Previous research examining heritage speaker bilingualism has suggested that interfaceconditioned properties are likely to be affected by crosslinguistic influence (e.g., Montrul & Polinsky, 2011; White, 2011). It is not clear, however, whether the core syntax can also be affected to the same degree (e.g., Cuza, 2013; Depiante & Thompson, 2013). Departing from Cuza’s (2013) and Depiante and Thompson’s (2013) research, the present study seeks to determine the extent to which this is possible in the case of Spanish as a heritage language. With this goal in mind, a total of thirty-three Spanish heritage speakers (divided into sequential and simultaneous bilinguals) and a comparison group of eleven late Spanish-English bilinguals completed a battery of off-line tasks that examined knowledge and use of preposition stranding (i.e., a syntactic construction whereby the object of the preposition is fronted while the preposition itself is left stranded), an understudied core syntactic phenomenon that is licit in English but precluded in Spanish. Overall findings reveal that the sequential heritage speakers
pattern with participants from the control group. The simultaneous heritage speakers, on the other hand, seem to have a grammar that is not so restricting as they accept and produce ungrammatical cases of preposition stranding. Herein, we argue that these results do not obtain the way they do due to incomplete acquisition or L1 attrition but crucially because of the timing of exposure to the societal language. We propose that this property was completely acquired, although differently acquired due to the structural overlap observed between the two languages involved (e.g., Müller & Hulk, 2001), and most importantly, to the timing of acquisition of English (e.g., Putnam & Sánchez, 2013).
Auditory Processing of Gender Agreement across Relative Clauses by Spanish Heritage Speakers
