Leopard spot variation: what dialects have to say about variation, change and acquisition.

In this work I intend to discuss some general properties of dialectal variation which have import on the theory of language acquisition. I will first briefly discuss whether dialectal variation is qualitatively different from typological variation, a prediction made by a theory which assumes that there exist macro and microparameters, as dialectal variation should only display microparameters. As this is not the case, I argue against the idea of a hierarchical order of parameters that the child has to set progressively in language acquisition, and in favor of an idea of a constellation abstract properties connected by the fact that they share some primitive components that cluster to form what we call “formal features” in a way similar to articulatory features cluster to combing phonemes. The second part of the article deals with what is indeed peculiar to dialectal variation, namely the type of distribution called by traditional dialectologists “leopard spots” and puts forth the idea that it can be used to formulate more precise hypotheses on the structure of functional items inside a framework like the cartographic/nano-syntax one.