Homer and Comparative Mythology

By Dr. Gregory Nagy
Francis Jones Professor of Classical Greek Literature
Professor of Comparative Literature
Director, Center for Hellenic Studies
Harvard University

Still under the spell of Heinrich Schliemann’s rediscovery of Troy, students of ancient Greece have been accustomed to regard the Greek epic tradition of Homer as a reporting of events that really happened in the second millennium B.C., the Mycenaean Bronze Age. [1] This view must be modified by the perspective of comparative mythology, as most clearly articulated in a three-volume series, Mythe et épopée, by Georges Dumézil. [2] This perspective takes the methodology of Indo-European linguistics beyond the level of pure language and applies it on the level of myth as expressed by language. In this sense, it is appropriate to think of comparative mythology, more broadly, as comparative philology:

One of the services that “comparative philology” can render the “separate philologies” [as, for example, Classical philology] is to protect them against their own unchecked attitudes concerning “origins,” to orient them toward the kind of empirical process, positive or negative, that goes beyond the uncertainty and consequent arbitrariness that can result from evaluating facts purely from a Greek or Roman or Indic or Scandinavian point of view. [3]

A manuscript illustration (18th c.?) of the Battle of Kurukshetra, fought between the Kauravas and the Pandavas, recorded in the Mahabharata epic / Wikimedia Commons

Just as the Greek language, is cognate with other Indo-European languages, including Latin, Indic, and Old Norse, so also various Greek institutions are cognate with the corresponding institutions of other peoples speaking other Indo-European languages. In other words, such {7|8} diverse groups as the ancient Greek and Indic peoples have a common Indo-European heritage not only on the level of language but also on the level of society. To appreciate the breadth and the depth of this Indo-European heritage in Greek institutions, one has only to read through the prodigious collection of detailed evidence assembled by Emile Benveniste in Le vocabulaire des institutions indo-européennes. [4] For now, however, we shall concentrate on Dumézil’s argument that one such Indo-European institution is the tradition of epic as reflected, for example, in the Indic Mahābhārata. The comparative approach, as we shall see, gives a vision of epic that is significantly different from the picture emerging from a “separatist” approach that restricts the field of vision to Homeric standards.

What comparative philology teaches us is that epic is a reflection not so much of historical events as of myth. According to this scheme, epic allows myth to take precedence over reality as we know it. Even where epic utilizes the raw material of real events, the argument goes, it will reshape these events to accommodate the requirements of myth.

This insight from comparative mythology concerning myth is a far cry from our own contemporary usage of the word “myth,” which conveys the opposite of reality. Myth, in societies where it exists as a living tradition, must not be confused with fiction, which is a matter of individual and personal creativity. Rather, myth represents a collective expression of society, an expression that society itself deems to be true and valid. From the standpoint of the given society that it articulates, myth is the primary reality. [5] For the purposes of the present argument, then, “myth” can be defined as “a traditional narrative that is used as a designation of reality. Myth is applied narrative. Myth describes a meaningful and important reality that applies to the aggregate, going beyond the individual.” [6]

The links between myth and epic are explored from another angle in the researches of Milman Parry and Albert Lord on the nature of traditional oral epic poetry. [7] From their fieldwork in the living South Slavic oral poetic traditions, we can witness directly how the composition and performance of epic are aspects of the same process, how myth is literally re-created in each new performance, and how there can be {8|9} countless variations in tradition without any deviation from tradition itself. Parry and Lord have successfully applied to the Homeric Iliad and Odyssey the criteria that they developed from their experience in field-work, and their demonstration that these marvels of Western literature are composed in the traditions of oral poetry may further justify applying the criteria that Dumézil has developed from his experience in the systematic comparison of surviving epic texts. [8] As we can ascertain independently with Parry and Lord, Homer is a master of mythe et épopée.

There are serious problems, however, in connecting the epic traditions of ancient Greece with those of other societies belonging to the Indo-European language family. Dumézil himself gives the clearest account of these problems, which can be summarized as follows: [9]

1. generally, it is difficult to connect the narrative of the Homeric poems with the basic patterns of Indo-European society as reconstructed from comparable narratives in languages related to Greek;
2. specifically, the themes associated with the major Homeric heroes seem not to match the themes associated with gods—at least, they do not match as they do, for example, in the clearly Indo-European epic traditions of the Indic Mahābhārata.[10]

The Delphic Tholos, seen from above / Photo by Tamara Semina, Wikimedia Commons

A solution to these problems may be found in archaeology—not in the evidence of the second millennium B.C., the represented era of the Homeric heroes, but in the evidence of the eighth century B.C., the era of the incipient Homeric audience. Whereas the archaeology of the second millennium has encouraged students of Hellas to concentrate on the historical realities found in Greek epic, the archaeology of the eighth century may lead them to perceive the mythmaking framework that integrates these realities.

