Finalist: Zeynep Tufekci

We can learn from our failures.

When the polio vaccine was declared safe and effective, the news was met with jubilant celebration. Church bells rang across the nation, and factories blew their whistles. “Polio routed!” newspaper headlines exclaimed. “An historic victory,” “monumental,” “sensational,” newscasters declared. People erupted with joy across the United States. Some danced in the streets; others wept. Kids were sent home from school to celebrate.

One might have expected the initial approval of the coronavirus vaccines to spark similar jubilation—especially after a brutal pandemic year. But that didn’t happen. Instead, the steady drumbeat of good news about the vaccines has been met with a chorus of relentless pessimism.

The problem is not that the good news isn’t being reported, or that we should throw caution to the wind just yet. It’s that neither the reporting nor the public-health messaging has reflected the truly amazing reality of these vaccines. There is nothing wrong with realism and caution, but effective communication requires a sense of proportion—distinguishing between due alarm and alarmism; warranted, measured caution and doombait; worst-case scenarios and claims of impending catastrophe. We need to be able to celebrate profoundly positive news while noting the work that still lies ahead. However, instead of balanced optimism since the launch of the vaccines, the public has been offered a lot of misguided fretting over new virus variants, subjected to misleading debates about the inferiority of certain vaccines, and presented with long lists of things vaccinated people still cannot do, while media outlets wonder whether the pandemic will ever end.

This pessimism is sapping people of energy to get through the winter, and the rest of this pandemic. Anti-vaccination groups and those opposing the current public-health measures have been vigorously amplifying the pessimistic messages—especially the idea that getting vaccinated doesn’t mean being able to do more—telling their audiences that there is no point in compliance, or in eventual vaccination, because it will not lead to any positive changes. They are using the moment and the messaging to deepen mistrust of public-health authorities, accusing them of moving the goalposts and implying that we’re being conned. Either the vaccines aren’t as good as claimed, they suggest, or the real goal of pandemic-safety measures is to control the public, not the virus.

Five key fallacies and pitfalls have affected public-health messaging, as well as media coverage, and have played an outsize role in derailing an effective pandemic response. These problems were deepened by the ways that we—the public—developed to cope with a dreadful situation under great uncertainty. And now, even as vaccines offer brilliant hope, and even though, at least in the United States, we no longer have to deal with the problem of a misinformer in chief, some officials and media outlets are repeating many of the same mistakes in handling the vaccine rollout.

The pandemic has given us an unwelcome societal stress test, revealing the cracks and weaknesses in our institutions and our systems. Some of these are common to many contemporary problems, including political dysfunction and the way our public sphere operates. Others are more particular, though not exclusive, to the current challenge—including a gap between how academic research operates and how the public understands that research, and the ways in which the psychology of coping with the pandemic have distorted our response to it.

Recognizing all these dynamics is important, not only for seeing us through this pandemic—yes, it is going to end—but also to understand how our society functions, and how it fails. We need to start shoring up our defenses, not just against future pandemics but against all the myriad challenges we face—political, environmental, societal, and technological. None of these problems is impossible to remedy, but first we have to acknowledge them and start working to fix them—and we’re running out of time.

The past 12 months were incredibly challenging for almost everyone. Public-health officials were fighting a devastating pandemic and, at least in this country, an administration hell-bent on undermining them. The World Health Organization was not structured or funded for independence or agility, but still worked hard to contain the disease. Many researchers and experts noted the absence of timely and trustworthy guidelines from authorities, and tried to fill the void by communicating their findings directly to the public on social media. Reporters tried to keep the public informed under time and knowledge constraints, which were made more severe by the worsening media landscape. And the rest of us were trying to survive as best we could, looking for guidance where we could, and sharing information when we could, but always under difficult, murky conditions.

Despite all these good intentions, much of the public-health messaging has been profoundly counterproductive. In five specific ways, the assumptions made by public officials, the choices made by traditional media, the way our digital public sphere operates, and communication patterns between academic communities and the public proved flawed.

Risk Compensation

One of the most important problems undermining the pandemic response has been the mistrust and paternalism that some public-health agencies and experts have exhibited toward the public. A key reason for this stance seems to be that some experts feared that people would respond to something that increased their safety—such as masks, rapid tests, or vaccines—by behaving recklessly. They worried that a heightened sense of safety would lead members of the public to take risks that would not just undermine any gains, but reverse them.

The theory that things that improve our safety might provide a false sense of security and lead to reckless behavior is attractive—it’s contrarian and clever, and fits the “here’s something surprising we smart folks thought about” mold that appeals to, well, people who think of themselves as smart. Unsurprisingly, such fears have greeted efforts to persuade the public to adopt almost every advance in safety, including seat belts, helmets, and condoms.

But time and again, the numbers tell a different story: Even if safety improvements cause a few people to behave recklessly, the benefits overwhelm the ill effects. In any case, most people are already interested in staying safe from a dangerous pathogen. Further, even at the beginning of the pandemic, sociological theory predicted that wearing masks would be associated with increased adherence to other precautionary measures—people interested in staying safe are interested in staying safe—and empirical research quickly confirmed exactly that. Unfortunately, though, the theory of risk compensation—and its implicit assumptions—continue to haunt our approach, in part because there hasn’t been a reckoning with the initial missteps.

Rules in Place of Mechanisms and Intuitions

Much of the public messaging focused on offering a series of clear rules to ordinary people, instead of explaining in detail the mechanisms of viral transmission for this pathogen. A focus on explaining transmission mechanisms, and updating our understanding over time, would have helped empower people to make informed calculations about risk in different settings. Instead, both the CDC and the WHO chose to offer fixed guidelines that lent a false sense of precision.

In the United States, the public was initially told that “close contact” meant coming within six feet of an infected individual, for 15 minutes or more. This messaging led to ridiculous gaming of the rules; some establishments moved people around at the 14th minute to avoid passing the threshold. It also led to situations in which people working indoors with others, but just outside the cutoff of six feet, felt that they could take their mask off. None of this made any practical sense. What happened at minute 16? Was seven feet okay? Faux precision isn’t more informative; it’s misleading.

All of this was complicated by the fact that key public-health agencies like the CDC and the WHO were late to acknowledge the importance of some key infection mechanisms, such as aerosol transmission. Even when they did so, the shift happened without a proportional change in the guidelines or the messaging—it was easy for the general public to miss its significance.

Frustrated by the lack of public communication from health authorities, I wrote an article last July on what we then knew about the transmission of this pathogen—including how it could be spread via aerosols that can float and accumulate, especially in poorly ventilated indoor spaces. To this day, I’m contacted by people who describe workplaces that are following the formal guidelines, but in ways that defy reason: They’ve installed plexiglass, but barred workers from opening their windows; they’ve mandated masks, but only when workers are within six feet of one another, while permitting them to be taken off indoors during breaks.

Perhaps worst of all, our messaging and guidelines elided the difference between outdoor and indoor spaces, where, given the importance of aerosol transmission, the same precautions should not apply. This is especially important because this pathogen is overdispersed: Much of the spread is driven by a few people infecting many others at once, while most people do not transmit the virus at all.

After I wrote an article explaining how overdispersion and super-spreading were driving the pandemic, I discovered that this mechanism had also been poorly explained. I was inundated by messages from people, including elected officials around the world, saying they had no idea that this was the case. None of it was secret—numerous academic papers and articles had been written about it—but it had not been integrated into our messaging or our guidelines despite its great importance.

Crucially, super-spreading isn’t equally distributed; poorly ventilated indoor spaces can facilitate the spread of the virus over longer distances, and in shorter periods of time, than the guidelines suggested, and help fuel the pandemic.

Outdoors? It’s the opposite.

There is a solid scientific reason for the fact that there are relatively few documented cases of transmission outdoors, even after a year of epidemiological work: The open air dilutes the virus very quickly, and the sun helps deactivate it, providing further protection. And super-spreading—the biggest driver of the pandemic— appears to be an exclusively indoor phenomenon. I’ve been tracking every report I can find for the past year, and have yet to find a confirmed super-spreading event that occurred solely outdoors. Such events might well have taken place, but if the risk were great enough to justify altering our lives, I would expect at least a few to have been documented by now.

And yet our guidelines do not reflect these differences, and our messaging has not helped people understand these facts so that they can make better choices. I published my first article pleading for parks to be kept open on April 7, 2020—but outdoor activities are still banned by some authorities today, a full year after this dreaded virus began to spread globally.

We’d have been much better off if we gave people a realistic intuition about this virus’s transmission mechanisms. Our public guidelines should have been more like Japan’s, which emphasize avoiding the three C’s—closed spaces, crowded places, and close contact—that are driving the pandemic.

Scolding and Shaming

Throughout the past year, traditional and social media have been caught up in a cycle of shaming—made worse by being so unscientific and misguided. How dare you go to the beach? newspapers have scolded us for months, despite lacking evidence that this posed any significant threat to public health. It wasn’t just talk: Many cities closed parks and outdoor recreational spaces, even as they kept open indoor dining and gyms. Just this month, UC Berkeley and the University of Massachusetts at Amherst both banned students from taking even solitary walks outdoors.

Even when authorities relax the rules a bit, they do not always follow through in a sensible manner. In the United Kingdom, after some locales finally started allowing children to play on playgrounds—something that was already way overdue—they quickly ruled that parents must not socialize while their kids have a normal moment. Why not? Who knows?

On social media, meanwhile, pictures of people outdoors without masks draw reprimands, insults, and confident predictions of super-spreading—and yet few note when super-spreading fails to follow.

While visible but low-risk activities attract the scolds, other actual risks—in workplaces and crowded households, exacerbated by the lack of testing or paid sick leave—are not as easily accessible to photographers. Stefan Baral, an associate epidemiology professor at the Johns Hopkins Bloomberg School of Public Health, says that it’s almost as if we’ve “designed a public-health response most suitable for higher-income” groups and the “Twitter generation”—stay home; have your groceries delivered; focus on the behaviors you can photograph and shame online—rather than provide the support and conditions necessary for more people to keep themselves safe.

And the viral videos shaming people for failing to take sensible precautions, such as wearing masks indoors, do not necessarily help. For one thing, fretting over the occasional person throwing a tantrum while going unmasked in a supermarket distorts the reality: Most of the public has been complying with mask wearing. Worse, shaming is often an ineffective way of getting people to change their behavior, and it entrenches polarization and discourages disclosure, making it harder to fight the virus. Instead, we should be emphasizing safer behavior and stressing how many people are doing their part, while encouraging others to do the same.

Harm Reduction

Amidst all the mistrust and the scolding, a crucial public-health concept fell by the wayside. Harm reduction is the recognition that if there is an unmet and yet crucial human need, we cannot simply wish it away; we need to advise people on how to do what they seek to do more safely. Risk can never be completely eliminated; life requires more than futile attempts to bring risk down to zero. Pretending we can will away complexities and trade-offs with absolutism is counterproductive. Consider abstinence-only education: Not letting teenagers know about ways to have safer sex results in more of them having sex with no protections.

As Julia Marcus, an epidemiologist and associate professor at Harvard Medical School, told me, “When officials assume that risks can be easily eliminated, they might neglect the other things that matter to people: staying fed and housed, being close to loved ones, or just enjoying their lives. Public health works best when it helps people find safer ways to get what they need and want.”

Another problem with absolutism is the “abstinence violation” effect, Joshua Barocas, an assistant professor at the Boston University School of Medicine and Infectious Diseases, told me. When we set perfection as the only option, it can cause people who fall short of that standard in one small, particular way to decide that they’ve already failed, and might as well give up entirely. Most people who have attempted a diet or a new exercise regimen are familiar with this psychological state. The better approach is encouraging risk reduction and layered mitigation—emphasizing that every little bit helps—while also recognizing that a risk-free life is neither possible nor desirable.

Socializing is not a luxury—kids need to play with one another, and adults need to interact. Your kids can play together outdoors, and outdoor time is the best chance to catch up with your neighbors is not just a sensible message; it’s a way to decrease transmission risks. Some kids will play and some adults will socialize no matter what the scolds say or public-health officials decree, and they’ll do it indoors, out of sight of the scolding.

And if they don’t? Then kids will be deprived of an essential activity, and adults will be deprived of human companionship. Socializing is perhaps the most important predictor of health and longevity, after not smoking and perhaps exercise and a healthy diet. We need to help people socialize more safely, not encourage them to stop socializing entirely.

The Balance Between Knowledge And Action

Last but not least, the pandemic response has been distorted by a poor balance between knowledge, risk, certainty, and action.

Sometimes, public-health authorities insisted that we did not know enough to act, when the preponderance of evidence already justified precautionary action. Wearing masks, for example, posed few downsides, and held the prospect of mitigating the exponential threat we faced. The wait for certainty hampered our response to airborne transmission, even though there was almost no evidence for—and increasing evidence against—the importance of fomites, or objects that can carry infection. And yet, we emphasized the risk of surface transmission while refusing to properly address the risk of airborne transmission, despite increasing evidence. The difference lay not in the level of evidence and scientific support for either theory—which, if anything, quickly tilted in favor of airborne transmission, and not fomites, being crucial—but in the fact that fomite transmission had been a key part of the medical canon, and airborne transmission had not.

