I am perplexed by a request for article revision received from the referees of a journal. Many of their criticisms are justified and their suggestions useful. The perplexity comes from comments on the paper's style, summarized three lines into the associate editor's revision-request cover message:
The reviewers agree that the presentation could be improved. Specifically, it reads more like a tutorial or textbook than a journal paper.
What does one do with such a comment?
The paper is not being blamed for being a tutorial; it is a research contribution. The reproach is that in its form of writing it looks like a tutorial. Indeed it is, especially in its first few pages, the result of a considerable effort to make the problem understandable, relying on metaphors and examples from everyday life. In other words, it is designed to be readable. The editor's and referees' remarks seem to suggest that the result violates an implicit rule that a scientific paper must be stern and boring.
My difficulty with this requirement is that it is completely against my view of how one should write scientific papers. If, as a referee myself, I ever wrote "the paper is written in tutorial style," I would mean it as a compliment.
Worse yet, over the years I have been giving lectures on scientific and technical writing, with the central message that technical writing is at the core writing, subject to many of the same rules as any kind of good writing. In this view, one should make the message clear, grab the reader's attention at the beginning, and endeavor to retain it throughout. The result is at the antipodes of the high-priest style of scientific writing, the austere and inscrutable kind, which turns the reading of an article into an obstacle course. If it turns out my view is out of tune with today's world, I feel terrible for anyone who has taken my advice and risks getting texts rejected for "reading more like a tutorial."
I also wonder what I should do for my article. It is, of course, not much work to follow the suggestion. Obscuring is far, far easier than clarifying, and it might actually be fun to see how fast I can learn to obfuscate a text down to an officially sanctioned standard of impenetrability. But should I really do it? Advice appreciated.
Bertrand Meyer is professor and provost at the Schaffhausen Institute of Technology (Switzerland), a visiting professor at Innopolis University, and chief technology officer of Eiffel Software (Goleta, CA).