Mary R. Lefkowitz, First-Person Fictions. Pindar's Poetic 'I'. Oxford: Clarendon Press, 1991. Pp. ix + 226. ISBN 0-19-814686-8.
Reviewed by Kathryn A. Morgan, Ohio State University.
Of all the mystifying aspects of Pindar's poetry, perhaps none is as intriguing as the identity and nature of the speaker(s) of his first-person statements. First-Person Fictions. Pindar's Poetic 'I' is a collection (more or less chronologically arranged) of the articles on the "various aspects of the fictions of Pindar's identity" (p. vi) that Mary Lefkowitz has written over the last thirty years. Common to all of them is the assertion that the first person voice in the odes is that of the poet alone and that the notion of multiple identities is fundamentally misguided, a result of modern reliance on ancient scholia which were themselves historical fiction.
L. has been encouraged to publish this collection because the idea of multiple identities still persists (p. vi), but the book is timely for another reason. One of the more interesting recent controversies in Pindaric studies has been the battle waged between the critical orthodoxy that Pindar's epinician odes were performed by a chorus singing in unison ("the choral hypothesis"), and the radical suggestion that many, if not all, of the victory odes may have been sung by a soloist, albeit with a dance accompaniment ("the solo hypothesis"). Along with M. Heath, L. has been the chief exponent of the radical position, but her views are the product of a long intellectual evolution which this book documents. Students of, and participants in, this debate over performance are fortunate to have most of the relevant material collected here. 1 L. has taken the opportunity to rewrite and update her notes, consolidate the bibliography, and redo her early translations, which now seem less literal, more nuanced and interpretive.
L. aimed also "to eliminate repetition whenever possible, to correct mistakes, and to remove inconsistencies" (p. vi). Of course, in a book of this nature some repetition is inevitable; L. returns to the same passages more often than one might wish, and it may be disconcerting to those who have not read the original articles to encounter the various palinodes that the development of L.'s thought has dictated. For historians of scholarship, however, the process will be fascinating. Perhaps it is selfish to wish that L. could have reworked all the material into a unified treatment, but it might ultimately have been more satisfying for the reader. It is hard, for example, after struggling with the interpretation of GARU/EI at P. 5. 72 (pp. 61-64) to realize (p. 61, n. 102) that L. now reads GARU/EIN (correctly, I think), especially since her treatment of the former reading is problematic (see below). L. does, however, give her reason for leaving her original discussion intact (GARU/EI with KLE/OS as its subject presents her with a greater challenge), and it is a reasonable one. (The presence in the text of passages enclosed in square brackets -- which seem never to be explained -- is somewhat confusing. They often represent substantial alterations in the original text, but not all alterations are so treated.)
Chapter 1, "The First Person in Pindar," is the longest, taking up one third of the book. It examines from a literary rather than an historical point of view (i.e. using only the evidence of the personal statements themselves, not the scholia) the question of whether the chorus of epinician ever speaks for itself. L. first distinguishes three theoretical types of choral statement in Pindar: bardic (formal, professional statements clearly made by the poet), more personal and subjective statements by the poet, and statements clearly made by the chorus (as in Paean 4). All three types have similar functions, that is, they serve as transitions and as ways of describing the speaker, but the nature of the self-description varies. Bardic and personal "I" statements deal with the role of the poet, while choral first-person statements are more topical and concerned with physical detail (the nature of their homeland, their appearance). In pure choral poems, the chorus speaks in character throughout, without the intrusion of the poet's voice. Once L. has established these criteria, she applies them to the epinician odes, with unsurprising results: there is no choral "I" in epinician, since no epinician first-person statement has the content of a pure choral "I" statement. Passages which the scholia assign to choral speakers are better understood as being spoken in the voice of the poet. Indeed, the reader will feel that as far as the possibility of a choral speaker is concerned, the battle is lost by p. 25. If persona and tone must be uniform and univocal throughout a victory ode (as they seem to be in paeans and partheneia), if the epinician voice must be universalizing and not topical, and if a choral speaker can only be topical and not universalizing, then the epinician "I" cannot be choral. Surely, however, uniformity of tone and persona in epinician is precisely what must be established. No matter how uniform paeans and partheneia are, it does not follow that epinicians must be the same.
