Pindar's Homer
The Lyric Possession of an Epic Past
a machine readable edition
Gregory Nagy

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.

Go to Previous chapter; Next chapter; Table of Contents; Information

Chapter 2

The Poetics of Panhellenism and the Enigma of Authorship in Early Greece

§1. I have been attempting both to distinguish poetry from song and to broaden the current concept of oral poetry in such a way as to include song. Since the conventional semantic range of our word poetry tends to exclude song, however, I prefer not to use the actual term oral poetry in this broader sense. Instead, I confine myself for now to the more narrow concept of oral poetry as distinct from oral song. But even this more narrow concept is not accurate enough to account for Homeric and Hesiodic poetry, or old elegiac and old iambic poetry. There is another characteristic of such poetic traditions that makes it a special kind of oral poetry. That characteristic is the phenomenon of Panhellenism.

§2. Panhellenism can be most readily defined in terms of the distinctively Greek institution of the polis or 'city-state', the importance of which for definining the concepts of Hellenism and even civilization can be most simply illustrated with Aristotle's dictum, ho anthrôpos phusei politikon zôion 'man is by nature an organism of the polis' (e.g., Politics 1253a2-3). With the polis 'city-state' as frame of reference, the phenomenon of Panhellenism can be summed up as follows. 1

§3. On the basis of archaeological and historical evidence, A. M. Snodgrass applies the concept of Panhellenism to the pattern of intensified intercommunication among the city-states of Hellas, starting in the eighth century B.C., as evidenced in particular by the following institutions: Olympic Games, Delphic Oracle, and Homeric poetry. 1 I have extended the concept as a hermeneutic model to help explain the nature of Homeric poetry, in that one can envisage as aspects of a single process the ongoing recomposition and diffusion of the Iliad and Odyssey. 2 I have further extended the concept to apply to Hesiodic poetry 3 and to Theognidean poetry. 4 Finally, as we see in Chapter 3, the concept can be extended still further to apply to lyric poetry in general, that is, to song.

§4. Essentially, the hermeneutic model of Panhellenism must be viewed as an evolutionary trend extending into the Classical period, not some fait accompli that can be accounted for solely in terms of, say, the eighth century. In other words the concept of Panhellenism as I use it here is a relative one. Thus various types of Archaic Greek poetry, such as the elegiac tradition preserved by Theognis, make their bid for Panhellenic status considerably later than Homeric and Hesiodic poetry. Still I see in such poetry as the Theognidea a parallel pattern of ongoing recomposition concomitant with Panhellenic diffusion. The most obvious reflex of this ongoing recomposition in diffusion is the ultimate crystalization of a body of poetry like the Theognidea, composed not in the native Doric dialect of a polis like Megara but in an accretive Ionic dialect that is for all practical purposes the same as we see in the poetry of Archilochus, Callinus, Mimnermus, Tyrtaeus, Solon, Xenophanes, and so on. 1

§5. From here on I refer to this process, described here as crystalization, simply as textual fixation. I apply this notion of textual fixation to oral traditions with an emphasis on gradual patterns of fixity in an ongoing process of recomposition in diffusion, and without presupposing that the actual composition of the "text" required the medium of writing. 1 In applying this notion I would stress the interconnected development of traditions alongside each other. 2

§6. By Panhellenic poetry, then, I mean those kinds of poetry and song that operated not simply on the basis of local traditions suited for local audiences. Rather, Panhellenic poetry would have been the product of an evolutionary synthesis of traditions, so that the tradition that it represents concentrates on traditions that tend to be common to most locales and peculiar to none. 1

§7. Such a synthetic tradition would require a narrower definition than suitable for the kind of oral poetry and song described by Albert Lord on the basis of his field work in the South Slavic oral traditions. The difference is that such a tradition is in the process of losing the immediacy of the performer-audience interaction expected in the context of ongoing recomposition in performance. The teleology of this loss is attested: in the historical period Homeric and Hesiodic as well as old elegiac and iambic poetry is being performed verbatim by rhapsôidoi 'rhapsodes' at Panhellenic festivals. 1 In the same context of Panhellenic festivals, what we have been calling song or lyric poetry is being performed verbatim by kitharôidoi 'lyre singers' and aulôidoi 'reed singers'. 2 Each new performance is now aiming at a verbatim repetition--not at an act of recomposition. 3

§8. Earlier I argued that the rhapsodes were direct heirs to earlier traditions in oral poetry. 1 But we see that over a long period their role has become differentiated from that of the oral poet. Whereas the oral poet recomposes as he performs, the rhapsode simply performs. In contemplating the recitation of Homer by rhapsodes, I am reminded of the following description of the recitation of "historical" poetry in Rwanda society: "Unlike the amateur, who gesticulates with his body and his voice, the professional reciter adopts an attitude of remoteness, a delivery that is rapid and monotone. If the audience should react by laughing or by expressing its admiration for a passage that is particularly brilliant, he stops reciting and, with the greatest detachment, waits till silence has been reestablished." 2 Such a description need not rule out the potential for variation from performance to performance. Still, variation is counteracted by the ideology of fixity. To that extent we see at least the impetus toward the notion of textual fixation without writing. 3

§9. The differentiation of composer and performer is attested in many cultures, as illustrated most succinctly by the incipient semantic split of trobador as 'composer' and joglar as 'performer' in Old Provençal usage. 1 In this case there is still evidence for some overlap. 2 With Homeric poetry, in contrast, the notion of composer is drastically retrojected, from the standpoint of the performers themselves, to a protopoet whose poetry is reproduced by an unbroken succession of performers; Socrates can thus envisage the rhapsode Ion as the last in a chain of magnetized metal rings connected by the force of the original poet Homer (Plato Ion 533d-536d). More accurately we may say that Ion is the next to last in the chain with relation to his audience, who would be the last link from the standpoint of the performance (Ion 536a). The implication of Plato's construct is that the magnetic force of the poetic composition weakens with each successive performer. Ion then, by virtue of being the last or at least the latest reproducer of Homer, would also be the weakest.

§10. In contrast, during phases of a given tradition where both composition and performance can be "owned" by the same person at a given occasion, 1 the advantage of the immediate composer-performer, as conferred by the occasion at hand, can be conventionally contrasted with the relative disadvantage of his predecessors, who are at this point deprived of their own occasion. Such an attitude is expressed this way in one particular Eskimo song: "All songs have been exhausted. He picks up some of all and adds his own and makes a new song." 2 We may note the juxtaposition made by Ferdowsi, poet of the Shâhnâma, with his predecessor, Daqiqi: 'In transmitting, his [= Daqiqi's] words became weak. \| Ancient times were not renewed by him' (VI 136.18). 3 A variant reads: 'Now I will tell what he [Daqiqi] has told. \| For I am alive and he has become united with the dust' (same citation). 4 The poet is presenting himself as owning the composition on the grounds that he, not his predecessors, now has access to the occasion, which is stylized as a series of performances before assemblies: 'These stories, grown old, will be renewed by me in all assemblies' (III 6.9). 5

§11. In the case of Plato's Ion, the myth-making retrojection of Homeric composition back to the strongest protopoet belies the evolutionary progression of a tradition where the aspect of recomposition gradually diminishes in the process of diffusion entailed by performance in an ever-widening circle of listeners. 1 The wider the diffusion, the deeper the tradition must reach within itself: the least common denominator is also the oldest, in that a synthesis of distinct but related traditions would tend to recover the oldest aspects of these traditions.

§12. A key to such Panhellenic synthesis is the ever-increasing social mobility of the poet or aoidos 'singer' by virtue of his being a dêmiourgos 'artisan in the community [dêmos]', as he is described in Odyssey xvii 381-385. 1 In Homeric poetry other professions besides the aoidos belong to this category of dêmiourgoi: the mantis 'seer', the iêtêr 'physician', the tektôn 'carpenter' (Odyssey, ibid.), and the kêrux 'herald' (Odyssey xix 135). 2 The dêmiourgoi are socially mobile, traveling from one dêmos to another. 3 For an example of a cognate institution, I cite the Old Irish áes cerd 'people of crafts [cerd]' the designation for artisans, including poets, who enjoyed a legal status even outside their own tuath 'tribe' as they traveled from one tuath to another. 4 Old Irish cerd 'craft' is cognate with Greek kerdos 'craft, craftiness; gain, advantage, profit'; the same Greek word, in the diction of poetry, can refer to the craft and the potential craftiness of poetry. 5

§13. Given the social mobility of the poet, who is teleologically evolving into the rhapsode, his cumulative exposure to multiform traditions in many places is analogous to the experience of an ethnographer who attempts to reconstruct back to a prototype the distinct but cognate versions of traditions in different but neighboring locales. What I am offering in effect is a hermeneutic model for explaining how the myth-making mind can become critical of variants in myth. I am arguing that such a faculty for criticism arises primarily from the factor of the poet's social mobility, which I offer as a substitute for what has been posited by others, namely, the factor of the poet's ability to write. 1

§14. A synthetic and critical tradition purports to represent a prototype of variant traditions, and the diachrony of its evolution thus becomes its own synchrony. Homeric synchrony, to take the clearest example, operates on the diachronically oldest recoverable aspects of its own traditions. 1

§15. The synthetic tradition, in order to survive, must prevail over the countless variant traditions from which it was constituted. And in order to prevail, the tradition must be performed. We turn to an observation of Claude Lévi-Strauss in his book on masks, where he confronts, perhaps more explicitly than elsewhere, the question of the relationships between localized myth-variants and localized ritual-variants, with particular reference to ritual objects such as masks: each performance of a myth, he notes, entails a re-creation of that myth, so that the latest version of the myth, in the context of its performance in the here and now, takes precedence over all previous versions. 1

§16. The latest performance is by necessity a crisis point for the traditions of myth, in that the latest performance determines what continues to be transmitted and what does not. There may be at any given time a multitude of latest performances by a multitude of performers in a multitude of places. Still, each latest performance is a crisis for what has been said in all previous performances, and the cumulative trends of latest performances determine what ultimately survives and what is lost. The crisis can be expected to deepen whenever the number of performances decreases or the occasions for them become progressively restricted. In any case each latest performance helps determine what is highlighted and what is shaded over, with the ever-present possibility that the shading will lapse, with the passage of time, into total darkness.

