Aims and objectives: In Dagestan, Russian is the language of education, urban way of life, and upward social mobility, and the means of communication between speakers of different languages. This is a result of a quick and drastic change. At the end of the 19th century, Russian was spoken by less than 1% of the population. The aim of this paper is to understand how such rapid spread of Russian as an L2 became possible. Approach: The study uses quantitative data on Dagestanians’ language repertoires. We relate the command of Russian to certain facts from people’s biographies, such as the level of education, migration, warfare and military service, and other professional experience, and run regression analysis. Data and analysis: The data were collected by the method of retrospective family interviews during numerous field trips to highland Dagestan. We use information on 3519 individuals collected in 27 villages. Findings/conclusions: We conclude that the compulsory school education introduced in Dagestan in the 1930s is the social mechanism that resulted in the spread of Russian and its later development into a lingua franca. Russian was imposed from above and supported by the ideology that associated it with future and progress. Originality: This is the first attempt to apply quantitative methods to a large collection of field data to reveal social mechanisms underlying the spread of a single L2 instead of local bilingualism. Significance/implications: The spread of one lingua franca across a large territory is attested in many areas. We suppose that lingua francas of different origin result from different constellations of social factors and show that in Dagestan lingua franca was imposed by the authorities via a systematic educational campaign. We also suggest it was the extreme linguistic diversity of Dagestan that brought Russian from a widely known L2 to a lingua franca.

Приводятся результаты социолингвистического эксперимента 2015-2016 гг., целью которого было выяснить степень влияния экстралингвистических факторов на распознавание и использование лексики города Твери местными жителями. Предпосылкой для исследования стало предположение о возможном изменении в составе региональной лексики и ее постепенном разрушении в речи молодых образованных тверичей под влиянием внешних социальных факторов (высокая мобильность, образование, интернет-коммуникация). Ключевые слова: региональная лексика; лексика города; регионализмы; региолект; социолингвистика.

The paper raises an issue of possibilities that are provided by corpus-based approach in analysis of metaphorical transfer based on the aspectual pair upast’ / padat’ (‘to fall’). The author reviews the structure of metaphorical meaning of predicates, which enforce the Lakoff ’s metaphor ‘LESS IS DOWN’ and copy some parts of valency structure from the primary meaning’s structure, and also analyses how collocations correlate with valency structure. The empirical base for analysis was extracted from the Russian corpus ruskell 1.6 in Sketch Engine. It consists of 1516 collocates in total used both in primary and metaphorical contexts. Formal corpus analysis of collocations demonstrate that unlike the primary meaning, metaphorical meaning is more strict and less variable when we talk about syntagmatic rules of filling patient (valency A1), source, or primary point (A2) and goal, or final point (A3), and interval (A4). As a result of analysis, we could extract several syntagmatic groups that are correlated with the metaphorical meaning only. Corpus-based analysis helps with identifying typical and marginal types that fill the valencies, nevertheless analysis of collocates that trigger metaphorical shift is more important. In the research, we focus on the syntagmatic group “noun(nom) + predicate” that marks patient’s valency. Semi-automated tagging of patient and semantic analysis of patient let us determine some specific lexical and semantic features and classify semantic groups (as parameter, property) that are typical for the metaphorical meaning. The author suggests that the mixed corpus-based approach, which is demonstrated in the paper, could be used in analysis of more general predicate metaphorical field as “get more <less>”