Porter, Andrew. 2019. Agamemnon, the Pathetic Despot: Reading Characterization in Homer. Hellenic Studies Series 78. Washington, DC: Center for Hellenic Studies. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_PorterA.Agamemnon_the_Pathetic_Despot.2019.
1. Introduction
1.1 Characterizing Agamemnon
Homer [1] and his audience knew Agamemnon as the primary leader of well over 1,000 Achaian ships that sailed against Troy. [2] He led an intense ten-year struggle, only to return home and die a miserable death at his wife’s hands. His presence in person or name impacts the poetic narratives of the Iliad and Odyssey in ways exceeded by few other epic personalities, and yet his full character has not been adequately considered. How are we to understand what Agamemnon was like as a character for Homer’s audience? More fundamentally, how are we to approach the topic of characterization itself since the discoveries of Parry and Lord and their successors?
The central goal of this book is to explain how characterization works in Homer and to focus on Agamemnon as a test case. To accomplish this goal, we will consider the characterization of Agamemnon in the Iliad and Odyssey from an oral traditional point of view. We turn especially to one branch of oral poetics, “traditional referentiality,” to consider the topic of characterization in general and the character of Agamemnon in particular. The goal is to discover and interpret traditional cues within the oral tradition Homer was accessing and creatively employing. We will capture Agamemnon’s personality by a careful reading of the epic narratives, where the traits and implications of his personality can be heard against known back stories through the resonance of words, themes, and type scenes, and by considering various other poetic devices. [3]
1.2 Epic Characterization
While Agamemnon as a whole character has proved to be of somewhat less interest to modern scholars than other epic personalities such as Achilles, Helen, and Odysseus, [4] particular aspects of his life or his political position have been considered in some depth. [5] Further, a limited amount of work has been published solely devoted to a consideration of epic characterization itself since the initial findings of Parry and Lord. Most scholars have not attempted to consider what oral traditional research might mean for the study of character. [6] There are a few exceptions in general studies of characterization. These include Martin’s (1993) foundational discussion of particular points that undergird epic characterization, but also Dué’s (2002) detailed study of Briseïs. [7] While not a study of characterization per se, Dué shows, though her application of oral poetics, that epic characterization is very much dependent on a character’s shared participation in common themes from past stories. It is a point my own study will develop in detail. Further, Minchin (2011a) outlines a convincing theory of how an epic singer organizes, stores, and accesses characterization (for “first-rank heroes”). She uses an appealing model from cognitive science and argues not only for shared, but also individual “themes” (what I call “traits”). She finds that it is the prevalence of “particular themes (that is, individual themes)” that make a character distinctly identifiable and memorable. [8] My own findings parallel hers in certain respects, particularly in her conclusion that “the poet has stored in memory” what is “appropriate” for each character. The present study will outline and then read traditional cues that access this stored knowledge, what I refer to as “tradition.”
We might, however, ask why, since the findings of Parry and Lord, there has been such limited research focused on epic characterization in so large a field as Homeric Studies. It may be that many feel that the question of how to approach epic characterization has changed from a literary to a quasi-folktale style inquiry. The question may now be: “How does a character fit a character type or story pattern?” The acceptance of the generally formulaic and oral nature of the Homeric epics can make their narratives appear, at least at first glance, “stereotypical.” [9] One could wonder whether the typical portrait of a character leaves any room for individual character depth. Further, to ask what the character of Agamemnon was like for the singer, audience, and even “the tradition,” may seem like the wrong question. [10] Yet, it is just this question that I am asking here and in Chapter 2: “Is individuality possible, or must all characters speak and act alike?” [11] It is a question initially raised in many ways by Notopoulos, who had urged that what is needed “is the study of the relationship of the formulaic technique to human characterization … the extent to which the oral technique can go beyond the typological ‘Man’ of Geometric vases” (Notopoulos 1964a:65). [12] In short, what does it mean to speak of Homeric characterization? Related to this is the question of how we can “read” character. We must answer these questions before we consider what sort of character Agamemnon himself was for Homer and his audience, in Chapters 3 and 4.
1.3 Reading Characterization Traditionally
The work of one pioneering scholar is of particular interest for our present study. John Foley developed a fuller methodology for reading traditional background for Homeric narrative by showing how tradition affects the way Homeric poetry is heard. [13] Foley’s approach of finding meaning through oral “patterning and context” (Kelly 2007a:5) was indebted to previous work carried out by a number of scholars. [14] Foley’s “traditional referentiality” argues that the narrative instant carries the most meaning when heard against the larger backdrop of the oral traditional register. The title of Foley’s 1991 book, Immanent Art, encapsulates the argument laid out in greater detail inside and in subsequent work. In short, the meaning that a formula, type scene, or story pattern carried with it was “immanent” when heard. Immanence is attained through “metonymy,” which means that a singular instance stands pars pro toto (Foley 1991:7–9). [15] As he described it in this and subsequent works, the singer’s system of tradition-based expressivity entails the “invoking of a context that is enormously larger and more echoic than the text of the work itself” (Foley 1991:7; cf. 1990:121–157, 1999:18–25, 2002:109–124, and 2005). We must read behind the surface of the immediate text to its connotative meaning by looking to the tradition to which it is referring, starting at the level of traditional diction and continuing up to the level of story pattern (Foley 1990, 1991, 1999:32–34 passim).
Foley’s central point is that the larger tradition is a very necessary place to look to discern how Homer’s audience understood meaning in any passing narrative moment:
Intone the formulas, instance the typical scenes, and, as long as the audience is reasonably fluent in the register, you accomplish far more than a superficially coherent narrative or metrically acceptable lines. Singular moments are implicitly framed as familiar situations. [16]
Foley 1999:33 (italics mine)
Epic narrative moments cannot be fully heard and appreciated without an understanding of the tradition as context (“familiar situations”). Although Wyatt employs a simple analogy, his comments meant to support Foley’s general approach are apt: “Modern television is opaque or silly to me because I am not fully aware of the conventions of the genre that the merest child internalizes early on” (Wyatt 2000). To see the bigger picture, so Foley suggests, we must seek to “read behind and between the signs,” tapping into “their idiomatic and traditional implications” (Foley 1999:7). [17] When heard in its entirety, Foley’s arguments are persuasive and formative. [18]
Foley’s methodology has proven influential in a number of recent works. [19] Central to Foley’s argument is what he termed the “resonant background” of an active oral tradition (Foley 1999:32), an idea taken up by other scholars. [20] Di Benedetto (1983), in his consideration of Homeric phrases, had presaged and paralleled some of Foley’s ideas, especially in his desire to move beyond metrics to read meaning. [21] It is this emphasis on resonant background that forms the core of the present study. As later readers, we must seek to hear the now written word in Homer’s oral register if we wish to understand a narrative description more like the traditional audience would have.
Resonance is especially important, so I am arguing, for a proper understanding of how the poets presented, and how audience members heard and thought of, epic personalities in any given narrative moment. I am consequently extending Foley’s ideas to a consideration of Homeric characterization. [22] I am arguing that epic characterization relies upon deep and extensive traditional resonances adhering to each character from his or her participation in past stories. [23] Resonance through traditional referentiality allows the poet to create richer and more suggestive moments of characterization, since words and idioms have taken on tradition-based meaning. Further, resonance from back stories creates audience expectations for the sorts of traits an individual might manifest in a given narrative moment (cf. Minchin 2011a:328–342). Of course, as I will argue in a moment about the Odyssey, even the memorialized version of the story we are reading from a written text has been performed before. We cannot, then, simply comprehend characterization in the Homeric text through chronologically linear strategies of character creation. Rather, we must also consider the whole of each epic’s language and story as parallel history that can inform any particular moment of character portrayal. As Dué articulates it, we need to consider the “large-scale” and “long distance” associations from within conventions that formed part of the epic song tradition (Dué 2002:19).
