Despite the pandemic, the University of Arizona enrolled a record 45,000 students in the fall of 2020. The university estimated that 30,000 of them returned to Tucson for the start of the school year. (Jordan Putt)

The Risk Universities Can’t Not Take

Inside the University of Arizona's $8-million, 117-day fight to reopen amid a pandemic

Ian Pepper and Charles Gerba have been waiting 30 years for the chance to use sewage to save the world. And in the last week of August, it looked like they might have done just that—or at least saved the sunbaked corner of Tucson that is the University of Arizona campus, at least for a little while.

Pepper, 74, is a microbiologist. Gerba, 75, is a virologist. They have spent a combined 82 years on the faculty at Arizona, and they are world-renowned experts on, among other subjects, the germs found in human waste. Their shared lab isn’t on campus, but on the grounds of a county sewage-treatment plant, where they have “sewage on tap,” as Gerba puts it. They’ve worked for years to find ways to use wastewater testing to get an early warning about the spread of disease. “Sewer mining,” they sometimes call it.

So when the University of Arizona’s president, Robert C. Robbins, announced in April that he was determined to reopen the campus in the fall, with students in dorms and some in-person classes—the first president of a major public university in the country to do so—Pepper and Gerba saw their opportunity. Over the summer, they set up a system to collect sewage at 16 manholes around campus, dropping a scoop 10 feet down into a stream of wastewater from a specific building, or set of buildings, then racing the samples back to their lab to test for the coronavirus. They collect samples three times a week, always in the morning, “around 8:30,” Pepper says. “That’s when people do their business.”

After a spring and summer of intense preparation—and not a little controversy—Arizona started moving students into dorms on Friday, August 14. Returning students were given antigen quick-tests, and could move in only if they tested negative for the coronavirus. Pepper and Gerba started testing the wastewater the following week, looking for early signs of infection. At first, all the dorms tested negative, as did the other buildings on campus.

But on Tuesday, August 25, the day after classes began, the wastewater from Likins Hall, home to 311 students, tested positive. Pepper and Gerba quickly ran verification tests: all positive. On Wednesday, every Likins resident was quick-tested again. Two came up positive for COVID-19. They were asymptomatic, but almost certainly contagious. They were then whisked to one of the two quarantine dorms that Arizona had set aside for just this scenario. A facilities team swooped into Likins Hall and cleaned their rooms.

And Pepper and Gerba? The next morning they were back, sampling the wastewater from Likins Hall. It was once again negative, and the dorm’s remaining 309 residents returned to the coronavirus-campus version of college.

The episode perfectly captured Arizona’s far-reaching effort to bring students back to campus—a multimillion-dollar ballet of precision logistics, campus reengineering, and unpredictable human behavior. In the 16 weeks leading up to reopening, everything at the university bent to accommodate the coronavirus. The food-service staff cut holes in the walls of four of the school’s 29 on-campus restaurants, so students could get their food through a walk-up window, without coming in the building. University faculty teamed up with a Silicon Valley start-up to develop a smartphone-based “virus exposure” app. Facilities staff re-tuned the air-conditioning system of every building on campus to bring in more fresh air, and upgraded the filters to those used in hospital operating rooms. They fabricated and installed 1,755 plexiglass shields and sneeze guards, replaced 3,000 paper-towel dispensers with the touchless, battery-operated variety, and mounted 1,530 hand-sanitizer dispensers in 326 campus buildings.

In six or 12 or 24 months, all of this might look like a triumph of creativity, cutting-edge science, and the willingness to try anything that could matter, including using a 10-foot pole to collect the raw sewage of college students. Or it might look like an ill-conceived attempt to control the transmission of a deadly and highly communicable disease among a group of people not known for following rules or keeping social distance. The difference between those two outcomes may turn on the finest details: an air-conditioning filter overlooked, a building entrance without hand sanitizer, an unrestrained Labor Day–weekend party. And the stakes aren’t just the health of Arizona’s students and staff, or the reanimation of the campus in the fall of 2020; that’s not what drives you to collect and analyze the excrement of your students.

University staff sample and test raw sewage flowing out of Arizona’s dorms in order to help monitor the spread of the coronavirus on campus. Sixteen manholes are sampled three times a week, as part of a program designed by the Arizona scientists Ian Pepper and Charles Gerba.

