James Bradley Wells, Pindar's Verbal Art: An Ethnographic Study of Epinician Style
Introduction. Philology as Perspective on the Interaction of Language and Social Life
1. Text and Sign
2. Epinikion as Event
3. Ways of Epinician Speaking I
4. Ways of Epinician Speaking II
5. Novelistic Features of Epinician Style
Introduction. Philology as Perspective on the Interaction of Language and Social Life 
Proverbially difficult  and commissioned by the elite, Pindar’s victory songs have the reputation of being “high art.” Epinician song, however, served the interests of the elite rulers and families whom it commemorated, not solely by becoming symbols of exclusive power and prestige, but by making itself and the laudandus’s achievement a popular possession through the inclusive appeal of entertainment, folk wisdom, and even laughter.  Accordingly, Pindar’s Verbal Art lays the philological foundations for approaching Pindar’s epinikion as a form of popular entertainment. My subtitle, “An Ethnographic Study of Epinician Style,” aligns my approach to Pindar’s stylistics with that of scholars in the overlapping fields of linguistic anthropology, folklore, and oral tradition studies. As this allegiance may suggest, speech and performance are the fundamental descriptive and interpretive foci of Pindar’s Verbal Art. Since the central findings of this book—Pindar is an oral poet; the epinician text records a speech (not text) event; epinikion is a centrifugal, stylistically diverse form of traditional art—hinge upon a rigorous, ethnographic description of Pindar’s language, this Introduction addresses questions of philological methodology—where we have been in the study of epinician performance and where this study will go. In other words, what makes Pindar’s Verbal Art a new and necessary contribution to scholarship on Archaic Greek poetry and poetics?  I will address this question, first, in terms of linguistics and, second, by comparing my ethnographic approach to Pindar’s style to current studies of epinician performance.
Existing scholarship describes epinician language primarily in terms of grammar (e.g. Des Places 1947, Hummel 1993) or content (e.g. Bischoff 1938, Lefkowitz 1963, Bowra 1964, and Boeke 2007) or some combination of the two (e.g. Forssman 1966, Hamilton 1974, Race 1990, and Mackie 2003). From the point of view of linguistics, Pindarists’ focus on grammatical form and textual content privileges the referential function of epinician language over its poetic function. This characterization draws upon Roman Jakobson’s model of communication (1960), which linguistic anthropologist Dell Hymes (1974:10) incorporates into the Ethnography of Speaking.  Jakobson identifies the following “constitutive factors in any speech event, in any act of verbal communication”: addresser, message, addressee, “a context referred to,” a code, and contact, “a physical channel and psychological connection between the addresser and the addressee, enabling both of them to enter and stay in communication” (1960:353). Jakobson associates each of the constituent factors of a speech event with a particular function, respectively: emotive, poetic, conative, referential, metalingual, and phatic (1960:354–357; cf. Hymes 1974:146). To summarize the relationships between a constitutive factor of a speech event (in CAPS) and its function (lower case):
CONTEXT/referentialIn contrast to the referential function of language, Jakobson defines the poetic function as “[t]he set (Einstellung) toward the message as such, focus on the message for its own sake” (1960:356; cf. Jakobson 1956:84–85 and Waugh 1980:58).  Linguistic anthropologists Richard Bauman and Charles Briggs explain that “the poetic function manipulates the formal features of the discourse to call attention to the formal structures by which the discourse is organized” (Bauman and Briggs 1990:73). Since “[t]he verbal structure of a message depends primarily on the predominant function” (Jakobson 1960:353) and since Pindar’s songs are forms of verbal art, we should expect that, in terms of Jakobson’s model, the poetic function is dominantly constitutive of epinikion.  Since, as Bauman and Briggs go on to say, performance is “the enactment of the poetic function,” the appropriate focus for a study of epinician style, which embraces performance and speech, is the poetic function of language (Bauman and Briggs 1990:73).  In Pindar’s Verbal Art the referential function of language, which existing studies of epinician language privilege, is subordinate to the poetic function of language. Such methodological appropriateness inheres in the ethnographic thrust of Pindar’s Verbal Art.
I can further qualify how Pindar’s Verbal Art differs from the existing studies of epinician language in terms of the three subfields of linguistics: (1) syntactics is the study of the relationships among signifiers; (2) semantics is the study of the relationships between signifiers and signifieds; and (3) pragmatics is the study of the relationships among signifiers and the people who use them.  Whereas approaches to Pindar’s language that depend upon grammatical form methodologically foreground syntactics and those that depend upon message content foreground semantics, the ethnographic stylistics of Pindar’s Verbal Art is firmly committed to the perspective of linguistic pragmatics.  Although such a stylistics necessarily cannot be purely ethnographic in the sense that it involves participant-observation field research or in situ documentation of communicative practices, linguistic pragmatics offers a set of descriptive and interpretative resources that enable us to penetrate the silence of the text-artifact and adopt the perspective of the community members who participated in epinician performances.
The difference that foregrounding the poetic function of language and rigorously applying the methods of linguistic pragmatics makes for our understanding of Pindar’s art will become clear as this study progresses. For the moment, I hope that this Introduction so far provides a theoretical foothold for understanding what I mean when I claim that I explore epinician performance from an ethnographic perspective, whose descriptive and interpretive charter embraces the methodological criteria of inductive analysis, culture-specificity, and the application of analytical controls that ensure, to the degree possible, an agent-centered approach to analysis of the epinician text. To locate such an ethnographic perspective vis-à-vis existing scholarship, I will briefly review the three most important developments in recent studies of epinician performance: (1) a series of articles that I refer to as the choral-monody debate (Lefkowitz 1963, 1988, 1991, and 1995; Heath 1988; Burnett 1989; Carey 1989 and 1991; Heath and Lefkowitz 1991; and Morgan 1993); (2) Hilary Mackie’s book Graceful Errors: Pindar and the Performance of Praise (2003); and (3) New Historical studies of Pindar and epinikion (Kurke 1991, Dougherty 1993, and Nicholson 2005). The common denominator of these various studies of epinician performance is methodological, to approach the context of epinician performance in an outside-in, deductive fashion: after defining context, Pindarists interpret epinikion in terms of how context so conceived influences epinician performance. By contrast, Pindar’s Verbal Art promises to present an inside-out, inductive description of epinician speech and performance that will embrace Pindar’s entire epinician corpus.