The Homeric poet is aware, then, that his audience depends upon tradition-based characterization for a fuller comprehension of a particular narrative moment. After all, as Minchin notes, “the stories themselves (their sequence of action) held this [character] information together” (Minchin 2011a:330). Both the plot and the character are there in skeletal form in the tradition. To miss the cues found in the register of Homer’s traditional language (as we will see in our study of Agamemnon’s appeal in the next chapter) or the larger back story is to miss a great deal. [24] Further, at any moment in the story, the poet is able to emphasize certain traditional aspects of character, and context does make a difference in what Homer’s audience was hearing during aoidic performance. As Kelly remarks, the semantic expressiveness of traditional language involves a “creative dynamic between denotative and connotative” [italics mine] meaning (Kelly 2007a:6). It is not a binary. This is true at all levels of Homer’s use of traditional elements.
As we will see in Chapters 3 and 4, literary questions and context also have a role to play, even when one accepts the priority of traditional background for “reading” character. [25] Nor would one expect audience members, for example, to have the same view of Agamemnon as a specter in Odyssey 11 that they did of him in the thick of the battle for Troy in Iliad 11. Yet, the fuller picture of a character from the larger tradition is always somewhat imminent, even if not brought fully “on stage.” [26] Consequently, as we will see when we look at Odyssey 11, even in Hades the poet assumes our comprehension of what Agamemnon once was, but is no longer. Then again, in Iliad 1, when Agamemnon arrogantly refuses to give back Chryseïs, the poet also makes some reference to Agamemnon’s own nostos and pitiful death. [27]
Yet, what can be said about the composition of the audience, since a referential reading is not always necessary for a limited understanding of the immediate action of a particular scene (Scodel 2002:19). [28] Danek offers a reasonable perspective when he speaks of an “ideally competent audience” (Danek 2002:4, 15, 19). I agree that Homer has a principal audience in mind as he sings. [29] It is not first the uninformed listeners, those members on the periphery of his traditional audience, but those at the core. It is these, for instance, that could really understand the proem of the Odyssey, or many other moments in Homer’s narrative. This realization does not of course mean that debutants or even strangers to the Homeric stories could not understand much what was sung. The retelling a traditional story, in the very act of its narration, will include much that needs little explanation for a general understanding. Yet, Homeric poetry also reveals that much would be missed without a traditional and informed audience. [30]
We must, moreover, it needs to be emphasized, be cautious here as armchair academics talking about live performance in an oral culture. Field work on the South Slavic epic tradition shows that the performance arena “is more vivid, more arresting, more demanding, more contingent” than we might expect as a reading audience (Foley 2002:84). [31] An eyewitness report from an early (pre-Parry-Lord) researcher in the late 1920’s shows that the South Slavic singer often competed with his peers with great drama and enthusiasm, in a performance that could last into the night. The event could leave the audience members on the edge of their seats:
L’auditoire écoute le chanteur avec le maximum d’attention, d’intérêt et de sympathie pour le héros, et il est parfois extrêmement touché du poème tout entier ou de certains épisodes. [32]
Murko 1928:341
Alternatively, the audience could be full of reproach (Murko 1928:340–342). Upon returning from a performance break, one less capable singer even found “les cordes et l’archet de son instrument [graissés] avec du suif” (Murko 1928:342), which made it impossible for him to continue his song. Apparently the South Slavic singer and his audience took epic performance as intently as Americans take their football, since it was seen as a competition between “warring” sides. Murko describes the intensity of competition:
En Herzégovine, un jeune homme de dix-neuf ans m’a dit: “Tous autant que nous sommes, ici, nous sommes ennemis les uns des autres. C’est un tourment pour moi quand j’en vois un autre qui en sait plus que moi.”
Murko 1928:341
That Homeric epic was once such a living oral tradition implies that it was heard intently and sung by singers that could feel “tourment” and other emotions. The same intensity of performance involvement is pictured for rhapsodic competition, if Plato’s (albeit somewhat ironic) portrait of Ion is any guide. Ion describes his performance:
“For whenever I speak anything woeful, my eyes with tears are filled: Whenever [I speak] anything fearful or awesome, my hair stands on end from fear, and my heart leaps!”
ἐγὼ γὰρ ὅταν ἐλεινόν τι λέγω, δακρύων ἐμπίμπλανταί μου οἱ ὀφθαλμοί: ὅταν τε φοβερὸν ἢ δεινόν, ὀρθαὶ αἱ τρίχες ἵστανται ὑπὸ φόβου καὶ ἡ καρδία πηδᾷ.
Ion 535c
Not only is the rhapsode emotionally involved, but the audience is described as responding with an equal fervor (τοὺς πολλοὺς ταὐτὰ ταῦτα, 535d) when “struck” (ἐκπλήσσω, 535b) by the power of the performed epic narrative. [33] Every time Ion looks down upon his audience, they are constantly crying (κλάοντας) and grimacing (δεινὸν ἐμβλέποντας), wonder struck (συνθαμβοῦντας) by what they are hearing. [34] While a later rhapsode’s recitative performance can be contrasted in kind with the earlier aoidos’ composition-in-performance, it is clear that Ion undertook his poetic presentation as a living and emotionally charged event.
Homer’s “core audience” members, then, were not only the better informed, but also the more personally and passionately involved. [35] They were actively thinking about what the next line might be, attuned to the “moment-to-moment reality of the evolving song” (Foley 2002:84). The proof that Homer’s concern was first with core audience members who were quite actively engaged and generally “in the know” should not be missed. There are moments where Homeric poetry (as it stands in our texts) cannot be fully understood, even at a surface level, without a metonymic reading of the poet’s presentation. Certain epithets and expressions make little sense without a metonymic comprehension of the greater tradition. [36]
Such is the case with Achilles’ first Iliadic self-reference to an abbreviated life (1.352): “Mother, since in fact you bore me to be short lived” (μῆτερ ἐπεί μ’ ἔτεκές γε μινυνθάδιόν περ ἐόντα). In this case the comment is bewildering to the reader unaware of the larger tradition. Thetis makes a second (but more extensive) reference to Achilles’ short-lived life, after she has heard the complaint of her son over the hubristic treatment that he received from Agamemnon:
Then Thetis answered him, pouring down tears:
“O my child, why did I raise you, having borne you terribly?
Would that you, by the ships, without tears and without harm,
would sit, since now your allotment of life [is] short and not very long;
But now altogether swift-fated and woeful beyond all men
you have turned out to be; therefore I bore you with a bad allotment of life in my house.
But indeed, in order to speak this word to Zeus who delights in thunder,
I am going myself to snow-clad Olympos, in case he can be persuaded.
But you now, sitting by the swift-going ships,
vent your wrath on the Achaians, but stay away from the war altogether.” [37]
“O my child, why did I raise you, having borne you terribly?
Would that you, by the ships, without tears and without harm,
would sit, since now your allotment of life [is] short and not very long;
But now altogether swift-fated and woeful beyond all men
you have turned out to be; therefore I bore you with a bad allotment of life in my house.
But indeed, in order to speak this word to Zeus who delights in thunder,
I am going myself to snow-clad Olympos, in case he can be persuaded.
But you now, sitting by the swift-going ships,
vent your wrath on the Achaians, but stay away from the war altogether.” [37]
τὸν δ’ ἠμείβετ’ ἔπειτα Θέτις κατὰ δάκρυ χέουσα·
ὤ μοι τέκνον ἐμόν, τί νύ σ’ ἔτρεφον αἰνὰ τεκοῦσα;
αἴθ’ ὄφελες παρὰ νηυσὶν ἀδάκρυτος καὶ ἀπήμων
ἧσθαι, ἐπεί νύ τοι αἶσα μίνυνθά περ οὔ τι μάλα δήν·
νῦν δ’ ἅμα τ’ ὠκύμορος καὶ ὀϊζυρὸς περὶ πάντων
ἔπλεο· τώ σε κακῇ αἴσῃ τέκον ἐν μεγάροισι.
τοῦτο δέ τοι ἐρέουσα ἔπος Διὶ τερπικεραύνῳ
εἶμ’ αὐτὴ πρὸς Ὄλυμπον ἀγάννιφον αἴ κε πίθηται.
