Abstract
It is estimated that more children in the world are exposed to multiple languages than are exposed to only one language (Crystal, 1997), and recent increases in immigration rates worldwide suggest the number of children in this circumstance will grow (United Nations and OECD, 2013). Against that backdrop, six empirical articles and a final commentary in this special section address questions of how languages are learned and used in by children living in multilingual environments. In truth, all the articles are actually about children in the simplest form of multilingual environment—a bilingual environment. The title of this section is intended to indicate the larger topic these articles begin to address. Given that research on children’s language was almost exclusively the study of monolingual development not so long ago (Bhatia & Ritchie, 1999), the increasing attention to bilingual development is progress toward studying language development in the range of environments that exist.
The articles in this special section are about children in different countries, children at different points in the language acquisition process, and children who are acquiring different pairs of languages. The articles are also about different components of what children learn in acquiring language. The language outcomes in these studies include acquisition of the sound categories of the target language(s) (Fennell & Byers-Heinlein; Silvén, Voeten, Kouvo, & Lundén), vocabulary learning and growth (Bosch & Ramon-Casas, Fennell & Byers-Heinlein; Jia, Chen, Kim, Chan, & Jeung; Silvén et al.), receptive and expressive language skill (Ribot & Hoff), the acquisition of grammatical devices (Paradis & Kirova), skill at using language to tell a structured, coherent story (Paradis & Kirova), and one outcome unique to speakers of multiple languages: the choice of which language to use when speaking (Ribot & Hoff).The articles in this special section ask how children acquire these multiple skills in two languages, how skills in one domain influence skills in other domains, and how skills in one language are related to skills in the other language. Many of these articles also tackle the question of how to characterize children’s language environments in multilingual contexts: How much of each language do children hear (Ribot & Hoff)? Is the speech children hear produced by native speakers or non-native speakers (Paradis & Kirova)? By bilingual or monolingual speakers (Fennell & Byers-Heinlein)? Is it heard at home from parents or outside the home from peers (Paradis & Kirova)? Or in heritage language lessons (Jia et al.)? These articles ask how these features of children’s language exposure influence the children’s development of skill in two languages.
Fennell and Byers-Heinlein studied word learning in French-English bilingual and English monolingual infants at 17 months and found evidence that, at this age, children’s representations of the sounds of their language are very tied to the nature of the speakers familiar to them. Infants were better able to learn new words spoken in the accent of familiar speakers than new words spoken in an unfamiliar accent. For monolingual children, the accent of monolingual speakers is more familiar, but for bilingual children it is bilingual speakers who are more familiar. But the full story is more complicated than that. Bilingual children who were English-dominant performed better with monolingual English stimuli than bilingual children who were French-dominant, and there was a significant positive correlation between the amount of English infants heard from monolingual English parents and their task performance, while there was no significant correlation between their task performance and the amount of English they heard from parents who were not raised as English monolinguals.
Bosch and Ramon-Casas used parents’ reports of their Spanish-Catalan bilingual children’s vocabularies at 18 months to ask whether form similarity between a bilingual child’s two languages influences early lexical growth. Spanish and Catalan share many cognates, and to the extent that these children had acquired words that do double duty, they had larger total vocabularies early in development than did the monolingual comparison group.
Silvén et al. studied Finnish-Russian bilingual children and Finnish monolingual children from infancy to three years and found, in contrast, that early in development the bilinguals’ total vocabularies were comparable to the monolinguals’ vocabularies, and that their skills in Finnish lagged behind those of monolingual Finnish children. They found that early speech perception skills in Finnish predicted Finnish-Russian bilingual children’s later acquisition of vocabulary in both languages. And, similar to other findings in this special section, they found that bilingual children quickly became dominant in the language of the community.
Ribot and Hoff studied Spanish-English bilingual 2½-year-olds in the U.S., focusing on a particular feature of language use that characterizes some bilinguals: they appear to understand two languages but tend to speak in only one. Ribot and Hoff found that such a pattern of language choice can be associated with a particular profile of bilingual skill, namely equal levels of receptive skill in two languages but stronger expressive skills in one compared to the other.
Paradis and Kirova also described what are termed “profile effects” in bilingual children (Oller, Pearson, & Cobo-Lewis, 2007). In the case of the 4½ to 5½-year-old children of immigrant families acquiring English in Canada that Paradis and Kirova studied, the nature of the profile effects was that the narrative story grammar scores of the bilingual children were close to monolingual levels, although their sentence-level skills were substantially below monolingual levels. The patterns of language skills in the bilingual children in both these studies show that the multiple domains of language proficiency (receptive and expressive skills in Ribot & Hoff, narrative and grammatical skills in Paradis & Kirova) do not have to develop in lock step and, as a result, bilingual children’s skills are not monolithic. Of course, monolingual children also can have relative strengths and weaknesses, but these data suggest that the multiple components of language development potentially diverge more in bilinguals than in monolinguals. Paradis and Kirova suggest that the skills that are more advanced in children’s second language are those that can make use of understandings already acquired in the first language, such as the understanding of narrative structure, while skills that are less advanced in the second language are skills that need to be learned anew for each language.
Jia et al. describe age-related differences in heritage language lexical skill among school-age children of Asian immigrant families in the U.S. They found, as have others, that heritage languages are difficult to maintain. The children’s English dominance increased across age groups, and they found that explicit heritage language instruction did not change this outcome. Jia et al. also found that use of English at home benefitted the English skills of younger children, but not older children. Instead, among older participants, more English use at home was associated only with weaker heritage language skills.
In a final section, Gathercole provides commentary. As her title suggests, she finds that the articles in this special section, while adding to our understanding of bilingual environments and bilingual development, also add to our appreciation of the heterogeneity of bilingual environments and the corresponding heterogeneity of bilingual children’s language skills. Gathercole brings order to this heterogeneity, identifying common themes in the results of these studies, connecting them to the larger literature on bilingual development, and drawing lessons for the design of future research.