Bilingual Literacy and Input Negotiate Knowledge: Tracing Heritage Language Bilingual Development   

Principal Investigator: Fatih Bayram (UiT)

Supervisors: Jason Rothman (University of Reading, UiT), Marit Westergaard (UiT, NTNU)

Funding: Marie Skłodowska-Curie actions (MSCA)

Duration: August 15, 2018 – August 15, 2020

BLINK significantly adds to a burgeoning—roughly 2 and half decades long—sub-field of bilingualism studies, namely Heritage Language Bilingualism (HLB). We propose a novel approach that combines offline and online methods and focuses on a severely understudied, yet crucially important age group (age 9-15) of heritage speaker (HS) bilinguals in an effort to fill crucial gaps in the available literature. We will examine the role and weight of key variables that (potentially) contribute to ubiquitous differences between these populations and matched monolingual counterparts. To this end, we capitalize on the relatively unique opportunity Turkish provides as a heritage language acquired in its diaspora in the EU (the snapshot will be Norway and Germany). Having access to balanced populations in both Norway and Germany and their respective multiple subgroups that differ in key variables (e.g., access to formal literacy in the HL, the context of immersion, age at time of testing, influence of the specific majority language) we will for the first time have a fairer chance to converge on an understanding of how and why HLB presents so differently from monolingualism and even other cases of bilingualism. To do so, we (i) investigate knowledge/production of word order and Case marking in HS Turkish and whether/how their relationship is employed for predictive processing and (ii) examine the extent to which input-experience factors that can vary considerably at the HLB individual level interact to explain HS performance variances. Our battery of experiments also fills a methodological gap in the HL literature by using oral production, online (eye-tracking) and offline comprehension experiments with the same HL populations, thus enabling one to see if modality of testing (production vs. processing in real time) brings anything to bear on HS performances.

Comments are closed.