Bilingualism reading group
Speaker: Cristina Flores (Universidade do Minho)
Title: How stable is an early-acquired language?
Abstract: This talk will address the topic of ‘language stabilization’ and ‘attrition’ in the context of heritage language acquisition and reversal by discussing empirical evidence from research on bilingual speakers of Portuguese and German. In a first part, I will show that heritage speakers of European Portuguese who live in a German-speaking host country develop very stable knowledge of their heritage language, due to continued exposure to Portuguese over their life span. Nevertheless, Portuguese as a heritage language shows the same particularities that have been thoroughly described for other heritage languages (see overview in Montrul 2016), namely delayed development of certain (late) structures, amplification of variation that characterizes vernacular Portuguese and absence of properties that occur mainly in formal registers (Flores, 2015, Flores et al., 2017, Rinke & Flores, 2014). I argue that these particularities are not the result of language attrition, a process that is claimed to be involved in the development of a heritage language (Polinsky, 2011), because heritage language development implies a steady grow of linguistic knowledge (even if it develops at a slower pace than monolingual grammars). In a second part, I will review evidence from Portuguese returnees, i.e. bilinguals who grew up in German-speaking country and moved, during childhood or adolescence, to their (parents’) country of origin, showing that in cases of very severe input reduction the linguistic knowledge of a bilingual speaker may destabilize (Flores, 2010, 2012, 2015, 2019). Destabilized knowledge (=language attrition) leads to high degree of variability in language production, i.e. the speaker produces grammatical utterances in parallel to ungrammatical ones in the language that is no longer present in daily interaction (German, in our cases).
The degree of attrition effects depends mainly on the returnees’ age of input loss and length of exposure to German during emigration. Data from a case study on re-immersion (Flores, 2019) show that ‘destabilized knowledge’ is the result of a process of strong language inhibition due to lack of use rather than to restructuring or erosion of grammatical knowledge.