Reading Bakhtin Reading the Classics: An Epic Fate for Conveyors of
 the Heroic Past

Gregory Nagy
[[Originally published in Bakhtin and the Classics (ed. R. B. Branham; Evanston, IL 2002) 71–96. Copyright, Northwestern University Press. In this online version, the original page numbers of the printed version are indicated within braces (“{” and “}”). For example, “{73|74}” indicates where p. 73 of the printed version ends and p. 74 begins.]]
In the title of my essay, there is no punctuation
 between reading Bakhtin and reading the Classics because there is
 meant to be no break in the syntax. In the essay itself, there is no
 break between reading Bakhtin and reading the way he reads the
 Classics. The only break is between the Classical and the
 non-Classical, as seen by Bakhtin. In this essay, even that break
 will in the end disappear. The actual distinction, however, between
 the Classical and the non-Classical will remain. My thesis is that
 this distinction is part of a larger system that highlights the fact
 of distinctness in some contexts and shades it over in others.
My thesis, in other words, is that the system of
 distinguishing the non-Classical from the Classical is itself
 Classical. In differing with Bakhtin on this one point, it is
 possible to build arguments on the basis of his own arguments, as
 deployed in his seminal and irreplaceable essay, "Epic and Novel"
 ([1975]). [1] That essay of Bakhtin will help highlight a
 feature of ancient Greek epic that is central to my essay: this
 genre of epic, I will argue, conveys the heroic past to the present
 in two different ways, one of them seemingly closed off from our
 present and the other, manifestly open to it. The closure or the
 opening depend on whether or not the epic highlights a break between
 the fate of its heroes in the past and the life of its audience in
 the present.
For my essay, the case in point is the fate of
 the Phaeacians in the Homeric Odyssey, which the epic
 narrative connects inextricably with the fate of the hero Odysseus.
 By conveying Odysseus back to his homeland Ithaca on one of their
 magical ships, the Phaeacians are fated to lose their status as
 conveyors from - or even of - one kind of "reality" to another.
 Their fate prophesies the possible fate of epic itself in the
 conveying of the Classical past to the non-Classical present. The
 subtitle of my essay conjures this fate.
Let us begin with an overview of Bakhtin's
 understanding of the Classical and the non-Classical as articulated
 in another work of his, Rabelais and his World ([1965]). [2] I cite a key formulation in this book: "The
 object that has been destroyed remains in the world but in a new
 form of being in time and space; it becomes the 'other side' {71|72}
 of the new object that has taken its place." [3] Here we see the essence of Bakhtin's
 distinction between the Classical and the non-Classical.
The context of Bakhtin's formulation is this: he
 is arguing that the principle of negation in "popular-festive
 imagery" is never abstract but always concrete. [4] The social forces that Bakhtin subsumes
 under the single hermeneutic term carnival embody this
 negation: "carnival celebrates the destruction of the old and the
 birth of the new world." [5] In Bakhtin's essay "Epic and Novel," we see
 a prime example of the "new form," that is, the novel, while the
 older "other side" is embodied in the "genre" that he calls
In my work on the social forces of
 canon-formation as culminating in the ancient Greek Classics, [6] 
 I link Bakhtin's distinction between the
 "new form" and the older "other side" with an axiom known in
 linguistic theory as "Kuryłowicz's fourth law of analogy," which I
 paraphrase as follows: "when two forms come into competition for one
 function, the newer form may take over that function while the older
 form may become relegated to a subcategory of its earlier
 function." [7]
This axiom is pertinent to what Prague School
 linguists describe as the distinction between "marked" and
 "unmarked" members of a given opposition:
The unmarked category is the general
 category, which can include the marked category, whereas the
 reverse situation cannot hold. For example, in an opposition
 of the English words long and short, the
 unmarked member of the opposition is long because
 it can be used not only as the opposite of short ("I am reading a long book, not a short one") but also as
 the general category ("how long is the book?").
The unmarked member is inclusive, in that
 the marked member can be an aspect of the unmarked. It can
 be exclusive, however, if it negates the marked member, as
 when we say "that is not short, it is
 long." The negation of the marked by the unmarked
 has been called the minus interpretation of the
 unmarked (for example, "long, not short") as distinct from
 the zero interpretation (for example, "long"); the
 assertion of the marked member is the plus
 interpretation (for example, "short," or "short,
 not long"). The term plus interpretation designates
 not "positive" but "marked, either negatively or
The zero interpretation of the
 unmarked member includes, as an overarching principle, both
 the minus interpretation of the unmarked member and the plus
 interpretation of the marked member. The opposition of
 long and short is a matter of length.
 Further, the opposition of unmarked order and marked
 disorder is a matter of overall order. [8] {72|73}
Applying this distinction between unmarked and marked
 categories to "Kuryłowicz's fourth law of analogy," we can say that
 the newer form, in replacing the older form, acquires the unmarked
 function of the older form, which in turn develops a marked
 function. By "older form" I mean the form that is already assigned to a given function, whereas by "newer form" I mean the
 form that is about to be assigned to that function. For
 example, English ‘quick’, cognate of Latin vîvus 'alive',
 lost the meaning 'alive, living' and became semantically specialized
 in the sense of 'lively' and, eventually, 'quick'; the older meaning
 survives residually in such expressions as the quick and the
 dead or bite the nails to the quick. [9] 
 If we contrast "newer" ‘alive’ with “older"
 ‘quick’, we can say that the newer form has an older meaning, which
 is undifferentiated or unmarked, while the older form has a newer
 meaning, which is differentiated or marked.
If we extend this linguistic analogy to Bakhtin's
 hermeneutics in contrasting the "newer" form of the novel with the
 "older" form of the epic, we can say that the newer form has an
 older function, which is undifferentiated or unmarked, while the
 older form has a newer function, which is differentiated or marked.
 This formulation, however, does not match Bakhtin's overall model:
 for him, the newer form, novel, has the newer function, while the
 older form, epic, has the older function, showing the older "other
This aspect of Bakhtin's description of the novel
 leads to an overly narrow formulation of epic as genre. [10] 
 A case in point is the ancient Greek
 evidence. Bakhtin's typologies of epic, as juxtaposed with his
 typologies of the novel, may indeed suit in some ways the Homeric
 Iliad, but they cannot be reconciled with the Odyssey, an
 epic that features characteristics that Bakhtin associates
 explicitly with characteristics of the novel, most prominently
 "heteroglossia" and "centrifugal" narrative. As John Peradotto
 notes, "I would venture to say that close readers of Homer are far
 more likely to recognize the Odyssey in Bakhtin's characterization
 of the novel than in his account of epic." [11]
To be sure, Bakhtin's own formulations of the
 contrast between epic and novel are hardly monolithic - or, let us
 say, monological. Whereas his essay "Epic and Novel" features these
 two forms "in a relation of opposition," another essay (no. 22 in
 Todorov's list of Bakhtin's papers: dated at 1936-38) treats the
 novel as a "species" of "the great epic form"; later, in the 1963
 version of his work on Dostoevsky, Bakhtin seems to reverse himself:
 now epic is one of three aspects of the "novelistic." [12] 

These three different formulations may yet be
 reconciled if we apply diachronic as well as synchronic
 perspectives. In using the terms synchronic and
 diachronic, I am following a linguistic distinction
 made by Ferdinand de Saussure. [13] For Saussure, synchrony and diachrony
 designate respectively a current state of a language and a phase in
 its evolution. [14] These terms diachronic and
 synchronic are not to be used as synonyms for
 historical and current {73|74} respectively:
 "Diachrony refers to the potential for evolution in a
 structure. History is not restricted to phenomena that are
 structurally predictable." [15]
It is easiest to start with Bakhtin's middle
 formulation: that the novel is a "species" of the epic. Before the
 emergence of the novel as a distinct form that was set apart from
 epic, we can say that the earlier form, epic, had already contained
 "novelistic" as well as "non-novelistic" functions. In terms of
 marked and unmarked categories, the "novelistic" was a marked
 function within epic. The novelistic function could be an aspect of
 epic and still be opposed to epic in special contexts. The
 novelistic Odyssey could be an aspect of the epic tradition
 of Homer even if it opposed in some ways that same tradition as
 defined by the Iliad. [16] It is only when the novelistic became the
 novel, that is, when it became a new genre that rivaled or even
 replaced the old genre of epic, that the function of the novel was
 no longer marked but unmarked.
It is not enough, though, to say that epic is
 marked while the novel is unmarked. Rather, epic becomes marked when
 it is set in opposition to the novel as a new form that has taken
 over the old unmarked function of the epic.
Also, it would be misleading to infer that the
 Classical is marked while the non-Classical is unmarked. My thesis
 is that "Classical" is the real unmarked category. It is opposed not
 to the "non-Classical" but to the "anti-Classical," which is marked.
 The problem is, Bakhtin concentrates on a minus-interpretation of
 the Classical. In terms of minus-interpretation, the only way that
 unmarked "Classical" can become marked is by opposing it to the
 already marked anti-Classical. Here, then, is a corollary to my
 thesis: Bakhtin's models for the anti-Classical, as embodied in his
 hermeneutic terms "novel" and "carnival," are really
 plus-interpretations of marked categories.