In the choral-monody debate there are two recurrent analytical movements: first, to reconstruct a hypothetical contextual backdrop of historical events and circumstances putatively associated with the composition of an individual victory song; second, to interpret individual passages or songs on the basis of such a reconstructed contextual backdrop. These two analytical movements constitute literal exegesis as defined by Tzvetan Todorov in Introduction to Poetics: literal exegesis is a method of literary study that “consists in elucidating the meaning of such and such an incomprehensible word, in supplying the references for such and such an allusion, in explicating such and such a syntactical construction” (1981:xxii). Todorov identifies two branches of literal exegesis, “one linguistic and the other historical” (1981:xxix). Linguistic analysis “distinguishes the true from the false,” and “[t]hesame is true of historical analysis” (1981:xxix–xxx).  Scholars participating in the choral-monody debate apply these branches of literal exegesis to the word kômos ‘revel’.  Burnett (1989) and Carey (1989 and 1991), who argue in favor of a choral epinician performance, maintain the view handed down by ancient commentators, that kômos refers to a group of performers and, so, to a chorus.  Heath and Lefkowitz, who advocate for the possibility of a monodic mode of performance, claim that kômos identifies not specifically epinician performers or performance but the celebration of the athlete’s victory in general, one component of which would have been the performance of the victory ode (1991:176). If we sever the long-assumed connection between kômos and a chorus, then, according to scholars in the monody camp, a solo performance of epinikion is possible. For both sides of the debate, the analytical process involves an outside-in approach to context: first, historical analysis in the form of reconstructing a conjectured sequence of events in order to define kômos (see especially Heath 1988); second, linguistic analysis in the form of interpreting individual passages of epinikia on the basis of that reconstructed context. 
To turn to more recent scholarship, Hilary Mackie’s Graceful Errors is the first book-length study of epinician song as oral performance (2003). Although Mackie does not methodologically define “the perspective of oral performance” (2003:2) that she adopts, it is possible to glean what her perspective is on the basis of her treatment of social interaction between composer and audience. Mackie approaches genre and convention from an outside-in perspective, from which the relationship between Pindar and his audience is a matter of fulfilling prefabricated roles: “I interpret the persona constructed and projected by the epinician poet as a convention specific to the genre and its function. I read the concerns the narrator expresses (both directly and indirectly) about the responses of his audience the same way” (2003:3). By formulating the performer-audience relationship in such terms, Mackie seems to conceive of human interaction, much like break-offs (9–37) and wishes and prayers for the future (77–106), as a category of formal convention rather than an intersubjective and emergent process.  Interpretation, then, is a matter of identifying how passages or songs reflect Mackie’s analytical template. What proves to be missing in Mackie’s study is Pindar’s point of view, which surfaces when we apply linguistic pragmatics and, more generally, ethnographic methods to the study of epinician speech and performance.
New Historicist Leslie Kurke has revolutionized how students of epinikion understand the contextual backdrop for the art form, and the work of Carol Dougherty and Nigel Nicholson further proves how productive New Historicism can be for our understanding of the social functions of Pindar’s victory songs. In her important book The Traffic in Praise Kurke demonstrates that “[t]o understand the social dimension of Pindar’s odes we must make sense of their economics, for he composed at a time when the economy was largely embedded in various social, political, and religious structures and institutions” (1991:7). To illustrate how Kurke’s study informs our understanding of the context for epinician performance, she draws a cultural map for an athlete’s departure and return that spatially delineates a loop whose starting and end point is the athlete’s house, or oikos (1991:15–82). Participation in athletic competitions that figure in Panhellenic religious festivals implicate, not only the athlete’s oikos, but also his polis and social class, institutional and ideological domains that the athlete’s loop of departure from and return to his oikos embrace. Kurke’s cultural map also charts how the athlete’s return creates a temporal loop that recycles the past kleos ‘renown’, which distinguishes his oikos, as well as augmenting the social capital of kleos in the present and for the future. The athlete’s return involves reintegration, a process of simultaneously welcoming the athlete to his hometown and actualizing the trade in social capital.  Since the traffic in praise embraces the domains of the polis and the aristocracy, the returning athlete must be reintegrated into his oikos, his polis, and his social class. This process of reintegration provides a contextual backdrop for identifying the social function of epinician performance: “it is the task of Pindar’s poem in performance to accomplish this reintegration.”  In an outside-in manner, Kurke’s elegant analysis locates epinikion as a coordinate on a cultural map and explains how the economics of aristocratic gift-exchange conditions the traffic in praise. 
Carol Dougherty’s important study of colonial discourse, The Poetics of Colonization (1993), has provided another view of the contextual surround for the composition and performance of epinikion. Colonial discourse is characterized by a pattern of colonial narrative, whose elements include crisis, Delphic consultation, colonial foundation, and resolution (1993:15–30), and by the cultural metaphors of purification (31–44), riddling (45–60), and marriage (61–80). Given this definition of colonial discourse, Dougherty explores “the extent to which the narrative pattern, cult, and metaphors of archaic colonization participate in the rhetoric of drama and the epinician ode” (10). For Dougherty, then, the performance of epinician song provides a public forum for transmitting colonial discourse, and “[t]he poet’s role is that of teacher; his poetry transmits a society’s ethical value system, and the community participates in this civic education both as audience and insofar as it is represented in the composition of the chorus itself” (95–96). On the basis of this outside-in approach to epinician performance, Dougherty’s discovery of strategies for representing the ideological armature of colonization provides a context for understanding epinikion as a medium of colonial discourse.
