Referent Reintroduction in the Japanese Narratives of Bilingual Children: The Relationship Between Referent Accessibility and Explicitness

Studies of children’s narratives have shown that selecting appropriate forms to reintroduce a referent compared to retaining a referent is challenging because it requires the integration of different accessibility features. Bilingual children are more explicit than their monolingual peers when it comes to referent selection, especially in the context of reintroduction in null-argument languages. Whether different accessibility features influence referent choice in the context of reintroduction in bilingual and monolingual children remains to be investigated. Japanese narratives were elicited from Japanese–English school-age early bilinguals (n = 13) and their monolingual peers (n = 8) using a wordless picture book and video clip, and the linguistic means of referent reintroduction were analyzed in terms of recency, ambiguity, and pragmatic predictability. The analysis revealed that in terms of recency, bilinguals used more noun phrases (NPs) than null forms when the referent was highly accessible, thus exhibiting overexplicitness, whereas in terms of ambiguity, bilinguals used more NPs for less accessible referents, while monolinguals were not as explicit. Both groups were sensitive to accessibility. We argue that bilinguals are selectively redundant, suggesting that the overproduction is not due to the load of processing two languages but is a manifestation of cross-linguistic influence modulated by accessibility features. The results emphasize the importance of considering discourse features in identifying overexplicitness.