Chapter 5 Auxiliary selection in heritage speakers of Italian

Abstract

The primary focus of many investigations into heritage languages is to provide an explanation about how and why heritage grammar may differ from the standard variety as its language of origin. Various theories (Scontras et al. 2015) share the common goal of understanding what contributes to the features of certain areas of the heritage grammar, often defined as “particularly vulnerable to reanalysis” (Scontras et al. 2015: 9). Simplified features of the heritage system are viewed as evidence of incomplete acquisition (Polinsky 2006), or of divergent attainment (Scontras et al. 2015), due to reduced input (Montrul 2008). The present study moves away from these perspectives and adopts the notion of heritage grammar as an independent linguistic system, which was born from the source language of input but then developed its own set of rules. The study specifically investigates the behavior of heritage speakers of Italian living in the US in auxiliary selection between essere (to be) and avere (to have), within the Italian compound past tense passato prossimo. Data is collected through a Forced-Choice (FC) Judgment Task administered to parents and children of the same family nucleus, for a total of seven children as input receivers and six parents as native input providers of Standard Italian. The language of the parents is compared to the language of the children in order to identify possible similarities and/or differences between two independent linguistic systems, the heritage language (HL) and the source language (SL). Results show that heritage speakers of Italian select the auxiliary according to the semantic properties of the verb and that they are sensitive to the unaccusativity gradient, as identified by Sorace (2000) in native monolingual speakers of Italian. The study contributes to highlight the notion that heritage speakers are guided by a native linguistic sensitivity, which operates within the structures of their heritage grammar. More studies are needed to support the idea that the process of reanalysis to which heritage grammar is subject in specific grammatical areas may not be triggered by the reduced properties of input, but by the specific nature of the source language and its inherent degree of variation, as well as language specific properties