Nigel Nicholson’s book Aristocracy and Athletics in Archaic and Classical Greece (2005) is a trenchant, interdisciplinary study of how victory memorials come to represent aristocratic ideology. Nicholson confronts the fact that, “although these victory memorials could speak of the victors at length, they were almost entirely silent concerning the drivers, jockeys, and athletic trainers who helped them win” (2005:1). The practice of commemorating athletic victories with a memorial provided a medium for representing those achievements “in ways that made them accord with aristocratic ideology” (2005:3).  Against the backdrop of ideology and exchange systems, victory memorials, like Pindar’s epinician songs, “proved an excellent tool for fixing the meaning of victory to [competitors’] advantage” (2005:11). Nicholson’s study, then, enables us to interpret epinician performance from the outside-in as one medium among several for memorializing athletic victories. 
As David Young has written, “we usually search history to explain Pindar, not vice versa” (1968:113). New Historical approaches to the study of epinician performance give us insight into what Bauman describes in his Verbal Art as Performance as the “context of performance as a generalized cultural system in a community” (1977:37). Indeed, Kurke’s description of epinikion as a “communal drama” is the premier example of approaching epinikion at this level of context (1991:257–262).  To draw from Pierre Bourdieu’s distinction between structured structures and structuring structures, however, the innovative scholarship of Kurke, Dougherty, and Nicholson describes how epinician performance is structured by the cultural schemas entailed in ancient Greek athletics and their commemoration in the victory song, but does not explore the communicative strategies that structure the event of epinician performance itself.  In terms of Jakobson’s model of a speech event, by treating the epinician text as a discursive realization or ramification of its contextual surround, outside-in approaches to the study of epinician performance implicitly assume the dominance of the referential function of language in the constitution of the speech event that the epinician text records. As Bauman goes on to say, “one wants to be able to appreciate the individuality of each [performance], as well as the community-wide patterning of the overall domain” (1977:37). My approach to the context of epinician performance looks at this more local level of “the emergent quality of performance” (Bauman 1977:37–45), where the focus is upon the verbal means used to produce a unique work of verbal art in the situated context of interaction among participants in the event of performance.
An inside-out understanding of epinician performance depends upon a method of philology that collaborates with a text artifact in order to describe and interpret the cultural practices that the artifact records.  Through a performance-centered, ethnographic approach to our textual artifacts of epinician song, fortified by linguistic pragmatics, it is possible to discover how Pindar’s art represents its ambient cultural world and the persons who populate it. To pursue this ideal, I adopt the overarching principle of dialogue as a methodological premise. In the first instance, epinician art entails dialogical relationships among participants in the event of epinician performance. This site of communicative practices is the ultimate object of my analysis. Second, I approach the various analyses constituting the whole of my study as vantage points comprising dialogically a composite picture of Pindar’s art; each dimension of analysis depends upon and informs others. Finally, by making explicit my methodological predispositions, I hope to enable a dialogue of shared inquiry. Such explicitness often takes the form of substituting methodological statements for the more conventional review of secondary literature on specific Pindaric questions. As mentioned, the strategy that I adopt to achieve these ends is to develop an ethnographic description and analysis of epinician song. In Hymes’s terms, I conceive of authentic understanding of Pindar’s art as a matter of “intersubjective objectivity”: “[e]thnographic objectivity is intersubjective objectivity, but in the first instance, the intersubjective objectivity in question is that of the participants in the culture” (1974:11).  The methodological ideal of intersubjective objectivity captures my analytical focus upon approaching context from the inside, adopting an “agent-centered line of analysis” (Bauman 1992:142) and pursuing a philological description of communicative resources that shape “the emergent quality of performance” (Bauman 1977:37–45).
Pindar’s Verbal Art is principally a study of epinician style. In his essay “Ways of Speaking” Hymes demonstrates that culture-specific patterns among the sociolinguistic features of speaking—for example, participants, linguistic code, rules for appropriate speaking, and social functions of speech—constitute ways of speaking, or styles, understood in terms of sociolinguistics: “one can characterize whatever [sociolinguistic] features go together to identify a style of speech in terms of rules of co-occurrence among them” (1989:434).  This conception of style has a crucially ethnographic orientation in that it refers to “a way or mode of doing something” (Hymes 1989:434); speech styles are ways of doing social life, and the study of style, conceived of from an ethnographic point of view, is concomitantly the study of human social practice. Thus I do not conceive of Pindar’s Verbal Art as a(nother) recapitulation of the long-standing quarrel between formalist (or just literary-critical) and historicist approaches to Pindar, so that I would be giving voice to the formalist cause.  First, I conceive of this project as more radical, as an attempt to reformulate fundamentally how we think about Pindar’s language and his entextualized records of epinician performance. Second, my method is radically integrative: since registers are “major speech styles associated with recurrent types of situations” (Hymes 1989:440), when we approach Pindar’s language as a way of speaking or register, a speech style dedicated for use in a particular context for speech practices, then we undertake the kind of linguistic analysis that is firmly grounded in practice and, so, history. 
