Distribution Patterns of Boosters and Hedges in English and Russian Spoken Scientific Discours

Abstract

We report on the work in progress aimed to define incidence and distribution patterns of boosters and hedges used in scientific conference presentations on medical biotechnology. We employ our corpus consisting of approximately 10 hours of recorded English and Russian presentations delivered in London, UK; Silver Spring, USA; Kazan and Moscow, Russia, along with corresponding TEI-based transcripts. Presentations date from 2015 to 2016 and were posted on NIH (genome.gov), Cell and Gene Therapy Conference, UniverTV, FutureBiotech, and RusOncoWeb video channels. The mean presentation length is 30 minutes, with reports ranging from 25 to 39 minutes. The corpus features discourse of 18 English and Russian researchers with at least 15 years of work in Oncology or Cardiology. The overall word count of the corpus amounts to 45018 words. Both English and Russian scientific discourses offer a formulaic way of introducing research and they are generally hedged. Though the texts studied are mostly fact oriented and impersonal, in introduction parts of presentations in both languages researchers prefer using pronoun I significantly more often than pronoun we, while in experimental parts the frequency of pronoun we is three times higher than that of pronoun I. Lexical analysis of the transcripts confirmed the hypothesis of numerous differences in distribution patterns of metadiscourse markers of English and Russian researchers. The difference lies in the tendency of Russian scholars to argue more explicitly, while this is less the case in the English spoken scientific domain. A detailed examination of frequency rates and range of hedges point to the fact that Russian researchers employ a limited range of hedges with the frequency two times lower than that in English. The most hedged parts in English presentations are literature reviews and experiment descriptions, while in Russian introductions hedges exceed in number all other parts. The underlying reasons may be found in cultural and linguistic differences between academic communities and presenters' choice of discourse markers. The revealed distribution patterns of metadiscourse markers in spoken scientific discourse can be used in natural language processing as well as in descriptive and comparative studies