Patterns of agentivity and narrativity in early science discourse

Between the 15th century and today there have been significant changes in the organisation of scientific discourse, previously identified as developments towards more literate styles (Biber & Finegan 1997) or less narrativity (Atkinson 1992, 1996). This paper argues that such changes affect the linguistic expression of agentivity, in particular. Corpus-based evidence for Early Modern and Modern English shows that early science texts moved from narratives in non-primary use (serving a non-narrative function) to a more function-based, argumentative text pattern. While, originally, both narrative and argument in science were based on personal reference and preserved agentivity, modern scientific discourse nominalises the experience and impersonalises the argument. This explains why the language of science has generally been characterised as static and impersonal.