The Can WAR and CRISIS be predicted from linguistic data? Preliminary proposal for a linguometric procedure

Abstract

In this article, I discuss a linguometric method/procedure that I am developing myself, with which it will be possible to forecast upcoming changes in the extra-linguistic reality. Econometric analyses are a model for this procedure. Underlying the proposed linguometric method, I made a cognitive assumption related to the linguistic picture of the world (or more precisely here: textual worldview) of a some kind of communicative community (here: Polish communicative community) and conducted corpus-based frequency-lexical analyses with reference to three exemplary historical events: a) the global economic crisis of 2008–2009; b) the outbreak of World War II in 1939; c) the Russian aggression against Ukraine in 2022. In the corpus‑based linguistic data, I tracked the frequency-lexical symptoms of the coming of the mentioned events. It was possible to establish by means of statistical data (frequency and textual frequency of synonymic words from the thematic fields CRISIS and WAR) that language ‘foretells’ an imminent change in reality 2 to 4 years before it occurs (it is therefore possible to predict future events on the basis of linguistic data). It seems, therefore, that the proposed linguometric procedure can be complementary to econometric, political science, sociological, and many other analyses. I also think that the method developed can be implemented into a digital linguometric tool operating on language corpora, which – for the purposes of linguometrics – should be continuously developed. Such a linguometric tool could be developed in the information and linguistics team.