Decomposing habituals

Abstract The goal of this chapter is to provide a Nanosyntactic hierarchy of features for imperfective categories in natural language. Building on Starke’s (2021) proposal to derive progressives from habituals, the authors propose that habituals are not uniform but come in two shapes. Reporting data from languages that provide two formally distinct categories of habituals, they identify their semantic correlates. In particular, the authors argue that the habitual category that expresses incidental generalizations is structurally more complex, containing the habitual category that expresses rule-like generalizations that project into the future. Hence, the chapter provides an empirically well-grounded enrichment of the fseq, making predictions about possible types of syncretisms in the domain of imperfectives.