Edited by Jeremy King and Sandro Sessarego
[Current Issues in Linguistic Theory 340] 2018
► pp. 299–333
This study adopts a quantitative methodology to investigate variation between null and overt expressions of Spanish complementizer que. I focus on two varieties of Spanish from Mexico City and Los Angeles, California. This research uses a corpus linguistics–inspired methodology, relying on Twitter as a database to extract tokens and Rrbul (Johnson 2008) to perform the multivariate analysis. I extracted and analyzed 1,505 tokens and coded for 9 linguistic and 2 extra-linguistic factors. I showed the internal factors ‘verb modality’ and ‘embedded subject status’ to significantly affect the alternation, favoring the null option. Contrary to what was expected, the external factor ‘city’ did not have any significant effect on the distribution. I discuss the results in light of formal and sociolinguistic studies on the nature of subordinated constructions in Spanish (Brovetto 2002; Etxepare 1996; Subirats-Rüggeberg 1987), English (Tagliamonte & Smith 2005) and Spanish in contact with English (Silva-Corvalán 1993, 1994, 1998).