A 1971 book by Anthony Snodgrass, The Dark Age of Greece, has made it plain that the eighth century B.C., the era in which the Iliad and Odyssey were reaching their ultimate form, is as important for our understanding of Homeric poetry as is the late second millennium B.C., the era that provides the overt subject matter for both of these epics. [11] Granted, Homeric poetry draws on details that archaeologists can indeed assign to the late second millennium. [12] But the point is that it also reflects the {9|10} overall orientation of the eighth century, which is a watershed for the evolution of Hellenic civilization as we know it: in this era, alongside the phenomenon of the pólis ‘city-state’, the heir to localized traditions in cult, law, and so on, there emerged a complementary phenomenon of pan-Hellenism, formalized in such institutions as the Olympic Games, the Delphic Oracle, and the Homeric poems themselves. [13]

In fact, Homeric poetry is a formalization of both these phenomena; it synthesizes the diverse local traditions of each major city-state into a unified pan-Hellenic model that suits the ideology of the polis in general, but without being restricted to the ideology of any one polis in particular. [14] Perhaps the clearest example is the Homeric concept of the Olympian gods, which incorporates yet goes beyond the localized religious traditions of each polis; the pan-Hellenic perspective of Homeric poetry has transcended local phenomena such as the cult of gods, which is functional only on the local level of the polis. [15] By “cult” I mean a set of practices combining elements of ritual as well as myth. For a working definition of “ritual,” I choose the following formulation: “Ritual, in its outward aspect, is a programme of demonstrative acts to be performed in set sequence and often at a set place and time—sacred insofar as every omission or deviation arouses deep anxiety and calls forth sanctions. As communication and social imprinting, ritual establishes and secures the solidarity of the closed group.” [16] The insistence of ritual on a set order of things should not be misunderstood to mean that all rituals are static and that all aspects of rituals are rigid. Even in cases where a given society deems a given ritual to be static and never changing, it may in fact be dynamic and ever changing, responding to the ever-changing structure of the society that it articulates.

Besides the cult of gods, another example of interplay between polis and pan-Hellenism in Homeric poetry is its attitude toward the cult of heroes. Erwin Rohde’s monumental book Psyche remains one of the most eloquent sources for our understanding the hḗrōs ‘hero’ as a very old and distinct concept of traditional Greek religion, requiring ritual practices that were distinct from those associated with the gods. [17] What archaeology now tells us is that this Hellenic institution of hero cults, without much difference from what we see attested in the Classical period of the fifth century, is shaped in the eighth century B.C., the same {10|11} era that shaped the Iliad and Odyssey. [18] It is of course tempting to explain the upsurge of hero cults throughout the city-states as a phenomenon motivated by the contemporaneous diffusion of Homeric poetry, [19] but it would be better to follow Snodgrass in looking for a more comprehensive explanation. [20] Again, the key is the twin eighth-century phenomena of the polis on one hand and pan-Hellenism on the other. I cite Rohde’s thesis that the cult of heroes was a highly evolved transformation of the worship of ancestors—a transformation that took place within the social context of the polis. [21] This thesis, perhaps most appealing from the viewpoint of cultural anthropology, [22] allows room for considering the constituent elements of hero cults to go back far beyond the eighth century. [23] In other words, we can posit a lengthy prehistory for not only the epics of heroes but also the cults of heroes, with this qualification: the ultimate forms of the epics and of the cults were definitively shaped in the eighth century. The strong eighth-century upsurge in the local cults of heroes can thus be viewed as a phenomenon parallel to—rather than derivative from—the pan-Hellenic epics of heroes, namely, the Iliad and Odyssey.

Thus the ideological heritage of Greek heroes may still in principle be reconstructed as Indo-European in character. But there are problems in extracting comparative evidence about the hero from Greek epic, especially about the religious dimension of the hero. It is worth stressing that the hero as a figure of cult must be local because it is a fundamental principle in Greek religion that his supernatural power is local. [24] On the other hand, the hero as a figure of epic is pan-Hellenic and consequently cannot have an overtly religious dimension in the narrative. Still, we may expect to find at least latent traces of this religious dimension within the Homeric poems. I have in fact produced a book with this expectation in mind. [25] Even further, it can be argued that these latent traces have cognates in the comparative evidence of other Indo-European epic traditions.

