Nature-the science journal-endorsed Biden - Trust in Science went down; Votes*** didn't change
Nature & Science editors still don't get it
Science is not politics. It a method to draw reliable and reproducible conclusions about the world. Ideally, science informs policy. It articulates the tradeoffs of different decisions, but science alone cannot set policy because science cannot tell you what to value, and what tradeoffs to accept. Only politics— in all its messiness— can set policy. This is why Follow the Science was a nonsense slogan, as I argued at the time.
In 2020, Nature— a scientific journal— did something unusual. It actively endorsed a political candidate. On Oct 14, 2020, it endorsed Joe Biden. At the time, I argued on Plenary Session that this action would almost certainly not change votes— Which undecided voter is reading Nature?— and would just result in loss of trust in the journal.
Well, early results are in. Zhang conducted a randomized survey, where he presented readers (in 2021) with information about Nature’s endorsement of Biden (treated) or a placebo (aka irrelevant info), and surveyed participants. What do you think happened?
Results are clear. Trump supporters lost trust in Nature a fair bit, while Biden supporters gained trust a little— aka Nature becomes seen as more partisan.
And of course, Nature’s endorsement didn’t change views about the candidates. Specifically how they handled COVID19 (these 3 Qs largely unchanged). As for the last Q
After the endorsement of Biden, Trump supporters were much less likely to seek vaccine info in Nature, and Biden supporters were v modestly more likely. In other words, the loss of trust cost them the ability to reach the audience of Trump supporters.
What a colossal mess!
Overall, the results paint the picture that Nature’s endorsement might make editors feel good, and get some high fives in liberal circles, but just polarizes and alienates people, and increases distrust in science. Worse, it is unlikely to change votes, as attitudes towards Biden/Trump on COVID did not change.
**Full caveat: this paper is not the same as studying actual voting impact, but that remains unstudied, and it is irrelevant for the argument in my essay, as Nature editors do not dispute the implications, nor offer alternative data their endorsement actual was of value. ***
Nature issued a defense of their actions. Their arguments are:
Other journals did it to
Trump is super bad (he pulled out of Paris climate & Iran nuclear deal) (PS: didn’t know Nature editors were experts on Iran foreign policy).
Inaction has costs to
But this was illogical. 1. It is irrelevant what other people do. If Timmy jumps off a cliff… 2. These are political preferences, not scientific facts and 3. What are we talking about - The paper shows that inaction would have been better! You did not change minds!
The very paper in question shows that you aren’t changing minds, even on COVID policy, and just losing trust. So what are you talk about? Others see the logical error instantly.
Even the author of the study skewered the Nature editorial board for their reply.
He decimated the ‘inaction has costs’ argument
Enter Holden Thorpe, the editor of Science. In a thread, he agreed with Nature’s endorsement, and went further.
He argued that Science itself says governments must tackle climate change. Specifically that disagreeing with this view was unacceptable from a scientific standpoint.
Of course, that is a political question. Science can articulate the tradeoffs but people have to make the hard decisions.
I have followed Holden Thorpe for some time, and I can’t help but feel that his view of science is that only Democrats can practice it. But that is an incredibly destructive idea. See these other e.g.s.
What is the solution?
As a progressive and scientist, I think the solution is clear. Science and politics should be kept apart. We should limit our scientific views to science, and not impute our values. Yes, we can all wear policy hats, but we have to acknowledge that our partisan preferences might not be shared by everyone. And of course, we need democrats and republicans to be a part of science.
Making political endorsements is not the role of a journal, and it is also not useful or helpful. Stick to publishing good papers. Honestly, both Nature and Science could do a better of job of ensuring the science inside is of high quality.
Finally, it is off-putting to the see the editors of top journals so nakedly partisan. You can be a pundit or you can be EIC of Science, but you can’t really be both. I am not the EIC of NEJM or Nature, but if I were, I would stay off of twitter, and keep my mouth shut about politics.
This is not rocket science.