Intervocalic voiced stops in Yucatan Spanish: A case of contact-induced language change?

Abstract

This study examines the possible role of language contact on the realization of /b d g/ in Yucatan Spanish (YS).Whereas standard Spanish displays an alternation between stops [b d g] and fricatives [β δ γ], YS shows a preference for stops in contexts that would require a fricative in other varieties.The extended use of [b d g] in YS has been attributed to influence from the contact language, Mayan, by some researchers (Nykl 1938, Mediz Bolio 1951, Lope Blanch 1987), while others prefer a language-internal explanation (Cassano 1977, Yager 1982).Therefore, the present study addresses the following research questions: Could the observed pattern for voiced stops be the result of language contact with Mayan?Is Mayan in a position to have influenced YS, thereby making a contact-based explanation plausible?Is there quantitative evidence of a link between speaking Mayan and higher rates of stop variants?Using the criteria established by Thomason (2001) for determining the possibility of contact-induced change, the study finds that there has been sufficient contact on both an individual and a societal level to warrant a contact-based explanation for [b d g].Especially important was the previous trend to hire Mayan-speaking nannies in middle and upper class homes.Based on data from 40 sociolinguistic interviews, this study finds a significant effect for Mayan-Spanish bilingualism on the production of stops.Speakers over the age of 30, exposed to more Mayan and Mayaninfluenced Spanish, also produce significantly more stops.Based on data from this study and the presence of similar patterns in other bilingual regions, this study concludes that the higher rates of [b d g] in YS are due to language contact via shifting second language (L2) speakers of Spanish, but not due to specifically Mayan influence on the dialect.

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 — 38 citations exist upstream.

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