Sometimes, experts and the public discussion failed to emphasize that we were balancing risks, as in the recurring cycles of debate over lockdowns or school openings. We should have done more to acknowledge that there were no good options, only trade-offs between different downsides. As a result, instead of recognizing the difficulty of the situation, too many people accused those on the other side of being callous and uncaring.

And sometimes, the way that academics communicate clashed with how the public constructs knowledge. In academia, publishing is the coin of the realm, and it is often done through rejecting the null hypothesis—meaning that many papers do not seek to prove something conclusively, but instead, to reject the possibility that a variable has no relationship with the effect they are measuring (beyond chance). If that sounds convoluted, it is—there are historical reasons for this methodology and big arguments within academia about its merits, but for the moment, this remains standard practice.

At crucial points during the pandemic, though, this resulted in mistranslations and fueled misunderstandings, which were further muddled by differing stances toward prior scientific knowledge and theory. Yes, we faced a novel coronavirus, but we should have started by assuming that we could make some reasonable projections from prior knowledge, while looking out for anything that might prove different. That prior experience should have made us mindful of seasonality, the key role of overdispersion, and aerosol transmission. A keen eye for what was different from the past would have alerted us earlier to the importance of presymptomatic transmission.

Thus, on January 14, 2020, the WHO stated that there was “no clear evidence of human-to-human transmission.” It should have said, “There is increasing likelihood that human-to-human transmission is taking place, but we haven’t yet proven this, because we have no access to Wuhan, China.” (Cases were already popping up around the world at that point.) Acting as if there was human-to-human transmission during the early weeks of the pandemic would have been wise and preventive.

Later that spring, WHO officials stated that there was “currently no evidence that people who have recovered from COVID-19 and have antibodies are protected from a second infection,” producing many articles laden with panic and despair. Instead, it should have said: “We expect the immune system to function against this virus, and to provide some immunity for some period of time, but it is still hard to know specifics because it is so early.”

Similarly, since the vaccines were announced, too many statements have emphasized that we don’t yet know if vaccines prevent transmission. Instead, public-health authorities should have said that we have many reasons to expect, and increasing amounts of data to suggest, that vaccines will blunt infectiousness, but that we’re waiting for additional data to be more precise about it. That’s been unfortunate, because while many, many things have gone wrong during this pandemic, the vaccines are one thing that has gone very, very right.

As late as april 2020, Anthony Fauci was slammed for being too optimistic for suggesting we might plausibly have vaccines in a year to 18 months. We had vaccines much, much sooner than that: The first two vaccine trials concluded a mere eight months after the WHO declared a pandemic in March 2020.

Moreover, they have delivered spectacular results. In June 2020, the FDA said a vaccine that was merely 50 percent efficacious in preventing symptomatic COVID-19 would receive emergency approval—that such a benefit would be sufficient to justify shipping it out immediately. Just a few months after that, the trials of the Moderna and Pfizer vaccines concluded by reporting not just a stunning 95 percent efficacy, but also a complete elimination of hospitalization or death among the vaccinated. Even severe disease was practically gone: The lone case classified as “severe” among 30,000 vaccinated individuals in the trials was so mild that the patient needed no medical care, and her case would not have been considered severe if her oxygen saturation had been a single percent higher.

These are exhilarating developments, because global, widespread, and rapid vaccination is our way out of this pandemic. Vaccines that drastically reduce hospitalizations and deaths, and that diminish even severe disease to a rare event, are the closest things we have had in this pandemic to a miracle—though of course they are the product of scientific research, creativity, and hard work. They are going to be the panacea and the endgame.

And yet, two months into an accelerating vaccination campaign in the United States, it would be hard to blame people if they missed the news that things are getting better.

Yes, there are new variants of the virus, which may eventually require booster shots, but at least so far, the existing vaccines are standing up to them well—very, very well. Manufacturers are already working on new vaccines or variant-focused booster versions, in case they prove necessary, and the authorizing agencies are ready for a quick turnaround if and when updates are needed. Reports from places that have vaccinated large numbers of individuals, and even trials in places where variants are widespread, are exceedingly encouraging, with dramatic reductions in cases and, crucially, hospitalizations and deaths among the vaccinated. Global equity and access to vaccines remain crucial concerns, but the supply is increasing.

Here in the United States, despite the rocky rollout and the need to smooth access and ensure equity, it’s become clear that toward the end of spring 2021, supply will be more than sufficient. It may sound hard to believe today, as many who are desperate for vaccinations await their turn, but in the near future, we may have to discuss what to do with excess doses.

So why isn’t this story more widely appreciated?

Part of the problem with the vaccines was the timing—the trials concluded immediately after the U.S. election, and their results got overshadowed in the weeks of political turmoil. The first, modest headline announcing the Pfizer-BioNTech results in The New York Times was a single column, “Vaccine Is Over 90% Effective, Pfizer’s Early Data Says,” below a banner headline spanning the page: “biden calls for united front as virus rages.” That was both understandable—the nation was weary—and a loss for the public.

Just a few days later, Moderna reported a similar 94.5 percent efficacy. If anything, that provided even more cause for celebration, because it confirmed that the stunning numbers coming out of Pfizer weren’t a fluke. But, still amid the political turmoil, the Moderna report got a mere two columns on The New York Times’ front page with an equally modest headline: “Another Vaccine Appears to Work Against the Virus.”

So we didn’t get our initial vaccine jubilation.

But as soon as we began vaccinating people, articles started warning the newly vaccinated about all they could not do. “COVID-19 Vaccine Doesn’t Mean You Can Party Like It’s 1999,” one headline admonished. And the buzzkill has continued right up to the present. “You’re fully vaccinated against the coronavirus—now what? Don’t expect to shed your mask and get back to normal activities right away,” began a recent Associated Press story.

People might well want to party after being vaccinated. Those shots will expand what we can do, first in our private lives and among other vaccinated people, and then, gradually, in our public lives as well. But once again, the authorities and the media seem more worried about potentially reckless behavior among the vaccinated, and about telling them what not to do, than with providing nuanced guidance reflecting trade-offs, uncertainty, and a recognition that vaccination can change behavior. No guideline can cover every situation, but careful, accurate, and updated information can empower everyone.

Take the messaging and public conversation around transmission risks from vaccinated people. It is, of course, important to be alert to such considerations: Many vaccines are “leaky” in that they prevent disease or severe disease, but not infection and transmission. In fact, completely blocking all infection—what’s often called “sterilizing immunity”—is a difficult goal, and something even many highly effective vaccines don’t attain, but that doesn’t stop them from being extremely useful.

As Paul Sax, an infectious-disease doctor at Boston’s Brigham & Women’s Hospital, put it in early December, it would be enormously surprising “if these highly effective vaccines didn’t also make people less likely to transmit.” From multiple studies, we already knew that asymptomatic individuals—those who never developed COVID-19 despite being infected—were much less likely to transmit the virus. The vaccine trials were reporting 95 percent reductions in any form of symptomatic disease. In December, we learned that Moderna had swabbed some portion of trial participants to detect asymptomatic, silent infections, and found an almost two-thirds reduction even in such cases. The good news kept pouring in. Multiple studies found that, even in those few cases where breakthrough disease occurred in vaccinated people, their viral loads were lower—which correlates with lower rates of transmission. Data from vaccinated populations further confirmed what many experts expected all along: Of course these vaccines reduce transmission.

And yet, from the beginning, a good chunk of the public-facing messaging and news articles implied or claimed that vaccines won’t protect you against infecting other people or that we didn’t know if they would, when both were false. I found myself trying to convince people in my own social network that vaccines weren’t useless against transmission, and being bombarded on social media with claims that they were.

What went wrong? The same thing that’s going wrong right now with the reporting on whether vaccines will protect recipients against the new viral variants. Some outlets emphasize the worst or misinterpret the research. Some public-health officials are wary of encouraging the relaxation of any precautions. Some prominent experts on social media—even those with seemingly solid credentials—tend to respond to everything with alarm and sirens. So the message that got heard was that vaccines will not prevent transmission, or that they won’t work against new variants, or that we don’t know if they will. What the public needs to hear, though, is that based on existing data, we expect them to work fairly well—but we’ll learn more about precisely how effective they’ll be over time, and that tweaks may make them even better.

A year into the pandemic, we’re still repeating the same mistakes.

The top-down messaging is not the only problem. The scolding, the strictness, the inability to discuss trade-offs, and the accusations of not caring about people dying not only have an enthusiastic audience, but portions of the public engage in these behaviors themselves. Maybe that’s partly because proclaiming the importance of individual actions makes us feel as if we are in the driver’s seat, despite all the uncertainty.

Psychologists talk about the “locus of control”—the strength of belief in control over your own destiny. They distinguish between people with more of an internal-control orientation—who believe that they are the primary actors—and those with an external one, who believe that society, fate, and other factors beyond their control greatly influence what happens to us. This focus on individual control goes along with something called the “fundamental attribution error”—when bad things happen to other people, we’re more likely to believe that they are personally at fault, but when they happen to us, we are more likely to blame the situation and circumstances beyond our control.

An individualistic locus of control is forged in the U.S. mythos—that we are a nation of strivers and people who pull ourselves up by our bootstraps. An internal-control orientation isn’t necessarily negative; it can facilitate resilience, rather than fatalism, by shifting the focus to what we can do as individuals even as things fall apart around us. This orientation seems to be common among children who not only survive but sometimes thrive in terrible situations—they take charge and have a go at it, and with some luck, pull through. It is probably even more attractive to educated, well-off people who feel that they have succeeded through their own actions.

You can see the attraction of an individualized, internal locus of control in a pandemic, as a pathogen without a cure spreads globally, interrupts our lives, makes us sick, and could prove fatal.

There have been very few things we could do at an individual level to reduce our risk beyond wearing masks, distancing, and disinfecting. The desire to exercise personal control against an invisible, pervasive enemy is likely why we’ve continued to emphasize scrubbing and cleaning surfaces, in what’s appropriately called “hygiene theater,” long after it became clear that fomites were not a key driver of the pandemic. Obsessive cleaning gave us something to do, and we weren’t about to give it up, even if it turned out to be useless. No wonder there was so much focus on telling others to stay home—even though it’s not a choice available to those who cannot work remotely—and so much scolding of those who dared to socialize or enjoy a moment outdoors.

And perhaps it was too much to expect a nation unwilling to release its tight grip on the bottle of bleach to greet the arrival of vaccines—however spectacular—by imagining the day we might start to let go of our masks.

The focus on individual actions has had its upsides, but it has also led to a sizable portion of pandemic victims being erased from public conversation. If our own actions drive everything, then some other individuals must be to blame when things go wrong for them. And throughout this pandemic, the mantra many of us kept repeating—“Wear a mask, stay home; wear a mask, stay home”—hid many of the real victims.

Study after study, in country after country, confirms that this disease has disproportionately hit the poor and minority groups, along with the elderly, who are particularly vulnerable to severe disease. Even among the elderly, though, those who are wealthier and enjoy greater access to health care have fared better.

The poor and minority groups are dying in disproportionately large numbers for the same reasons that they suffer from many other diseases: a lifetime of disadvantages, lack of access to health care, inferior working conditions, unsafe housing, and limited financial resources.

Many lacked the option of staying home precisely because they were working hard to enable others to do what they could not, by packing boxes, delivering groceries, producing food. And even those who could stay home faced other problems born of inequality: Crowded housing is associated with higher rates of COVID-19 infection and worse outcomes, likely because many of the essential workers who live in such housing bring the virus home to elderly relatives.

Individual responsibility certainly had a large role to play in fighting the pandemic, but many victims had little choice in what happened to them. By disproportionately focusing on individual choices, not only did we hide the real problem, but we failed to do more to provide safe working and living conditions for everyone.

For example, there has been a lot of consternation about indoor dining, an activity I certainly wouldn’t recommend. But even takeout and delivery can impose a terrible cost: One study of California found that line cooks are the highest-risk occupation for dying of COVID-19. Unless we provide restaurants with funds so they can stay closed, or provide restaurant workers with high-filtration masks, better ventilation, paid sick leave, frequent rapid testing, and other protections so that they can safely work, getting food to go can simply shift the risk to the most vulnerable. Unsafe workplaces may be low on our agenda, but they do pose a real danger. Bill Hanage, associate professor of epidemiology at Harvard, pointed me to a paper he co-authored: Workplace-safety complaints to OSHA—which oversees occupational-safety regulations—during the pandemic were predictive of increases in deaths 16 days later.

New data highlight the terrible toll of inequality: Life expectancy has decreased dramatically over the past year, with Black people losing the most from this disease, followed by members of the Hispanic community. Minorities are also more likely to die of COVID-19 at a younger age. But when the new CDC director, Rochelle Walensky, noted this terrible statistic, she immediately followed up by urging people to “continue to use proven prevention steps to slow the spread—wear a well-fitting mask, stay 6 ft away from those you do not live with, avoid crowds and poorly ventilated places, and wash hands often.”

Those recommendations aren’t wrong, but they are incomplete. None of these individual acts do enough to protect those to whom such choices aren’t available—and the CDC has yet to issue sufficient guidelines for workplace ventilation or to make higher-filtration masks mandatory, or even available, for essential workers. Nor are these proscriptions paired frequently enough with prescriptions: Socialize outdoors, keep parks open, and let children play with one another outdoors.

