The Best of the Achaeans
Concepts of the Hero in Archaic Greek Poetry
Revised Edition
Gregory Nagy

Copyright © 1980, 1997 The Johns Hopkins University Press. All rights reserved. First edition 1979. Revised edition 1999. This document may be used, with this notice included, for noncommercial purposes within a subscribed institution. No copies of this work may be distributed electronically outside of the subscribed institution, in whole or in part, without written permission from the JHU Press.

Go to Next chapter; Table of Contents; Information


A Word on Assumptions, Methods, Results

§1. My approach to archaic Greek poetry is based on two major working assumptions. One, the mechanics and artistry of a given poem are traditional not only on the level of form--let us call it diction--but also on the level of content--let us call it theme. Two, the diction is a most accurate expression of the theme.

§2. The basis for my understanding of Greek poetic diction is the work of Milman Parry on Homeric phraseology, which can be summed up in his concise definition of the formula: "a group of words which is regularly employed under the same metrical conditions to express a given essential idea."[1] The mechanical nature of the formula is reflected by what Parry called the principle of economy.[2] Denys Page restates the principle: "Generally speaking, for a given idea within a given place in the line, there will be found in the vast treasury of phrases one formula and one only."[3] Page goes on to offer an illustration by examining all the Homeric expressions for the concept of "sea":[4]

For this one idea, "the sea," and for its expression in noun + epithet phrases only, he [the poet] relied upon his memory to provide him with a ready-made formula for almost every requirement; and the traditional vocabulary was now so highly developed, so refined and reduced, that for each requirement he found never, or hardly ever, more than one single formula. He has no freedom to select his adjectives: he must adopt whatever combination of words is supplied by tradition for a given part of the verse; and that traditional combination brings with it an adjective which may or may not be suitable to the context.
There is, however, something troublesome here about the insistence on the poet's lack of freedom to say accurately whatever he means. It seems as if the factor of metrics were in control of what can or cannot be said. In this particular case of adjectives describing the sea, for instance, we are being told that the poet had no choice but to accept the various epithets that tradition had thrust upon him to fill out the various metrical positions of the Greek hexameter.

§3. In short, Parry's work on the mechanics of Homeric diction has caused a serious problem of esthetics for generations of Hellenists reared on the classical approaches to the Iliad and Odyssey: how can compositions that have always seemed so deliberate and integral in their artistry result from a system of diction that is so mechanical--one might almost say automatic? For various Homeric experts the solution lies in objecting to various aspects of Parry's findings: the genius of Homer must somehow be rescued from the workings of a formulaic system.[1] For me, however, it is easier to accept Parry's work and to proceed from there by looking for a solution in the factor of tradition itself.

§4. Let us take another glance at the Homeric deployment of epithets. Granted, Parry's descriptive approach shows us that the choice of epithets is regulated by metrical factors. On the other hand, the historical approach of comparative reconstruction reveals much more.[1] From my own previous studies using this approach, I have learned that certain fixed noun + epithet combinations in Homeric diction go back to a time that predates the very existence of the Greek hexameter; [2] further, that the choice of epithet is ultimately determined by themes that can be reconstructed all the way to a period when Greek was not yet differentiated from its sister languages in the Indo-European family.[3] With the help of such findings, I have developed the theory that Greek meter itself is a long-range result of regularizations in the formal patterns of traditional poetic diction.[4] Granted, diction is indeed regulated by meter from the descriptive point of view, but this regulation is from the historical point of view only the result of a more basic principle, namely, that diction is ultimately regulated by theme.[5]

§5. My theory, then, has it that theme is the overarching principle in the creation of traditional poetry like the Iliad and the Odyssey; also, that the formulaic heritage of these compositions is an accurate expression of their thematic heritage. Such a theory helps account for the problems raised by Parry's theory of the formula. Did the poet really mean this or that? Did he really intend such-and-such an artistic effect? My general answer would be that the artistic intent is indeed present--but that this intent must be assigned not simply to one poet but also to countless generations of previous poets steeped in the same traditions. In other words, I think that the artistry of the Homeric poems is traditional both in diction and in theme. For me the key is not so much the genius of Homer but the genius of the overall poetic tradition that culminated in our Iliad and Odyssey.