Hymes’s ethnographic conception of stylistics informs the performance approach to the study of entextualized records of verbal art. In his seminal work Verbal Art as Performance (1977), Bauman imports the Ethnography of Speaking into a philological method for the description and analysis of artistic texts, highlighting the relationship between ways of speaking, or registers, and performance as a matter of descriptive method:Since Pindar’s victory songs survive as an entextualized record of the epinician register, that text also bears information about the speech event of epinician performance, the context to which the use of the epinician register is dedicated.  A crucial sociological dimension of the use of the epinician register is its artistic style of communication with culture-specific patterns of participant interaction. By treating Pindar’s language as a register, a sociological conception of style, the epinician text turns out to be a record of interaction among the participants in performance, so that it is possible to move analytically from entextualized words to historically embodied voices.
Performance, as we conceive of it…is a unifying thread tying together the marked, segregated esthetic genres and other spheres of verbal behavior into a unified conception of verbal art as a way of speaking. Verbal art may comprehend both myth narration and the speech expected of certain members of society whenever they open their mouths, and it is performance that brings them together in culture-specific and variable ways, ways that are to be discovered ethnographically within each culture and community. 
In broad strokes, I describe Pindar’s art in terms of what Hymes identifies as the “fundamental notions” of the Ethnography of Speaking: “ways of speaking, fluent speaker, speech situation, speech event, speech act, components of speech events and acts, rules (relations) of speaking, and functions of speech” (1974:45). Along the way I apply a new approach to the most salient problems in the study of Pindar and his art: genre, epinician performance, the relationship between epinikion and Archaic Greek poetic tradition, and the unity of the victory ode. Whereas “Ways of Speaking” is the dominant analytical perspective in the Ethnography of Speaking, “speech community” designates the fundamental object of analysis. Hymes stresses the speech community as a “necessary, primary concept” because “it postulates the unit of description as a social, rather than linguistic, entity” (1974:47).  In the intellectual-historical context of his book Foundations in Sociolinguistics, Hymes here formulates a counterstatement to structural linguistics. In an analogous vein, my ethnographic approach to epinician performance is intended as a counterbalance to outside-in approaches to the study of the context of epinician performance. Just as structuralism in the study of language runs the risk of generating theories and descriptions of language that do not account for what actually happens in social practice, literal exegesis and New Historicism depend upon structural rules that can over-determine how we interpret ancient texts.  Central to my study of epinician style and performance is to step back and reconsider: What is the primary field of cultural description in the study of Pindar’s art? Accordingly, in Chapter One, “Text and Sign,” I demonstrate that Pindar represents his art as a spoken form of communication. The communication rules that obtain in Pindar’s epinikia are specifically rules for speaking. My further description of the epinician way of speaking depends upon establishing conclusively that Pindar’s art represents itself as an orally transmitted and aurally received mode of communication. Analogous to the place of the analytical category of speech community in Hymes’s Foundations in Sociolinguistics, the first tier of my ethnographic study of epinikion, then, is to demonstrate that speech practices recorded in the epinician text are the primary object of philological analysis in the study of epinician performance.
In Chapter Two, “Epinikion as Event,” I move from the speech community stratum of analysis to that of speech situation and speech event. Hymes applies the concept of speech event specifically “to activities, or aspects of activities, that are directly governed by rules or norms for the use of speech” (1974:52). Speech situations, however, “may comprise both verbal and nonverbal events, and the verbal events may be of more than one type” (1974:52). The empirical record for epinikion is fuzzy here. We possess the text-artifact of epinician performance, which records the poetry, but so much more is missing: information about music, dance, the locations of epinician performances, and the events (e.g. procession or feast) associated with such performances. Due to the status of the evidence, I have decided to define the speech situation, the context of situation in which epinician performance occurs, in a circumspect way as the typical frame of epinician performance. Relative to such a frame, the individual epinician song is a framework, an actual speech event. Here I am applying linguistic anthropologist William F. Hanks’s working definition of frame and framework: “frames are relatively static resources defined at the level of schematic structure, whereas frameworks are dynamic productions defined at the level of local usage” (1993:128).  These definitions of frame and framework will apply at various points in Pindar’s Verbal Art. In Chapter Two I also explore the strategies by which epinikion delineates itself as a conventional speech situation through the use of certain performance keys.  I describe performance keys at the schematic level to show how the epinician way of speaking typically marks out the frame for epinician performance. On the basis of our understanding of the typical keys to epinician performance, it is possible to identify how they function in an individual epinician song to key the framework of a particular performance, a level of analysis where we are dealing with a speech event, a “level of local usage” of the epinician way of speaking, as Hanks defines framework.
In Chapters Three and Four, “Ways of Epinician Speaking I” and “Ways of Epinician Speaking II,” I describe the most stable and recurrent stylistic patterns occurring in Pindar’s epinikia—that is, the styles associated with, for example, prayers, mythological narrative, and the gnomic statement. Each of these stylistic patterns is a speech genre, in Mikhail Bakhtin’s terms, and I draw upon his essay “The Problem of Speech Genres” (1986:60–102) as a principal theoretical and methodological point of departure for my analysis of the ways of epinician speaking that constitute the epinician way of speaking. This tier of analysis provides a partial practical (versus syntactic, semantic, or referential) taxonomy of speech acts, components of speech events and acts, and functions of speech, to refer again to Hymes’s fundamental notions of the Ethnography of Speaking.  Such a taxonomy enables us, first, to develop a participant-centered analysis of epinikion and, second, to demonstrate that the epinician way of speaking is a composite of speech styles, the ways of epinician speaking.
