Small-scale multilingualism through the prism of lexical borrowing
Aims and Objectives: We assess whether data on lexical borrowings obtained through field elicitation may point back not only to a specific donor language but also to its specific regional variety, and hence whether these data are a reliable tool for reconstructing unknown historical patterns of interaction between ethnic (sub-)groups.
Methodology: We use quantitative analysis of the data obtained by loanword probing — elicitations of short word lists from speakers of minority languages — to calculate the amount and source of lexical transfer. We compare the results across several areas with varying degrees of bilingualism and different contact varieties of the donor language to see how this influences the results of our analysis.
Data: The data for this study come from a large-scale collection of field data in Daghestan (72 speakers, 19 villages), with four Lezgic languages spoken as L1 and varying degrees of traditional bilingualism in Azerbaijani, a major Turkic language of the area.
Findings: Our method suggests that the speech communities clearly indicate one of the regional varieties of Azerbaijani as the donor, the Azerbaijani of Qax for the villages in the southwest of Daghestan (Akhty and Rutul districts) and the Azerbaijani of Daghestanian lowlands in the villages of the southeast of Daghestan (Tabasaran and Khiv districts). We also observe that the amount of lexical convergence with the donor depends not only on the level of bilingualism observed in the specific village but also on the L1 of this village, suggesting language borders as natural constraint to the spread of lexical borrowing.
Originality: The study is novel in that it is fully based on analysis of data on lexical convergence obtained through fieldwork on minority languages and provide quantitative results that can be compared across speech communities in the survey.
Implications: Our conclusion is that the method is sensitive enough to trace donorship to one specific regional variety of the donor language.
Limitations: While we were able to clearly detect not only a donor language but also its regional variety, our observations on the relative weight of the degree of bilingualism and language affiliation of a speech community as predictors of the amount of lexical convergence require more data obtained both from other linguistic environment and by different methods.