀσκηθῆ) in a fast ship for a good homecoming (nostos) to his own land.
[ back ] 34. For the reverse concept—that is, Mycenaean Greeks telling Minoan stories—see Nagy 2015c with further references ad loc.
[ back ] 35. Cf. Betancourt 2007:122: “The scenes are certainly specific enough to suggest that this is a narrative, and those familiar with the story could probably have identified individual ships with known captains and familiar crew members.… Pictorial cycles like this one can best succeed if they are supported by some type of oral or written narratives, like epic poetry.”
[ back ] 36. See also Burgess 2009:72–97 and passim. On the interrelationships between Hektor, Patroklos, and Achilles see especially Nagy 1979 and 1997, as well as Whitman 1958:199–203, Sinos 1980, and Lowenstam 1981.
[ back ] 37. On Patroklos as an invented character, see Howald 1924:411–12 as well as Dihle 1970:159–160 and bibliography ad loc. More recent discussions include Allan 2005 and Burgess 1997 and 2006.
[ back ] 38. The translation is based loosely on that of Samuel Butler.
[ back ] 39. See Watkins 1995:499 citing Wackernagel 1953:1116n1 and 1170n1 (originally published in 1909), Leumann 1950:221n16, and West 1982:15.
[ back ] 40. Watkins 1995:499, citing Ruijgh 1967:69, Wathelet 1970:171ff., Watkins 1987, and West 1988:156–157. Recently Timothy Barnes (2011:10) has argued against this interpretation of the history of the formula, arguing instead for *ἀμ(β)ροτῆτα καὶ ἥβην “as the model upon which the attested ἀνδροτῆτα καὶ ἥβην was coined, at a relatively recent date, by a poet aiming at an impressive line to close an important scene: the death of Achilles.” (Barnes reconstructs this formula, not attested in surviving Homeric epic poetry, on the basis of an Avestan counterpart.) One of Barnes’s primary objections to viewing ἀνδροτῆτα καὶ ἥβην as a vestige of Mycenaean hexameter poetry is the implications for our understanding of the composition and history of the Iliad. I do not share Barnes’s objections and fully embrace the implications. Contra Barnes (2011:1–2), I assert the continuity of the Iliadic tradition over more than seven hundred years. The song persisted precisely because it was traditional, while simultaneously flexible, multiform, and ever evolving. These concepts will be explored in more detail in the ensuing chapters.
[ back ] 41. On the antiquity of the poetic traditions that resulted in the poems we now know as the Epic Cycle (which survive only in fragments and the testimony of ancient witnesses), see especially Burgess 2001 and 2009. For more on the concept of competing Iliads—which is to say, competing epic versions of the mythological story of Achilles’ wrath told in our Iliad, see Dué 2002:21–36.
[ back ] 42. Sherratt makes a distinction between the pre-palatial and early palatial periods, which she considers periods of active creation, and the palatial period, which she describes as being one of “active maintenance of inherited tradition with little modification,” for which see her arguments at Sherratt 1990:817–819.
[ back ] 43. On the Iron Age reflections in the Homerics epics see Finley 1954, Sherratt 1990:819–820, Morris 1997, Dougherty 2001, Whitley (forthcoming), and Antonaccio (forthcoming).
[ back ] 44. See Powell 1991 and 1997a for the view that the Greek alphabet was adapted for the purposes of writing down the Homeric epics. The problem with Powell’s thesis, as noted for example by Nagy in his exchange with Powell in the Bryn Mawr Classical Review (97.4.18), is that: “Turning an ‘oral poem’ into a text does not by itself stop the oral tradition that created the ‘oral poem.’ The oral traditions of composition-in-performance can be independent of a writing technology that turns compositions into texts. This fundamental thesis is evident throughout Albert Lord’s last book, The Singer Resumes the Tale (1995, especially chapters 1, 8 and 10). It is supported by a wealth of comparative evidence drawn by various scholars from various historical contexts, some of which are adduced in PP [= Nagy 1996a].” On Homer and the alphabet see also González (forthcoming [b]).
[ back ] 45. Sherratt 1990:821. For the role of performance at festivals in this process see Ebbott (forthcoming) and Nagy (forthcoming). For the Panionia’s role in the shaping of the Homeric epics as we now have them see Frame 2009:551–620, §4.20–4.71.
