Classifiers and the mass-count distinction in Uzbek

The paper argues for the existence of the (NP-level) mass-count distinction in Tashkent Uzbek, an obligatory classifier dialect. Evidence is provided based on the distribution and interpretation of modifiers, classifiers and quantificational suffixes of different types, as well as properties of flexible and object mass nouns. A formal analysis is further provided that treats classifiers as sensitive to the mass-count distinction but not uniformly serving as “individuators” of otherwise mass nouns. Sortal classifiers are argued to differ from mensural ones in that only the latter contribute a measure function.