The chapter ends with the germ of the idea that will later develop into the solo hypothesis: Pindar's complex and metaphorical diction and his emphasis on his poetic abilities lends his odes a subjective tone akin to elegy. As the importance of athletic victory increased, L. argues, so did the need for a greater poetic authority than that of a chorus. Thus Simonides, Bacchylides, and Pindar treat epinician (which may have had its origin in choral poetry) almost as a form of monodic poetry, where the only voice is that of the poet.
Since this chapter is the oldest (the original article in HSCP was based on L.'s dissertation), it is the most extensively revised. As L. later comments (p. 127), the article was written before the publication of Bundy's Studia Pindarica, when she still believed that parts of the odes looked to specific historical situations. The amount of historical discussion in the present chapter has, then, been considerably reduced. In a note (p. 8, n. 8) L. distances herself from the political interpretation of N. 8 offered by Brown and Finley, although she leaves this interpretation untouched in the text. On other occasions L. eliminates or decreases historical references in the text itself. So, for example, her treatment of N. 5 no longer alludes to strained relations between Athens and Aegina (p. 32), and the pages on I. 8 cease to emphasize so strongly Thebes' questionable position shortly after the battle of Plataea (pp. 44-47). Along the same lines, the discussion of N. 4. 33-43 no longer includes speculation (based on the scholia) on the identity of Pindar's poetic rival (pp. 48-49).
The distinction between a bardic and a personal "I" in the odes is another position which L. has ceased to support; Pindar's subjectivity is rather a pose and a fiction. Accordingly, phrases in the original article such as "intense subjectivity" have become "apparent (though illusory) subjectivity," and in general, adjectives such as personal and subjective are now put in quotation marks. Other changes worthy of note are an updated discussion of Alcman 1 PMG (pp. 17-20), a new emphasis on the role and importance of the komos (obviously a reflection of her most recent work), and most important, the dismissal of Hermann's conjecture E(A=| at N. 7. 85 (which she had previously accepted) in favor of her more recent interpretation that takes patra to mean "clan" rather than "fatherland," thus making Pindar a member of the Aegidae. This removes one of the biggest obstacles to seeing Pindaric self-reference in this passage, namely that on the standard reading it is inappropriate for Pindar to address Aeacus as guardian of his fatherland.
Pindar the Aegid makes several appearances in this book, and since L. gives this hypothesis such explanatory power, it is worthwhile to comment on it briefly. We learn from the scholia that the Aegidae were Thebans who helped the Spartans (alluded to in I. 7) and afterwards settled in Lacedaemonia. Subsequently they were involved in the colonization of Thera, and came from there to Cyrene (as narrated in P. 5). The reference to them in P. 5 confused the Hellenistic commentators (Schol. P. 5. 96a), who were uncertain whether the words were spoken in the person of the poet or of the chorus. Declaring that Pindar is an Aegid certainly solves one problem, but the cost is great. There is no evidence, even in the scholia, that Pindar was a member of that clan, and only the prior assumption that the poet must be speaking justifies this move. Even if Pindar is speaking here, this could just as easily be as a Theban rather than as a member of a particular clan. L. again introduces the Aegidae to explain N. 7. 84-86b, citing Schol. I. 7. 18a as evidence that they were originally an Aeginetan clan. But this piece of lore is not mentioned in the scholia to N. 7, and in the scholia to I. 7 an Aeginetan origin for the clan is merely one of several possibilities that include Athenian provenance. We may well be wary of a hypothesis so inventive that even the scholia did not come up with it. Moreover, in order to construct this theory, L. must use scholiastic information that she elsewhere castigates for its lack of historical foundation.
This practice will recur in the next chapter, where L discusses the problematic encounter with Alcmeon in P. 8. Again, the scholiasts offer various explanations unsatisfactory to L.: there was a heroon of Alcmeon near the victor's house, or there was a shrine of Amphiaraus near Pindar's house. L.'s solution is to postulate a similarly unattested shrine to Alcmeon near Pindar's house. Yet all these solutions are based upon the dreaded eikos and no one is inherently more plausible than another, unless, that is, we have already decided Pindar must be speaking. Now, it is true that L. claims only that problematic passages are as easily understood on her hypothesis as they are on any other, not that the text must dictate her reading, but in this instance her interpretation again involves her in the kind of elaborate historical fiction we find in the scholia. Thus Pindar must either have had an encounter with Alcmeon of the same type as Hesiod's with the Muses (although Pindar's has no trace of the poetic investiture motif), or the hero communicated with him in a dream. If the latter, however, are we to assume a true "prophetic" dream or a retrospective fiction? Both solutions are problematic, and I do not pretend to know the answer, but I believe the final word on such difficult passages has yet to be spoken.