§6. To my mind there is no question, then, about the poet's freedom to say accurately what he means. What he means, however, is strictly regulated by tradition. The poet has no intention of saying anything untraditional. In fact, the poet's inherited conceit is that he has it in his power to recover the exact words that tell what men did and said in the Heroic Age.[1]

§7. These theoretical underpinnings have fostered a general attitude of literal mindedness in my approach to the concept of the hero in archaic Greek poetry. In the pages that follow, I will as a policy assume that the application of an epithet--whether it be fixed or particularized--is thematically appropriate as well as traditional.[1] Moreover, my working assumption extends from the usage of epithets in particular to the usage of words in general: the entire formula, to repeat, is an accurate response to the requirements of traditional theme.[2] I stress this point now in order to prepare the reader for the oncoming plethora of transliterated Greek words that I will be continually citing in my discussion of central poetic themes. My reliance on key words in context cannot be dismissed as a reductive and oversimplified method of delving into the thematic complexities of archaic Greek poetry, if indeed the words themselves are functioning elements of an integral formulaic system inherited precisely for the purpose of actively expressing these complexities. The words should not be viewed merely as random vocabulary that passively reflects the themes sought by the poet. The semantic range of a key word in context can be expected to be as subtle and complex as the poetry in which it is encased.

§8. In the course of confronting the diverse problems entailed by my overall inquiry, I have found that the most striking confirmation of my literal readings has been the remarkable pattern of correspondences between the deployment of key words on the one hand and, on the other, the artistic unity of the Iliad and the Odyssey as compositions. I should emphasize that the positing of a unitary Iliad and a unitary Odyssey has been for me not an end in itself, one that is continually threatened by contextual inconsistencies in this Homeric passage or that. Rather, it has been a means for solving the problems presented by these inconsistencies. Whatever Homeric passages seem at first to be inconsistent in the short range may in the long range be the key to various central themes of the overall Iliad or Odyssey--central messages that are hidden away from those of us, such as we are, who have not been raised by Hellenic society as the appreciative audience of Epos.

§9. Unlike most Homerists who perceive an artistic unity in the Iliad and the Odyssey, however, I would still prefer to reconcile what I see with what Parry has discovered about formulaic composition, with all that his discoveries imply about the traditional predetermination of diction. In this respect I find myself in the congenial company of Michael Nagler,[1] although my work lays less emphasis on the poet's thought processes and more on the poet's tradition. From my point of view, the way to reconcile the factor of formulaic composition with the factor of artistic unity is to infer that both are a matter of tradition. The unity of a masterpiece like the Iliad may itself be the product of a lengthy evolution in the artistic streamlining of form and content.[2]

§10. If indeed tradition is a principal factor in the artistic integrity of an archaic Greek poem, it follows that we need not simply attempt to ascribe an Iliad or an Odyssey to the creativity of one genius, the poet Homer. I prefer to follow the same line of reasoning in the case of Hesiodic poetry. Whatever unity we may discover in the Theogony and in the Works and Days need not lead us to the certainty that we have just found the "author" called Hesiod. Nor can we with any certainty recover an "author" by the name of Homer (or by any other name) on the basis of the Homeric Hymns. Granted, the Theogony itself names Hesiod as its composer (verse 22);[1] or again, the Homeric Hymn to Apollo actually presents itself as a poem composed by a blind poet from Chios whose songs are heard throughout the city-states of the Hellenes--surely the figure of Homer himself (verses 166-176).[2] Nevertheless, we will have a chance to see that the references made by an archaic poem to its composer, or "author," are not so much a personal attempt by the poet to identify himself but rather a formal reflection of the poetry upon its own importance: the archaic poem presents itself retrospectively as something transmitted by the ultimate poet.[3] Even the poems of a historical figure like Pindar tend to present their composer as a mere function or instrument of the poetry itself. In short, an archaic poem establishes its authority primarily by asserting the traditions upon which it is built.

