This paper will discuss some differences between the Homeric technique of verse-making and the technique of later epic poetry by analyzing constructions for the verb προσέειπεν in the epics. The concept of construction, borrowed from Usage-Based Linguistics and Language Acquisition studies, is a refinement and expansion of Parry’s formula, and proves to be a very powerful tool for analyzing a poet’s technique (Bozzone 2010). Unlike formulas, constructions can capture expressions of different levels of fixity, ranging from fixed formulas to general syntactic templates used for building a line.
Comparison between Homer and later hexametric poets is not new to the field. Parry (1928) famously demonstrated that Homeric language is traditional by constrasting the use of noun-epithet formulas in Homer to the use of the epithet in Virgil and Apollonius Rhodius. While epithets in Homer were embedded in an extensive and economic system of noun-epithet formulas, epithets in later authors were not. Parry’s demonstration was the first step in the development of Oral-formulaic theory.
The aim of this study is to extend this kind of comparison to the constructions for the high-frequency verb προσέειπεν in Homer and other Greek epics, namely Hesiod, Batrachomyomachia, Apollonius Rhodius, and Quintus Smyrnaeus. Even such a small case study proves revealing of the poets' technique, and suggests further directions of research. Both oral and writing poets can rely on constructions to build their lines (just like speakers rely on constructions in natural language production), but close observation reveals stark differences, both within the Homeric corpus and without. These differences are not simply in terms of orality vs. literacy, but reveal the gradual evolution of the poetic technique.
Homer knows three general construction types for the verb-form προσέειπε, each with several subtypes (often specialized for highly specific narrative or thematic material). One such construction, exemplified in (1), is notated in (2):
(1) Τὸν δ’ αὖτε προσέειπε θεὰ γλαυκῶπις Ἀθήνη (Il. I.206)
(2) [Object.Pronoun] [Adverb] προσέειπε [Subject.NounPhrase]
(Where each bracketed item has one fixed metrical shape).
The Iliad shows a strong a preference for constructions that place προσέειπε immediately before the feminine caesura, with a full subject following the verb, as in (2) above. These constructions often work as “staging formulas,” introducing a new referent to the scene (Bakker 1997: Ch.7).
In the Odyssey, although this pattern is still prevalent, another is quickly gaining ground, where the verb is line-final, and is often preceded by the predicative participle ἀμειβόμενος (the subject can be expressed or not):
(3) ἐξαῦτίς μιν ἔπεσσιν ἀμειβόμενος προσέειπεν (Od. XVI.193)
This pattern is often used for characters that are already on the scene, and need no introduction; it can also convey a higher density of information. A smaller, yet denser pattern, taking up only half a line, seems to be an experiment limited to Odyssey XIV-XVII:
(4) στῆ δὲ πάροιθ’ αὐτῆς. τὸν δὲ προσέειπεν Ἀθήνη· (Od. XVI.146)
Yet it is this rare pattern that the poet of the Batrachomyomachia (177) seems to reproduce faithfully. Can this small detail tell us something about the relationship between the Homeric tradition and the writing poet of Batrachomyomachia?
As opposed to Homer, Hesiod only shows instances of pattern (2) above, and no trace of the more innovative types.
Among later hexametric poets, the comparison between Apollonius and Quintus proves especially interesting. In Quintus Smyrnaeus, we can see how even a literate poet can develop a number of semi-fixed constructions much like Homer’s (though of a different breed than the Homeric ones). Such semi-fixed constructions are largely absent from Apollonius, who seems to operate with a largely free approach to word-choice. This points to a marked difference in compositional technique even between poets who are literate.