Page 125 - Tallinn University

HARIDUSTEADUSED Tallinna Ülikooli üliõpilaste 2015/2016. õppeaasta PARIMAD TEADUSTÖÖD / Artiklite kogumik
MoDELLIng
THe CompeTiTion BeTween THree linguisTiC groups
KOLME LINGVISTILISE GRUPI DÜNAAMIKA MODELLEERIMINE
III koht magistritööde kategoorias
SAnDER PAEkIvI
bacKgrOund
Speaking languages as an intrinsic capability of humans, has ties to almost all facets of our lives. For- merly a topic researched largely by linguists, has over time become increasingly interdisciplinary. To an extent, this has been motivated by the prediction of a die-o of languages on a massive scale, by prominent research in linguistics [1].
e prediction of extinction was also the motivator of a seminal paper by Abrams and Strogatz [2], published on 2003 in Nature, which unleashed a surge of interest in mathematical modelling of lan- guages, being o en cited as the source of inspiration for a wide range of similar research [3, 4, 5, 6, 7, 8, 9, 10, 11].
Work on language dynamics via mathematical modelling has so far only addressed systems with two languages. However, it is also relatively common for societies to see competition between three lan- guages. For example, international languages spread quickly into distant places where local languages already compete. We constructed a model to describe this system and analysed in depth.
Besides application to linguistic groups, the model could be applied to many similar competition pro- cesses. Examples of this can be found in [12, 13], where similar models were applied toward describing the spread of religion, tobacco use [14] and might even be applied for bacterial competition [15].
the mOdel
e model describes the dynamics of a society with three distinct linguistic groups, which are treat- ed as entities which do not eliminate their prey, but rather convert it. e model assumes that births, deaths and migration cancel each-other out, thus the total population size does not change over time. e society is then made up of agents (speakers) of linguistic groups X, Y and Z, who change their respective group based on a transition probability function P. e latter describes the attrac- tiveness of a language to the population, determining the general outcome of the competition.
125