Mood Variation Among English-Spanish Bilinguals in Southern Arizona

Abstract

Variation in Spanish spoken in the United States has been explored through phonological, morphosyntactic, and discourse variables in different communities across the country. Among the different morphosyntactic variables studied, mood selection has been widely investigated not only in bilingual communities in the US but also monolingual communities elsewhere.Mood selection has been traditionally approached from a prescriptive perspective and a normative effort has been made for dictating when subjunctive should and should not be used. Based on obligatory and variable contexts of subjunctive, several studies claim that US Spanish speakers overextend the use of indicative in contexts where subjunctive is expected. These studies (e.g., Lynch, 1999; Montrul, 2007, 2009; Silva-Corvalán, 1994a, 1994b) often conclude that the overextension of indicative is due to contact with English, claiming that US Spanish speakers’ use of subjunctive is the result of simplification or even incomplete acquisition. Nevertheless, mood selection in Spanish is much more variable than prescribed and is conditioned by both linguistic and extralinguistic factors. As argued by Poplack et al. (2013), there tends to be very little overlap between prescription and praxis. Via a variationist lens, studies (e.g., Bessett & Carvalho, 2022; Otheguy, 2016) have discussed that these claims are grounded in the comparison of bilingual data to idealized monolingual standards since variation in mood selection is in fact not explained by the binary between optional and obligatory contexts. Uncovering the usage patterns usually hidden by prescriptive and deficit-oriented accounts, this dissertation examines the linguistic and extralinguistic conditions in which mood selection takes place in US Spanish, specifically in the variety spoken in Southern Arizona. Following the quantitative methods of Language Variation and Change (LVC) research, it investigates the behavior of mood selection in US Spanish in complement clauses according to linguistic (i.e., lexical identity of the governor, governor frequency, polarity, coreferentiality, and morphological form of the embedded verb) and extralinguistic factors (i.e., immigrant generation, language dominance, and language use in the family). Results reveal the systematicity and complexity of mood variation in Southern Arizona, which is determined by the lexical identity of the governor and significantly conditioned by governor frequency, polarity, and coreferentiality. Considering the comparison of such results to previous research, findings highlight the similarities across Southern Arizona and monolingual and other bilingual Spanish varieties. In terms of extralinguistic conditioning, all factor groups analyzed are shown to be statistically significant to the variable output. However, language use in the family is found to be the one that best explains variance in the data set, calling attention to the role of language socialization to mood variation in US Spanish. Ultimately, this dissertation provides evidence for the integrity of the variable grammar of US Spanish, specifically regarding mood selection.