Vaccines are the tool that will end the pandemic. The story of their rollout combines some of our strengths and our weaknesses, revealing the limitations of the way we think and evaluate evidence, provide guidelines, and absorb and react to an uncertain and difficult situation.

But also, after a weary year, maybe it’s hard for everyone—including scientists, journalists, and public-health officials—to imagine the end, to have hope. We adjust to new conditions fairly quickly, even terrible new conditions. During this pandemic, we’ve adjusted to things many of us never thought were possible. Billions of people have led dramatically smaller, circumscribed lives, and dealt with closed schools, the inability to see loved ones, the loss of jobs, the absence of communal activities, and the threat and reality of illness and death.

Hope nourishes us during the worst times, but it is also dangerous. It upsets the delicate balance of survival—where we stop hoping and focus on getting by—and opens us up to crushing disappointment if things don’t pan out. After a terrible year, many things are understandably making it harder for us to dare to hope. But, especially in the United States, everything looks better by the day. Tragically, at least 28 million Americans have been confirmed to have been infected, but the real number is certainly much higher. By one estimate, as many as 80 million have already been infected with COVID-19, and many of those people now have some level of immunity. Another 46 million people have already received at least one dose of a vaccine, and we’re vaccinating millions more each day as the supply constraints ease. The vaccines are poised to reduce or nearly eliminate the things we worry most about—severe disease, hospitalization, and death.

Not all our problems are solved. We need to get through the next few months, as we race to vaccinate against more transmissible variants. We need to do more to address equity in the United States—because it is the right thing to do, and because failing to vaccinate the highest-risk people will slow the population impact. We need to make sure that vaccines don’t remain inaccessible to poorer countries. We need to keep up our epidemiological surveillance so that if we do notice something that looks like it may threaten our progress, we can respond swiftly.

And the public behavior of the vaccinated cannot change overnight—even if they are at much lower risk, it’s not reasonable to expect a grocery store to try to verify who’s vaccinated, or to have two classes of people with different rules. For now, it’s courteous and prudent for everyone to obey the same guidelines in many public places. Still, vaccinated people can feel more confident in doing things they may have avoided, just in case—getting a haircut, taking a trip to see a loved one, browsing for nonessential purchases in a store.

But it is time to imagine a better future, not just because it’s drawing nearer but because that’s how we get through what remains and keep our guard up as necessary. It’s also realistic—reflecting the genuine increased safety for the vaccinated.

Public-health agencies should immediately start providing expanded information to vaccinated people so they can make informed decisions about private behavior. This is justified by the encouraging data, and a great way to get the word out on how wonderful these vaccines really are. The delay itself has great human costs, especially for those among the elderly who have been isolated for so long.

Public-health authorities should also be louder and more explicit about the next steps, giving us guidelines for when we can expect easing in rules for public behavior as well. We need the exit strategy spelled out—but with graduated, targeted measures rather than a one-size-fits-all message. We need to let people know that getting a vaccine will almost immediately change their lives for the better, and why, and also when and how increased vaccination will change more than their individual risks and opportunities, and see us out of this pandemic.

We should encourage people to dream about the end of this pandemic by talking about it more, and more concretely: the numbers, hows, and whys. Offering clear guidance on how this will end can help strengthen people’s resolve to endure whatever is necessary for the moment—even if they are still unvaccinated—by building warranted and realistic anticipation of the pandemic’s end.

Hope will get us through this. And one day soon, you’ll be able to hop off the subway on your way to a concert, pick up a newspaper, and find the triumphant headline: “COVID Routed!”

This article has been updated.

A few sentences have shaken a century of science.

A week ago, more than a year after the World Health Organization declared that we face a pandemic, a page on its website titled “Coronavirus Disease (Covid-19): How Is It Transmitted?” got a seemingly small update.

The agency’s response to that question had been that “current evidence suggests that the main way the virus spreads is by respiratory droplets” — which are expelled from the mouth and quickly fall to the ground — “among people who are in close contact with each other.”

The revised response still emphasizes transmission in close contact but now says it may be via aerosols — smaller respiratory particles that can float — as well as droplets. It also adds a reason the virus can also be transmitted “in poorly ventilated and/or crowded indoor settings,” saying this is because “aerosols remain suspended in the air or travel farther than 1 meter.”

The change didn’t get a lot of attention. There was no news conference, no big announcement.

Then, on Friday, the Centers for Disease Control and Prevention also updated its guidance on Covid-19, clearly saying that inhalation of these smaller particles is a key way the virus is transmitted, even at close range, and put it on top of its list of how the disease spreads.

There was no news conference by the C.D.C. either.

But these latest shifts challenge key infection control assumptions that go back a century, putting a lot of what went wrong last year in context. They may also signal one of the most important advancements in public health during this pandemic.

If the importance of aerosol transmission had been accepted early, we would have been told from the beginning that it was much safer outdoors, where these small particles disperse more easily, as long as you avoid close, prolonged contact with others. We would have tried to make sure indoor spaces were well ventilated, with air filtered as necessary. Instead of blanket rules on gatherings, we would have targeted conditions that can produce superspreading events: people in poorly ventilated indoor spaces, especially if engaged over time in activities that increase aerosol production, like shouting and singing. We would have started using masks more quickly, and we would have paid more attention to their fit, too. And we would have been less obsessed with cleaning surfaces.

Our mitigations would have been much more effective, sparing us a great deal of suffering and anxiety.

Since the pandemic is far from over, with countries like India facing devastating surges, we need to understand both why this took so long to come about and what it will mean.

*

Initially, SARS-CoV-2 was seen as a disease spread by respiratory droplets, except in rare cases of aerosol transmission during medical procedures like intubation. Countertops, boxes and other possible fomites — contaminated surfaces — were seen as a threat because if we touched them after droplets fell on them, it was believed the virus could make its way to our hands, then our noses, eyes or mouths.

The implications of this were illustrated when I visited New York City in late April — my first trip there in more than a year.

A giant digital billboard greeted me at Times Square, with the message “Protecting yourself and others from Covid-19. Guidance from the World Health Organization.”

First, “Hygiene” flashed, urging me to wash my hands, “practice respiratory hygiene,” avoid touching my face and wear a mask when necessary. Next, “Social distancing” told me to avoid close contact with people (illustrated by people separated by one meter), avoid shaking hands and stay home if unwell. Then “Medical help” advised me to follow local medical protocols.

I was stunned that the final instruction was “Stay informed.”

That billboard neglected the clearest epidemiological pattern of this pandemic: The vast majority of transmission has been indoors, sometimes beyond a range of three or even six feet. The superspreading events that play a major role in driving the pandemic occur overwhelmingly, if not exclusively, indoors.

The billboard had not a word about ventilation, nothing about opening windows or moving activities outdoors, where transmission has been rare and usually only during prolonged and close contact. (Ireland recently reported 0.1 percent of Covid-19 cases were traced to outdoor transmission.)

The omission is not surprising. Throughout the pandemic, the W.H.O. was slow to accept the key role that infectious particles small enough to float could be playing.

Mary-Louise McLaws, an epidemiologist at the University of New South Wales in Sydney, Australia, and a member of the W.H.O. committees that craft infection prevention and control guidance, wanted all this examined but knew the stakes made it harder to overcome the resistance. She told The Times last year, “If we started revisiting airflow, we would have to be prepared to change a lot of what we do.” She said it was a very good idea, but she added, “It will cause an enormous shudder through the infection control society.”

This assumption that these larger droplets that can travel only a few feet are the main way the disease spreads is one of the key reasons the W.H.O. and the C.D.C. didn’t recommend masks at first. Why bother if one can simply stay out of their range? After the C.D.C. recommended masks in April 2020, the W.H.O. shifted last June, but it first suggested ordinary people generally wear masks if physical distancing could not be maintained, and still said health care workers performing screenings in the community did not need to wear masks if they could stay that single meter away from patients. The W.H.O. last updated its mask guidance in December but continued to insist that mask use indoors was not necessary if people could remain separated by that mere meter — this time conceding that if ventilation might not be adequate, masks should be worn indoors, regardless of distancing.

In contrast, if the aerosols had been considered a major form of transmission, in addition to distancing and masks, advice would have centered on ventilation and airflow, as well as time spent indoors. Small particles can accumulate in enclosed spaces, since they can remain suspended in the air and travel along air currents. This means that indoors, three or even six feet, while helpful, is not completely protective, especially over time.

To see this misunderstanding in action, look at what’s still happening throughout the world. In India, where hospitals have run out of supplemental oxygen and people are dying in the streets, money is being spent on fleets of drones to spray anti-coronavirus disinfectant in outdoor spaces. Parks, beaches and outdoor areas keep getting closed around the world. This year and last, organizers canceled outdoor events for the National Cherry Blossom Festival in Washington, D.C. Cambodian customs officials advised spraying disinfectant outside vehicles imported from India. The examples are many.

Meanwhile, many countries allowed their indoor workplaces to open but with inadequate aerosol protections. There was no attention to ventilation, installing air filters as necessary or even opening windows when possible, more to having people just distancing three or six feet, sometimes not requiring masks beyond that distance, or spending money on hard plastic barriers, which may be useless at best. (Just this week, President Biden visited a school where students were sitting behind plastic shields.)

This occurred throughout the world in the past year. The United States has been a bit better, but the C.D.C. did not really accept aerosol transmission until October, though still relegating it to a secondary role until its change on Friday, which put the risk infection from inhaling these tiny particles first on its list of means of transmission.

*

The scientific wrangling, resistance and controversy that prevented a change in guidance stem from a century of mistaken assumptions whose roots go back to the origins of germ theory of disease in the 19th century.

Until germ theory became established in the 19th century, many people believed that deadly diseases like cholera were caused by miasma — stinking fumes from organic or rotting material. It wasn’t easy to persuade people that creatures so small that they could not be seen in a seemingly innocent glass of water could be claiming so many lives.

This was a high-stakes fight: Getting the transmission mechanisms of a disease wrong can lead to mitigations that not only are ineffective but also make things worse. During the 19th century, fearing miasma, Londoners worked hard to direct their stinky sewers into the nearby Thames River, essentially spreading cholera even more.

But clear evidence doesn’t easily overturn tradition or overcome entrenched feelings and egos. John Snow, often credited as the first scientific epidemiologist, showed that a contaminated well was responsible for a 1854 London cholera epidemic by removing the suspected pump’s handle and documenting how the cases plummeted afterward. Many other scientists and officials wouldn’t believe him for 12 years, when the link to a water source showed up again and became harder to deny. (He died years earlier.)

Similarly, when the Hungarian physician Ignaz Semmelweis realized the importance of washing hands to protect patients, he lost his job and was widely condemned by disbelieving colleagues. He wasn’t always the most tactful communicator, and his colleagues resented his brash implication that they were harming their patients (even though they were). These doctors continued to kill their patients through cross-contamination for decades, despite clear evidence showing how death rates had plummeted in the few wards where midwives and Dr. Semmelweis had succeeded in introducing routine hand hygiene. He ultimately died of an infected wound.

Disentangling causation is difficult, too, because of confusing correlations and conflations. Terrible smells frequently overlap with unsanitary conditions that can contribute to ill health, and in mid-19th-century London, death rates from cholera were higher in parts of the city with poor living conditions.

Along the way to modern public health shaped largely by the fight over germs, a theory of transmission promoted by the influential public health figure Charles Chapin took hold.

Dr. Chapin asserted in the early 1900s that respiratory diseases were most likely spread at close range by people touching bodily fluids or ejecting respiratory droplets, and did not allow for the possibility that such close-range infection could occur by inhaling small floating particles others emitted. He was also concerned that belief in airborne transmission, which he associated with miasma theories, would make people feel helpless and drop their guard against contact transmission. This was a mistake that would haunt infection control for the next century and more.

In modern medical parlance, respiratory transmission routes are divided between the larger droplets, associated with diseases that spread at close distance, and the smaller aerosols (sometimes also called droplet nuclei), associated with diseases like measles that we know can spread at long distance and are usually highly contagious. Indeed, studies showing that respiratory diseases spread more easily in proximity to infected people seemingly confirmed the role of droplets.

It was in this context in early 2020 that the W.H.O. and the C.D.C. asserted that SARS-CoV-2 was transmitted primarily via these heavier, short-range droplets, and provided guidance accordingly.

But from the beginning, the way the disease was spreading around the world did not fit this theory well. In February 2020, after an infected person was found to have boarded the cruise ship Diamond Princess, hundreds of people trapped on board for weeks were infected, including 567 of the 2,666 passengers, who were largely confined to their rooms and delivered food by masked personnel — hard to explain solely with droplet-driven transmission. (Hitoshi Oshitani, a Japanese virologist who played an important role in his country’s response to the epidemic, said it was this ship outbreak that helped convince him this was airborne — and it’s why Japan planned around airborne transmission assumptions from as early as February 2020.)