Chapter 2, "The Influential Fictions in the Scholia to Pindar's Pythian 8," is an examination of how the scholia (on which most modern commentaries are based) were influenced by Hellenistic cultural aesthetics. L. deals first with the political scholia (which attempt to reconstruct an historical background for the odes) and then with the choral ones (which resolve difficulties in interpretation by positing a choral "I"). Both types, L concludes, rely on eikos to deduce facts from the poem itself (a common Hellenistic biographical practice); the presence of alternate and mutually exclusive scholiastic explanations proves that the Hellenistic commentators worked without the benefit of external information. Typical also is the explanation of mythological or imagistic obscurity by recourse to allegory (thus the jackdaws of O. 2 become Simonides and Bacchylides).
In chapter 3, "Pindar's Lives," L. identifies four different constructs of Pindar, ancient and modern. Common to all four is the use of Pindar's poetry as the sole source. The first, the Pindar of the ancient lives, is a saintly figure in close contact with the divine, a product of historical fiction whose life notably conforms to ancient biographical topoi. The Pindar of the scholia is an isolated, competitive, and heroic figure whose characterization arises from interpretation of the dramatic metaphors of the odes. The scholiastic Pindar's poetic rivalries are, however, a reflection of Hellenistic professional hostilities. L. comments that although this version of Pindar has been widely accepted by subsequent scholars (since it makes his poetry more comprehensible), it distorts the meaning of the odes. The third Pindar is Bundy's technician, but he is an impersonal and abstract figure and the approach is too reductive. The last version of the poet is L.'s own and is constructed by examining what makes Pindar's response to tradition distinctive: his direct involvement with the process of success, his identification of his own achievement with that of the athlete, and his awareness of the risk of failure.
In "The Poet as Hero," L. continues on the same track. The ancient lives of the poets are considered as examples of a new fifth-century mythology wherein the hero, a figure of reserve and moderation, uses words as his weapon and serves the commonwealth. The source of this mythology is poetic first-person statements (as in Pindar and Xenophanes) and works such as the Frogs of Aristophanes; the presupposition was that poet and work represented each other. L. describes this heroism as "constructive and popular." Moral courage, however, is balanced in the lives by normalizing details that bring the poetic hero down to earth (such as degrading deaths). Ultimately, L. concludes, the new heroism is trivial and anti-intellectualizing.
Chapter 5, "Autobiographical Fiction in Pindar," looks at the tone of Pindar's longer "I" statements as expressions of his own mythology of poetic behavior. The poet is combative and defensive, sharing the victor's determination and isolation, but also acting as his model. Pindar's "autobiographical" passages are meant as examples and do not tell us anything about the real person; rather, they describe the general meaning of the victor's achievement. "The Poet as Athlete" (Chapter 7) will expand upon these observations: Pindar sometimes pictures his own art in terms of athletic victory. Athletic metaphors stress Pindar's skill and superiority and his appreciation of the risk involved in endeavor; thus they add excitement to his song.
In chapter 6, "The Pindar Scholia," L. characterizes the kind of mistakes made by the Hellenistic commentators. They simplify and specify where Pindar is complex and abstract, and assume that poetry is irrational. Taking the poet at his word, they see myths as digressions and Pindar in danger of losing control. Confused by unusual metaphor they think him deliberately obscure. The result is a Pindar (familiar to us from Horace) who is "impulsive, impressive, but enigmatic" (p. 160).
