The 3 Weeks That Changed Everything

The modern aviation system works. From the dawn of commercial aviation through the 1990s, 1,000 to 2,000 people would typically die each year in airline crashes. Today, the worldwide total is usually about one-10th that level. Last year, before the pandemic began, more than 25,000 commercial-airline flights took off each day from airports in the United States. Every one of them landed safely.

In these two fundamentally similar undertakings—managing the skies, containing disease outbreaks—the United States has set a global example of success in one and of failure in the other. It has among the fewest aviation-related fatalities in the world, despite having the largest number of flights. But with respect to the coronavirus pandemic, it has suffered by far the largest number of fatalities, about one-quarter of the global total, despite having less than one-20th of the world’s population.

James Fallows: Is this the worst year in modern American history?

Consider a thought experiment: What if the NTSB were brought in to look at the Trump administration’s handling of the pandemic? What would its investigation conclude? I’ll jump to the answer before laying out the background: This was a journey straight into a mountainside, with countless missed opportunities to turn away. A system was in place to save lives and contain disaster. The people in charge of the system could not be bothered to avoid the doomed course.

The organization below differs from that of a standard NTSB report, but it covers the key points. Timelines of aviation disasters typically start long before the passengers or even the flight crew knew anything was wrong, with problems in the design of the airplane, the procedures of the maintenance crew, the route, or the conditions into which the captain decided to fly. In the worst cases, those decisions doomed the flight even before it took off. My focus here is similarly on conditions and decisions that may have doomed the country even before the first COVID-19 death had been recorded on U.S. soil.

What happened once the disease began spreading in this country was a federal disaster in its own right: Katrina on a national scale, Chernobyl minus the radiation. It involved the failure to test; the failure to trace; the shortage of equipment; the dismissal of masks; the silencing or sidelining of professional scientists; the stream of conflicting, misleading, callous, and recklessly ignorant statements by those who did speak on the national government’s behalf. As late as February 26, Donald Trump notoriously said of the infection rate, “You have 15 people, and the 15 within a couple of days is going to be down close to zero.” What happened after that—when those 15 cases became 15,000, and then more than 2 million, en route to a total no one can foretell—will be a central part of the history of our times.

But what happened in the two months before Trump’s statement, when the United States still had a chance of containing the disease where it started or at least buffering its effects, is if anything worse.

1. The Flight Plan

The first thing an airplane crew needs to know is what it will be flying through. Thunderstorms? Turbulence? Dangerous or restricted airspace? The path of another airplane? And because takeoffs are optional but landings are mandatory, what can it expect at the end of the flight? Wind shear? An icy runway? The biggest single reason flying is so much safer now than it was even a quarter century ago is that flight crews, air traffic controllers, and the airline “dispatchers” who coordinate with pilots have so many precise tools with which to anticipate conditions and hazards, hours or days in advance.

And for the pandemic? Since at least the early years of the George W. Bush administration, the U.S. government has devoted scientific, military, and intelligence tools toward refining its understanding of what diseases might be emerging and where, and what might be done about them. One reason for this increased emphasis was the overall heightened (and sometimes overhyped) domestic-security awareness after the 9/11 terrorist attacks. Another was the series of anthrax attacks soon after 9/11, in which envelopes containing toxins were mailed to media and political figures on the East Coast.

But the most important event was the H5N1 “bird flu” outbreak, in 2005. It originated in Asia and was mainly confined there, as the SARS outbreak had been two years earlier. Bush-administration officials viewed H5N1 as an extremely close call. “We were deeply and genuinely concerned about the potential for human-to-human transmission of the bird flu,” John R. Allen, now president of the Brookings Institution, told me. Allen is a retired four-star Marine Corps general who during the Bush administration led an effort to assess the lessons of the H5N1 threat. “We realized that if it had spread worldwide, the numbers would have been enormous. So the national-security system was pulled right into the process of improving our awareness mechanisms, and developing a national pandemic strategy.”

