Analysing discourse in research genre: the case of biostatistics.

The paper examines a small corpus of biostatistics texts, a discipline whose discourse has, as yet, not been explored, from the point of view of its evolution in terms of textual organisation and models. The research aims to explore the diachronic variations in the conceptual encoding of the discipline, its methodology and the grammatical structures used in the presentation, argumentation and interpretation of numerical data applied to the biosciences. It contrastively examines texts from three historical periods focussing in particular on evolutions in foregrounding structures (morphological and syntactic arrangements), figurative language and the typical characteristics of scientific registers (depersonalization and thematizing). The approach is both qualitative (semantic, pragmatic and rhetorical characteristics) and quantitative (keywords, phraseology and collocation) and signals the similarities and differences over time in the texts in terms of conceptual and lexical choices, and the discursive construction of the identity of the scientific community in communicating disciplinary theories.