Input effects and unidirectional transfer in Mandarin–English heritage bilingual children: The case of spatial prepositional zai-phrases

Aims and objectives: This study investigates the effects of fine-grained aspects of input and cross-linguistic influence on the placement of prepositional zai-phrase ( zai-PP) in heritage Mandarin. Methodology: Utterances with zai-PPs were extracted from transcripts of an elicited narration task administered with child heritage speakers of Mandarin ( n = 27, aged 4–14) and their parents ( n = 18) in the United Kingdom, and compared with those produced by child and adult Mandarin speakers in Beijing in a similar narration task, selected from the Child Language Data Exchange System (CHILDES). Data and analysis: The zai-PPs were coded for pre/postverbal placement and event semantics of co-occurring verbs, as well as innovative uses. Comparisons were made between groups of participants, and correlations were performed among child and input variables. Findings: In Mandarin, the heritage children overused postverbal zai-PPs across measures and comparisons, producing innovative uses of postverbal zai-PPs not found in the input or Beijing baselines. Their production of preverbal zai-PPs increases with age (proxy for cumulative amount of input) rather than structural frequency in their parental input. In English, they correctly placed spatial PPs postverbally. Our findings revealed protracted development of zai-PP placement in heritage Mandarin due to inherent complexity of the structure, reduction of input in the heritage language at school age, and strong unidirectional transfer from English to Mandarin. Originality: This study is among the first to understand a heritage structure by analysing speech samples of both heritage children and their parents and examine input–outcome relations between them directly. Significance: The placement of zai-PP in relation to individual verbs with different event semantics is subject to protracted development and vulnerable to majority language influence in heritage language acquisition, but further development at the school age is likely, if sufficient amounts of input are provided.