[[I schematize as follows:]]
unmarked marked
non-Classical vs. Classical
Classical vs. anti-Classical, e.g. "novel,"
epic vs. "novelistic" function within epic
novel as genre vs. epic as genre
Bakhtin's own academic career re-enacts the
 interaction of the non-Classical and the Classical. His
 formation as a student at St. Petersburg owes much to
 the influence of a leading professor at the university, Faddej F.
 Zelinkij - that is the way this Classics scholar is known to experts
 in Russian intellectual history. [17] Classicists in America and Western Europe
 know him by the polonized form of his name, Tadeusz Zieliński. [18] 
 Among Classicists, Zieliński is highly
 regarded for his research in the classicized forms of Comedy; his
 student, Bakhtin, is associated with what at first seems to be a
 quintessentially non-classicized form or non-form, the "novel" as
 realized in the Gargantua and Pantagruel of
 François Rabelais. {74|75}
Whereas the teacher defines a genre, comedy, the
 student explores the apparent defiance of generic categories in the
 case of whatever it was that Rabelais created in the historical
 context of the sixteenth century. [19] In the "preclassical" period of the
 seventeenth century, preceding the reign of Louis XIV, "Rabelais did
 not yet appear exceptional." [20] Soon thereafter, however, "the atmosphere
 in which Rabelais was understood disappeared almost entirely, and he
 became a strange and solitary author who needed special
 interpretation and commentary." [21] Not only did the literary form of Rabelais
 recede, with the passage of time, from self-definition as a genre:
 it was also progressively demoted from higher to lower art. To put
 it in extreme terms: Rabelais slipped from canonical to
 anti-canonical status. [22] As Bakhtin argues, a parallel slippage is
 evident in the early reception-history of the Don Quixote of Cervantes. [23]
In my work, I re-apply Bakhtin's model by
 comparing the genre-amorphousness of Rabelais with that of the
 archaic Greek figure Archilochus, who is represented in relatively
 later archaic Greek traditions as a proto-poet of not only blame
 (invective) but also praise (encomium): in Pindar's
 Olympian 9, for example, Archilochus is credited with
 the invention of a "spontaneous" kind of victory song or ode,
 pictured as some kind of primordial praise poem. [24] 
 The prototypical victory ode of Archilochus
 is pictured by the Classical victory ode of Pindar as something
 non-Classical or even anti-Classical.
Bakhtin himself applies a similar model of
 genre-amorphousness to what he reconstructs as the prototypical
 phases of Greek tragedy: "In antique culture tragedy did not exclude
 the laughing aspect of life and coexisted with it. The tragic
 trilogy was followed by the satyric drama which complemented it on
 the comic level." [25] Citing Albrecht Dieterich's
 Pulcinella, [26] Bakhtin continues: "antique tragedy did not
 fear laughter and parody and even demanded it as a corrective and a
 complement." [27]
In my work I have counterargued that
 pre-Classical "literary" forms like those represented by Archilochus
 and by early drama (as it was evolving before the Classical phases
 of the City Dionysia at Athens) can be explained in terms of earlier
 phases of the "Classical," and that the evolving differentiations
 remain part of a "Classical" system. [28]
In this regard, it is essential to distinguish
 between historical and hermeneutic usages of the term "Classical."
 Historically, we may say that "Classical" represents the second half
 of the fifth century BCE, coming after what we may call the
 "pre-Classical" period. Hermeneutically, however, "Classical"
 represents whatever system is current, which is opposed by what we
 may call not so much the "non-Classical" or pre-Classical but the
 "anti-Classical." My point remains that the "anti-Classical," as
 embodied in the hermeneutic terms "novel" and "carnival," is part of
 the overarching system that is the "Classical." The terms
 "non-Classical" and even "pre-Classical" mask the markedness of
 "novel," of "carnival." The Classical is dynamic, ever-evolving, and
 inclusive of an anti-Classical element.
This formulation puts into perspective Bakhtin's
 observations about "narrow {75|76} genrism." [29] He uses this term to describe various
 failed attempts to categorize what I now call the anti-Classical. My
 formulation also reinforces Bakhtin's arguments against the
 "historic allegorical method" in approaching the phenomenon of
 allusion: "The image is always deeper and wider, it is linked to
 tradition, it has its own aesthetic logic independent of the
 allusion." [30] Using the term "image," Bakhtin is here
 coming to terms with the underlying system that is the Classical,
 this time in an unmarked sense. The return to the Age of
 Saturn, [31] the return to an Aesopic vision of
 verbal communication among gods and men and
 animals, [32] and even the return to a world view where
 death is a part of life [33] - all these things can be seen as
 explorations, through time, of the Classical by the
The collapsing of distinctions in the
 retrospective self-references of the Classical to the imagined
 proto-Classical is not just a matter of reverting to the
 "novelistic" form in the "carnivalesque" sense that Bakhtin gives
 it. There are also other forms that accentuate in other ways the
 undifferentiated aspects of the Classical. A case in point is the
 word ainos, which applies not only to the carnivalesque and
 "low art" form of the fables of Aesop but also to the serious and
 "high art" form of the victory odes of Pindar: ainos means
 'fable' in the world of Aesop but 'praise poem' in the world of
 Pindar's victory odes. [34]
As a genre, the victory ode conventionally calls
 itself ainos (as in Pindar Olympian 2.95), and
 such self-references make it distinct from epic, which does not call
 itself ainos but which contains, within its own narrative
 frame, explicit references to ainos as a genre of its own
 (as in Odyssey xiv 508). [35]
The victory ode also calls itself kleos 'glory' (as in Pindar Nemean 7.63), and so too does epic
 (as in Iliad II 486). In this respect, the difference
 between the two genres is this: whereas the kleos of epic
 centers on the glorification of heroes in the remote past, the
 kleos of the victory ode extends from the heroes of the
 remote past to the humans of the immediate present. [36] 
 A case in point is the following
 juxtaposition of the honorand in the present with the hero in the
 past, who is also an honorand in his own right:
λέγεται μὰν Ἕκτορι μὲν κλέος ἀνθῆσαι
 Σκαμάνδρου χεύμασιν | ἀγχοῦ, βαθυκρήμνοισι δ᾿
ἀμφ᾿ ἀκταῖς
 Ἑλώρου, | ... δέδορκεν | παιδὶ τοῦθ᾿
Ἁγησιδάμου φέγγος ἐν
 ἁλικίᾳ πρώτᾳ
It is said that kleos bloomed [37] for Hektor near the streams of
And near the steep cliffs that rise above [the
 stream] Heloros, […] this light shone
upon the coming of age
 of [Chromios] the son of Hagesidamos.
Pindar Nemean 9.39-42
In the
 victory ode, we see the collapsing of distinctions between the
 glorification of idealized men in the past as heroes and the
 glorification of idealized men in the present as participants in the
 world of heroes. In this way, both hero and latter-day man
 participate in both past and present, in both epic and victory ode.
 {76|77} Other distinctions too are collapsed: the ordeals of heroes
 as warriors of the past are expressed in the same language as the
 ordeals of men as warriors of the present. Moreover, the ordeals of
 latter-day men as warriors who fight for their native land are
 expressed in the same language as the stylized ordeals of these same
 men as athletes competing for victory at Panhellenic or local
 athletic contests. Martial and athletic ordeals were considered
 parallel rituals. [38] Finally and most importantly for the
 victory ode, the hero's mythical ordeal in the past is expressed in
 the same language as the athlete's ritual ordeal in the present,
 leading to victory in athletic competitions. [39]
I disagree, then, with the comment of Bruce
 Braswell on the pertinence of lines 39-42 in Pindar's
 Nemean 9, as just quoted, to the contemporary honorand
 Chromios, who is being praised by the words of the victory ode:
 "Chromios' kleos here like Hektor's came from martial, not
 athletic prowess." [40] I counterargue that Pindar's victory odes
 programmatically shade over rather than highlight the distinction
 between martial and athletic ordeals. In the case of Nemean 9, the defining occasion that is being celebrated is the victory of
 Chromios at an athletic event, a chariot race. [41] The reference at lines 39-42 to another
 victory of Chromios, this one martial rather than athletic,
 complements the overriding reference to the athletic victory at
 hand, which defines the occasion for the whole song. In the victory
 odes of Pindar, the kleos of martial achievements cannot be
 separated from the overall kleos of athletic victory. [42] 

I argue further that the parallelism of hero and
 athlete in the genre of the victory ode emphasizes the ordeals of
 the hero, not his victories per se. [43] There is a religious motivation for this
 emphasis: the ritual activity of athletic competition was predicated
 on the central mythological fact of a hero's ultimate ordeal,
 death. [44] This mythical ordeal of the hero is
 notionally re-enacted in the ritual ordeals of the latter-day
 athlete or warrior as he struggles to achieve victory. [45] 
 Whenever the athlete or warrior wins, his
 victory is conceived as emerging from the merged ordeals of both the
 heroes of the past and the athletes or warriors of the
I also disagree with the following comment of
 Braswell on Chromios the honorand: "Pindar, by comparing him to
 Hektor, seems to suggest that the victory [of Chromios in war] was
 in no small part due to him [Chromios]." [46] Yes, the victory ode is making a reference
 here to a martial exploit of Chromios in the historical "present" -
 more precisely, in the glory days of the honorand's youth. Still,
 the well-known fate of Hektor in the distant epic past—his gruesome
 death at the hands of Achilles—makes it unlikely that this
 comparison with Hektor was intended merely to give credit to the
 latter-day honorand for a victory in battle. The focus, rather, is
 on the ordeal of the hero, which is likened to an ordeal in a
 formative stage of the honorand's career. It is against this
 backdrop of ordeals that the ultimate success of Chromios can be
In short, the genre of the victory ode, by
 calling itself kleos, blurs the distinctness of the genre of epic as
 kleos precisely because it collapses distinctions that
 are maintained in epic (as we shall see) between the heroes of the
 past who are {77|78} inside the heroic narrative and the men of the
 present who are the outside audience of that narrative. If epic and
 praise poetry are respectively the unmarked and marked members of an
 opposition, then kleos conveys a zero interpretation of the
 marked member, praise poetry.
