Social variation of code-switching acceptability

Aims and objectives: Empirical research on code-switching acceptability tends to focus on structural differences and has only recently begun to focus on social motivations of what is considered grammatical/acceptable in bilingual speech. We study perceptions of code-switching grammaticality with a focus on the social identity of the perceiver. Methodology: A modified Acceptability Judgment Task was designed and implemented via Qualtrics, and data were collected from Latinx Spanish–English bilinguals across five U.S. states (California, Florida, Illinois, New York, Texas). Participants responded to four grammaticality judgment questions for 40 code-switched sentences adapted from reggaetón music lyrics. Half the grammatical and ungrammatical sentences were labeled “(source: reggaetón).” Data and analysis: Data were analyzed using linear mixed-effects regression in R. The fixed effects were place of birth, state of residence, family national origin, gender, age, and contextual label, while the random effects were participant and item (sentence). Findings: U.S.-born participants had significantly lower AJT ratings for ungrammatical sentences, which were significantly lower when the label was present. Participants from California demonstrated the highest AJT ratings for grammatical and ungrammatical sentences. Interaction effects between state of residence and label status are found only for participants from Florida in the ungrammatical category. Originality: This paper presents one of the first studies to address the social and regional variation of AJT results in the U.S. concerning Spanish–English code-switching, and it is one of the first to emphasize the role of context in acceptability judgments. Implications: The results of this study make connections to sociolinguistic theory on the listener’s identity and its effects on language perception. Furthermore, the study highlights the importance of considering context-based information in acceptability judgment testing.