The rise of articles in Indo-European
One of the central goals of historical linguistics is to distinguish probable from improbable linguistic changes. This includes not only individual changes, but also interactions among changes (i.e., whether one change makes another more or less probable). Until recently, there was no method for assessing dependencies among linguistic changes. The advent of Bayesian phylogenetics in historical linguistics has delivered just such a method. In this paper, I take up the diachronic interaction of definite and indefinite articles in Indo-European. It has been claimed (e.g., Irslinger 2013:46) that their development was independent—that is, that the rise of one did not impact the rise of the other. I demonstrate that there is in fact a dependence among the rates at which the two types of articles develop. Indefinite articles developed at a faster rate among languages that had already developed a definite article. The predominant diachronic trajectory in the history of Indo-European was accordingly for definite articles to precede indefinite.