§11. My criteria, then, for determining the integrity of poetic composition do not directly involve questions of authorship. In the case of something like Hesiod fragment 204MW, for example, it does not matter for my purposes whether this piece of archaic poetry was or was not composed by a person who may be identified as Hesiod: all that matters to me is whether I have comparative evidence to show that the given poetry is traditional in theme and diction and that its traditions are cognate with the ones we find in the Theogony or Works and Days.[1]

§12. I should add that any single composition may well be built from multiple traditions. In fact, the Homeric poems are prodigiously versatile in integrating a plethora of various different traditions in epic narrative.[1] Moreover, they even adapt and then integrate a variety of traditions in poetic genres other than epic.[2] My point remains, however, simply that the unity achieved by the Iliad and the Odyssey in their integration of various different traditions is itself an overall tradition.

§13. In archaic Greek poetry, the principle of unity in composition may be the result of social as well as artistic factors. In the Homeric Hymn to Apollo, for example, the integrity of the poem results from the fusion of two traditions about Apollo, the Delian and the Pythian, but the artistic fusion of the two distinct traditions implies a corresponding social fusion of two distinct audiences. The worship of Delian Apollo is the founding principle uniting the city-states on the Aegean Islands and on the coast of Asia Minor--precisely those Hellenic areas that are not included in the vast affiliation of city-states united in the worship of Pythian Apollo at Delphi.[1] Since the Homeric Hymn to Apollo is appropriate to the city-states under the sway of the Delian as well as the Pythian Apollo, its range of audience is truly Panhellenic in scope.

§14. Mention of the Panhellenic orientation that we find in the Homeric Hymn to Apollo brings us now to a vital contribution to our understanding of Homeric composition--from the field of archaeology. A recent archaeological synthesis by Anthony Snodgrass has made it clear that the eighth century B.C., the very era in which the Iliad and the Odyssey approached their ultimate form, was a watershed in the evolution of Hellenic civilization; alongside the emergence of the polis 'city-state' as a general institution with a strong trend of localized traditions (cult, law, etc.), there emerged a commensurately strong trend of intercommunication among the elite of the city-states--the trend of Panhellenism.[1] Some specific manifestations of the latter trend are:

Such institutions as the Olympic Games and the Delphic Oracle, both stemming from the eighth century, are of course monumental feats of intersocial organization and also of intercultural synthesis.[2] Significantly, the same can be said of Homeric Epos itself. From the internal evidence of its contents, we see that this poetic tradition synthesizes the diverse local traditions of each major city-state into a unified Panhellenic model that suits most city-states but corresponds exactly to none; the best example is the Homeric concept of the Olympian gods, which incorporates, yet goes beyond, the localized religious traditions of each city-state.[3] We also know that the Iliad and the Odyssey had proliferated throughout the city-states at the time that they reached their present form; it may be, then, that the Panhellenic nature of Homeric Epos is due not only to its composition but also to its proliferation.[4]

§15. Moreover, composition and proliferation need not necessarily be related as an event followed by a process: the evolution of the fixed texts that we know as the Iliad and Odyssey may be envisaged as a cumulative process, entailing countless instances of composition/performance in a tradition that is becoming streamlined into an increasingly rigid form as a result of ever-increasing proliferation.[1] Again we come to the image of that blind singer from Chios, the poet in the Homeric Hymn to Apollo (166-176).[2] If indeed such a figure amounts to an idealized retrojection based on the poetic tradition's sense of its own glory,[3] then we may also see the actual factor of proliferation reflected in the poet's boast that his songs are heard throughout the city-states of mankind:

hêmeis d' humeteron kleos oisomen hosson ep' aian
anthrôpôn strephomestha poleis eu naietaôsas

And we [the poet] will carry on your glory [kleos][4] wherever on earth we go,
throughout the well-inhabited city-states [polis plural] of men.
H.Apollo 174-175

