Recent and Upcoming Talks/Workshops

The social embedding of lectal coherence among mobile speakers

Abstract

Lectal coherence has long been a topic of interest in sociolinguistics (e.g., Labov 2006 [1966]), and recent research has delved into more comprehensive investigations (e.g., Beaman & Guy 2022; Erker 2017; Guy & Hinskens 2016; Tamminga 2019). This study examines the social embedding of lectal coherence among 35 Brazilian Portuguese mobile speakers from rural Northeast to an urban area in the Southeastern state of São Paulo, to discuss whether internal migrants in a dialect contact situation simultaneously adopt typical host community variants. Four sociolinguistic variables were analyzed in the corpus, balanced for speakers’ gender, age of arrival and length of residence in São Paulo: (i) /r/-fronting–the realization of coda /r/ (e.g., porta ‘door’) as aspirated or tap/retroflex (N=2,847); (ii) /s/-fronting–the realization of coda /s/ (e.g., festa ‘party’) as post-alveolar or alveolar (N=1,787); (iii) /r/ non-deletion–deletion or realization of non-morphemic /r/ (e.g., amor ‘love,’ excluding the infinitival morpheme as in amar ‘to-love’) (N=3,372); and (iv) /s/ non-deletion–deletion or realization of non-morphemic coda /s/ (e.g., mas ‘but,’ excluding the plural morpheme as in gatos ‘cats’) (N=3,822). The second variants are more common in the host community, with the first two variables reflecting North-South identities and the latter two indicating rural-urban differences. Data were analyzed in Spearman correlation tests in R, utilizing usage rates and probabilities in logodds of Southern/urban variants from multivariate logistic regression analyses. The models included Speaker and internal variables as fixed effects, lexical item as a random effect, and excluded social variables. The consistent findings reveal significant positive correlations among three out of the six possible pairings: /r/- and /s/-fronting (rho=.578, p<0.001); /r/ and /s/ non-deletion (rho=.545, p<0.001); and /s/-fronting and /r/ non-deletion (rho=.431, p<0.01) (Figure 1). Further analyses based on social groups show that, regarding Gender, women tend to be more coherent than men, as women exhibit covariation for these three pairs and between /s/-fronting and /s/ non-deletion; in contrast, men only demonstrate coherence in the use of /r/- and /s/-fronting (Figure 1a). Regarding Age of Arrival, speakers who arrived earlier (before 19 y.o.) display higher coherence than those who arrived later (after 20 y.o.), as the former exhibit significant covariation for four pairings (the above three pairs and between /r/-fronting and non-deletion) (Figure 1b). Lastly, regarding Length of Residence, speakers who have lived longer in the new community (10+ years) are more coherent than those with a shorter length of residence (-9 years), the former exhibiting covariation for three pairs and the latter for none (Figure 1c). The overall findings demonstrate that migrants exhibit a coherent pattern of adopting or rejecting Paulista (fronted /r/ and /s/) and urban traits (/r/ and /s/ non-deletion), and suggest that similar phonological processes (fronting and non-deletion) promote lectal coherence. The results also indicate that dialect contact initially disrupts lectal coherence, as migrants don’t adopt new variants simultaneously. Lectal coherence is restored over time and is contingent on speakers’ gender and age of arrival. This study thus enhances our comprehension of how coherence is preserved, disrupted, and attained.

Date
Location
Queens College, New York City-NY, USA