Geographic variation of voseo on Spanish Twitter

Abstract

This study uses the social media platform Twitter to analyze geographic variation in the use of voseo throughout the Spanish-speaking world. It forms part of the first-ever voseo census (Morgan et al., 2017a, 2017b), which triangulates multiple sources to calculate the number of voseantes worldwide and to document their morphological and attitudinal preferences. We compiled a corpus of 32 million geocoded tweets, searched within it for ten linguistic variables – the pronoun vos and associated verbal morphology – and estimated usage rates by country. We found widespread voseo in the River Plate basin and Central America, but with important differences in rates of indicative, subjunctive, and imperative forms. We also address methodological questions in the use of big data for dialectological research.