June 22, 2020

On 5 June, the World Health Organisation changed its mind:

“In light of evolving evidence, WHO advises that governments should encourage the general public to wear masks where there is widespread transmission and physical distancing is difficult, such as on public transport, in shops or in other confined or crowded environments.”

This is in stark contrast to their previous advice — issued just two months earlier:

Like what you’re reading? Get the free UnHerd daily email

Already registered? Sign in

“… the wide use of masks by healthy people in the community setting is not supported by current evidence and carries uncertainties and critical risks.”

As Tom Chivers points, out the UK government is also in the process of U-turning:

“…the UK government began by telling people they didn’t work, then ‘advising people to consider‘ using them in enclosed spaces, now mandating them on public transport.”

Late last week, the mayor of London, Sadiq Khan, wrote to the Prime Minister calling for mask-wearing to be made compulsory in shops.

So, battle-by-battle, the war for masks-for-all is being won. But it’s not over yet.


Go to a supermarket in the UK and the majority of people are not wearing masks. Prominent politicians are failing to lead by example — appearing maskless for photo-ops. Donald Trump, despite his own government’s advice, makes a special point of it.

A wide range of commentators have declared their support for masks, from Polly Toynbee to Tom Harwood, but there’s no shortage of anti-mask contrarianism. For instance, last week, Toby Young retweeted this pearl of wisdom:

“On average, the gap in weave of (non-medical) cotton masks is around 1/10th mm. The virus is 1/10,000th mm, so roughly like trying to stop mosquitos with hula-hoops…”

A vivid image, but an incomplete one. The coronavirus is spread through what are delicately termed ‘respiratory droplets’ — i.e. bio-weaponised mucus missiles that are big enough to be impeded by an ordinary face mask. That’s especially true when they’re being propelled out of an infected, super-spreading individual who’s sneezing, coughing or otherwise breathing heavily.

Of course, were not talking about complete effectiveness here. Ordinary masks don’t provide the same degree of protection as medical-grade PPE — but then they don’t have to. At a population level, any significant reduction in R (the rate of transmission) can still make a difference — whether it’s in flattening the curve at the most dangerous stage of an epidemic, or crushing it when R drops below 1.

Any extra help is especially important when lockdown restrictions are lifted and there’s a risk of R rising above 1 again (as appears to be happening in America). Mask-wearing is of particular relevance at this stage, because unlike a lot of other social-distancing measures it is compatible with re-opening and re-mobilising the economy.


Yet, if that’s all true, why didn’t more governments adopt a masks-for-all policy from the outset — or, at least, from the start of lockdown? Why was the official guidance from the WHO and elsewhere decidedly unenthusiastic? Why is our own government’s U-turn still incomplete (masks required on public transport, but not yet in shops)?

The answers to these questions not only reveal a deep misunderstanding of risk at the highest levels, but also a fatal conceit in our culture of politics — and a childish self-centredness in the soul of modern man.

To begin with, risk. The main excuse given for not recommending masks was the lack of a conclusive evidence base. However, governments didn’t have proof that the lockdown measures that they did take would work either — but they went ahead with them anyway. Putting entire populations under house arrest? No probs! Pitching millions of people out of work? Needs must! Bringing whole industries crashing down? We’ll work it out later! But requiring people to wear a small piece of cloth over their mouths? No, for some reason, that was crossing the line. A measure with almost no downside risk, but a potentially huge upside in terms of saving lives and shortening lockdown was not only not taken, but actively discouraged.

It is, of course, important not to react to one systemic risk by taking another systemic risk — for example, the disastrous error of trying to protect hospitals by moving potentially infected patients into care homes. Perhaps that’s why the WHO speculated at length as to what it called the “uncertainties and critical risks” of wearing masks. Among the scenarios floated were “self-contamination”, “diversion of resources from… hand hygiene” and “potential breathing difficulties”. But where was the evidence for any of that?

If there was a systemic downside of wearing masks, wouldn’t we have noticed it by now? After all, habitual mask-wearing in a modern setting is not a novel practice. It’s a long-established feature of 21st century life in a number of East Asian countries where masks are commonly used as protection against infectious disease and air pollution. Indeed, in the course of the current pandemic it quickly became clear that these countries have done a lot better than most of western Europe and North America in containing the spread.

As for the argument that mask-wearing might lead to a “false sense of security” (another WHO assertion), that is a question that could have been settled very quickly by the simple act of observing people. In fact, when researchers did observe they found that, if anything, the effect was in the other direction, i.e. mask-wearers taking more care than non-wearers. That’s not so surprising. The slight discomfort of wearing a mark, the deliberate act of putting one on and the sight of other people wearing them are all reminders of the pickle we’re in.


