PERFECT microvariation in Madrid Spanish: a corpus-experimental approach

Abstract

Romance languages vary considerably in how broadly they use the have-perfect, forming a continuum from highly restricted to highly liberal systems. Within Spanish, significant microvariation exists across varieties, but the contextual factors that influence this variation are not yet fully understood. This study investigates perfect use in Madrid Spanish, a Peninsular variety with a relatively liberal distribution of the perfect, through a two-stage design combining a parallel corpus comparison with an experimental task. The corpus data show that Madrid Spanish licenses the perfect in a wider range of contexts than Standard Peninsular Spanish, especially in hodiernal and narrative contexts, aligning more closely with Catalan in its licensing of the perfect. The experimental findings confirm that perfect selection is systematically sensitive to core perfect readings, hodiernality, narrativity, lexical aspect, and temporal marking. Notably, none of these factors categorically block or require the perfect, rather they influence the strength of speakers' preferences, suggesting that perfect use reflects flexible, context-sensitive preferences rather than rigid grammatical constraints. Overall, these results contribute to a more precise understanding of perfect microvariation within Spanish and demonstrate the value of context-sensitive, speaker-oriented approaches to tense-aspect research.