Pindar’s Homer is not “our” Homer

2015.12.24 | By Gregory Nagy

The Monteleone Chariot (see below for image credit)


0§1. The author of an article on Pindar and the Epic Cycle makes the following claim (Rutherford 2015:451–452):

“No extant Pindaric Ode uses as its primary narrative the story of the Iliad [. . .] or that of the Odyssey.”

In the spot that I have marked here with a sign for ellipsis (“[. . .]”), however, this same author adds, within parentheses, a qualification:

“although the death of Hector is mentioned alongside Achilles’ other conquests”

In the context of this claim, the writer lists the references in Pindar’s Odes to the killing of Hector by Achilles, and I quote (Rutherford 2015:452n16):

At Ol[ympian] 2.81 alongside Cycnus and Memnon, at Isthm[ian] 5.39–41 alongside Cycnus, Memnon and Telephus, and at Isthm[ian] 8.55 alongside Memnon. Notice also that the funeral of Achilles, described in Isthm[ian] 8.56–8, was in the Odyssey (24.60) as well as the Aethiopis.

As we can see from what I have quoted so far, this writer thinks that “Homer,” in Pindar’s poetics, is the poet of the Iliad and the Odyssey, while the so-called “Epic Cycle” is non-Homeric. I disagree with this line of thinking, and I will argue here that Pindar’s Homer, unlike “our” Homer, is not only the poet of the Iliad and the Odyssey: Pindar’s Homer is also the poet of the Epic Cycle.

0§2. Let me return to the basic facts as described in the statement I quoted a moment ago (Rutherford 2015:452n16). The killing of Hector by Achilles, as mentioned three times in the victory odes or epinicia of Pindar, is attested not in the Epic Cycle but only in the Iliad as we have it, while the killings of the heroes Kyknos (Cycnus) and Memnon, as mentioned in the same Pindaric context, are attested almost exclusively in the Epic Cycle but not in the Iliad and the Odyssey (though Memnon is mentioned, in passing, at Odyssey 11.522).

0§3. The claim that I already quoted, “No extant Pindaric Ode uses as its primary narrative the story of the Iliad,” is followed up by an additional claim (Rutherford 2015:452):

“Here we may contrast Bacchylides, whose Ode 13 contains a sort of mini-Iliad.”

So, the thinking here is that such a “mini-Iliad” in Ode 13 is Homeric, as it were, and not Cyclic. In other words, Ode 13 of Bacchylides is supposedly based on “our” Homeric Iliad. My own work on Ode 13 (Nagy 2011), however, has led me to think the opposite: that this composition of Bacchylides refers to an Iliad that does not correspond to “our” Homeric Iliad. My thinking is shaped by the details in Ode 13 concerning the role of the hero Ajax in fighting off the fire of Hector, which threatens to burn down the beached ships of the Achaeans. In Ode 13, this role of Ajax is prominently highlighted. In “our” Iliad, by contrast, what we see being highlighted is instead the role of Patroklos in fighting off the fire of Hector, while the corresponding role of Ajax is ostentatiously shaded over. Meanwhile, in Ode 13 of Bacchylides, there is not even any mention of Patroklos. So, the “mini-Iliad” of Ode 13 is not really “Homeric,” and Bacchylides cannot really be contrasted here with Pindar.

0§4. I now return once again to the claim that I quoted at the beginning (Rutherford 2015:451–452):

“No extant Pindaric Ode uses as its primary narrative the story of the Iliad [. . .] or that of the Odyssey.”

This claim is contradicted especially by the evidence of Pindar’s Isthmian 8, which features as one of its primary narratives a story that tells about a decision made by the Olympian gods to arrange for the goddess Thetis to be married off to the mortal hero Peleus instead of being impregnated by the immortal god Zeus himself. As Laura Slatkin has shown, this story is embedded in the central plot of the Homeric Iliad as we have it.[1]

0§5. The claim that I just quoted once again a moment ago is followed up with a further claim:

“If Pindar avoided the Iliad and Odyssey, it may have been because his aim in the epinicia was to create for each patron a unique memorial, and that could only be done with a myth that had not achieved such a fixed form in the Panhellenic tradition.” (Rutherford 2015:452)

