Introduction
According to Newsweek, as of August 2023 in the United States, hospitals and colleges in a few states – California, New York and Ohio – announced renewed mask mandates to address a recent uptick in Covid-19 cases and hospitalizations. While that was but a handful of organizations with mask mandates compared to the height of the pandemic a few years prior, it nonetheless highlights the value of thinking better about the morality and implications of mask mandates as policy. Further, the effectiveness of mask mandates should be distinguished from the effectiveness of the wearing of a mask – these two issues are distinct and should be treated as such. The aim here is solely on the effectiveness of mask mandates as policy, and not at discouraging mask use by individuals who use masks properly. It is possible to think that mask mandates as policy do not work, and yet think that an individual could benefit from wearing a mask.
Spelman’s Argument
In autumn of 2022, the moral philosopher Jonathan Spelman argued that states should ban maskless shopping via legal mandates due to the fact that maskless shopping is worse than drunk driving. And since states outlaw drunk driving to save lives, then by analogy, states should outlaw maskless shopping too, in order to save lives. As Spelman argues,
[While] drunk driving’s potential to cause harm shouldn’t be understated, mask-less shopping’s potential for harm is even greater. This is because when mask-less shoppers infect others, those who are infected can go on to infect other people, and so on. Imagine, for example, a world in which those who are injured by drunk drivers … contract a disease that turns them into drunk drivers. … Imagining that world gives us a better sense of the kind of threat that mask-less shoppers pose and helps us see why it is even more important to outlaw mask-less shopping than it is to outlaw drunk driving.
As Spelman acknowledges, the reason that drunk driving should be outlawed, and why allegedly maskless shopping should too, is because drunk driving ‘does significantly endanger non-consenting third parties’. And thus, because maskless shopping, especially while one is infected, can spread among third parties exponentially, wreaking havoc on third parties, then there is even better reason to ban maskless shopping than to ban drunk driving – call this the argument from analogy. The flip side of this argument, though, is that if there are reasons to doubt that mask mandates, as a policy, would reduce harm to third parties, then that would undercut Spelman’s argument from analogy in favour of mask mandates. Unfortunately, the argument from analogy relies on several highly disputed assumptions. We examine those assumptions next.
Theoretical Reasons for Doubt
In a world where everyone properly used their mask when a mandate was in place, and where individuals didn’t engage in increasingly risky behaviour to offset the safety conferred by the use of masks, mask mandates would be tightly linked to saving lives. One could imagine that in such a world mandating masks would be justified by the lives such policies saved. However, for the argument from analogy to work, there must be several highly questionable assumptions made that are not adequately defended in Spelman’s article. Let’s unpack each of the assumptions underlying the argument from analogy.
1. The first assumption by the argument from analogy: mostly untrained folks can properly apply and discard masks. One may object that applying and discarding masks is so simple that most people, without training, should be up to the task. However, one could say the same of driving a car: it is easy enough to drive a car or truck without any prior training since one need only know how to break, accelerate, rotate the steering wheel. And yet, if one distributed cars to wholly untrained individuals and allowed them to operate the vehicles at will, the results would probably be disastrous.
A second reason to doubt that usage and discarding is simple enough not to need training: OSHA (Occupational Safety and Health Administration), insists that individuals need to be trained in the use of personal protective equipment including masks. To quote OSHA’s handbook,
Employers shall make sure that each employee demonstrates an understanding of the PPE training … as well as the ability to properly wear and use PPE before they are allowed to perform work requiring the use of the PPE. If an employer believes that a previously trained employee is not demonstrating the proper understanding and skill level in the use of PPE, that employee should receive retraining.
‘[T]he moral philosopher Jonathan Spelman argued that states should ban maskless shopping via legal mandates due to the fact that maskless shopping is worse than drunk driving.’
The claim here is not that (untrained) people would be unable to properly use and discard masks, but that one must assume that enough people have the ability to properly use and discard masks for the argument from analogy to work. And the fact that organizations specialize in the use of such equipment is at least some reason to doubt that enough untrained people have the ability to properly use and discard masks such that mask mandates would save lives.
2. The second assumption by the argument from analogy: shoppers will sufficiently comply with mask mandates such that the mandates will save lives. Unfortunately, the assumption is dubious: for one thing, many people resented and highly disliked having to wear a mask to a grocery shop, not only because many people often don’t like being forced to do something they didn’t choose, but because wearing a mask can be uncomfortable. Whatever one thinks of their reasons, there will likely be many people who appear to comply with mask mandates but who do not actually comply with them. Indeed, there are many ways to look compliant without actually complying. For example, one could wear a mask under their nose, have a face covering that looks like a mask but is too thin to offer protection, or they could have it pulled away from their face just enough to render the mask largely ineffective. I leave it to the readers to concoct other ways that people could (and did) pass as compliant without the compliance.
