For a long time, the assumption prevailed that all scientific texts should conform to a common academic style. Then some papers started to emphasise the specificity of a discourse into scientific disciplines. Now, under the influence of globalization and the shift to teaching in the lingua franca languages, there is a question about the peculiarities of a national style of academic discourse. The article continues a series of studies in the field of sociological discourse and its changing after the introduction of SFL-based CLIL approach in non-western sociological doctoral programmes. The current paper is focusing specifically on tools for the structuring of science written discourse that are significantly different in different cultures. The research is interdisciplinary since it is performed at the intersection of sociology and ethnolinguistics. The method is an analysis of discourse markers as one of the widely recognized approaches in ethnolinguistics to the identification of differences in scientific writing in Russian and English. Data collected from doctoral theses in Russian and in English from the field of sociology. It is shown that the average number of discourse markers at 1000 words-3.89 in Russian theses and 1.75 in doctoral theses written in English. The authors suggest that these variations are associated with the structure and goals of a scholarly paper. English academic genres are more empirical, whereas Russian focused on the development of theory. The results of the study clarify the reasons for refusal of Russian professors of scientific advising in English, and their negative attitude towards the English-speaking model of a thesis, traditional for Western science.