In making this particular claim, the author cites in a footnote (Rutherford 2015:452n17) an article by Rupert Mann (1994) stemming from a DPhil dissertation that extends over 994 pages (Mann 1993).[2] The claim, however, that “Pindar avoided the Iliad and Odyssey” (again, Rutherford 2015:452) does not quite mesh with the subtle thinking of Mann (1993, 1994), with whom I in fact generally agree. Mann (1994, especially p. 335) sees the poetics of the Iliad and the Odyssey as qualitatively different from the poetics of the Epic Cycle, and he thinks that this difference is reflected in the poetic infrastructure of Pindar, who tends to keep Iliadic and Odyssean themes in the background while foregrounding alternative themes that predominate in the Cycle.

0§6. I concentrate here, however, not on my general agreement with Mann but on a disagreement I have with an inference taken from the work of Mann. I now quote the wording of this inference, where Mann (1994) is cited as an antidote, as it were, to what I had argued in my book Pindar’s Homer (Nagy 1990a):

“By contrast Nagy [1990a] argued that there is no qualitative difference between the way Pindar uses ‘Homeric’ and ‘Cyclical’ material, on the grounds that the lyric tradition, represented by Pindar, had from earliest times contained within itself prototypes of all the mythical narratives of the Trojan Cycle; the epic realizations of these myths would on this model be later, as would be the process whereby some versions achieve Panhellenic status (i.e. Homeric ones) and others remained local (i.e. Cyclical ones). The improbable implication of Nagy’s somewhat mystical approach is that the lyric versions are always primary and epic ones secondary, perhaps even that epic ones adapt/allude to lyric ones.” (Rutherford 2015:452n17)

0§7. I find that this summary of the model that I had built in my book is full of misreadings, the most salient of which is the claim that my approach to the poetics of Pindar is “mystical.” In contradicting this claim, I suggest that my approach is based on a pragmatic reading of Pindar’s poetics, and that the idea conveyed by the title of my 1990 book, Pindar’s Homer, is an accurate description of Pindar’s metaphorical world as expressed by the rhetoric of his songmaking. In terms of this rhetoric, the lyric form of Pindar’s songmaking generates, from within itself, the epic form to which this songmaking regularly refers. Such a relationship of lyric and epic, as expressed in Pindar’s poetics, is actually validated by the etymology, as it were, of epic meter: as I argue in the Appendix of Pindar’s Homer, the dactylic hexameter of Homeric poetry—and of epic in general—originates from a fusion of the two basic lyrical patterns that we see at work in the metrical infrastructure of Pindaric songmaking (those lyrical patterns are the so-called Aeolic and dactylo-epitrite meters). But I concentrate here not on the relationship of epic and lyric—which is a topic that I hope to explore further in another posting—but on what I see as a misinterpretation, in the summary I just quoted, of the way I use the terms “Cyclical,” “Homeric,” and “Panhellenic.” In what follows, I offer working definitions of these terms as I actually understand them.

On the term “Cyclical” and its relation to the name of Homer


§1. For an understanding of the term “Cyclical,” I start with the earliest reconstructable meaning of the word kuklos (κύκλος) as applied to the Epic Cycle.[4] In terms of such an application, kuklos refers to all poetry composed by Homer.[5] Such a meaning of kuklos as the sum total of Homeric poetry goes back to a metaphorical use of the word in the sense of ‘chariot wheel’. In Homeric diction, kuklos actually means ‘chariot wheel’ (Iliad 23.340, plural κύκλα at 5.722). The metaphor of comparing a well-composed song to a well-crafted chariot wheel is explicitly articulated in the poetic traditions of Indo-European languages (as in Rig-Veda 1.130.6); more generally in the Greek poetic traditions, there is a metaphor comparing the craft of the tektōn (τέκτων) ‘joiner, master carpenter’ to the art of the poet (as in Pindar Pythian 3.112–114).[6]

