The paper is an attempt at a quantitative corpus related approach to the subject of multilingualism in contemporary Czech poetry (published both in books as well as on literary servers). The authors of the paper examine the frequency and distribution of foreign (i.e. non-Czech) lexical units, raise questions about the forms and functions of individual means of expression; three selected poets (T. Kafka, M. Šanda, M. Torčík) are analysed more in-depth. In addition, the paper is a piece of news about a database being developed – contemporary poetry corpus – and the possibilities of using it. It suggests how beneficial the quantitative data analysis in the first phase of linguistically oriented literary research can be, or it points to the necessity of interconnecting the quantitative and qualitative approaches; it is only the researcher´s interpretative competence that can actually define the boundaries of the research field and the significance of the present elements. When conducting text-centred analyses, language corpora should begin to play a role similar to that of other tools of scientific infrastructure, e.g. bibliographic database.