The awareness mechanisms were a combination of military and civilian, structured and informal, open-source and classified, with a heavy emphasis on the then-infant tools of artificial intelligence, or AI. For instance, in Bush’s second term, an unclassified government-funded project called Global Argus—named for the all-seeing giant of Greek mythology—began sifting through news reports, radio broadcasts, road-traffic patterns, business data, and other kinds of open-source information for signs of abnormalities that, in turn, could be early indicators of disease.

Read: How the pandemic will end

“Epidemics cause social disruption,” the program’s creators explained, in a PowerPoint presentation from that era that I have seen. “Social disruption is a common feature that can be tracked and used in lieu of direct reporting of disease.” As a person involved in the process explained to me, the direct and indirect indicators of social disruption could range from reports of hospital-admission rates to unexplained changes in food prices. “Suddenly the price of chicken goes down in Thailand, and it gets your attention,” a man who worked on Global Argus told me. “It may mean that farmers have seen that their flock is sick, and they slaughter them all at once and send them to market.” This project aspired to process 250,000 bits of news per day, in nearly three dozen languages, for advance warning of anomalies that could possibly indicate disease.

Fifteen years later, in the age of autonomous tracking of everything, that scale might seem quaint. A current app like Waze, for instance, is at any given moment combining readings from tens of millions of cellphones to gauge current traffic conditions on roads across the country. But Bush-era programs like Global Argus predated the introduction of the very first iPhone, and naturally algorithmic powers have increased at least as fast as civilian technology has. These days, “AI has the capacity to ingest virtually all open-source media around the world, all day every day,” a person with direct experience in the process told me. “That can provide us with the early warning that would give the opportunity for the U.S. to move out quickly with civilian medical specialists, and military-logistics teams if necessary.” Then, with these early warnings in place, this person said, “we could focus our advanced national-intelligence assets there and be able to go at a moment’s notice. We would prepare to go to ground zero, help them understand what was happening, and do everything to keep the disease from spreading.”

What might such help entail? The metaphor several people used was of firefighters from Oregon and Idaho traveling to help contain a forest fire in California before it can spread. The U.S. has many times in the past 20 years deployed scientists, doctors, and logistical-support teams to Africa, Asia, and the Middle East during the disease outbreaks.

The U.S. military excels in logistics: mobile hospitals, teams of medics, food and water, masks and gowns. American scientists, at leading universities as well as the Centers for Disease Control and Prevention (despite its well-documented recent hollowing-out and politicization) are still seen as world leaders in many fields.

Read: ‘How could the CDC make that mistake?’

Shortly before Barack Obama left office, his administration’s Pandemic Prediction and Forecasting Science and Technology Working Group—yes, that was a thing—released a report reflecting the progress that had been made in applying remote-sensing and AI tools since the early days of Global Argus. The report is freely available online and notes pointedly that recent technological advances “provide opportunities to mitigate large-scale outbreaks by predicting more accurately when and where outbreaks are likely to occur, and how they will progress.”

James Giordano, a biosecurity expert at Georgetown University Medical Center who has been extensively involved in pandemic-response planning, told me this spring: “Absolutely nothing that has happened has been a surprise. We saw it coming. Not only did we see it, we ran the models and the gaming exercises. We had every bit of the structure in place. We’ve been talking about a biohazard risk like this for years. Anyone who says we did not see this coming has their head in the sand, or is lying through their teeth.”

The system the government set up was designed to warn not about improbable “black swan” events but rather about what are sometimes called “gray rhinos.” These are the large, obvious dangers that will sooner or later emerge but whose exact timing is unknown. Did the warning system work this time, providing advance notice of the coronavirus outbreak? According to everyone I spoke with, it certainly did. A fascinating unclassified timeline compiled by the Congressional Research Service offers a day-by-day and then hour-by-hour chronology of who knew what, and when, about developments in central China. By at least late December, signs were beginning to show something seriously amiss—despite foot-dragging, lies, and apparent cover-up on the Chinese side. A different kind of Chinese government might have done a different job, calling for help from the rest of the world and increasing the chances that the coronavirus remained a regional rather than global threat. But other U.S. leaders had dealt with foreign cover-ups, including by China in the early stages of the SARS outbreak in 2002. Washington knew enough, soon enough, in this case to act while there still was time.