§2. Connected with this idea of kuklos as a chariot wheel is the meaning of the name ‘Homer’, that is, of Homēros (Ὅμηρος). This name is a nomen loquens derived from the noun homēros, to be explained etymologically as a compound *hom-āros (*ὅμ-ᾱρος) meaning ‘the one who fits/joins together’, composed of the prefix homo- ‘together’ and the root of the verb ar-ar-iskein (ἀρ-αρ-ίσκειν) ‘fit, join’. In terms of this etymological explanation, Homēros (Ὅμηρος) is a metaphor: the poet Homer is ‘the one who fits [the song] together’—as if the song were ‘the Cycle’ in the sense of a kuklos or ‘chariot wheel’. Homer as the master poet ‘fits together’ pieces of song that are made ready to be parts of an integrated whole just as a master carpenter or joiner ‘fits together’ or ‘joins’ pieces of wood that are made ready to be parts of a chariot wheel.[7]

§3. But the metaphorical world of Homer extends beyond song, as we see from the meaning of homēros as ‘hostage’. A homēros in the sense of a ‘hostage’ is notionally ‘the one who fits [people] together’.[8] This meaning can be connected with a myth that aetiologizes the figure of the poet Homer himself as a hostage who is blinded by the Lydians because he displays his solidarity with the Greek-speaking people of Asia Minor.[9] So, the notion of integrity that is built into the art of Homer as ‘the one who fits [the song] together’ extends to the society that is integrated by this art, since Homer is also ‘the one who fits [people] together’.

§4. It is in the context of such a societal metaphor for the name Homēros that we can appreciate the relevant analysis of this name by Marcello Durante: he argues that Greek Homēros is cognate with the Indic noun samaryám, which refers to an ‘assembly’ of people engaged in various contests, including poetic competitions.[10] I compare here the semantics of the Greek noun agōn, derived from the root ag- of the verb agō as it is used in the compound formation sun-agein, which means ‘bring together, assemble, gather’.[11] Basically, an agōn is a ‘bringing together’ of people; and the occasion of such a ‘bringing together’ is a ‘competition’. This meaning, ‘competition’, is still evident in the English borrowing of a compound formation involving the word agōn, that is, antagonism. We can see a comparable idea embedded in the meaning of the Latin word that gives us the English borrowing competition: basically, the meaning of Latin com-petere is ‘to come together’, and to come together is to compete.[12]

On contrasting the terms “Cyclic” and “Homeric”


§5. The Epic Cycle was in earlier times considered to be part of the Homeric tradition.[14] In these earlier times, the Epic Cycle was not anti-Homeric or even non-Homeric: it was Homeric. In these earlier times, further, Homer was the poet of an epic Cycle that included the earlier forms of what we know as the Iliad and Odyssey. Only in later times were the Homeric Iliad and Odyssey differentiated from the epic Cycle, which thus became non-Homeric. In terms of my explanation, then, Homer in earlier times was considered to be the poet of an Epic Cycle that included what we now know as the Iliad and Odyssey, but these two epics gradually became differentiated from the Cycle in later times. In the course of such a differentiation, the Iliad and the Odyssey eventually became the only epics that were truly Homeric, while the Cycle became non-Homeric.

§6. Such a differentiation between the Homeric Iliad and Odyssey on the one hand and a non-Homeric Cycle on the other hand is most clearly visible in sources dating from the fourth century BCE. For example, when Aristotle in his Poetics (1459a) refers to the kuklos (κύκλος) in the sense of ‘Epic Cycle’, he is referring to a body of epic poetry that was explicitly not composed by Homer.[15]

§7. In the second half of the sixth century BCE, the epics of the Epic Cycle were still being performed in Athens: evidence for Athenian performances at that time can be found in patterns of Athenian accretions embedded in both the form and the content of such epics as the Aithiopis and the Iliou Persis and the Little Iliad.[16] For example, in the case of the Iliou Persis attributed to Arctinus of Miletus, there is mention of the rescue of the mother of Theseus by the Athenian hero’s two sons Akamas and Demophon after the capture of Troy (Proclus summary p. 108.10–11 ed. Allen 1912); there is another such mention of these figures in the Little Iliad attributed to Lesches of Lesbos (F 18 ed. Allen 1912 via Pausanias 10.25.8).[17]