What emerges from this description of the ways of epinician speaking is that epinician art, like Bakhtin’s characterization of the novel, is “a phenomenon multiform in style and variform in speech and voice. In it the investigator is confronted with several heterogeneous stylistic unities, often located on different linguistic levels and subject to different stylistic controls” (Bakhtin 1981:261). In Chapter Five, “Novelistic Features of Epinician Style,” I advance the thesis that, based upon the fact that the epinician way of speaking is a complex blend of registers, epinikion is novelistic discourse. Perhaps it is just a statement of the obvious to stress that this is not to say that epinikion belongs to the genre of the novel. Rather, I am identifying Pindar’s art as a novelistic form of discourse in order to challenge the more conventional perception that Pindar’s art is strictly a “high genre” possessed of “hierarchical distance” projecting “the events, victors and heroes of ‘high’ contemporary reality” into a “heroic past and tradition.”  Rather than being a high form of art characterized by distance from its own present, Pindar’s art form exhibits the basic characteristics that Bakhtin identifies with the novel (1981:11):In Chapter Five I will explore each of these stylistic dynamics in the course of a sociolinguistic analysis of Pindar’s Olympian 1.
(1) its stylistic three-dimensionality, which is linked with the multi-languaged consciousness realized in the novel; (2) the radical change it effects in the temporal coordinates of the literary image; (3) the new zone opened by the novel for structuring literary images, namely the zone of maximal contact with the present (with contemporary reality) in all its openendedness.
Although Pindar’s Verbal Art ventures a new approach to the study of Pindar and epinikion, my work does not proceed in isolation from existing Pindaric scholarship. Elroy Bundy’s Studia Pindarica exhibits the method-ological tenets of an ethnographic approach to the study of texts. Bundy urges the importance of understanding “the conventional aspects of choral communication,” indicating that he regards the epinician text as a record of practices organized by social rules for communication (1962:2, my emphasis). His observation that “a community of poets working within well-recognized rules of form and order” produced epinician art and, in general, his both praised and maligned emphasis upon genre, show that for Bundy epinikion is to be understood as a special way of communicating between a performer subject to critical assessment by an audience that was fluent in those “rules of form and order” (1962:3 and passim). Two of Bundy’s often-cited sources, Franz Dornseiff and Wolfgang Schadewaldt, greatly influenced the sociological dimension of Studia Pindarica. In the preface to his Pindars Stil Dornseiff stresses what Hymes would come to call “ways of speaking”: “Es gibt für veile griechische Dichter Arbeiten de genere dicendi , Programme über einzelne Tropen und Figuren” (1921:iii). Schadewaldt defines the scope of his work on the structure of epinikion as having stylistic-formal, objective-historical, and subjective-personal standpoints, but asserts that the stylistic-formal standpoint is the dominant consideration (1928:263).  Christian Gottlob Heyne, who produced a 1798 edition of Pindar, is the source for the earliest attested use of the expression “ways of speaking.”  Heyne observes, anticipating Hymes’s ethnographic conception of style, that ways of speaking, or registers, are defined by the events with which they are conventionally associated: “Hence follow all the precepts for the interpretation of mythic narrations—or the words, and sentences, and expressions of mythical usage. First, 1): we know that in very ancient mythic or symbolic ways of speaking, the senses and thoughts were communally uttered…” 
Beyond these traces of intellectual heritage shared by classical philology and linguistic anthropology, Bundy’s philological method corresponds in important ways to the performance method of approaching a verbal art form as a way of speaking. His innovation in the study of Pindar is to describe epinician language from the standpoint of literary genre. Genre and style are inextricably related in communicative practices. Accordingly, Bundy asserts the importance of pursuing “a thorough study of conventional themes, motives, and sequences in choral poetry—in short, a grammar of choral style that will tell us what systems of shared symbols enabled the poet and his audience to view the odes as unified artistic wholes” (1962:32). Whereas Bundy adopts a formalist approach to establish that epinikion has a functional unity comprised of conventional elements, I adopt a practice-centered description of epinician language. This difference can be seen as integrative: Pindar’s Verbal Art resumes Bundy’s project of developing a grammar of choral style.
Following Young’s advice that “we must learn the language” of epinikion, the notion of fluency perhaps best captures what I hope will be the ultimate outcome of my study (1968:107). In a dialogical fashion, fluency cuts both ways: as we develop our understanding of what makes for fluency from the point of view of participants in the event of epinician performance, we simultaneously become a more fluent audience to epinician art (cf. Bonifazi 2004a:392). I further maintain that the description of and fluency in the communicative means used by participants in epinician performance and recorded in the text-artifact methodologically precedes the study of the relationship between Pindar’s art and the cultural contexts in which it is embedded. For a practice-centered, ethnographically grounded philology, meaning is not a tautological function of the relationship between words on a page and their lexicographic essence; it emerges from the social actions—the deeds—that the speaking of those words entails. A corresponding mode of philology takes the description of communicative practices as its point of departure and acknowledges that meaning is socially situated and a product of human interaction. By adopting a practice-centered philology that looks to the event of social interaction, namely performance, between Pindar, the composer-performer of epinikion, and his audience, and by employing an empirical analysis grounded in linguistic pragmatics, we can discover the historical actualities implicated in the art form of epinikion, beginning crucially with its poetics understood as a special way of communicating that is associated with a particular event by ancient Greek cultural conventions. While the historical circumstances determining the current status of the evidence for epinician performance present obstacles to understanding the social practices entailed in that community event, a practice-based, participant-centered approach to the study of epinician language nevertheless enables us to discover how people interacted in the event of performance through the ways of speaking that they used. And when we can identify the actual social moves employed through the use of the epinician way of speaking by participants in the event of epinician performance, a description of epinician art as performance becomes also a historiographic project.
[ back ] 1. I programmatically adapt the title to my introduction from the title of Hymes 1972.