The Rage of Achilles by Giovanni Battista Tiepolo / Wikimedia Commons

Both the Iliad and the Odyssey reveal a pervasive theme that implicitly tells of a hero’s parallelism, not only in character but also in action, with a corresponding god. This theme is particularly manifest in the case of {11|12} Achilles and Apollo in the Iliad. For example, both the hero and the god have mênis ‘anger’ (I 1/I 75) that inflicts álgea ‘pains’ (I 2/I 96) so as to be a loigós‘devastation’ for the Achaeans (I 341/I 97). [26] Moreover, the hero and the god are traditionally represented as look-alikes, for example both appearing unshorn in the manner of a koûros ‘uninitiated male’. Their physical parallelism has led Walter Burken to describe Achilles as a Doppelgänger of Apollo. [27] We see here a remarkable analogue to the five epic heroes known as the Pāṇḍava-s in the Indic Mahābhārata, heroes whose parallelism with corresponding gods (Dharma, Vāyu, Indra, and the two Aśvin-s) has been traced in detail by Dumézil. [28]

this pattern of antagonism on the level of myth is matched by a pattern of symbiosis on the level of cult. [29] As a case in point, I cite the relationship of Apollo with Pyrrhos/Neoptolemos, son of Achilles: this hero is killed by the god himself within his own divine precinct (e.g. Pindar Paean 6.117-120), the very place where Pyrrhos/Neoptolemos is believed to be buried and where he is worshiped as the chief cult hero of Delphi (e.g. Pindar Nemean 7.44-47). [30] This and many other similar themes of god-hero antagonism suit an Indo-European pattern documented by Dumézil, who links the story patterns of the Old Norse hero Starkaðr, the Indic hero Śiśupāla, and the Greek hero Herakles. With the myth of Herakles, Dumézil himself has admirably demonstrated that the Indo-European pattern of god-hero antagonism is indeed attested in the realm of Greek myth. What I would add here is that this same pattern is central to the narrative traditions of Greek epic in particular, as exemplified in the Iliad and Odyssey. [31] In fact, the compressed retelling of the Herakles story in Iliad XIX 95-133 is a clear attestation of the same Indo-European pattern that Dumézil has reconstructed from such nonpoetic retellings as in Diodorus Siculus (4.8-39).[32]

The claim can be made, then, that the themes associated with the major Homeric heroes do indeed match the themes associated with {12|13} gods, and that Dumézil’s doubts about the applicability of his reconstructions to Greek epics can in the end be dispelled. Moreover, since the Greek evidence shows parallelisms of god and hero attested even on the level of cult, Dumézil’s vision of epic as structured by myth may even be extended one level further: from epic to myth to ritual.

There is a striking attestation of all three levels in the Homeric Hymn to Herakles (Hymn 15), which can be described as a brief and stylized prayer in worship of the hero Herakles, invoked as the son of Zeus (verses 1 and 9) and implored to grant success and wealth (9). Within this prayer Herakles is described as áristos ‘best’ among the epikhthónioi ‘earth-bound men’ of Thebes (1-2), and each of these words conjures up a specific heroic theme. Considering epikhthónioi first, we note that this word can mean more than ‘earth-bound men’ pure and simple: it is also attested in the context of designating heroes as worshiped in cult (e.g. Hesiod Works and Days 123). [33] As for áristos ‘best’, it serves as a formal measure of a given hero’s supremacy in his own epic tradition, as we see from the deployment of the expression áristos Akhaiôn ‘best of the Achaeans’ in the Homeric poems: the Iliad and the Odyssey each appropriate this epithet to fit the central figures Achilles and Odysseus, respectively.[34]

There are other epic touches as well in Homeric Hymn 15, as for example in these verses describing the Labors of the hero Herakles:

πλαζόμενος πομπῇσιν ὑπ’ Εὐρυσθῆος ἄνακτος
πολλὰ μὲν αὐτὸς ἔρεξεν ἀτάσθαλα, πολλὰ δ’ ἀνέτλη

Homeric Hymn to Herakles 15.5-6

Set off-course on missions at the direction of Eurystheus the king,
many are the reckless [atásthala] things that he did, many the things that he endured.

Let us compare the verses beginning the Odyssey:

ἄνδρα μοι ἔννεπε Μοῦσα πολύτροπον ὃς μάλα πολλὰ
πλάγχθη, ἐπεὶ Τροίης ἱερὸν πτολίεθρον ἔπερσε.
πολλῶν δ’ ἀνθρώπων ἴδεν ἄστεα καὶ νόον ἔγνω,
πολλὰ δ’ ὅ γ’ ἐν πόντῳ πάθεν ἄλγεα…

Odyssey i 1-4

About the man sing to me, Muse, the one of many turns [polú-tropos] the {13|14} one who many times
was set off course after he destroyed the holy citadel of Troy.
Many are the men whose cities he saw, and he came to know their way of thinking,
and many are the pains that he suffered at sea.