§17. In this evolutionary vision of change in oral traditions, I have deliberately used the imagery of light in order to bring home a point that is actually made by the poetic traditions of Greek myth making, although there the view is episodic, not evolutionary. As Marcel Detienne has documented in his survey of Archaic Greek poetics, the poetic power of mnêmosunê 'remembering' is traditionally associated with light, which is in conflict with the darkness of lêthê 'forgetting'. 1 What is illuminated or obscured by poetry is what is respectively preserved or lost in the tradition.

§18. The concept of lêthê 'forgetting', however, is not only negative. As Detienne points out, lêthê is not only the opposite of mnêmosunê 'remembering': it can also be an aspect of mnêmosunê. 1 For example, the goddess Mnêmosunê is described in the Theogony of Hesiod as giving birth to the Muses, divine personifications of the poet's power, so that they, through their poetry, may provide lêsmosunê 'forgetting' of sadnesses and of worries for humankind (53-55); whoever hears the Muses no longer memnêtai 'remembers' his own ills (Theogony98-103). By implication the highlighting of the glory of poetry is achieved by shading over anything that detracts from it. A bright light needs a background of darkness.

§19. Such a concept of mnêmosunê can be achieved only through an ever-present awareness of its opposite, lêthê. Without the obliteration of what need not be remembered, there cannot be memory--at least, from the standpoint of Archaic Greek poetics.

§20. Let us reformulate these thought patterns in terms of an opposition between unmarked and marked categories. 1 In an opposition of mnê- 'remember' vs. lêth- 'forget', mnê- would be the unmarked member and lêth-, the marked, in that lêth- can be included by mnê- as an aspect of mnê-. Besides the passage just considered from the Theogony, I cite another striking illustration, from a different source: in the ritual of incubation connected with the cult of Trophonios, the initiate drinks from the springs of both Lêthê and Mnêmosunê; this way the undesired mental state can be shaded over while the desired mental state is highlighted (Pausanias 9.39.8). 2

§21. To pursue the subject of these thought patterns even further, I cite an example of unmarked and marked opposition in the English language. In an opposition of the pronouns he and she, he is the unmarked member and she, the marked, in that she is included by he as the feminine aspect of being he. The masculine aspect of being he, by contrast, has to be achieved through an ever-present negation of the feminine. We may say something like: this is not a she, this is a he. Otherwise he does not, of and by itself, convey a masculine aspect. In generalizing statements, for example, he can stand for both he and she, as in "everyone may interpret as he chooses." 1

§22. Where the unmarked member excludes the marked member through a negation of the marked, the unmarked member receives a minus interpretation; where the unmarked member includes the marked, it receives a zero interpretation. 1 The minus interpretation of the unmarked member is ever-present in the context of a given Archaic Greek poem's references to itself as absolute truth, conveyed by a specialized mnêmosunê 'remembering' that excludes lêthê instead of including it. These relationships can be visualized as a larger circle of mnêmosunê 'remembering' that includes an inner area of lêthê 'forgetting' surrounding a smaller circle of specialized mnêmosunê 'remembering' that excludes the outer area of lêthê 'forgetting'. The area of forgetting is visualized as the ongoing erasure of things not worth remembering, erasure by way of lêthê 'forgetting'; the smaller circle of remembering, within the larger circle, is highlighted by the area of darkness surrounding it, the area of forgetting. In fact, a special word in the diction of Archaic Greek poetry formalizes this specialized and exclusive kind of remembering: that word is the negation of lêthê 'forgetting', namely a-lêtheia, normally glossed in English as 'truth'. A comparable case of minus interpretation in English can be seen in the word unforgettable. The alêtheia 'truth' of the poet is the nonerasure of the poetic glory that is his to confer. 2 The same concept is evident in the periphrastic expression oude me/se/he lêthei 'it does not escape my/your/his-her mind', which conventionally reinforces injunctions to be memnêmenos 'mindful, remembering'. 3

§23. Besides contrasting with negative thoughts about human ills, 1 in the periphrastic expression oude me/se/he lêthei 'it does not escape my/your/his-her mind', which conventionally reinforces injunctions to be memnêmenos 'mindful, remembering'. 2

§24. Besides contrasting with negative thoughts about human ills, 1 or erroneous thoughts that lead to injustice, 2 the alêtheia 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. 3 This argument brings us back to the observation of Lévi-Strauss that the latest performance of myth is in principle an occasion for selecting from and thereby potentially erasing versions available from countless previous performances. 4 In what survives of Archaic Greek poetry--and now I am using the word poetry in the broadest sense--what we keep finding is the ultimate extension of this principle, to the point where the latest version becomes the last version, a canonization that brings to a final state of crystalization what had been becoming an ever-less fluid state of variation in performance. 5 I attribute this canonization not so much to the phenomenon of incipient literacy as to the broader social phenomenon of Panhellenization. 6 I reiterate that this phenomenon is relative from the standpoint of an outsider to the tradition, in that some compositions are more Panhellenic in scope than others. From the standpoint of the insider to the tradition, however, in the here and now of performance, the Panhellenic perspective is the absolutist perspective of alêtheia 'truth'. 7

§25. This notion of canonization, as I have just outlined it, 1 is analogous to a concept used by scholars associated with the Museum housing the great library of Alexandria. 2 This concept is krisis, in the sense of separating, discriminating, judging (verb krînô) those works and those authors that are worthy of special recognition and those that are not. 3 The Alexandrian scholars who inherited the legacy of this process of separation, discrimination, judgment were the kritikoi 'critics', 4 while the Classical authors who were recognized as the survivors of this process of krisis were called the enkrithentes, 5 a term that corresponds to the Roman concept of the Classics, the classici, who are authors of the 'first class', primae classis. 6 The krisis of the enkrithentes, however, starts not with the Alexandrian scholars, nor even with the likes of Aristotle. 7 The crisis of this krisis happens to be already under way in the Archaic period. We must remind ourselves that songs and poetry were traditionally performed in a context of competition. 8 A striking example is the tradition of dramatic festivals at Athens, with the krisis 'judgment' of winners by kritai 'judges'. 9 But the criteria of the crisis are different. The very evolution of what we know as the Classics--as both a concept and a reality--was but an extension of the organic Panhellenization of oral traditions. In the Archaic period, I argue, the general principle that determines what is worthy of special recognition and what is not can be formulated as a question: what is Panhellenic, alêtheia, and what is not?

§26. For illustration, let us turn to the "Days" portion of the Works and Days of Hesiod, which begins with the following injunction:

êmata d' ek Diothen pephulagmenos eu kata moiran
pephrademen dmôessi
Hesiod Works and Days 765-766

Take care to mark the days 1 [of the month], which come from Zeus, giving each day its due, for the household-servants

The very first day of the month to be mentioned is a crisis point for the Panhellenic perspective, since it is the day when each polis is most idiosyncratic, with local traditions prevailing:

...triêkada mênos aristên
erga t' epopteuein êd' harmaliên dateasthai,
eut' an alêtheiên laoi krinontes agôsin.
Hesiod Works and Days 766-768

The thirtieth day of the month is best
for inspecting different kinds of work that have to be done and for apportioning food-supplies.
This is the day that people spend by sorting out [= verb krînô] what is truth [alêtheia] and what is not.

A commentator on the Works and Days remarks: "Civil calendars often fell out of step with the moon..., and it was on the 30th that errors arose. Each month had to be allowed either 29 or 30 days, but the last day was called triâkas (or in Athens henê kai neâ [meaning 'the old and the new']) in either case, the preceding day (?) being omitted in a 'hollow' month. So it was always a question of when to have the 30th." 2 In other words each polis had its own traditions about the calendar. At the 30th, there is a crisis about arriving at a Panhellenic norm from the standpoint of each polis. This norm is conveyed here by the notion of alêtheia 'truth', which, I argue, is the criterion of Panhellenism. Then the poet embarks on a catalogue of those days of the month that share the highest degree of consensus in local traditions, with the catalogue proceeding in a descending order of consensus. The 30th may be a crisis point, varying from polis to polis, but the crisis leads to a shared Panhellenic perspective. The poet has blotted over the differences, simply noting that alêtheia 'truth' is being 'sorted out' [= is in a crisis: the verb is krînô] on the 30th. After the 30th it is possible to arrive at a fixed sequence of given days traditionally spent in given ways by all Hellenes. 3

§27. The poet now highlights this fixed sequence, which is the Panhellenic perspective. Zeus, the god who is the planner of the universe, is an appropriate symbol for the organizing principle that underlies the Panhellenic perspective. With Zeus the poet begins the catalogue, as he then proceeds to present a synthetic overview of the days of the month:

haide gar hêmerai eisi Dios para mêtioentos:
prôton henê tetras te kai hebdomê hieron êmar
têi gar Apollôna chrusaora geinato Lêtô
ogdoatê t' enatê te. duô ge men êmata mênos
exoch' aexomenoio brotêsia erga penesthai
en de tetartêi mênos agesth' eis oikon akoitin,
oiônous krinas hoi ep' ergmati toutôi aristoi.
pauroi d' aute isasi triseinada mênos aristên
arxasthai te pithou kai epi zugon aucheni theinai
bousi kai hêmionoisi kai hippois ôkupodessin,
nea <te> poluklêida thoên eis oinopa ponton
eirumenai: pauroi de t' alêthea kiklêskousin.
tetradi d' oige pithon--peri pantôn hieron êmar--
. pauroi d' aute meteikada mênos aristên
êous geinomenês: epi deiela d' esti chereiôn.
haide men hêmerai eisin epichthoniois meg' oneiar:
hai d' allai metadoupioi, akêrioi, ou ti pherousai,
allos d' alloiên ainei, pauroi de t' isasin:
...eudaimôn te kai olbios, hos tade panta
eidôs ergazêtai anaitios athanatoisin,
ornithas krinôn kai huperbasias aleeinôn
Hesiod Works and Days 769-774, 800-801, 814-824, 826-828,

For what I now tell you are the days of Zeus the Planner.
To begin with, the first, 1 the fourth, 2 and the seventh 3 are each a holy day
(it was on the seventh that Leto gave birth to Apollo of the golden sword).
So too the eighth 4 and the ninth. 5 And yet, these two days of the waxing part of the month
are particularly good for various kinds of work by mortals. 6
On the fourth of the month bring home your wedded wife,
having sorted out [verb krînô] the bird omens, 7 which are best for doing this.
Further, few people know that the thrice-nine of the month is best
for opening a wine jar and for putting yokes on the necks
of oxen, mules, and swift-footed horses,
or for hauling a swift ship with many oars down to the wine-colored sea.
Few give it its true [alêthês] name. 8
Open your jar on the fourth. The fourth of the midmonth is the most holy of them all.
Again, few do it [=give it its true name]. 9 I mean the after-twenty [= the twenty-first], 10 which is best
when dawn comes. As evening approaches, it is less good.
These, then, are the days, a great blessing for earth-bound men.
The others fall in between. There is no doom attached to them, and they bring nothing.
Different people praise different days, 11 but few really know. 12
With respect to all of these days, fortunate and blissful is he who
knows all these things as he works the land, without being responsible to the immortals for any evil deed,
as he sorts out [= verb krînô] the bird omens, 13 and as he avoids any acts of transgression.