In this connection, we cannot afford to ignore the actual existence of poetic organizations like the Homêridai of Chios and the Kreôphuleîoi of Samos--both of which had a heritage of strong Panhellenic affiliations.[5] The very concept of Homêros may be reflected by the inherited function of the Homêridai.[6] In sum, I think of Homeric poetry as a masterpiece of organization not only in an artistic but also in a social dimension.[7]

§16. The Panhellenic character of the Iliad and the Odyssey is actually reflected--albeit indirectly--by the two Panhellenic institutions that we have considered, the Sanctuary of Apollo at Delphi and the Olympics. As we shall see later in detail, the death of Achilles is a theme officially celebrated in the paiân 'paean', a form of song performed in worship of Apollo at Delphi (Pindar Paean 6).[1] Also, Achilles was traditionally mourned by the women of Elis in a ceremony that inaugurated the holding of the Olympics every four years (Pausanias 6.23.3).[2] As we shall also see, these traditional practices concerning Achilles reflect a latent dimension of the prime figure of Panhellenic Epos: even in classical times and beyond, he was also a figure in cult.

§17. We will have ample opportunity to examine the religious dimension of the Hellenic hero in cult. For now I wish only to insist on the most fundamental aspect: that the hero must experience death. The hero's death is the theme that gives him his power--not only in cult but also in poetry. We as readers of Hellenic poetry can still sense it. When a hero enters combat in the Homeric Epos, we are fully aware of the intense seriousness of it all: he will confront death. Not even the lofty Olympians can match that, since they cannot die; when the pro-Achaean gods enter combat with their pro-Trojan counterparts in Iliad XXI, the results cannot be fatal--and they cannot be serious either. For the Achilles of Homeric Epos, on the other hand, I will argue that the reality of death has a religious dimension that corresponds to the traditional ideology of hero cults.

§18. In this connection, it would be apt for me to quote a particularly intuitive observation linking the factor of hero cults with the factor of artistic unity in Homeric composition:[1]

It was only natural that the zeal of our specialists, be they philologians, historians, or archaeologists, should have led them far too frequently to proceed as if the Homeric poems were a rudis indigestaque moles. But in so doing they have tried, quite unconsciously and with the best intentions, to break the spiritual law which decrees that no human speech or communication, in prose or in verse, shall have any real meaning for those who fail to pay attention to the whole, or for those who are bored and inattentive whenever an author says something which is foreign to their personal and private interests. The poems respond to such students by promptly falling into fragments; they decay into masses of unrelated symbols. It is therefore the duty of the historian and of the archaeologist to expand their definitions of history to include the history of Greek religion and of Greek poetry; it will then become clear that Homer's transformation of history is founded upon hero worship, and that the Homeric poems deliberately and on the whole successfully suppress the post-Mycenaean aspect of Greece, and magnify the glory of the heroes in a most unhistorical but most poetical manner.
So much for intuition; what about evidence? Here again we get a vital contribution to our understanding of Homeric poetry from the field of archaeology. The Greek religious institution of hero cults, in much the same form that we see even in the classical era, can be traced back all the way to the eighth century B.C.--the same archaic era in which the Iliad and the Odyssey were attaining their ultimate form.[2]

§19. Accordingly, I have set as my main goal the answer to this vital question about the Achilles of our Iliad: does this Panhellenic figure possess the religious dimension of a cult hero even within Epos? There are other questions that are related: how is myth stylized in epic, and how does poetry in general express the connections between myth and ritual? In the course of my lengthy inquiry into the problem of heroes in cult, heroes in epic, I will be presenting a wealth of evidence in the form of passages from archaic Greek poetry in general and from Homeric Epos in particular.[1] From a careful reading of these passages, we will, I hope, enhance our overall understanding of the many-sided heroes who appear in them. In some cases, we will even discover a heroic dimension in the figures who are said to be the actual makers of Greek poetry--including Homer.[2] But the focus is not on Homer but on Achilles and Odysseus. My prime concern is that each of us may arrive at his own understanding of whoever is "best of the Achaeans."


§2n1. Parry 1971 [=1930] 272.

