To see how far the textbook stereotype of impersonality corresponds with actual practice, I interviewed
expert writers and examined 240 published journal articles, 30 from each of eight disciplines.
Using Wordpilot 20001 (Milton, 1999), a text analysis and concordance programme, I electronically
searched the corpus for the pronouns I, me, my, we, us, and our, and then examined each
case to ensure it was an exclusive first person use, i.e. that it referred only to the writer(s) and was
therefore a genuine author pronoun. My results suggest that academic writing is not the uniformly
faceless prose it is often thought to be, but displays considerable differences between disciplines
(Table 1). Broadly, writers in the hard sciences and engineering prefer to downplay their personal
role to highlight the issue under study, while a stronger identity is claimed in the humanities and
social sciences papers.