The Week In Doom — Bad Science Edition

Abbreviated doom this week since I’m on the road.

Item

Mathematicians divided over faculty hiring practices that require proof of efforts to promote diversity (Thanks to Kip Hansen for the tip.)

When Abigail Thompson, chair of the math department at the University of California (UC), Davis, penned an essay in the December 2019 issue of the Notices of the American Mathematical Society criticizing mandatory “diversity statements” in university hiring, simmering frictions in math boiled over. Researchers rushed to author op-eds and joint public letters both supporting and opposing Thompson. The reactions reflect a tension between mathematicians who see efforts to promote diversity as an intrusion of politics into research, and those who see opening their field to historically marginalized communities as the surest way to advance research. As befits the field, each side claims numerical data support their view.

Academic math skews overwhelmingly white and male.

Now one would ordinarily take this as evidence that white and male persons were better at becoming top mathematicians than non-white non-males. Yet it is taken as evidence of equality, as if every group could produce the same rate of top mathematicians. The opposite is thus proved by the contrary observation.

To improve diversity in all fields, eight of the 10 UC campuses as well as an increasing number of other universities across the country require faculty job applicants to submit a statement explaining steps they have taken to enhance diversity and equity…

“This is a contentious issue whose discussion has been suppressed,” Thompson tells Science. Her essay applauds inclusivity efforts of recent decades, including encouraging people from diverse backgrounds to pursue a math career. But she calls diversity statements a “political litmus test” that she compares to the McCarthyist “loyalty oaths”…

They are indeed loyalty oaths, as we have demonstrated many times. If you do not swear to uphold progressive anti-Reality ideology, you will be banned, purged, isolated, ignored, and pushed out. Everybody knows this, which is why those wanting to keep their jobs and who are still in touch with Reality lie on these oaths. This is entirely pragmatic, and even excusable, because those who push the oaths have no right to expect the truth. But every lie is corrupting, even good ones.

Thompson says she has received “strong support” from colleagues at UC Davis and elsewhere. But to Chad Topaz, a mathematician at Williams College, requiring such diversity statements isn’t political; it’s a recognition that math generates better research when everyone can participate. “We want math to be the best it can be, so we have to make it accessible to everybody.”

This is not so. Better research is generated when better mathematicians, and fewer than those we have now, purse mathematics. Worse research, far worse research, is generated when the field is stacked with people hired because they meet demographic quotas.

When the field is stuffed with mediocre and worse talents, it bogs down, it chokes out the best who are forced to deal with the worst.

The next item fits right in.

Item

Russian journals retract more than 800 papers after ‘bombshell’ investigation

Academic journals in Russia are retracting more than 800 papers following a probe into unethical publication practices by a commission appointed by the Russian Academy of Sciences (RAS). The moves come in the wake of several other queries suggesting the vast Russian scientific literature is riddled with plagiarism, self-plagiarism, and so-called gift authorship, in which academics become a co-author without having contributed any work.

Gift authorships, incidentally, are far from rare even in the West, especially in certain fields, where it is not uncommon for the department chair’s name to find its way onto papers he had little or nothing to do with.

And Russian authors frequently republish their own work, says Yury Chekhovich, CEO of Antiplagiat, a plagiarism detection company. In September 2019, after sifting through 4.3 million Russian-language studies, Antiplagiat found that more than 70,000 were published at least twice; a few were published as many as 17 times. Chekhovich believes most instances are due to self-plagiarism.

That 17 cracks me up, but it also leads me to wonder. I’ve seen plenty of papers, including some of my own, where I use the same phrases. Academics must publish or they will perish. And given the rapid expansion of lesser talents, and the subsequent flood of poor papers, the value of any single paper on average decreases. Which in turn pushes quantity to become a measure over quality.

That leads to academics to reusing their work, churning out multiple works on the same subject when one or none would have been sufficient.

Point is, I wonder how American or other Western papers would fare if run through the same test as the Russians had to do.