“Objective” science and the “subjective” I
The other day my son told me that in his school essays he is not allowed to use a phrase like “I will conclude that…” I agreed that such announcements are inelegant – just do it, and write clearly enough so that the reader knows what it is you’re doing. But that was not the problem. He is not allowed to use “I”, because “it is unscientific”. My heart sank. Why must 15-year olds be corrupted with this nonsense? It is bad enough that this dogma is stamped into our first-year students’ brains. Each year I need to reassure my bachelor thesis students that with me, they can use any personal pronoun they like. And then I have to reassure them again, because after all, they insist, using “I” is unscientific, everyone knows that. When it finally gets through to them that it’s really okay, they giggle.
I used to think that the Publication Manual of the APA must be the source of the no-“I” dogma, but I was wrong. For all its faults, the Manual is remarkably sensible about pronouns. Avoid passive constructions, it says for example (§ 3.18). You should write “we conducted the survey in a controlled setting”, rather than “the survey was conducted in a controlled setting”. Don’t use the third person to describe the steps you took in your experiment, use a personal pronoun: “we reviewed the literature”, not “the authors reviewed the literature” (§ 3.09). (And if you are the sole author, don’t use “we”, use “I”.) The Manual notes that authors use the third person because they think it is “objective”, and that, I think, hits the nail on the head. The real source of the prejudice against the first person singular is the mistaken belief that you can make science objective by pretending it is the work of anonymous people (“the authors reviewed the literature”) or by erasing all traces of human interference altogether (“surveys were conducted”).
We need to use the first person singular and plural more often. Science only gets better if researchers are explicit about their role in the research process, the decisions they made and the work they did. Don’t take my word for it, Henk Kiers (professor of statistics and dean of the faculty) said the same thing in the recent colloquium on “Healthy research practices”. To make your research maximally informative, he said, you should tell the full story of a study: your line of reasoning, the choices you made, the dead ends, the outliers you removed, in short: the full journey you travelled to arrive at your results. There’s nothing wrong with making choices, even after you’ve looked at the data, as long as you’re transparent about it. I think Henk Kiers is right; and with him Tom Postmes, who made much the same point in his presentation at the colloquium (see also his recent post on Mindwise). The way to make science better, more “objective” (whatever that means, but that’s for another day) is not to remove the human fingerprints, it is to examine each other’s work, the journeys we travelled, and explore alternative routes. Science is a product of human labour, of people making decisions and choices, and pretending that it isn’t, is just wrong. In fact, it’s unscientific.