Then there were the many superspreader events around the world that defied droplet explanations. In March 2020 in Mount Vernon, Wash., 61 pandemic-aware people showed up to a choir practice and sang with some distance between them in a large space, were provided hand sanitizer and left the doors open, reducing the need for people to touch the handles. But 53 of them were confirmed or strongly suspected to have contracted Covid-19 anyway, and two died. Long-distance transmission was being documented as well: One study from China in April 2020, clearly documenting transmission from beyond one meter, had video evidence showing the initially infected person had not come very close to those he infected, and there were no common surfaces touched.

Epidemiological studies and examples kept pouring in, too, all of them showing that Covid-19 was spreading primarily indoors and clusters were concentrated in poorly ventilated spaces. And when outdoor transmission did occur, it was often when people were in prolonged close contact, talking or yelling, as with construction workers on the same site.

The disease was also greatly overdispersed, sometimes being not very contagious and other times dramatically so. Large-scale studies showed that more than 70 percent of infected people did not transmit to any other person, while as few as 5 percent may be responsible for 80 percent of transmissions through superspreading events. Despite databases documenting thousands of indoor superspreader incidents, I’m not aware of a single confirmed outdoor-only case of superspreading.

None of this could be explained easily if the disease were “primarily transmitted between people through respiratory droplets and contact routes,” as the W.H.O. had said, since those larger, heavier particles would behave the same indoors as outdoors, would be largely indifferent to ventilation and would not be conducive to so much superspreading.

Finally, it was clear from early on that people who weren’t yet sick or coughing or sneezing — which produce a lot more droplets — were transmitting and that things correlated with aerosol emissions like talking, yelling and singing were associated with many of the outbreaks.

Amid the growing evidence, in July, hundreds of scientists signed an open letter urging the public health agencies, especially the W.H.O., to address airborne transmission of the coronavirus.

That month, after the open letter, the W.H.O. updated its guidance to say that “short-range aerosol transmission” from infected people in poorly ventilated spaces over time “cannot be ruled out” but went on to say that “the detailed investigations of these clusters suggest that droplet and fomite transmission could also explain human-to-human transmission within these clusters” and that close contact could still be the reason, “especially if hand hygiene was not performed and masks were not used when physical distancing was not maintained.”

Evidence kept accumulating. Transmission was documented in adjacent rooms in a quarantine hotel where people never interacted. Several hospital workers were proved to have been infected despite strict contact and droplet precautions. Viable virus was found in air samples from hospital rooms of Covid-19 patients who hadn’t had aerosol-generating procedures and in an air sample from an infected person’s car. The virus was found in exhaust vents in hospitals, and ferrets in cages connected only via shared air infected each other. And so on.

There were quibbles with each study: Was the sampled virus infective enough? (It is hard to catch the viruses from the air without destroying them.) Could some fomite connection have been missed? Still, it kept getting harder to deny the role of aerosols as a major factor.

Last October, the C.D.C. published updated guidance acknowledging airborne transmission, but as a secondary route under some circumstances, until it acknowledged airborne transmission as crucial on Friday. And the W.H.O. kept inching forward in its public statements, most recently a week ago.

*

Linsey Marr, a professor of engineering at Virginia Tech who made important contributions to our understanding of airborne virus transmission before the pandemic, pointed to two key scientific errors — rooted in a lot of history — that explain the resistance, and also opened a fascinating sociological window into how science can get it wrong and why.

First, Dr. Marr said, the upper limit for particles to be able to float is actually 100 microns, not five microns, as generally thought. The incorrect five-micron claim may have come about because earlier scientists conflated the size at which respiratory particles could reach the lower respiratory tract (important for studying tuberculosis) with the size at which they remain suspended in the air.

Dr. Marr said that if you inhale a particle from the air, it’s an aerosol. She agreed that droplet transmission by a larger respiratory particle is possible, if it lands on the eye, for example, but biomechanically, she said, nasal transmission faces obstacles, since nostrils point downward and the physics of particles that large makes it difficult for them to move up the nose. And in lab measurements, people emit far more of the easier-to-inhale aerosols than the droplets, she said, and even the smallest particles can be virus laden, sometimes more so than the larger ones, seemingly because of how and where they are produced in the respiratory tract.

Second, she said, proximity is conducive to transmission of aerosols as well because aerosols are more concentrated near the person emitting them. In a twist of history, modern scientists have been acting like those who equated stinky air with disease, by equating close contact, a measure of distance, only with the larger droplets, a mechanism of transmission, without examination.

Since aerosols also infect at close range, measures to prevent droplet transmission — masks and distancing — can help dampen transmission for airborne diseases as well. However, this oversight led medical people to circularly assume that if such measures worked at all, droplets must have played a big role in their transmission.

Other incorrect assumptions thrived. For example, in July, right after the letter by the hundreds of scientists challenging the droplet paradigm, Reuters reported that Dr. John Conly, who chairs a key W.H.O. infection prevention working group, said that there would be many more cases if the virus was airborne and asked, “Would we not be seeing, like, literally billions of cases globally?” He made similar claims last month. And he is not the only member of that group to assert this, a common assumption in the world of infection control well into 2021.

However, Dr. Marr pointed out, there are airborne diseases, like measles, that are highly contagious and others, like tuberculosis, that are not. Moreover, while SARS-CoV-2 is certainly not as infectious as measles on average, it can be highly infectious in the superspreading events driving the pandemic.

Many respiratory viruses carried by aerosols survive better in colder environments and lower relative humidity, Dr. Marr said, again fitting the pattern of outbreaks around the world, for example, in many meatpacking plants. Plus, some activities produce more aerosols — talking, yelling, singing, exercising — also fitting the pattern of outbreaks globally.

Why did it take so long to understand all this?

*

One reason is that our institutions weren’t necessarily set up to deal with what we faced. For example, the W.H.O.’s Infection Prevention and Control (I.P.C.) global unit primarily concentrates on health care facilities. Many of the experts they enlisted to form the Covid-19 I.P.C. Guidance Development Group were hospital-focused, and some of them specialized in antibiotic-resistant bacterial infections that can spread wildly in health care facilities when medical personnel fail to regularly wash their hands. So this focus made sense in a prepandemic world. Hospitals employ trained health care workers and are fairly controlled, well-defined settings, with different considerations from those of a pandemic across many environments in the real world. Further, in some countries like the United States, they tend to have extensive engineering controls to dampen infections, involving aggressive air-exchange standards, almost like being outdoors. This is the opposite of modern office and even residential buildings, which tend to be more sealed for energy efficiency. In such a medical environment, hand hygiene is a more important consideration, since ventilation is taken care of.

Another dynamic we’ve seen is something that is not unheard-of in the history of science: setting a higher standard of proof for theories that challenge conventional wisdom than for those that support it.

As part of its assessment of the virus’s spread, the W.H.O. asked a group of scientists last fall to review the evidence on transmission of the coronavirus. When reviewing airborne transmission, the group focused mostly on studies of air samples, especially if live virus was captured from the air, which, as mentioned above, is extremely hard. By that criterion, airborne transmission of the measles virus, which is undisputed, would not be accepted because no one has cultivated that pathogen from room air. That’s also true of tuberculosis. And while scientists, despite the difficulties, had managed to capture viable SARS-CoV-2 in three studies that I’m aware of, the review noted that the virus was detected only intermittently in general, disputed whether the captured live virus was infective enough and ultimately said it could not reach “firm conclusions over airborne transmission.” The lead author and another senior member of the research group previously said they believed transmission was driven by droplets.

The skepticism about airborne transmission is at odds with the acceptance of droplet transmission. Dr. Marr and Joseph Allen, the director of the Healthy Buildings program and an associate professor at Harvard’s T.H. Chan School of Public Health, told me that droplet transmission has never been directly demonstrated. Since Dr. Chapin, close-distance transmission has been seen as proof of droplets unless disproved through much effort, as was finally done for tuberculosis.

Another key problem is that, understandably, we find it harder to walk things back. It is easier to keep adding exceptions and justifications to a belief than to admit that a challenger has a better explanation.

The ancients believed that all celestial objects revolved around the earth in circular orbits. When it became clear that the observed behavior of the celestial objects did not fit this assumption, those astronomers produced ever-more-complex charts by adding epicycles — intersecting arcs and circles — to fit the heavens to their beliefs.

In a contemporary example of this attitude, the initial public health report on the Mount Vernon choir case said that it may have been caused by people “sitting close to one another, sharing snacks and stacking chairs at the end of the practice,” even though almost 90 percent of the people there developed symptoms of Covid-19. Shelly Miller, an aerosol expert at the University of Colorado Boulder, was so struck by the incident that she initiated a study with a team of scientists, documenting that the space was less full than usual, allowing for increased distance, that nobody reported touching anyone else, that hand sanitizer was used and that only three people who had arrived early arranged the chairs. There was no spatial pattern to the transmission, implicating airflows, and there was nobody within nine feet in front of the first known case, who had mild symptoms.

Galileo is said to have murmured, “And yet it moves,” after he was forced to recant his theory that the earth moved around the sun. Scientists who studied bioaerosols could only say, “And yet it floats.”

*

So much of what we have done throughout the pandemic — the excessive hygiene theater and the failure to integrate ventilation and filters into our basic advice — has greatly hampered our response. Some of it, like the way we underused or even shut down outdoor space, isn’t that different from the 19th-century Londoners who flushed the source of their foul air into the Thames and made the cholera epidemic worse.

Righting this ship cannot be a quiet process — updating a web page here, saying the right thing there. The proclamations that we now know are wrong were so persistent and so loud for so long.

It’s true that as the evidence piled on, there was genuine progress and improvement, especially as of late. Even before the change in language last week, for example, the W.H.O. published helpful guides on ventilation, first in July and updating it in March. Recently, though the organization’s documents have lagged, more of its officials have started giving advice compatible with aerosol transmission, emphasizing things like close mask fit — which matters little for droplet transmission — and ventilation — which matters even less. All this is good, but nowhere near enough to change the regulations and policy bundles that had already been put in place around the world.

And the progress we’ve made might lead to an overhaul in our understanding of many other transmissible respiratory diseases that take a terrible toll around the world each year and could easily cause other pandemics.

So big proclamations require probably even bigger proclamations to correct, or the information void, unnecessary fears and misinformation will persist, damaging the W.H.O. now and in the future.

Scientists have responded. In just the past few weeks, there has been a flood of articles published about airborne transmission in leading medical journals. Dr. Marr and other scientists told me the situation was very difficult until recently, as the droplet dogma reigned. I co-wrote one of those papers, published in The Lancet last month, arguing that aerosols may be the predominant mode of transmission for SARS-CoV-2, a step farther.

I’ve seen our paper used in India to try to reason through aerosol transmission and the necessary mitigations. I’ve heard of people in India closing their windows after hearing that the virus is airborne, likely because they were not being told how to respond. Plus, there are important questions for what this means for higher-risk settings, like medical facilities.

The W.H.O. needs to address these fears and concerns, treating it as a matter of profound change, so other public health agencies and governments, as well as ordinary people, can better adjust.

The past year has revealed how crucial the agency is, despite being hampered by chronic underfunding, lack of independence and attempts to turn it into a political football by big powers. Like other public health organizations, many of its dedicated staff members work tirelessly under difficult conditions to safeguard health around the world. Maintaining its credibility is essential not just for the rest of this terrible pandemic but in the future.

It needs to begin a campaign proportional to the importance of all this, announcing, “We’ve learned more, and here’s what’s changed, and here’s how we can make sure everyone understands how important this is.” That’s what credible leadership looks like. Otherwise, if a web page is updated in the forest without the requisite fanfare, how will it matter?

If world leaders don’t act now, the end of the Covid pandemic may come with a horrible form of herd immunity, as more transmissible variants that are taking hold around the world kill millions.

There’s troubling new evidence that the B.1.617.2 variant, first identified in India, could be far more transmissible than even the B.1.1.7 variant, first identified in Britain, which contributed to some of the deadliest surges around the world.

In countries with widespread vaccination, like the United States and Britain, we can expect that Covid cases, hospitalizations and deaths will continue to decline or stay low, especially because lab tests and real world experience show that vaccines appear to defend recipients well against the severe effects of both variants.

For much of the rest of the world, though, this even more transmissible new variant could be catastrophic.

The evidence is not yet conclusive because the data is preliminary and figuring out if a variant is more transmissible isn’t easy. It could be spreading rapidly in an area because of chance. Maybe it got there before other variants and found a susceptible population, or got lucky and seeded a mega-cluster. If a variant is seen more frequently in a country’s genomic databases it could just be because travelers, who are often tested more routinely, are bringing it in from another country where it is already dominant.

One key measure that’s been used in previous outbreaks to figure out if a variant was more transmissible was to look at “secondary attack rates” in non-travel settings — how many people who come into close contact with an infected person get the virus themselves. The greater the number of these contacts getting the virus, on average, the greater the likelihood that a variant’s transmissibility is greater.

Data on secondary attack rates released on Saturday by a British public health agency similar to our own Centers for Disease Control and Prevention suggests that this variant first seen in India may be substantially more transmissible among close contacts than even the already highly transmissible B.1.1.7. A report published by the same agency on Thursday further supports last week’s findings. It was just such early data that raised alarms about B.1.1.7, with later information confirming those early fears.