By contrast, the genre of the victory ode
 accentuates its distinctness from epic by calling itself
 ainos in the sense of 'praise poetry'. In this way,
 ainos conveys a plus interpretation of the marked
 member, praise poetry. Praise poetry, as a marked member in
 opposition to unmarked epic, signals something that the epic is not.
 Epic is not ainos. As we have seen, epic can refer to
 instances of ainos, but it does not call itself
 ainos as praise poetry calls itself
In the poetics of Pindar, praise poetry presents
 itself as older than epic and even as the prototype of epic. [47] 
 Such an ontogeny of epic, as proclaimed by
 the praise poetry of Pindar, is recapitulated by the phylogeny
 posited by Aristotle, who in his Poetics actually derives epic from
 a prototypical form of praise poetry (1448b32-34). [48] 

The implications of Aristotle's insight are
 enormous. My own attempts to follow through on this insight
 culminated in a book that exceeds 400 printed pages in length,
 Pindar's Homer, which deals with the formal (including
 metrical) parallelisms between epic and the praise poetry of
 Pindar's victory odes. [49] Even a work of this size, however, is a
 mere beginning. For purposes of the present argumentation, I simply
 quote the final paragraph that sums up my whole book, paraphrasing
 Pindar's poetic program in terms of his own traditional poetics:
From the lofty vantage point of Pindaric
 song, Homer is Pindar's Homer. Pindaric song is both staying
 in the present and reaching back into the past within
 itself. It does not have to go outside for the purpose of
 bringing the epic inside. Epic is within it, and from it
 epic shall forever flow. [50]
As an undifferentiated older alternative to epic,
 the praise poetry of the victory ode is analogous to the epic as an
 undifferentiated older alternative to the novel. In contrasting the
 "newer" form of epic with the "older" form of praise poetry as
 retrojected by Aristotle, we can say that the newer form, epic, has
 appropriated the older function of praise poetry as an
 undifferentiated or unmarked medium of expression, while the older
 form, praise poetry, now has a newer function as a differentiated or
 marked medium.
As a differentiated medium, the new marked
 function of praise poetry as opposed to epic is to privilege the
 present with reference to the heroic past, which remains an absolute
 point of reference. As we have seen, Pindar's victory odes
 consistently emphasize the victories stemming from the ordeals of
 athletes in the present, with reference to the archetypal and
 definitive ordeals of heroes in the past.
By contrast with praise poetry, the unmarked
 function of epic is simply to privilege the heroic past with
 reference to the non-heroic present. The present, which is
 ever-shifting, is needed only as a foil for the privileging of the
 permanent {78|79} and absolute heroic past: to claim, as epic does,
 that the heroes of the past were larger than life is to look at this
 past from the shifting and all-too-human perspectives of life in the
 present. The present is needed merely as a contrast, a background,
 for the past. As a background, the present needs to be shaded over
 in order to highlight the foreground of the past.
In the specific case of Homeric poetry as epic,
 however, the shading over of the present becomes disproportionate to
 the highlighting of the past. It is almost as if there existed only
 a foreground in the past and no background at all in the present. As
 I have argued at length elsewhere, there are special historical
 reasons for this near-absolute detachment of epic, in the form of
 Homeric poetry, from the occasionality of the present. These reasons
 have to do with a strong trend of Panhellenism in the Homeric
 tradition, which promotes a near-blackout of local cultural and
 political concerns that traditionally link the world of the present
 to the world of the heroes in the past:
In the epic poetry of Homer the gap that
 separates the heroes of the past and the men of the present
 could not and would not be bridged. Little wonder, then,
 that heroes could lift stones that not even two of us
 "today" could even manage to budge (Iliad XII
 445-449). [51]
Epic is a problematic genre in the history of Greek
 literature. Diachronically, epic as a form is newer than praise
 poetry; historically, however, the form of epic as attested in
 Homeric poetry is far older than the form of praise poetry as
 attested in the victory odes of Pindar. [52]
The complications increase with the two-way
 opposition of epic: it is the unmarked member in its opposition with
 praise poetry but it is the marked member in its opposition with the
 novel. Moreover, the unmarked function of epic, to privilege the
 past with reference to the present, is symmetrical with the marked
 function of praise poetry, to privilege the present with reference
 to the past. This symmetry helps further shape the opposition of
 epic and novel. In terms of this new opposition, the privileging of
 the past can become a marked function for epic as set apart from the
 novel, which privileges the present without having to depend on any
 absolute point of reference in the past. Now the privileging of the
 past can become an absolute for epic, by way of excluding the
 present altogether. In its restriction to the past, the newer
 function of epic seems older (but is not), while the older function
 of praise poetry seems newer (but is not) in its free access to the
[[I schematize as follows:]]
unmarked genre marked genre
epic, with unspecific reference to the present vs. praise poetry, with specific reference to the present
novel, with unspecific reference to the past vs. epic, with specific reference to the past
Here, finally, we return to Bakhtin's essay
 on Epic and Novel. I submit the following list of ten of his
 formulations (the numbers in square brackets indicate the pages
 of the 1981 printing), supplemented with recapitulations of my
 earlier comments. Each of Bakhtin's formulations about "epic"
 can be reinterpreted as complementary insights into the
 "Classical" - as an unmarked and inclusive concept.
1. On the "absolute past" of epic: "The epic was never a
 poem about the present, about its own time (one that became
 a poem about the past only for those who came later)." [13]
As a point of reference, the world of heroes in ancient
 Greek traditions is absolutized not only for epic but also
 for praise poetry. The praise poem refers specifically to
 the absolute heroic past as well as the present. Epic
 concerns this same past - but with unspecific reference to
 the shifting non-heroic present. The elimination - or, more
 accurately, the tendency toward near-elimination - of
 references to the present in Homeric poetry is a special
 case, as we will see later on, and it cannot be generalized
 for epic writ large. There are other kinds of epic where
 references to the agenda of the present are clearly in
 evidence. A locus classicus is the reference to Marcellus in
 Virgil Aeneid 6.883 (cf. Horace Odes 1.12.46). [53]
2. "… epic discourse is infinitely far removed from
 discourse of a contemporary about a contemporary addressed
 to contemporaries." [13-14]
This formulation applies, at least on the surface, in the
 case of Homeric poetry. In other kinds of epic, however, it
 does not necessarily apply (as in the case of Virgil, cited
 above). In oral traditions, moreover, we find cases of epic
 where the given discourse of the contemporary performer is
 about his contemporaries as well as about heroes - and is
 addressed to the heroes as well as to the
 contemporaries. [54]
3. "[The epic past] is walled off absolutely from all
 subsequent times, and above all from those times in which
 the singer and his listeners are located." [15f]
Even the explicit instances of "walling off" in Homeric
 poetry can signal that something needs to be
 "walled off" - that there is indeed an implicit break, a
 breach of the barrier between past and present, and that
 this breach needs to be mended or emended. In Iliad XII 17-33, for example, the "Achaean Wall" is leveled by the
 direct intervention of the gods precisely because the epic
 needs to obliterate features of a landscape
 that must no longer be there in the "present" time of its
 narration. [55] In this retrospective context,
 where the vantage point becomes exceptionally the "present"
 time of epic narration, Homeric poetry refers exceptionally
 to heroes as hêmitheoi (XII 23); this exceptionally
 non-epic usage of "post-epic" or "present-day" terminology
 stands in sharp contrast with the conventional epic
 terminology, hêrôes. [56] In speaking here about the
 "present" time of epic, I must stress a central {80|81}
 point in my cumulative work on Homeric poetry: the "present"
 time of Homeric narration is an ever-shifting one. To the
 extent that the epic of Homeric poetry is an evolving
 medium, [57] its "present" time is relative, not
 absolute. [58] I will return to this point in my
 extended comments on the tenth in this list of ten selected
 formulations of Bakhtin.
4. "… the epic past is absolute and complete. … There are no
 loopholes in it through which we glimpse the future." [16]
Even in the Homeric Iliad, prophecies of a hero's
 future can allow glimpses of alternative traditions that
 make contact between the hero of the past and the political
 or cultural realities of the epic's "present." A case in
 point is the prophecy uttered by Poseidon to Aeneas in
 Iliad XX 306-308. [59] In general, explicit Homeric
 references to details about heroes in the past may indicate,
 implicitly, details about the political and cultural agenda
 connected with the cults of these heroes in the epic's
 "present." A case in point is the description of the
 territory ruled by the hero Ajax in the Catalogue of
 Ships, Iliad II 557-570: this 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" (as opposed to Hesiod F 204.44-51,
 which is relatively more advantageous to Megara). [60] 

5. "Everything incorporated into this past [of epic] was
 simultaneously incorporated into a condition of authentic
 essence and significance, but therefore also took on
 conclusiveness and finality, depriving itself, so
 to speak, of all rights and potential for a real
 continuation." [16; highlighting mine]
Such finality is merely implicit in epic as an unmarked
 category, while the continuation of the heroic world into
 the present is explicit in praise poetry as a marked
 category in opposition to epic.