ἀλλὰ σὺ μὲν νῦν νηυσὶ παρήμενος ὠκυπόροισι
μήνι’ Ἀχαιοῖσιν, πολέμου δ’ ἀποπαύεο πάμπαν·
Iliad 1.413–422
Why the tears? “Therefore I bore you with a bad allotment of life” (τώ σε κακῇ αἴσῃ τέκον)? Why is Achilles’ “allotment of life” (αἶσα) short and mal? [38] “Cease from the war altogether (πολέμου δ’ ἀποπαύεο πάμπαν)!” Why should he wait by the ships and keep himself completely out of the war; and what would happen were he to reenter battle? Thetis just leaves without explaining herself! What does all this portend for the traditionally informed audience member?
A significant traditional cue occurs in line 413. Here, Thetis’ sorrowful crying (1.413, κατὰ δάκρυ χέουσα) seems to derive from the register of a funerary lament [39] —“O my child, why did I raise you, having born you terribly” (ὤ μοι τέκνον ἐμόν, τί νύ σ’ ἔτρεφον αἰνὰ τεκοῦσα, 414)? This is likely if we compare Thetis’ with Hecuba’s later lament over her son Hector’s corpse (Iliad 22.431–32): “Child, I am wretched; Why should I live now, suffering terribly, / since you have died” (τέκον ἐγὼ δειλή· τί νυ βείομαι αἰνὰ παθοῦσα / σεῦ ἀποτεθνηῶτος)?” We observe that both Thetis and Hecuba address their “child” in the first half of the line (1.414 and 22.431). Both ask a rhetorical question brought about by their son’s death, or in Achilles’ case, portended death. [40] In the case of Thetis, the lament is proleptic, and is, as with other “déplorations anticipées, généralement plus courtes” (Arnould 1990:189). [41] Yet, although brief, it cues the audience to what will eventually transpire—Achilles’ death—making the future imminent through the impinging tradition. That Thetis’ address is posed as a question proves significant. As Alexiou notes, the opening question of a lament can “emphasize the plight of the mourner” (Alexiou 1974:162). Moreover, a formulaic participle with adverb is common to both laments (Kirk 1985:96): “having born you terribly” (αἰνα τεκοῦσα, 1.414) / “suffering terribly” (αἰνα παθοῦσα, 22.431). [42] Elements often occurring in a lament, such as distress and anger or an immediate and expressed sorrow (Alexiou 1974:162) are less prolonged in our scene when compared to Hecuba’s threnody, but they are there. [43]
Hopefully now the implications are clear. The singer assumes the core audience’s knowledge of the looming fate of Achilles when Thetis speaks; and he employs traditional cues that resonate against an oral traditional background for his hearers. [44] Of course the much fuller lamentation by Thetis is found in Iliad 18, where it is clear that she thinks of Achilles’ death. [45] Yet, Achilles’ impending death is already found through metonymy in Iliad 1, very early on in the text.
The foregoing exegesis is meant simply to describe what the traditional audience is aware of already. The poet is signaling the death of Achilles, about which the Iliad knows much but says little directly. As Schein articulates it: “The most important later event referred to but not told in the Iliad is the death of Achilles” (Schein 1984:25; cf. Whitman 1958:201–203). The entrance of Achilles into the war signifies that “Achilles is virtually dead from the beginning of book 18 on.” Further, Schein notes that although Achilles’ death is not expressly related in the Iliad, nevertheless Homer links his death with that of Hector (Schein 1984:128–167; cf. Dué 2002:14). [46] Dué’s remarks concerning lamentation support the centrality not just of Achilles’ mortality, but also of his death: “The Iliad quotes within its narration of Achilles’ kleos many songs of lamentation that serve to highlight the mortality of the central hero” (Dué 2006:66). Schadewaldt paints a vivid picture of Achilles’ comprehension of his mortality, noting that a large vista opens up for Achilles, but one “unter der Herrschaft des Tods” (Schadewaldt 1965:240). Clearly Homer is not the first singer to weave a tale of Achilles’ life and death. The audience is aware of Achilles’ fate in a story they have heard in some form many times already. The poet’s mental “script” assumes a dependence on a common oral tradition. The actual explication of Achilles’ and Thetis’ comments is not forthcoming ostensibly (without a referential reading of each instant) until Iliad 9.410–416, where Achilles, through knowledge given from his mother Thetis, also outlines his choice. [47]
Then there is the matter of certain epithets that clearly contradict the action of the immediate context, something Charles Perrault noted already in the seventeenth century: “Le plus souvent ces épithets vaines et vagues, non seulement ne conviennent point au fait qui est raconté, mais y sont directment opposées.” [48] Two instances stand out vividly. I speak of “swift-footed Achilles” (πόδας ὠκὺς Ἀχιλλεύς Iliad 1.489) [49] sitting by his ships in anger (Iliad 1.488–492; cf. 9.307) and “laughter-loving Aphrodite” (φιλομμειδὴς Ἀφροδίτης Iliad 5.375), [50] troubled (382), enfolded in the comforting arms of her mother, complaining before her father Zeus how “arrogant Diomedes” (ὑπέρθυμος Διομήδης 376) has stabbed her (336–417). How can such instances make sense without a traditional comprehension of the character’s activities?
Parry’s theory was that the epithet should be read as a metrically convenient formula, whose usefulness lies in its essential theme. In his 1930 paper he formulated his famous dictum:
The formula in the Homeric poems may be defined as a group of words which is regularly employed under the same metrical conditions to express a given essential idea. [51]
Parry 1971:272 (italics are his; cf. 13; and Lord 1960:49–53)
As subsequent research has pointed out, however, absolute metrical thrift is confined (but even here with exceptions [52] ) to epithet formulas. More significantly for our present study, “the idea of thrift has nothing to do with idiomatic meaning” (Foley 1999:211). [53] We must instead read behind the sign to what is being referred to in the tradition. Whallon had earlier shown that epithets can reveal the “essential and unchanging character” of individuals in the epic tradition (Whallon 1969:2) [54] ; that an epithet, especially one that is particular to a character (or even mostly particular as Di Benedetto suggests [55] ), instances qualities that stem not first from the immediate moment in the singer’s song. They retain their semantic force within the larger tradition. [56] The epithet must be heard against a broader and deeper oral history, or, in Nagy’s apposite description: “A distinctive epithet is like a small theme song that conjures up a thought-association with the traditional essence of an epic figure, thing, or concept” (Nagy 1990a:23). The epithet “swift footed” (πόδας ὠκύς) in “swift footed Achilles” (πόδας ὠκὺς Ἀχιλλεύς) was heard as a whole “word.” [57] For the ancient audience, it referred to Achilles as a character “using a telltale sign,” his swiftness of foot (Foley 1999:210). [58] An epithet, I suggest, brings to the minds of the audience members a character’s traditional personality.
The epithet has traditional meaning. It is heard by the singer’s audience against the larger tradition, but also within a local narrative context as a singer weaves the tapestry of the traditional story. [59] Di Benedetto’s study of Hector’s epithets suggests further that the more specialized epithets had even more pronounced local meaning for Homer’s audience in cases where a singer had a choice between two metrically equivalent epithets (Di Benedetto 1994:131–139). [60] There may even be irony, then, in such situations as “swift-footed Achilles’” idleness. [61]
The foregoing observations concerning epithets support my approach for considering epic characterization more generally. Like the particular epithet, character is not something first created in a singular occurrence within a narrative moment. Nor can the fuller nature of Homeric characterization be heard in isolation from the larger tradition. Character is something profoundly connected to the tradition the poet has received and is referring to, as it is being heard and understood within a local narrative setting. Although much is lost to us, as later readers we have to try to recreate the web of character traits known to the singer and his core audience.
All that I have said is not intended to deny the poet freedom to employ recognizable tradition in creative ways or for the tradition to change over time and absorb new elements. [62] Indeed, the poet’s individual presentation can play dynamically upon the audience’s knowledge or expectation (cf. Morrison 1992a, Porter 2014). This is a point that we will return to again and again as we consider particular moments of Homeric characterization. Further, the tradition is itself always developing. The very existence, however, of a recognizable history attached to a particular character frees the poet from the necessity of pure invention, carte blanche. A distinct character already invokes a particular ēthos for the hearer. As I will note in Chapter 2, significant departures from the inherited picture can be used to the poet’s advantage as he unfolds his story.
