This study on borrowing is founded on the analysis of free-speech corpora from four Slavic minority languages spoken in Austria, Germany, Greece, and Italy. Analysis of the 34,000 word tokens demonstrates that speakers from Italy generated significantly more borrowings and noun borrowings than speakers from the other three countries. A Random Forests analysis determines ‘language’ as chief forecaster for the ratio of both borrowings and noun borrowings, pointing to the existence of borrowing patterns that individual speakers follow. Finally, it is argued that the borrowing patterns prevailing in the examined communities are linked to contact intensity in the past, and to the existence or shortage of literary traditions for the minority languages.