In the Hymn to Herakles, the anaphora of πολλᾲ…πολλᾲ ‘many things…many things’ at verse 6 in the context of πλαζόμενος ‘set off-course’ at verse 5 is parallel to the anaphora of πολύτροπον…πολλὰ/πολλῶν…/πολλὰ ‘the one of many turns’…’many’/’many’ …/’many’ at Odyssey 1/3/4 in the context of πλάγχθη ‘was set off course’ at I 2. Further, the expression πολλὰ…ἀνέτλη ‘he endured many things’ at Hymn 15.6 describing Herakles is parallel to πολλὰ…πάθεν ἄλγεα ‘he suffered many pains’ at Odyssey i 4 describing Odysseus, while the ἀτάσθαλα ‘reckless’ deeds of Herakles at Hymn 15.6 correspond to the characterization of Achilles in his own dark moments of savagery as ἀτάσθαλος (e.g. XXII 418). [35]

The study of the diction in Hymn 15 could be taken much further, but the point has already been made: this composition is not only a functional prayer to a hero in his capacity as a cult figure but also a glorification of his epic attributes—the very same epic attributes that seem to be divided up between Achilles and Odysseus in the overall epic diction of the Iliad and Odyssey. [36]

Achilles and Odysseus on a Greek amphora / Wikimedia Commons

It is as if the epic figures of Achilles and Odysseus resulted from a split of characterizations that we can find in one Homeric figure, Herakles. In light of Dumézil’s work on Herakles in Mythe et épopée II, [37] we see that Herakles is in fact far closer to an Indo-European model of the hero than is either Achilles or Odysseus. We may even add another factor, this time in light of Dumézil’s work in Mythe et épopée I. [38] Such heroes as the five Pāṇḍava-s in the Indic Mahābhārata are not only each parallel to specific gods in traits and actions: they are also sired by these very same gods. [39] So also with Herakles: the best of heroes, as he is described in the Homeric Hymn To Herakles (15.1-2), is sired by the best of gods. By {14|15} contrast, most Homeric heroes are several generations removed from divine parentage, just as they are several stages removed from the Indo-European model of the hero.

There is a striking reflex of this state of affairs on the formal level of poetic diction. The word in question is hēmí-theoi (ἡμίθεοι) ‘demigods’, which as we see from Hesiod F 204.100 MW and elsewhere clearly denotes direct divine parentage on one side(ἡμίθεοι at verse 100 = τέκνα θεῶν at verse 101), not simply semidivine status. [40] In Hesiod Works and Days 160, this same word hēmí-theoi designates the heroes of the generations that fought at Thebes and Troy. [41] Yet Homeric diction consistently refers to this same generation of heroes as hḗrōes (ἥρωες) ‘heroes’, not hēmí-theoi (ἡμίθεοι) ‘demigods’. The only exception is at Iliad XII 23, where hēmí-theoi does indeed refer to the Achaean heroes who fought at Troy—but the reference here is made from the standpoint of the Homeric audience, as it looks back, centuries later, at the remains of the Trojan War (see especially Iliad XII 26-32). [42] The point remains, then, that Homeric poetry—unlike other traditional forms of poetry—cannot as a rule designate its own heroes as hēmítheoi ‘demigods’, being restricted instead to the word hḗrōes. It is as if the Indo-European model of hero were no longer appropriate for the Homeric tradition of epic narrative, whereas it remained so for other poetic traditions such as the Hesiodic. [43]

I return to Hesiod F 204 MW for a remarkable illustration of this principle. At verses 95 and following, we find a tightly compressed narrative about the beginnings of the Trojan War: how the Olympian gods were split into pro-Achaean and pro-Trojan factions ever since the éris ‘strife’ at the Judgment of Paris (verses 95-96), [44] how the Will of Zeus ordained the deaths of heroes in the Trojan War (96-123). By good fortune, a corresponding passage is attested in a fragment from the epic Cycle, specifically from the beginning of the epic Cypria (F 1 Allen). Here, too, we find a reference to éris ‘strife’, this time designating the Trojan War itself (Cypria F 1.5), and how it was the Will of Zeus that heroes should die at Troy (F 1.6-7). [45] So much for the convergences. One major divergence, however, is that the heroes of the Trojan War are called hḗrōes in the Cyclic version (F 1.7 Allen) but hēmítheoi ‘demigods’ in the Hesiodic (F 204.100 MW). It appears that the epic format is more specialized, {15|16} more restricted, than other forms of poetry, and that it cannot easily tolerate the semantics of an Indo-European model that contradicts the genealogies of its own more specialized, more restricted heroes.

