Disinformation Crossing Spaces and Language Borders: A Contrastive Analysis of English and Lithuanian

Abstract

This paper examines the language of disinformation on the topic of the COVID-19 pandemic in English and Lithuanian using the methods of contrastive corpus linguistics. The study not only reports research results but also addresses some methodological issues encountered in contrastive analysis of disinformation, a main one being the absence or limited amount of original content in Lithuanian disinformation texts. Since most of the Lithuanian content is translated or adapted from other sources, an important question is how likely it is that some distinct language-specific features will emerge in disinformation published in a lesser- used language. The content modifications in the Lithuanian texts range from very close translations of the source texts to highly abridged summaries of the original. A general trend is that almost all the texts are shorter in Lithuanian. Regarding the analysis of linguistic properties, the type-token ration (TTR) is very low in English texts but considerably higher in Lithuanian, which could be a result of typological differences between the two languages. Emphatics are almost equally distributed in both datasets; however, tentative language is more frequent in English. Such trends suggest that the language of disinformation tends to be simple, but Lithuanian false news aims at sensationalism by retaining the same frequency of emphatic wording but reducing the tentative tone.