Linguistics Graduate Student Colloquium

Linguistics PhD students Wil Gonzales and Felicia Bisnath will present their research.

Wilkinson Daniel Wong Gonzales, University of Michigan
Sociolinguistic variation in a mixed language? A corpus-based analysis of Lánnang-uè conjunctions and prepositions


ABSTRACT
For several decades, research has shown that sociolinguistic factors play an important role in language variation (Weinreich et al. 1968; Eckert 2005). However, prior sociolinguistic research has primarily focused on well-documented varieties in Western contexts (e.g., American English).

In this presentation, I analyze the variation in a low-resource, previously undocumented “mixed language” in the Philippines called Lánnang-uè – a variety that systematically derives linguistic elements from Hokkien (Southern Min), Mandarin, English, and Tagalog (Gonzales 2018; Gonzales and Starr 2020; Gonzales 2022a). Specifically, I focus on the patterns of variation in two lexical categories: conjunction and prepositions – two categories that show higher rates of variation compared to other features in Lánnang-uè.
Using a mix of quantitative (i.e., corpus-based, computational) and qualitative (i.e., ethnographic) approaches, this analysis investigates the impact of four factors – age, sex, self-reported language proficiency in the source languages, and language attitudes – on the variation observed. I pre-processed, machine-tagged, and statistically analyzed conjunction and preposition data from the Lannang Corpus (LanCorp) (Gonzales 2022b) – a self-compiled 375,000-word corpus of Lánnang-uè, acquired from 135 Lánnang-uè speakers. I also analyzed metalinguistic commentary from a subset of these speakers in an attempt to provide a more holistic explanation for potential sociolinguistic patterns.

The findings indicate that variation in the use of conjunctions and prepositions can be explained by at least one of the four enumerated sociolinguistic factors, corroborating my previous work on Lánnang-uè (Gonzales 2018; Gonzales and Starr 2020) and other research on related contact varieties in East Asia (Hansen Edwards 2019; Starr and Balasubramaniam 2019; Lee 2014). However, I also found that the effects of age, sex, language proficiency, and attitudes varied depending on many context-specific factors (e.g., degree of awareness, stylistic practices unique to a particular social group). I discuss the sociolinguistic patterns uncovered in my presentation in light of cognitive, sociolinguistic, and contact linguistics theories, and conclude by briefly identifying potential avenues for future research.


Felicia Bisnath, University of Michigan
Mouthing constructions in 37 signed languages: typology, ecology and ideology

ABSTRACT
Sign languages – like creoles and other contact languages– are minoritised in their communities and in linguistics. This makes perspectives on creoles potentially illuminating to the study of sign languages. A common way that sign languages are categorised, based on social criteria, is into deaf and rural sign languages. This distinction highlights relationships between social and linguistic properties. This paper investigates one such relationship motivated by the literature: namely whether the extent of contact with spoken language(s) via institutionalised education translates into a higher prevalence of the silent articulation of spoken words, mouthing. Across 37 sign languages (26 deaf; 11 rural), mouthing was found to be prevalent regardless of language type, having been reported in 35 languages (25 deaf; 10 rural). This suggests that differences in contexts of language emergence that have been used to motivate a typological separation between deaf and rural sign languages does not equate to a structural difference in terms of the structural property, mouthing.