The coronavirus pandemic left America’s colleges and universities, particularly its public ones, in an impossible bind. Shutting down campuses was the safe choice when it came to the virus itself, but carried the risk of alienating students and devastating brittle university finances: After Harvard announced that all its classes would be online, almost a quarter of undergrads opted to take the year off—a tuition hit only the richest institutions can withstand.

But opening campuses—even with rules, even with unlimited supplies of face masks and hand sanitizer, with testing, with dorms limited to one student per room—could create a runaway on-campus outbreak: dozens or hundreds of students catching the coronavirus, and then perhaps passing it not just to their classmates, but to the thinned-out ranks of campus food workers and instructors, not to mention people in the surrounding community.

Open a university’s campus, and you could be knocked back to Option 1—all virtual—within days, as happened at the University of North Carolina at Chapel Hill, having wasted the work of thousands of staff over months of preparation, having wasted the money and effort and expectation that students put into going back in person. Not to mention having opened administrators to waves of criticism and mockery for their foolishness and mismanagement in trying to reopen in the first place.

The choice is existential. People’s lives are at risk. But the institutions themselves are also at risk—the programs and faculties and reputations, the quality of instruction and the campus culture that took decades to build.

At the University of Arizona, President Robbins’s decision to reopen the campus conveyed a particular authority and reassurance, because for most of his career he was a physician—a cardiac transplant surgeon. When you signed up as Robbins’s patient, you were signing up to have him cut the still-beating heart from your chest, and replace it with something better, something healthier.

Robbins knows something about disease and immunity, about weighing choices and risk when lives are at stake. If you need the aortic valve in your heart replaced, you have to evaluate how you think about survival.

“Your best chance of being alive five days from now is definitely not to let me replace it,” Robbins told me. “You might go into cardiac arrest on the table. You might get an infection and die. It’s a small chance—0.5 percent or 0.2 percent—but possible. But if you do nothing, you’re going to die within a year, maybe two years. If you want to be alive five years from now, you need me to replace your aortic valve.

“This is risky.” But not doing it might well be riskier.

The University of Arizona campus in Tucson. To be ready to bring back students in person, faculty and staff had a little more than 100 days to change everything about how the campus worked, including the way organic-chemistry lab was taught and how bagels were served.

Arizona is one of those places that seem to have popped up fully formed: appealing in a certain way, fast growing, and most especially new. A place of sunshine and fresh starts. More than half the people in Arizona have been there only since 1990.

But the University of Arizona was founded in 1885—27 years before Arizona itself became a state. It is a land-grant university, with 36,500 undergraduates and almost 9,000 graduate students, and a history of serving Arizona’s Latino and Indigenous communities. It has a law school, a business school, a school of public health, two separate medical schools, a school of agriculture, and a wide reach into the far corners of the state—a cube of Western red-rock geography that is bigger than New York and Pennsylvania combined but that is home to just 7 million people.

Arizona, like most big universities, closed its campus quickly in mid-March, moving all its classes online. A few weeks later, on April 28, Robbins launched his campaign to reopen, on CNN’s talk show OutFront with Erin Burnett. The spring term still had a week of online classes to go, but Robbins was there to talk about the fall. In-person classes would surely have to be combined with online learning, he said, “but we’re excited about having our students come back to campus”—including, he told Burnett, living in dorms.

Robbins, silver-haired, was casual in a bright-red Wildcats pullover, and reassuring (the onscreen chyron identified him as a “cardiac surgeon”). When he talked about bringing students back in person, it didn’t seem like such a crazy possibility.

There was only one problem: Robbins hadn’t exactly told his own faculty and staff—16,500 people—that Arizona was reopening. Five days earlier, he’d sent an email to the campus community, but it had been a good deal more circumspect than he was with Burnett. Thousands of staff, and thousands of students, found out about the reopening from CNN, or from people who had seen Robbins on CNN.

It was an odd fumble. Universities are devilishly hard places to run, and reopening in the middle of a pandemic was going to require every ounce of energy and effort from everyone. Robbins had already announced temporary salary cuts to cope with revenue losses he expected from the pandemic, and now he’d taken his reopening plan public before rallying the troops.

Two mornings after his CNN appearance, Robbins held an on-campus press conference to make the reopening official. But he was able to offer few details. “Ultimately, there’s going to be meticulous attention to every detail that we can,” he said. And as would become typical, he added a qualifier: “There’s going to be things we can’t control.”

That was Thursday, April 30. The following Tuesday morning, Robbins was up at 4:30 a.m. for an interview on NBC’s Today show; that afternoon, he was on MSNBC.

