The complexity of knowledge sharing in multilingual corporations: evidence from agent-based simulations
This article discusses the possibility of adopting a complexity theory approach to the study of language policy and planning (LPP). Besides, it argues that agent-based modelling provides a significant support in this sense. Indeed, while agent-based modelling has become a major ally of researchers in the social sciences, it remains largely unexploited in the study of language-related issues in society. As a central tool of complexity theory, agent-based models (ABMs) lend themselves particularly well to the study of all sorts of complex systems. To provide justification for the use of ABMs in LPP, I show how language issues display the typical traits of complex systems and how ABMs can easily translate ideas and notions from the literature into computer-simulated processes. To support my argument, I discuss communication within multinational corporations as an example of a highly complex language matter. In particular, I focus on how language skills impact the process of knowledge creation and knowledge sharing among employees. By means of a model based on a number of straightforward rules, I show how poor language skills (or an utter lack thereof) risks creating an unbalanced distribution of knowledge (and, consequently, of power) across language groups and how this unbalanced distribution is very sensitive to initial conditions. On the contrary, average language skills seem to support communication well enough to avoid skews that favour even slightly more numerous language groups.