‘We believe that … ’: Changes in an academic stance marker

This paper explores changes in the use of an important pattern used by writers in all disciplines to present an authorial stance: the structure Hyland and Tse call evaluative that. This construction allows writers to front-load utterances with attitudinal meanings and offer an explicit evaluation of the proposition which follows. Linguists have tended to regard this as separate patterns, but seeing it as a single structure of a matrix clause [evaluation] + that clause [evaluated entity] enables us to recognize a single evaluative purpose with a variety of rhetorical options for writers. Here we examine the contribution of this explicit that pattern to the key genre of the academy, the research article, and map changes in its use and frequency over the past 50 years, drawing on a corpus of 2.2 million words taken from four disciplines. We find that this structure is widely employed in these papers, with an average of 53 cases per paper in the 2015 data, but occurrences per 10,000 words have declined by about 20% with fairly uniform falls across disciplines. We track these diachronic and disciplinary changes and seek to explain them in terms of changing rhetorical practices