Chapter 8, "Pindar's Pythian 5," is a reading of one ode, triad by triad, that attempts to specify what makes the poem distinctive, identifying instances where Pindar has used conventional motifs with special force, either through bold phrasing (e.g. Arcesilaus' prosperity as an ancestral O)FQALMO/S in 18) or reiteration (e.g. the need for divine support in order to achieve, or the power of wealth). Again, L. affirms that the voice of the speaker, both here and elsewhere, is Pindar's own. This chapter is further notable for the first appearance of the suggestion that no ode of Pindar need have been sung by a chorus. This "solo hypothesis," is developed at greater length in the final chapter, "Who Sang Pindar's Victory Odes?" Here, L. argues against the scholarly assumption that choral and monodic lyric were different genres and proposes that, since recent work on Stesichorus has suggested that his longer triadic poems (such as the Geryoneis) were not chorally performed, triadic structure need not imply unison choral performance of Pindar's epinicians either. L. examines some of the passages which have been adduced as evidence for the choral hypothesis (particularly in O. 6 and N. 3) and concludes that references to young men singing, or waiting to sing, reflect not a formal chorus, but the informal komos that was part of the victory celebration (it is telling that Pindar never calls such young men a chorus, only a komos). Here at last is the solution to the problem L. first formulated in chapter 1: "it is difficult to explain why epinikia should have been performed by choruses, in spite of their subject matter, and in spite of the fact that the poet speaks in his own person throughout" (p. 70). The answer? The victory ode may in fact be monodic.
It would not be useful for me to rehearse here the possible arguments against the solo hypothesis, since this task has been undertaken by Burnett and Carey,2 but I will mention one additional consideration. L. states that we should question the scholiastic assumption of choral performance because the scholiasts have been so often proved to have fabricated information on the basis of eikos in order to solve a problem. Yet distrusting the scholia on the question of choral performance is a step fundamentally different in kind from distrusting their historical and biographical reconstructions. Scholiastic reconstructions were a response to problems in interpretation, but choral performance as such does not seem problematic in the scholia and is taken for granted. It is a choral voice that the commentators propose as the way out of their interpretive difficulties.
The book's conclusion acknowledges that Pindar's language is vague enough that no decisive answer to the problem of choral versus solo performance is possible, but rightly insists that the questioning of old assumptions is at least a valuable exercise. Even more important, L. briefly returns to the implications of the solo hypothesis for larger issues in Pindaric interpretation (first addressed on p. 201). If the idea of choral performance is rejected, it is no longer so easy to see the performance of epinician as a public celebration and reflection of communal ethos (L. displaces the communal function of epinician onto the komos of which the song is a part), and the perception of Pindar's song as mediator between victor and community is weakened.3 L. refers with some acerbity to the appealing "political currency" of the communal view; paradoxically, her radical reinterpretation of the conditions of performance results in a deeply conservative perception of Pindar's poetic task.
L.'s picture of Pindar and his poetry is clear and readily comprehensible. The poetry is univocal (in the most literal sense) and its tone is uniform. This is both attractive and disturbing, and the relative proportion of these feelings in the reader will be determined by the degree to which he/she desires a uniform and univocal poet. Such uniformity is precisely what needs to be established; it cannot be assumed. The portrait of a univocal Pindar performing monody lacks some of the tension (I believe, a fruitful tension) inherent in the choral hypothesis between the voice of the poet and the multiple presenters of that voice. It will be clear from this comment that I agree with L. in seeing the poet as the sole speaker, but I do not think that the difficulties inherent in this view (such as Pindar the Aegid and his encounter with Alcmeon) can be dismissed quite as neatly as L. manages to do. Perhaps L. herself points us towards a way of resolving these difficulties in her stress on xenia. If expressions such as "I suffered a sorrow unspeakable" in I. 7 can be seen as an expression of sympathy for those who have suffered a loss, we might take such imaginative identification further and conceive that the poet's xenia may by its very nature generate what seem to be multiple and confusing voices.
The major achievement of this series of articles is the establishment of Pindar's identity in his epinicians as a rhetorical construct, and the subjection of the scholia to a rigorous and enlightening (although sometimes extreme) examination; we are now less likely to trust them on matters of historical fact. The question of whether Pindar's is the only voice we encounter in the epinicians, and whether these poems were chorally performed or not, cannot be regarded as settled. Disagreement in an area of such interest and importance is inevitable. What is more certain is that L.'s work has had the valuable function of stimulating new thought, making us more aware of our interpretive methodology, and forcing us to reevaluate and justify our assumptions even if we do not entirely agree with her.
 Heath and Lefkowitz's response to the objections of C. Carey (CP 86 ) was still forthcoming when First Person Fictions went to press.  Burnett: CP 84 (1989). Carey: AJP 110 (1989); CP 86 (1991).  Contrast the scholars referred to on p. 201, n. 42; p. 205, n. 9. To this list we should now add Leslie Kurke, The Traffic in Praise (Ithaca, 1991).