Read: Coronavirus researchers tried to warn us

Through routine work or personal emails and other means of contact, U.S. and other international scientists began hearing from their Chinese colleagues very late last year about a new outbreak of what was initially referred to as pneumonia or flu. On December 31, the open-source platform ProMED—the Program for Monitoring Emerging Diseases—carried a translated “Chinese media report about the outbreak.” According to all of the intelligence-community veterans I spoke with, signals like this would certainly have been enough to alert U.S. officials to a significant development. “From these early indications, a pattern would have been discernible, and we would have slewed the rest of the system to find out more about it,” one of these people said. “Particularly since we’d know what to look for. If Martians were invading, we wouldn’t know what that would look like. But we have been down this road before, with MERS and SARS and Ebola, and we know the indications that are visible and detectable.”

With cues like these, the intelligence apparatus directed more attention at the area around the city of Wuhan. “China is a very hard target,” a man who recently worked in an intelligence organization told me. “We have to be very deliberate about what we focus on”—which in normal times would be military developments or suspected espionage threats. “The bottom line is that for a place like Wuhan, you really are going to rely on open-source or informal leads.” During the Obama administration, the U.S. had negotiated to have its observers stationed in many cities across China, through a program called Predict. But the Trump administration did not fill those positions, including in Wuhan. This meant that no one was on site to learn about, for instance, the unexplained closure on January 1 of the city’s main downtown Huanan Seafood Wholesale Market, a so-called wet market where wild animals, live or already killed, were on sale along with fish and domesticated animals. It was at this market that the first animal-to-human transfer of the virus is generally thought to have occurred, probably from a bat. But by that time, as Marisa Taylor of Reuters first reported, the Trump administration had removed dozens of CDC representatives in China.

Nonetheless, information came in. By the final days of December, and no later than January 1, a warning would have appeared in the President’s Daily Brief—the classified summary of international developments distilled from all intelligence agencies and passed to the president and a handful of advisers. “It was in the briefings by the beginning of January,” a person involved in preparing the president’s briefing book told me. “On that there is no dispute.” This person went on: “But knowing it is in the briefing book is different from knowing whether the president saw it.” He didn’t need to spell out his point, which was: Of course this president did not.

To sum up: The weather forecast showed a dangerous storm ahead, and the warning came in plenty of time. At the start of January, the total number of people infected with the virus was probably less than 1,000. All or nearly all of them were in China. Not a single case or fatality had been reported in the United States.

2. The Air Traffic Controllers

From the sky you see only the natural features that separate countries and continents—mountains, water—and not the political demarcation lines. The system that makes flying safe has done so by means of a thoroughgoing, borderless internationalism.

Controllers and flight crews around the world are supposed to be competent in the same spoken language—English—and use the same formulaic instructions that serve as an unambiguous code. For instance: Aviation English prescribes “tree” as the pronunciation for three, in part because the th- sound can be difficult for non-native speakers. Controllers around the world say “Climb and maintain 4,000 feet” rather than “Climb to 4,000 feet,” because to could be misheard as two. Controllers in Paris sequencing a Korean Air plane to land between ones from Lufthansa and Aeromexico at Charles de Gaulle Airport must be sure that all the nationalities involved will follow the same procedures in the same way.

In cases of disease outbreak, U.S. leadership and coordination of the international response was as well established and taken for granted as the role of air traffic controllers in directing flights through their sectors. Typically this would mean working with and through the World Health Organization—which, of course, Donald Trump has made a point of not doing. In the previous two decades of international public-health experience, starting with SARS and on through the rest of the acronym-heavy list, a standard procedure had emerged, and it had proved effective again and again. The U.S, with its combination of scientific and military-logistics might, would coordinate and support efforts by other countries. Subsequent stages would depend on the nature of the disease, but the fact that the U.S. would take the primary role was expected. When the new coronavirus threat suddenly materialized, American engagement was the signal all other participants were waiting for. But this time it did not come. It was as if air traffic controllers walked away from their stations and said, “The rest of you just work it out for yourselves.”