6. "… the epic past is … isolated … from that eternal
 present of children and descendants in which the epic singer
 and his listeners are located, which figures as an event in
 their lives and becomes the epic performance." [17]
Bakhtin implies that the genealogies of the audience of epic
 in the present, reaching backward in time, are cut off from
 the genealogies of the heroes of epic in the past, which do
 not reach forward in time. Homeric poetry seems to be this
 kind of epic, where any access from the present to the
 heroic past is only implicit. Such access, however, becomes
 explicit in the praise poetry of Pindar's victory odes. This
 medium conventionally establishes contact between heroes or
 ancestors at one extreme and latter-day men as their proud
 descendants on the other extreme. Praise poetry blurs the
 distinction between immediate ancestors and remote ancestors
 or even heroes: when a victorious athlete is said to have
 done his ancestors proud, mention of his immediate ancestors
 is conventionally linked to a followup mention, in {81|82}
 elaborate narrative, of remote heroes of the past who may or
 may not be ancestors of the honored athlete. A case in point
 is Pindar Pythian 8.35-60: the glorification of the
 athletic victor, Aristomenes, is linked with the
 glorification of his patriliny, which in turn is immediately
 linked with the glorification of the hero Alkmaon who had
 done his father Amphiaraos proud in the sequel to the heroic
 narrative of the Seven against Thebes. [61]
7. "But the events, victors and heroes of 'high'
 contemporary reality are, as it were, appropriated by
 the past as they enter into these high genres (for
 example, Pindar's odes or the works of Simonides)." [18;
 highlighting mine]
It can also be said that the heroes of the past and their
 world are appropriated and even possessed by the present
 occasions of praise poetry composed by poets like Pindar and
 Simonides, commemorating contemporary events and victories.
 Besides the examples already cited from Pindar, I draw
 attention to a remarkable elegiac composition of Simonides,
 F 11 [West], commemorating the contemporary Hellenic victory
 at Plataea by linking the kleos of the victorious
 warriors (line 28) with the kleos (line 15),
 conferred by Homer himself (lines 15-17), of the heroes who
 destroyed the city of Troy.
8. "It is impossible to achieve greatness in one's own
 time." [18]
Praise poetry makes such an achievement possible. It
 connects contemporary reality with the distant world of
9. "Contemporaneity was reality of a 'lower' order in
 comparison with the epic past." [19]
The contemporaneity of praise poetry, as distinct from the
 contemporaneity of the novel, was reality of a 'higher'
 order. [62]
10. "The present … by its very nature … demands
 continuation, it moves into the future. … Therefore, when
 the present becomes the center of human orientation in time
 and in the world, time and world lose their completedness as
 a whole as well as in each of their parts. The temporal
 model of the world changes radically; it becomes a world
 where there is no first word (no ideal word), and the final
 word has not yet been spoken." [30]
The passage in question is Odyssey xiii 146-184. As we join the narrative, we find that the god
 Poseidon is very angry at the Phaeacians for providing Odysseus
 {82|83} with one of their ships to convey the hero back to his
 home in Ithaca. The god now plans to take revenge, and he asks
 Zeus to approve his plan, which has two parts: (1) to smash the
 ship as it sails back home to the Phaeacians and (2) to make a
 huge mountain "envelop" their city:
νῦν αὖ Φαιήκων ἐθέλω περικαλλέα νῆα
ἐκ πομπῆς ἀνιοῦσαν
 ἐν ἠεροηδέι πόντῳ
ῥαῖσαι, ἵν ἤδη σχῶνται, ἀπολλήξωσι δὲ
ἀνθρώπων, μέγα δέ σφιν
 ὄρος πόλει
So now I want to
 smash the very beautiful ship of
 the Phaeacians
when it comes back, in a misty
 crossing of the sea, from its conveying mission,
so that these people [= the
 Phaeacians] will hold off, at long last, and stop their
 practice of conveying
humans. And I want
 to make a huge mountain envelop [63] their city.
Odyssey xiii 149-152
 Zeus gives his approval, he modifies the terms of Poseidon's
 two-part plan for vengeance. In the case of the first part, as
 we are about to see, the Will of Zeus is not that the ship be
 smashed but only that it be turned into a rock at the very
 moment that it sails into the entrance to the harbor - a rock
 destined to be a famous landmark for all time to come. In the
 case of the second part of the sea god's plan, it seems that
 Zeus will indeed allow Poseidon to make a huge mountain
 "envelop" the city. Here is the precise wording of these two
 parts of the Will of Zeus, addressed as commands to Poseidon:
ὁππότε κεν δὴ πάντες ἐλαυνομένην
λαοὶ ἀπὸ πτόλιος, θεῖναι λίθον ἐγγύθι
νηὶ θοῇ ἴκελον, ἵνα θαυμάζωσιν
ἄνθρωποι, μέγα δέ σφιν
 ὄρος πόλει
When all the people of the city look
 out and see the ship sailing in,
turn it into a rock, just as it is
 about to reach land.
Make it look like a swift ship, so
 that people will look at it with wonder
- all of humanity will do so;
 and make the huge mountain envelop their city.
Odyssey xiii 155-158
I print
 the last verse here, xiii 158, as it is printed in most modern
 editions of Homer. [64] In this verse, the god Poseidon is
 commanded to seal off the Phaeacians forever within the confines
 of the epic past.
There is another version of this verse,
 however, adduced by the Alexandrian editor Aristophanes of
 Byzantium, which reads: {83|84}
ἄνθρωποι, μηδέ σφιν ὄρος πόλει ἀμφικαλύψαι
- all of humanity will do so; but do
 not make the mountain envelop their city.
Odyssey xiii 158 (variant) [65] 

This different version was
 disputed by the later Alexandrian editor Aristarchus of
 Samothrace: he preferred the version of xiii 158 that I printed
 earlier above, which is the one that survives in the medieval
 manuscript tradition. [66]
According to the version that survives only
 by way of Aristophanes, the future of the Phaeacians is not at
 all closed off. It remains open-ended, extending into the
 "present" when the epic is being narrated.
Two questions immediately come to mind.
 First, how could this different version fit the overall
 narrative of the Homeric Odyssey? Second, is the
 textual basis of this version "legitimate"? Addressing the first
 question first, I start by taking a close look at how the
 immediate narrative proceeds from here.
Complying with the reaction of Zeus to the
 original two-part plan of revenge, Poseidon proceeds to turn the
 returning ship into a rock (xiii 160-164). The first part of
 Poseidon's two-part plan has now been accomplished, although in
 modified form, in compliance with the Will of Zeus. [67] 
 The ship has been petrified at the
 approach to the harbor, instead of being "smashed" at
 midsea. [68]
At this midpoint in the ongoing narrative
 about the fate of the Phaeacians, we hear their reaction to the
 petrifaction of their ship. They are in shock: they cannot
 understand how this disaster could have happened to them (xiii
 165-169). But Alkinoos, their king, has comprehended what is
 still in the process of happening. He explains to the Phaeacians
 that he now understands a prophecy that his father Nausithoos
 had once told him: it must have been this present disaster,
 Alkinoos says, that his father had prophesied to him - along
 with that other disaster still waiting to be narrated in the
 Odyssey. Here is the precise wording of the explanation given by
 King Alkinoos:
φῆ ποτε Φαιήκων
 ἀνδρῶν περικαλλέα νῆα
ἐκ πομπῆς ἀνιοῦσαν ἐν ἠεροηδέι
ῥαισέμεναι, μέγα δ᾿ ἧμιν ὄρος πόλει
 ἀμφικαλύψειν [69]
He [my father] once said that he
 [Poseidon] will smash the very
 beautiful ship of the Phaeacian men
when it comes back, in a misty
 crossing of the sea, from its conveying mission,
and that he will
 make a huge mountain envelop our city.
Odyssey xiii 175-177
 audience of the Odyssey already knows this prophecy as
 recapitulated in xiii 173-177, because Alkinoos had already
 "quoted" it to Odysseus {84|85} at viii 565-569. [70] 
 At that earlier point in the narrative,
 however, Alkinoos had said something in addition, which he does
 not say now:
ὣς ἀγόρευ᾿ ὁ γέρων. τὰ
 δέ κεν θεὸς ἢ τελέσειεν,
ἤ κ᾿ ἀτέλεστ᾿ εἴη, ὥς οἱ φίλον
 ἔπλετο θυμῷ
That is what the old man said. And
 the god [Poseidon] could either bring these things to
or they could be left unfulfilled,
 however it was pleasing to his
Odyssey viii 570-571
 instead of "repeating" this part of the old man's prophecy,
 Alkinoos commands the Phaeacians to take immediate action:
ὣς ἀγόρευ᾿ ὁ γέρων. τὰ δὲ δὴ νῦν
 πάντα τελεῖται.