[ back ] 2. On Pindar’s perceived obscurity see Most 1985:11–25 and Hamilton 2003. Pfeijffer’s treatment of Pindar’s style hinges upon attributing the poet’s obscurity to the qualities of “implicitness,” an absence of explicit cues about how the variety of features in epinician song work together (1999a:23–34), and the “fiction” of extempore composition (1999a:34–41). I discuss Pindaric poikilia ‘stylistic variation’ in Chapter 5 and in Chapter 1 critically evaluate the claim that internal evidence for the extempore composition of epinician song is a fiction.
[ back ] 3. With choral performance of Pindar’s epinikia in mind, it is interesting to note that, as Burnett 2005:6n20 observes, Plato (Laws 654a) etymologically derives the word khoros ‘chorus’ from khara ‘joy’. A caution is in order: for Pindar laughter can be derisive, even annihilating, as at Isthmian 1.68. What I mean by “laughter” is suggested by Plato’s juxtaposition of khoros and khara, by Bundy’s description of euphrosuna ‘delight’ (1962:2), and by Pindar Pythian 8.85 (the absence of laughter implies here a regrettable absence of joy). Newman and Newman 1984:38–50 conceive of Pindar’s style as carnivalesque, tapping into the resonance between kômos ‘revel’ and kômôdia ‘comedy’ or ‘kômos song’ (cf. Nicholson 2001). Dickie’s study of hêsukhia gives a more tempered, even genteel, characterization of euphrosunê (1984). Robbins’s perception of an absence of festive mood in Pythian 3 supports his argument (1990) that the song is not a true epinikion.
[ back ] 4. Young 1970 is the benchmark study of the history of Pindaric scholarship. Most 1985:11–41 offers a fine critical survey. See also Lloyd-Jones 1973, Kopff 1981, and Heath 1986. Morrison 2007:5–7 reviews literature on epinician performance, concluding that upon closer inspection existing studies prove in his view to focus upon audience more than performance itself, and focus specifically on the audience to an original performance. Morrison’s project sustains this audience-centered trend, but expands the conception of audience to include “overlapping audiences,” which “have heard another Pindaric ode when they hear a given ode,” “secondary audiences,” “the audiences of reperformances of Pindar’s odes” (19) and “tertiary audiences,” which “include all later audiences of the odes, including readers and, ultimately, us modern Pindarists” (21). As my methodological statements below make clear, I see performance and audience as integrated, not “separable,” as Morrison maintains. I address the reperformance of Pindar’s epinikia in Chapter 5.
[ back ] 5. See Hanks 1996a:104 for the context of Jakobson’s model of language in the history of linguistics. Martin 1989:14–15 applies Jakobson’s model of a speech event to his study of Homeric speech acts.
[ back ] 6. Jakobson’s famous, more qualified definition of the poetic function is: “The poetic function projects the principle of equivalence from the axis of selection into the axis of combination” (1960:358).
[ back ] 7. To observe Jakobson’s important stress upon the relationality of components of speech events and their corresponding functions, I am in more accurate terms striving for a correspondence between the dominantly constitutive language function and the dominant mode of analysis. Thus Jakobson: “Poetic function is not the sole function of verbal art but only its dominant, determining function, whereas in other verbal activities it acts as a subsidiary, accessory constituent” (1956:85). For the concept of the dominant, see Jakobson 1935, and cf. Waugh: “Dominance presumes a hierarchization of functions, not an absolutization of functional differences” (1980:58).
[ back ] 8. Cf. Bauman’s observation that “performance may be understood as the enactment of the poetic function” (1986a:3).
[ back ] 9. I have adapted this delineation of subfields of linguistics from Hanks (1996a:57, emphasis in original): “The relation between the sign form and its designatum/denotatum is the focus of semantics. The relation between the sign form and the interpretant is the focus of pragmatics. The study of the relations among sign forms is syntactics.” Hanks 1996a:54–64 contextualizes Charles Morris’s theory of semiotics (1938), upon which he draws in the quoted passage. For the sake of simplicity I have recast the characterization of these relations in terms of signifier, signified, and sign-users.
[ back ] 10. Anna Bonifazi (2000, 2001, 2004a, 2004b, and 2004c) and Egbert Bakker (1997 and 2005) admirably model the application of linguistic pragmatics to ancient Greek poetics. Essays collected in Felson 2004a apply deixis to the study of Archaic Greek poetry, with significant focus on Pindar.
[ back ] 11. Note that Todorov’s characterization corresponds closely with Saussure’s criticism of grammar which “aims solely at providing the rules which distinguish between correct and incorrect forms,” as distinguished from the study of linguistic structure (1983:1). Saussure’s criticism of grammar is a particular criticism embedded in his general critique of philology (1983:1): because “[p]hilology seeks primarily to establish, interpret and comment upon texts” the philological method is “too slavishly subservient to the written language, and so neglects the living language.” The juxtaposition of Todorov’s characterization of exegesis and Saussure’s criticism of grammar is relevant because, from a historical point of view, literal exegesis in Pindar scholarship entails interpretive methods that predate and motivate Saussure’s counterstatement to grammar, which in turn motivates the counterstatement to structural linguistics that is represented by linguistic anthropologists who adopt an ethnographic approach to linguistics (see Gumperz and Hymes 1972 and Hymes 1974). In other words, the choral-monody debate employs descriptive and interpretative methods that predate major intellectual advances in the study of language, texts, and literature during the twentieth century.
[ back ] 12. Lefkowitz 1963 and Davies 1988 raise issues that inform the choral-monody debate, in which Heath 1988, Burnett 1989, Carey 1989 and 1991, and Heath and Lefkowitz 1991 engage. Kômos occurs fifteen times in Pindar’s epinikia: Olympian 4.9, 6.18, 6.98, 8.10, 14.16; Pythian 3.73, 5.22, 5.100, 8.20, 8.70; Nemean 3.5, 9.50; Isthmian 2.31, 6.58, 8.4 (cf. Slater 1969b:296–297).