Read: America’s patchwork pandemic is fraying even further

From the U.S. point of view, news of a virulent disease outbreak anywhere in the world is unwelcome. But in normal circumstances, its location in China would have been a plus. Whatever the ups and downs of political relations over the past two decades, Chinese and American scientists and public-health officials have worked together frequently, and positively, on health crises ranging from SARS during George W. Bush’s administration to the H1N1 and Ebola outbreaks during Barack Obama’s. As Peter Beinart extensively detailed in an Atlantic article, the U.S. helped build China’s public-health infrastructure, and China has cooperated in detecting and containing diseases within its borders and far afield. One U.S. official recalled the Predict program: “Getting Chinese agreement to American monitors throughout their territory—that was something.” But then the Trump administration zeroed out that program.

“We had cooperated with China on every public-health threat until now,” Susan Shirk, a former State Department official and longtime scholar of Chinese affairs at UC San Diego, told me. “SARS, AIDS, Ebola in Africa, H1N1—no matter what other disputes were going on in the relationship, we managed to carve out health, and work together quite professionally. So this case is just so anomalous and so tragic.” A significant comparison, she said, is the way the United States and the Soviet Union had worked together to eliminate smallpox around the world, despite their Cold War tensions. But now, she said, “people have definitely died because the U.S. and China have been unable to cooperate.”

From May 2020: H. R. McMaster: What China Wants

What did the breakdown in U.S.-Chinese cooperation mean in practice? That the U.S. knew less than it would have otherwise, and knew it later; that its actions brought out the worst (rather than the merely bad) in China’s own approach to the disease, which was essentially to cover it up internally and stall in allowing international access to emerging data; that the Trump administration lost what leverage it might have had over Chinese President Xi Jinping and his officials; and that the chance to keep the disease within the confines of a single country was forever lost. “If Trump had been following the norm of previous presidents, we would have known about this informally, because our people would have been on the ground in China,” Shirk said. “But the Trump administration pulled them out, and the last epidemiologist who worked for the U.S. government left last year.”

In addition to America’s destruction of its own advance-warning system, by removing CDC and Predict observers, the Trump administration’s bellicose tone toward China had an effect. Many U.S. officials stressed that a vicious cycle of blame and recrimination made public health an additional source of friction between the countries, rather than a sustained point of cooperation, as it had been for so many years. Through Trump’s time in office, official American attitudes toward China have been a mixture of servility and truculence. Trump himself has been almost as personally flattering and subservient to Xi Jinping as he has been to Vladimir Putin. In his speeches and tweets he has emphasized that Xi is a “great leader” and his personal friend. (And if former National Security Adviser John Bolton’s account is to be believed, Trump told Xi that he liked the idea that Xi was holding Muslim Uighurs in concentration camps in Xinjiang.) But at the official level, Trump’s administration has been as hostile to China as Trump sounds in his rally speeches, when he utters “Chy-nah” as if the word itself were profane. Visa allowances have been tightened; long-standing cooperative arrangements have been cut; “thought leaders” of the administration, from Vice President Mike Pence and Secretary of State Mike Pompeo onward (but not including Trump, who is a “tone leader”), have suggested that it’s time for a new Cold War, with China as the existential foe.

“The state of the relationship meant that every U.S. request was met with distrust on the Chinese side, and every Chinese response was seen on the American side as one more attempt to cover up,” Paul Triolo, a former U.S. government official with extensive experience in Asia (and who is now with the Eurasia Group), told me. “There was a huge distrust of China as a malign actor on all levels, that you would never want to help them in any way. At the working level, this had a significant impact.”