ἀλλ᾿ ἄγεθ᾿, ὥς ἂν ἐγὼ εἴπω, πειθώμεθα
That is what the old man said. And
 now you and I see that all these things are being
 brought to fulfillment. [71]
But come, let us all comply with exactly what I am about to say.
Odyssey xiii 178-179
 Alkinoos had first "quoted" the prophecy of his father at viii
 570-571, the "quotation" had left a loophole: Poseidon may or
 may not bring "these things" to fulfillment, as he wishes. But
 now at xiii 178-179 there is the greatest urgency, and Alkinoos
 exclaims hyperbolically that "all these things are being brought to fulfillment." The rhetorical point of this
 hyperbole is to motivate the Phaeacians to take immediate
 action. Even though the half-hopeful words of Alkinoos at viii
 570-571 are not repeated but are replaced by the increasingly
 desperate words of xiii 178-179, there is still a trace of hope
 - provided that the Phaeacians take immediate action by
 following the emergency orders of Alkinoos, which are formulated
 in the verses that immediately follow, xiii 180-182.
King Alkinoos orders the Phaeacians to do two
 things without delay: to resolve never again to engage in the
 otherworldly pompê 'conveying' (xiii 180) of mortals
 back to their real world and to offer a sacrifice of twelve
 bulls to Poseidon (xiii 180-182). [72] The Phaeacians must do these two things
 before the second of the two disasters should happen. The hope,
 Alkinoos says, is that Poseidon may still take pity and stop his
αἴ κ᾿ ἐλεήσῃ
μηδ᾿ ἥμιν
 περίμηκες ὄρος πόλει ἀμφικαλύψῃ
… in hopes that he [Poseidon] will
 take pity
and will not make the tall mountain
 envelop our city.
Odyssey xiii 182-183 {85|86}
 Phaeacians immediately proceed to make sacrifice to the sea god,
 supplicating him (xiii 184-187). At this sacrifice, we may
 presume that they do indeed resolve never again to engage in the
 otherworldly "conveying" of mortals back to their "real"
 world. [73] Such a resolution by the Phaeacians
 would of course cancel their own otherworldly status as
 mediators between the inner world of the narrative and the outer
 world of "reality" as implicit in the "present" time when the
 narration of epic is actually happening.
In terms of the mythological hermeneutics
 developed by J. Gordon Howie, the Phaeacians are hereby being
 shifted from the "Spatium Mythicum" to the "Spatium
 Historicum." [74] But questions remain. Are they being
 shifted merely in the sense that they have been removed, as of
 now, from the narrative? If the price of their survival is the
 loss of their status in the "Spatium Mythicum," will we ever get
 to see them again in the so-called "Spatium Historicum"?
But the most basic of all remaining questions
 is really this: what will happen to the Phaeacians according
 to the narrative? We cannot be completely certain. The
 Homeric narrative about the Phaeacians breaks off at
 Odyssey xiii 187, at the very moment when they are
 offering sacrifice and praying to Poseidon to take pity on them.
 As Peradotto points out, the narrative break takes place most
 abruptly, dramatically, and even exceptionally—at
 mid-verse. [75] In the first part of the verse at xiii
 187, the Phaeacians are last seen standing around the
 sacrificial altar; in the second part of the verse, Odysseus has
 just woken up in Ithaca. A new phase of the hero's experiences
 has just begun in the "real" world of Ithaca. [76] 

The narrative, then, ultimately leaves it
 open whether the Phaeacians will or will not be enveloped by the
 huge mountain. [77] Peradotto describes this uncertainty as
 a way for Homeric poetry "to avoid saying 'yes' to one system
 and 'no' to another, in a higher and more complicated system,
 the poem, that only precariously maintains them both." [78] 
 For Peradotto, the two competing
 systems that are subsumed "precariously" by the overriding
 Homeric system are, on the one hand, the element of fairy-tale
 or "Märchen" and, on the other, the element of "tragically
 oriented myth." [79]
In what follows, I offer a different
 explanation for whatever competing "systems" may be at work in
 this narrative. To anticipate my conclusions: Homeric poetry has
 left here an opening not only for two different outcomes but
 also for two different ways of thinking of an outcome.
I start my explanation by stressing again the
 importance of the loophole of viii 570-571, as formulated by
 Alkinoos: the god Poseidon may or may not bring his threat to
 fulfillment: he may do as he pleases. Moreover, we have already
 seen that even the first disaster did not quite happen in the
 way that the father of Alkinoos had prophesied - or the way that
 the god Poseidon had originally wanted it to happen before Zeus
 went ahead and modified the original terms in the process of
 formulating the eventual Will of Zeus.
Still, despite such tentatively hopeful
 signs, the plot of the Odyssey {86|87} accumulated many
 other signs that point toward the inevitability of disaster for
 the Phaeacians. [80] Can we really be sure, then, that there
 is still a way out? It all depends ultimately on whether Zeus
 had modified the terms for the second part of Poseidon's plan,
 not just for the first part. And that depends on whether we read
 the version featuring the variant μηδέ = mêde as
 adduced by Aristophanes instead of the variant μέγα δέ =
 mega de as preferred by Aristarchus and as
 transmitted by the medieval manuscripts.
Here I return to the second of my two initial
 questions about Odyssey xiii 155-158: is the textual basis of
 this different version featuring mêde really
 "legitimate"? We can now add a related question: if it is
 legitimate, then does that delegitimize the version featuring
 mega de?
For Erwin Cook, the outcome of the epic
 narrative depends on our making an actual choice between two
 variants, mega de vs. mêde at xiii 158, and he
 proceeds to choose μηδέ = mêde in line with his
 interpretation of the epic narrative's treatment of Poseidon's
 interactions with Zeus. [81] I agree with Cook's interpretation, but
 it leaves unanswered the question of legitimacy. How can we
 justify the textual transmission of the form mêde in
 this context? [82] Further, how can we justify the meaning
 of this variant in terms of Homeric poetry?
The actual need to choose one or the other
 variant depends on the way we look at Homeric poetry. If this
 poetry is merely a static text, then we are indeed forced to
 make a choice. If, however, we view Homeric poetry as a living
 system - an oral tradition that evolves ultimately into the
 textual tradition inherited by the Alexandrian editors - then we
 do not have to choose whenever we see a variation. Rather, as I
 will now go on to argue, the choices were already being made by
 Homeric poetry itself, which could opt for different variants in
 different phases of its own evolution.
My reasoning here derives from an overall
 "evolutionary model" that I have worked out as a general way to
 account for the making of Homeric poetry. [83] In terms of this model, as I now plan
 to argue, the living and evolving oral tradition of Homeric
 poetry itself allowed for a choice either to seal off its own
 past from the present time of narration or to reach into this
 present time and thereby make its presence fully manifest.
According to the narrative option linked with
 the first of our two variants from Odyssey xiii 158,
 mega de, the outlook is hopeless for the
 Phaeacians, since Poseidon's plan to seal off the city of the
 Phaeacians has been restated by Zeus and is therefore tantamount
 to the Will of Zeus, which the Homeric tradition conventionally
 equates with the way things ultimately turn out in epic
 narrative, as in Iliad I 5. [84] At the beginning of the
 Odyssey, however, Zeus himself undercuts the
 equation of epic plot with the Will of Zeus (i 32-34). [85] 
 That is, there are {87|88} differences
 in shades of meaning between the Iliadic and the Odyssean
 perspectives on the Will of Zeus as the plot of epic. [86] 

According to the narrative option linked with
 the second variant mêde, the outlook is still hopeful.
 After all, at an earlier point in the narrative, xiii 144-145,
 we can see a way out when Zeus tells Poseidon to exact any
 punishment he pleases "if any human dishonors you not at all"
 (ἀνδρῶν δ᾿ εἴ περ τίς σε . . . | οὔ τι τίει xiii 143-144). The
 context is this: Poseidon has been angrily questioning Zeus,
 calling on him to explain the Will of Zeus (Διὸς δ᾿ ἐξείρετο
 βουλήν xiii 127) - that is, to explain the overall plot of the
 narrative - now that the Phaeacians have conveyed Odysseus back
 home to Ithaca. How can I be honored among the gods, Poseidon
 plaintively asks Zeus, "when the Phaeacians do not honor me at
 all?" (ὅτε με βροτοὶ οὔ τι τίουσι | Φαίηκες xiii 129-130). But
 then, as we have already seen, the story goes on to say that the
 Phaeacians will indeed initiate a remedy after the first
 disaster by proceeding to honor Poseidon with sacrifice in order
 to avert the second disaster.
The narrative option that I link with the
 variant mêde, according to which the Phaeacians are to
 be spared the second disaster of an all-enveloping mountain,
 depends on whether this variant as adduced by Aristophanes of
 Byzantium in place of mega de at xiii 183 is a genuine
 formulaic variant or only a textual variant. If it is the latter, then mêde may be just an
 editorial conjecture. [87] That possibility would severely reduce
 the chances for arguing that mêde is a genuine
 alternative to mega de. In what follows, however, I
 will argue against that possibility on several levels.
From an analysis of the formulaic system in
 which mêde is embedded, this form can be justified as a
 functioning element in that system, just as the form mega
 de is a functioning element: in other words,
 mêde and mega de can be considered
 compositional alternatives in the formulaic system of Homeric
 diction. [88]
Moreover, there is immediate contextual as
 well as formulaic evidence to support the argument that
 mêde is a functioning compositional variant in the
 formulaic system. Let us consider the wording of Zeus in his
 answer to Poseidon's angry questioning:
ἔρξον ὅπως ἐθέλεις καί τοι φίλον
 ἔπλετο θυμῷ
Do as you wish and as was pleasing to
 your heart.