§28. For further illustration of the concept of alêtheia as a Panhellenic truth-value, I offer five additional passages. This truth-value is associated not just with poetry in the narrower sense but also with song, as in the lyric poetry of Pindar, and even with prose, as in the Histories of Herodotus. The five passages that follow, then, are selected from the widest possible range of Greek verbal art and range all the way from song to prose:

§29. Here we see the juxtaposition of what purports to be a unique and true Panhellenic version with a plethora of false versions, described as mûthoi 'myths'. 1 The mûthoi 'myths' are the outer core, containing traditions that are apista 'untrustworthy', while alêtheia 'truth' is the inner core, containing traditions that are pista 'trustworthy'. In referring to itself, the alêtheia of Panhellenic poetics represents mûthoi as if they were additions to the kernel of truth as formulated by alêtheia. 2 I would argue, however, that mûthoi 'myths' stand for an undifferentiated outer core consisting of local myths, where various versions from various locales may potentially contradict each other, while alêtheia 'truth' stands for a differentiated inner core of exclusive Panhellenic myths that tend to avoid the conflicts of the local versions.

§30. If we symbolize the exclusive sort of mnêmosunê 'memory' (which excludes lêthê: so a-lêth-eia) in terms of a smaller circle surrounded by an area of lêthê within a larger circle of inclusive mnêmosunê (which includes lêthê), we may imagine an ongoing erasure of mûthoi by lêthê within the outer circle, resulting in a-lêtheia. Thus a-lêtheia 'truth' is 'un-forgettable'. It cannot be emphasized enough that such a model of Panhellenic tradition is dynamic, not static. Through time, the inner and outer circles, along with the area of lêthê between them, keep shifting.

§31. Myths that are epichoric, that is, local, are still bound to the rituals of their native locales, whereas the myths of Panhellenic discourse, in the process of excluding local variations, can become divorced from ritual. The word mûthos 'myth' is associated with the epichoric rather than Panhellenic phase of myth making; its remaining links with ritual can be seen even in its etymology, if indeed mûthos is to be derived from the verb muô in the sense of 'I say in a special way' or 'I see in a special way', where the special way is the marked procedure of ritual. 1 Local traditions in ritual, and the myths that go with them, seem to be unfit for Panhellenic discourse. Thus Hecataeus of Miletus, at the beginning of his discourse, dismisses the local tales of the Greeks as polloi te kai geloioi 'many and laughable [geloioi]', as distinct from the things that he has to say, which are alêthea 'true [alêthea]' (FGH 1 F 1). Pollux uses the same notion of 'laughable' (geloio-) in referring to such distinctly epichoric concepts as the herm or the evil eye (7.108). All of which helps account for the negative implications of mûthos in the discourse of figures like Pindar (again Olympian1.29; also Nemean7.23, 8.33) and Herodotus (2.23.1, 2.45.1). 2 Moreover, earlier versions that claim Panhellenic authority can be dismissed by later versions as mûthoi: for example, the authority for the mûthos that is discredited by Herodotus at 2.23.1 is his own predecessor, Hecataeus. 3

§32. All this is not to say that a local or epichoric version, as distinct from a Panhellenic version, can be equated with the version that is supported and promoted by the polis. As an institution, the polis mediates between the epichoric and the Panhellenic: although it contains what is epichoric, it also promotes what is Panhellenic. 1 In the development from tribe to polis, certain older institutions, no longer compatible with any individual polis, coalesce to form Panhellenic institutions in which a variety of city-states may participate. A prime example is the institution of athletic games, which preserve aspects of tribal initiation patterns no longer suited to the ideologies of any Greek city-state. 2 Another case is the institution of poetry and song. The polis can best promote its prestige by promoting its own traditions in poetry and song on a Panhellenic scale. I have already cited the example of Theognis of Megara. 3 What is particular to Megara alone, grounded in Megara's own rituals and its own myths, tends to be shaded over; what is shared by Megara and by a wide variety of other city-states is highlighted. 4 Thus the polis is in such cases incompatible with mûthoi, in the narrow sense of "myths" that reflect the given city's diversity from other cities. For such reasons the implied concept of poliêtai 'people of the polis, citizens' is explicitly opposed by the neologism mûthiêtai 'people of mûthos' (Anacreon PMG 353). 5

§33. These words are spoken to Hesiod by the Muses, who are conferring upon him his power as a poet. To 'announce things that are true [alêthea]' is a model for the evolution of a unique Panhellenic Theogony out of an untold number of local theogonies, which seem etuma 'genuine' but are here set apart as a plethora of falsehoods. 1

§34. The tradition that informs this song is realized only in the here and now of performance, which is visualized as blossoming from the irrigating waters of the tradition. 1 The 'kharis of the past' is the cumulative response to all other potential realizations of the tradition, with the adjective palaio- 'of the past' implicitly contrasting the present performance. 2 In the diction of Pindar the present performance is conventionally described as neo- or nearo- 'new', which refers not to the novelty of a theme but to the ad hoc application of a myth to the here and now of those who attend and are the occasion of performance. 3

§35. Telemachus says these words to his mother Penelope, who had tried to stop the singer Phemios from singing what is described as the nostos 'homecoming' of the Achaeans (i 326). 1 Here too, as in the diction of Pindar, the concept neo- 'new' refers to the appropriateness of the myth to the situation in the here and now. 2 In this case Odysseus is both a prime figure in the myth and about to become a prime figure in the here and now narrated by the poem. His nostos is literally in the making, which is precisely the subject of the singer. Naturally the Panhellenic Nostos of Odysseus in the Odyssey is one to end all nostoi. 3

§36. The word apodeixis 'public display' refers to the medium for performing what we see written in Herodotus. 1 Since this medium makes public the deeds of men, what men do is also apodeixis, in that their deeds are being publicly witnessed. 2 In this case the apodeixis of the priests--and of Herodotus in turn--erases the deeds of all kings except one. To put it in terms of the passage, the making public of a tradition by way of performance--or at least by way of a written record that simulates performance--highlights the deeds of an exceptional figure as it darkens over the deeds of other figures. 3

§37. In light of these illustrations, let us return to the notion of a single Panhellenic tradition as opposed to a plethora of local traditions. It should be clear that this notion of Panhellenic is absolute only from the standpoint of insiders to the tradition at a given time and place, and that it is relative from the standpoint of outsiders, such as ourselves, who are merely looking in on the tradition. Each new performance can claim to be the definitive Panhellenic tradition. Moreover, the degree of Panhellenic synthesis in the content of a composition corresponds to the degree of diffusion in the performance of this composition. Because we are dealing with a relative concept, we may speak of the poetry of the Iliad and Odyssey, for example, as more Panhellenic than the poetry of the Epic Cycle. To put it conversely: a Cyclic poem like the Aithiopis, attributed to Arctinus of Miletus, is clearly less Panhellenic and more regional, focusing on the local traditions of Miletus. 1 Whereas both the Iliad and Odyssey refer to the immortalizing kleos 'glory' of Panhellenic epic that is to serve as compensation for the death of Achilles, 2 the Aithiopis is concerned rather with the personal immortalization of Achilles after death, on the island of Leuke (Aithiopis/Proclus, p. 106.12-15 Allen; Pindar Nemean4.49-50). This myth, as espoused by the Aithiopis, is anchored in local cult: Leuke is not only a mythical place of immortalization for Achilles but also the ritual place of his hero cult, localized in the territory of Olbia, daughter city of Miletus. 3 I have argued elsewhere that "the Cyclic epics are so different from the two Homeric epics not because they are more recent or more primitive but rather because they are more local in orientation and diffusion." 4

§38. To explain the superior prestige of the Homeric poems on the basis of their greater Panhellenic orientation and diffusion would also help account for the dependency of the whole Cycle on the narrative structure of the Iliad and Odyssey combined: the Cycle fills in the portions of the tale of Troy that had not already been told in its own way and on its own terms by the poetry of the Iliad and Odyssey. 1 In other words the text of the Epic Cycle is built around that of the Iliad and Odyssey. 2

§39. Clearly the poems of the Cycle, as also the Homeric Hymns, become fixed texts at various times that are all later than the textual fixation of the Iliad and Odyssey. 1 So we must assume either that poems like the Aithiopis, Little Iliad, and The Destruction of Ilion developed parallel to the Iliad and Odyssey or that they are derived from them. The latter alternative is an impossibility. The basic achievement of the "neoanalytic" approach as advocated by Wolfgang Kullmann and others is their cumulative demonstration that any given Homeric treatment of a given tradition may entail a refinement of that same tradition as attested in the Cycle. 2 In terms of my argument, such refinements are a reflex of further Panhellenization in Homeric poetry as distinct from other related poetry. And Such refinements cannot be accounted for merely by invoking the genius of a poet who stands out from among the rest.

