The linguistic and social factors that drive the choice between a velarized and a non-velarized variant are explored through a series of GoldVarb analyses (Rand and Sankoff 2001) in three Salvadoran communities. Results showed that the probabilistic weighs that favored a velar realization (syllable stress, preceding segment, following segment, and age of arrival) align neatly across the groups under review. Frequency and statistical analyses results showed that while velarization rates differ across communities, the constraints are uniformly maintained. Salvadoran speakers in Houston have lower frequencies of word final nasal velarization than in San Sebastián, but much lower frequencies of velarization occur in Mexican-Salvadoran exchanges in Houston when compared to more intimate Salvadoran-Salvadoran exchanges.
Selected Proceedings of the 13th Hispanic Linguistics Symposium
edited by Luis A. Ortiz-López
Table of contents