Annette Teffeteller
Recent work on the poetic grammar of Pindar has emphasized the interaction of syntactic and metrical constituency and prosodic features of the language in relation to the syntactic and metrical boundaries of the verse. This work has also taken into account the value of comparison of these aspects of Pindar’s lyric verse with the hymns of the Rigveda (Watkins 2002a, 2002b), following an early account of parallels in the metrical systems of the Greek and Indic traditions (Meillet 1923).
The poetic structures of the Vedic hymns have been seen by some as consisting of ‘loose syntactic bonds’ (“subjects, verbs, and objects range freely over the course of several verses”, Patton 1996) or of covert as opposed to overt constituency (Watkins 2002a); a different theoretical perspective sees these structures rather as the product of skilled exploitation on the part of the Vedic rishis of the syntactic licensing allowed by a nonconfigurational syntactic typology. Similarly, in the syntax of Pindar many apparent instances of ‘covert’ constituency are, on this account, not instances of constituency at all but are rather adjunctions or, in the case of adjectives, separate secondary predications.
T.S. Eliot’s view of the poet’s task as “to force, to dislocate if necessary, language into his meaning” reflects an Indo-European tradition that continues in the hymns of the Vedic and Greek poets. It is a tradition that valorizes complexity of thought, systematic ambiguity, multivalency, multiple levels of literal and metaphorical meaning held in equipoise, and a deeply-felt interaction between the world of men and the realm of divinity. And all of this complexity of meaning is encoded in the intricate interplay between the syntactic, the metrical, and the stylistic structures of the poetic grammar.
This paper focuses on Pindar’s manipulation of syntactic structures in his poetic art, a practice which clearly shares a tradition with the poets who composed the Rigvedic hymns, as also of other Indo-European traditions. In these poetic traditions verse boundaries (line-end, line-break, etc.) often coincide with syntactic boundaries (phrase/clause/sentence). As Calvert Watkins emphasized already in 1995, “syntactic phenomena sensitive to sentence boundary are frequently found adjoining metrical boundary, both external (e.g. line boundary) and internal (e.g. caesura). Metrical boundaries frequently coincide with formula boundaries. The resultant interplay or counterpoint between syntax and meter is a very distinctive characteristic of the earliest Indo-European poetry, and presumably of the poetic grammar of the proto-language as well” (Watkins 1995:20).
Expanding on these observations in his 2002 study of ‘word order and metrical structure in the Odes of Pindar’, Watkins stated that his concern was “the contribution of the poetry of Pindar to the study of Indo-European syntax, with particular attention to the question of word order” (Watkins 2002b:319). The concern of the present study is the reverse: the contribution that our analysis of Indo-European syntax can make to our understanding of the poetry of Pindar. The theoretical framework builds on Meillet’s analysis of Greek and Indo-European syntax (Meillet 1903 and 1913 (echoed by Renou 1952 for Vedic, noted by Elizarenkova 1995)) and the study of hyperbaton in Greek by Devine and Stephens, who state that “it is unlikely that subject conjunct hyperbaton originated from the splitting of a coordinated noun phrase either in regular conjunct hyperbaton with a singular verb or in the schema Alcmanicum with a plural verb. In fact, more than any other single piece of evidence, the schema Alcmanicum requires us to take seriously the idea that in its prehistory Greek was not only a nonconfigurational language but one that made at least some use of pronominal arguments” (Devine and Stephens 2000:158-159). The nonconfigurational syntactic typology of early Greek and the pronominal argument status of the early Indo-European languages has been the subject of recent work by Teffeteller (2015; cf. 2001). The present study investigates the poetic syntax of Pindar within this framework.