§40. Paradoxically the textual fixation of the Homeric poems is older than that of the Cycle, in that the overall narrative of the Cycle is built around the Iliad and Odyssey, and yet the inherited themes of the Cycle appear consistently older than those of the Homeric poems. 1 I propose to account for this state of affairs by arguing that the Panhellenization of the Homeric tradition entailed a differentiation from older layers of Panhellenic epic tradition, and that 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. 2 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. 3

§41. This sloughing off of older layers could have been expressed in terms of myth as a break in genealogical continuity with the Homeridai, the descendants of Homer. 1 We earlier witnessed the explicit articulation of such a break in the case of Kynaithos the rhapsode: according to the scholia to Pindar (Nemean2.1c, III 29.9-18 Drachmann), Kynaithos of Chios, credited with the final form of the Homeric Hymn to Apollo, could no longer trace himself to Homer. In other words the tradition continued by Kynaithos is here being discredited by the sources as no longer authorized by the Homeridai. 2

§42. Myth can provide for indirect as well as direct heirs to the Homeridai. According to one version Stasinus, reputedly the poet of the Cypria (e.g., scholia A to Iliad I 5), was not the author of that poem but instead received the Cypria as a dowry from his father-in-law, Homer (Tzetzes Chiliades 13.636-640; Pindar F 265 SM, in the fuller context of Aelian Varia Historia 9.15). According to another myth Homer had taught the composition called the Capture of Oikhalia to Kreophylos of Samos, founder of the corporation of rhapsodes known as the Kreophyleioi, as a gift in return for the hospitality of Kreophylos when Homer had traveled from Chios to Samos (Strabo 14.1.18 C638; cf. Callimachus Epigram 6 Pfeiffer). 1 From the standpoint of the Homeridai, the Kreophyleioi were left out of Homer's genealogy, perhaps because they were rivals. The rivalry would concern the basic question: who is the authoritative representative of the epic tradition in a given community? In Archaic Sparta, for example, it appears that the Kreophyleioi of Samos were more authoritative than the Homeridai of Chios: tradition has it that the introduction of Homeric poetry took place in Sparta by way of the Kreophyleioi of Samos (Aristotle F 611.10 Rose). 2

§43. Wherever a given succession of rhapsodes is left out of the direct genealogy of Homer, we can expect that they may have to relinquish the central repertory of the Homeridai and make do with what is left over. Such is the case, it seems, with the distinct repertories attributed to Arctinus of Miletus, Lesches of Mytilene, Stasinus of Cyprus, and the other poets of the Cycle. 1

§44. To judge by the repertory of the Cycle, what was indeed left over was already finite 1 in that the Panhellenism of the Cycle, however less developed than that of the Iliad and Odyssey still entails the preservation of a few variants at the expense of the extinction of many others. Just as the Iliad and Odyssey had prevailed over the Cycle, so also the Cycle had teleologically prevailed over countless other epic traditions. In terms of myth, however, as long as a given tradition can somehow survive after losing to another tradition, the loser can be presented as the winner: it is as if the surviving tradition, deprived of the repertory of the prevalent tradition, had won its own remaining repertory as a concession from the prevalent one. 2 Such is the case in the myth that tells of the contest between Arctinus of Miletus and Lesches of Mytilene, which is won by Lesches (Phaenias F 33 Wehrli, in Clement Stromateis 1.131.6). 3

§45. A similar point may be made about the Contest of Homer and Hesiod tradition (Certamen, pp. 225-238 Allen). 1 Here Hesiod is represented as the victor on the basis of the specialization of Hesiodic poetry: telling of peacetime pursuits is deemed superior to telling of wartime ones (Certamen, p. 233.205-210 Allen). By implication, the telling of wartime pursuits is not open to Hesiod, whereas it is to Homer. The myth of this victory compensates for the fact that the poetry of Hesiod is relatively less Panhellenic than that of Homer. For example, we hear only of Homeric poetry, not of Hesiodic, in connection with reports about recitations that are privileged by law at the Panathenaia. 2

§46. This point about relative Panhellenism brings us back to the Epic Cycle. The story about the contest between Arctinus and Lesches, which Lesches wins (Phaenias F 33 Wehrli), may have a symbolic bearing on the interrelationship of narratives in the poetry attributed to Arctinus and Lesches. Both the Aithiopis and the Destruction of Ilion are attributed to Arctinus, whereas the Little Iliad is attributed to Lesches. On the basis of this story of a contest, combined with the fact that the story line of the Little Iliad of Lesches intervenes between that of the Aithiopis and that of the Destruction of Ilion, the two compositions attributed to Arctinus, it has been supposed that Arctinus, who "has the lion's share of the Cycle," had been "forced by Lesches' rising merits to yield him the [Little Iliad]." 1 This interpretation assumes that the story of a contest between Lesches and Arctinus is a historical fact, whereas I argue that it is a myth reflecting the historical relationship between the poetry attributed to these two figures. As we look at the narrative coverage of the Little Iliad as attributed to Lesches, this poet from the island of Lesbos, it seems at first to be an intrusion into the narrative of Arctinus of Miletus. But it would be more accurate to say that the narrative of Arctinus envelops the narrative of Lesches at both ends, almost engulfing it. Just as the Epic Cycle is built around the Homeric Iliad and Odyssey, so also, within the Cycle, the repertoire of Arctinus seems to be built around that of Lesches. There seems to be a stratification here, as if an earlier repertoire represented by Lesches of Mytilene were being enveloped by a later repertoire represented by Arctinus of Miletus. 2

§47. This observation, made on the level of content, has an analogy on the level of form: the Ionic diction of Homer, as also of the Cycle, envelops an earlier stage of Aeolic diction, so that Aeolic forms tend to survive only where they are not replaceable, in the metrical frames that they occupy, by corresponding Ionic forms. 1 From the linguistic point of view an even earlier layer of Homeric diction is closely akin to the surviving local dialect of Cyprus. 2 In this connection, we may take note of the fact that the Cypria makes explicit localized references to Cyprus (see, e.g., Lysimachus FGH 382 F 12; cf. "Apollodorus" Epitome 3.3-4); also, Stasinus, the reputed poet of the Cypria, is conventionally described as 'the Cypriote' (e.g., at Suda s.v. Homêros; cf. Athenaeus 334b). 3 Since the themes of the Cypria constitute the narrative basis for the entire epic tradition of the Trojan War, its Cypriote associations may be connected to an early phase in the Panhellenic diffusion of epic traditions on the island of Cyprus. 4

§48. In sum the Cycle may be viewed as a vestigial recapitulation in content of the chronological layering of the entire Homeric tradition, even though much of this layering has been sloughed off by the actual Iliad and Odyssey. The traditions of these sloughed-off layers, as represented by the Cycle, could have kept growing alongside the Homeric tradition, becoming even more Ionian in diction than Homeric diction itself. 1 In the few remaining fragments of the Cycle, an Ionic layer is clearly superimposed on the arguably Aeolic traditions represented by Lesches of Mytilene and the Cypriote traditions represented by Stasinus of Cyprus. 2

§49. The overarching Panhellenism of the Homeric poems, as we have seen, is evident from the differentiation of these poems from those of the Cycle. But the differentiation must be asserted on the basis of whatever distinct traditions are offered by the Cycle. Without such a pattern of assertion, the distinction between the Homeric poems and those of the Cycle poems can lapse into indifference in face of Homeric Panhellenism. Thus whereas the poems of the Epic Cycle could be attributed to individual poets like Arctinus and Lesches, they could also be attributed to the central figure of textual fixation, Homer. 1 According to one particular myth, for example, Homer himself was commissioned to "dictate" the Little Iliad, along with another composition called the Phokais, when he traveled to Phokaia (Herodotean Life of Homer 15, pp. 202-203 Allen). In this version any attribution of the Little Iliad to Lesches of Mytilene is wanting. For another example, I cite again Pindar F 265 SM, referring to the myth that told how the composition of the Cypria was Homer's dowry for his daughter (who was married to Stasinus: Aelian Varia Historia 9.15, Tzetzes Chiliades 13.636-640). Further, Herodotus goes out of his way to argue, apparently against certain traditions in his own time, that the poet of the Cypria is not Homer (2.116-117). 2 The words of Callinus apparently referred to the Seven against Thebes as Homer's poem (F 6 W); or again, Herodotus feels bound to express doubt whether the poet of the Epigonoi, the sequel of the Seven against Thebes, is indeed Homer (4.32). 3 By the time of Aristotle the safest thing was to say simply 'the author of the Cypria' or 'the author of the Little Iliad', as opposed to the prototypical and idealized Homer (Poetics 1459b). One latter-day critic puts it this way: "If Homer had a kind of claim to all this epic literature--a rather strong claim to the Hymns, a weaker one to the Cycle--and the alternatives to admitting his claim were either anonymity or naming a definite poet, what explanation can be given of the phenomenon except that the whole literature was the work of a school?" 4 For the phrase the work of a school, elsewhere deemed the Homeric canon by the same author, 5 I would substitute a Panhellenic tradition. Whereas the Aithiopis and the Destruction of Ilion are claimed by Miletus by way of attribution to Arctinus of Miletus, and the Little Iliad is claimed by Mytilene by way of attribution to Lesches of Mytilene in Lesbos, no single polis has an unequivocal claim on Homer (though his cult as hero at Chios seems definable by way of the Homeridai at Chios). 6

§50. In offering this sketch of the synthetic tradition that produced the Homeric poems, I should close by stressing that I do not deny the notion of "poets within a tradition." 1 The oral composer in the context of performance can execute considerable refinements in the act of recomposition. 2 The composer can even appropriate the recomposition as his own composition, as if it emanated exclusively from an owned authority: "This is my song." 3 But the gradual replacement of divergences in local oral traditions by convergences in Panhellenic oral tradition leads to an internal idealization of the very concept of the composer. If indeed Panhellenization gradually eliminates opportunities for recomposition in performance, we should then expect a commensurate elimination of opportunities for successive generations of performers to identify themselves as composers. I therefore do not argue generally that tradition creates the poet. 4 Rather I argue specifically that the Panhellenic tradition of oral poetry appropriates the poet, potentially transforming even historical figures into generic ones who represent the traditional functions of their poetry. The wider the diffusion and the longer the chain of recomposition, the more remote the identity of the composer becomes. Extreme cases are Homer and Hesiod. 5 To put it another way: the person of the poet, by virtue of being a transmitter of tradition, can become absorbed by the tradition. 6 Then the poet as an exponent of his poetry can become identified with and even equated with that poetry. Thus, for example, when Heraclitus (22 B 42 DK) says that Homer and Archilochus should be banned from contests in poetic performance, agônes, what is really being said is that rhapsodes should not be allowed to perform Homer and Archilochus. 7

§51. The appropriation of a historical person by the poetic tradition in which that person is composing can be visualized hypothetically in the following general schema of progressive phases, constructed from specific examples of performance conventions taken from a variety of traditional societies:

§52. The key to loss of identity as a composer is loss of control over performance. Once the factor of performance slips out of the poet's control--even if the performers of the poet's poetry have traditional comments about the poet as a composer--the poet becomes a myth; more accurately the poet becomes part of a myth, and the myth-making structure appropriates his or her identity. Such is the case with the poetry of a Homer or a Hesiod or an Archilochus, as performed by rhapsôidoi 'rhapsodes' like Ion of Chios. 1

§53. In sum, Panhellenism affects not only the form and the content of Archaic Greek poetry. It affects also the very identity of the poet. As the poet's composition is successively reperformed, the poet's identity is successively reenacted and thereby reshaped. 1


§2n1. The next paragraph is a reworking of N 1985.35 §17n3. From here on the Greek word polis is no longer printed in boldface.

