The effects of heritage language instruction on first language proficiency: a psycholinguistic perspective

Abstract This study investigates the effects of weekly heritage language (HL) classes on first language (L1) proficiency in speakers who arrived in the second language (L2)-dominant setting before the onset of puberty. Two groups of L1 Spanish – L2 Swedish bilingual high school students living in Sweden participated in the study. One group currently attended HL classes once a week, whereas the other group was no longer doing so. The two groups did not differ with regard to the total number of years of HL class attendance, age of arrival in Sweden, length of residence or degree of L1 contact. Results from a grammaticality judgement test and a cloze test showed that the group that currently attended HL classes outperformed the non-attending group. Using a framework that emphasises heightened attrition susceptibility among speakers who lost contact with the L1-dominant setting before puberty, the study suggests that HL classes function as a factor that, all other things being equal, may counterweigh attrition susceptibility. Moreover, it is suggested that the effects of HL classes on L1 proficiency are short term rather than long term. That is to say, once attendance ceases the counterweighing effect is less visible. Keywords: language attritionage and language learningheritage language educationSpanishSweden Acknowledgements We wish to thank Colin Baker, Christopher Stroud and two anonymous reviewers for insightful feedback on a previous version of this paper. Needless to say, any remaining errors are entirely our own. Notes 1. Note that the studies reviewed here relate to bilingual education programmes that may not always be labelled 'heritage language programmes'. These studies are reviewed nonetheless as they illustrate the effects that the presence of L1 in a L2 school environment has on L1 proficiency. 2. Note that this variable was computed on the basis of the number of completed school years of HL instruction up until – but not including – the current school year. 3. Through the background questionnaire, we also tried to collect information about the profession/education level of the parents of the participants. Since the majority of the participants left this information out, it was excluded from the analysis. 4. One background variable that was not controlled for here is language aptitude. Even though it is common practice in attrition (as well as Second Language Acquisition) studies to assume that experimental groups and control groups do not differ in terms of aptitude, this should ultimately be subject to empirical assessments. Admittedly, the claim made in the current study about group compatibility would have been even stronger if language aptitude data had been collected. We are thankful to an anonymous reviewer for pointing this out. 5. It should be kept in mind though, as pointed out by an anonymous reviewer, that despite this not entirely unproblematic implementation policy the results of the current study clearly suggest positive effects of HL class attendance. 6. This does not, however, mean that literacy-training was the only factor for test performance, and that other activities that are not literacy-based do not have an impact on the type of language proficiency measured here. It should be recalled that medium-strong correlations were found between GJT and cloze test scores, and the variable Spanish spoken at home. Oral use of Spanish thus also played a role for test performance.