In January, Trump-administration officials asked to send back into China some of the CDC observers they had previously withdrawn. The Chinese declined. “One of the puzzles has been why the Chinese initially said no when we finally offered to send people there,” Susan Shirk said. “I think they must have been alienated by our having pulled those people out.” Several weeks later, some observers did get in.

Read: Don’t believe the China hype

In normal circumstances—three words I heard often as a qualifier in these conversations, sometimes also phrased as “in a normal administration” or “with a normal president”—the president’s national security adviser would have called his counterpart in Beijing, and worked out a quiet modus vivendi for dealing with the pandemic. Or the president himself would have called his Chinese counterpart. What the U.S. would have wanted early in the process—at the beginning of January, while the Chinese were still covering up the extent of the disease, or in February, when the disease was beginning to spread more rapidly—would be (ironically) more U.S. scientists on the scene, and more Chinese openness to the world.

Would Xi Jinping have been willing to consider such requests, if he had received a call from the president? “I think there would have been leverage,” Ryan Hass, now of Brookings, who was the senior NSC staffer for China policy under Barack Obama, told me. “Not out of goodwill. Just pure self-interest. If we would have privately brought to the Chinese leadership’s attention that they had a potential pandemic outbreak in one of their provinces and we wanted to provide assistance in locating the source and scale of its spread, they would have answered the phone call.”

Several officials who had experience with China suggested that other presidents might have called Xi Jinping with a quiet but tough message that would amount to: We both know you have a problem. Why don’t we work on it together, which will let you be the hero? Otherwise it will break out and become a problem for China and the whole world.

These calls never happened. Donald Trump has claimed that impeachment proceedings, which ran through much of January, preoccupied him. That didn’t keep him from making five separate campaign trips to rallies during that same month, or from watching television (and tweeting about it) for several hours every day.

Beginning with Jimmy Carter’s administration and continuing through Obama’s, the U.S. and Chinese governments had woven an ever-denser web of institutional and personal connections. U.S. Treasury officials met regularly with officials from the Chinese ministry of finance; the Pentagon and the People’s Liberation Army had exchange programs; the Federal Aviation Administration trained Chinese air traffic controllers; and on through a long list, whose combined intention was to buffer inevitable superpower strains. Under Trump, most of these stopped. The only influential U.S. officials who had regular contact with Chinese counterparts were Steven Mnuchin, the Treasury secretary, and Robert Lighthizer, the trade representative. They were intent on getting the “phase one” U.S.-China trade deal signed, and all other business ran a distant second to that. Mnuchin and Lighthizer “didn’t want to be pressing Xi Jinping with anything else,” a former intelligence official told me.

Read: China has dominated the West before

“CDC asked for access, and was denied it [by the Chinese government],” Ron Klain, who coordinated efforts against the Ebola pandemic during the Obama administration, told me. “In normal circumstances, that request would have gone up the chain, and you would have had senior-level people in the NSC pressing at senior levels. My guess is that it wasn’t pressed in this case because the senior people were Mnuchin and Kudlow, and they had other priorities.” (Larry Kudlow, the director of the National Economic Council at the White House, was also pushing aggressively for a trade deal.)

“It would have taken diplomatic pressure on the Chinese government to allow us to insert our people” into Wuhan and other disease centers, Klain said. “The question isn’t what leverage we had. The point is that we gave up leverage with China to get the trade deal done. That meant that we didn’t put leverage on China’s government. We took their explanations at face value.”

Trump flattered Xi Jinping in public statements until the trade deal was signed, on January 15, and for a while kept on flattering him. On January 22, the U.S. had its first diagnosed case—a traveler who had arrived from Wuhan a week earlier. On that day, Trump referred to this traveler and said, “It’s one person coming in from China. We have it under control.” Eight days later, on January 30, he said, “We’re working very closely with China and other countries, and we think it’s going to have a very good ending for us.” The next day, Trump issued his partial ban on travel from China, but through February he was still publicly complimenting Xi Jinping. “He is strong, sharp, and powerfully focused on leading the counterattack on the coronavirus,” Trump said of Xi on February 7.