Odyssey xiii 145
This open-ended wording
 of Zeus matches formulaically the wording of Alkinoos, when he
 had originally "quoted" the prophecy of his father:
ὣς ἀγόρευ᾿ ὁ γέρων. τὰ δέ κεν θεὸς ἢ
ἤ κ᾿ ἀτέλεστ᾿ εἴη, ὥς οἱ φίλον ἔπλετο θυμῷ
That is what the old man said. And
 the god [Poseidon] could either bring these things to

or they could be left unfulfilled,
 however it was pleasing to his heart.
Odyssey viii 570-571
 formulation of Zeus, then, in leaving it still undecided whether
 or not the Phaeacians are to be "enveloped," can be used as
 evidence to argue that mêde is indeed a genuine
 compositional alternative to mega de.
As for the possibility that mêde is
 an emendation based on an editorial conjecture, my own
 cumulative work on Homeric variants as adduced by the three
 great Alexandrian editors of Homer (Zenodotus, Aristophanes, and
 Aristarchus) leaves me skeptical, since I find that these
 editors normally do not make emendations without manuscript
 evidence. [89]
In making the specific argument that both
 variants mega de and mêde are genuine
 compositional alternatives, I return to my general argument that
 Homeric poetry is not a static text but a slowly evolving
 system. [90] In terms of this general argument, the
 variant mega de produces a narrative closure for the
 Phaeacians: their fate is sealed. The variant mêde,
 however, produces an outcome that is open-ended. [91] 

These two variants, I contend, reflect
 different phases in the evolution of Homeric poetry. Let us
 begin with the variant mega de, the context of which
 can be linked with a relatively more Panhellenic phase of
 epic. [92] I have defined this phase elsewhere as
 one that "concentrates on traditions that tend to be common to
 most locales and peculiar to none." [93] The Panhellenic phases of epic make
 contact with the "present" time of narration by shading over any
 "local color" that might distract from the widest possible range
 of ways to visualize this "present." [94] A Panhellenic version, then, will tend
 to universalize the concerns of the present.
But there are also other, less Panhellenic,
 ways for epic to make contact with the "present" time in which
 narration happens: the "local color" can be highlighted, though
 only at the cost of narrowing the range of ways to visualize
 this "present." The context of the variant mêde can be
 linked with such a relatively less Panhellenic phase of epic.
 This variant makes contact with the epic "present" in a less
 universalized and more localized way. The focus of localization
 is historical Corcyra, modern-day Corfu.
The fact is, the Corcyraeans of the Classical
 period claimed to be the descendants of the Phaeacians, as we
 know from a remark of Thucydides (1.25.4); from another remark
 of his, we know also that they worshipped King Alkinoos as their
 local cult hero (3.70.4). [95] As Howie surmises, "the value of the
 Phaeacians for the Corcyraeans was that they gave them a stake
 in the mythical past independent of their mother-city [Corinth],
 which was famous as a centre of the worship of the sea-god
 [Poseidon] and as site of the panhellenic Isthmian Games in the
 god's honour." [96]
In Odyssey xiii 155-158, we hear how
 the Phaeacians will one day look out at {89|90} their harbor and
 see their returning ship suddenly turn into a rock, and we hear
 also how that fabulous petrified ship will continue to be a most
 wondrous sight for future generations of humanity to see and to
 keep on seeing for all time to come. These epic verses of
 Homeric poetry, one commentator surmises, may be providing an
 aetiology "for the fact that the rock which rises from the sea
 just outside the harbour of Corfu was taken to be 'Odysseus'
 ship'." [97] There are references to this
 "real-life" rock in Pliny (Natural History) 4.53 and
 Eustathius (Commentary on Odyssey vol. II p. 44 line
 27), and to this day the "petrified ship" remains a most
 celebrated tourist attraction for visitors to Corfu. [98] 
 But the essential point is, the
 reference to this rock is already there in the Odyssey - that
 is, in a version of the Odyssey that says mêde instead
 of mega de at xiii 158.
The identity of the Corcyraeans as
 descendants of the Phaeacians depends on the Will of Zeus as he
 formulates it in Odyssey xiii 155-158, and it depends especially
 on the variant mêde of xiii 158, which yields an
 open-ended narrative that reaches directly into the "present" of
 the Classical period and beyond.
As a political and cultural fact of life, the
 self-identification of the Corcyraeans with the Phaeacians has
 been dated as early as the third quarter of the eighth century
 BCE. [99] The variant represented by
 mêde at xiii 158 may be just as early, and in fact
 it may be the vehicle for expressing just such a political and
 cultural fact of life. This is not to say that the other variant
 represented by mega de at xiii 158 may not be just as
 early. It is only to say that both variants were still available
 to the Homeric tradition of epic as it evolved during the
 pre-Classical period. In such an early period, the affirming -
 or the denying - of a claim of descent from the Phaeacians was
 essential not just poetically but also politically and
 culturally. [100] It really mattered then, and it
 continued to matter well into the Classical period of the fifth
 century and beyond, as we have seen from the remark of
 Thucydides (1.25.4, 3.70.4) about the Corcyraeans' claim to be
 descended from the Phaeacians of King Alkinoos, whom they
 worshipped as their local hero. [101]
In the Hellenistic period of the Alexandrian
 editors of Homer, by contrast, the question of choosing mega
 de or mêde would have mattered purely from a
 poetical rather than a political or cultural point of view. The
 question of the Corcyraeans' claims of descent from the
 Phaeacians would not be a major concern any more, at least not
 politically. But it would still really matter in another way:
 was the petrified ship of the Phaeacians a figment of the poetic
 imagination, walled off in the "Spatium Mythicum" of the epic
 past, or is it the same thing as the real-life rock at the
 entrance to the harbor of Corcyra, accessible to all humanity in
 the "Spatium Historicum" of the contemporary Hellenic world? The
 disagreement between Aristarchus and Aristophanes over the
 choice of mega de or mêde respectively must
 have centered on such questions. One way, we see a beautiful
 snapshot from the enchanted imaginary world of the epic past.
 The other way, we see a comparably beautiful vista in the
 enchanting touristic world of the non-epic present, still
 anchored in the {90|91} permanence of the epic past. Either way,
 petrified ship or scenic rock, what we see is a beloved cultural
 landmark of Hellenism.
All this is not to say that we must
 ultimately choose between these two versions of seeing things
 Homeric. It is only to say that both variants were still
 available to the Homeric tradition of epic as it evolved into
 the Classical period and beyond. And it is to ponder the power
 of epic - and of the Classical - either to close down or to open
 up its pathways to the present. The fate of the Phaeacians in
 conveying the heroic past to the present depends on that power
 of Homeric dimensions.

Works Cited

Bakhtin, M. [1975]/1981. "Epic and Novel." In
 The Dialogic Imagination. Ed. and introduction
 by M. Holquist. Austin.
———. [1965]/1984a. Rabelais and his World.
———. [1963]/1984b. Problems of
 Dostoevsky’s Poetics. Minneapolis.
Bakker, E.J. 1997. Poetry in Speech:
 Orality and Homeric Discourse. Ithaca.
Branham, R. B. 1995. “Inventing the Novel.”
 Bakhtin in Contexts: Across the Disciplines.
 Ed. Amy Mandelker. Evanston, IL. 79-87.
Braswell, B. K. 1998. A Commentary on
 Pindar Nemean Nine. Berlin.
Brelich, A. 1958. Gli eroi greci, un
 problema storico-religioso. Rome.
———. 1961. Guerre, agoni e culti nella
 Grecia arcaica. Bonn.
Carrière, J. 1979. Le carnaval et la
 politique. Paris.
Cook, E. 1992. “Ferrymen of Elysium and the
 Homeric Phaeacians.” The Journal of Indo-European
 Studies 20:239-267.
———. 1995. The Odyssey in Athens: Myths of
 Cultural Origins. Ithaca.
Dieterich, A. 1897. Pulcinella:
 pompejanische Wandbilder und römische Satyrspiele.
 Leipzig. Republished 1982, Aalen.
Dindorf, W., ed. 1855. Scholia graeca in
 Homeri Odysseam ex codicibus aucta et emendata.
Edwards, A. T. 1993. “Historicizing the
 popular grotesque: Bakhtin’s Rabelais and Attic Old
 Comedy.” In Theater and Society in the Classical
 World. Ed. R. S. Scodel. Ann Arbor. 89–117.
Finkelberg, M. 1988. “Ajax’s Entry in the
 Hesiodic Catalogue of Women.” The Classical
 Quarterly, n.s., 38:31-41.
Friedrich, R. 1989. “Zeus and the Phaeacians:
 Odyssey 13.158.” The American Journal of
 Philology 110:395-399.
Gavrilov, A. 1995. “Russian Classical
 Scholarship.” In The Classics in East Europe: Essays on
 the Survival of a Humanistic Tradition. Ed. V. Bers
 and G. Nagy. Worcester, MA. 61–81.
