I was a bit shaken to read this short article, Submit It Again! Learning From Rejected Manuscripts (Campbell et al. 2022), recently posted on Mastodon by open-access legend Peter Suber.

For example:

Journals may reject manuscripts because the paper is not in the scope of the journal, because they recently published a similar article, because the formatting of the article is incorrect, or because the paper is not noteworthy. In addition, editors may reject a paper expecting authors to make their work more compelling.

Let’s pick this apart a bit.

“Because they recently published a similar article”? What is this nonsense. Does the Journal of Vertebrate Paleontology reject a paper on, say, ornithopod ontogeny because “we published something on ornithopod ontogeny a few months ago”? No, it doesn’t because it’s a serious journal.

“Because the formatting of the article is incorrect”? What is this idiocy? If the formatting is incorrect, the job of the publisher is to correct it. That’s literally what they’re there for.

“Expecting authors to make their work more compelling”. This is code for sexing up the results, maybe dropping that inconvenient outlier, getting p below 0.05 … in short, fraud. The very last thing we need more of.

Elsewhere this paper suggests:

… adjusting an original research paper to a letter to the editor or shifting the focus to make the same content into a commentary or narrative essay.

Needless to say, this is putting the cart before the horse. Once we start prioritising what kind of content a journal would like to have ahead of what our work actually tells us, we’re not scientists any more.

Then there is this:

Most manuscripts can eventually Ynd a home in a PubMed-indexed journal if the authors continually modify the manuscript to the specifications of the editors.

I’m not saying this is incorrect. I’m not even saying it’s not good advice. But I worry about the attitude that it communicates — that editors are capricious gods whose whims are to be satisfied. Editors should be, and good editors are, partners in the process of bringing a work to publication, not barriers.

Next up:

Studies confirming something already well known and supported might not be suitable for publication, but looking for a different perspective or a new angle to make it a new contribution to the literature may be useful.

In other words, if you run an experiment, however well you do the work and however well you write the paper, you should expect to have it rejected if the result doesn’t excite the editor. But if you can twist it into something that does excite the editor, you might be OK. Is this really how we want to encourage researchers to behave?

I’ve seen studies like this. I have seen projects that set out to determine how tibia shape correlates with lifestyle in felids, find out the rather important fact that there is no correlation, and instead report the Principle Component 1, which explains 4.2% of the morphological difference, sort of shows a slight grouping if you squint hard and don’t mind all your groups overlapping. (Note: all details changed to protect the guilty. I know nothing of felid tibiae.) I don’t wish to see more such reporting. I want to know what a study actually showed, not what an editor thought might be exciting.

But here is why I am so unhappy about this paper.

It’s that the authors seem so cheerful about all this. That they serenely accept it as a law of the universe that perfectly good papers can be rejected for the most spurious of reasons, and that the proper thing to do is smile broadly and take your ass to the next ass-kicking station.

It doesn’t seem to occur to them that there are other ways of doing scientific communication: ways that are constructive rather than adversarial, ways the aim to get at the truth rather than aiming at being discussed in a Malcolm Gladwell book[1], ways that make the best use of researchers’ work instead of discarding what is inconvenient.

Folks, we have to do better. Those of us in senior positions have to make sure we’re not teaching out students that the psychopathic systems we had to negotiate are a law of the universe.

References

Campbell, Kendall M., Judy C. Washington, Donna Baluchi and José E. Rodríguez. 2022. Submit It Again! Learning From Rejected Manuscripts. PRiMER. 6:42. doi:10.22454/PRiMER.2022.715584

Notes

1. I offer the observation that any finding reported and discussed in a Malcolm Gladwell book seems to have about an 80% chance of being shown to be incorrect some time in the next ten years. In the social sciences, particularly, a good heuristic for guessing whether or not a given result is going to replicate is to ask: has it been in a Gladwell book?