Corpus-based analysis of speech errors in heritage Russian speakers: theoretical and methodological foundations of the study

The aim of the study is to identify how a corpus-based approach can offer new insights into the speech patterns of heritage Russian speakers in bilingual environments. The research draws on theoretical frameworks from cognitive linguistics, generative grammar, and sociolinguistics, and is based on data from the Russian Learner Corpus (RLC), which includes both written and spoken texts produced by heritage speakers. The scientific novelty of the study lies in the application of a multi-layered corpus methodology to the analysis of language erosion: for the first time, errors typical of heritage speakers – related to verb aspect and tense, case usage, prepositional choice, and sequence of tenses – are examined in correlation with metadata on speakers’ bilingual experiences. The results showed that while core syntactic structures tend to remain stable in heritage speech, interface areas – such as the interaction between syntax and pragmatics – are especially vulnerable to erosion. The findings also confirm the stratification of the language system into a core and a periphery, where the retention of linguistic elements depends on frequency, cognitive salience, and functional load. These conclusions have practical relevance for designing educational programs that focus on developing metalinguistic awareness rather than merely correcting deviations from the standard norm.