Adam Kucharski, an epidemiologist with the London School of Hygiene and Tropical Medicine, also told me that the faster spread in areas of Britain with higher levels of the variant suggests it has higher transmissibility. This point seems to be backed up by the terrible outbreaks in India and neighboring Nepal, where it is also widespread. Given how limited genetic identification is in those countries, the data from Britain is particularly useful for assessing the risk.

A variant with higher transmissibility is a huge danger to people without immunity either from vaccination or prior infection, even if the variant is no more deadly than previous versions of the virus. Residents of countries like Taiwan and Vietnam that had almost completely kept out the pandemic, and countries like India and Nepal that had fared relatively well until recently, have fairly little immunity, and are largely unvaccinated. A more transmissible variant can burn through such an immunologically naïve population very fast.

Increased transmissibility is an exponential threat. If a virus that could previously infect three people on average can now infect four, it looks like a small increase. Yet if you start with just two infected people in both scenarios, in total, after 10 rounds of infection, the more transmissible variant would cause 2.8 million cases, as opposed to 177,000 for the less transmissible one.

Morally and practically, this emergency demands immediate action: widespread vaccination of those most vulnerable where the threat is greatest.

Waiving vaccine patents is fine, but unless it’s tied to a process that actually increases the supply of vaccines, it’s little more than expressing thoughts and prayers after a tragedy. Officials from all nations that produce vaccines need to gather for an emergency meeting immediately to decide how to commandeer whatever excess capacity they have to produce more, through whatever means necessary. Because of the threat of increased transmissibility, and since the evidence at hand indicates that all of the vaccines, even the Chinese and Russian versions, appear to be highly effective against severe disease or worse, the focus should be on manufacturing and distributing the highest number of doses possible as fast as possible.

If the choice is between no vaccine and any vaccine, the precedence should go to whatever can be manufactured fastest, regardless of patents, nation of origin, or countries prioritizing their allies or wannabe allies.

Vaccine supplies need to be diverted now to where the crisis is the worst, if necessary away from the wealthy countries that have purchased most of the supply. It is, of course, understandable that every nation wants to vaccinate its own first, but a country with high levels of vaccination, especially among its more vulnerable populations, can hold things off, especially if they also had big outbreaks before. In addition, excess stockpiles can go where they are needed without even slowing down existing vaccination programs.

Right now, Covax, the global alliance for vaccine equity, does not have enough vaccines to distribute, and what supply it does get is allocated according to national population, not the seriousness of outbreaks. This needs to change. Our fire department needs more water, and should direct it to where the fire is burning, not to every house on the street.

The responses could vary. The elderly and health care workers could be prioritized wherever a crisis is worst. Dose-sparing strategies could be applied — delaying boosters has been successful in Britain and Canada when they faced surges. The details can be decided by the global health authorities.

If there is a reasonable possibility that the world faces a mounting threat, it’s best to intervene as early and aggressively as possible, because even a few weeks of delay could make a huge difference. Waiting for definitive evidence of transmissibility could allow the variant to rampage — besides, if fears prove unfounded then the world would still be better off for being more vaccinated. And any solution, even if imperfect, is better than waiting for the perfect setup or the most conclusive evidence.

As Dr. Kucharski told me, it’s now entirely possible that most Covid deaths could occur after there are enough vaccines to protect those most at risk globally. Britain had more daily Covid-related deaths during the surge involving B.1.1.7 than in the first wave, when there was less understanding of how to treat the disease and far fewer therapeutics that later helped cut mortality rates. Even after the vaccination campaign began, B.1.1.7 kept spreading rapidly among the unvaccinated. A similar pattern was observed in much of Eastern Europe as well.

Even if it is determined the transmissibility of B.1.617.2 isn’t as bad as feared, the emergency is still there.

The kind of catastrophic outbreak like the one in India can cause many more needless deaths simply by overwhelming our resources. Already, there are reports that countries ranging from Nepal to the Philippines to South Africa to Nigeria may face supplemental oxygen shortages of the kind seen in India. This pathogen has one fatality rate when oxygen is available as a therapy and one when it is not, and it would be an unspeakable tragedy to suffer the latter in the second year of the pandemic.

Like all pandemics, this one will end either with millions — maybe billions — being infected or being vaccinated. This time, world leaders have a choice, but little time to make that choice before it is made for them.

Correction: May 31, 2021

A previous version of this article referred imprecisely to the effects of a more transmissible variant. The calculations were for cases in the 10th round of infection, not total cases after 10 rounds of infection.

There were curious characteristics about the H1N1 influenza pandemic of 1977-78, which emerged from northeastern Asia and killed an estimated 700,000 people around the world. For one, it almost exclusively affected people in their mid-20s or younger. Scientists discovered another oddity that could explain the first: It was virtually identical to a strain that circulated in the 1950s. People born before that had immunity that protected them, and younger people didn’t.

But how on earth had it remained so steady genetically, since viruses continually mutate? Scientists guessed that it had been frozen in a lab. It was often found to be sensitive to temperature, something expected for viruses used in vaccine research.

It was only in 2004 that a prominent virologist, Peter Palese, wrote that Chi-Ming Chu, a respected virologist and a former member of the Chinese Academy of Sciences, told him that “the introduction of this 1977 H1N1 virus” was indeed thought to be due to vaccine trials involving “the challenge of several thousand military recruits with live H1N1 virus.”

For the first time, science itself seemed to have caused a pandemic while trying to prepare for it.

Now, for the second time in 50 years, there are questions about whether we are dealing with a pandemic caused by scientific research.

While the Chinese government’s obstruction may keep us from knowing for sure whether the virus, SARS-CoV-2, came from the wild directly or through a lab in Wuhan or if genetic experimentation was involved, what we know already is troubling.

Years of research on the dangers of coronaviruses, and the broader history of lab accidents and errors around the world, provided scientists with plenty of reasons to proceed with caution as they investigated this class of pathogens. But troubling safety practices persisted.

Worse, researchers’ success at uncovering new threats did not always translate into preparedness.

Even if the coronavirus jumped from animal to human without the involvement of research activities, the groundwork for a potential disaster had been laid for years, and learning its lessons is essential to preventing others.

*

Until the SARS outbreak, coronaviruses were considered fairly benign, causing only minor to moderate colds. Even five months after SARS emerged in southern China in November 2002, the Chinese government was covering up details about its threat, while the disease was spreading to other countries. By summer 2003, it had been contained, but not before infecting over 8,000 people and killing 774. Officials were able to suppress SARS because infected people spread it when visibly sick, making it easier to identify and isolate people. But it was a close call, and that roughly 10 percent case fatality rate raised alarms. Preventing the next coronavirus pandemic became a scientific priority.

By 2005, researchers — including Dr. Shi Zhengli, a virologist at the Wuhan Institute of Virology — had identified horseshoe bats as the likely primary host animal from which SARS had emerged. In the years that followed, scientists pursued bat coronaviruses in the field and studied them in the lab.

It is often assumed that SARS was spread to humans by palm civets, an adorable small mammal sometimes sold at wildlife markets, though by 2008, it was suspected that bat coronaviruses could directly infect human lung cells without needing an intermediary animal. By 2013, Dr. Shi’s lab experiments showed this could happen.

Still, scientists sometimes worked with bats, bat samples and bat viruses under conditions that have since raised eyebrows.

*

It is in the nature of viruses to continually mutate, with random accidents altering, adding or removing parts of its genome or bits of genetic code being exchanged with other viruses — recombination. This constant trial and error enables the emergence of features that can allow viruses to infect a new species.

In order to anticipate these jumps, humans have tried to steer this process. In what is sometimes called gain-of-function research, they genetically manipulate viruses to see how they can become more dangerous.

In an article in Nature Medicine in 2015, researchers from two of the major coronavirus laboratories in the world — Dr. Shi; Ralph Baric, a professor at the University of North Carolina at Chapel Hill; and others — wrote that they had bioengineered a coronavirus. The work was carried out in Dr. Baric’s laboratory at U.N.C. They took a spike protein, the “key” that coronaviruses use to unlock and infect cells, from a horseshoe bat virus and combined it with a human SARS virus adapted for mice. They reported that this “chimeric” virus could infect human cells, suggesting some bat viruses may be “capable of infecting humans without mutation or adaptation.” This was the second time since Dr. Shi’s 2013 experiments that a SARS-like bat coronavirus showed the ability in the lab to directly infect human airway cells.

This kind of genetic manipulation had already raised concerns, especially after laboratories in the Netherlands and the United States announced in 2011 that they had created strains of flu viruses using genetic material from the H5N1 influenza A virus, which is very deadly but generally can’t yet spread among people. These new strains could spread by air among ferrets, which have humanlike lungs. The uproar had been immediate.

In defense of the 2015 coronavirus experiment by Dr. Shi and her colleagues, Peter Daszak, whose organization, EcoHealth Alliance, has worked closely with her and has been granted tens of millions of dollars in the last decade from the U.S. government, said the findings would allow scientists to focus on the greatest risk because it would “move this virus from a candidate emerging pathogen to a clear and present danger.”

Others were more worried. “If the virus escaped, nobody could predict the trajectory,” said Simon Wain-Hobson, a virologist at the Pasteur Institute in Paris.

Recent history provided plenty of reason for such concern.

Nearly every SARS case since the original epidemic has been due to lab leaks — six incidents in three countries, including twice in a single month from a lab in Beijing. In one instance, the mother of a lab worker died.

In 2007, foot-and-mouth disease, which can devastate livestock and caused a massive crisis in Britain in 2001, escaped from a drainage pipe leak at an English lab with the highest biosafety rating, BSL-4.

Even the last known person who died of smallpox was someone infected because of a lab incident in Britain in 1978.

In its first published survey of the reporting systems in American labs working with dangerous pathogens, the Centers for Disease Control and Prevention in 2012 reported 11 laboratory-acquired infections across six years, often in BSL-3 labs — the category of safety reserved for pathogens like tuberculosis. In each instance, the exposure was not realized or reported until lab workers became infected.

In January 2014, the C.D.C. contaminated a benign flu virus sample with deadly A(H5N1) but didn’t discover the danger until months later. And in June 2014, it mistakenly sent improperly deactivated anthrax bacteria to labs, potentially exposing at least 62 C.D.C. employees who worked with the samples without protective gear. One month later, vials of live smallpox virus were found in a storage room at the National Institutes of Health.

In October 2014, after that string of high-profile incidents, the United States paused its funding of new gain-of-function research, with few exceptions. The moratorium was lifted in 2017.

Far more serious questions about scientific safety would soon arise.

*

On Dec. 30, 2019, a public email list run by the International Society for Infectious Diseases warned that an “unexplained pneumonia” had appeared in Wuhan, China, and reports connected the first cases to the city’s Huanan seafood market. On Jan. 10, 2020, a Chinese scientist posted the genome of the virus — soon to be named SARS-CoV-2 — on an open internet depository, confirming that it was a coronavirus. The Chinese government denied that the virus was spreading among humans until Jan. 19, 2020; three days later, it announced a complete lockdown of Wuhan, a city of 11 million people.

About a week after the lockdown, Chinese scientists published a paper in The Lancet medical journal that identified bats as the likely source of the virus. The authors noted that the outbreak happened during local bat hibernation season and “no bats were sold or found at the Huanan seafood market,” so they reasoned that it may have been transmitted by an intermediary animal.

Outbreaks can occur far from their source. The 2002 SARS outbreak started in Guangdong, about a thousand kilometers from the caves in Yunnan with the horseshoe bats from which SARS is believed to have emerged. Masked palm civets, farmed and traded across China, often in cramped, unsanitary conditions making them prone to outbreaks, were cited as the vehicle that SARS probably used to travel from Yunnan to Guangdong. Since SARS-CoV-2 was first detected at a market where live wild animals may have been sold, the wildlife trade was immediately suspected.

Social media users in China were among the first to be more skeptical. Did the spread of a disease from bats just happen to start in Wuhan, home to the Wuhan Institute of Virology, one of the few top bat coronavirus research facilities in the world? And what about the Wuhan Centers for Disease Control and Prevention, which also carries out bat research, a few hundred yards from the seafood market?

On Feb. 19, 2020, 27 prominent scientists published an open letter in The Lancet. They decried “conspiracy theories suggesting that Covid-19 does not have a natural origin.”

As we consider its origin, the question is not so much whether SARS-CoV-2 could have gotten out of a lab — accidents happen — but whether it could have gotten in and how it would have been handled there.

Shortly after Wuhan was locked down in January 2020, it became apparent that SARS-CoV-2 was related to a virus that scientists had been aware of for years.

On Feb. 3, 2020, Dr. Shi and co-authors announced in Nature that they had found a virus in their database, RaTG13, whose genome sequence was 96.2 percent identical to SARS-CoV-2 and was previously detected in horseshoe bats of Yunnan.

Suspicious internet sleuths combed through genomic databases and found that RaTG13 was an exact match for a bat coronavirus called 4991 retrieved from a cave implicated in an unexplained outbreak of pneumonia in 2012 among miners who collected bat guano from a mine in Yunnan. Three of the six miners died.

