A corpus investigation of the typology of code-switching between closely related languages: Data from Mandarin–Taiwanese code-switching

Aims and objectives: The study provides a documentation of the code-switching patterns in typologically similar languages and compares the code-switching patterns between typologically distant versus typologically similar languages. Design: The study examines the scale, type, matrix language, and syntactic categories of spontaneous Mandarin–Taiwanese code-switching as well as their interaction with the speakers’ language dominance and the main language of the conversation. Data and analysis: The study utilized data from a Mandarin–Taiwanese bilingual corpus of 16,186 utterances and conducted frequency analyses. Findings: Due to the large lexical gap in Taiwanese and the higher sociolinguistic status of Mandarin, code-switching between Mandarin and Taiwanese tends to be intra-sentential, classic—with a non-negligible proportion of composite code-switching, and in the direction of Taiwanese to Mandarin. When compared to the code-switching between typologically distant languages, Mandarin–Taiwanese code-switching tends to occur on a smaller scale as most of the translational equivalents in the two languages have a shared syntactic frame, which poses little constraint as to where code-switching can occur. Originality: This study presents a systematic corpus examination of the typological characteristics of code-switching between typologically similar languages. Significance: The results suggest differences in the code-switching patterns in typologically similar versus distant languages and invite future research to reconsider the applicability of the current models of code-switching when studying bilingual speech in typologically similar languages.