[ back ] 13. Cf. Morgan 1993 and Mackie 2003:1n2. Burnett more recently differentiates khoros and kômos, contrasting the choral performance of a victory song with spontaneous celebration by “[i]n-formal bands of kômos-singers” on the same night of an athlete’s victory at the victory site (2005:6).
[ back ] 14. Of the phrase tonde kômos ‘this revel’ (Olympian 4.9, 8.10, 14.16; Pythian 5.22), Bonifazi writes, “this deictic utterance provides little objective information about either the inclusion of the comiasts among the performers (thus a choral hypothesis) or their exclusion (thus a soloist hypothesis)” (2004a:398). Cole captures both the kômos’ exuberant recklessness and the threat it poses, “not simply ‘epic exhilaration’ and excitement, but the confused mixture of triumph and let-down, apprehension and expectation to be found in a band of revelers that might have included, not simply members of a king or victor’s immediate circle, but invited guests, admirers from afar, foreign visitors, and possibly even rivals and enemies” (1992:11–32). Nagy 1994–1995:24 identifies the kômos as the prototypical occasion for the genre of epinikion. In his study of antecedents for the animal choruses of Greek comedy, Rothwell assembles literary and plastic testimonia for kômos practices (2007:6–35). Reflecting Rothwell’s findings about the relationship between the kômos of Archaic Greece and comic choruses, Lowe 2007:168–169 traces the meaning of enkômion from its original synonymity with comedy to its fourth-century BCE application to any form of eulogy.
[ back ] 15. Corroborating this characterization of Mackie’s view is her qualification of what she means by referring to “the poet”: “I am speaking of the constructed, fictional narrator of the odes—I do not mean to imply anything about the real live poet Pindar” (2003:3n8). In line with Mackie, for Kurke the performed quality of lyric poetry gives rise to “role-playing” in which the performer adopts “a more or less fictive position from which to speak” (2007:143). While I share the concern to avoid conjecture about the life and times of the historical Pindar, I disagree with the idea of treating Pindar’s agency as a constructed fiction; from an agent-centered point of view, Pindar as a traditional artist is “the real live poet Pindar.” As Bonifazi rightly explains, “from the pragmatic [i.e. linguistic pragmatics] point of view, the ‘I’ is never fictive” (2004a:398). See also Nagy 1994–1995:20 on the limitations of a “fictional” first person, instead of which he proposes the concept of a “generic I.”
[ back ] 16. Dougherty expresses the same view: “the victory song both celebrates the victor as he is welcomed home and orchestrates his reintegration” (1993:103). Cf. Crotty 1982:121, to whom Kurke 1991:6 directly responds.
[ back ] 17. Kurke 1991:6. On the athlete’s return see also Crotty 1982:104–138, Slater 1984, and Kurke 1993:139–140. Dougherty describes how “the victorious athlete was met outside his city and led as part of a magnificent procession through the main streets and into the center of the city, the agora” (1993:88). See also Dougherty’s description (1993:96) of epinician performance during the ceremony commemorating the victorious athlete’s return as a “ritual occasion.” Nicholson relates that “[t]he odes were performed mainly in the victor’s city of residence, although other locations, including the site of victory were possible” (2005:11). Fränkel earlier suggested such a scenario: “A suitable occasion for the public performance of epinician odes was the return of the winner to his city. The successful athlete entered in a joyous procession, and the chorus of friends and compatriots that escorted him sang the epinician to welcome him and sing his praises during the procession” (1973:429).
[ back ] 18. I note here that in a recent study of the ritual context for the performance of choral song, Kurke signals a shift in her thinking about context by addressing the “critical pitfall” of “the assumption that the context of ‘religion’ or ‘ritual’ is real, preexistent, and inert, while the literary text in relation to it becomes a mere secondary reflection thereof; or in other terms, that a literary text can be entirely explained by reference to a preexistent historical or ritual context” (2005:82).
[ back ] 19. Burnett 2005:51 is skeptical of the view that Pindar’s alleged categorical alignment with aristocratic ideology would have been an incentive to suppress the role of trainers in athletic victories.
[ back ] 20. Cf. Steiner 1998 and Smith 2007, who compare plastic and literary media for victory commemoration. Thomas 2007:152–163 describes the historical development of epinikion in the context of the broader development of victory monuments.
[ back ] 21. While outside-in approaches to Pindar are the norm, I have focused on those most compelling (the New Historicists) and most relevant to this study of epinician speech and performance. Krummen’s highly regarded book is also exemplary of this norm; she describes her study in the following way: “Ein erster Teil der Untersuchung [31–151] beschäftigt sich mit der Rekonstruktion dieses kultisch-festlichen Erwartungshorizontes… In einem zweiten Teil [153–266] wird diese rezeptionsorientierte Fragestellung auf die Eingrenzung der mythisch-kulturellen ‘Erwartung’ des Publikums erweitert; dies drängt sich bei den hier zur Diskussion stehenden Gedichten (Olympie 1 und 3) besonders auf, da ihre mythischen Erzählungen als ‘Neuschöpfungen’ Pindars gelten” (1990:9). Burnett prefaces a detailed study of each of Pindar’s Aiginetan odes (2005:55–238) with an admirably detailed presentation of the documented and mythological history of Aigina (13–28), discussion of the Aphaia Temple sculpture and practices associated with the monument (29–44), and Aiginetan patterns for participation in the crown games (45–54)—dimensions of context that inform her interpretation of individual songs in an outside-in manner. For other studies that opt for an outside-in view of epinician performance, see also Bernardini 1983:9, 76–77, and 80, Slater 1984, Clay 1999, Pfeijffer 1999a:2–21 (with critical discussion of secondary literature on the context of epinician performance), Currie 2005, Carey 2007, and Morrison 2007.
