To the memory of C. J. Ruijgh and C. M. J. Sicking


Epic is concerned with the past. It depicts heroes that are larger than life and accomplish their exploits in a bygone age outside the reach of ordinary mortals. Often, and certainly in the case of the Homeric tradition, epic’s very language is a representation of the past, in the form of vocabulary, syntax, and grammar that have long vanished from the poet’s everyday vernacular. Epic not only remembers the heroic past, but also its own, linguistic, past.
The Homeric poems, like most epics that are amenable to literary appreciation or analysis, are texts that have attained canonical status in a literary tradition. Such a glorious survival is itself an epic accomplishment, which we may attribute to the poems’ literary quality and continuous capability to repay careful reading. Yet written reception, essential and inevitable as it is, may blind us to certain features of these erstwhile oral compositions whose audience were listeners at a performance. Reading the Iliad and the Odyssey, studying them as texts, may alter the peculiar relation these poems once had with the past. It turns into a reified memory of the remote past what once was deliberate activity in the present. And it shifts the burden from the act of reference to the object of reference, from the present to the past.
The chapters that follow attempt to reverse this shift; they argue in various ways for the importance of the act of reference that realizes the presence of the epic past in the present of the performance. The epic events in the past are inseparable from the speech events in the present. When we shift the perspective in this way, epic stops being a mere sign that points to the past; epic in action is something more dynamic and agentive: a performer’s pointing at a past in the very act of its being created in the present.
Such pointing requires the epic poet to use in his narrative the deictic markers with which ordinary speakers situate themselves in time and space. An analysis of the deictic features of characters’ discourse, the mimetic impersonation of the gods and epic heroes, serves accordingly as a basis for an account of these same features in the discourse of the narrator at the level of the performance. Even though I acknowledge the narratological differences between narrative and characters’ speech, I still find that stressing the similarities between them in some key respects is the best way to arrive at the distinctive quality of Homeric narrative.
The deixis performed by the Homeric narrator and his characters yields hard textual evidence (demonstrative pronouns and markers of temporal deixis in particular) that can be understood entirely within the text, since the reality to which they point is created in and through the text: in contrast to other performed poetic genres, such as choral lyric, epic is self-sufficient in its deixis. No reference is made to specifics of time or place. Epic’s here and now is the moment of its performance; it is instantiated whenever the singer starts singing.
A study of the grammar of deixis may also alter our ideas of the past of epic itself, the history of its language. The relation of the Homeric poems with their past as many scholars see it is, again, a matter of a non-agentive, impersonal “pointing to”: the poems are thought to carry previous stages of the epic tradition within them, in the form of, e.g., Aeolic elements that are typically analyzed as belonging to a previous Aeolic phase. Such forms may also be located in the poet’s present, in the utilitarian perspective created by Milman Parry’s conception of oral composition: they are useful metrical alternatives to words in the poet’s contemporary Ionic and will be retained as long as no Ionic metrically equivalent form is available. The present study suggests a third way in which epic can position itself with respect to its past. If the performance is a deliberate strategy to summon the past into the present, then the language used to achieve that goal may be part of that strategy: the past is summoned to the present in the form of a deliberate use of language that is part of the linguistic past of the poet’s contemporary vernacular. In fact, the language of deixis itself, as we will have occasion to see, involves linguistic features that are not used according to contemporary Ionic grammar.
The chapters that follow have originally been written over a period of some fifteen years and are presented here, in revised form, roughly in the chronological order in which they have been first published. They reflect the changes in my own interests, but, more significantly, they also reflect a broad change in the way in which we have been thinking about Homer in terms of “oral poetry,” a change that took us from “formula” to “performance” as the central concept in the historicizing study of Homer. Neither concept excludes the other in any way—indeed, they are interdependent—but they still can be used in such a way as to represent very different perspectives. Whereas “formula” typifies the production, the composition of the epic tale (and so came to characterize epic as qualitatively different from other, nonformulaic, poetry), “performance” is a hermeneutic tool that can be used for the study of the presentation and reception of Homeric narrative and the relation it entertains with the reality it evokes. In the present study “performance” is developed into a critical paradigm that brings to bear various aspects of spoken language (e.g., speech acts and deixis) on our understanding of Homeric epic poetics.
Chapters One and Two represent my personal situation before the formula-performance shift. Both pieces seek to modify and fine-tune Milman Parry’s original concept of “formula” in order better to account for the poems’ oral composition and semantic richness. Chapter One [1] places the ornamental element in the formulaic diction (best exemplified in the epithets studied by Milman Parry) in a wider framework of “peripheral semantics” and argues for a compositional strategy by which the redundancy that is inherent in ordinary language can be systematized and stylized for the sake of versification. The chapter studies in detail the expressions for “with the spear” as peripheral with respect to a “nuclear” verb of killing or hitting in the same way that the epithet is peripheral with respect to its name. Chapter Two, [2] working on much the same material, addresses the phenomenon of synonymy in Homeric diction, which according to Parry’s model plays an important role in epic formulaic diction: functional synonymy is often seen as a factor enhancing the diction’s adaptation to oral-formulaic composition (the same meaning can be conveyed in metrically different ways). I argue, however, that in epic diction synonymy is just as elusive as in the ordinary language of real life: there are always semantic factors over which the system has no control.
In Chapter Three [3] I offer a transition from “formula” to “performance” that starts with a discussion of the bias inherent in the term “oral” as it is used in Homeric studies: “oral poetry” and “oral composition” are literate concepts. The background against which I propose to study Homer is not “poetry” but “speech,” a concept that involves cognitive phenomena such as consciousness and memory. Understanding Homer as a stylized flow of ideas passing through the performer’s consciousness has consequences, I argue, for Homeric syntax, formulas, and verse, and their joining in enjambment. Chapter Four [4] continues this perspective in a rereading of Erich Auerbach’s well-known characterization of Homeric style as “external” and “foregrounded.” I suggest that Auerbach’s criticism is more in line with Homeric poetics than recent literary criticism has acknowledged, provided we replace “text” with “speech.”