§8. Still, we can expect Athenian accretions at a lower degree in the epic Cycle and at a higher degree in the Iliad and Odyssey, since the epics of the Cycle were phased out of the epic program that evolved at the festival of the Panathenaia in Athens by the time of the classical period, in the fifth century BCE, leaving the Iliad and Odyssey as the sole representatives of Homeric poetry at that festival. Even in the preclassical period, the epics of the Cycle were peripheral while the Iliad and Odyssey were central in the Homeric tradition, as we can see from the fact that the overall narrative of the Cycle is built around the Iliad and Odyssey.[18] This formulation holds not only for the preclassical era of epic as performed at the Panathenaia in Athens around the second half of the sixth century BCE but even more so for the earlier preclassical era of epic performance as it evolved at the festival of the Panionia at the Panionion of the Ionian Dodecapolis, in the late eighth and early seventh centuries. Already then, the two central epics performed at the festival of the Panionia were prototypical versions of the Iliad and the Odyssey. As Douglas Frame has shown, a lasting trace of this centrality is the fact that each of these two epics is divisible into six performance units, adding up to twelve performance units representing each one of the twelve cities of the Ionian Dodecapolis.[19] To be contrasted are the two Ionian epics attributed to Arctinus of Miletus, the Aithiopis and the Iliou Persis, which do not fit the broader social framework of the Ionian Dodecapolis but rather the narrower one of Miletus as a single city. And the themes that we find in such epics of the Cycle tend to be more localized and therefore more conservative than the more Panhellenized themes of the Iliad and Odyssey. In my book Pindar’s Homer (1990a), I offered this explanation:

The Panhellenization of the Homeric tradition entailed a differentiation from older layers of Panhellenic epic tradition, and [. . .] these older layers were gradually sloughed off in the process of Homeric streamlining. Such an explanation would account for not only the artistic superiority of the Iliad and Odyssey but also the thematic archaism of the Cycle. The older layers represented by the Cycle kept developing alongside the emerging core of the Homeric tradition and, being the more local versions, had the relative freedom to develop for a longer time, albeit at a slower pace, toward a point of textual fixation that still seems like a case of arrested development in contrast with the ultimate Homeric form.[20]

On the term “Panhellenic”


§9. The term Panhellenic is derived from the ancient Greek compound noun pan-Hellēnes ‘all Greeks’, which is attested in the Hesiodic Works and Days (528: πανελλήνεσσι) in the sense of referring to ‘all Greeks under the sun’ (526–528: ἠέλιος . . . πανελλήνεσσι φαείνει).[22] This archaic use of the compound noun pan-Hellēnes in the absolutizing sense of ‘all Greeks’ helps explain the later use of the non-compound noun Hellēnes ‘Hellenes’ to mean ‘Greeks’; earlier, that noun Hellēnes had been used to designate a sub-set of Greeks dwelling in Thessaly rather than any full complement of Greeks. As the linguistic evidence shows, the accentuation of the non-compound noun Hellēnes should be non-recessive (*῾Ελλῆνες), not recessive (῞Ελληνες), and the fact that Hellēnes acquired an innovative recessive accentuation proves that its innovative meaning of ‘Greeks’ was predicated on the accentuation of the compound noun pan-Hellēnes in the absolutized sense of ‘all Greeks’.[23] In other words, the linguistic evidence shows that the non-compound noun Hellēnes acquired the meaning of ‘Greeks’ from the built-in politics of the compound noun pan-Hellēnes, the basic meaning of which can be paraphrased this way: Hellenes (as a subset of Greeks) and all other Greeks (as a notionally complete set of Greeks).[24]

§10. In my analysis of both Homeric and Hesiodic poetry, I use the term Panhellenic in a relativized sense, despite its inherently absolutized meaning as ‘common to all Greeks’. To relativize Panhellenic is to recognize that the pan-Hellenization of Homer and Hesiod, just like other aspects of Panhellenism, cannot be described in absolute terms of universalization. Despite the totalizing ideology implicit in the term Panhellenic, the Panhellenization of Homer and Hesiod was not an absolute: it was merely a tendency toward a notional absolute.[25] And, just as the concept of Panhellenism was in fact relative, so also the concept of a Panhellenic Homer or a Panhellenic Hesiod was relative, since it depended on the various appropriations of these poetic figures by the various Greek communities that claimed them as their own.