In May 2020, a former science teacher from India, with the Twitter pseudonym TheSeeker268, found a 2013 master’s thesis, as well as a 2016 Ph.D. thesis, supervised by George Fu Gao, the current director of the Chinese Center for Disease Control and Prevention. The master’s thesis hypothesized that the miners’ illness was caused by direct transmission of a SARS-like coronavirus from a horseshoe bat. The Ph.D. thesis was more cautious but still called the outbreak “notable.” It also revealed that a team from the Wuhan Institute of Virology had collected bat samples from the cave. The dissertation noted that all four of the miners who were tested for SARS antibodies had them in their blood a few weeks after they became ill.

None of those crucial facts — the name change or the link to the previous fatal outbreak possibly from a SARS-like coronavirus — were mentioned in the original paper about RaTG13. In an interview published in March 2020, Dr. Shi said fungus was the pathogen that had sickened the miners, not a coronavirus.

The questions persisted.

Last July, Dr. Shi confirmed that RaTG13 was indeed 4991 renamed. In November 2020, her paper in Nature was finally updated, additionally acknowledging what sleuths had also uncovered: Her team genetically sequenced RaTG13 in 2018. (The possible bat coronavirus link to the miner deaths was still not acknowledged.)

The less than forthcoming disclosure — a virus with two names, the connection to a deadly outbreak, shifting diseases and inconsistent stories — fueled suspicions.

Some speculated whether RaTG13 had been subjected to gain-of-function-type manipulation to create SARS-CoV-2. But RaTG13 is more like a distant cousin of SARS-CoV-2, meaning it is unlikely to have produced SARS-CoV-2 as an offspring, either through recent evolution in the wild or manipulation in the lab.

Even if RaTG13 had no role in the Covid-19 outbreak, questions were raised about why Dr. Shi and others seemed so unforthcoming about it. Then more questions were raised.

For example, the same group of internet sleuths that linked RaTG13 to the mine also uncovered that a genomic database maintained by the Wuhan Institute of Virology, with information about thousands of bat samples and at least 500 recently discovered bat coronaviruses, went offline in September 2019. The official explanation — that it was taken offline because it had been subjected to hacking — doesn’t explain why it was never securely shared some other way with responsible independent researchers.

Such gaps made it harder to rule out worrying scenarios. If there had been a lab accident involving SARS-CoV-2 or a virus like it that had been collected in the wild or experimented on in the lab, the database might have been taken down so there would be less evidence that might help others connect the dots. Officials might have investigated possible lab cases and prematurely believed it was in the clear. However, cases can be asymptomatic, and they might have missed the one that started a transmission chain and allowed the virus to circulate quietly until a superspreader event in December.

The secrecy and the cover-ups have led to some frantic theories — for example, that the virus leaked from a bioweapons lab, which makes little sense, since, for one thing, bioweapons usually involve more lethal pathogens with a known cure or vaccine, to protect those who employ them.

But much more mundane threats lurked.

*

Dr. Shi’s scientific work was dependent on collecting and analyzing hundreds of bat samples. And it was her work that showed the dangers associated with this endeavor. The 2013 paper by Dr. Shi, Dr. Daszak and others demonstrated that a live bat coronavirus from a Yunnan sample could bind to human lung cell receptors, showing that “intermediate hosts may not be necessary for direct human infection.” That controversial 2015 experiment co-authored by a group of researchers that included Dr. Baric and Dr. Shi was carried out after they had found another bat coronavirus they suspected could infect humans, but it was difficult to cultivate. They then created that chimeric one using its spike. They showed that it, too, could infect human airway cells directly.

In October 2015, Dr. Shi’s lab sampled over 200 people living within a few miles of two Yunnan bat caves and found that six tested positive for bat coronavirus antibodies, indicating past infection. All six reported having seen bats and only 20 people in total had reported seeing bats flying close to their homes, suggesting exposure created a great risk of infection.

The research practices, however, may not have always incorporated these lessons.

While a 2017 Chinese article noted the caution of the Wuhan Institute of Virology’s workers and showed them hooded and some wearing N95 masks, later that year a Chinese state-TV story about Dr. Shi’s studies showed researchers handling bats or bat feces with their bare hands or with exposed arms. A person on her team likened a bat bite to “being jabbed with a needle.”

In a 2018 blog post that was later removed, Dr. Shi said that the job was “not as dangerous” as everyone thought. “The chance of directly infecting humans is very small,” she wrote. “In most cases only ordinary protection will be taken,” unless a bat was known to carry a virus that might infect humans. She repeated something similar in a 2018 TED Talk-style video, according to The Washington Post, noting that “simpler protection” — illustrated with slides of unmasked or surgically masked colleagues with bare hands — was appropriate because it was believed that bat pathogens usually required an intermediate host.

Dr. Shi said that all the research at the institute is done in strict accordance with biosafety standards and the lab is tested annually by a third-party institution.

The Wuhan C.D.C. also reportedly conducts research on bat-borne viruses.

One of its staff members, Tian Junhua, has developed a reputation for adventurous scientific discovery. A 2013 paper notes his team caught 155 bats in Hubei Province. The Washington Post reported that in a video released on Dec. 10, 2019, he boasted about “having visited dozens of bat caves and studied 300 types of virus vectors.” Previously, he also talked about having made mistakes in the field, like forgetting personal protective equipment and being splashed with bat urine or accidentally getting bat blood on his skin, according to The Post. And yet the World Health Organization reported that the agency denied ever storing or working with bat viruses in the lab before the pandemic.

This March the W.H.O. reported that the Wuhan C.D.C. lab “moved on 2nd December 2019 to a new location near the Huanan market.” The W.H.O. report said there were “no disruptions or incidents” during the move. Given the Chinese government’s lack of candor, that raises suspicions that lab samples, if not bats themselves, were being hauled around near the market at the time of the outbreak.

Many of these research practices weren’t deviations from international norms. A bat field researcher in the United States told me she now always wears a respirator in bat caves but that wasn’t standard practice before.

It isn’t a wild idea to suggest that field research risks setting off an outbreak. Dr. Linfa Wang, a Chinese-Australian virologist based in Singapore who frequently works with Dr. Shi and pioneered the hypothesis that bats were behind the 2003 SARS epidemic, told Nature there is a small chance that this pandemic was seeded by a researcher inadvertently getting infected by an unknown virus while collecting bat samples in a cave.

Bats could create further risks if housed in laboratories, like the risk posed by the sale of wildlife in urban markets.

On Dec. 10, Peter Daszak, who organized The Lancet letter denouncing the questioning of Covid-19’s natural origins and was announced as a member of the W.H.O. origins investigation committee last fall, insisted it was a conspiracy theory to suggest that there were live bats in labs he had collaborated with for 15 years. “That’s not how this science works,” he wrote in a tweet he later deleted. “We collect bat samples, send them to the lab. We RELEASE bats where we catch them!”

But evidence to the contrary has accumulated. An assistant researcher told a reporter that Dr. Shi took on the role of feeding the bats when students were away. Another news report in 2018 said a team led by one of her doctoral trainees “collected a full rack of swabs and bagged a dozen live bats for further testing back at the lab.” The Chinese Academy of Sciences website has listed the Wuhan institute as having at least a dozen cages for bats, and in 2018 the institute applied for a patent for a bat cage. Dr. Shi has talked about monitoring antibodies in bats over time — which would not be done in a cave. Recently, another video surfaced that reportedly showed live bats in the institute.

Just a few weeks ago, Dr. Daszak changed his claims. “I wouldn’t be surprised if,” he said, “like many other virology labs, they were trying to set up a bat colony.”

Meanwhile, no intermediary animal has yet been found, despite testing thousands of animals around Wuhan. Last month a former commissioner of the Food and Drug Administration, Scott Gottlieb, said this failure added to the evidence of a lab leak, although Dr. Daszak suggested that investigators look further, at wildlife farms in southern China.

But if bat-to-human transmission is how the spillover happened, no intermediary animal is necessary, since it could have been any interaction with a bat — by a villager or a field researcher.

Despite widespread assertions that bat viruses need an intermediary animal to spread to humans, research is not even settled on whether the palm civet spread SARS to humans from bats. We do know that palm civets amplified the outbreak once SARS arrived in the Guangdong market and that back-and-forth transmission between humans and civets was possible. However, the only widespread infected civet populations that researchers found were those at urban markets and sometimes at farms — where people are — and not in the wild. We know we can infect animals. Last year Denmark had to kill 17 million minks after they caught SARS-CoV-2 from people. It’s possible that humans were the initial intermediary animal for civets and that the cute little creatures were framed.

*

Other sources of risk were the lab activities themselves.

There has been a lot of speculation that SARS-CoV-2 was the result of genetic engineering. This hypothesis cannot be ruled out based on genomic analysis alone, and suspicion has grown because of the opaque response by Chinese authorities.

They have refused to share direct records from the lab. Dr. Shi echoed this stance in May when a group of scientists, including her co-author Dr. Baric, pushed for broader transparency. “It’s definitely not acceptable,” she emailed a reporter in response to the group’s request to see her lab’s records.

Meanwhile, throughout December 2019, Wuhan doctors suspected that a SARS-like virus was on the loose, and the local government arrested whistle-blowers, including at least one health care worker. The cover-up by Communist Party officials continued until the prominent SARS scientist Zhong Nanshan traveled to Wuhan on Jan. 18 and raised the alarm.

That said, circumstantial evidence casts some doubt on the claim that SARS-CoV-2 was bioengineered.

For instance, aspects of the virus that have made some suspect it was bioengineered could also be evidence that the virus evolved naturally. A lot of attention has been drawn to an unusual feature on its spike protein called a furin cleavage site, with which the virus can better infect a human cell. It’s one of several odd features of SARS-CoV-2 that are weird enough that even virologists who greatly doubt lab involvement told me they were shocked to see it. In fact, even beyond the furin cleavage site, SARS-CoV-2 was a virus that scientists had never seen before. Evolution can be a random accumulation of weird, novel features. For the research on viruses that scientists like Dr. Shi do for high-level scientific publications, such a combination would be incongruous. Their work usually involves examining or changing one element of a virus at a time to find out what each element does and can be made to do. If your computer conked out, for instance, you wouldn’t see what’s wrong by simultaneously changing the power source, the cable and the electrical outlet. You’d test each one individually. Having a variety of unusual elements leads to hard-to-assess results, not a paper in Nature.

But even if we put aside directed engineering, regular lab work at the Wuhan labs has raised concerns.

In 2016 the Wuhan institute reported experimenting on a live bat coronavirus that could infect human cells in a BSL-2 lab — a biosafety level that has been compared with that of a dentist’s office. Protective gear other than gloves and lab coats is usually optional at this level, and there’s often no airflow control sealing ventilation between the work area and the rest of the building. Michael Lin, an associate professor of neurobiology and bioengineering at Stanford, told me it was “an actual scandal, recorded in print,” that a SARS-like virus capable of replicating in human cells was worked on under such low safety conditions.

Just trying to culture bat viruses in the lab can create risks that the scientists may not even be aware of. While trying and failing to cultivate one strain, they might inadvertently culture another one they don’t even know about. It’s even possible, Dr. Lin told me, that viruses can coexist in a single sample and quietly recombine, giving rise to something novel but undetected. Under BSL-2 conditions or even sloppy BSL-3 conditions, researchers could get exposed to a pathogen they didn’t know existed.

Several scientists who signed The Lancet letter denouncing the consideration of anything but natural origins have since said they are more open to lab involvement. One, Bernard Roizman, an emeritus virologist at the University of Chicago with four honorary professorships from Chinese universities, said he was leaning toward believing there was a lab accident.

“I’m convinced that what happened is that the virus was brought to a lab, they started to work with it,” he told The Wall Street Journal, “and some sloppy individual brought it out.” He added, “They can’t admit they did something so stupid.”

Charles Calisher of Colorado State University, another signatory, recently told ABC News that “there is too much coincidence” to ignore the lab-leak theory and he now believes “it is more likely that it came out of that lab.”

Peter Palese, the virologist who wrote about the 1977 flu pandemic, said that “a lot of disturbing information has surfaced since The Lancet letter I signed” and that he wants an investigation to come up with answers.

Other scientists have also said they have changed their minds.

Ian Lipkin, the director of the Center for Infection and Immunity at Columbia University and a co-author of an influential article in Nature Medicine that argued in favor of a natural origin in March 2020, is also now more skeptical. “People should not be looking at bat viruses in BSL-2 labs,” he told the science reporter Donald G. McNeil Jr. last month. “My view has changed.”

Medical records of lab workers could help clarify such questions. Last July, Dr. Shi said “a possibility did not exist” that anyone associated with the institute may have gotten infected “while collecting, sampling or handling bats.” She added that it had recently tested all institute staff members and students for antibodies showing past infection by SARS-CoV-2 or SARS-related viruses and had found “zero infection” and insisted that she could rule out this possibility for all labs in Wuhan.

It’s hard to see how a careful scientist could dismiss even the slightest possibility for all labs, including those not her own. “Zero infection” would mean not a single case among the hundreds of people at the institute, even though a study found that 4.4 percent of the Wuhan population had been infected.

Later, the W.H.O. team asked for more information about the earliest Covid-19 cases in Wuhan, including anonymized but detailed patient data — something that should be standard in any outbreak origin investigation — and were denied access.

All this leaves a lot of possibilities open and a lot of confusion.

