Past Voices, Present Insights: Sociolinguistic Research through Literary Artifacts

Abstract

Oral speech, historically the foundational mode of human communication, has not been explored as extensively as its written counterpart. This disparity underscores the necessity of examining sociolinguistic characteristics of speech across time. Current analyses often rely on data from contemporary speech corpora, yet understanding historical speech patterns is equally vital. Literary works, particularly from periods with scarce speech resources, offer invaluable insights into the sociolinguistic landscape of the past. Extracting speech fragments from literary texts creates a sub-corpus, approximating the spoken word as “heard” by writers of the time. This study, leveraging the Corpus of Russian Short Stories of the early 20th century, provides preliminary statistical insights into the speech patterns of various social groups. Its contribution is twofold: firstly, in pioneering a methodology for investigating sociolinguistic variability through literary analysis; secondly, in laying the groundwork for a theoretical model of dynamic sociolinguistic variability. These findings not only enhance our understanding of historical speech patterns but also aid in forecasting sociolinguistic trends, thereby informing the development of future speech technology applications tailored to evolving language use.

Keywords

• nlp
• spoken speech
• everyday language
• sociolinguistics
• russian
• literary texts
• frequency lists