And yet the general themes shared by these Cyclic and Hesiodic passages are distinctly Indo-European in character. Even if the Cyclic Cypria avoids calling its heroes hēmítheoi ‘demigods’, these epic figures nonetheless share a vitally important theme with their distant Indic cousins, the divinely sired Pāṇḍava-s of the Mahābhārata: Zeus brings the éris ‘strife’ of the Trojan War because he intends to depopulate the Earth of the myriad heroes that weigh upon her (Cypria F 1.1-6 Allen). Similarly in the Mahābhārata, the war of the Pāṇḍava-s is a divine solution to the overpopulation of heroes weighing upon the earth. [46] In this way the major epic narratives of the Greek and the Indic peoples are inaugurated with a cognate theme, and it is hard to imagine more compelling evidence for the Indo-European heritage of the epic traditions about the Trojan War. [47]

A related theme, equally important as evidence for an Indo-European heritage in Greek epic, is apparent in the Iliad’s own compressed reference to the beginnings of the Trojan War—a theme elaborated in detail by the Cypria. We have already seen that the war itself is an éris ‘conflict’ (Cypria F 1.5 Allen), just as the Judgment of Paris is an éris (Hesiod F 204.96 MW). The éris that marks the beautiful shepherd’s judgment took the form of neîkos (νεῖκος) ‘quarreling’ (Cypria/Proclus summary p. 102.15 Allen), and the Iliad itself refers to the whole affair as éris (III 100) or neîkos (XXII 116). I emphasize the words éris‘strife’ and neîkos ‘quarreling’ not only because of their thematic importance as a plot motivation of epic but also because of their programmatic significance beyond epic. It so happens that the language of Greek praise poetry designates its own converse, blame poetry, with these very words, éris and neîkos. [48] Moreover, epic itself uses the word neikéō as ‘blame’ in opposition to ainéō ‘praise’ (e.g. Iliad 249-250). [49]

This opposition fits a larger pattern of a complementary interplay between praise and blame, particularly on the level of poetry, in such Indo-European societies as the Italic, Celtic, and Indic. [50] Marcel Detienne has extended the comparison to the Greek evidence, [51] and I in turn have described in detail the diction of praise and blame specifically {16|17} in Greek poetry. [52] The point to be made now, however, is more fundamental: as we can see from the Iliadic references to the Judgment of Paris, Greek epic presents its own genesis in terms of the opposition between praise and blame. As the Iliad puts it, the Judgment of Paris entailed the blaming of the goddesses Hera and Athena along with the praising of Aphrodite:

ὃς νείκεσσε θεάς, ὅτε οἱ μέσσαυλον ἵκοντο,
τὴν δ’ ἤνησ’ ἥ οἱ πόρε μαχλοσύνην ἀλεγεινήν

Iliad XXIV 29-30

[Paris] who blamed [verb neikéō ] the goddesses [Hera and Athena], when they came to his courtyard, but he praised [verb ainéō ] her who gave him baneful sensuality.

The Judgement of Paris, Peter Paul Rubens, c. 1636 (National Gallery, London) via Wikimedia Commons

Thus the primary narrative of Greek epic, which is the Trojan War, is self-motivated by the Indo-European social principle of counterbalancing praise and blame.

It is precisely in this theme, the Judgment of Paris, that Dumézil has found the most overt Greek example of a general Indo-European social ideology that he describes as trifunctionalism. In Dumézil’s formulation, society as reflected by the Indo-European languages tends to divide along the following three lines of ideological organization: (1) sovereignty and the sacred, (2) warfare, (3) agriculture, herding, and fertility in general. [53] The theme of the Judgment of Paris reflects this pattern of trifunctionalism. Paris the herdsman is being offered a gift from each of the three functions: (1) sovereignty, from Hera; (2) military supremacy, from Athena; and (3) sexual relations with Helen, the most beautiful woman on earth, from Aphrodite. [54]

Thus the Judgment of Paris, the ultimate point of departure for the narrative traditions that we know as Homer’s epics, can itself be judged as an epic theme with an Indo-European foundation. More broadly, Homer can be judged an authoritative source for the myths inherited by and through the Greek language. {17|18}

Footnotes

Originally published by The Center for Hellenic Studies under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported license. From “Greek Mythology and Poetics”, by Gregory Nagy (Ithaca, NY: Cornell University Press, 1992).