The rise of the verbal weak inflection in Germanic. An agent-based model

The verbal weak inflection, one of the defining innovations of Proto-Germanic, currently holds a dominant position in the verbal inventory of most remaining Germanic languages. This has not always been the case, though. This paper investigates how the weak inflection could have grown to overthrow its competitor, the strong inflection, even if (i) the strong inflection was still more regular and (ii) the weak inflection had to start from a position vastly inferior in frequency to any strong ablaut class. As opposed to earlier work, which focused on language acquisition in models of iterated learning, our focus lies on language usage, which is why we have composed an agent-based model. This enabled us to test a number of minimal assumptions needed to explain an ascent of the weak inflection, of which several have been proposed in the literature. It was found that the weak inflection's functional advantage of general applicability is sufficient by itself already. That is, the weak dental suffix is in principle applicable to all verbs, while each separate strong ablaut class is not. This is shown to put the strong inflection at a crucial disadvantage, even if (i) the strong system as a whole is applicable to all verbs, and (ii) each separate ablaut class starts out as dominant in both type and token frequency over the weak dental suffix. There is no need to assume that the strong system has irregularized for the weak inflection to get airborne; this irregularization may rather be the result and subsequent catalyst of the rise of the weak inflection.status: publishe