Speaking of the latest research, studies (e.g. here and here) are now coming in to confirm the common sense expectation that mask-wearing does indeed reduce transmission. The missing evidence base is missing no longer.

Still the question remains: why wasn’t common sense alone enough to prompt action on masks much earlier on? Nassim Taleb identifies numerous errors in the reasoning of those in power; not least mistaking absence of evidence for evidence of absence. With low-cost, low-risk measures like mask-wearing, the obvious thing to do is try them on the basis that they might work — because what have we got to lose?

Taleb has a long record of accusing the credentialed ‘expert class’ (and the decision-makers they serve) of having a poor philosophical understanding of risk — both epistemically and ethically. He has an equally long record of being proven right, including on the pandemic. He and his colleagues, Joe Norman and Yaneer Bar-Yam were early to realise the seriousness of the threat posed by Covid-19, early to argue for lockdown measures and early to criticise the initial errors of the UK response.

And yet while Taleb is right about the biases of the expert class, this doesn’t in itself explain why our system of government is biased to the expert class. In other words, as well as asking why the official advice was so negative about masks, we also need to ask why the decision-makers didn’t over-rule the experts and go ahead with masks-for-all on the basis of common sense.

It is sometimes said that the problem with common sense is that it isn’t common. But, in this case, it is precisely because it is. Unlike the hierophantic bestowal of official scientific advice, common sense doesn’t provide political cover if things go wrong. Furthermore, it tends to lend itself to the local and experimental. It requires epistemic humility on the part of our leaders — a willingness to admit what we don’t know and to own up to failure.

Unfortunately, our culture of politics, which systematically centralises decision-making — works in the opposite direction. Influence is exerted and reputations are made through the enacting of bold, confident, top-down solutions. Which is why the narrowness of the official expert advice that informs those plans is a feature not a bug of system.


But we can’t blame it all on our politicians. The resistance to mask-wearing is also cultural.

A lot of this is entirely understandable. To conceal one’s face in public — or cause someone else to do so — is a powerful statement. The controversy surrounding the Niqab and the Burqa is testament to that.

Face masks impede communication. Indeed that is one of the things that makes them effective — given what we know about how loud talking spreads the virus. The easing of lockdown will remobilise the economy, but with masks-for-all how we look and sound to one another will be transformed. Public spaces will fill with people again, but they will be weirdly silent, their smiles (and scowls) hidden away. It will, in its way, be more disconcerting than the deserted streets of deep lockdown — a cityscape of crowds without hubbub, faces without expression.

But it wouldn’t be forever. Indeed, as a price for accelerating the eradication of the epidemic — and for preventing an economically ruinous second wave — it is surely worth paying.

So, what’s stopping government from giving the order? Perhaps because the stakes are so high. If you think of most of the other things we’ve been universally required to do, there’s always been a bit of wiggle room. Unless you’re a public figure like Neil Ferguson or Dominic Cummings, the anonymity of modern life allows people to occasionally bend or break the rules with a low chance of detection, let alone punishment. Outside of our immediate circles, we don’t know who’s been washing their hands, keeping their distance or self-isolating should they need to. Masks, though, are different. Someone is either wearing one or they’re not. Both compliance and non-compliance would be immediately visible. Which is why a masks-for-all policy is such a high-profile test of a government’s authority (and, competence, should not enough masks he made available). 

The visibility of non-compliance is also what so sticks in the craw of civil libertarians — especially those who imagine their liberties to be ancient ones. I say “imagine”, but in some ways, they’re absolutely right. Our ancestors were allowed to do all sorts of things that we are not. At one time it was the right of every freeborn Englishman to throw the contents of his chamber pot out the window. He could spit in the street and no one cared. He could bait animals for ‘sport’, build a house without planning permission and, in later eras, drive a car without a licence… or being sober. Even in my own lifetime, one could still legally drive without a seatbelt, smoke on public transport or, in certain circumstances, beat a child with a weapon.

The history of civilisation is the history of regulation — the accumulation of rules that we are all expected to abide by for the common good. For all these impositions, we are not only safer, but fundamentally freer and happier too. Not that we admit it, of course. We seem to think that our comparatively carefree world is the result of us being able to do as we please, when in fact the opposite is true.

Mandatory mask-wearing would be a further catch-all imposition. Indeed, that may be the most effective thing about it  — the fact that it encompasses everyone, including super-spreaders. Personally, I’d be happy if masking-up while sick with a cold or the flu became a permanent norm, but the priority right now is killing off Covid without killing the economy. For most of us, wearing a mask would be a temporary inconvenience for a lasting victory.