Hoekstra, A. 1989. “Books XIII-XVI:
 Introduction and Commentary.” In A Commentary on Homer’s
 Odyssey. Vol. II: Books IX–XVI, by A.
 Heubeck and A. Hoekstra. Oxford.
Holquist, M. 1981. Introduction to Bakhtin
Hornblower, S. 1991. A Commentary on
 Thucydides. Vol. I: Books I-III.
Howie, J.G. 1989. “The Phaeacians in the
 Odyssey: Fable and Territorial Claim.”
 Shadow 6:23-34.
Jacopin, P. 1988. “Anthropological Dialectics:
 Yukuna Ritual as Defensive Strategy.” Schweizerische
 Amerikanisten-Gesellschaft 52:35-46.
Kuryłowicz, J. 1945-1949. “La nature des
 procès dits ‘analogiques’.” Acta Linguistica 5:15–37.
Martin, R. 1993. “Telemachus and the Last Hero
 Song.” In “Essays on Homeric Epic,” ed. H. Roisman and J.
 Roisman, special issue, Colby Quarterly 29(3):222–240.
Muellner, L. 1976. The Meaning of Homeric
 Euchomai Through its Formulas. Innsbruck.
Nagy, G. 1990. Pindar’s Homer: The Lyric
 Possession of an Epic Past. Baltimore.
———. 1996a. Poetry as Performance.
———. 1996b. Homeric Questions.
———. 1997. “The Shield of Achilles: Ends of
 the Iliad and Beginnings of the Polis.” In New Light on
 a Dark Age: Exploring the Culture of Geometric
 Greece. Ed. S. Langdon. Columbia, MO.
———. 1998. “Homer as ‘Text’ and the Poetics of
 Cross-Reference.” In Verschriftung und
 Verschriftlichung: Aspekte des Medienwechsels in
 verschiedenen Kulturen und Epochen. Eds. C. Ehler
 and U. Schaefer. ScriptOralia 94. Tübingen. 78-87.
———. 1999. The Best of the Achaeans:
 Concepts of the Hero in Archaic Greek Poetry.
Peradotto, J. 1990. Man in the Middle
 Voice. Princeton.
Reckford, K. 1987. Aristophanes’ Old and
 New Comedy. Chapel Hill.
Reynolds, D. F. 1995. Heroic Poets, Poetic
 Heroes: The Ethnography of Performance in an Arabic Oral
 Epic Tradition. Ithaca.
Rösler, W. 1986. “Michail Bachtin und die
 Karnevalskultur im antiken Griechenland.” Quaderni
 Urbinati di Cultura Classica 52:25–44.
Saussure, F. 1916. Cours de linguistique
 générale. Wiesbaden.
Seaford, R. 1994. Reciprocity and Ritual:
 Homer and Tragedy in the Developing City-State.
Segal, C. 1994. Singers, Heroes, and Gods
 in the Odyssey. Ithaca.
Thalmann, W. G. 1998. The Swineherd and
 the Bow: Representations of Class in the Odyssey.
Todorov, T. 1984. Mikhail Bakhtin: The
 Dialogical Principle. Manchester.
Waugh, L. R. 1982. “Marked and Unmarked: A
 Choice between Unequals in Semiotic Structure.”
 Semiotica 38(3-4):299-318.


[ back ] 1. Bakhtin [1975]/1981, ch.1, pp. 3-40.
 Hereafter abbreviated as EN.
[ back ] 2. Bakhtin [1965]/1984a. Hereafter abbreviated
 as RW.
[ back ] 3. 
 RW 410.
[ back ] 4. 
 RW 410.
[ back ] 5. 
 RW 410.
[ back ] 6. Nagy 1990, hereafter abbreviated as
[ back ] 7. 
 PH 5-6; see Kuryłowicz 1945-1949.
[ back ] 8. 
 PH 5-6, following Waugh 1982. In PH 6, I
 offer an illustration of the opposition unmarked/"order" vs.
 marked "disorder" by examining the mechanics and esthetics
 of the choriambic dimeter.
[ back ] 9. 
 PH 6n18.
[ back ] 10. See Todorov 1984:85-91.
[ back ] 11. Peradotto 1990:53n13.
[ back ] 12. Bakhtin [1963]/1984b:109: "one could say
 that the novelistic genre has three fundamental roots: the
 epic, the rhetorical, and the
 carnivalistic (with, of course, many
 transitional forms in between)." Cf. Todorov 1984:90.
[ back ] 13. Saussure 1916:117.
[ back ] 14. Saussure 1916:117: "Est synchronique tout
 ce qui se rapporte à l’aspect statique de notre science,
 diachronique tout ce qui a trait aux évolutions. De même
 synchronie et diachronie désigneront
 respectivement un état de langue et une phase
[ back ] 15. 
 PH 21n18, following Jacopin 1988:35-36, who adds:
 "Both synchrony and diachrony are abstractions extrapolated
 from a model of reality."
[ back ] 16. On the complementarity of the
 Iliad and Odyssey, see paragraphs
 15-18 in the new Preface of Nagy 1999.
[ back ] 17. Holquist 1981:xxii.
[ back ] 18. Gavrilov 1995:61.
[ back ] 19. For an analysis of some of the reasons for
 Bakhtin’s avoidance of Athenian Old Comedy (especially as
 exemplified by Aristophanes) as a focus of interest, see
 Edwards 1993, with important refinements on the earlier work
 of Carrière 1979, Rösler 1986, and Reckford 1987.
[ back ] 20. 
 RW 107.
[ back ] 21. 
 RW 107; cf. PH 398n91.
[ back ] 22. 
 RW 65.
[ back ] 23. 
 RW 65.
[ back ] 24. 
 PH 394, 397-401.
[ back ] 25. 
 RW 121.
[ back ] 26. Dieterich 1897.
[ back ] 27. 
 RW 121.
[ back ] 28. 
 PH 389-401.
[ back ] 29. 
 RW 52.
[ back ] 30. 
 RW 114.
[ back ] 31. 
 RW 48.
[ back ] 32. Nagy 1999:316.
[ back ] 33. 
 RW 50.
[ back ] 34. 
 PH ch.7.
[ back ] 35. Nagy 1999:235.
[ back ] 36. 
 PH 150.
[ back ] 37. With reference to kleos, the
 metaphor of anth- in the sense "blooming" or
 "blossoming" like a flower (cf. also Pindar Pythian 1.66) seems to be the antithesis of "wilting," as conveyed
 by phthi-: cf. PH 225 (also Nagy
 1999:175-189; pace Braswell 1998:122).
[ back ] 38. Brelich 1961; cf. PH 137, 139,
 141, 152.
[ back ] 39. See PH 152, with special reference
 to the poetic usage of words like ponos,
 kamatos, and athlos - all denoting the
 idea of "ordeal" in contexts of both war and
[ back ] 40. Braswell 1998:122.
[ back ] 41. Braswell 1998:51.
[ back ] 42. On the kleos of athletic victory,
 see Pindar Nemean 7.63, Isthmian 5.8, etc.
 As for kleos at line 29 of Pindar Isthmian 7, lines 31-36 go on to celebrate the martial achievements
 of the victorious athlete’s maternal uncle (who is compared
 to Hektor as well as Meleagros and Amphiaraos). Both here
 and at Nemean 9.39-42, the kleos of
 martial victory complements what I contend is an implicit
 overriding kleos of athletic victory. I do not
 think that it is "misleading" (pace Braswell
 1998:122) to use Nemean 9.39-42 as an example of
 Pindar’s equation of the kleos of victorious
 athletes of the present with that of heroes of the
[ back ] 43. 
 PH 150-152.
[ back ] 44. Brelich 1958:94-95; cf. PH 118-120
 and ch.4 in general; also Thalmann 1998:165-166.
[ back ] 45. Brelich 1958:94-95.
[ back ] 46. Braswell 1998:122.
[ back ] 47. 
 PH 192-193, especially with reference to Pindar
 Nemean 8.50-51. See also the formulation in
 PH ch.14, especially p. 437, quoted
[ back ] 48. Nagy 1999:253-254.
[ back ] 49. See n6 above.
[ back ] 50. 
 PH 437.
[ back ] 51. 
 PH 191-192.
[ back ] 52. On the distinction between diachronic and
 historical perspectives, see above.
[ back ] 53. On the possibility that Horace had in mind
 the passage of Pindar Nemean 9.39-42 as quoted
 above, see Braswell 1998:122.
[ back ] 54. Reynolds 1995, with reference to Arab epic
 traditions. See especially his p. 207, quoting Martin
 1989:xiv: "My central conclusion is that the Iliad takes shape as a poetic composition in precisely the same
 'speaking culture' that we see foregrounded in the stylized
 words of the poem's heroic speakers, especially those
 speeches designated as muthos, a word I redefine as
 'authoritative speech act.'"
[ back ] 55. Nagy 1999:160.
[ back ] 56. Nagy 1999:160: "Whereas hêrôes is
 the appropriate word in epic, hêmitheoi is more
 appropriate to a style of expression that looks beyond
[ back ] 57. 
 PH 23-24, 53-58, 197-198; Nagy 1999:xiv-xv =
 paragraphs 22-25 of the new Preface.
[ back ] 58. 
 PH 53, 60, 70.
[ back ] 59. Nagy 1999 ch.15, especially pp. 267-269.
