As I'd already hinted at in a previous post, over the last year or so, I've grown more and more interested in a possible interaction between quantitative and qualitative linguistics. The phrase I've been using to sum up my approach is "quantitative methods in theoretical linguistics" (or QMTL for short). Note the figure-ground relation between the two DPs in this phrase: theoretical linguistics is the background against which this research takes place. I am and remain first and foremost a formal linguist, and the goal of my writings is and remains to uncover and model the formal system underlying human language. At the same time, though, formal linguists cannot remain blind to the myriad of sophisticated and useful techniques that are out there and that are driving large portions of the field that are not under the rubric of 'theoretical (or formal) linguistics'. I believe that a genuine mututally beneficial interaction between the two is possible and intend to pursue this in the years to come. Last year I've given a number of talks about this topic, and now I've (finally) written them up into a paper. Here's the abstract:

"This paper combines quantitative-statistical and formal-theoretical approaches to language variation. I provide a quantitative analysis of word order variation in verb clusters in 267 dialects of Dutch and map the results of that analysis against hypotheses extracted from the theoretical literature on verb clusters. Based on this new methodology, I argue that variation in verb cluster ordering in Dutch dialects can be largely reduced to three grammatical parameters."

It's a first draft, so there's still a lot of room for improvement (if you have specific suggestions, let me know), but all in all I'm pretty happy with how it turned out and look forward to continuing along this path in future work.

  1. In Dutch we have the eminently more useful word wisselwerking to describe this concept.