§2n2. Parry, pp. 276, 279.

§2n3. Page 1959.224.

§2n4. Page, pp. 225-226.

§3n1. There seem to be two favorite modes of objection. One is to scoff at the primary typological parallel adduced by Parry and his successor Albert Lord, to wit, the living epic traditions preserved by the South Slavic peoples (on which see Lord 1960). The second is to worry about whether Homer was literate or illiterate. I will not stun the reader at this point with massive doses of bibliography documenting these objections.

§4n1. In the field of linguistics, this approach is designated simply as the "comparative method": Meillet 1925.

§4n2. Nagy 1974.229-261.

§4n3. Ibid.

§4n4. Nagy, pp. 140-149. Even from a descriptive point of view, I will consistently argue that Homeric epithets are indeed appropriate to the themes associated with the words that they describe.

§4n5. Nagy 1976b. Rewritten in 1990b Ch. 2.

§6n1. More at Ch.15§§7-8, where the factor of regional variation also is taken into account. It stands to reason that different poets on different occasions will draw their material from different local traditions and that the poetic versions of what exactly happened in the past will differ from tradition to tradition. The important thing to keep in mind, however, is that variant traditions function as multiforms (cf. Ch. 3§2). Regional variations are themselves an aspect of what we call traditional oral poetry (cf. Lord 1960 passim). What the poet tells is true or false, depending on where he tells it: the local traditions on which the poet's immediate audience has been reared constitute the ultimate criterion of "truth." Such an ideology is clearly documented in Radloff's study of Kirghiz poetry (1885) and is still visible in Homeric passages that allude to the poet's tailoring the contents of his song to the predilections of his audience; see Svenbro's illuminating discussion (1976.5-73). I should stress that such poetic tailoring need not be interpreted as untraditional: it could just as easily be a matter of adjusting to local traditions. In the case of Homeric Epos, however, the tendency is to avoid localized idiosyncrasies: see the suggestive remarks of Svenbro, pp. 42-43, who correlates this tendency with what he sees as an ongoing process of text fixation. Unlike Svenbro, however, I would emphasize the factor of the polis 'city-state' less than the factor of Panhellenism (see §§14-15 below); within the context of the polis, there seems to be ample opportunity for regional variations (§14n4).

§7n1. On the distinction between fixed and particularized epithets, see Parry 1971 [= 1930] 153-165. In a critique of Parry's formulation (Nagy 1976b.243-244), I made the strategic error of applying the term particularized also to fixed epithets that are restricted to describing one entity. see now Nagy 1990b.22-23.

§7n2. For examples of thematic accuracy in the deployment of epithets in particular and words in general, see Ch.2 and Ch.5 respectively. Consider also my comments on the epithet korunêtês 'club wielder' at Ch.20§11n1.

§9n1. Nagler 1974; see also Austin 1975 and Frame 1978.

§9n2. Cf. Pagliaro 1970.39-40 on the theories of Giambattista Vico; also Nagy 1974.11. I have developed the theory more fully at Ch.2§18 and Ch.5§§18-19. In fact, I was tempted to have those paragraphs here, but I finally decided to place them in specific contexts where the point could perhaps be made more strongly.

§10n1. Cf. Ch.17§9.

§10n2. Cf. §15 below.

§10n3. Ch.17§§8-9.

§11n1. More on Hesiod fr. 204MW at Ch.11§§13-15.

§12n1. Cf. Ch.3§§1-2, 19.

§12n2. Cf. Ch.6 on lamentation, Ch.11-Ch.15 on praise- and blame-poetry. To go one step further: Homeric Epos even adapts and integrates the formal conventions of actual prayers. See Muellner 1976.

§13n1. See Giovannini 1969.67, who also points out that the areas not included in the world of the Pythian Apollo correspond to the areas not included in the Homeric Catalogue of Ships in Iliad II.

§14n1. Snodgrass 1971.421, 435; cf. West 1973.182. See now Snodgrass 1987.

§14n2. Snodgrass, pp. 352, 376, 416-417, 421, 431; cf. West ibid.