[ back ] 22. As an illustration of Bourdieu’s ideas here, I cite the following: “The structures constitutive of a particular type of environment (e.g. the material conditions of existence characteristic of a class condition) produce habitus, systems of durable, transposable dispositions, structured structures predisposed to function as structuring structures, that is, as principles of the generation and structuring practices and representations which can be objectively ‘regulated’ and ‘regular’ without in any way being the product of obedience to rules, objectively adapted to their goals without presupposing a conscious aiming at ends or an express mastery of the operations necessary to attain them and, being all this, collectively orchestrated without being the product of the orchestrating action of a conductor” (1977:73, emphasis in original).
[ back ] 23. Russian Formalism (Èjxenbaum 1927, Tynjanov 1927, Brik 1923, and Jakobson 1935) and the Prague School of Linguistics (Bogatyrev and Jakobson 1929, Bogatyrev 1936, Havránek 1932, Bühler 1934, and Mukařovský 1940) are important points of reference for my understanding of the problem of meaning in artistic language and of how to conceive of a philological method equipped to take up the challenges of confronting that problem. Newman and Newman 1984:32–38 integrate Russian Formalism and Pindar criticism.
[ back ] 24. Cf. Hymes: “The ethnographer is likely to look at communication from the standpoint and interests of a community itself, and to see its members as sources of shared knowledge and insight” (1974:8).
[ back ] 25. Cf. Bakhtin’s discussion of “particular social dialects” (1981:293) and his reference to “different languages, even from the point of view of abstract socio-dialectological markers” (1981:295–296, emphasis in original).
[ back ] 26. For a historical perspective on this quarrel within Pindaric studies, see Young 1970. Recently Currie 2005:11–15, Hornblower and Morgan 2007, and Nicholson 2007:208 indicate that this quarrel is alive and well, with the historicists in the ascendancy.
[ back ] 27. Cf. Hymes: “We can understand the perspective [of speech styles] as applying to any and all organization of linguistic features, of verbal means, in relation to a social context” (1974:59). My approach to style, then, has a more inclusive scope than that of Race, for whom style involves “the poet’s selection and placement of words” (1990:1–2).
[ back ] 28. Bauman 1977:5. On the relationship between style and performance in Homer’s Iliad, see Martin 1989:160–161.
[ back ] 29. Given that linguistic registers are situated and that conventional speech situations are genred—and approaching genre with Bourdieu 1977 as fields of social practice (as Hanks 1987 does)—my observations here are highly resonant with Kurke: “we must correlate genre with performance: if we define genre as the set of audience expectations which shapes and constrains each individual composition, we must take into account the nature of the audience and the occasion that informed their expectations” (1991:1, emphasis in original). Fearn 2007:220n168 has recently challenged Kurke’s conception of genre as “too limiting,” but as his discussion of genre in general (219–225) and his description of generic features of Bakkhulides’ poetry in particular (224) make clear, a static conception of tradition hamstrings his own approach to genre.
[ back ] 30. In her study of the performance of Archaic Greek poetry, Stehle defines “community” as “a collection of families organized under a system of governance, laying claim to a demarcated territory, worshipping common cults, and recognizing one another as fellow members” (1997:18).
[ back ] 31. Here I call attention to Martin’s (2003:158–159) balanced critique of structuralist approaches to the analysis of myth.
[ back ] 32. See also Hanks 1990:78–80.
[ back ] 33. On performance keys see Bauman 1977:15–24 and Foley 2002:85–93.
[ back ] 34. See Hymes 1974:52–65 for a discussion of each of these notions of the Ethnography of Speaking.
[ back ] 35. Here I am quoting from Bakhtin’s description of epic specifically in contrast to the novel: “The three characteristics of the epic posited by us above are, to a greater or lesser extent, also fundamental to the other high genres of classical antiquity and the Middle Ages. At the heart of all these already completed high genres lie the same evaluation of time, the same role for tradition, and a similar hierarchical distance. Contemporary reality as such does not figure in as an available object of representation in any of these high genres. Contemporary reality may enter into the high genres only in its hierarchically highest levels, already distanced in its relationship to reality itself. 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); they are woven by various intermediate links and connective tissue into the unified fabric of their heroic past and tradition. These events and heroes receive their value and grandeur precisely through this association with the past, the source of all authentic reality and value. They withdraw themselves, so to speak, from the present day with all its inclusiveness, its indecisions, its openness, its potential for re-thinking and re-evaluating. They are raised to the valorized plane of the past, and assume there a finished quality” (1981:18, my emphasis). To illustrate the prevalence of such a view of Pindar’s poetry among scholars, consider the following: “Given Pindar’s intellectual archaism in which the past is insistently viewed as a collective and generalized model for the particular events of the present, each epinician ode necessarily moves back and forth between the two modes of description” (Greengard 1980:48). In a similar vein, Mackie writes that “in spite of [an] apparent emphasis on the present and its importance, most epinician odes set out to understand the significance of the present victory by setting it against the backdrop of the heroic past” (2003:40), and Athanassaki that “[a]lmost all extant epinicians display a common pattern: they commemorate a specific historical victory in light of memorable heroic and divine deeds in the remote past” (2004:317). As I will show in Chapter 5, the past-present relationship alone does not account for other and more dynamic temporal trajectories operating in epinician performance.
[ back ] 36. Cf. Schadewaldt 1928:265, where he describes his work as Stilforschung. See Hamilton 1974:4–6 for an assessment of Schadewaldt 1928.
[ back ] 37. Richard Bauman per litteras.
[ back ] 38. Cited in Feldman and Richardson 1972:222 (my emphasis).