§11. Pindar’s Homer, then, is a form of epic that is more “Cyclical” and thus relatively less “Panhellenic” than the form of “Homeric” poetry that we know. Pindar’s poetic infrastructure allows for a Homer who is the poet of the Epic Cycle, not only the poet of the Iliad and the Odyssey. Pindar could recognize the differences between Iliadic and Odyssean themes on the one hand and Cyclic themes on the other, but all these themes could still be seen as Homer’s creations. That is why I argue that Pindar’s Homer is not “our” Homer.



Allen, T. W., ed. 1912. Homeri Opera V (Hymns, Cycle, fragments). Oxford.

Burgess, J. S. 2001. The Tradition of the Trojan War in Homer and the Epic Cycle. Baltimore.

Chantraine, P. 2009. Dictionnaire étymologique de la langue grecque: histoire des mots (ed. J. Taillardat, O. Masson, and J.-L. Perpillou), with supplement “Chroniques d’étymologie grecque” (ed. A. Blanc, Ch. de Lamberterie, and Jean-Louis Perpillou) 1–10. Paris.

Debiasi, A. 2004. L’epica perduta: Eumelo, il Ciclo, l’occidente. Hesperìa 20. Rome.

Debiasi, A. 2012. “Homer ἀγωνιστής in Chalcis.” Homeric Contexts: Neoanalysis and the Interpretation of Oral Poetry (ed. F. Montanari, A. Rengakos, and C. Tsagalis) 471–500. Trends in Classics Supplementary Volume 12. Berlin and Boston.

DELG. See Chantraine 2009.

Durante, M. 1976. Sulla preistoria della tradizione poetic greca II. Risultanze della comparazione indoeuropea (Incunabula Graeca 64) Rome.

Frame, D. 2009. Hippota Nestor. Hellenic Studies 34. Cambridge, MA and Washington, DC.

Mann, R. 1993. Myth and Truth in Some Odes of Pindar. DPhil dissertation, University of Oxford.

Mann, R. 1994. “Pindar’s Homer and Pindar’s Myths.” Greek Roman and Byzantine Studies 35:313–337.

Nagy, G. 1979. The Best of the Achaeans: Concepts of the Hero in Archaic Greek Poetry. Baltimore. 2nd ed. with new introduction 1999. Baltimore.

Nagy, G. 1990a. Pindar’s Homer: The Lyric Possession of an Epic Past. Baltimore.

Nagy, G. 1990b. Greek Mythology and Poetics. Ithaca, NY. Revised paperback edition 1992.

Nagy, G. 1996a. Poetry as Performance: Homer and Beyond. Cambridge.

Nagy, G. 1996b. Homeric Questions. Austin.

Nagy, G. 1999. See Nagy 1979.

Nagy, G. 2009. “Hesiod and the Ancient Biographical Traditions.” The Brill Companion to Hesiod (ed. F. Montanari, A. Rengakos, and Ch. Tsagalis) 271–311. Leiden.

Nagy, G. 2009|2008. Homer the Classic. Printed | Online version. Hellenic Studies 36. Cambridge, MA and Washington, DC.

Nagy, G. 2010|2009. Homer the Preclassic. Printed | Online version. Berkeley and Los Angeles.

Nagy, G. 2011. “A Second Look at the Poetics of Reenactment in Ode 13 of Bacchylides.” Archaic and Classical Choral Song: Performance, Politics and Dissemination (ed. L. Athanassaki and E. L. Bowie) 173–206. Berlin.

Nagy, G. 2013. The Ancient Greek Hero in 24 Hours. Cambridge, MA.

Nagy, G. 2015. “Oral traditions, written texts, and questions of authorship.” Chapter 2 of The Greek Epic Cycle and its ancient reception: A companion (ed. M. Fantuzzi and Ch. Tsagalis) 59–77. Cambridge.