The timing—of the decision, the announcement, the publicity—wasn’t coincidental: May 1 is the traditional commitment date for new college students around the country, including Arizona. That day, a $425 registration deposit is due to reserve a spot at Arizona.

“We knew, if we were planning to open, we should get that message out before May 1,” Kasey Urquidez, Arizona’s dean of undergraduate admissions, says. Arizona had decided to give students until June 1 to pay their deposits, but it was competing for students with other universities. And as students and parents were weighing their choices, one of those universities had declared that it would make every effort to bring students back to college in person. In the week of the reopening news and Robbins’s media blitz, 1,000 fresh students paid their deposits, a jump of 40 percent over the previous weeks.

Robbins’s decision would affect the life and future of every instructor, professor, and employee of the university. And while he constantly mentioned the health and safety of the staff and faculty, his lack of consultation with them in advance, and failure to even give them notice of what he was about to announce, grated on many. “Our particular president doesn’t do that,” says Jessica Summers, an associate professor in the college of education and the president of Arizona’s faculty senate. “He doesn’t come to the faculty senate and run it by us.”

Despite efforts to keep the virus from Arizona’s campus, by the second week of classes an outbreak took hold that would ultimately infect 2,300 students, before starting to taper off. Left: Student Hannah Wolvos on campus. Right: A sign faces out from one of the University of Arizona's quarantine dorms.

To both supporters and critics, Robbins, 62, has been the character at the center of the reopening drama, because he’s ultimately making the decisions, but also because he has taken a high-visibility approach, doing hour-long televised coronavirus briefings, streamed live, every week—a total of 18 by mid-September. Although Robbins had been both the chief resident and the head of cardiothoracic surgery at Stanford University, he started his academic career by getting an associate’s degree at a small community college in tiny Ellisville, Mississippi, the rural town where he grew up. He got his MD from the University of Mississippi. Robbins has a low-key, accessible southern manner. Pre-pandemic, he would sometimes ride his bike to work, a habit he started at Stanford, and he likes to give students his cellphone number.

Early in the effort to reopen, Robbins brought on board a lieutenant to help manage the sprawling process, a fellow surgeon named Richard Carmona, whose role has been to help organize the cascade of data, reports, and suggestions from the university’s staff and faculty. Carmona, who has a faculty appointment at Arizona, was the surgeon general for George W. Bush, and has a cheerfully forceful manner. He joins Robbins for each of the video briefings.

The announcement of the plan to teach some classes in-person left many faculty feeling “unsure. Disconnected. Anxious,” Summers told me. How could they possibly know whether it would be safe to show up in a classroom, lab, or lecture hall three months into the future?

While Robbins might have seen the TV appearances as a way of raising the profile of the university and bolstering confidence and enrollment at a shaky moment, some faculty saw something else: the start of a campaign to pressure them to go back to the classroom in person whether they were ready or not.

University of Arizona President Robert C. Robbins, a former heart-transplant surgeon

To understand Robbins’s decision—not to mention the preparation, the politics and the media blitz—it’s crucial to understand the finances of the University of Arizona, which show starkly why the school, and others like it, so urgently needed students to return to school this fall.

According to Arizona’s constitution, the price of the state’s public universities “shall be as nearly free as possible.” The reality is not even close. For an Arizona resident, tuition and fees at the Tucson campus this year are $12,600. As recently as 2010, they were just $6,842. During a decade of insignificant inflation, the cost to attend Arizona has gone up 84 percent.

Here’s why: Arizona’s state legislature has cut funding for the state’s universities by more than half since the Great Recession, from $10,300 a student in 2008 to $4,800 a student last year. Arizona ranks 43rd in the nation in higher-education spending, per student.

The University of Arizona has an annual budget of $2.2 billion. (Harvard’s budget is twice that, and Harvard enrolls half as many students.) At this point, only 11 percent of that money comes from the legislature. More than 30 percent comes from tuition.

That crucial tuition funding is especially vulnerable in the pandemic because of another strategy that Arizona, and a lot of other prestigious state universities, relies on to fill the gap: wooing out-of-state students and charging them market rates. At Arizona, 40 percent of undergraduates are from out of state, and they pay $36,600 a year; as a group, they provide twice as much in tuition dollars as Arizona residents do.

“We depend on out-of-state students,” Robbins told me. “We need them to come back to make it work financially.”