1.4 The Relation of the Iliad and Odyssey
Before we consider Homeric characterization further, however, I need to state explicitly an underlying assumption affecting the treatment of the Iliad and Odyssey in Chapters 2–5. My present consideration allows for the possibility of different poets for the Iliad and Odyssey, but for a common tradition to which each refers, as, among other things, the underlying traditional vocabulary suggests. [63] Yet the history of the poems from oral performance(s) to written form is obscure. I therefore employ Foley’s description of Homeric epic as “oral-derived” (Foley 1990:5–8, 1991:38n42, 2011:603).
“Oral-derived” affirms the inherent oral-traditional essence of the poems. The use and adaptation of formulae, the local overuse of repetitions (clustering), as well as other elements of diction and metrical irregularities, suggest the essential oral nature of the poems. [64] As Di Benedetto concludes: “I poemi omerici presentano una caratteristica formale del tutto peculiare” unconcerned with any written text (Di Benedetto 1994:104). [65] Yet “oral-derived” also accepts and notes by the qualifier “derived” that the Homeric epics come to us through documents with a relatively unknowable textual history from Homer’s day to the time of the Alexandrian scholars and the establishment of the Vulgate text. The fact that the poems are written reminds us that they most likely represent an expanded presentation, as Lord observed when making his comparison with South Slavic epic (Lord 1956, 1960:124–128). Lord reminds us that our Iliad and Odyssey cannot ever be entirely identical with the sung text in the normal performance arena. The study of enjambement and other poetic elements may also suggest as much (Dukat 1991, Friedrich 2007). [66] The dictated manuscript will be longer and provide opportunity for the “exceptional singer” to expand traditional themes and story length and to “build his lines somewhat differently” (Lord 1960:128, 127, respectively). [67] The process of dictation itself could also have a negative effect on verse formation, something that Parry had earlier noted; but really, the variables in the details of the process of dictation are themselves uncertain and manifold, as Ready’s survey of the issue demonstrates. [68] The Iliad and Odyssey are representatives of songs built in a somewhat artificial environment. We cannot know exactly what the first, dictated texts looked like. Nor did the performance tradition suddenly stop with these memorializations, but continued to live on (cf. Dué 2001 and Scodel 2002:58).
Further, as I argue in greater detail in Chapter 2, the Odyssey poem and the character of Odysseus as we have it in the Iliad shows that our written version of the Iliad in fact postdates some oral version of the Odyssey. There must have been an oral story similar to our Odyssey, known to the Iliad poet composing his song. The question of the relation of the Iliad to the Odyssey and therefore of the relationship of individuals who appear in each epic is an important one. It affects how we should understand the character of Agamemnon in each epic. My findings suggest a different emphasis from that of Heubeck, who writes, “What is certain is that the figure of Odysseus as it appears in the Odyssey is shaped by what the poet found in the Iliad and took from there” (Heubeck 1988:19–20). I agree that the Iliad’s story does play a significant role in shaping the action of the Odyssey (including, as Heubeck notes, Odysseus’ movement from an Iliadic hero to a hero who experiences the loss of glory and humiliation). [69] Yet, while this is true, the opposite is also accurate. The Odysseus of the Iliad is referred to in ways that presume the audience’s knowledge of the sort of story one finds in the Odyssey, a point suggested earlier by Nagy: “The Iliad is recording the fact that Odysseus already has an Odyssey tradition about him” (Nagy 1979:2). [70] Each epic story had been told before in some similar fashion. Performed versions of the Odyssey predate our present written version of the Iliad. [71] The Iliad itself alludes to such traditions, however obliquely, when Homer narrates, at a moment of uncertainty in the war, the danger of a nostos for the Achaians beyond what was fated. [72] Their return stories had already been told.
Nor is it likely in the general illiteracy of late-eighth-century Greek society that earlier written versions would have served much of a practical purpose. So the influence took place orally, through aoidic performance. [73] As Lord observed through his study of the South Slavic comparison, even reading a song from a songbook to an illiterate guslar did not determine how the singer would render his own uniquely crafted performance (Lord 1960:79). This realization does not deny that writing and the process of textual transmission have influenced the form of our present manuscripts of the Homeric epics (e.g. see Bolling 1925 and Apthorp 1980). [74] Rather, it acknowledges the essential oral traditional nature of our poems.
The idea that the Iliad poet knew and had in mind an oral story similar to our Odyssey when composing his poem will be a point we will consider further. We will see in the course of this book that Homer may have had in mind not only the return of Odysseus, but also the kakos nostos (“dismal return”) of Agamemnon, as he characterized Agamemnon in the Iliad. This tenet forms the impetus for the way I structure Chapters 3 and 4, where we will examine the evidence from Homer, looking first at the Odyssey and then at the Iliad. [75]
Even the history of the Odyssey as a written poem continues to intrigue. It challenges any simple limitations we might put on its use and presentation in actual performance, for instance, within the genre of fifth-century drama. For example, in one speech from Euripides’ Trojan Women (lines 425–45), Cassandra presents the events of Odysseus’ wanderings out of order when considered against the sequence we have narrated at any point in our Odyssey. Are Euripides and his Classical Greek theatai more concerned with the homecoming events themselves than with their chronological sequence? [76] Are the events thought of as separate episodes from within the post-war period of Odysseus’ return? [77] The case of Euripides’ Trojan Women offers evidence of the freedom for variation in the fifth century, even after the creation of written versions of Homer for didactic purposes. [78] The freedom for chronological variation in Euripides is an analogue, I suggest, for the performative background of the Odyssey as epic, which, I am arguing, needs to be read not just after, but also alongside, the Iliad. [79]
Footnotes
[ back ] 1. I employ “Homer”/“Homeric” throughout to stand for the unknown oral poet (aoidos) or poets (aoidoi) who sang the Iliad and Odyssey (cf. Parry 1971:3n2, Martin 1993:227, Edwards 2005:302, and Minchin 2007:3), as well as for these epics as memorialized texts. By “poet” I mean to emphasize a singer’s thoughtful artistry through the purposeful use and arrangement of traditional elements that carry inherent meaning for the audience during epic performance. As with the more competent among the guslari (South Slavic “epic singers”), such as Salih Ugljanin, Stanko Pižurica, or Avdo Medjedović (see the CD-ROM of archival material in the updated [2000] edition of Lord 1960 edited by Mitchell and Nagy), I assume that the aoidoi who sang the Iliad and Odyssey were accomplished performers (cf. Dué and Ebbott 2010:10, and also Lord’s [1960:26] comments about the less competent guslari), but that the tradition first gives meaning to what is sung (Foley 1999:56–58, 2002:8–10, and Dué and Ebbott 2010:13, 20–28). As I hope to show throughout the chapters that follow, the Homeric poets were aware of a web of traditional associations and cues and used them to great advantage (cf. the experience of Lord 1960:26). For assumptions and difficulties in defining “Homer,” see also Graziosi and Haubold 2005:1–34. My transliteration choices for Greek names are similar to those of The Homer Encyclopedia and Powell’s recent translation of the Iliad and Odyssey.
[ back ] 2. The Iliad’s catalogue of ships was later rounded to an even 1,000 ships, preserved for us first in Aeschylus’ Agamemnon 45. The same count appears in five different Euripidean plays. The more well-known description, however, is derived from Christopher Marlowe, in Doctor Faustus, referring to ‘Helen of Greece’: “Was this the face that launch’d a thousand ships / And burnt the topless towers of Ilium? / Sweet Helen, make me immortal with a kiss.”
[ back ] 3. This involves the venerable task dating back to Aristarchus of explaining Homer from Homer: ἐξηγεῖσθαι τὸν ἄνδρα ἐξ ἑαουτοῦ / explicare Homerum ex Homero (Porter 1992).
[ back ] 4. For a list of some works before the mid-seventies on individual characters, see Heubeck 1974:197. There are, of course, a limitless number of works that touch on a particular character’s attributes in some way, without ever defining how characterization is created. My concern here is first with studies of characterization itself undertaken since the foundational work of Parry and Lord, and second, with the study of Agamemnon’s entire character in Homer.
