Author:

Keywords:

complex systems, FOS: Languages and literature, Linguistics, Linguistics/statistics & numerical data, Simulation software, 11P8324N#57179821

Abstract:

As researchers, we sometimes have questions that are rather difficult to test empirically. Often, these are bold, far-reaching questions which, while ambitious, are simply impossible to operationalise. Usually, these restraints are of practical, financial or some other concern. In my talk, I propose a methodology which sidesteps these issues: computer simulations. With computer simulations, you can build your own virtual societies which you can establish in any way you like, a practice conceptually analogous to “The Sims” games. The idea is that in these simulations, you can distill social relations between people to just the mechanisms you are interested in as a researcher. In this virtual world free of noise, budget contraints or ethical commissions, you are free to change the rules according to which agents, virtual inhabitants, behave. In this way, you are able to test what the theoretical limits are to the specific behaviour or theories you are interested in. While it is clear that computer simulations could be used for a plethora of different research questions within the humanities (i.e. history, archaeology), I will present how simulations can be used for the study of language change. To this end, I will explain the plans for a case study from my FWO project, which is centred around how computer simulations can help us understand how frequency of use shapes language change. I will give concrete examples of the types of questions I would like to answer with the computer simulations, which will give a better idea of the genre of questions this methodology is able to answer in the first place. At the same time, I will emphasise the shortcomings and pitfalls of computer simulations as a research tool, because unfortunately, unlike “The Sims”, good research has no cheat codes.