Chinese‐English code‐mixing: a case of matrix language assignment

Abstract

ABSTRACT: While it has been shown that bilingual code‐mixing (CM) is not random and that there exist rules that determine what code‐mixes are or are not allowed in bilingual speech, it has not been established yet whether the rules in question ‘come’ predominantly from one language, the matrix language, or whether they are evenly contributed by each of the languages participating in CM. More specifically, the issue has been whether there exists a matrix language to a code‐mixed sentence, that is, a language whose morphosyntactic rules govern the syntactic structure of CM, or whether it is the case that no feasible linguistic analysis can reliably assign a matrix language to a code‐mixed sentence and that, therefore, CM is acceptable only where it does not violate the syntactic integrity of either of the languages involved. This paper attempts to address this issue in light of data drawn from natural conversations involving Chinese–English bilinguals in Singapore. Based on this data it is argued that there exists a matrix language to a code‐mixed sentence; that bilingual speakers can intuitively identify the matrix language; and that it is this language that licenses the ‘mixability’ of linguistic elements of the other language, the embedded language. This argument receives support from the language assignment test that we conducted asking Chinese–English bilinguals to determine the matrix language in a set of Chinese–English code‐mixed sentences. This paper reports on the results of this test. It also discusses the implications of these results for the syntactic structure of CM in general, and for studies of constraints on CM in particular.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 25 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.