[ back ] 5. Agamemnon can be found as part of larger questions about particular themes in Collins 1988:69–102 (the relation of king and warrior), Taplin 1990 (Agamemnon’s status as leading basileus), Zanker 1994 (his relation to Achilles), Wöhrle 1999:49–61 (his role as a surrogate father in the “Hierarchie” of father-son type relationships), Haubold 2000 (his role as shepherd), Hammar 2002 (his political role), Wilson 2002 (Agamemnon and compensation), Heiden 2008 (his responsibility for the suffering of the army), Rinon 2008 (his relation to a tragic pattern and lost opportunity), Scodel 2008 (Agamemnon and the theme of saving face), Ammone 2010 (Agamemnon and the diapeira), Cairns 2011 (Agamemnon and atē), and Holway 2012 (his feud with Achilles). For short portraits of Agamemnon as a character, see Bonnéric 1986:7–11, Auffarth 2002, van Nortwick 2011, and Kanavou 2015:44–48. In terms of Agamemnon’s representation in iconographical, lyric, and tragic representation, the Oresteia in its narrowest sense has been the central focus (especially events surrounding Agamemnon’s death). See especially Prag 1985, Knoepfler 1993, Neschke 1986, Garvie 1986, and Raeburn and Thomas 2011.
[ back ] 6. There are various more traditional approaches to the study of characterization in the Homeric epics, however, from Griffin 1980:50–81, Collins 1988, Pelling 1990, and Race 1993. Collins reviews a group of scholars who study character through the lens of social institutions, and adopts their structural-functional approach to characterization (following Vernant 1966, Finley 1979, Queller 1981, and Donlan 1982). Griffin’s study finds both individuality and depth of presentation for Homeric characters. Pelling considers general psychological tendencies with Homeric characterization. Race’s concern is with characters in the Odyssey, and his article reveals something of each character’s traits. He is not concerned as such with providing a theoretical framework for understanding what characterization is or with the centrality of the background tradition influencing characterization in context. He consequently emphasizes first appearances, rather than considering the local expression of traits against the traditional character already known to Homer’s audience. Yet, his article helps demonstrate that many Odyssean characters have a recognizable persona.
[ back ] 7. Within his discussion, Martin emphasizes epithets, themes, phrases, and narrative focalization. The last point, as we will see, is parallel to my own emphasis throughout on the singer’s creative use of tradition and the relation of audience and performer.
[ back ] 8. Her study, although not meant first as a character study, nevertheless, shows that individual characterization is present in Homer, thanks to the presence of an individual’s “themes, goals, and plans.”
[ back ] 9. Minchin (2005:66) rejects the idea that oral poetry is “stereotypical,” noting that there is, “on the contrary, a remarkable rich vein of psychological insight in the poet’s recreation of everyday behavior patterns in his principal actors.” For earlier consideration of the question, see Donlan 1970, 1971, Griffin 1980:70, Martin 1993, and Dué (2002:8), who considers the “syntagmatic,” as well as the “paradigmatic,” in the Homeric picture of Briseïs. Both particular and typical character elements are found in Homer, as we will consider further in Chapter 2. For discussion of lesser Homeric characters and the question of invention, see Schein 1984:27, Nünlist 2009:240–241, and Kanavou 2015:134–150, although it would be difficult to prove invention in most cases, even for minor characters. Dué’s (2002:33–35 passim) findings for the multi-forms in the Briseïs tradition suggest the need for caution in making such claims.
[ back ] 10. Martin (1993) suggests the role of such factors as Russian Formalism (including questions of function) and the “deterministic” view of Homeric poetry that followed upon Parry’s initial discoveries, as the reason that many Homerists have avoided exploring the topic of characterization.
[ back ] 11. This is the question already posed, but left unanswered by Griffin (2011:158), whose principal concern is with character description.
[ back ] 12. Cf. the call of Martin (1993:227) for a book-length study of “ways of employing the tools of oral-formulaic analysis in the study of characterization.”
[ back ] 13. I had the pleasure of being a graduate student of John’s from 2002 to 2007. John was a very generous scholar and took the time to help me consider oral traditional patterns, beginning at the level of language. This made my own research experience quite diverse, since I had come to Missouri upon the completion of a graduate degree in Canada with another leading Homerist from a different school of thought, Prof. Rainer Friedrich of Dalhousie.
[ back ] 14. This includes, among other scholars after Parry and Lord, Whallon 1969, Muellner 1976, Nagy 1979, Holoka 1983, Lowenstam 1981, Slatkin 1986, 1991, and Sacks 1987.
[ back ] 15. Foley’s work on metonymy (1991:38–36) was especially influenced by Iser (1974). For a consideration of metonym as a linguistic feature, see Barcelona et al. 2011.
[ back ] 16. Cf. Foley 1991:54–55, on becoming a responsible reader of an oral-derived text.
[ back ] 17. Cf. the earlier comments of Anne Amory Parry (1973:7) that this appreciation begins at the level of poetic diction: “Most efforts to define Homeric words start with an etymological hypothesis; ancient scholia abounded with suggestions ranging from the patently ridiculous to the more plausible, and the latter are still accepted. But etymology is still essentially irrelevant to the problem of meaning in a literary context. Whatever the etymology of a word may have been conjectured to be by the early grammarians, or whatever modern scholars may assert that it is, the meaning of a word for the composer of the Iliad and Odyssey, and therefore its proper meaning for us as readers of the poems can only be determined by its usage in the text of Homer” (italics mine). This realization does not rule out the sort of connections that form the basis for etymological or comparative studies (such as the Indo-European meaning noted for Diomedes by Benveniste 1969; cf. Schnapp-Gourbeillon 1982:46). It does, however, act as a needed reminder that one must first judge any Homeric meaning by a word’s (or idiom’s) use in Homer. Further, while traditional meaning begins at the level of words, it continues through theme, type scene, story pattern, and various other poetic devices.
[ back ] 18. Cairns (2011:113n26) reacts to Foley’s methodology in what he reads as extreme claims. He sees “no evidence for the view that audiences always activate knowledge of the totality of a multiform tradition” (italics mine). His words come as part of a response that questions Wilson’s (2002) associations of poinē with particular scenes in the Iliad (on Wilson’s methodological assumptions, see Wilson 2002:185n40). While some of Foley’s statements when taken in isolation, could, I suppose, be read in absolute terms as Cairns does, Foley’s overall claims for traditional referentiality and audience knowledge as outlined above, should not be. Wyatt (2000) is more measured: “Foley has overdetermined some of the ‘words’ with which he deals. … Nonetheless he asks the important question and has gone some way to answering it.”
[ back ] 19. A representative sampling from Homeric Studies includes Ledbetter 1993:485, Bakker 1995:102–103, Dunkle 1997, Danek 1998:13, 509, Danek 2002, Dué 2002, Tsagalis 2004:20, Turkeltaub 2005, Barker and Christensen 2006, Dué 2006, Brown 2006, Kelly (2007a, who includes a referential lexicon), Gaca 2008:159, Sammons 2008:360, Scott 2009:20, Létoublon 2009:37, Dué and Ebbott 2010:13–29 (with a referential commentary for Iliad 10), Barker 2011, Porter 2011, Ready 2011, Alden 2012:117, Bierl 2012, Elmer 2013:21–47, Barker and Christensen 2014, Arft 2014, Karanika 2014, Kelly 2014, Porter 2014, and Ready 2015. Foley’s ideas have also proven influential in Biblical Studies, Folklore, Linguistics, Old English, South Slavic, and other studies.
[ back ] 20. Foley’s idea of resonant background has been particularly foundational in Dué’s (2002:9 et passim) study of what she calls “micro” and “macro” narratives that influence audience comprehension of Briseïs, and Graziosi and Haubold’s (2005:48–55) consideration of Homeric themes as they relate to the wider epic traditions.