If the University of Arizona is all online, why would out-of-state students attend class on their laptop, for $36,600, when they could attend their own state universities—also via laptop—for one-quarter or half that price?

Students Becca Worden (left) and Emma Harris (right) on the University of Arizona's mall. When school started, the only classes allowed to meet in person were those for labs or performing arts—and they were limited to 10 people. Still, that brought 5,000 students into campus buildings the first week of school.

College and university presidents were often accused of announcing their reopening plans in the spring in an act of winking cynicism—they weren’t committed to in-person education, or bringing life back to their campuses, they were just trying to get kids to pay up. The presidents were mocked for their mercenary attitude, as if the tuition dollars flowed directly to their bank accounts.

But in the case of universities, when you say “money,” what you’re talking about is sustaining institutions and communities whose longevity belies how fragile they are. A geology department, a music school, a computer science department, a business school take years, even decades, to build; they involve complicated intellectual, managerial, and personal relationships. They have momentum and purpose. And all of those things can be shattered. Universities have deep resources of learning and experience, history and heritage, but most don’t have much in the way of savings accounts. The University of Arizona’s endowment is $1 billion—half of one year’s budget. If the University of Arizona were to lose more than 20 percent of its undergrads for a year—as Harvard has—it would face financial armageddon. If 8,750 students took the year off—if half were in state and half were out of state—that alone would create a $215 million gap in revenue, just from tuition.

Robbins clearly takes seriously the health of the students, staff, and faculty he was trying to bring back in the fall. But he takes just as seriously his responsibility to protect the institution. The way Robbins sees it, no one was suggesting sacrificing people on behalf of reopening. Just the opposite: The summer was spent implementing every precaution that could be imagined—short of not opening at all.

And the damage of not trying to open could have been quick and brutal. If Robbins faced an immediate $200 million or $300 million deficit, even for just a year, he’d have to layoff significant numbers of faculty and staff. The very best professors wouldn’t have to flee—they would be recruited away by more prestigious institutions. Not to mention the impact on the health and careers of the hundreds of faculty and staff who would be laid off, those people who were ostensibly being protected by not opening

So of course trying to bring people back to the University of Arizona—and to Purdue and North Carolina State, to Harvard and Yale—was about money. But only insofar as the health of the university itself is about money. In this pandemic, in this moment in the United States, it was about health and it was about money, and ignoring either would have been equally rash.

The only way to reopen a university is to set the goal months in advance, work furiously toward it, and be ready to not reopen if circumstances change. If you wait to know whether reopening will be safe, you won’t have time to do it.

Universities are deliberative, with widely dispersed power. But in confronting a crisis such as the pandemic, they also have a significant advantage: More than perhaps any other organization, they are like their own small cities, with in-house IT services, health clinics, and maintenance shops—not to mention public-health and epidemiology experts. Robbins and his senior staff spent their days in Zoom meetings, trying to settle on mask policies and testing policies and what kinds of classes should meet in person; the rest of the university had barely 100 days to ready the campus for a semester unlike any other.

Chris Kopach, the university’s director of facilities, marshaled a staff of 600 to adapt the HVAC systems, triple-source disinfectant, PPE, testing supplies, and hand sanitizer, and stockpile 5,000 spray bottles, 400,000 surgical masks, and 120,000 cloth masks with Arizona’s logo. They manufactured and posted 40,000 signs  (about social distancing, elevator occupancy, one-way stairwells), the sign-making machine working so hard, Kopach says, “it has smoke coming out of it.”

Left: Chris Kopach, the director of facilities at the University of Arizona, with some of the 5,000 spray bottles for disinfectant on hand for fall term. Right: Jennie Norris prepares a sign in the University of Arizona's sign shop. Norris estimated that she and her colleagues have created about 40,000 signs in preparation for the reopening of the University of Arizona.

With testing bottlenecks across Arizona delaying routine test results for days, the university built its own ability to quickly process PCR tests in-house—1,700 a week. It also bought antigen quick-test machines from Quidel, with the ability to do 1,600 tests a day. Walk in Monday through Friday, get swabbed, and the result comes to your phone in an hour.

