The evidence for word order correlations

Dunn et al. (2011) argue for the following four theses. First, they argue that there is no evidence of a crosslinguistic word order correlation between certain pairs of elements for which a correlation has been claimed in the previous literature. Second, they argue that although some correlations do exist, they are specific to individual families and not universal. Furthermore, they argue that, at least for the data they consider, these correlations are never found in more than half of the families they examine and, in the majority of cases, in only one family. Third, they argue that there is evidence in some families for correlations that they claim the existing literature predicts should not exist. Fourth, they argue that cognitive factors do not play a major role in determining linguistic structure, at least as far as word order is concerned, and that (in their words) “cultural evolution is the primary factor that determines linguistic structure, with the current state of a linguistic system shaping and constraining future states”. For reasons of space and time, I will only be able to address some of their claims in this commentary and will address other issues elsewhere. I will argue in Section 2 that Dunn et al. are mistaken in the first of their claims; I will argue that there is good evidence for at least two crosslinguistic correlations; data in Dryer 1992 provides evidence for other correlations. I will also argue in Section 2 against their third claim: I will argue that the correlations which they find and which they claim contradict existing claims in the literature are in fact fully consistent with what is claimed in the existing literature. I will argue in Section 3 that Dunn et al. in fact cannot account for the existing data on word order correlations. In Section 4 I will address their second claim and will argue that the number of families they claim exhibit a correlation, as defined by their methodology, is approximately what we would expect under the view that