Yes, Masks Work: Debunking the Pseudoscience
A Reply to Denis Rancourt's "Masks Don't Work."
Posted Jul 14, 2020
Note: A special thanks to my colleague Jennifer McClinton-Temple for inspiring, researching, and helping me write this article.
The science is clear: we know masks work. But with adults throwing produce and tantrums in the middle of supermarkets during a pandemic, the issues surrounding “wear-a-mask mandates” could not be a hotter topic. In my last article, I argued legal mask mandates cannot be a threat to freedom any more than burn bans during a drought; laws aimed at keeping reckless individuals from putting others in danger during times of special concern do not threaten liberty. And if you have the legal freedom to not wear a mask during a pandemic, businesses have the legal freedom to refuse to serve you. But if, unlike burn bans, masks don’t actually help prevent the spread of COVID-19—well, then, wear-a-mask mandates wouldn’t make much sense, would they? Do masks actually work?
It was in this context that a colleague of mine came to me asking about common mistakes of pseudoscientists. She knew the scientific explanation for why wearing masks reduces the spread of COVID-19. Coronaviruses typically spread through respiratory droplets, and masks hold those back, so infected people are less likely to spread them to others. Since it’s easy for people to have COVID-19 and not know it, making sure everyone is wearing a mask helps prevent accidental contamination. On a neighborhood message board, however, a poster claimed there was no scientific evidence of this.
When my colleague asked for scientific evidence to back this denial, the poster directed her to an article by Denis Rancourt, entitled “Masks Don’t Work.” And, indeed, Rancourt’s paper cited eight peer-reviewed essays, all from reputable journals. But when she actually clicked on the links provided, she found something very curious. None of the studies cited concluded what Rancourt says they did. For example, six of the eight studies measured the effectiveness of N95 respirators compared to surgical masks—not, as Rancourt implied, the effectiveness of wearing a mask vs. not wearing a mask.
Further, the quotes he provided from these articles misrepresented their findings. For example, his quote from a 2012 study in the journal Influenza and Other Respiratory Viruses read “None of the studies established a conclusive relationship between mask/respirator use and protection against influenza infection.” This implies that there is no benefit to wearing masks. In reality, however, the slash in the “mask/respirator” phrase is meant to indicate a comparison between the two types of facial coverings. In other words, the study is saying that masks and respirators are equally effective; it is not lumping them together and declaring them both ineffective. Several of the sentences before and after the one he quotes demonstrate this.
“Eight of nine retrospective observational studies found that mask and ⁄ or respirator use was independently associated with a reduced risk of severe acute respiratory syndrome (SARS).”
The text accompanying the other five studies comparing masks with respirators contain similarly misleading quotes, and the conclusions of others (e.g., Cowling et. Al, 2010) just outright contradict his thesis.
“There is some evidence to support the wearing of masks or respirators during illness to protect others, and public health emphasis on mask wearing during illness may help to reduce influenza virus transmission.”
Of the eight studies Rancourt cites, none of them provides any evidence to support his claim that masks do not work in reducing the spread of viruses, and several of them provide evidence that they do. And all that happens just in the first section of his article.
In the lengthy narrative section, Rancourt cites what he calls a 2012 “landmark” study, by Jeffery Shaman, et.al, which explored the effects of relative humidity on the spread of influenza. In layperson’s terms, this study concluded that higher humidity makes it harder for the virus to spread. Rancourt then concludes, with no evidence whatsoever, that any “second wave” of a virus is due to changes in humidity, not to lax human behaviors (such as not wearing masks). The recent surge of the virus, however, in places like South Carolina and Texas (which are both more humid in July than they are in April) is enough to call this thesis into question. Indeed, Jeffery Shaman himself, quoted on May 20, 2020 in an article on WHYY.org, says that we don’t know enough about the coronavirus to make assumptions about humidity’s effect on it.
A full debunking of Rancourt’s entire article would take more space and time than I have here (I have published one here), but at this point one would be justified in concluding that his entire argument is bunk. As I pointed out in “The Countless Counterfeits Fallacy,” once you realize that a few pieces of proposed evidence for something is pseudoscience, you are justified in concluding that all such evidence will be similarly flawed.
But how could a person make such a mistake? How could a Ph.D. physicist, like Rancourt, say that a study says one thing, when it actually says another—or, even, say a study supports his thesis when it actually contradicts it?
In online discussions, this is usually due to someone misreading the headline and not reading the study. Other times, it is due to expectation bias, the human tendency that can cause people to see or experience what they expect, despite the fact that it is not there. Such things are possible because the nature of our perception is constructive. We don’t see the world as it is; our brain constructs a world for us to experience. Because it often does so based on what it expects to be there, we will often literally see just what we expect to see. One can easily read a study and literally see something that is not there.
But when it comes to articles written by pseudoscientists defending false conclusions, they are usually just outright lying, hoping that you won’t bother to read the study yourself. It’s impossible, of course, to know what’s in Rancourt’s head and heart.
What’s more important, however, is why readers like our neighborhood message board poster will so readily embrace arguments, like Rancourt’s, that are simply not grounded in truth, and share them as if they are. Part of it is certainly confirmation bias; readers will pay attention to the parts of Rancourt’s argument that support what they want to believe, and ignore the parts that don’t (like the mistakes in Rancourt’s argument that invalidate it). But expectation bias is also active; the poster already believes masks don’t work, and so expects Rancourt’s argument to be good. He therefore sees it that way. He might even read the studies and think they support Rancourt’s thesis because Rancourt has made him expect exactly that.
As further evidence of the fact that people can read articles and literally see something that is not there, I submit to you an email that I recently received in response to my article “Confirmation Bias and Police Brutality.”
“Your article on confirmation bias is all well and good. But you are so eager to point the finger at cops for every death you listed, that you failed to realize that Travon [sic] Martin wasn't killed by a cop. Good job proving your own bias.”
Go read the article. I never mention Trayvon Martin once.
And so, yes, masks work to help prevent the spread of COVID-19; thus laws that require them are completely justified. And those who cite articles and studies “to the contrary,” are very likely, but quite literally, seeing something that is not there.
David Kyle Johnson, Copyright 2020,
Get the help you need from a therapist near you–a FREE service from Psychology Today.
- Atlanta, GA
- Austin, TX
- Baltimore, MD
- Boston, MA
- Brooklyn, NY
- Charlotte, NC
- Chicago, IL
- Columbus, OH
- Dallas, TX
- Denver, CO
- Detroit, MI
- Houston, TX
- Indianapolis, IN
- Jacksonville, FL
- Las Vegas, NV
- Los Angeles, CA
- Louisville, KY
- Memphis, TN
- Miami, FL
- Milwaukee, WI
- Minneapolis, MN
- Nashville, TN
- New York, NY
- Oakland, CA
- Omaha, NE
- Philadelphia, PA
- Phoenix, AZ
- Pittsburgh, PA
- Portland, OR
- Raleigh, NC
- Sacramento, CA
- Saint Louis, MO
- San Antonio, TX
- San Diego, CA
- San Francisco, CA
- San Jose, CA
- Seattle, WA
- Tucson, AZ
- Washington, DC