The food-service operation at Arizona, which typically serves 25,000 people a day, had to be completely reinvented. Every meal served on campus now—whether a bagel or a full-course dinner—is prepared to go. None of the 21 open restaurants have indoor seating. Locations with self-serve drink fountains have assigned a staff member to the fountain, to get the drink each student requests. (No refills.) Cameras and artificial intelligence monitor the traffic of customers in the campus food court. The credit-card readers at every cashier stand at every restaurant have been repositioned to face the customers, so they can slide their card themselves. “Everything has changed,” Todd Millay, the senior director of Arizona’s student unions, says. “Except we’re still serving food.”

Millay wants to serve every meal someone on campus wants to buy, and he wants to be able to do that as quickly as possible; social distancing is vital, but it slows everything down.

To help get food to students quickly, and without human contact, Arizona’s food services installed two banks of “smart lockers.” A courier delivers your food to one of the lockers, which pops open after scanning a code on your phone.

He expects delivery service to surge—he contracts out to Grubhub, which mostly employs students to bring meals to other students—and he’s turned two of his closed restaurants into staging areas just for student pickup and for delivery couriers. He’s also experimenting with digital “smart lockers,” for quick, contactless meal pickup: The door opens on your food after the locker scans a code on your phone. Millay ordered and installed two banks of smart lockers—15 lockers each, at a cost of $10,000 a piece. If they work well, he’ll order seven more. He’s also deployed a robotic salad maker, which he says is “amazing.”

As a substitute for the spaces where students used to gather, Millay converted the university’s grand ballroom to an indoor picnic space. Each round banquet table seats four instead of eight, and each seat has a three-sided plexiglass enclosure. “You have your own spot at the table,” he told me. “You can take your mask off. You are sitting behind your plexiglass shield; the people you’re eating with are behind their plexiglass shields.” He also purchased and erected five tents across campus, also with distanced tables and chairs, although those areas may have more appeal a few weeks further into the semester, when the daytime temperature drops below 100 degrees.

The grab-and-go meals may keep on-campus students fed, Millay told me, but they’re no substitute for the food court, the student union, the Starbucks, and the deli with their seating areas. “This is their living room. This is their space of relaxation,” he said. “We are an essential part of campus life. And also a super-spreader event waiting to happen.”

Arizona’s student union, which normally serves 35,000 students a day. For the reopening, the university turned to cameras and artificial intelligence to help make sure the food court didn’t get too crowded.

Tucson’s summer was almost rainless, and punishingly hot—the hottest in the city’s history. As the days ticked down toward August, the university’s reopening efforts faced two distinct challenges. The first was the sprawling effort to prepare the campus itself. The second was a growing, and sometimes angry, opposition to the leadership of Robbins and Provost Liesl Folks, and to the reopening itself, from faculty and staff.

Magnifying the stakes of both was the pandemic, which unleashed itself across Arizona, and around the university in Tucson and Pima County, right in the middle of the reopening effort.

In the space of weeks, Arizona went from placidity to the center of one of the most out-of-control coronavirus outbreaks in the country. On April 28, the day Robbins announced his determination to reopen, Arizona reported 232 new cases statewide. But for much of the summer, Arizona’s seven-day average of new cases was 2,300 a day or more. It was so bad that on June 15—before new cases soared to 4,000 a day—Robbins said in an interview, “If this keeps going, we’re in big trouble. If it doesn’t relent in the next six weeks, we’ll have to pull the plug on this plan.”

Two things happened. Astonishingly, the number of new cases in Arizona fell almost as steeply as it had risen. And the outbreak never got a foothold on campus.

Starting in June, about 7,000 people a day were coming to the University of Arizona—people getting the campus ready, and faculty and students working in some of Arizona’s 800 research labs, which were allowed to reopen if they could follow strict protocols. Over six weeks, not a single case of on-campus transmission of the coronavirus occurred, even amid the soaring cases in Arizona and Pima County. That gave Robbins and his team a sense of hope, if not quite confidence, as August 24 approached. The preparation and the protocols, if followed, actually worked: Everyone on campus wore masks, the labs and offices were all socially distanced, people worked staggered shifts, the air-conditioning system had been adjusted, and spaces that needed them were equipped with plexiglass shields. There was plenty of disinfectant.

But then, on July 30, just two weeks until students would start arriving, a little more than three weeks until classes would begin, it began to look as if all the work of the summer might unravel.

That Thursday, at his regular weekly coronavirus briefing, Robbins had announced that not only would Arizona be reopening as planned on August 24, but the in-person classes would roll out across campus according to a crisp, week-by-week schedule.

The lobby of one of the University of Arizona’s isolation dorms. Arizona set aside two dorms, with a total of 418 rooms, in which to isolate students who tested positive for COVID-19. They would need all 418.

