CATEGORISING ENGLISH TEXT FRAGMENTS BY LENGTH IN LITHUANIAN “TWITTER” MESSAGES: A TRANSLANGUAGING PERSPECTIVE

Abstract

Translanguaging research has predominantly focused on spoken interaction in educational contexts, leaving written multilingual practices outside classrooms comparatively underexplored. This article addresses that gap by examining translanguaging in written,
asynchronous computer-mediated discourse by analysing the length of English text fragments embedded in 2,100 Lithuanian “Twitter” messages. Adopting computer-mediated discourse analysis within a translanguaging framework, this study operationalises a 
length-based typology, reports relative frequencies of each length category, and discusses their characteristic features, thus linking structural distribution, pragmatic function, and sociocultural meaning. A total of 2,493 English fragments were identified and categorised as word-length (1,459, or 58.5%), phrase-length (603, or 24.2%), and sentence-length (431, or 17.3%). 
This corresponds to approximately 118.7 fragments per 100 tweets, or roughly 75.6 fragments per 1,000 words. Empirically, the research shows that English single-word insertions dominate, functioning as compact, high-salience resources for stance-taking, identity indexing, technological literacy signalling, and memetic reference, whereas longer forms are comparatively rare. Methodologically, this paper demonstrates how the structural and pragmatic dimensions of computer-mediated discourse analysis can be adapted to capture the distribution, form, functions, and meaning of multilingual features in social media texts. These findings
extend beyond the Lithuanian-English context, as they highlight how brevity-driven platforms, such as “Twitter”, foster highly lexicalised translanguaging, with implications for understanding language contact, digital literacy, lexicography, and language technology. Limitations include the single-corpus scope and the absence of user metadata. This study concludes with recommendations for comparative, cross-platform, and mixed-methods research to further theorise short-form translanguaging as a distinct phenomenon in global digital communication.