Pfeiffer, R. 1968. History of Classical Scholarship: From the Beginnings to the End of the Hellenistic Age. Oxford.

Rutherford, I. 2015. “Pindar’s Cycle.” Chapter 24 of The Greek Epic Cycle and its ancient reception: A companion (ed. M. Fantuzzi and Ch. Tsagalis) 450–460. Cambridge.

Schmitt, R. 1967. Dichtung und Dichtersprache in indogermanischer Zeit. Wiesbaden.

Slatkin, L. 1991. The Power of Thetis: Allusion and Interpretation in the Iliad. Berkeley and Los Angeles.

Slatkin, L. 2011. The Power of Thetis and Selected Essays. Hellenic Studies 16. Cambridge, MA and Washington, DC.

West, M. L. 1999. “The Invention of Homer.” Classical Quarterly 49:364–382.

West, M. L. 2013. The Epic Cycle: A Commentary on the Lost Troy Epics. Oxford.



[1] Slatkin 1991, 2011. See also Nagy 1979:346 = 20§28 on Pindar Isthmian 8.31–35. Rutherford 2015:456n33 cites Slatkin 1991:74–75, among others, in arguing that the story about the decision of the gods to marry off Thetis to Peleus may have originated in some lost epic. But this argument is missing the point that Slatkin is making in both her books, which is, that this story is embedded in the Homeric Iliad as we have it. Rutherford p. 456 also entertains an alternative idea, that Pindar may have “invented” this story about Thetis.

[2] Specifically, Rutherford 2015:452n17 cites Mann 1994:325, 334–335.

[3] What follows recapitulates what I say in Nagy 2015:60–61.

[4] See also Nagy 2010|2009:255–256|II§§330–333, based on a more extensive analysis in Nagy 1990a:70–81 = 2§§36–52.

[5] For more on this earlier sense of kuklos with reference to all poetry composed by Homer, see Pfeiffer 1968:73 and Nagy 1996b:38. See also West 2013: 22–23, with reference to Proclus, Life of Homer 9. I agree with West (2013: 1, 8) that Proclus is to be dated to the second century CE.

[6] Nagy 1979:296–300 = 17§§9–13, interpreting the evidence assembled by Schmitt 1967:296–298.

[7] Nagy 1996a:74–75.

[8] DELG s.v. ἀραρίσκω.

[9] Nagy 2010|2009:254–264|II§§327–351.

[10] Durante 1976:194–197, as cited already by Nagy 1979:296–297 = 17§9; also cited 20 years later by West 1999:372, 375–376, who offers his own explanation based on the argumentation of Durante. I comment in detail on West’s explanation in Nagy 2009|2010:60|I§142n1. See also Debiasi 2012:474n21, who points out that West’s explanation can be reconciled with the etymology that I propose.

[11] Nagy 2013 8b§4.

[12] Nagy 1990a:136–137 = 5§2.

[13] What follows recapitulates what I say in Nagy 2010|2009:69–70; 96; 320–321 | §§169–170; 231; E§§31–32.

[14] Nagy 1990a:70–81 = 2§§37–53.

[15] Nagy 2010|2009:320|E§§30–31.

[16] Details of such Athenian accretions in the transmission of the epic Cycle are surveyed by Debiasi 2004:206–207.

[17] Debiasi 2004:132n58, 207; for further examples of such Athenian accretions, see Burgess 2001:152, 247n75.

[18] Nagy 1990a:72 = 2§40.

[19] Frame 2009 ch. 11, who shows that each one of these twelve performance units corresponds to four rhapsōidiai ‘rhapsodies’ or ‘books’ of the Homeric Iliad and Odyssey as we know them (‘books’ 1–4, 5–8, 9–12, 13–16, 17–20, 21–24).

[20] Nagy 1990a:73 = 2§40.

[21] What follows is derived from Nagy 2009:274–275.

[22] Nagy 1990b:37.

[23] DELG s.v. ῞Ελληνες.

[24] Nagy 1996b:39n40.

[25] Nagy 1996b:38–40.

Image by Eric Chan via Wikimedia Commons under a Creative Commons Attribution 2.0 license