Aspectual features in Role and Reference Grammar: A layered proposal

Abstract

The kernel of the semantic representation of a predicate in Role and Reference Grammar (RRG) is based on its characterization in terms of an Aktionsart typology based on Vendler’s (1957) classes plus some additional elements from Smith (1997) and Dowty (1979). This means that event structures are mainly considered a lexical phenomenon pertaining to predicates, and only occasionally higher predicational structures are considered in event construction. Even though this approach is adequate to a great extent, there are still some problems in the approach taken in RRG. The most significant drawback is that non-lexical aspects appear intermingled with predicate-only features, which leads to misinterpretations and misclassifications of predicates. Consequently, it sees more sensible to bring a functional model of grammar like RRG to a compromise position and, thus, consider in what ways different units identified as belonging to the different layers in RRG’s syntactic projections ‘conspire’ in the final aspectual characterization of events. In this line, this paper will propose a classification of aspectual features in terms of the levels found in the functional projection of the clause as devised in RRG, namely the Predicate Level (the domain of Aktionsart typology), the Nucleus (where morphological aspect has scope) and the Core (the locus for what will be described as ‘aspectuality’ features).