An exploratory analysis of code-switching and borrowing in a corpus of Zimbabwean English

Abstract

• Focusses on code-switching and borrowing between Zimbabwean English and Shona. • Intrasentential code-switching was the most frequent type of switching. • Spoken registers had more switches due to their informal and spontaneous nature. • Multi-word switches occurred more often due to their frequency in public dialogues. • Social functions namely accommodation and solidarity had the highest frequencies. This paper explores patterns of English-Shona code-switching and borrowing in a corpus of Zimbabwean English to determine the types, frequencies, and functions of these features. Results indicate that intrasentential code-switching was the most frequent type of switching followed by intersentential code-switching and then lexical borrowing. The least frequent type of switching was loanword adaptations. The analysis revealed that spoken registers had the highest frequency of switching compared to written registers. The informal and spontaneous nature of spoken registers were cited as motivations for the high frequency. Subsequently, multi-word switches occurring in the same place were attested more, which can be linked to their higher frequency of occurrence in public dialogues. Social functions, namely accommodation and solidarity, had the highest frequencies whilst switching at a time of vagueness or uncertainty was the second highest function of code-switching. Other functions of code-switching attested in the Zimbabwean English corpus, in the order of frequency, include repetition or reiteration, asking questions, identity marking, code-switching to show emotions, code-switching as discourse marking, clarification, and emphasis. Analyzing the functions of code-switching in the Zimbabwean English corpus and determining the frequency of each function is crucial for understanding the linguistic-, cognitive-, and sociocultural dimensions of bilingualism in Zimbabwe.

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 — 1 citations exist upstream.

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