Face Masks May be Necessary and Sufficient to Stop Covid-19
An entrepreneur’s perspective on Nassim Taleb’s conclusions about face masks
I’ve been following the writing and thinking of philosopher Nassim Nicholas Taleb since the start of the pandemic (and really, for a solid decade prior to that). As with much of the 2008 financial crisis and many other “shocks”, Taleb was either on the cutting edge of anticipating impacts, or several months or years ahead of everyone else.
In the case of Covid-19, he was advocating for measures like social distancing and lockdowns in the West as early as January 2020. I based large chunks of my own writing about strategies for reopening on his thinking and writing.
Taleb has now turned his focus to a new aspect of the Covid-19 pandemic: face masks.
This is exciting to me both because I appreciate his thinking on the topic, and because masks are something I’ve been thinking about intensely since the start of the pandemic. I’ve been approaching them from more of a scientific/practical entrepreneurship side, whereas Taleb takes this approach but also mixes in more political/moral discussion.
To each of Taleb’s points, allow me to add in my own perspectives as an entrepreneur and someone who has been actively testing masks for months now — often on myself.
One and Two : Missing the Compounding Effect, Missing the Nonlinearity of the Risk of Infection
Taleb argues that mask-wearing doesn’t just reduce the risk of transmission by a fixed percentage. There are non-linearities at play, so a reduction in probability of transmission of x% actually reduces transmissions by another, larger percentage — perhaps as large as 95%.
For me, this jibes with the idea of looking at viral transmission not as a threshold event (“Get exposed to virus, get the virus”) but as a cumulative exposure based on your total viral load, with some inflection point (likely personal) where you become infected if a particular exposure value is exceeded.
This fits with Taleb’s other work. Everything is dose dependent — poisons can be beneficial at small doses, but fatal at higher ones. Covid-19 is probably not beneficial at any dose. But it’s also likely that infection (and perhaps the severity of infection) is dependent on total viral load (also likely in a non-linear fashion). That explains why workers, like front line medical staff, that receive higher viral loads get sick more and die more from the virus.
Masks, in reducing the wearer’s total viral load (both by protecting them partially from virus in the environment and protecting others from spreading the virus if infected) reduce the viral load substantially. They might even reduce it past an inflection point at which infection occurs in all but the most susceptible parties.
If that’s the case, then masks may be both necessary and (shockingly) sufficient to control environmental exposure and infections from Covid-19.
Three: Mistaking Absence of Evidence for Evidence of Absence
There is not yet compelling evidence that masks prevent the spread of Covid-19. That does not mean there is evidence that masks do not prevent the spread of Covid-19.
This is a subtle difference when you first begin to look at this particular fallacy, and a glaringly obvious one once you come to see it more fully and in more contexts.
Saying that there’s no evidence I have cancer is not the same as saying there’s evidence I don’t have cancer. The difference feels semantic, but it’s actually wildly important for avoiding flawed and dangerous thinking about evidence.
Taleb also addresses the idea of “false confidence” from masks, saying that there’s a “strong argument that masks makes one more alert to the risks and more conservative in behavior.” I would add the more pragmatic item that masks stop you from touching your face, cutting out another major vector for transmission in addition to the airborne route.
Four: Misunderstanding the Market and People
As an entrepreneur, this one especially speaks to me. Taleb argues that governments essentially treated us like children, telling us that masks weren’t effective not because that’s what the science said, but because they feared there would be a run on masks, depriving healthcare workers of the masks they need to do their jobs.
This is idiotic, and dangerous. Obscuring the evidence about masks’ effectiveness not only costs lives, but also erodes trust in government.
And more to the point, and indulges in the fallacy that the future will be essentially like the present. In reality, as soon as governments advocated mask wearing, the conditions around scarcity changed: markets stepped in and made cheap, effective cloth masks available to everyone (even with fancy or ironic slogans), basically overnight. I
I made my other cloth masks, received them from bored sewing enthusiasts in my area, bought them on Amazon from the American Mask Project, and even had them custom made on Etsy.
There was also the assumption that governments were the best party to obtain masks in bulk. This is patently not true. As an entrepreneur, even in the height of “shortages”, we were able to obtain masks from China (which we donated to front line nurses), with offers to buy tens of thousands more.
Suppliers want to do business with trusted, known counter-parties, not governments that renege on deals or fight with each other about payment, receiving shipments, etc.
Five: Missing Extremely Strong Statistical Signals
Sometimes you need overwhelming evidence to prove a point. An n of 1 trial of a vaccine for Covid-19 would be unacceptable — it wouldn’t tell us enough about how a potential vaccine impacted different kinds of people, prevented different strains of the virus from spreading, etc.
Sometimes, though, an n of 1 (or a small n) is enough to prove a point, or at least provide a very strong signal in favor of a conclusion.
Taleb cites the example of two hairstylists, both Covid-19 positive, who together saw 140 clients while infectious. Both the stylists and the clients wore masks. There was no transmission.
If masks didn’t prevent transmission, it would be extremely unlikely that we’d see zero transmissions in this scenario, given how infectious Covid-19 is known to be. It’s a small sample, but sufficiently large to reach meaningful conclusions, or at least to see a meaningful signal.
In my own Neuroscience background, we often relied on n of 1 studies. One patient with a specific deficit was often enough to make meaningful conclusions about how the brain is wired — often through negative evidence.
It’s not exactly the same, but remember that n of 1 studies are a big part of what Taleb is all about. A single black swan is enough to disprove the hypothesis that “all swans are white.”
Six: The Non-Aggression Principle
This is where things get more political or philosophical. If we accept the scientific aspects of mask wearing, should we accept governments’ or businesses’ ability to mandate the practice?
Taleb argues that liberty is based on the Silver Rule: do not harm others; they in turn should not harm you.
To me, this relates to basic questions we as a society have grappled with before. Protecting your own liberties gives you the right to harm yourself, if that’s what you really want to do. But it doesn’t give you the right to harm others.
Governments can tax the bejesus out of smoking to make it unappealing, but they can’t prevent you from doing it on your own property. They can, and should, prevent you from smoking in public areas, though, since in this case you’re harming others who have not chosen to accept the risks of your actions.
To me, mask wearing is no different. I’m a photographer. I rely on the First Amendment to conduct my work, and the protections it affords me. But I accept the very old, very fundamental argument that free speech doesn’t allow one to run into a crowded symphony (whenever those return) and shout fire.
Likewise, I could choose not to wear a mask if it was only harming me. But all the evidence points to the fact that masks exist mainly to protect others from you. Given this, governments should be able to tell me where I have to wear a mask, just as they tell me where I can smoke (I don’t). Doubly so for private businesses.
Coda: Practicalities of Mask Wearing
I always like to back up conclusions with experience, preferably personal.
Here’s a stannic chloride fit test I performed on myself, testing one of Taleb’s mask-wearing strategies: doubling up on face masks for extra protection in risk environments like an airplane cabin.
And a demonstration of the principle behind masks protecting others from you — again using stannic chloride.
I’ve also put together a basic primer on the different types of masks and how to distinguish them.
To conclude, Taleb discusses how one act of altruism (wearing a mask) can have a cascading effect, greatly reducing systemic risk. If you’re infectious and your mask prevents the spread of the virus to one other person, you’re potentially saving hundreds or thousands of lives as you avoid the impact of your infection cascading through the entire system.
Taleb’s final conclusion? “Wear a mask. For the Sake of Others.”