*

Since most pandemics have been due to zoonotic events, emerging from animals, is there reason to doubt lab involvement? Maybe if you look at all of human history. A better period of comparison is the time since the advent of molecular biology, when it became more likely for scientists to cause outbreaks. The 1977 pandemic was tied to research activities, while the other two pandemics that have occurred since then, AIDS and the H1N1 swine flu of 2009, were not.

Plus, once a rare event, like a pandemic, has happened, one has to consider all the potential paths to it. It’s like investigating a plane crash. Flying is usually very safe, but when a crash does happen, we don’t just say mechanical errors and pilot mistakes don’t usually lead to catastrophes and that terrorism is rare. Rather, we investigate all possible paths, including unusual ones, so we can figure out how to prevent similar events.

Perhaps the biggest question has been what to read into the location of the outbreak, a thousand miles from the closest known viral relatives yet close to a leading research institution.

Sometimes the curiosity around the location has been waved away with the explanation that labs are set up where viruses are. However, the Wuhan Institute of Virology has been where it is since 1956, doing research on agricultural and environmental microbiology under a different name. It was upgraded and began to focus on coronavirus research only after SARS. Wuhan is a metropolis with a larger population than New York City’s, not some rural outpost near bat caves. Dr. Shi said the December 2019 outbreak surprised her because she “never expected this kind of thing to happen in Wuhan, in central China.” When her lab needed a population with a lower likelihood of bat coronavirus exposure, they used Wuhan residents, noting that “inhabitants have a much lower likelihood of contact with bats due to its urban setting.”

Still, location itself is not proof, either. Plausible scenarios implicating research activities don’t rule out other options.

This week, Jesse Bloom, an associate professor at the Fred Hutchinson Cancer Research Center, told me that when he recovered and analyzed a set of partial early Wuhan genetic sequences that had been removed from a genomic archive, it supported “substantial existing evidence that SARS-CoV-2 was circulating in Wuhan prior to the seafood market outbreak.” Both the early reports from Chinese scientists and the more recent W.H.O. investigation this winter found many of the early cases had no connection to the seafood market, including the earliest acknowledged case so far, on Dec. 8, 2019. So the seafood market may not have been the original location of the outbreak.

It’s also plausible that an outbreak could have started someplace else and was detected in Wuhan simply because it was a big city. Testing blood banks from across China, especially in areas near wildlife farms and bat caves, would help, but with limited exceptions, the Chinese government has not carried out such research — or allowed the sharing of the results if it has.

People are understandably wary that these claims might demonize scientists from other countries, especially given the anti-Asian racism that has abounded. But why would perpetuating this state of events be to their benefit?

After a lab accident with anthrax bacteria in the Soviet Union in 1979 that killed dozens, leading Western scientists accepted the Soviet government’s excuses, which all turned out to be lies. That doesn’t help lead to better safety standards, including those that would benefit scientists in authoritarian countries.

But a better path forward is one of true global cooperation based on mutual benefit and reciprocity. Despite the current dissembling, we should assume that the Chinese government also doesn’t want to go through this again — especially given that SARS, too, started there.

This means putting the public interest before personal ambitions and acknowledging that despite the wonders of its power, biomedical research also holds dangers.

To do this, government officials and scientists need to look at the big picture: Seek comity and truth instead of just avoiding embarrassment. Develop a framework that goes beyond blaming China, since the issues raised are truly global. And realize that the next big thing can simply mean taking great care with a lot of small details.

Globally, the Covid-19 pandemic is a threat because of scarcity of vaccines, with the highly transmissible Delta variant threatening millions around the world who can’t get vaccinated.

In the United States, the threat is dysfunction, with unwanted vaccines ready to expire on the shelves as desperate people around the world die for lack of them.

“This is becoming a pandemic of the unvaccinated,” Dr. Rochelle Walensky, the director of the Centers for Disease Control and Prevention, recently said, as the data shows that almost everyone who died from Covid-19 recently was unvaccinated.

Certainly the severe consequences will fall mostly on the unvaccinated. But the dysfunction affects all Americans.

To start with, not everyone is unvaccinated by choice. Children under 12 are not eligible to get the shots — that’s about 50 million young people. Plus, the immunocompromised may not respond as well to vaccines, which is at least about five million more people.

Then there’s the risk that, especially over time, the elderly, whose immune systems are not as robust, may lose some of their vaccine protection — as occurs with other illnesses, and as we’ve seen with Covid, to some extent, in Britain and Israel.

Finally, initial data from Britain suggests that Delta may lead to more-severe cases than previous variants. That question remains unsettled, but the possibility greatly increases the urgency for a powerful public health response.

In addition, Delta seems to be able to evade some immunity, so, compared to earlier variants, vaccinated or previously infected people are more likely to have infections break through their immunity, even ones leading to mild or moderate symptoms, which means they may be more likely to transmit onward, too. Thus even the vaccinated may pose a danger to unvaccinated children or vulnerable co-workers.

Our vulnerability isn’t small, even among those eligible for vaccines. About 34 percent of Americans over the age of 12 and about 44 percent of the entire U.S. population haven’t been vaccinated at all. Some of those may have had Covid, so they have some level of protection, but that is not as good as being fully vaccinated, especially against Delta.

Even an optimistic projection leaves tens of millions of unvaccinated people exposed to higher risks of hospitalization and death from Covid.

Already, we are seeing rises in hospitalizations and deaths, almost entirely among the unvaccinated, and the Delta surge has just begun. Based on what we’ve seen elsewhere, we can expect it to rapidly accelerate and wash over the country fairly quickly.

The most important thing we can do now is to increase vaccination coverage.

One important step would be to implement and broaden vaccine mandates. There’s plenty of precedent for mandating vaccines in health care, the military and schools, so it wouldn’t be some novel step to do so for one of the safest vaccines we’ve ever had.

Some large employers already mandate flu vaccines. Kentucky legally requires everyone working at a long-term care facility to be vaccinated against the flu and pneumococcal disease, unless they have a medical or religious exemption. Mandates for Covid vaccines, too, should be issued, especially for people who work with high risk or vulnerable populations — children, the elderly, the incarcerated and those in medical settings — and possibly for employees in workplaces where large numbers of people congregate indoors. Exemptions, too, should be re-examined so that they aren’t overly broad.

A staggering 40 percent of workers at nursing homes and other long-term care facilities remain unvaccinated. This is terrible, considering that the elderly, even if vaccinated, would be expected to have more breakthrough infections.

People may have a right to take their chances with an infection but not to risk transmitting the virus to vulnerable others.

An obstacle to extending mandates is that the Food and Drug Administration has approved the vaccines only for emergency use and has yet to fully approve them. Typically a company can apply for and receive full approval after six months of compiling safety data. Pfizer submitted its application in early May, and the F.D.A. just accepted it, so two months have passed. Moderna applied in early June and is still waiting for acceptance of its application.

The F.D.A. has until January to decide whether to formally approve Pfizer’s vaccine. The acting F.D.A. commissioner, Janet Woodcock, said she hopes it will get done more quickly. Let’s hope so. It would be a grave mistake for the agency to take another six months.

The lack of formal approval has allowed some anti-vaxxers to claim the vaccines are experimental. Polls show that a lack of trust in the vaccine is a rationale for some of the vaccine hesitancy. And without full approval, it might be easier to fight vaccine mandates, both legally and politically.

Lack of full approval also helps feed a misunderstanding. Adverse reactions to some drugs can indeed occur long after we start taking them, especially after long-term use. However, vaccination is a one-time event (or two, if you count the booster). Immediate side effects, like allergic reactions, would occur fairly soon after inoculation. That’s why people are asked to spend 15 minutes in a waiting room postvaccination or 30 minutes if they’ve had a history of allergic reactions. There are also questions about immune system reactions that can take longer to appear. However, as the immunologist Andrew Croxford explained, medical experts have learned to expect such problems within the first few weeks and months after vaccination. Regulators require six months of safety data, not more. They have that now.

The F.D.A. said it needs to complete “a high-quality review and evaluation,” and it does. But it has the data to do that quickly. The safety data on these vaccines is comparable to that of many fully approved vaccines. This is not to mention the evidence from hundreds of millions of doses administered in the United States alone.

Given the severe threat the Delta variant poses to the unvaccinated, the risk-benefit calculation is especially stark, and the need for full approval is even more urgent.

Some might feel less concern for the unvaccinated, viewing them as hard-core anti-vaxxers or eager consumers of extremist propaganda on social media or Fox News. It’s true that attitudes toward vaccination have become affected by our political polarization. At a major conservative event recently, a mention of low vaccination rates was met with cheers. As horrifying as this was, we still have to try to reach this population.

In general, it is easier to convince people if the message comes from a trusted intermediary, and the current polarization makes this even more important. For conservatives, one good place to look might be Republican governors who face more direct accountability to the affected people.

The effort isn’t necessarily hopeless. Almost 90 percent of Americans 65 or older have received a vaccine. Since many of them are Republicans and heavy consumers of misinformation, this demonstrates that the harsh reality of the risk calculation can, under the right circumstances, cut through the misinformation. Delta’s terrible effects may provide that tragic tipping point for more.

And the vaccination gap runs through ideological, racial and social divisions. Black and Hispanic people are still less likely to be vaccinated, despite having suffered disproportionately throughout the pandemic. This is probably due to lack of access, especially early on, when getting an appointment could be complicated; a fear of side effects, especially if working without paid sick leave; and historic distrust of the medical system, in which, studies show, they continue to face discrimination.

Reaching underrepresented and historically mistrustful communities requires resources and, crucially, local action: a public health system in which trusted local leaders can work with the medical system to take both the vaccine and the persuasion to the people. In New York City, about 20,000 people who were offered a free weekly MetroCard or train ticket got vaccinated at pop-up clinics in subway and train stations. We can try versions of this in neighborhoods, shopping malls, supermarkets and workplaces as much as possible. It’s fine to offer incentives, but it might help just to have health care workers present to answer questions.

The federal government should deploy epidemiologists, pollsters and ethnographers to figure out what arguments, incentives and approaches work best now and even carry out local experiments.

Meanwhile, schools will open in the coming weeks, and yet so much remains unclear and confusing about protecting younger children until they can get vaccinated. Even if they seem to be largely spared from severe outcomes of Covid-19, they are not invulnerable. Plus, they can bring the infection home.

C.D.C. guidelines make clear that good ventilation is essential for lowering the risk of airborne transmission. The federal government allocated more than $120 billion for K-12 schools in the latest relief package for improving ventilation and other mitigations, but rules for using these funds are flexible, and local implementation remains haphazard.

Parents need concrete information. Can windows in classrooms be opened, and will they be? Are HEPA filters, which cost a few hundred dollars each, being used in classrooms? Has the air-conditioning, if used, been adjusted to bring in outside air, and have its filters been upgraded to better catch pathogens? Have classroom ventilation rates been evaluated, for example by using carbon dioxide monitors?

Teachers and other staff members who work with kids younger than 12 should be subject to vaccine mandates, just like those who work with the elderly. We should also increase the use of rapid tests in schools, to try to catch outbreaks early.

Masking policies in schools remain confusing and muddled as well. The American Academy of Pediatrics is recommending that all children, even if vaccinated, wear masks in schools, which contradicts C.D.C. guidelines saying only the unvaccinated should wear them. Meanwhile, places like Texas and Iowa have barred schools from requiring masks.

How are parents supposed to negotiate a situation like this, especially if their children are under 12?

The problem with selective masking indoors, as the C.D.C. recommends now, is that it’s impossible to enforce. This could get especially tricky in schools, where peer pressure can play a large role.

A sensible school policy would be to mandate masks for all elementary school children, at least until a vaccine becomes available to them, and tie it to local infection and vaccination rates for those 12 and over. Parents who can vaccinate their kids are less at the mercy of everyone else’s choices, making mandates less crucial.

In the meantime, rules on indoor masking and crowding limits should be kept in place or restored until local vaccination rates rise to a set benchmark — say, at least 70 percent of those eligible — and case numbers remain low and steady. Too many people remain vulnerable, and in many places with low vaccination rates, hospitalizations are soaring. Even if we can’t expect people to keep masking indoors forever while others refuse vaccination, too many remain vulnerable now to throw up our hands. One lesson in this pandemic has been that waiting too long to respond or relaxing rules too early can create serious problems.

There is also a question about how likely the vaccinated are to transmit to others if they are infected with Delta. In May the C.D.C. stopped tracking breakthrough cases unless people became hospitalized. That was a mistake. Its explanation was that such cases were rare, and they may well have been when the decision was made. Delta is a reminder that the reason for surveillance is to be able to quickly notice changes in an outbreak.

Since Delta came onto the scene, I’ve acquired a pile of anecdotes from social media and a few examples from countries, like Singapore, that are doing a better job of tracing such cases that indicate vaccinated people are sometimes transmitting the virus. In comparison, I had hardly heard of any such examples from the Alpha variant, and studies did not indicate it could evade immunity in a substantive way. We also know that Delta has a higher viral load and infected people start shedding viral particles much earlier, so it makes sense why all this is happening.

Shouldn’t we have better evidence, though, so we can have better guidance, especially for those who live with unvaccinated children or have immunocompromised family members?

Given the ongoing nature of the threat and many remaining unknowns, the Occupational Safety and Health Administration should provide rules for rapid testing in workplaces, paid sick leave and, as appropriate, high-filtration masks and even ventilation standards.