[ back ] 46. Scholars differ in their dating of the text fixation, depending on how they conceptualize “Homer.” Some imagine a dictating Homer as early as 800 BCE, and others date the text considerably later. The editors of the Homer Multitext, following Nagy’s arguments discussed here, understand the process to have been a gradual one. On the concept of a dictating Homer see also Dué 2006b, González 2013:15–70, Ready 2015, and below, note 52.
[ back ] 47. See also the introduction, pp. 11–12.
[ back ] 48. On the “Panathenaic bottleneck” see especially Nagy 2001 and 2002:3–35. See also Seaford 1994:151–153. The term “crystallization” is Nagy’s, but others have employed the metaphor as well; see, e.g., Sherratt 1990:820 and Cook 1995:5.
[ back ] 49. The ancient evidence comes from the pseudo-Platonic dialogue Hipparchus 228–229.
[ back ] 50. See Anecdota Graeca 1.6 and Cicero, De oratore 3.137. For a parallel myth concerning the reassembly of the Homeric poems by Lycurgus, lawgiver of Sparta, see Plutarch, Life of Lycurgus 4.4. For more on the myth of the Peisistratean recension see Nagy 1996b:73–75.
[ back ] 51. Summaries of these poems, made in antiquity by a scholar named Proklos, are preserved on folios 1, 4, and 6 of the Venetus A manuscript of the Iliad, discussed below. The few, meager surviving fragments of the poems of the Epic Cycle have been edited by Bernabé (1987), Davies (1988), and West (2003). For a discussion of the remaining fragments see Davies 1989. On the relationship of the Epic Cycle to the oral epic tradition in which the Iliad and the Odyssey were created see Burgess 2001 and Marks (forthcoming).
[ back ] 52. It is important to understand that in Nagy’s model the presence of such “transcripts” does not put an end to oral tradition or cause the text of the Iliad and the Odyssey to become fixed. González 2013:15–70 and Ready 2015 are two comprehensive critiques of the so-called “dictation theory” (as set forward first by Lord [1953], then by Jensen [1980], Janko [1982 and 1998], West 1990, and Powell [1991 and 1997a], among others), which attributes text fixation of the Iliad and the Odyssey to a process of dictation by a master poet. See also Nagy 1996b:30–35, Dué 2006b, and the conclusion.
[ back ] 53. Some papyri also contain scholia. On the relationship between the scholia that survive on papyri and those of the medieval manuscripts see McNamee 1981, 1992, 1995, and 2007.
[ back ] 54. On Aristarchus’ editorial practice see also van Thiel 1992 and 1997, Nagy 1996a:107–152 (with extensive references to earlier scholarship ad loc.), Montanari 1998 and 2002, and Schironi 2018 and forthcoming (a).
[ back ] 55. There is debate about the extent to which Aristarchus and other Alexandrian scholars consulted other available texts and editions in the preparation of their commentaries, as well as the extent to which they relied upon their own conjectures about the text; for a succinct overview of the debate, see Schironi (forthcoming [a]). In subsequent chapters I argue that attested multiforms in the ancient scholia that are attributed to Aristarchus and others are often just as “Homeric,” by any number of criteria, as the readings that survive in the main text in our medieval manuscripts. I will approach them, therefore, as at least potentially performance derived, and will try to understand them as such.
[ back ] 56. For more on the Venetus A, see Allen 1899, Dué 2009a, and Dué and Ebbott 2014. For an overview of the work of medieval scribes and the medieval transmission of Greek literature in general see Reynolds and Wilson 2014. For more on the differences between papyrus scrolls and parchment codices and the reasons for the transfer, see Ebbott 2009.
[ back ] 57. For more on the manuscript’s history see Blackwell and Dué 2009.
[ back ] 58. I have explored this idea in depth in Dué 2006b.
[ back ] 59. What I have described here can be termed the “intentional” model, where “finding an authoritative text [is] based on ‘final intentions’” (Price 2008; see also Greetham 1992:347–372, Tanselle 1995:15–16, and Shillingsburg 1996:74–102). Robinson (2009) speaks of this type of editing as being largely in the past, but this is not yet the case in Homeric studies.