§3n1. Snodgrass 1971.421, 435; cf. Snodgrass 1987, especially pp. 160, 165, and Morris 1988, especially pp. 754-755.

§3n2. N 1979.5-9.

§3n3. N 1982.43-49, 52-57, 59-60.

§3n4. N 1985.34-36.

§4n1. Ibid. Gentili 1985.45 discusses a poem for the Corinthian dead at Salamis: the inscription is Doric (CEG 131; Simonides EG 11), but the transmission is Ionic (Plutarch On the Malice of Herodotus 870e).

§5n1. For the related notion of intertextuality, in a specialized sense as applied to Homeric poetry, see Pucci 1987.29n30.

§5n2. This interconnected development of traditions is reflected in cross references from one tradition to another. I suggest that the notion of "cross reference" is indeed workable in the study of oral poetics, provided we understand that any references to other traditions in any given composition / performance would have to be diachronic in nature. On such cross referencing between the Iliad and the Odyssey traditions, see N 1979.35-58; also Pucci, pp. 240-242. For analogous cross referencing in Hesiodic poetry, I cite Theogony87, where the assertion that an ideal king can resolve even the greatest possible neikos 'quarrel' seems to presuppose a thematic association with the neikos between Hesiod and Perses at Works and Days35, which is treated by the Works and Days as an ultimate criterion, as the quarrel to end all quarrels; cf. N 1982.58- 59. In fact cross references can serve to distinguish one tradition from another. To cite an example: the description of the funeral of Achilles in Odyssey xxiv makes references to Patroklos and Antilokhos (77-79) in such a way as to signal that the Odyssey follows the Iliad tradition, not the Aithiopis tradition: see Edwards 1985b.223-225 (cf. N 1979.21; also Ch.7§15). A veritable network of cross references establishes the complementarity of the Homeric Iliad and Odyssey traditions, of the Hesiodic Theogony and Works and Days traditions (cf. Slatkin 1987). It may be that the distinctness of two separate major compositions within each of the two traditions resulted from evolutionary differentiations within the Homeric and the Hesiodic traditions. Moreover, the Contest of Homer and Hesiod tradition (Certamen, pp. 225-238 Allen), and the myth behind it (see further at Ch.2§45), implies an even more fundamental pattern of evolutionary differentiation between the Homeric and the Hesiodic traditions.

§6n1. See N 1982.48-49.

§7n1. See Ch.1§10, where I discuss the notion of Panhellenic as applied to international (that is, inter-polis) festivals like the Panathenaia. On the performance of old elegiac and iambic poetry by rhapsodes, see Ch.1§16.

§7n2. References and further discussion in Ch.3, where we shall see that the patterns of Panhellenization in song are even more complex than they are in poetry.

§7n3. That the rhapsodes may not be able to accommodate the compositions that they perform to the current political requisites of the audience is suggested in Herodotus 5.67.1, as interpreted by Svenbro 1976.44; for more on this passage, see Ch.2§43n1.

§8n1. Cf. Ch.1§21 and Ch.1§22.

§8n2. Coupez and Kamanzi 1962.8, quoted by Finnegan 1970.6.

§8n3. See Ch.2§5.

§9n1. Cf. Boutière and Schutz 1950, p. xii.

§9n2. Cf. Stevens 1986.43. Cf. Zwettler 1978.84-88 on the Arabic concept of the râwî.

§10n1. On categories of ownership of song (and / or dance), see Kunst 1958.2.

§10n2. Boulton 1954.4-5. Note here the oral performance's reference to the newness of its composition. It would be deceptive, here as elsewhere in oral traditions, to equate such "newness" with our own general notions of innnovation. We must be on guard against projecting into oral traditions an anxious modernist vision of the creative self, which can lapse all too easily into romantic scenarios of creation out of self-contained genius. We may achieve a more balanced formulation from the vantage point of anthropology: I cite Barnett 1953.39-95 on the concept of innovation, where he offers a cross cultural survey of nine possible social factors that promote innovation. Barnett's book has strongly influenced ethnomusicologists like Merriam (e.g., 1964.312-313), especially with his observation that whereas innovation in oral traditions may be initiated by individuals, the cultural background must allow it (for example, by way of collaboration of effort, expectation of change, and the like). On the relativity, from society to society, of the descriptive term improvisation, see the useful discussion of Merriam, p. 179.

§10n3. Davidson 1985.110; also in general pp. 103-142. The textual references here to the Shâhnâma follow the volume and page numbers of Bertel\'s 1966-1971. Again I draw attention to the "renewal" claimed by the composer.

§10n4. Ibid.

§10n5. Davidson, p. 109. Note again the notion of "renewal."

§11n1. I note the interesting ethnographic typologies discussed in Bausinger 1980.52.

§12n1. Commentary in N 1979.233-234. On the dêmos 'administrative district, population' in Archaic Greek poetic diction in the sense of 'local community', with its own traditions, customs, laws, and the like, see N, p. 149 §11n6; also Ch.9§2n4.

§12n2. Cf. Hesiod Works and Days25-26, where the aoidos 'singer' is juxtaposed with the tektôn 'carpenter' and the kerameus 'potter'; also with the itinerant ptôkhos 'beggar', ibid. Such a juxtaposition of aoidos and ptôkhos is also built into Odyssey xvii 381-385.

§12n3. There is clearly a hierarchy of professions within the category of dêmiourgoi, supplemented by the notion that the ptôkhos 'beggar' is at the bottom. The association of the ptôkhos with the category of dêmiourgoi relates perhaps to the notion that beggars, like dêmiourgoi, could appeal for immunity as they traveled from dêmos to dêmos, or perhaps to a poetic topos concerning the social scale as ranging all the way from the king at the very top to the beggar at the very bottom. It could be argued that such a mention of beggars on one end of of the social scale is intended as a symmetrical implication of kings on the other end: just as beggars can be listed at the bottom of a hierarchy of dêmiourgoi, so also kings can be listed at the top. The sliding scale in the social status of Odysseus from king to beggar back to king in the Odyssey may be connected with a poetic topos concerning the relationship between the king and the dêmiourgoi. On the ainos 'fable' of the Hawk and the Nightingale in Hesiod Works and Days 202-212, where the hawk is to a king as the nightingale is to a poet as dêmiourgos, cf. N 1979.238-241.

§12n4. Old Irish tuath 'tribe' (as ruled by a king) is cognate with Umbrian touto 'civitas' and German Deutsch. On the áes cerd, see [J. F.] Nagy 1985.33, 35 and 239n48; cf. also Meid 1974.

§12n5. Cf. Ch.6§77.

§13n1. For the theory that literacy is the primary impetus toward a critical faculty, see, for example, Goody and Watt 1968.

§14n1. N 1982.47-49, 52. For a particularly compelling formulation with regard to Arabic traditions, I cite Zwettler 1978.221.

§15n1. Lévi-Strauss 1979.153-163, especially pp. 162-163.

§17n1. Detienne 1973.22-27. For example, Lêthê or 'Forgetting' personified is descended from Night in Hesiod Theogony227/224; Mnêmosunê 'Remembering' is contrasted with darkness in Pindar Nemean7.12-16.

§18n1. Detienne, pp. 69-80.

§20n1. For these terms, see the Introduction, Intro. §12.

§20n2. See the discussion by Detienne, p. 74.

§21n1. Waugh 1982 compares the French usage of the masculine gender as the unmarked member of an opposition with the feminine, in that the masculine can stand for the category as a whole: thus an adjective describing both masculine and feminine categories will be put into the masculine: des hommes et des femmes intelligents.

§22n1. For these terms, as used by Waugh 1982 following Jakobson 1939, see again the Introduction. Waugh, p. 302, pictures the marked-unmarked relationship as "a subset-set relationship where the marked category is the subset and the unmarked category is the set," or alternatively as "a figure-ground relationship where the marked pole is the figure and the unmarked pole is the ground."

§22n2. Given that the smaller circle within the larger circle symbolizes the specialized sort of mnê-, that is, a-lêtheia, I would say that the larger circle that contains lêth- would correspond to the function of the Muses, who help humans forget some things so that they may remember others. The root *mnâ- of mnê- 'remember' may in fact be related to the root *mon-t- (or *mon-th-) of Mousa 'Muse' (Hesiod Theogony53-55, 98-103). The etymological connection is certain if Mousa is to be derived from the root *men-, expanded as *mon-t- (or *mon-th-), which is one of several possibilities entertained by Chantraine DELG 716. The relationship of the root *men- with the expanded form *mnâ-, as in mnê-, is clear: Chantraine, p. 703.

§22n3. The pertinent passages are discussed in N 1983.44. This expression oude me/se/he lêthei 'it does not escape my/your/his-her mind' implies a synchronic understanding of the word alêtheia as a compound consisting of privative a- and the root lêth-. In the formulation of Cole 1983.12, the reference of alêtheia is "not simply to non-omission of pieces of information...but also to not forgetting from one minute to the next what was said a few minutes before, and not letting anything, said or unsaid, slip by without being mindful of its consequences and implications." (For a critique of Heidegger's celebrated explanation of alêtheia, see Cole, pp. 7-8.) Cf. also Detienne 1973.48n107.

§23n1. I cite again Hesiod Theogony53-55, 98-103). The etymological connection is certain if Mousa is to be derived from the root *men-, expanded as *mon-t- (or *mon-th-), which is one of several possibilities entertained by Chantraine DELG 716. The relationship of the root *men- with the expanded form *mnâ-, as in mnê-, is clear: Chantraine, p. 703.

