Formal syntax as a phylogenetic method.

Since the beginning of the historical-comparative paradigm and the early attempts at scientific language phylogenies, the study of syntax has hardly ever played any role in these investigations. Until a few decades ago, this attitude could be justified with the lack of syntactic models with the characters of modern science: pattern regularity, universality, precise and quantifiable hypotheses. However, for at least the past 40 years, formal linguistic theories have attempted to remedy these shortcomings. Especially since the development of parametric models (Chomsky 1981), the theory of generative grammars, in the search for descriptive and explanatory adequacy (Chomsky 1964), has successfully addressed many issues of cross-linguistic syntactic diversity. Yet, even this broadly comparative framework has remained largely synchronic, or diachronic only in the sense of trying to explain individual instances of local changes (i.e. between two contiguous stages of the ‘same’ language: Lightfoot 1979, 1998, 2006, Roberts 2007, a.o.): in other words, it has so far continued the century-long tradition of regarding syntax as a domain irrelevant for capturing generalizations about the historical relatedness of languages.