As both vaccinations and infections increase, we get closer to the point that this virus joins the other four endemic coronaviruses that commonly circulate among humans. Though it is too early to predict what exactly will happen with this one, those mostly cause nuisance colds.

But the acute phase of this pandemic is far from over. Our moral obligation is to make vaccines available for as many people as possible and as fast as possible and to do everything we can to save the lives and health of people who don’t yet have that protection.

Back when a viral pandemic killing millions around the world was just the plot of a scary movie, the film “Contagion” was lauded for how accurately it depicted the way such an outbreak would occur.

On the science of viral contagion, it was quite sharp, clearly explaining things like R0 (the measure of how widely one infection could spread to others, on average).

Of the human dimension of contagion, it did not prove as prescient. In the movie, fearful nurses walked off the job at the start of the pandemic, which begins to end as soon as vaccines become available, with people lining up eagerly for their turn.

The opposite happened in real life. Despite enormous personal risk, almost all health care workers stayed on the job in the first months of the Covid pandemic. Despite vaccines being widely available since spring in the United States, tens of thousands of people are dying every month because they chose not to be inoculated.

The failure of the United States to vaccinate more people stands out, especially since we had every seeming advantage to get it done. As early as the end of April of this year, when vaccines were in dire short supply globally, almost every adult who wanted to get vaccinated against Covid-19 in the United States could do so, free of charge. By June, about 43 percent of the U.S. population had received two doses while that number was only about 6 percent in Canada and 3 percent in Japan.

Now, just a few months later, these countries, along with 44 others, have surpassed U.S. vaccination rates. And our failure shows: America continues to have among the highest deaths per capita from Covid.

Science’s ability to understand our cells and airways cannot save us if we don’t also understand our society and how we can be led astray.

There is a clear partisan divide over vaccination — Republicans are more likely to tell pollsters that they will not get vaccinated. Some Republican politicians and Fox News hosts have been pumping out anti-vaccine propaganda. The loud, ideological anti-vaxxers exist, and it’s not hard to understand the anger directed at them. All this may make it seem as if almost all the holdouts are conspiracy theorists and anti-science die-hards who think that Covid is a hoax, or that there is nothing we can do to reach more people.

Real-life evidence, what there is, demonstrates that there’s much more to it.

Almost 95 percent of those over 65 in the United States have received at least one dose. This is a remarkable number, given that polling has shown that this age group is prone to online misinformation, is heavily represented among Fox News viewers and is more likely to vote Republican. Clearly, misinformation is not destiny.

Second, reality has refuted dire predictions about how Americans would respond to vaccine mandates. In a poll in September, 72 percent of the unvaccinated said they would quit if forced to be vaccinated for work. There were news articles warning of mass resignations. When large employers, school districts, and hospital systems did finally mandate vaccines, people subject to mandates got vaccinated, overwhelmingly. After United Airlines mandated vaccines, there were only 232 holdouts among 67,000 employees. Among about 10,000 employees in state-operated health care facilities in North Carolina, only 16 were fired for noncompliance.

The remarkable success of vaccine mandates shows it is not firm ideological commitments that have kept everyone from getting vaccinated, and that the stubborn, unpersuadable holdouts may be much smaller than we imagine.

Let’s start with what we do know about the unvaccinated.

There has been strikingly little research on the sociology of the pandemic, even though billions of taxpayer dollars have been spent on vaccines. The assumption that some scientific breakthrough will swoop in to save the day is built too deeply into our national mythology — but as we’ve seen, again and again, it’s not true.

The research and data we do have show that significant portions of the unvaccinated public were confused and concerned, rather than absolutely opposed to vaccines.

Some key research on the unvaccinated comes from the Covid States Project, an academic consortium that managed to scrape together resources for regular polling. It categorizes them as “vaccine-willing” and “vaccine-resistant,” and finds the groups almost equal in numbers among the remaining unvaccinated. (David Lazer, one of the principal investigators of the Covid States Project, told me that the research was done before the mandates, and that the consortium has limited funding, so they can poll only so often.)

Furthermore, its research finds that the unvaccinated, overall, don’t have much trust in institutions and authorities, and even those they trust, they trust less: 71 percent of the vaccinated trust hospitals and doctors “a lot,” for example, while only 39 percent of the unvaccinated do.

Relentless propaganda against public health measures no doubt contributes to erosion of trust. However, that mistrust may also be fueled by the sorry state of health insurance in this country and the deep inequities in health care — at a minimum, this could make people more vulnerable to misinformation. Research on the unvaccinated by KFF from this September showed the most powerful predictor of who remained unvaccinated was not age, politics, race, income or location, but the lack of health insurance.

The Covid States team shared with me more than a thousand comments from unvaccinated people who were surveyed. Scrolling through them, I noticed a lot more fear than certainty. There was the very, very rare “it’s a hoax” and “it’s a gene therapy,” but most of it was a version of: I’m not sure it’s safe. Was it developed too fast? Do we know enough? There was also a lot of fear of side effects, worries about lack of Food and Drug Administration approval and about yet-undiscovered dangers.

Their surveys also show that only about 12 percent of the unvaccinated said they did not think they’d benefit from a vaccine: so, only about 4 percent of the national population.

In law, “dying declarations” are given special considerations because the prospect of death can help remove the motivation to deceive or to bluster. The testimony we’ve seen from unvaccinated people in their last days with Covid, sometimes voiced directly by them from their hospital beds, gets at some of the core truths of vaccine hesitancy. They are pictures of confusion, not conviction.

One woman who documented her final days on TikTok described being uncertain about side effects, being worried about lack of F.D.A. approval, and waiting to go with her family to get the shot — until it was too late.

Or consider Josie and Tom Burko, married parents who died from Covid within days of each other, leaving behind an 8-year-old daughter. They hadn’t taken the pandemic lightly. They were “100 percent pro-vaccination,” a close friend told The Oregonian afterward, but Josie reportedly had a heart murmur and chronic diabetes and worried about an adverse reaction. Tom reportedly had muscular atrophy, and similar worries. Afraid, they had not yet gotten vaccinated.

It’s easy to say that all these people should have been more informed or sought advice from a medical provider, except that many have no health care provider. As of 2015, one quarter of the population in the United States had no primary health care provider to turn to for trusted advice.

Along with the recognition of greater risk, access to regular health care may be an important explanation of why those over 65 are the most-vaccinated demographic in the country. They have Medicare. That might have increased their immunity against the Fox News scare stories.

One reason for low vaccination rates in rural areas may be that they are “health care and media” deserts, as a recent NBC report on the crises put it, with few reliable local news outlets and the “implosion of the rural health care system” — too few hospitals, doctors and nurses.

Plus, let’s face it, interacting with the medical system can be stress-inducing even for many of us with health insurance. Any worry about long-term side effects is worsened by a system in which even a minor illness can produce unpredictable and potentially huge expenses.

Then there is the health system’s long-documented mistreatment of Black people (and other minorities) in this country. Black people are less likely to be given pain medication or even treatment for life-threatening emergencies, for instance. I thought of those statistics while reading the poignant story of a Black physician who could not persuade her mother to get vaccinated because her mother’s previous interactions with the medical system included passing out after screaming in agony when a broken arm got manipulated and X-rayed without sufficient care for her pain.

While the racial gap in vaccination has improved over the last year — nonwhite people were more likely to express caution and a desire to wait and see rather than to be committed anti-vaxxers — it’s still there.

In New York, for example, only 42 percent of African Americans of all ages (and 49 percent among adults) are fully vaccinated — the lowest rate among all demographic groups tracked by the city.

This is another area in which the dominant image of the white, QAnon-spouting, Tucker Carlson-watching conspiracist anti-vaxxer dying to own the libs is so damaging. It can lead us to ignore the problem of racialized health inequities with deep historic roots but also ongoing repercussions, and prevent us from understanding that there are different kinds of vaccine hesitancy, which require different approaches.

Just ask Nicki Minaj.

About a month ago, the rap artist made headlines after tweeting that she was worried about vaccines because she had heard from her cousin that a friend of his had swollen testicles after being vaccinated. (Experts pointed out that, even if this had happened, it was most likely caused by a sexually transmitted disease.) She was justifiably denounced for spreading misinformation.

But something else that Minaj said caught my eye. She wrote that she hadn’t done “enough research” yet, but that people should keep safe “in the meantime” by wearing “the mask with 2 strings that grips your head & face. Not that loose one.”

“Wear a good mask while researching vaccines” is not the sentiment of a denier. She seemed genuinely concerned about Covid, even to the point that she seemed to understand that N95s, the high-quality masks that medical professionals wear, which have the “2 strings that grips your head & face,” were much safer.

Lazer said that the Covid States Project’s research showed that unvaccinated people who nonetheless wore masks were, indeed, more likely to be Black women. In contrast, those who were neither vaccinated nor masked were more likely to be Republicans, and more likely to be rural, less educated and white. (Among the vaccinated, Asian Americans were most likely to be still wearing masks.)

Lazer also highlighted an overlooked group with higher levels of vaccine hesitancy: young mothers. They were hesitant, both for themselves and their children, an alarming development especially if it starts affecting other childhood vaccinations. Similarly, from real-life data, we know that only a little more than one-third of pregnant women are vaccinated, which has led to many tragic stories of babies losing their mothers just as they are being whisked into the neonatal intensive care unit after an emergency cesarean section.

It may well be that some of the unvaccinated are a bit like cats stuck in a tree. They’ve made bad decisions earlier and now may be frozen, part in fear, and unable to admit their initial hesitancy wasn’t a good idea, so they may come back with a version of how they are just doing “more research.”

We know from research into human behavior but also just common sense that in such situations, face-saving can be crucial.

In fact, that’s exactly why the mandates may be working so well. If all the unvaccinated truly believed that vaccines were that dangerous, more of them would have quit. These mandates may be making it possible for those people previously frozen in fear to cross the line, but in a face-saving manner.

Research also shows that many of the unvaccinated have expressed concerns about long-term effects. Consider an information campaign geared toward explaining that unlike many drugs, for which adverse reactions can indeed take a long time to surface, adverse effects of vaccines generally occur within weeks or months, since they work differently, as the immunologist Andrew Croxford explained in The Boston Review. Medical professionals could be dispatched to vaccination clinics, workplaces and stores to get that point across. (Yes, medical professionals are overwhelmed, but the best way to reduce their burden is to vaccinate more people.) This would let some hesitant people feel like they had “done their research,” while interacting with a medical professional — the basis for more trust.

Finally, consider something hidden amid all the other dysfunction that plagues us: fear of needles.

Don’t roll your eyes. Prepandemic research suggests that fear of needles may affect up to 25 percent of adults and may lead up to 16 percent of adults to skip or delay vaccinations. For many, it’s not as simple as “suck it up”: It’s a condition that can lead to panic attacks and even fainting. During the pandemic, a study in Britain found that as many as one in four adults had injection phobia, and that those who did were twice as likely to be vaccine-hesitant. Research by Covid States shows that about 14 percent of the remaining unvaccinated mention fear of needles as a factor.

Countries with far higher rates of vaccination, Canada and Britain, have responded by mobilizing their greatest strength: a national health care system. Cities in Canada held clinics aimed especially at people with such anxiety, which included privacy rooms and other accommodations. Britain’s national health care system offers similar accommodations.

I’ve yet to find a systematic program in the United States addressing this fear. Worse, many of our public communications around the vaccines feature images of people getting jabbed with a needle, even though that can worsen anxiety.

In researching, I was inundated with stories from people who struggled with this fear and were often unable to find help. Some women said they were treated like drug seekers because they asked for a single anti-anxiety pill to get through a shot. (They also said their male family members and friends had an easier time.) It may seem hard to believe that people might risk their lives over seemingly small fears, but that’s exactly how people behave in many situations.

Of course, there are some people who it seems will never be persuaded. One strategy that has been shown to work is to highlight deceptive practices. In campaigns to keep teens from smoking, advertisements pointed out how the tobacco industry manipulated people. For Covid, the unvaccinated could be shown that they have been taken in by people who have misled them even while those people themselves got vaccinated.

Just recently, there was a brief glimpse at how Fox News actually looks behind the camera: Everyone in the office was wearing masks, even as the hosts have often talked about the alleged tyranny of it all. Stars like Tucker Carlson rant against vaccines, even as their network says that more than 90 percent of full-time employees have been vaccinated. Realizing that one may have been conned and manipulated by opportunists who do not practice what they preach may — just may — be the breakthrough for some.

Responding to our societal dysfunctions has been among the greatest challenges of this pandemic, especially since this includes a political and media establishment stirring up resentment and suspicion to hold on to power and attention in an increasingly unresponsive political system.

Anger — and even rage — at all this may be justified, but deploying only anger will not just obscure the steps we can and should try to take, it will play into the hands of those who’d like to reduce all this to a shouting match.

Instead, we need to develop a realistic, informed and deeply pragmatic approach to our shortcomings without ceding ground to the conspiracists, grifters and demagogues, and without overlooking the historic inequities in health care and weaknesses in our public health infrastructure. It’s not all fair, and it is not a Hollywood ending, but it’s how we can move forward.