[ back ] 21. Di Benedetto’s earliest work (1983) stressed only resonance within the Iliad, while his subsequent research included the Odyssey, where he suggested that there can exist not just “formularità esterna (in riferimento alla ripresa di espressioni tradizionali fisse preesistenti)” but also “formularità interna” (1994:ix). He argues that “formularità interna” are essentially recurring formulae that carry meaning, reused (but with replacement, extension, etc., 1994:115–120) by the poet, rather than simply being repeated without meaning (see esp. 1994:122–139). Significantly, Di Benedetto (e.g. 1994:103–121, 1999:210 et passim) suggested that one epic idiom was dependent on its other occurrences, albeit not from a strictly oral traditional perspective (his approach includes the neoanalytical, e.g. 1994:104), and without the idea of a larger epic register of a living oral tradition. He envisions not just a singular “risonanza,” but also “una serie di risonanze” both within the Odyssey, but also from the Iliad to the Odyssey. My own emphasis stresses a common oral traditional background for both the Iliad and Odyssey and importantly assumes, and in this I differ with Di Benedetto, the influence of the Odyssey tradition upon the Iliad for Agamemnon’s character traits.
[ back ] 22. Martin (1993) noted the need for such a study and his article harbingered some of the questions I seek to consider here and in Chapter 2. Dué (2002) has applied Foley’s traditional referentiality to her consideration of Briseïs’ involvement in particular themes (as captive, prize, girl, daughter, and wife). Scott (2009:20) and Ready (2015:89) have applied Foley’s traditional referentiality to the study of the simile.
[ back ] 23. Cf. the comments of Dué 2002:15–19 and Dué and Ebbott 2010:9. It may even be, as Griffin (1980:73–75) boldly argues, that the character’s presence in a narrative drives the plot, rather than merely being subservient to it.
[ back ] 24. The importance of traditional cues becomes apparent in ironic type-scene inversions of the sort noted by Reece 1993:130–143, Foley 1999:181–183, and Kelly 2012:241–245. Cf. Pantelia’s (2002) comments about expectations for the order of mourners in Iliad 24. As we will see throughout our present study, the tradition often points beyond the ostensible facts of the immediate moment to larger traditional elements. We will return to consider type scene inversions in Chapter 2, s.v. 2.1.1 Character Consistency.
[ back ] 25. Janko (1998:10) notes questions of theme as one example. Janko’s (1994:xi) comments regarding literary approaches seem sensible.
[ back ] 26. Scodel (2002:15) argues that sometimes the poet intentionally does not actively “bring on stage” mythic elements known to the audience, because of his intended emphasis at a given narrative moment. Scodel’s concern here, through example, includes the thornier issue of how much of the Cyclic Epic material was known to Homer.
[ back ] 27. We will return to Odyssey 11 and Iliad 1 in Chapters 3 and 4, respectively.
[ back ] 28. Scodel does not deny the importance of traditional referentiality to narrative content, but she sees it as a spectrum (Scodel 2002:13): “Traditional referentiality does not mean that everyone knows all the same stories, in the same variants.” Cf. 16–17, 24, 173 et passim. Cf. Scodel 2005:402.
[ back ] 29. Further, Nagy (1979:7, 1990:43, 1996b:38–43) and Dué (2002:2n5, 23–36) suggest the mechanism of Panhellenism (local traditions are de-emphasized) for reducing less recognizable elements for the Homeric audience.
[ back ] 30. Danek (1998:31) notes Homer’s assumption of an informed audience, arguing that from the beginning, the proem “suggeriert daß die Handlung an einem bekannten Punkt innerhalb eines bekannten Rahmens einsetzt.” Scodel’s (2002:19) views are more cautious about suggesting as much: “Even young children probably knew the names of Achilles and Odysseus. That there was a canon of acts, however, does not define its extent.” While it is true that it would be difficult (often impossible; cf. Dué 2002:36) to define the full extent of what is traditional, yet, as I will show, Homer did very often assume for his core audience a certain canon of acts and many other traditional elements.
[ back ] 31. The memorialized copy of an oral performance cannot fully represent the performance (Powell 2000:107, Foley 2002:82–108, Finnegan 2007:79, Jensen 2011:290, and Ready 2015:47–51; cf. the earlier experience of Notopoulos 1964a:48, with Greek heroic poetry).
[ back ] 32. Lord (1960:280) praised the work of his predecessor. Further, one can also hear a proudly competitive tone in the response of another singer recorded by Parry, Đemo Zogić, who feels (he claims all the audience was more pleased with his rendition) that he can sing a particular song better than another singer (Suljo Makić, on which see Lord 1960:27). On the importance of competition in early Greek culture, see Nagy 1992:79.
[ back ] 33. Cf. the comments of Lada-Richards (2002:413), who notes that Ion was “totally submerged in the roles he incarnates, with the ebb and flow of his emotions entirely congruent with the succession of the passions encoded in the text.”
[ back ] 34. 535e: καθορῶ γὰρ ἑκάστοτε αὐτοὺς ἄνωθεν ἀπὸ τοῦ βήματος κλάοντάς τε καὶ δεινὸν ἐμβλέποντας καὶ συνθαμβοῦντας τοῖς λεγομένοις. Of course, we are receiving this picture, as the dialogue goes on to show, through the rather jaundiced eyes of Plato (cf. Republic 10.606e–608b1), on which see Halliwell 1986:1–6, 19–27, 331–336; Murray 1997:14–32; and Rijksbaron 2007:9–14.
[ back ] 35. If, as I hold as quite likely, the recording of Homeric epic took place by dictation, then the singer may have “imagined” (Jensen 1980:170) his audience, unless the memorialization of the epics took place, as usual, in an aristocrat’s residence, with the host and audience members in attendance (though the performance dynamics would have changed, as shown by Ready 2015:1–7, 13–33).
[ back ] 36. Cf. the presence of allusions noted by Kullmann 1960:13–17. For other examples, see Edwards 1987:52–53. I am not suggesting here, however, that there was a particular written song in the audience’s mind, fixed in one exact, static form, since, as Murko (1928:337) noted early on, “Toutes les tentatives faites pour reconstituter un chant dans sa forme originaire sont vaines.”
[ back ] 37. All translations from Greek texts throughout are my own. They are meant, in the case of Homer, to retain, as much as possible, the cola of the Greek lines, rather than create a polished English translation. It is nearly impossible, however, as Austin (2009:73) notes, to fully represent Homeric formularity. For a note about the metrical assumptions of this study, see Appendix A.
[ back ] 38. Chantraine (1968:38, s.v. αἶσα) connects the term αἶσα to an allotment of life by Zeus and close to “destiny,” a meaning equivalent to μοῖρα. The common origin of these two nouns from a single verbal form was suggested by Leitzke 1930; Cf. Dietrich 1965:184.
[ back ] 39. Cf. the tears enveloping the Muses’ threnody in Odyssey 24.61: οὔ … ἀδάκρυτον.
[ back ] 40. “Since now your allotment of life [will be] short and not very long” (ἐπεί νύ τοι αἶσα μίνυνθά περ οὔ τι μάλα δήν, 416). For a discussion of the meaning of this knowledge for Achilles, see Zanker 1994:77–82.
[ back ] 41. Kelly (2012) calls such laments “prospective,” when outlining their typical elements. Neither Arnould nor Kelly discusses this admittedly highly syncopated example in Iliad 1. Tsagalis (2004:139) mentions it in passing. For discussion of the classification of laments, see Tsagalis 2004:27–28n101.
[ back ] 42. Both are in the adonean clausula. For my metrical assumptions, see Appendix A.
[ back ] 43. They last a mere four lines, cut short by Thetis’ resolve to go to Zeus and keep Achilles, at least immediately, from his looming fate. The proleptic lament of Thetis is extended even farther in Iliad 18.54–60. Lamentation language (πένθος ἄλαστον; Arnould 1990:147) is also used when Zeus sends for Thetis in Iliad 24.105. She is already mourning the future loss of her son. The basis of my textual research throughout is the TLG database (cf. Edwards 1991:55, Di Benedetto 1999:217).
