Ilja Serzants

Lecture: Referential, discursive and syntactic properties of the bare partitive genitive in Ancient Greek
Abstract:
The present paper aims to investigate the main semantical functional, discursive and syntactic properties of partitives on the bases of the bare (independent) partitive genitive in Ancient Greek. Contrary to the previous views that the bare partitive genitive (PG) primarily encodes a partitive relation I argue that the PG marks a participant as being un(der)determined as to reference, quantity and, partially, as to its theta role, though this indeterminacy of the PG is only a default one. I assume that the core function of the bare PG in Ancient Greek is discursive: it makes the participant it refers to discursively inherently backgrounded while, at the same time, it promotes the underlying set to which the denoted participant belongs to in the discourse. Thus, the underlying set may be a focus or topic while the actual participant encoded by the bare PG cannot. As regards morphosyntax, the PG in Ancient Greek has a number of atypical, typologically rare features such as its ability to trigger verbal agreement (while being in the subject position) along its semantic number. Furthermore, the PG in the subject position behaves also otherwise as a nominative marked constituent: it can be coordinated with nominatives and it can agree with nominative case-marked participles. These properties can be readily accounted for in terms of a covert head that both triggers the (formal) agreement and gets case.
Furthermore, the PG, in a way, “disturbs” the nominative-accusative alignment in that it can equally encode both intransitive subjects and transitive objects, levelling out, thereby, the morphological distinction between S and O (in Dixon’s terms).