§23n2. The pertinent passages are discussed in N 1983.44. This expression oude me/se/he lêthei 'it does not escape my/your/his-her mind' implies a synchronic understanding of the word alêtheia as a compound consisting of privative a- and the root lêth-. In the formulation of Cole 1983.12, the reference of alêtheia is "not simply to non-omission of pieces of information...but also to not forgetting from one minute to the next what was said a few minutes before, and not letting anything, said or unsaid, slip by without being mindful of its consequences and implications." (For a critique of Heidegger's celebrated explanation of alêtheia, see Cole, pp. 7-8.) Cf. also Detienne 1973.48n107.

§24n1. I cite again Hesiod Theogony53-55, 98-103.

§24n2. On this theme, see Detienne 1973.29-50.

§24n3. This is not to say, of course, that the convergent version may not be complex, containing multiformities within its overarching uniformity.

§24n4. Cf. Ch.2§15; cf. also Ch.2§51n1.

§24n5. Royce 1977.104 points out, with reference to traditions of dance, that various structures of performance, as they become progressively more rigid, can suffer "abrupt confrontation and loss."

§24n6. The threat of "abrupt confrontation and loss," to use the expression quoted immediately above, could help promote an impetus for recording by way of writing. But a critical attitude toward myth is caused not by the technology of writing but rather, more fundamentally, by the crisis of confrontation between variants of myth. See Ch.2§13.

§24n7. We may well ask: how does the local perspective contribute to the Panhellenic, and to what degree does the Panhellenic perspective recognize the local? From the standpoint of the local tradition, the best chance for self-assertion is a process of self-selection that accommodates the Panhellenic tradition. Note the discussion by Royce 1977.164 of the repertory of some 90 sones (dances) among the Zapotec of the Isthmus of Tehuantepec: in asserting their identity to outsiders, the Zapotec tend to select just three of these 90 sones. Royce notes (ibid.) that "these three are the dances that any non-Zapotec would name if asked about 'typical' dances of the Isthmus," and that it is these three dances that are synthesized by the Ballet Folklórico in its suite "Wedding in Tehuantepec."

§25n1. For a history of the usage of canon to designate a selective listing of authors and works, see Pfeiffer 1968.207.

§25n2. For an introduction to the era of Alexandrian scholarship, see Pfeiffer, pp. 87-233.

§25n3. For a survey of this usage, see Pfeiffer, p. 117.

§25n4. See Pfeiffer, pp. 89, 242.

§25n5. Pfeiffer, pp. 206-207. Cf. Horace Odes I 1.35, and the comments of Pfeiffer, p. 206.

§25n6. Pfeiffer, p. 207. The canon as conceived by the Alexandrian scholars is not to be confused with the actual collection of works housed in the great library of the Museum at Alexandria. The Pinakes or 'Tables' of Callimachus, in 120 books, was intended not as a selection but as a complete catalogue of the holdings of the Museum, generally organized along the lines of formal criteria, including meter. For an informative discussion, see Zetzel 1983.98-100, who stresses that the Alexandrian system of classification was "eminently suitable for describing the literature of pre-Alexandrian Greece" (p. 99).

§25n7. On the principles of selection, from Aristotle to the Alexandrians, see Pfeiffer, pp. 117, 205. This is not to assume that there was an ongoing process of actual selecting of Classical (as opposed to current) authors in the period of Alexandrian scholarship; I cite Page 1953.68, who doubts that "any ancient lyrical poet whose works were in circulation up to the Alexandrian era was omitted by the Alexandrian editors from their collection" (for a critique of this formulation, see Ch.3§3n2). In the case of epic, Quintilian Institutio oratoria 10.1.55 notes explicitly that the Alexandrian editors Aristophanes and Aristarchus included no contemporary poets into the ordo, or canon, barring even Apollonius of Rhodes.

§25n8. See Ch.1§21n5.

§25n9. For the wording, see, for example, the description in Plato Laws 659ab. There is a stylization of this institution in the Frogs of Aristophanes, in the form of a competition between Aeschylus and Euripides (see Ch.13§41); in this context, we may note the usage of the word krisis and the corresponding verb krînô at Frogs 779, 785, 805, 873, 1467, 1473. Cf. also Dunkel 1979.252-253.

§26n1. First we had the "works"; now we have the "days."

§26n2. West 1978.351.

§26n3. For the apparent exception on the island of Keos, see the passages quoted by West, p. 351.

§27n1. In the Odyssey, the new moon is the context for a festival of Apollo (xiv 162 = xix 307; xx 156, 276-278, xxi 258): West 1978.352.

§27n2. For example, Aphrodite was specially worshipped on this day: sources in West, ibid.

§27n3. The most important holy day of Apollo: sources ibid.

§27n4. For example, the 8th at Athens was the day for honoring Poseidon and Theseus: West, p. 353.

§27n5. For example, the 9th at Athens inaugurated the City Dionysia: ibid.

§27n6. That is, they may be holy days, but they are not necessarily holidays. This hedge suggests that the 8th and the 9th are less "Panhellenic" than the 1st, 4th, and 7th. This reading differs from that of West, p. 132, whose punctuation indicates that he takes duo ge men êmata as referring to what follows (the 11th and 12th at line 774) rather than to what precedes (the 8th and the 9th at line 772). I take duo ge men êmata at 772 and amphô ge men at 774 to be parallel in referring to what precedes in the syntax.

§27n7. Note the parallel expression concerning bird-omens at Works and Days828, as discussed at Ch.2§27n1.

§27n8. The Hesiodic name 'thrice-nine' would be the Panhellenic designation, as implied by the word alêthês. Note the observations at Ch.2§26 about alêtheia at Works and Days768. Local designations of this day may have been subject to tabu. The number thrice-nine is particularly sacred: see the references collected by West, p. 361.

§27n9. This interpretation differs from what is found in the standard editions.

§27n10. Note again the periphrasis, as in the case of thrice-nine at line 814.

§27n11. Here we see the localized perspective.

§27n12. Here we see the Panhellenic perspective. 'Know' is in the sense of histôr 'the knowledgeable one', as at Works and Days792.

§27n13. Note the parallelism between verse 828 here and verse 801, Ch.2§27, where again the verb krînô 'sort out' is used with reference to divination by birds. The crisis of sorting out the right and the wrong bird-omens is implicitly parallel to the crisis (again, verb krînô) of sorting out what is alêtheia 'truth' and what is not. In order to appreciate the importance of ornithomanteiâ 'divination by birds' in the whole poem, we may note that Works and Days828 had served as a lead-off for a concluding stretch of verses, now lost, containing instructions on the interpretation of bird omens (West 1978.364-365). A bird omen is central to the entire ethical message of the Works and Days, that is, the ainos 'fable' (202) of "The Hawk and the Nightingale" (202-212), on which see Ch.9§7n6.

§28n1. That is, kharis personified. For the specific purposes of this book, I consistently interpret kharis as a 'beautiful and pleasurable compensation, through song or poetry, for a deed deserving of glory'. This word conveys both the beauty ("grace") and the pleasure ("gratification") of reciprocity.

§28n2. West 1978.49 observes: "The aorist of mêdomai, unlike the imperfect, means not 'planned' but 'wrought'." In the range of meaning from 'planned' to 'wrought', I submit, we see the range of meaning within the single word kerdos 'craft', on which see Ch.2§12.

§29n1. I shall argue in Ch.4 below that Pindar's "unique and true Panhellenic version" represents the official aetiology of the Olympics in Pindar's time.

§29n2. In visualizing an outer core of 'falsehoods' and an inner core of 'truth', I am following the interpretation of Young 1986, who adduces, besides Pindar Olympian1.28-32, Plato Republic 377a, Pausanias 8.2.6 and Strabo 1.2.9 C20. I would add Thucydides 1.21.1: oute hôs poiêtai humnêkasi peri autôn epi to meizon kosmountes mallon pisteuôn. These parallels help put Pindar Nemean7.20-23, also adduced by Young, in a new light; note the singularity of the Pindaric patha at Nemean7.21 as opposed to the plurality of the Homeric polla...pathen algea in the prooemium of the Odyssey (i 4). See Ch.7§5n2.

§31n1. Cf. Ch.1§29.

§31n2. The verb mûtheomai 'say', derivative of mûthos, seems less susceptible to such negative implications: see Pindar Pythian4.298 and Hecataeus FGH 1 F 1. On Aristotle's rehabilitation of the word mûthos, see Halliwell 1986.57-58, especially n16. On mûthos in Plato: Brisson 1982.

§31n3. How and Wells 1928 I 170.

§32n1. This point is developed further at Ch.5§15. It is no coincidence that the decline of the polis in the fourth century and thereafter coincides with the decline of Panhellenism.

§32n2. At Ch.5§15 and following, I refer to this phenomenon in shorthand as the exoskeleton of the polis.

§32n3. Cf. Ch.2§4.

§32n4. Detailed discussion in N 1985.

§32n5. Cf. Detienne 1981.92-94. As the scholia to Odyssey xxi 71 make clear, the mûthiêtai 'people of mûthos' in the island-polis of Samos are the people who represent stasis 'discord' (stasiastai). On the theme of stasis as a negative way of achieving a definition of the polis, see Ch.12§51 and following.

§33n1. See N 1982.47-49. The phrase alêthea gêrusasthai 'announce things that are true [alêthea]' at Theogony28 is one of a set of variants, including alêthea muthêsasthai 'tell [verb mûtheomai] things that are true' at Iliad VI 382, Homeric Hymn to Demeter121, etc. (also attested as a textual variant at Theogony28) and etêtuma muthêsasthai 'tell [verb mûtheomai] things that are real [etêtuma]' at Homeric Hymn to Demeter44. I suggest that these variations result from a chain of differentiations setting off a marked Panhellenic version from unmarked versions that are ostensibly local or at least more local. The variant gêrusasthai 'announce' represents a differentiation of marked gêrûsasthai 'announce' from unmarked muthêsasthai 'tell'; also, the variant alêthea 'things that are true [alêthea]' represents a differentiation of marked alêthea 'things that are true' from unmarked etêtuma 'things that are real'. In each case, the marked member differentiates a concept that is Panhellenic (alêthês, gêrûsasthai) from an earlier concept that is perceived as obsolete (etêtumo- [or etumo-], muthêsasthai) with reference to the new marked member. At each stage of differentiation, we must allow for the probability that the unmarked member of the opposition had once been the marked member in earlier sets of opposition.

§33n2. Cf. Ch.2§28n1.

§33n3. Here I am following the interpretation of Race 1986.108.

§34n1. On water as a symbol of poetry or song, see Ch.10§9n2.