 The name of Aeneas, derived from ainos, is
 pertinent: see pp. 274-275.
[ back ] 60. 
 PH 73n106, following Finkelberg 1988.39-40.
[ back ] 61. 
 PH 194-195, 203-204.
[ back ] 62. For a modification of Bakhtin's formulation
 of the novel, with reference to the ancient Greek novel, see
 Branham 1995, especially pp. 84-86.
[ back ] 63. I choose this translation of ἀμφικαλύψαι in
 light of the observations of Merry 1887 on xiii 152:
 "Poseidon does not propose to bury the city, but to shut it
 off from the use of its two harbours by some great mountain
 mass." See also Peradotto 1990:78n18.
[ back ] 64. For example, the editions of Allen 1919 and
 van Thiel 1991.
[ back ] 65. This variant, adduced by Aristophanes of
 Byzantium, is reported by the scholia (Ἀριστοφάνης δὲ
 γράφει, μὴ δέ σφιν: H at xiii 152, evidently with reference
 to xiii 158). The scholia go on to say that Aristarchus
 opposed (ἀντιλέγει) this reading in his hupomnêmata or commentaries (evidently preferring μέγα δέ σφιν over μηδέ
 σφιν). See Dindorf 1855:566; cf. Hoekstra 1989:174 and
 Friedrich 1989:396n2. Conceivably, Aristophanes adduced
 πόλιν ἀμφικαλύψαι instead of πόλει ἀμφικαλύψαι. At xiii 158
 and 177 there is variation in the medieval manuscripts:
 either πόλιν ἀμφικαλύψ- or πόλει ἀμφικαλύψ-.
[ back ] 66. See the previous note.
[ back ] 67. On the systematic subordination of the Will
 of Poseidon to the Will of Zeus in the Odyssey, see
 Segal 1994:210; see also his analysis, p. 219, of Zeus as
 "the most detached of all the gods." For important further
 elaboration on these themes, see Cook 1995:123-127.
[ back ] 68. The formulaic language of epic is quite
 precise here in making a distinction between a "smashing" of
 the ship at midsea and a petrifaction of the ship at the
 approach to the harbor; see Cook 1995:124.
[ back ] 69. There is variation in the medieval
 manuscripts: either ἀμφικαλύψαι or ἀμφικαλύψειν.
[ back ] 70. The textual transmission of viii 565-569
 and xiii 173-177 leaves the two passages matching almost
 exactly, word for word. There is some degree of
 non-matching, though: thus the ship is εὐεργέα 'well-built'
 in most manuscripts at viii 567 vs. περικαλλέα 'very
 beautiful' in most manuscripts at xiii 175, while the
 mutually alternative forms are attested in a minority of
 manuscripts at both places. In terms of oral poetics, such
 variation may be justified even where the "quoting" of a
 character's words happens to be a narrative requirement of
 the composition, as it is here.
[ back ] 71. On the "evidentiary" function of δή, see
 Bakker 1997:75-76, 78-79. This particle δή is used by a
 speaker when he or she "assumes that the listeners are
 willing to see the evidence produced, so that conducting the
 discourse becomes an activity aimed at shared seeing, a
 being together in the situation created by the speaker's
 phrasing" (Bakker p. 76). In the present context, I
 translate this "evidentiary" function of δή by adding "And
 now you and I see that" to "all these things are being
 brought to fulfillment."
[ back ] 72. For an incisive analysis of the
 otherworldly aspects of the Phaeacaians' activity of
 pompê 'conveying' (xiii 180) by way of their
 supernatural ships, see Cook 1992, especially pp. 240-241,
 245. See also in general the valuable interpretations of
 Segal 1994:12-64.
[ back ] 73. Cf. Cook 1995:124n36, who comments:
 "Poseidon's essential aim has been achieved with his
 transformation of the ship: the Phaiakes cease to offer
 escort to mortals." On the hermeneutics of the "reality" of
 Ithaca as opposed to the "world apart" that is Phaeacia, see
 Segal 1994:12-25.
[ back ] 74. Howie 1989:25 and 28. His model of the
 "Spatium Mythicum" is comparable to Peradotto's model of
 "Märchen" (1990:82-83), on which I have more to say
[ back ] 75. Peradotto 1990:81.
[ back ] 76. On the return of Odysseus to Ithaca as the
 notional end of the heroic age and the notional beginning of
 the "present" time of Homeric composition, see Martin
[ back ] 77. Peradotto 1990:80-81.
[ back ] 78. Peradotto 1990:83.
[ back ] 79. Peradotto 1990:83.
[ back ] 80. Howie 1989:31 speaks of "the inevitability
 of the second phase of the prophecy."
[ back ] 81. Cook 1995:124n36; also Friedrich
[ back ] 82. Cf. Friedrich 1989:398-399: "Aristophanes'
 reading has against it the whole weight of the [medieval]
 manuscript tradition, and Aristarchus' authority to boot."
 He leaves it open whether Aristophanes conjectured
 mêde or whether he found it attested in the
 ancient manuscript tradition (p. 396). Still, he argues
 strongly for the upgrading of mêde "from the
 apparatus to the text" (p. 399).
[ back ] 83. Nagy 1996a:109-114 and 1996b:29-112; cf.
 Seaford 1994:144-154. For an incisive overview, see Thalmann
[ back ] 84. 
 PH 238, with further bibliography.
[ back ] 85. Extensive commentary, with bibliography, in
 PH 241-242.
[ back ] 86. 
 PH 241-242.
[ back ] 87. Peradotto 1990:79 argues that mêde at xiii 183 is just that, an editorial conjecture:
 "Aristophanes, scandalized by a pusillanimous Zeus who would
 make himself accessory to the destruction of the Phaeacians,
 alters μέγα δέ in line 158 to μηδέ." As Peradotto points out
 (ibid.), "With few exceptions modern critics generally tend
 to reflect Aristophanes's tender-mindedness." At pp. 79-80,
 he quotes some interesting examples.
[ back ] 88. As Leonard Muellner points out in a message
 written 3/10/1998 to me and to Chad E. Turner, the variant
 μηδέ at xiii 177 is syntactically and formulaically parallel
 to the μηδέ of xiii 183. In a message written 2/3/1998,
 Turner had pointed out to me that the metrical placement of
 the variant μηδέ at xiii 198 is singular (although there are
 cases where this word straddles the last syllable of a
 spondee and the first syllable of a dactyl in the third and
 fourth feet, he finds no other cases in the second and third
 feet). But the formulaic system is capable of generating
 rare forms and combinations. For a striking example, we may
 compare the singular attestation of μηδέν at Iliad XVIII 500: here is a word that is found this one and only
 time in the Iliad and the Odyssey put
 together, and yet it can be shown to be formulaic.
 See Muellner 1976:101-102, 106.
[ back ] 89. Nagy 1996a:107-152. This finding is a
 source of ongoing debate, some of which I survey in Nagy
[ back ] 90. Nagy 1996b:29-112, where I stress that the
 pace of evolution in Homeric poetry as a system slows down
 markedly after the eighth century BCE.
[ back ] 91. We may compare the open-endedness conveyed
 by the word μηδέν = mêden in Iliad XVIII
 500, centering on the moral dilemma of an aggrieved man in a
 litigation that is pictured on the Shield of Achilles. The
 unnamed man in the picture is locked into a stance of
 eternal refusal, extending indefinitely into the future: see
 again Muellner 1976:101-102, 106. With reference to this
 picture, see also Nagy 1997:195: "the Iliad need
 not end where the linear narrative ends, to the extent that
 the pictures on the Shield of Achilles leave an opening into
 a virtual present, thus making the intent of the
 Iliad open-ended."
[ back ] 92. On the relativity of Panhellenism (despite
 the absolutist implications of the term) as a cultural
 impulse, see PH 53.
[ back ] 93. 
 PH 54, with further discussion of Panhellenic
[ back ] 94. Further discussion in PH 57, where
 I describe Panhellenism as "a hermeneutic model for
 explaining how the myth-making mind can become critical of
 variants in myth."
[ back ] 95. Hornblower 1991:70 and 469; see also Howie
[ back ] 96. Howie 1989:27.
[ back ] 97. Hoekstra 1989:174.
[ back ] 98. Cf. Howie 1989:32, Hoekstra 1989:174. Here
 is the wording of Pliny 4.53: … a Phalacro, Corcyrae
 promuntorio, scopulus in quem mutatam Ulixis navem a
 simili specie fabula est. Note too Pliny 4.52:
 Corcyra … Homero dicta Scheria et
[ back ] 99. Hoekstra 1989:174: "A possible terminus
 post quem is the third quarter of the eighth
 century when Eretrians, soon followed by Corinthians,
 settled there." For a critical survey of testimonia, see
 Howie 1989:29.
[ back ] 100. On the impact of prevailing political and
 cultural forces on the evolution of Homeric poetry before
 the Classical period, see again my comments on the fifth of
 Bakhtin's ten selected formulations.
[ back ] 101. For parallel claims in the pre-Classical
 period, see PH 153-155, especially with reference
 to (1) the Peisistratidai of Athens, who claimed descent
 from Peisistratos, son of Homeric Nestor; (2) the
 Penthilidai of Mytilene in Lesbos, claiming descent from
 Penthilos, son of Orestes; (3) the Neleidai of Miletus,
 claiming descent from Neleus, father of Nestor.