Psycholinguistic Computerized Tools of Linguistic and Translation Studies Discourse Analysis

ABSTRACT

Recently, researchers in the field of linguistics, psycholinguistics, psychology of language, translation studies and other related fields have shown an increased interest in studying linguistic features of discourse. An increasing amount of studies on deception, means of suggestion and manipulation in discourse clearly indicate that there is a relationship between the use of function words, discourse cognitive compexity and speakers’ emotional states, his/her hidden intentions. Function words as ‘style words’ (Tausczik & Pennebaker, 2010) including pronouns, conjunctions, prepositions, articles are processed automatically due to their procedural meaning. This paper is focused on the application aspects of two computerized methods – Linguistic Inquiry and Word Count (LIWC), and Textanz that are sensitive to function words. Designed by J. Pennebaker, social psychologist from University of Texas in Austin, and his colleagues (Pennebaker et al., 2007), LIWC as a text analysis program that counts words in linguistic and psychological categories helps a researcher detect meaning in a wide variety of experimental settings, including to demonstrate information distortion in political discourse and translation. Our previous investigation on deception validated LIWC’s potential in identifying information distortion in English political discourse. The current study was aimed at verifying the above mentioned programs’ ability to detect deviations in translating. The authors found that such translation universals as simplification, normalization, and explicitation are markers of information distortion, or the “third code” (Frawley, 1984), in the target versions. Another important finding was that  they can be easily traced in English-Ukrainian parallel and comparable corpora through the use of LIWC and Textanz – reliable detectors of linguistic means with mostly procedural meaning.