Tid: Torsdag 24 september 2015, kl 15:00-17:00
Plats: C397, Södra huset, Frescati

Postseminarium följer direkt efter seminariet i institutionens pentry.


Language contact is widely believed to accelerate change in one or more of the languages involved. In a language shift situations, change is generally attributed to innovations that may occur when speakers adopt a new language, during the second language (SLA) acquisition process. If this hypothesis is correct, then the rate of language change should be related to the rate at which L2-speakers enter the population. This idea has until recently been based on mostly historical reconstructions of contact situations where data on the numbers of L1- and L2-speakers is scant or nonexistent. In this presentation I suggest a way to model the mechanism that makes SLA accelerate language change on a population level in a way that is formally testable. In order to do this I use a simulation model of linguistic interaction, where we let L2 speakers introduce new linguistic forms while they acquire the language, in a growing population constituted of L2 as well as L1 speakers. I test the model by comparing its predictions to diachronic linguistic and demographic data from the ongoing language shift from Bantu Languages to Portuguese in Maputo, Mozambique.

I will also discuss typological differences between contact-induced and non-contact-induced language change. Earlier models have shown that specific conditions, in terms of populational biases and/or network position of the innovator, are required for a single linguistic innovation to spread through a larger population. I propose that a characteristic that distinguishes contact-related language change driven by L2 speakers from language change where no contact in involved is that, in the contact situation, a novel variant is introduced on multiple occasions by different individuals, independently of one another.


Ljuba Veselinova