[ back ] 44. Schadewaldt (1965:264), from a neoanalytical point of view, recognized that the lament of Thetis was also seeing into the future. He argues, however, to my mind the less likely point, that the poet’s first reason for including lamentation in Iliad 1 was “der täglichen Plage seines Daseins.”
[ back ] 45. On Thetis’ lament in Iliad 18, see Kakridis 1949:66–68, Schadewaldt 1965:248–251, Tsagalis 2004, Dué 2006:65–66, and Kelly 2012:246–255. There is no antiphonal element in the present prospective lament, however, on which see Tsagalis 2004:48–51.
[ back ] 46. As Nagy (1979:113; cf. Schein 1984:25) has observed, however, the ritual of a real funeral is reserved for Achilles’ surrogate, Patroklos.
[ back ] 47. Willcock (1978:276, followed by Hainsworth 1993:116) suggests that Homer created the choice of Achilles introduced (and only overtly mentioned) here. I reserve decision on this. I hope that I have shown, however, that the actual fate of Achilles is too pervasive and necessary not to have been part of the poet’s received tradition. Other references to Achilles’ fate (instanced in all but the last of the following examples by some form of ὠκύμορος) include Iliad 1.505, 18.95, 18.458, and 21.277. (Less direct allusions to Achilles’ fate are made, too, in Iliad 11.793–794 and 16.36–37; more direct is Achilles’ own words at Patroklos’ funeral, Iliad 23.144–151.) It is obvious that the poet’s core audience has heard this story before.
[ back ] 48. Perrault (vol. 3, p. 111, from the 1979 reprint of his 1692–1697 work, also cited at length in Shive 1987:6) briefly lists a number of examples, including the first two that I give here.
[ back ] 49. On the epithet πόδας ὠκὺς Ἀχιλλεύς, see Nagy 1979:326–327.
[ back ] 50. The possible pun of Hesiod’s Theogony 200, which connects Aphrodite with μήδεα (on which see Chantraine 1968:677, s.v. μειδιάω), even if accepted, sounds equally out of place in the context of Iliad 5. On the importance of not reading the Homeric tradition through the lens of the Hesiodic, see Porter 2014.
[ back ] 51. I employ the term “formula” and “formulaic” throughout to speak of recurring traditional idioms or patterns of words with one or more important parts recurring as component(s), but with a varying amount of replacement of parts within the traditional idiom (which can also cause variation in length). Cf. Fenik (1968:5), who lists as “typical” any type scene elements that recur two or more times. It is not my intention, however, to limit what is formulaic. Much more work has been done to expand our consideration of what may constitute formulaic elements: Nagler 1967, Hainsworth 1968, Higbie 1990:152–198, Bakker 1997, Clark 1997; cf. the overviews of Russo 1997, 2011 and Edwards 1997. Further, I do not, in contrast to Peabody (1975:97) and Parry (1971:275n1), attempt to limit the minimum length of formulae; nor do I attempt with Sale (1993:101) to define what is a “frequent” or “infrequent formula.” I do think that Erbse (1972:180) was unnecessarily hesitant about calling repeated elements formulaic (and so part of the tradition inherited by Homer).
[ back ] 52. See Shive 1987; on Shive, see Friedrich 2007:30–66, 128–146; see also Visser 1988 and Di Benedetto 1994:122–139.
[ back ] 53. Cf. the comments of Edwards (1988:27): “Because the epithets are chosen for metrical convenience does not mean that they lack meaning”; and Austin’s (2009) arguments about νεφεληγερέτα Ζεύς, extending arguments made by Friedrich 2007. Parry (Parry 1971:46) did show some awareness of this possibility (cf. Martin 1993:225). Particularized epithets, he noted, unlike generic epithets, designated, “a particular feature,” but the implications of this idea were not developed (cf. Amory Parry 1973:2–7). Lord (1960:66) thought further about this issue, however, on which see Dué and Ebbott 2010:10.
[ back ] 54. But see Whallon’s larger argument on pages 1–33. While Di Benedetto (1994:123) finds some of Whallon’s arguments “troppo generica,” I find his central thesis persuasive. See also the comments of Anne Amory Parry 1973:4, 163 et passim.
[ back ] 55. Di Benedetto (1994:123; cf. 109): “Il fatto che all’interno del poema un epiteto era riferito molto frequentemente a un singolo personaggio era sufficiente per creare negli ascoltatori il sensa di un collegamento reale ed inequivocabile tra quell’ epiteto e quel personaggio” (italics mine).
[ back ] 56. Whallon (1969:29–32) answers the question “Would epithets really apply as well to one man as another?” His examples of epithets’ appropriateness to particular characters (but inappropriateness to other characters) are telling. The point is made through imaginative, dissonant combinations of particular epithets. Imagine, he says, “Paris staunch in battle” (Ἀλέξανδρος μενεχάρμης), “Telemachos the spear-famed” (Τηλέμαχος δουρικλυτός), or “much devising Achilles” (πολύμητις Ἀχιλλεύς). He is not trying to say that every epithet bears significant qualities, but “only that many of them [do].” Cf. the comments of Vivante 1982:13–14 passim; Di Benedetto’s (1994:123) observations “di una tendenziale accentuata specializzazione” concerning epithets and his conclusions about Hector’s (126) and Agamemnon’s (129) epithets; and the arguments of Heath (2001) about Telemachos maturing into his oft-used epithet πεπνυμένος, “the ‘final’ version of the adult Telemachus who is hinted at with each speech introduction along the way.”
[ back ] 57. On the concept of a phrase as a “word,” see Foley 1999:201–237.
[ back ] 58. Dunkle (1997:228–233) has shown that even Achilles’ horses mirror this epithet; and that the traditional noun-epithet combination, “swift footed Achilles,” defines Achilles’ personality even when he merely supervises the foot and chariot races of Iliad 23.
[ back ] 59. Foley (1999:22) comments that we should discount “neither traditional nor situation-specific meaning.” Cf. the comments of Ledbetter (1993:490n20) about the importance of both the horizontal (intra-textual) and vertical (metonymic) readings of Homeric repetition. Anne Amory Parry (1973:2) had complained about her father-in-law’s work, that when it came to epithets (especially those deemed contextually inappropriate), “He never considers the possibility that it is the meanings traditionally assigned to such epithets that make them seem inappropriate.”
[ back ] 60. Cf. the earlier discussion of Sacks (1987:105–214). On choice between alternatives and the influence of the local context on the singer’s choice, see also Beck 1986, Friedrich 2007, and the important questions raised by Martin 1993:229.
[ back ] 61. Cf. Aphrodite as discussed earlier by Perrault, and the comments of Graziosi and Haubold 2005:52; cf. Martin’s comments about irony in Homer’s portrayal of Paris 1993:231–233. We will return to the subject of irony in Chapter 2.
[ back ] 62. See Finkelberg 2012 on the use of non-formulaic language. On the creative adaptation of formulae see Paraskevaides 1984, Richardson 1987:165–184, and Clark 1997:24. Di Benedetto (1994:117) shows that partial modifications of formulae are numerous (cf. 1998, passim), and he concludes that (1994:136), “l’atto creativo del poeta non si annulla pur nella utilizzazione di un sistema epitetico già preesistente”; Cf. the salient comments of Scodel 2002:19 and the experience of Murko 1928:337. The poet’s creativity is tradition-based, however. Further, John Foley once told me an anecdote of how, when carrying out field research, he witnessed a South Slavic guslar turn a grocery list into decasyllabic lines of South Slavic epic on the spur of the moment as he stood there. Cf. the effect of collectors on other epic traditions in Ready 2015:27–33. Yet, variation depends upon the importance of the tradition, rather than on novelty per se, as Dué (2002:1–2, 16–36, following Lord 1960:65) notes.
[ back ] 63. Beyond the shared traditional vocabulary (note the sensible comments of Bowra 1962:62–64), the changes in epic diction over time are mapped by Janko (1982:47), whose findings support the close dates of memorialization for the Iliad and Odyssey. I find Janko’s findings convincing for each of the (longer) epic traditions. For critical responses to Janko’s glottochronologic approach, see Fowler 1983, Jones 2010, Olson 2012:10–15, and Vergados 2013:142–145. On the similarity of the gods’ activities in the Iliad and Odyssey, see Reinhardt 1948:86–88 and Allan 2006. Di Benedetto (1999:217) sees the role of the deities in the Odyssey as “semplificato,” except that Athena’s role is greatly expanded. He suggests that the difference between epics is one of emphasis, through expansion or contraction. For further similarities and differences between the Iliad and Odyssey, see Van Duzer 1996:313–318.
