On the substantive primitives of morphosyntax and their parametrization: Northern Italian subject clitics

We illustrate our general approach in relation to an apparently trivial variation problem, involving the lexicalization of subject clitic pronouns in Northern Italian varieties. We show the empirical inadequacy of views based on implicational generalizations (Renzi and Vanelli 1983) despite the fact that they are upheld by recent formal work (Cardinaletti and Repetti 2008). Empirical problems are not completely resolved under the Distributed Morphology view of Calabrese (2008b, to appear) either, involving markedness hierarchies and repairs. We argue that a simple Boolean organization of categorial (i.e. conceptual) space is best at modeling variation.