Bilingual and monolingual children’s productions of German word stress, grammatical gender and verb placement : The role of language exposure and other linguistic and sociological predictors and mechanisms

Abstract

Growing up multilingual is an advantage (e.g. Bialystok et al. 2012, Han 2010) and daily reality for a large number of children in Germany (Autorengruppe Bildungsberichterstattung 2020). At the same time, children with a non-German home language have on average lower German skills than children who grow up with German as their only first language (e.g., Hußmann et al. 2017b, among others). Aspects related to the home language, such as characteristics of the heritage language (HL), the age of onset of acquisition of the target language (AOA), and quantity and quality of language experience (exposure), together with the socio-economic status (SES), such as education and cultural capital, are important predictors for language acquisition and language competence (e.g., Baumert et al. 2003, Meisel 2018, Müller et al. 2002, Unsworth 2013, among many others). Sociological and linguistic language acquisition models associate influences of the HL, AOA and SES with mechanism of efficiency or internal mechanisms (Chiswick & Miller 1995, Esser 2006, Paradis 2011). However, the influence of these efficiency-indicators can also relate to language exposure (Hart & Risley 1995, Hoff 2003a, 2003b, Jia & Fuse 2007). The dissertation examines whether and how the HL, AOA, SES and exposure influence bilingual and monolingual children’s acquisition of three linguistic phenomena in German: syllable durations in word stress, gender marking on the definite article, and verb placement in main and relative clauses. 161 children aged between 3 and 10 years who were either Italian-German-bilingual, Turkish-German-bilingual or German monolingual participated in playful elicitation procedures in which productions of the phenomena were recorded. Information on the predictor variables was collected in a parental questionnaire. Language exposure was measured and operationalized as input and output, in quantitative and qualitative, in current and cumulative form, as well as within and outside the family (e.g., Klein & Becker 2017, Paradis 2011, Unsworth 2013). While the investigated predictors did not strongly affect variation in syllable duration differences, the predictors, particularly the indicators of language exposure, showed more visible effects on gender marking and verb placement. Certain results suggest an influence from the characteristics of the HL. However, differences found between the speaker groups could also be explained by their different quantity and/ or quality of exposure. Similarly, many effects of AOA or SES seem to relate to language exposure. Linguistic input or interaction is therefore the most influential predictor for the phenomena investigated in the three speaker groups. This result supports and complements previous research on the high effectiveness of exposure for language acquisition, which can be helpful for the development and evaluation of measures to support language skills.