[ back ] 64. On the use and adaptation of formulae, see Hainsworth 1968 and Di Benedetto 1994:103–121, 177–231; on clustering, see Di Benedetto 1994:108–115 and Beye 2006:82; on metrical irregularities and other indications of the oral nature of the Homeric epics, see Janko 1990 and 1998: esp. 7–9. For other observations suggesting the oral nature of Homeric poetry, see Austin 2009:70–73. For the possibilities of influence by the scribe and collector in dictation and editing, see Ready 2015:13–45.
[ back ] 65. Cf. the earlier comments of Nagy (1979:42): “when we are dealing with the traditional poetry of the Homeric (and Hesiodic) compositions, it is not justifiable to claim that a passage in any text can refer to a passage in another text.” Cf. n. 21 in this chapter.
[ back ] 66. For a critique of Friedrich’s reaction to M. Parry, see Austin 2009:88–96; on the challenges and salient questions posed by the process of dictation itself, see Ready 2015:13–20.
[ back ] 67. Cf. the comments of Powell 2000:107 and Jensen 2011:290.
[ back ] 68. Parry 1971:450–451. On dictation, see M. Parry in A. Parry 1971:451, Janko 1990, 1998, Powell 1997, Haslam 1997:80–84, and Kelly 2007a:10–11. Ready (2015) reviews the variables in the process of dictation by comparing other oral and oral-derived epics and reminds us that neither Homer nor his patron possessed electronic recording equipment. His emphasis is therefore on pre-Parry collectors. Ready presents some intriguing possibilities for the dictation process and suggests a possible range of variables (and outcomes) for the undoubtedly changed performance and recording dynamics (speed, pause, input, and influence by scribe or collector, etc.). Nagy (1990b:52–81; 1996:69–77, 2002, 2004:36; cf. Bakker 2005) suggests that fixation took place through the performance tradition and occurred at the feast of the Panathenaia. Ready’s (2015:60–63 passim) recent work supports this as a possibility and reminds us of the importance of the term “oral-derived.” While I think it is less likely that the poets who gave us the Homeric epics were themselves especially literate, Foley’s work (2002:22–57), outlining the diversity of worldwide oral poets, has made me more agnostic on the issue, although the contrast made by Austin (2009:92–96) between Homeric and Virgilian lines must still be reckoned with.
[ back ] 69. Heubeck (1988:3–23), while accepting an oral background for the Homeric epics, privileges writing. Cf. Friedrich 2007:140–144. For suggestions of the Odyssey’s allusions to the Iliad, see Rinon’s (2006:209–211) discussion of “mise en abyme” and the first song of Demodokos in Odyssey 8. For the history of “mise en abyme,” see Van Duzer 1996:45n110.
[ back ] 70. Cf. similar conclusions by Olson 1990:5, Danek 1998:2, cf. 12, Dué 2002:12–17 passim, Scodel 2002:21, Lentini 2006:13–14, and Barker 2009:59; but also the neoanalyst view that some oral version of Cyclic epics predated the oral version of the Homeric epics (Burgess 2001:33). For an extended discussion of “cycle,” see Fantuzzi and Tsagalis 2015:1–7; see 28–34, on the creation of the term “epic cycle.” On the difficulty of arriving at definitive conclusions about the “Epic Cycle,” see Burgess 2015a.
[ back ] 71. Lord, as Dué (2000:21n2) points out, preferred to speak of Homeric tradition as being “multi-form,” rather than containing “variants.”
[ back ] 72. ὑπέρμορα νόστος ἐτυχθη, Iliad 2.155. The Iliad, unsurprisingly, ostensibly foregrounds little of the post-Trojan War events. Much of the audience, however, would have been aware of “what happened later” all the same. This lack of foregrounding is necessary, nevertheless, since the plot of the Iliad itself is set before the actual return of Agamemnon. Consequently, when, through his embassy, Agamemnon offers Achilles a home in Pylos (Iliad 9.291–295), he is acting as anyone would who was not yet aware that he would fall into the snare of his unfaithful wife upon his arrival home. There is a fiction that is kept up here. After all, the Iliad aoidos, unlike the aoidos of the Odyssey, is not singing a nostos tale. There may be oblique references to the Odyssey story, however, in such instances as Priam’s reply to Helen in the Teichoskopia, where Odysseus is likened to a ram (cf. Iliad 3.197 and Odyssey 9.432, etc.). Cf. the observations of Maronitis 2004:133–146. We will consider other potential references in Chapters 3–5.
[ back ] 73. Nagy (1979:42): “it is not justifiable to claim that a passage in any text can refer to another passage in another text”; Di Benedetto (1994:103): “È certo che il destinatario dell’ Iliade era un pubblico che il poeta raggiungeva esclusivamente attraverso una comunicazione orale”; cf. Scodel 2012:27, Barker and Christensen 2014:250, Kelly 2012, and Martin 2013. On the question of literacy, see Enos 2002 and Wilson 2009.
[ back ] 74. For a consideration of editing by the collector or scribe involved in the dictation process itself, see Ready 2015.
[ back ] 75. We need not be concerned that the language of our present copies of the two epics, as Janko (1982, 2012) has argued, makes the Iliad a bit earlier than the Odyssey. His findings are not meant to address oral Ur-forms. Further, his findings suggest relative memorialization in writing, rather than the actual date that the Iliad was recorded in printed form. Cf. n. 63 in this chapter.
[ back ] 76. Cf. the variable order of Palnadu and Sirat Bani Hilal epics, which only kept a strict chronological order when versions were artificially induced (Jensen 2011:37–41); and also Dué’s (2002:32–36) findings for multi-forms of the Briseïs tradition in the iconographical tradition.
[ back ] 77. Cf. Malkin (1998:52), who also suggests that “episodes from Homer were sung separately” in the fifth century. He notes as evidence Herodotus 1.116, Thucydides 1.110, and Plato Ion 539b. What exactly is meant by Plato’s later comments about Hipparchus demanding that the rhapsodes “run through” Homer’s poetry “alternately in order” (Hipparchus 228b–c: καὶ ἠνάγκασε τοὺς ῥαψῳδοὺς Παναθηναίοις ἐξ ὑπολήψεως ἐφεξῆς αὐτὰ διιέναι, ὥσπερ νῦν ἔτι οἵδε ποιοῦσιν) is not clear. In any case, it is really quite impossible to know whether Plato reflects earlier or just contemporary conventions when he makes his temporal comparison ὥσπερ νῦν ἔτι οἵδε ποιοῦσιν. Martin (2015:21) suggests “on-demand” performance of epic episodes; and Burgess (2012:286; cf. 2001:111) sees evidence in early Polyphemos iconography that “episodes from the Odyssey may have circulated independently.” On the existence of various travel narratives in the Odyssey, see Burgess 2012.
[ back ] 78. See Beck (1964:72–146), who suggests that Homer became a standard “school” text. On early schools, see Herodotus 6.27, Thucydides 7.29, Aristophanes Clouds 961–983, Aristotle fr. 233 PCG, Plutarch Life of Alcibiades 7.1.
[ back ] 79. In this regard, the Odyssey itself begins near the end of events, with the devastation that has inflicted Odysseus’ house during his absence (books 1–2), and moves back into events of the past through various retrospective narratives of the Trojan War (books 3 and 4, pre-nostos tales), such as those told by Nestor and Menelaos to Telemachos, and by Odysseus himself of his own absence during his nostos (books 9–12). On types of retrospective narrative (prolepsis and analepsis), see Genette 1980:48–85; cf. De Jong 2007a:3–8. On epic time, see also De Jong 2007b:17–37. Cf. Foley (1995:115–142) on the chronology of Odysseus’ account to Penelope in Odyssey 23.310–341.