§34n2. The word palaio- 'of the past' implies a contrast specifically with neo- 'of the present, new': cf. Chantraine DELG 851.

§34n3. Cf. [A. M.] Miller 1982.114. Cf. Ch.2§10n2.

§35n1. On nostos as both 'homecoming' and 'song about homecoming', see N 1979.97 §6n2.

§35n2. Cf. again [A. M.] Miller 1982.114.

§35n3. Cf. N 1979.98. Telemachus is "wrong" in not understanding that the myth applies to the situation in the here and now. For him, the "newness" of the song has the surface-meaning of mere novelty (cf. the interpretation in Plato Republic 424bc). But he is "right" in insisting that the singing proceed. This way, the nostos sung by the singer may ultimately be fulfilled in the nostos of Odysseus, which is the "novelty" of the Odyssey--the "news" of what finally happened in the Odyssey. Penelope, by contrast, is "right" in understanding that the song applies to the present, but she is "wrong" in interpreting it at this particular moment in the overall narrative of the Odyssey. What is absolutely right, not wrong, can emerge only from the overall narrative in progress.

§36n1. Cf. Ch.8§5 and following.

§36n2. Ibid.

§36n3. As the narrative of Herodotus proceeds, the spotlight keeps shifting. At the beginning of this account, the spotlight is on a figure described as the very first king of Egypt (2.99.1-4); after he is named and his deeds are accounted for, it is said that the priests who were the informants of Herodotus had records of a sequence of 330 other names that followed the first pharaoh, including one woman (2.100.1). This woman is then highlighted, with a recounting of her name and some of her deeds (2.100.2-4). Then follows the statement just quoted: 'About the other kings, they [= the Egyptian priests] had no public statement [apodeixis] to tell of their deeds, since there was nothing distinguished [= literally 'bright'], except for the last [king]' (2.101.1). At this point, the spotlight falls on the last in this sequence of 330 pharaohs, with a recounting of his name and some of his deeds (2.101.1-2), capped off by a reaffirmation that this king at least had these deeds to 'show for himself' (2.101.2; the verb is apo-deik-numai, on which see Ch.8§5 and following), whereas the other kings did not (again 2.101.2). Then the spotlight shifts again, to the king who came after this last one, and there follows a particularly lengthy and detailed accounting of this pharaoh's name and some of his notable deeds (2.102.1-2.111.1).

§37n1. On the Milesian orientation of the Aithiopis, see Pinney 1983, who argues convincingly that the iconographic theme of Scythian archers on Attic late sixth-century vases is akin to local epic traditions specifically associated with Milesian colonization on the northern coast of the Black Sea, and that these local epic traditions are reflected in the Aithiopis of Arctinus of Miletus. In the mythological traditions of the mother city, Miletus, the notion of "the Other" apparently became particularized as "the Scythian" in the social context of the daughter cities on the northern coast of the Black Sea (on the subject of Milesian colonization in this area, notably at Olbia, see Bravo 1974). Just as the mother city tends to replicate its social structure, divisions and all, in the daughter city, so also the new social experiences of the daughter city, such as contacts with "new" kinds of barbarians (in this case, Scythians), become incorporated into the ideology of the mother city (see Figueira 1981.192-202, especially p. 199). In light of the fact that Archaic Miletus and Megara were as a rule linked together in rivalry against Corinth and were both predominant as the colonizers of the coastline of the Propontis and the Black Sea (Figueira 1985.276), I note that the poetic traditions of Megara, like those of Miletus, draw attention to the theme of Scythians. These Megarian poetic traditions are preserved in the corpus attributed to Theognis of Megara, where the ideology of Megara incorporates the ideologies of the daughter cities of Megara along with those of the mother city (N 1985.51 §38n1 and Figueira 1985.127-128), and I cite here in particular the Scythian reference at Theognis825-830 (with the commentary of Figueira 1985.146). Even the Theognidean vision of the kakoi, the ethically inferior, as sociopolitically inferior savages who threaten the polis from the outside (Theognis 53-68; cf. N 1985.44 §29n4, 51 §39n2, 54), may convey a colonial point of view adopted from a daughter city on the coast of the Black Sea (cf. Figueira 1985.129). In contemplating the partial "Scythian" characteristics of the Achilles figure in the mythological traditions of Miletus/Olbia (as surveyed by Pinney 1983, especially pp. 133-139; cf. Alcaeus F 354 V), I see a typological parallelism in the figure of Rostam in the Shâhnâma of Ferdowsi: this national hero of Iranian epic traditions has partial "foreign" characteristics that give form to his function as both "the Other," an outsider, and "the Self," an insider to the body politic as represented by the ruling shâh (see Davidson 1985, especially pp. 61-103). On the equation of the ephêboi 'pre-adults' of Elis with Scythians in Photius Lexicon s.v. sunephêboi, see Hartog 1980.59-79 (especially pp. 71-72) and Vidal-Naquet 1986.133.

§37n2. Discussion in N 1979.29, 35-36, 38-40, 184, especially with reference to Iliad IX 413 and Odyssey xi 489-491.

§37n3. Details in Pinney 1983.143n56 and 145n94, who accepts as early a dating as the late sixth century B.C. (p. 133); cf. N 1979.167 (I agree with the reservations expressed by Pinney, p. 144n64, about the thesis that Achilles was originally a god of the underworld). On names like Êlusion 'Elysium' and Makarôn Nêsoi 'Islands of the Blessed' as simultaneous designations of a mythical place of immortalization and a ritual place of hero cult, see N, pp. 189-192.

§37n4. N 1979.8 §14n1: there I make clear that I part company with Griffin 1977, who thinks that the Homeric epics have screened out most of "the fantastic, the miraculous, and the romantic" (p. 40) elements characteristic of the Cycle because Homer was a superior or "unique" poet. (For a useful critique of Griffin's position, see Young 1983.166n32.) Instead I would stress that the fantastic and the miraculous elements in the Cycle characterize the religious ideology of local cults, reflecting the more localized interests of individual city- states or groups of city-states. The same goes for the romantic element of love stories, again for the most part screened out by the Homeric epics: it goes without saying that love affairs lead to conceptions of heroes, a basic theme of genealogical poetic traditions that promote the localized interests of the status quo. On the relationship of Panhellenic poetic traditions with the more localized ktisis ('foundation, colonization') poetic traditions of various city-states, see N 1979.8 §14n1 (with cross references) and especially pp. 139-141; also N 1982.63-64 and 1985.51 §38n1 and 63 §51n2.

§38n1. For a helpful survey, see Allen 1924.72-75.

§38n2. I use the concept of text in the broadened sense outlined at Ch.2§5.

§39n1. For a comprehensive survey of linguistic and other criteria that can be applied for an overall relative chronology, see Janko 1982.

§39n2. Kullmann 1985, especially pp. 17-18n37.

§40n1. A survey of Archaic Greek iconographical evidence, as assembled by Fittschen 1969 and juxtaposed by Kannicht 1982 with the evidence of Archaic Greek poetry, shows that the earliest identifiable pictorial responses to epic concern predominantly the themes of the Cycle, not those of the Iliad or Odyssey.

§40n2. This explanation differs from the one offered by Kannicht, p. 85, who accounts for the relative absence of early pictorial references to the Homeric Iliad and Odyssey by arguing that the artistic limitations of early Greek iconographical traditions made it too difficult for these traditions to react to the great artistic achievements of the Iliad and Odyssey.

§40n3. A similar point can be made in the case of the contrast between Iliad II 557-570 and Hesiod F 204.44-51, where both passages describe the extent of the dominion of the hero Ajax. As Finkelberg 1988 argues, the Homeric passage from the Iliad, part of the Catalogue of Ships, is more innovative than the Hesiodic passage in drastically restricting the realm of Ajax, even though the text fixation of the Homeric passage is presumably earlier than that of the Hesiodic. As Finkelberg also argues (pp. 39-40), the Homeric version is politically advantageous to Athens under the Peisistratidai and, secondarily, to Argos in the era of Pheidon, as also to Corinth and even to Sparta, while it is disadvantageous primarily to Megara. Such a version, which suits the politics of the more powerful city-states, is clearly more Panhellenic in scope than the Hesiodic version (which itself is distinct from the overtly pro-Megarian version: Finkelberg, p. 40). I should add that the parallelisms between Iliad II 557-570 and Hesiod F 204.44-51 (as illustrated by the underlinings in Finkelberg, pp. 32-33) suggest that the Homeric version reduces the realm of Ajax not so much by deleting elements found in the Hesiodic version but by augmenting the traditional elements and then reassigning the greater portion to figures other than Ajax.

§41n1. On the Homeridai, see Ch.1§11.

§41n2. Cf. Ch.1§12.

§42n1. The basic testimonia are conveniently available in Allen 1924.228-229 and Burkert 1972.76n10. Cf. N 1979.165-166.

§42n2. Tradition also has it that it was Lycurgus, lawgiver par excellence, who brought to Sparta the Homeric poems, which he acquired from the descendants of Kreophylos at Samos, according to Plutarch Life of Lycurgus 4. It is said of the poems that Lycurgus 'had them written down', egrapsato, and that he then 'assembled' them (ibid.). I draw attention to a further detail in the narrative of Plutarch (ibid.):

ên gar tis êdê doxa tôn epôn amaura para tois Hellêsin, ekektênto de ou polloi merê tina, sporadên tês poiêseôs, hôs etuche, diapheromenês: gnôrimên de autên kai malista prôtos epoiêse Lukourgos
Plutarch Life of Lycurgus

'for there was already a not-too-bright fame attached to these epics among the *greeks, and some of them were in possession of some portions, since the poetry had been scattered, carried here and there by chance, and it was *lycurgus who was the first to make it [= the poetry] well-known' (ibid.)

