The internal syntax of Q-words

Abstract This paper aims at describing Q(uantity)-words, i.e. many/much and few/little , from a typological perspective, and presenting typological generalisations based on it. The typological sample provides support for a mass-count and positive-negative dimension in the domain of Q-words. Both dimensions also intersect. Along the negative dimension, it seems that languages fall into two groups: those having an opaque strategy for few/little and those having only an analytic strategy ( not-much/many ). Four patterns can be discerned on the basis of the sample, which are each exemplified by means of one language, i.e. English, Dutch, Wolof and Western Armenian. In addition, I make an attempt at developing a nanosyntactic analysis of the data, which aims to show how language variation in the domain of Q-words can be accounted for in terms of varying the size of lexically stored trees ( Starke 2014 ). Finally, I show how one missing type of pattern is underivable on the basis of the analysis proposed.