§14n3. Cf. Rohde I 125-127.

§14n4. In this connection, it is vital to point out that the Homeric Iliad and Odyssey are radically different in scope and artistry from the epics of the so-called Cycle-- namely, the Cypria, Aithiopis, Little Iliad, Iliou Persis, Nostoi, and Telegonia (the fragments of which will be cited consistently from Allen 1912). I rely on the definitive article by Griffin 1977, who demonstrates convincingly the uniqueness of the Iliad and the Odyssey in relation to the Cyclic poems. Griffin implicitly ascribes this uniqueness to "Homer." Instead, I prefer to stress the factor of Panhellenism: the Iliad and the Odyssey seem to be the only epics that ultimately achieved a truly Panhellenic status. To put it another way: I suggest that the Cyclic epics are so different from the two Homeric epics not because they are either more recent or more primitive but rather because they are more local in orientation and diffusion. For example, consider the myth in Vita Herodotea 15 (Allen, pp. 202-203) that tells how Homer was commissioned to dictate not only the Little Iliadbut also a composition called the Phokais--when he traveled to Phokaia! On the relationship of the Cycle with the local ktisis (`colonization') poetry of various city-states, see Ch.7§§27-29 (esp. §28n3; cf. also Ch.8§12n2). On the relationship of the Cycle with the Iliad and the Odyssey, cf. Ch.3§§1-2.

§15n1. It is significant that the proliferation of the alphabet and of the Homeric poems seems to be contemporaneous. As for the context of performance, I cite the international format of the institution known as the panêguris 'gathering, festival', on which see Wade-Gery 1952.2-6; one example is the Delian festival as reflected in lines 146-150 of the Homeric Hymn to Apollo and as discussed by Thucydides 3.104. I agree with Wade-Gery's argument that there is also internal evidence for the existence of such institutions within the Iliad and the Odyssey (ibid.), although I cannot agree with other aspects of his presentation.

§15n2. Cf. §§13-14 above.

§15n3. Cf. Ch.17§§8-9; cf. also Ch.18§4.

§15n4. On kleos in the sense of "glory" as conferred by poetry, see Ch.1§§2-4. The poet is referring to the kleos that he will make for the Deliades (named in H.Apollo 157); note that their kleos is destined never to perish (verse 156).

§15n5. On the subject of the Homêridae/Kreôphuleîoi in particular and rhapsôidoi in general: Burkert 1972b. Further details at Ch.9§25. On the expression used by rhapsôidoi to designate their inherited function, "to recite Homer," see Ch.6§6n4. On the meaning of rhapsôidos 'rhapsode', see Ch.17§10n5.

§15n6. More at Ch.17§§9-13.

§15n7. Even the root *ar- in Homêros and Homêridai (on which see Ch.17§9 and n2) is thematically appropriate for designating both social and artistic cohesion: Ch.17§12, esp. n5. Here as elsewhere, questions of etymology will enter the discussion. I should note at the outset that I intend to avoid building my arguments on the meanings of names; still, they frequently serve as convenient points of departure for any overall examination of traditional themes associated with the names of mythical figures (cf. Ch.5§1, Ch.8§9, etc.).

§16n1. Ch.4 (esp. §§4-6); also Ch.5§9, Ch.7 (esp. §§4, 24-30).

§16n2. Ch.6§§26 and 30.

§18n1. Hack 1940.481.

§18n2. Snodgrass 1971.191-193, 398-399. Further discussion at Ch.6§28.

§19n1. The important testimony of Athenian drama has been as a rule left out of consideration in this phase of my research; I hope to undertake a separate treatment of this vast area in a future project.

§19n2. Ch.17§§10-13.

Go to Next chapter; Table of Contents; Information

Copyright © 1980, 1997 The Johns Hopkins University Press. All rights reserved. This document may be used, with this notice included, for noncommercial purposes within a subscribed institution. No copies of this work may be distributed electronically outside of the subscribed institution, in whole or in part, without written permission from the JHU Press.