This project seeks to bridge the gap between formal-theoretical and quantitative-statistical linguistics. It combines a quantitative methodology with hypotheses taken from the formal-theoretical literature. On the one hand it investigates to what extent theoretical analyses of language phenomena can narrow down the hypothesis space created by quantitative-statistical analyses of large digital collections of language data, but on the other hand the quantitative findings also serve as a testing ground for the linguistic hypotheses. The project thus tries to establish a fruitful bidirectional collaboration between two subdisciplines of linguistics that traditionally do not communicate or collaborate much with one another. The research domain of the project will be verb clusters in Dutch dialects.