We are not supposed to use first-person pronouns, and contractions aren’t allowed. These rules also discourage unattended anaphoric pronouns and say that split infinitives should be rarely used. And to start a sentence with an initial conjunction is as bad as to include a listing expression, and so on. Exclamation marks are forbidden!

The rules of academic writing are many, but they have one intention: to avoid informal language, in all its forms. Blogs and social media may encourage authors to write it as they say it, but much of what passes for scholarly and scientific prose is simply not designed for human ears. Academic writing is code, with freedom of expression and emotional range curtailed in favour of explicit meaning and a necessary lack of ambiguity. If nothing else, it (by which we mean academic writing, for those still on the watch for unattended pronouns) is writing that knows its audience and gives them what it (the audience) expects.

But, to use a direct question, another stylistic tool on the banned list, is this academic supply and demand still in place? Do the academics of the Internet age still communicate as stiffly as their colleagues did at the time of the Apollo programme? Or, heaven forbid, has some scruffy informality crept into scholarly discourse?

Yes, and no, according to an illuminating new analysis. Formal language is largely intact, the study finds, give or take a mildly more tolerant attitude to split infinitives and initial conjunctions. Yet there has been an explosion in the use of the first-person pronouns in academic papers by biologists. What, we wondered, is that all about?

The analysis, published online in the journal English for Specific Purposes, looked at the language of academic papers selected at random from several high-impact journals published across a range of disciplines in 1965, 1985 and 2015 (K.HylandandF.JiangEngl.Specif.Purp.http://doi.org/bssn;2017). If anything, academic publishing in applied linguistics and sociology has become slightly more formal. The number of informal features included in papers in the major electrical-engineering journals went up by 9% over the 40 years. But it was the eye-catching increase of 24% in biology journals that stood out, dominated by a headline threefold increase in words such as I and we.

Personal pronouns are frowned on in academic text, with many guidelines to help novice writers avoid them, chief of which is the use of the passive voice (so we did not see something — instead, it was seen). One explanation for the rise is that as the passive voice becomes less fashionable, one obvious way to restructure sentences is to reach for a personal pronoun.

The passive voice is encouraged in scholarly prose precisely for the reasons that dramatists and journalists try to avoid it: it introduces distance between the action and the protagonist and between writer and reader. This, convention suggests, lends an air of detached objectivity to observations and conclusions. It, perhaps, just feels more scientific. The increased use of I and we, the study authors suggest, could also reflect wider language changes in society, or perhaps is down to the increased number of articles written by people for whom English is not their first language. They may not feel so acutely the sense that writing I or we makes a statement of projected authority.

Another explanation is more subtle. Perhaps modern biologists, under increased pressure and competition, do not feel confident that merely stating their case is enough. Personal language builds a connection to the reader and helps, ultimately, to persuade. We think so. Don’t you?