Periodic Reporting for period 1 - HeLA (Heritage Language Acquisition: The qualitative nature of input and cross-linguistic influence)
Reporting period: 2022-01-03 to 2024-01-02
This MSCA project, entitled “Heritage Language Acquisition: The Qualitative Nature of Input and Cross-Linguistic Influence (HeLA)” examined which are the key factors contributing to the development of the heritage language in bilinguals. The field of heritage language bilingualism is a relatively young field within the more general field of multilingualism. It focuses on a specific group of bilinguals: heritage speakers, who typically grow up with a heritage language only spoken in the home or with other members of the (often immigrant) community, which is not an official language of the country of residence. Due to the increased input in the societal language, particularly after the onset of schooling, these bilinguals often show considerable divergencies relative to the baseline language (i.e. how the language is spoken in the home country) by the time they are adults, while typically becoming dominant in the majority language of the society over time. Although formative work over the past two decades has demonstrated that adult HLs tend to diverge in various ways from their baseline languages as spoken in the country of origin, we do not (yet) have a full understanding of how precisely these grammars come to be the way they are, or how we can explain the continuum of differences between individual speakers of the same HL in the same linguistic and societal context.
In particular, little is known about the qualitative nature of the input HSs receive. Input quality refers to the richness of the input (i.e. the number of interlocutors, the amount of education in the HL, and exposure to written text, etc.), but also to the potential deviances in the input from parents and other community members due to language loss as a result of long term exposure en use of a second language. Furthermore, it is not clear yet how much of the differences found in HL grammars can be attributed to cross-linguistic influence from the majority language. Project HeLA addressed several gaps in the existing knowledge by investigating child HSs of a severely understudied age range (7-14), in which environmental and input-related circumstances are likely to have important effects on the development of the HL.
HeLA has high societal relevance, given the increasing rate of migration in Europe and the world, as well as the many native minority languages in European states (e.g. Italian dialects, Basque, Sami). Unfortunately, European policies towards minorities rarely consider language preservation as one of the tasks. The findings from this project can ge used to better inform parents, teachers and other stakeholder for heritage language communities of how to best maintain the heritage language, as well as to promote their value and elevate these languages to a position of higher importance in society.
In particular, little is known about the qualitative nature of the input HSs receive. Input quality refers to the richness of the input (i.e. the number of interlocutors, the amount of education in the HL, and exposure to written text, etc.), but also to the potential deviances in the input from parents and other community members due to language loss as a result of long term exposure en use of a second language. Furthermore, it is not clear yet how much of the differences found in HL grammars can be attributed to cross-linguistic influence from the majority language. Project HeLA addressed several gaps in the existing knowledge by investigating child HSs of a severely understudied age range (7-14), in which environmental and input-related circumstances are likely to have important effects on the development of the HL.
HeLA has high societal relevance, given the increasing rate of migration in Europe and the world, as well as the many native minority languages in European states (e.g. Italian dialects, Basque, Sami). Unfortunately, European policies towards minorities rarely consider language preservation as one of the tasks. The findings from this project can ge used to better inform parents, teachers and other stakeholder for heritage language communities of how to best maintain the heritage language, as well as to promote their value and elevate these languages to a position of higher importance in society.
In total 5 groups of approximately 30 participants have completed the experiment: 1) bilingual children in the Netherlands, 2) their Spanish-speaking parents, 3) bilingual children in the UK, 4) their Spanish-speaking parents, and 5) monolingual Spanish-speaking children in Chile. The experiment entailed a combination of oral speech production and online comprehension through eye tracking using the visual world paradigm. A background questionnaire was used to capture bilinguals’ experiences regarding their language use habits/patterns. Recently, I decided to also include a group of monolingual adults in Chile, for which data is currently still being collected. Most of the data production data have been transcribed and coded for the relevant linguistic features, and the eye-tracking data have been analyzed using (linear) mixed effects models and growth curve analyses.
Preliminary analyses reveal a high degree of variation between different heritage language bilinguals, as expected. This variation can be traced back to not only the amount of input that children receive, but also the quality of that input, as measured by the number of different interlocutors, as well as education literacy in the heritage language, confirming one of HeLA’s main hypotheses. Moreover, it was observed that the children’s parents, while they were completely target-like in terms of their production, deviated from the monolingual children regarding their eye-tracking patterns, indicating that they do not process language in the same way as monolinguals do, likely as a result of the vast amount of exposure and use of the newly learned language (Dutch / English). This is an important finding both theoretically and methodologically, because it shows that conventionally utilized offline tasks such as elicited production tasks, are not necessarily capable of capturing deviation that may be going on under the surface. Online tasks such as eye-tracking, which tap into the processing of sentence in real time, are able to detect subtle differences that may otherwise go unnoticed.
Given the large size of my dataset, I expect a total of at least six papers to result from project HeLA.
Preliminary analyses reveal a high degree of variation between different heritage language bilinguals, as expected. This variation can be traced back to not only the amount of input that children receive, but also the quality of that input, as measured by the number of different interlocutors, as well as education literacy in the heritage language, confirming one of HeLA’s main hypotheses. Moreover, it was observed that the children’s parents, while they were completely target-like in terms of their production, deviated from the monolingual children regarding their eye-tracking patterns, indicating that they do not process language in the same way as monolinguals do, likely as a result of the vast amount of exposure and use of the newly learned language (Dutch / English). This is an important finding both theoretically and methodologically, because it shows that conventionally utilized offline tasks such as elicited production tasks, are not necessarily capable of capturing deviation that may be going on under the surface. Online tasks such as eye-tracking, which tap into the processing of sentence in real time, are able to detect subtle differences that may otherwise go unnoticed.
Given the large size of my dataset, I expect a total of at least six papers to result from project HeLA.
HeLA has gone beyond the state of the art in various ways. While the majority of research has focused on adult HSs, HeLA tested HSs who are in the continued process of development, precisely in the missing period (10-12yo) that bridges the reported successes of bilingual children to the reported outcomes of arrested development in adult HSs.) Moreover, HeLA is the first study to connect children's linguistic behavior to that of their parents by comparing one-to-one dyads. Third, to isolate the role of cross-linguistic influence, HeLA has been one of very few studies to compare the same HL in different language contact situations (Spanish as a HL in the Netherlands and the UK), and targeting linguistic phenomena that are instantiated differently in English and Dutch. Finally, HeLA has applied online measures of language processing, which thus far been relatively unexplored territory in HL research, especially in this age group.
The main findings, if maintained in the final analyses, have important implications, both theoretically and societally, as they underline that variation between different heritage speakers cannot just be reduced to how much the parents speak their HL to their children; it is also the quality of that input, as measured by the number of different interlocutors, education, and literacy in the heritage language. This is vital information for member of heritage language communities, as it can be used to provide clear and specific recommendations to parents and teachers of these children.
The main findings, if maintained in the final analyses, have important implications, both theoretically and societally, as they underline that variation between different heritage speakers cannot just be reduced to how much the parents speak their HL to their children; it is also the quality of that input, as measured by the number of different interlocutors, education, and literacy in the heritage language. This is vital information for member of heritage language communities, as it can be used to provide clear and specific recommendations to parents and teachers of these children.