Human languages can be split into two groups based upon how they mark the grammatical subject and object in a sentence. Accusative languages like English, Spanish and Japanese mark subjects distinctly from objects; ergative languages like Kaqchikel, Inuktitut and Samoan mark the direct object of a transitive verb in the same way as the subject of an intransitive verb. Ergativity is thought to be a fragile feature of language since ergative languages may lose their ergative marking and become accusative languages, but accusative languages rarely become ergative. This shift occurs because children acquire ergativity quite late in childhood, making ergativity susceptible to loss. Second language (L2) learners have great difficulty with it, and heritage speakers whose childhood home language is ergative have particular difficulty with ergativity if the dominant societal language is accusative. Thus many ergative languages are in danger of losing this defining characteristic. This dissertation project investigates how ergativity might be strengthened in the minds of heritage Samoan speakers by presenting them with a targeted protocol designed to increase their knowledge of ergativity. This project contributes to our understanding of an important though understudied property of human language, ergativity, as well as a large (and understudied) learning population, namely heritage speakers. It also furthers our understanding of an important language of the Pacific: Samoan. The research protocol and results will have important implications for language pedagogy that target languages with fragile grammatical features such as ergativity.
The experiments will be conducted by Grant Muagututiʻa, a doctoral student at the University of Hawaii, and will involve three groups of learners (native, heritage, L2) in a series of experiments targeting morphological ergativity (case) and syntactic ergativity (relative clauses, wh-questions) in both comprehension and production. The durability of the recovered ergativity is tested by increasing the length of time between instruction and testing. Comparing L2 speakers and heritage speakers addresses whether the increase in ergativity is attributable to knowledge gained as (heritage) children, while testing case, relativization and wh-questions addresses whether morphological ergativity transfers to syntactic ergativity.
