Scholars are ten times more likely to be targeted for right-wing speech

Anyone who doesn’t have his head in the sand knows that the atmosphere in Anglo-American universities has changed dramatically over the last five years. Although certain topics have always been taboo, the threshold for what’s considered “problematic” has fallen considerably, and the number of scholars winding up cancelled has correspondingly risen. Yet until recently, anyone pointing this out was open to the charge that “the plural of anecdote is not data”. Critics would allege that the free speech crisis is a “right-wing myth”, and that in most cases the supposed target was merely being held “accountable” for his opinions.

Jeffrey Sachs, a left-wing political scientist, made an important contribution by assembling the ‘The US Faculty Termination for Political Speech Dataset’. However, as the name suggests, this only dealt with cases where faculty were actually terminated. And these are of course a subset of all the instances of cancel culture at university. Another important contribution was the ‘Disinvitation Database’, compiled by researchers at the Foundation for Individual Rights in Education’s (FIRE). Once again, however, the focus of this dataset was somewhat narrow; it only concerned cases in which activists sought to get someone disinvited from speaking. And it included non-academic speakers like Ben Shapiro and Donald Trump Jr.

Now a comprehensive study of academic cancel culture has been published. The ‘Scholars Under Fire Database’, assembled by researchers at FIRE, lists 426 cases in which scholars were targeted for ideological reasons between 2015 and 2021. Although it only includes cases in the United States, it’s the most comprehensive and systematic assemblage to date. The researchers defined a “targeting incident” as one involving “efforts to investigate, penalize or otherwise professionally sanction a scholar for engaging in constitutionally protected forms of speech”. They therefore excluded cases of harassment or intimidation that didn’t result in formal sanctions.

What does the new database tell us about cancel culture at US universities? For a start, it confirms that the number of incidents has risen substantially over just the last five years. While there were 24 cases in 2015, there were 67 in 2019, and 111 in 2020. (Though 2020 may have been an unusually censorious year, due to acrimony surrounding the election and the mostly peaceful riots). Another interesting but not surprising result is that ‘race’ was the most frequent subject of controversy. The next most frequent was ‘partisanship’ (i.e., Republicans dissing Democrats and vice versa), followed by ‘gender’. Interestingly, ‘Israel/Palestine’ was relatively infrequent by comparison.

Who is doing the cancelling? The researchers coded each incident for whether it came “from the left” or “from the right” of the target. So a self-identified liberal being targeted by woke activists for questioning “systemic racism”, say, would count as an attack “from the left”, even though the target himself is not right-wing. In total, there were 264 attacks “from the left” versus 147 “from the right”. (The remaining 15 could not be assigned to either category.) However, simply comparing these two figures understates how much more likely those expressing “right-wing” speech are to be targeted for cancellation. That’s because there are many fewer right-wing scholars on campus. (Incidentally, it would be more accurate to say “speech to which left-wing activists object”, but that’s a mouthful.)

The most recent nationally representative survey of which I’m aware was carried out by Sam Abrams in 2016. He observed a 6 to 1 ratio of liberal to conservative faculty. (The skew may well have increased since then.) If liberals and conservatives were equally likely to be targeted, we’d expect about 6 times more attacks “from the right”. (Note: attacks can come from outside the university as well as inside; and in fact, the researchers found that attacks “from the right” were most likely to be instigated by politicians or the general public.) However, what we actually observe is about 1.8 times more attacks “from the left”. This means that, overall, scholars are more than 10 times more likely to be targeted for expressing “right-wing” speech.

What about the consequences of being targeted? The researchers found that fully three quarters of incidents led to some kind of sanctions against the target – whether that was an investigation, suspension or termination. Refusal to concede wrongdoing was the most common response on the part of targets. However, individuals who took this approach were twice as likely to be terminated as those who expressed regret or apologised. Though it should be noted: this simple comparison doesn’t control for the “severity” of the incident; perhaps scholars who say something really “controversial” then go for broke, as they expect to get fired.

If true, the finding that termination is more likely when targets stand their ground is interesting because it seems to contradict some previous evidence. In a 2019 study, Richard Hanania found that survey respondents were more likely to say that Larry Summers should have faced negative consequences for his statements about women in STEM when they were told that he’d apologised for them. And Cass Sunstein made a similar finding in a survey of his own. However, these results aren’t necessarily inconsistent with what the ‘Scholars Under Fire Database’ shows. It’s possible that apologising incriminates you in the minds of the general public, but demonstrates servility to activists and university administrators.

The FIRE researchers have done excellent work compiling the ‘Scholars Under Fire Database’. Sceptics can no longer claim that this or that example of cancel culture is “just an isolated incident”. Nor can they fall back on the truism that “students have always protested things”, since the numbers have gone up massively in a short space of time. And of course, the 426 cases listed are only part of the broader phenomenon. As Steven Pinker notes, “it’s the regime of intimidation that silences many more and warps our knowledge.” As far as I’m concerned, the debate over whether cancel culture exists is now settled. It does.

Image: Daderot, Massachusetts Hall, Harvard University, 2007

More on Japan’s COVID death rate

My previous newsletter examined the puzzle of why there have been so few COVID deaths in Japan. I concluded that lack of obesity can’t explain more than a small part of the difference between Japan’s death rate and the UK’s. And I said that the absence of a major epidemic in Japan may be due to some combination of voluntary behaviour change and greater prior immunity.

By far the most common response on Twitter was that lack of obesity is what explains Japan’s apparent success, which suggests many people didn’t bother to read the post. This is not to say that the “obesity hypothesis” has been conclusively disproven. But simply pointing to Japan’s low obesity rate, and then saying “case closed”, isn’t a very convincing argument, for the reasons I mentioned.

The second most common response was that I didn’t mention mask-wearing, and this is what explains the absence of a major epidemic in Japan. Personally, I’m not convinced that community masking (i.e., members of the public wearing cloth or surgical masks in public) has any effect at all. But if it does, it’s almost certainly not large enough to explain the situation in Japan. After all, there are numerous countries where mask mandates were followed by major epidemics.

The third most common response was that higher vitamin D intake is the crucial factor. While there is evidence that vitamin D protects against COVID-19, it’s somewhat mixed. But more importantly, there’s no consistent evidence that Japanese people are less likely to be deficient than the citizens of other wealthy countries. For example, a 2007 study notes, “vitamin D deficiency is common in … Japan”. And a 2015 study with a sample size of ~9,000 found that the “prevalence of vitamin D sufficiency” was low in the Japanese population.

The Daily Sceptic

I’ve written four more posts since last time. The first summarises an article by the reporter Josh Rogin, who discovered that “a large swath of the government” already believed the lab leak theory last spring. The second argues that lockdowns were motivated by “politician’s logic”, a fallacy popularised by the British sitcom Yes, Prime Minister. The third argues that vaccine passports could have harmful unintended consequences. The fourth summarises a recent study finding that Scottish teachers did not face an elevated risk of severe COVID-19 when schools were open.

Thanks for reading. If you found this newsletter useful, please share it with your friends. And please consider subscribing if you haven’t done so already.