Week 1, essential in-person classes—organic chemistry, gross anatomy, classes that simply couldn’t be done remotely—would begin, limited to 10 people, bringing 5,000 students back to campus buildings that first week.

Week 2, the university would add classes of up to 30 students. Week 3, the university would add classes of up to 250 people, in lecture halls with 1,000 seats. That third week, Phase 3, would bring a total of 14,000 students to campus at the same time at some points during the week.

Robbins didn’t suggest any latitude for adjustment if the rollout was bumpy. Typically good-humored and affable, on this Thursday morning, Robbins seemed a little testy and impatient. When asked by reporters specifically about the lack of flexibility in his plan, Robbins said, “We discussed this idea of going slowly to move the campus forward. I’ve chosen to go faster. Because I think we’ve hit a lot of the things that the experts have told us we need to do.”

He alluded to his critics. “This accelerated two-week period of rolling into the semester will obviously not please everyone,” he said. “Some people have said we don’t have our plan together. I would say, I have listened to our epidemiologists. I have listened to the experts.”

But a group of university epidemiologists and public-health faculty listening to Robbins’s briefing were swallowing hard, and trying not to panic. Messages flew in real time on a texting group they shared. Despite Robbins’s reference to epidemiologists and experts, it wasn’t a plan they endorsed. “I went ‘Holy cow,’” Joe Gerald, an associate professor at Arizona’s Zuckerman School of Public Health, told me. A key problem with the coronavirus response across the country—at every level—had been exactly what Robbins was suddenly endorsing: a speedy determination to reopen bars, gyms, hair salons, whole states. And now, apparently, the classrooms at UArizona.

Masks and social distancing, one-way stairways, and elevators limited to two people were all great. But this plan to get on with it once classes started looked like a prescription for failure. Every seven days would bring thousands of new people to campus without any certainty about what the impact of the previous week had been. Were infections spreading? Could testing keep up? Were there enough quarantine beds? Were instructors and students comfortable and safe seeing one another in person?

The doctors and scientists watching and listening to the weekly briefing thought this unbending schedule might well end up wasting everyone’s efforts in getting ready, not to mention wasting the hopes and efforts of the students who had returned to Tucson.

“I told my epi colleagues, ‘We have to do something to slow this down,’” Gerald said. The informal group—seven physicians, epidemiologists, and public-health experts, all colleagues from Arizona’s own faculty—met on Zoom the next morning, Friday, July 31, and came up with a plan to try to change Robbins’s approach.

Joe Gerald, one of an informal group of public-health experts at the University of Arizona who worried that a move to reopen too fast would overwhelm all the safeguards the university had set up.

They wanted to give their boss a series of quantitative benchmarks—new cases on campus, testing capability, capacity of the quarantine dorms—which would make clear whether moving from one style of in-person classes to the next was safe. The group didn’t oppose the reopening; they just wanted it to be done right. “We want the university to be successful,” Gerald told me. “We want the students to be safe.”

Gerald is a faculty senator, and a senate meeting was scheduled for that Monday, August 3. So he wrote a brief but direct senate resolution, asking the administration to use public-health criteria—not calendar dates—for moving from 10-person classes to 30-person classes to 250-person classes, and to make those criteria public. Robbins and Provost Folks dialed into the faculty meeting, along with more than 800 senators, faculty members, and observers. Gerald’s nonbinding resolution was approved without a single “no” vote.

Not long after the meeting ended, the public-health group got an email: The president wanted to meet Wednesday to hear more. They spent two hours the next day strategizing their pitch to Robbins. “We were really worried about this perception that the university was going to bait-and-switch students,” Gerald said. “We would start to reopen with face-to-face classes; then we’d recognize we couldn’t contain the virus; students would get the blame—they didn’t wear masks; they partied too hard on weekends—then we’d shut down and go online.

“And if their tuition wasn’t refunded, that was going to leave a terrible public-relations stain on the university. And it could jeopardize our credibility.”

Robbins ran Wednesday’s conversation over Zoom. “He has a real way of putting you at ease, which allows you to speak your mind fairly comfortably,” Kacey Ernst, an epidemiologist who was an outspoken member of the group, told me. “He said, ‘I think we’re on the same page; I think we have creative differences,’ slightly tongue-in-cheek. That sense of humor is helpful. He kind of walks that line between being warm and open, but making sure you know that the decision maker is him.”

