Criticism ≠ bullying

I’ve never been the victim of bullying, either in science or social media. I have, in fact, been called a bully (and worse) for some of the things I’ve written on this blog (Responses from both Davids (I’m Goliath)). But I know I’m among the privileged (white, cisgender, middle-aged heterosexual male) few who are least likely to experience bullying. So when Ed Yong tweeted his approval of a Slate article by Simine Vazire, “Criticizing a Scientist’s Work Isn’t Bullying. It’s Science,” I checked it out.

The backlash Yong received on Twitter would lead one to believe that the actual article was much more extreme than it really was:

The story of Amy Cuddy, as told in a recent New York Times Magazine story, illustrates why self-correction is so rare in science. In painting a moving portrait of Cuddy’s life over the past few years, it conflates Cuddy’s experience as the target of scientific criticism with her experience as the target of something much more vicious and universal: actual bullying.

Vazire argues that criticizing the content of a scientist’s work is not equivalent to a personal attack on the scientist:

Cuddy’s story is an important story to tell: It is a story of a woman living in a misogynistic society, having to put up with internet bullies who I have no doubt have cruelly and unreasonably criticized her life and career. But it is also a story of a woman experiencing completely appropriate scientific criticism of a finding she published. Conflating those issues, and the people delivering the “attacks,” does a disservice to the fight for gender equality, and it does a disservice to science.

From the comments, you would think Vazire hadn’t acknowledged the misogyny that pervades both academia and social media. Here’s (part of) what she actually said:

We live in a world where women who rise to the top almost invariably must endure systematic harassment and threats of assault along the way. Being a public figure demands heroic levels of bravery from them, particularly given that women are still underrepresented in positions of leadership.

The actual article is thoughtful and balanced, making a crucial distinction between legitimate criticism motivated by a desire to get the science right and personal attacks motivated by a desire to belittle and intimidate.

It cannot be controversial for one scientist to say to another, “I think you’re wrong,” or, “Show me your evidence.” Indeed, the motto of the Royal Society of Science is nullius in verba, which means “take no one’s word.” There should be no sacred findings in science.

I couldn’t agree more (though I think she means the Royal Society of London). Of course, I’m biased to agree; I sometimes use this blog to criticize scientific and philosophical papers. I do, however, try to keep the focus on the results, not the people. Which is exactly what the authors of the blog post in question did. There is not the slightest hint of a personal attack in that post; in fact, the post doesn’t mention Amy Cuddy and colleagues by name at all, only calling them “the authors.” It’s all about the data.

I’m not saying it’s impossible for scientific criticism to cross the line into bullying, or even that I know where the dividing line is located (if in fact there is a sharp dividing line). Power dynamics matter, tone matters, and even strictly scientific arguments get heated. There’s also no doubt that Amy Cuddy has been subjected to some vile attacks, as the Slate article points out. But if you assume from the title that the Slate article is dismissive of online bullying, I suggest you take Ed Yong’s advice:


  1. Curt Sampson says

    Well, there seems to be a little more to the story of that blog post than you include here. According to the NYT article, Cuddy and Simmons had corresponded about Cuddy and Carney’s response to the failed replication:

    In the wake of Ranehill’s failed replication, Cuddy and Carney set to work on a response. Carney…tried to chart a P-curve of all 33 studies they were mentioning in their paper (which was already under review). Carney sent the paper and the P-curve to Nelson for some feedback, but he sent it on to Simmons and Simonsohn, as they were the experts.

    The letter Simmons wrote back to Carney was polite, but he argued that her P-curve had not been executed correctly. He and Simonsohn had each executed P-curves of the 33 studies, and each found that it was flat, suggesting that the body of literature it reflected did not count as strong evidence. He did write that “conceptual points raised before that section are useful and contribute to the debate” but that they should take the P-curve out. “Everybody wins in that case.” According to Cuddy, she and Carney thought the P-curve science was not as settled as Simmons believed it to be. But afraid of public recrimination, they did exactly as he said — they took out the P-curve.

    But then, a key point (in fact, the substance) of the blog post is that one should run a P-curve before considering moderators:

    One approach would be to run a series of studies to systematically manipulate these hypothesized moderators to see whether they matter.

    But before spending valuable resources on that, it is necessary to first establish whether there is reason to believe, based on the published literature, that power posing is ever effective. Might it be instead the case that the original findings are false-positive?

    P-curve is just the tool for this.

    I agree that the blog post, standing alone, looks fine. But in light of the above, where Simmons suggests Cuddy take out the P-curve out of her response and then goes and skewers her response with it, I can see how someone could consider that a bit of a sucker punch.

    As as it turns out, even Simmons himself can see it that way. Again from the NYT article:

    When Simmons and I met, I asked him why he eventually wrote such a damning blog post, when his initial correspondence with Carney did not seem particularly discouraging. He and Simonsohn, he told me, had clearly explained to Cuddy and Carney that the supporting studies they cited were problematic as a body of work — and yet all the researchers did was drop the visual graph, as if deliberately sidestepping the issue. They left in the body of literature that Simmons and Simonsohn’s P-curve discredited. That apparent disregard for contrary evidence was, Simmons said, partly what prompted them to publish the harsh blog post in the first place.

    But the email that Simmons and Simonsohn had sent was, in fact, ambiguous: They had explicitly told her to drop the P-curve and yet left the impression that the paper was otherwise sound. At my request, Simmons looked back at his original email. I watched as he read it over. “Oh, yeah,” he said quietly. He had a pained look on his face. “We did say to drop the graph, didn’t we?” He read it over again, then sat back. “I didn’t remember that. This may be a big misunderstanding about — that email is too polite.”

    Cuddy and Carney had taken their advice literally. Simmons stood by his analysis but recognized that there was confusion at play in how they interpreted the events that transpired. Simmons says he harbored no ill will toward Cuddy before criticizing her paper; if anything, he remembered her warmly. “She was great,” he said, smiling at the memory. “We published the blog post despite my history with Amy. Because I realized that once we pulled the trigger on this. … ” He did not finish the sentence. Cuddy had, in fact, become the poster girl for this kind of work, which even he thought was not fair. “The original study wasn’t particularly egregious,” he said. “It was published in 2010 before anyone was thinking about this.”

    For a moment, the scientist allowed the human element to factor into how he felt about his email response to that paper. “I wish,” he said, “I’d had the presence of mind to pick up the phone and call Amy.”

Leave a Reply