This paper considers the temporal relationship between account and in the passages of Pindar's odes. Highly selective, each covers vast extent of story-time, creating rapidly juxtaposed tableaux. The account typically begins near the end of the story, proceeding backwards by stages until it reaches the beginning. Thence it returns through all, some, or none of the intermediary stages until it reached its starting point, beyond which it may then continue forward in time. Prophecy and foreshadowing provide counterpoint to this basically analeptic structure. Different lines, simultaneous events, and alternative possibilities all find their place in the narrative. The length of account-time is generally less than that of storytime and diminishes as the proceeds. Repetition is necessitated by the backward and then forward order of combination as well as serving other purposes. Elaboration, when it occurs, serves to explain events that are improbable or inconsistent with the maxims contained in the non-narrative passages. Since Pindar's is the most complex art that we possess from classical antiquity, critical description of it is clearly desideratum. This need notwithstanding, recent critical descriptions of the odes, such as those by Richard Hamilton (1974) and Carola GreenAn earlier version of this article formed chapter of my doctoral dissertation, University of Toronto, 1987. I am grateful to Professors Gloria DAmbrosio-Griffith, Hugh Parry, Michael J. O'Brien, and especially Emmet Robbins, for their many helpful comments. Much further improvement is to be credited to the editor and the anonymous referees of Poetics Today. Poetics Today 14:4 (Winter 1993). Copyright ? 1993 by The Porter Institute for Poetics and Semiotics. CCC 0333-5372/93/$2.50. This content downloaded from 207.46.13.149 on Mon, 03 Oct 2016 05:30:10 UTC All use subject to http://about.jstor.org/terms 608 Poetics Today 14:4 gard (1980), discuss the ode as totality, inevitably failing thereby to give sufficiently detailed accounts of its component. The only monograph devoted entirely to the study of Pindaric narrative, meanwhile, is the important book by Leonhardt Illig (1932). Illig's work antedates the theoretical studies of by Roland Barthes (1966), A. J. Greimas (1971), Gerard Genette (1972, 1988), Mieke Bal (1977), Meir Sternberg (1978), Shlomith Rimmon-Kenan (1983), and others, who have attempted to create narrative (Greimas 1971: 794). The concept of is familiar to Pindarists, who were introduced to the notion of thematic motivational grammar of the odes by Elroy Bundy (1986 [1962]: 33; cf. Fraenkel 1950: 305; Lattimore 1964: 6; and Nisetich 1980: 25), and some Pindarists, such as Andre Hurst (1983, 1985), have adapted these theorists' grammars to the study of Pindar in piecemeal way. Nevertheless, acquires its strength from its successful totalization, and it is precisely such totalizing description of Pindaric that is lacking. The remarks that follow take step toward meeting this need. The narratological study of begins with the distinction among story, account, and text (Rimmon-Kenan 1983: 3). Between account and there are two kinds of relationships: those of narrational time (the selection, combination, and elaboration of elements), and those of narrational mood (the point of view whence the is recounted). There is, moreover, relationship among account, story, and narration-that is, the act of narrating itself-namely, narrational (Todorov 1966), or voice (Genette 1972: 225-67). It is my hope to one day to the consideration of mood and aspect in Pindaric narrative; the present article begins this project by considering the temporal relationship between account and story. With respect to time, the account differs from the in that it is selection of, combination or arrangement of, and possibly an elaboration or midrash-that is, commentary-on the events of the story. Let us consider each of these three differences in turn. Que n'oseje lui raconter de meme toutes les petites anecdotes de cet heureux age, qui me font encore tressaillir d'aise quandje me les rappelle! Cinq ou six surtout.... Composons. Je vous fais grace des cinq; mais j'en veux une, une seule, pourvu qu'on me la laisse conter le plus longuement qu'il me sera possible, pour prolonger mon plaisir. (Rousseau 1964 [1782]: 22) The selection of events is the product of the inevitable decision concerning what to state, what to suppress by paralipsis, and what to This content downloaded from 207.46.13.149 on Mon, 03 Oct 2016 05:30:10 UTC All use subject to http://about.jstor.org/terms Griffith * Pindaric Narrative 609 imply. Failed selection would result in a long story (Pyth. 8.30),1 total surrender to the encyclopaedic drive (Hardie 1986: 67), which Pindar avoids with scrupulousness usually associated with the later Alexandrian writers, whose motto was book big evil (Callimachus frag. 465 [= Pfeiffer 1949: 353]). Yet this ability to select the telling instance is mark of all Greek art and literature (Race 1986: 32) and is especially appropriate in an epinician context (e.g., Sophocles Electra 688-89). We need offer no examples of what Pindar states; the mythic narratives are composed almost entirely of such things. Where Pindar's suppression of events is tacit, there will be no agreement on whether suppression is even taking place. Of such cases may be said what J. B. Hainsworth (1988: 278) says of Homer, that he relates what men said and did. If for time they perform no deed worthy of record, there is nothing to tell; if it were not for his visitors we would be told nothing of Achilles' feelings between [Iliad] i and xvi. In some instances, however, it is clear that an event has been passed over in silence (Isth. 1.63; cf. Nem. 5.16-18). Important elements of myth, such as the name of the main character, may be suppressed by selection, as in Olympian 2.38-42, where the name of Oedipus is omitted either in order to touch lightly on the catastrophe (Gildersleeve 1979 [1885]: 147) or, less likely, because he is just tool of fate (Mezger 1880: 158); or in Aeschylus's Agamemnon 217, where Agamemnon's decision to sacrifice his daughter is suppressed. Some events are implied. For example, the phrase dinners given in return (01. 1.39) implies other, earlier banquets upon Olympus; remedy (Aeschylus Agamemnon 198) implies other, unspecified remedies. Selection is often motivated by the desire to model one scene on another by making the characters in given conform to the characters in another or other, inexplicit subtexts. For Greek lyric in general, Homer is the source of many such debts. The extent of Pindar's debt to him as model is uncertain. Pindar never directly alludes to the Iliad or the Odyssey (Wilamowitz-Moellendorff 1922: 173); his use of them as inexplicit subtexts is more difficult to measure. An example of selection that takes previous literary text as its model is to be found in the Geryoneis of Stesichorus, which is modeled on various passages in the Iliad. In that poem, the three-headed monster, Geryon, engages in debate about his mortality (frag. 11 [= Page 1974: 7-8]), as does Sarpedon (Iliad 12.322-28), and is visited by his 1. The following abbreviations are used in referring to Pindar's text: Pind. = Pindar; 01. = Olympian odes; Pyth. = Pythian odes; Nem. = Nemean odes; and Isth. = Isthmian odes. Pindar's odes are quoted following the Snell and Maehler (1984) edition, and his fragments according to that of Maehler (1989); all translations are