DYNAMICAL PREFERENCES AND SELF-ACTUATED LANGUAGE CHANGES JEROME MICHAUD *1,2 * Corresponding Author: jerome.michaud84@gmail.com 1 Department of Sociology, Uppsala University, Uppsala, Sweden 2 Department of Physics and Astronomy, Uppsala University, Uppsala, Sweden A puzzling fact about linguistic norms is that they are mainly stable, but the conventional vari- ant sometimes changes. These transitions seem to be mostly S-shaped and, therefore, directed. Previous models have suggested possible mechanisms to explain these directed changes, mainly based on a bias favoring the innovative variant. What is still debated is what is the mechanism that leads to such a bias. In this paper we propose a refined taxonomy of mechanisms of lan- guage change and identify a family a mechanisms explaining self-actuated language changes. We exemplify this type of mechanism with the preference-based selection mechanism that re- lies on agents having dynamical preferences for different variants of the linguistic norm. The key point is that if these preferences can align through social interactions, then new changes can be actuated. We present results of a multi-agent model and demonstrate that the model produces trajectories that are typical of language change. 1. Introduction An important question asked by Blythe and Croft (2012) is: ‘How many quali- tatively distinct possible mechanisms of language change are there?’ (Blythe & Croft, 2012, p. 270). Based on the generalized theory of selection by Hull (1980, 2001, 2010) adapted to language evolution by Croft (2000) and by Baxter, Blythe, Croft, and McKane (2006), Blythe and Croft (2012) proposed a classification of mechanisms influencing language change. They distinguished four categories of mechanisms, namely neutral evolution (NE), neutral interactor selection (NIS), weighted interactor selection (WIS) and replicator selection (RS). The RS mecha- nism groups all mechanisms in which the variants are treated in a different manner by agents. Blythe and Croft (2012) argued that only RS can reliably account for S- shaped trajectories of change. In their model of RS, the innovative variant is given a selective advantage that causes its directed and S-shaped propagation through the population, but they did not provide an explanation for the origin of the shared advantage of the innovative variant and assumed it as given. It is unlikely that so- cial, linguistic and cognitive factors, which can all induce a selective advantage for a variant, influence the dynamics of a language in the same way and lead, there- fore, to qualitatively different selection mechanisms. The taxonomy proposed by