. For an alternative tradition, according to which Lycurgus met Homer directly, see Ephorus FGH 70 F 103 and 149 (by way of Strabo 10.4.19 C482). The notion of a disassembled book, scattered here and there throughout the Greek world, and then reassembled for one particular time and place by a wise man credited with the juridical framework of his society, is parallel to the story about the making of the Book of Kings in the Iranian epic tradition. According to Ferdowsi's Shâhnâma I 21.126-136, a noble vizier assembles mubad-s, wise men who are experts in the Law of Zoroaster, from all over the Empire, and each of these mubad-s brings with him a "fragment" of a long-lost Book of Kings that had been scattered to the winds; each of the experts is called upon to recite, in turn, his respective "fragment," and the vizier composes a book out of these recitations. As Davidson 1985.123 points out, "It would seem from this passage that the authority of the unified Empire and of the unified Book of Kings is one." The vizier reassembles the old book that had been disassembled, which in turn becomes the model for the Shâhnâma 'Book of Kings' of Ferdowsi (Shâhnâma I 23.156-161). We see here paradoxically a myth about the synthesis of oral traditions that is articulated in terms of written traditions, as Davidson argues in detail (pp. 111-127). For a comparable myth in Irish traditions, concerning the recovery of the "lost" Cattle Raid of Cúailnge, see [J. F.] Nagy 1986.292-293.

§43n1. So also, perhaps, in the case of Kynaithos of Chios: it may well be that his "Homeric" repertory was not the Iliad and Odyssey. This Kynaithos, as we have seen, is said to have been the 'first' rhapsode to recite Homeric poetry at Syracuse, in the 69th Olympiad (504/1 B.C.), according to Hippostratus (FGH 568 F 5). By implication Kynaithos was the first recorded winner in a seasonally recurring festival at Syracuse that featured a competition of rhapsodes. Cf. Ch.1§12. As another possible example of a distinct repertory, see Ch.1§10n4 on Herodotus 5.67.1, where the reference to the "Homeric" repertory of the rhapsodes who were banned from Sikyon implies not the Iliad and Odyssey but rather an overall Seven against Thebes epic tradition. In the Contest of Homer and Hesiod 287-315 Allen, a myth tells how Homer visited the people of Argos, gave a performance there, and was subsequently given great honors by that city-state. In Callinus F 6 W, the Seven against Thebes is attributed to Homer. In contrast Herodotus 4.32 may be taking a stance that is detrimental to Argos when he expresses doubt whether the poet of the Epigonoi, the sequel of the Seven against Thebes, is indeed Homer.

§44n1. Eleven books of the Cypria (Proclus summary, p. 102.10 Allen), five of the Aithiopis (p. 105.21), four of the Little Iliad (p. 106.19), two of the Destruction of Ilion (p. 107.16), five of the Nostoi (p. 108.15), and two of the Telegonia (p. 109.7).

§44n2. A clear example is the Homeric Hymn to Hermes, where the functions that Hermes "wins" from Apollo correspond to earlier stages in patterns of differentiation involving the aoidos 'singer' and the mantis 'seer'. The later stages of these patterns were taken over by the figure of Apollo. Discussion in N 1982.56-57.

§44n3. See Ch.1§21n5. On the associations of the Little Iliad of Lesches with the local traditions of Lesbos, see the discussion and bibliography in Aloni 1986.120-123. In general, Aloni's book offers an interesting case in point illustrating the interdependence between two distinct narrative traditions owned by two distinct communities in conflict. The communities in question are the cities of Mytilene in Lesbos on the one hand and on the other Athens in the era of the Athens Athenian dynasty of tyrants, the Peisistratidai. The focal point is "Trojan" Sigeion (cf. Herodotus 4.38.2), an outpost of Athenian power, founded by the Peisistratidai as an intrusion into a geographical area controlled by Mytilene, whose rival outpost was the Akhilleion (Herodotus 5.94.1-2; Sigeion was at the mouth of the river Skamandros: Herodotus 5.65.3). Aloni argues that the contemporary winners of the conflict, the Athenians, needed to appropriate at least part of the narrative traditions owned by the losers, the Mytilenaeans, in order to legitimize their own expansionistic presence in Sigeion (see especially pp. 65n8, 91). In the context of the conflicting claims of the Athenians and Mytilenaeans, note the use of the word apo-deik-numi at Herodotus 5.94.2; cf. Ch.6§31 (also Ch.11§5). Note too the story of the duel to the death between the Athens Athenian Phrynon, an Olympic winner, and the Mytilenaean Pittakos, tyrant and lawgiver, as recorded in Diogenes Laertius 1.74 and Strabo 13.1.38 C599-600 (and omitted in Herodotus 5.95: cf. Plutarch On the Malice of Herodotus 858ab). Both sources agree that the Mytilenaeans won in this conflict, only to lose later in an arbritration undertaken by Periandros, tyrant of Corinth (this aspect of the tradition is not omitted in Herodotus 5.95.2). The accretion of narrations concerning an earlier victory and a later loss by the Mytilenaeans recapitulates, it seems, a hierarchy of accommodations between rival narrative traditions: the ultimately losing side is pictured as having won first.

§45n1. For a possible allusion to Contest of Homer and Hesiod tradition in Hesiod Works and Days657, see N 1982.66.

§45n2. Cf. Ch.1§10 and following.

§46n1. Allen 1924.64.

§46n2. Cf. Kurylowicz's fourth law of analogy, as discussed at Intro. §13.

§47n1. Cf., for example, Householder and Nagy 1972.785. For a restatement of the facts that necessitate the positing of an Aeolic phase in the evolution of Homeric diction: West 1988.162-163 (with bibliography).

§47n2. Householder and Nagy, pp. 783-785.

§47n3. As Huxley 1969.134-135 points out, there are also traces of Cypriote localization as the setting for Homeric Hymn10, where the poet, in praying to Aphrodite as the queen of Salamis in Cyprus (10.5), treats her as a local Muse in asking her to give him a song that brings gratification (ibid.); also in Homeric Hymn6, addressed to Aphrodite as queen of all Cyprus (6.2-3), where the poet prays that the goddess grant him victory in the competition: dos d' en agôni | nikên tôide pheresthai 'grant that I carry away victory in this contest [agôn]' (6.19-20).

§47n4. Janko 1982.176 gives reasons for estimating 750 B.C. or thereabouts as the terminus post quem for any possible proliferation of early phases of the Cypria tradition on the island of Cyprus, but he doubts that Cyprus was "the area in which this tradition grew to maturity."

§48n1. The case is clear in Hesiodic poetry: despite its ultimate local provenience, Aeolic Boeotia, this poetry is more Ionic in diction than even Homeric poetry. See Janko 1982.85, 197; cf. N 1982.70-72.

§48n2. In Little Iliad F 12 Allen, however, as quoted by Clement Stromateis, the occurrences of a in place of ê may suggest an Aeolic layer of transmission: see West 1971.308n3. Alternatively such occurrences may reflect editorial aeolicisms.

§49n1. This point is stressed by Allen 1924.75.

§49n2. It is the more complex pattern of the Homeric poems, where one level of narrative is being subordinated to another, as contrasted with the more simple pattern of the Cypria, that convinces Herodotus that the poet of the Cypria cannot be Homer. Further discussion in Ch.14§16.

§49n3. We have already seen that the city-state of Argos apparently attributed to Homer the entire Seven against Thebes tradition. See Ch.2§43n1.

§49n4. Allen 1924.71, who adds that the "work" of such a poet could have been "gradually taken from him" by the "survival and revelation of local tradition" (ibid.). I disagree with this additional point to the extent that I interpret the "revelation of local tradition" not as something that is taken away from Homeric poetry but rather as something that is generally rejected by Homeric poetry.

§49n5. Allen, p. 76, where he speculates that, by the time of Peisistratos, "the Cycle was all but finished and the Homeric canon all but closed."

§49n6. For references on the Homeridai, cf. N 1979.165 §25n4.

§50n1. As Griffith 1983.58n82 suggests that I do in N 1979.5-6, 296-297.

§50n2. To call such refinements "innovation," however, can be deceptive: see Ch.2§10n2.

§50n3. Cf. Ch.2§10. For a survey of conventions expressing the simultaneous appropriation of authority and authorship in Iranian poetic traditions, see Davidson 1985.103-142. In the conventions of Greek oral poetic traditions, self-identification is particularly appropriate in the context of the prooimion or 'prelude': brief discussion in N 1982.53.

§50n4. As Griffith 1983.58n82, again, suggests that I do in N 1979.5-6, 296-297.

§50n5. Cf. N 1979.295-300.

§50n6. I explore this topic at length in N 1985.

§50n7. See Ch.1§10n4.

§51n1. On categories of ownership of song (and/or dance), see Kunst 1958.2. See also the examples cited at Ch.2§10. On the ownership, in North American Indian traditions, of personal songs obtained in the vision quest, see Merriam 1964.83. Cf. also Merriam and d'Azevedo 1957.623: "Most songs seem to have been embellished, consciously or unconsciously altered over time, combined, improvised, forgotten and 'caught' again in new form as one's own. The last are thought of as 'new' or 'my' songs, but the singer has no inclination to hide the fact that he was influenced by another song, and that 'I just changed it a little'. Nevertheless, it does become a 'new' song." For bibliography on the relativity of the descriptive term improvisation, see Ch.2§10n2.

§51n2. As in Rwanda praise poetry, with memorization and remembering of the "original" composer by name: Finnegan 1977.79; cf. also p. 75. In Somali poetry: "A poet's composition...becomes his own property, under his own name, and another poet reciting them has to acknowledge from whom he has learnt them" (Finnegan, p. 74). For a possible trace of this type of attribution in the South Slavic traditions, see Lord 1960.19-20.

§51n3. Rwanda and Somali examples: Finnegan, p. 83. Cf. Merriam 1964.83 on the ownership of songs by kinship groups. See Boutière and Schutz 1950.xii on the Provençal convention that requires the joglar 'performer' to narrate the vida 'life story' of the trobador 'composer' whose composition he is about to perform. On the related genre of the razo (from Latin ratiô) as a sort of prelude, see Boutière and Schutz, p. xiii. The purpose is to recover the context of composition.

§51n4. Kirk 1962.88-98 offers a different model of Homeric transmission, where his division into various successive stages presupposes a general pattern of decline. For a critique, see Jensen 1980.113-114.

§52n1. There are instances where we have specific evidence that the transmission is regulated in the context of a hero cult in honor of the poet. See N 1979.304 §4n3 on the cult of Archilochus; cf. also p. 124 §9n1 on what appears to be a cult of a clearly historical figure, Pindar himself. See also the discussion in N 1982.49-51 on the cult of Hesiod, to be supplemented with the comments at Ch.1§22n4. In this discussion it is argued that the traditions of Archaic Greek poetry already contain, as a built-in program, so to speak, the ideology that makes cult heroes of poets.

§53n1. On the reenactment of the poet through reperformance, see Ch.12§68.

Go to Previous chapter; 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.