Robbins questioned each of the metrics the public-health scientists were suggesting. “Why this metric? Where does the data threshold come from? What’s the rationale? He went through those very methodically with us,” Ernst said.

Over the course of nearly two hours, the physicians and scientists made one point consistently: If you push too hard, too fast, the systems you’ve set up can break. This slower approach doesn’t undermine the plan—just the opposite. “We told them, ‘We need to do this, so we don’t have to pull the rug out from under the students four weeks in,’” Gerald said.

It worked. Robbins agreed to slow down.

One thing never came up: money.

“Everybody knows money is the 400-pound gorilla in the room,” Gerald said. “But it was never even a minor element in the discussion ... I don’t think they want to bring students back in any other way but safely.”

The next day, Robbins had his usual Thursday-morning public briefing. “We’re taking an on-ramp approach to all in-person and flex in-person classes,” he said at the very opening. “As I shared last week, we expect the on-ramp to span several weeks.” Its length, he said, slipping in the policy change without quite acknowledging it, would depend on how things developed. “We will be data driven,” he said.

Robbins went on to thank the faculty members who’d provided input. “They convinced me we needed to go a little bit slower,” he said. “Because we need to make sure.”

With that commitment from Robbins, Ernst said, “I feel like everyone has done the best they possibly can.” Across the university, she told me, opinions have ranged widely about the decisions themselves, and also the style of the decisions. But everyone has been committed to trying.

“Reopening has kind of taken over,” she said. “It’s been exhausting. Has it been enough? That remains to be seen.”

Students sit outside the University of Arizona's Likins Hall, where a coronavirus outbreak was caught early.

Robbins maintained from the beginning that students wanted to come back to campus. The numbers bore that out. The university estimated that about 30,000 of its 45,000 undergraduate and graduate students returned to Tucson as school started. The dorms did not, however, fill to their pandemic capacity—only 5,000 students decided to live on campus, of the 6,500 spaces they could accommodate.

But the university was more popular than ever. It got more applications for its freshman class and more transfer applications than in any previous year, and overall enrollment set a record. The entering class is also the most diverse Arizona has ever admitted: 48 percent nonwhite and 31 percent first-generation college attendees.

Students moving in were required to get quick-tested on arrival at campus, and be negative in order to move in. As was the case at many universities across the country, Arizona’s pre-entry testing produced astonishingly low rates of infected students. Of the first 4,274 students tested as they arrived back at school, only nine tested positive.

Things went well the first couple weeks. That’s when the sewage testing of Ian Pepper and Charles Gerba caught two infections in Likins Hall.

For the first month of school, Arizona has kept just the smallest, essential classes in person. Phase 1. Everything else remained online.

But in the second week of classes, the week starting August 31, the virus came to the University of Arizona in quick waves.

The previous 31 days had seen a total of 73 positive tests. That Monday, the wastewater in three dorms tested positive for the coronavirus, and over the next few days, Arizona tested all 1,225 students from those dorms.

On Monday, 31 people on campus tested positive.

On Tuesday, 63.

On Wednesday, 104.

The next week, 402 people in four days.

Arizona’s systems held up well. At one point, 400 students were in quarantine on campus. Wastewater testing was expanded from twice a week to three times a week. Every dorm resident had to be tested every week. And only 20 faculty and staff, out of 16,500, tested positive. As of September 30, no students from the thousands who eventually tested positive have gotten seriously ill, or even required hospitalization. One faculty member died in the spring from COVID-19, the university said, but not from a case caught on campus. Since then, no staff member has gotten seriously ill or required hospitalization.

Those numbers, of course, could change before the term ends. And across the country, colleges reopening had consequences. In late September, the CDC reported that cases among college-aged people had jumped 55 percent between August 2 and September 2. According to a New York Times analysis of 1600 American colleges and universities, these schools have seen 130,000 cases—about 80 cases per school—and 70 deaths, mostly among faculty and staff, since the start of the pandemic, with the majority of cases coming during the fall term.

As infections were spiking at Arizona, Robbins took to walking the campus and nearby neighborhoods at night, to see what the students were up to. They weren’t catching the virus in class or on trips to the dining hall, according to contact tracing. They were getting it from parties, from close living arrangements in off-campus high-rises, from fraternity and sorority houses. Sixty percent of the students who tested positive lived off-campus.

