An experimental analysis of embedded Bare Argument Ellipsis in Italian

Abstract

Ellipsis appears to challenge modern theories of language as it would appear to convey meaning in absence of linguistic material. Throughout the last few decades, scholars have tried to delineate a cohesive theory of ellipsis in order to account for the exceptionality of the phenomenon and its varieties: in fact, elliptical constructions are numerous and the differences among them are notable. Crucially, every language that has been studied so far appears to display some kind of ellipsis. This has given rise to a plethora of studies of the matter in the various languages, and the present work is to be set in such tradition. In fact, this thesis is an experimental study of subordinate Bare Argument Ellipsis (BAE) in Italian. The study has been carried out starting from the data in Wurmbrand (2017), according to which in some languages embedded BAE is possible only when the complementizer is absent, whereas in others, the same phenomenon can occur despite the presence of a complementizer. The question to be answered via the present work is whether Italian belongs to the first or second category of languages, and what are the peculiarities of embedded BAE in such language. The data has been collected via survey, whereby the subjects have been presented with 20 embedded contexts containing Bare Argument Ellipsis. By using a Lickert (1932) scale, an acceptability judgment has been provided for each sentence, rating from 0 (completely unacceptable) to 5 (perfectly acceptable). The variables under consideration are the presence or absence of the complementizer, the syntactic nature of the remnant (subject or object), and the nature of the main verb, as regards whether it requires the indicative or the subjunctive.