Here a critic may object that even if people lack the competence necessary to wear and dispose of masks properly, and some people will only appear to comply with mask mandates, but will not comply rendering their masks less effective, at the margins, widespread masked shopping is better at saving lives than widespread unmasked shopping. The improvement at the margins is still good reason to favour mask mandates – call this the marginal objection.
3. An appeal to the marginal effectiveness of masks, even those poorly worn, is a reasonable point, except for a minor detail: it ignores the Peltzman Effect. The rough idea underlying the effect is that safety improvements – that is, seat belts, American football helmets – reduce the expected cost of risky behaviour. And as a result, individuals can be expected to increase risk taking behaviour because such behaviour is less costly overall because of safety improvements. People tend to adjust their behaviour in response to perceived risks: they become more careful when they sense greater risk, and less careful as they feel safer and better protected. The history of the Peltzman effect goes back to the first seat belt laws: although the use of seat belts and other safety gear had the effect of reducing the risk of death if an accident occurred, there was no actual decrease in the number of automotive deaths because the improved safety features were offset by an increase in the number of accidents that occurred due to increased risky behaviour. And while the application of the Peltzman effect was first applied to seat belts, the effect is noticeable across a wide range of safety improvements and risky behaviour, such as sunscreen and health insurance.
The Peltzman effect, then, looks like major worry for the marginal objection. After all, if we have good empirical evidence that people tend to engage in riskier behaviour when they feel safer, then we should expect that people would take more risks overall when required to wear masks than they would otherwise as a result of an increased sense of safety. We would thus predict, on the basis of the Peltzman effect, that people would take more risks when they are aware that a mask mandate is in place in their state by, for example, not switching masks as often, not paying as much attention to fitting the mask properly, going out more often to bars and restaurants, and so forth. We can easily imagine various scenarios where people engage in riskier behaviour than they would otherwise because a mask mandate is in place, reassured into thinking other people will be more likely masked up, and so there will be less risk of infection from others.
While the Peltzman effect doesn’t prove that mask mandates wouldn’t do much to decrease the threat of Covid-19, because people would engage in risky behaviour that offsets the safety gains of wearing a mask, but instead appealing to the effect is good reason to be highly sceptical that mask mandates (as a policy) would result in a net increase in safety. And since a decrease in Covid-19 deaths and hospitalization supposedly justifies state-level mask mandates – just like a reduction in traffic deaths and injuries justifies state laws against drunk driving – we should doubt that justification. Perhaps, though, the empirical evidence vindicates the case for mask mandates.
The Empirical Evidence Isn’t Better
Perhaps, though, the theoretical reasons we highlighted aren’t the end of the matter. After all, we can have compelling theoretical reasons to expect something which isn’t borne out by the empirical evidence. Examples of the mismatch between theory and evidence abound. Indeed, in the course of making the argument from analogy, Spelman appeals to empirical evidence:
‘The history of the Peltzman effect goes back to the first seat belt laws: although the use of seat belts and other safety gear had the effect of reducing the risk of death if an accident occurred, there was no actual decrease in the number of automotive deaths because the improved safety features were offset by an increase in the number of accidents that occurred due to increased risky behaviour.’
This suggests that increasing mask-wearing is crucial to reducing the Covid-19 infection rates, especially in places where people cannot physically distance. And since outlawing mask-less shopping increases mask-wearing, it should come as no surprise that a study by Wei Lyu and George L. Wehby published in Health Affairs found that ‘US states mandating the use of face masks in public had a greater decline in daily COVID-19 growth rates after issuing these mandates compared with states that did not issue mandates.’
While this study is supportive of the argument from analogy, it is undermined by further, better studies. A recent Cochrane metastudy found that, ‘Compared with wearing no mask in the community studies only, wearing a mask may make little to no difference in how many people caught a flu‐like illness/COVID‐like illness …’ And that’s what we should expect in virtue of the theoretical reasons canvassed earlier, especially given the Peltzman effect. We shouldn’t expect there would be positive effects to mask mandates if they gave people a false sense of security, thereby lowering the perceived cost of risky behaviour, and incentivizing them to engage in offsetting risky behaviour. Of course, the Cochrane metastudy is too thin to conclude that mask mandates don’t work, but it does provide reason to be sceptical that mask mandates are justified on the basis of lives saved.
While it is natural to assume that mask mandates would increase mask wearing, and this increase would decrease casualties and hospitalization, the evidence just doesn’t bear this out. However, it bears repeating: the point is to highlight that policy sometimes fails to have the intended effect – people’s behaviour is often hard to predict – but that doesn’t mean an individual couldn’t benefit others by wearing a mask properly, without engaging in offsetting risky behaviour. It just means that one should be careful when mixing public policy and moral philosophy.