In fact, by the fourth week of classes, the rate of infection among Arizona students was out of control. Robbins himself had dinner with a student who tested positive the next day, but the president’s PCR test came back negative. That week, Robbins and the Pima County health director, Theresa Cullen, announced a “shelter in place” policy, to last 14 days. All Arizona students—whether they lived in dorms or housing adjacent to campus—were requested to stay home, except to go to class, work, the supermarket, or the doctor. Robbins made clear that the fate of the reopening was in the hands of “the small subset of individuals … who are not following the rules.” The shelter-in-place request—along with an effort to tamp down weekend parties—worked dramatically: Daily cases dropped by two-thirds or more.

The reopening had cost $8 million, just to get to August 24. The effort might cost $20 million through the end of the academic year. It was a 117-day sprint for everyone involved, from the April 30 announcement to the start of classes on August 24. Arizona, like many big universities around the country, plans to end in-person classes—if they are still going—just before Thanksgiving, and finish the term online. So in-person classes could last 93 days. The preparation will have taken longer than the actual class participation.

The process was bumpy. The communication across campus was sometimes confused or imprecise. Although the university had dozens of task forces—every one of Arizona’s 20 colleges had its own reopening task force, along with the ones at the university level—and although many of the task forces had their own subgroups and working groups, hundreds of faculty felt they weren’t adequately consulted.

Bobby Robbins and Liesl Folks came to the crisis at a bit of a disadvantage. Robbins had been the university’s president for only two and a half years when the pandemic hit; Folks had been its provost for only seven months. Neither had had the time to develop a well of trust, or even a wide group of allies.

If things unravel, plenty of people will say that the whole process was destined to unravel. But two of the most experienced and most concerned—Joe Gerald and Kacey Ernst—wouldn’t agree.

“Not only would I not criticize them; I would commend them,” Gerald said before the reopening started. “They have been listening to everyone and anyone who can contribute. They have made an incredible good-faith effort to do everything they can to bring students back in a safe manner.”

Students Blake Thomas, Kayden Thomas, Kayla Wright, and Izzy Ehrlich practice social distancing outside the University of Arizona's School of Information.

The most important question you can ask about the universities that tried to welcome back students this fall is the one almost no one is asking:

What if money weren’t a factor at all?

What if the only concerns were safety and the most engaging education possible?

Bobby Robbins and the University of Arizona—and every college and university in the country—could be living in that world. Congress and the state legislatures could have said in April that they would cushion higher education. They could have said, You all do the safest and the best things for your campus. We will provide the money necessary to make sure you keep your faculties, your facilities, and your programs intact. Everyone might have to take a salary cut. You might have to let the lawns go, and adjust the thermostats. But you won’t have to fire professors or dining-hall workers. That is, in fact, what Congress did for the country’s airlines, providing them with $25 billion over about seven months.

If money weren’t a factor, everyone would have taken a different approach. Many more colleges and universities would have moved everything online and waited for a vaccine or an effective treatment. Some surely would have tried to do some in-person teaching—engineering, anatomy, organic chemistry, dance, theater are difficult to learn over Zoom. With small numbers on campus, with good testing and good safety precautions, it would certainly have been possible. Arizona’s own experience, with 7,000 people on campus over the summer, shows that.

Taking money into consideration doesn’t prove the cynicism of the effort to bring students back. It underscores that college and university presidents were explicitly put in the position of weighing the safety of their students and staff against the survivability of their schools. All but the very richest schools had to protect both.

The University of Arizona owes its existence to the generosity of three Tucson residents, a saloonkeeper and two gamblers.  In 1885, Arizona’s territorial legislature gave the city $25,000 to start the university, but only if it provided the land. Just before the grant was to expire, the three were persuaded to donate the land to get things started. Robbins knows that story; his office sits on the land they provided. And he knows about weighing odds, about taking chances.

Despite an easygoing manner, Robbins has a surgeon’s competitiveness and a surgeon’s bias for action. The whole point of harnessing the entire campus to get ready, he said, “is so we can control our own destiny.” Which is also, of course, the point of heart transplants.

Robbins stopped doing transplants a decade ago.

“I love it,” Robbins told me. “I miss it. But there is nothing more exhilarating and important than running a public university. It’s one of the highest callings you can be entrusted with.”

Robbins doesn’t have a precise count of the number of transplants he did, but for 20 years, he scrubbed in on 30 or 40 a year: more than 500 new hearts.

“It’s a lot easier than the job I have now,” he said.