Chapter Five [5] turns from the introverted operations of consciousness and memory to the performance as an extroverted pointing to the reality it produces. I will point in particular at the demonstrative pronoun οὗτος (houtos) whose use and distribution in Homer is different from that in Classical Greek. Being exclusively used in Homer to “point” at someone or something in a speaker’s immediate present, the pronoun is most at home in the speech of the characters of the epic tale. This makes the infrequent occurrences of this deictic element in the discourse of the Homeric narrator all the more significant, and the chapter discusses these cases as moments that typify the nature of the Homeric performance.
Chapter Six [6] introduces time and temporality, a dimension for deixis that will remain a central interest in the remainder of the book. Homeric epic, I suggest, is less interested in locating the epic events in time than in positioning itself, its own present, as the moment of activation of the past. The present can thus be seen as the past’s future, a moment of recognition that can be projected by the epic character. I argue that epic’s narrative present can be seen as a moment of understanding that is modeled on the way in which characters in the tale reach insight about the past by drawing conclusions from the physical evidence before their eyes. In the interplay between past ignorance and present insight (at the level of the performance a source of dramatic irony) two “evidential” linguistic items, the particle ἄρα (ara) and the verb μέλλειν (mellein) play a central role.
Chapter Seven [7] is a study of verbal augment against this background and at the cross-roads of linguistics and poetics. I believe that Homeric narrative offers us a glimpse of a function of this element of verbal morphology prior to its generalized use as marker of past tense. Augment’s original function, I argue, was to relate an event to the speaker’s speaking present, either by marking its proximity in time or its continuing relevance for the present. The fact that aorists in similes are without exception augmented provides a new opportunity for addressing the relation of epic with its own linguistic past.
The results obtained so far are used in Chapter Eight [8] for a new interpretation of the problematic opening scene of the Homeric Hymn to Apollo, which I propose to read as the verbalization of a vivid mental image. The Hymn’s opening phrase “I shall remember” (μνήσομαι) is taken as point of departure for a discussion of the importance and nature of remembering by characters within the epic, which is then used, as earlier in the study, to typify the performance as such.
Chapter Nine brings together tense and deixis in a comparison of Homer and Thucydides in terms of enargeia (“vividness”). When speakers adopt a deictic vantage point with respect to a reality that is imagined, they can either imagine the imagined reality close to them or themselves close to the imagined reality. Greek narrative and Greek grammar, I argue, have in the opposition between the imperfect and the aorist an essential tool for the articulation of this difference. A discussion of Thucydides’ use of the imperfect in narratives that unambiguously displace the reader in the imagination to the past is used as foil for the demonstration that Homer does the very opposite. Epic can thus be contrasted as a narrative whose now is its own narrative present with narratives whose now is a shift to a vantage point in the past.
Finally I wish to express my gratitude to Gregory Nagy for his invitation to bring out this collection and for his unfailing enthusiasm and encouragement throughout. I also thank Leonard Muellner for expert and prompt help with styles, conversions, and electronic publishing and Casey Dué for perceptive copy-editing. The book’s dedication expresses my debt to two Dutch scholars who, in their very different ways, contributed through their teaching to my own thinking about Greek language and literature. It is sad that the power of memory had to be invoked so soon.
New Haven, Connecticut, April 2005


[ back ] 1. Originally published as “Peripheral and Nuclear Semantics in Homeric Diction. The Case of Dative Expressions for ‘Spear’” (with Florence Fabbricotti). Mnemosyne 44 (1991), 63–84. Permission to reprint from Brill Academic Publishers.
[ back ] 2. Originally published as “Aspects of Synonymy in Homeric Formulaic Diction: An Investigation of Dative Expressions for ‘Spear’” (With Nina van den Houten:): Classical Philology 87.1 (1992): 1–13. Permission to reprint from The University of Chicago Press.
[ back ] 3. Originally published as “How Oral is Oral Composition?” in Signs of Orality, ed. E. A. Mackay, 29–47. Leiden 1999: Brill. Permission to reprint from Brill Academic Publishers.
[ back ] 4. Originally published as “Mimesis as Performance: Rereading Auerbach’s First Chapter.” Poetics Today 20 (1999): 11–26. Permission to reprint from Duke University Press.
[ back ] 5. Originally published as “Homeric ΟΥΤΟΣ and the Poetics of Deixis.” Classical Philology 94.1 (1999): 1–19. Permission to reprint from The University of Chicago Press.
[ back ] 6. Originally published as “Storytelling in the Future: Truth, Time, and Tense in Homeric Epic.” In Written Voices, Spoken Signs: Tradition, Performance, and the Epic Text, ed. E. J. Bakker and A. Kahane, 11–36. Cambridge 1997: Harvard University Press. Permission to reprint from Harvard University Press.
[ back ] 7. Originally published as “Similes, Augment, and the Language of Immediacy.” In Speaking Volumes: Orality and Literacy in the Greek and Roman World, ed. J. Watson, 1–23. Leiden 2001: Brill. Permission to reprint from Brill Academic Publishers.
[ back ] 8. Originally published as “Remembering the God’s Arrival.” In Epos and Mythos: Language and Narrative in Homeric Epic, ed. M. Malamud and C. Higbie, 63–81. Arethusa 35 (2002): 63–81. Permission to reprint from the Johns Hopkins University Press.