The Parametric Comparison Method (PCM, Guardiano & Longobardi 2005, Longobardi & Guardiano 2009) grounds on the assumption that syntactic parameters are more appropriate than other traits for use as comparanda for historical reconstruction, because they are able to provide unambiguous correspondences and objective measurements, thus guaranteeing wide-range applicability and quantitative exactness. This paper discusses a set of experiments explicitly designed to evaluate the impact of parametric syntax in representing historical relatedness, and performed on a selection of 26 contemporary Indo-European varieties. The results show that PCM is in fact able to identify correctly genealogical relations even from modern languages only, performing as accurately as lexical methods, and that its effectiveness is not limited by interference effects such as ‘horizontal’ transmission. PCM is thus validated as a powerful tool for the analysis of historical relationships not only on a long-range perspective (as suggested by Longobardi & Guardiano 2009), but even on more focused, though independently well-known domains.
