“It was like God’s message,” she said. “My parents and I were suffering from a bad situation mentally, but suddenly we found hope.”

So the family filed the necessary paperwork and waited for an interview at the US Embassy, a process that usually took just a few months before the pandemic and the Trump administration. Weeks turned into months, and still they waited. Now, more than a year later, the Taliban has taken over their country. With the US Embassy closed, Najma and her family fear they may never be able to get out.

Najma’s family is one of millions from all over the world who applied for the US diversity visa lottery last year. The program hands out about 55,000 green cards to people chosen from countries such as Nepal, Afghanistan, Albania, and Somalia, which don’t typically send many immigrants to the United States. Approximately 1 percent of applicants get selected. This year, 60,602 people in Afghanistan applied, and 2,189 were chosen.

The number of diversity visas allotted in 2021 is among the lowest the United States has ever issued. As of July 2021, the State Department had issued around 3,506 visas out of 55,000 lottery winners, or a meager 6 percent of the visas. As of June 30, they hadn’t issued even a single visa to any of the 2,189 winners from Afghanistan in 2021. And time is running out: If the State Department doesn’t issue the remaining 50,000 visas by the end of September, they’ll expire. The winners will lose their shot at coming to the United States forever. 

It’s somewhat of a miracle that the diversity visa lottery still exists at all. President Trump hated it and tried to dismantle it. First, he said the program brings immigrants from “shithole countries.” Then, citing COVID-19 related reasons, he shut down consulates across the world and froze diversity and a few other types of visas entirely. Then, claiming that the pandemic was taking jobs from Americans, for many months last year, the administration stopped issuing diversity visas altogether. After multiple lawsuits challenged his policies and pressure from federal judges, the administration restarted the program but placed diversity visas in a bottom priority tier, which further slowed processing.

Curtis Morrison, an immigration attorney who is a litigator of one of these lawsuits, said that even though Trump put these policies in place, the Biden administration hasn’t tried very hard to undo them.  

“Absent judicial intervention, everybody's going to lose because this administration has proven that justice and immigration policy is just not where their priority is,” he said.

A State Department spokesperson told Mother Jones that the agency was "making every effort to process as many Diversity Visa cases as possible, consistent with other priorities, despite the severe operational constraints and backlog resulting from the COVID-19 pandemic." The spokesperson added that it's highly unlikely that they'll be able to issue all the allocated diversity visas for the year, but that those who've applied for their visas from Afghanistan could request their visas be transferred to a nearby embassy.

Last week, I spoke with several families in Afghanistan who won the diversity visa lottery—and are now afraid that their dream of a better life in the United States may never come true. I’ve changed all the names of Afghan citizens in this story to protect them from retaliation by the Taliban. 

Jamshed, a civil engineer from Kunduz, applied for the diversity visa for two years in a row and didn’t get selected. The third time, he applied on behalf of his wife, hoping she’d bring him luck. Their application was selected in 2020. Jamshed made multiple trips to Kabul and spent many months collecting documents in hopes that they’d be invited for an interview at the embassy, but it never happened.

Now, he and his family have escaped from their home and are hiding in fear of the Taliban. Jamshed worries that his experience working in US-backed firms and publicizing that he was a lottery winner with intentions to immigrate to the US may put him at risk. He has already started scrubbing his social media profiles.

His wife—a midwife and a teacher—worries that she won’t be able to go out to buy groceries, let alone continue to work.

Winning the diversity lottery was the best news for Jameel and his family in 2020 amid the pandemic. A member of the Ismaili Muslim community, a religious minority group in Afghanistan that has been targeted by the Taliban in the past, Jameel hasn’t stepped out of his home since he learned about the Taliban's takeover of Kabul. He fears he’ll be persecuted because of his faith. He considered escaping from the country by running to the airport but decided against it.

“Where can I escape?” he said. “I cannot. The borders are all controlled by the Taliban.”

Multiple lawsuits challenge the rapidly approaching September 30 expiration date on diversity visas and request the courts to consider carrying forward the quota beyond the deadline. In a lawsuit challenging these rules in 2020, a judge ordered that a portion of the visas wouldn’t be counted as expired. Even if the lawsuits provide some relief for diversity visa winners from other countries, it may be too late for winners from Afghanistan to get out. Given how hard it is for anyone to get out of Afghanistan right now, even if the visas were granted, there’s no guarantee they would be able to leave.

The political climate in Afghanistan is rife with uncertainty. Two major major money-transfer services, Western Union Co. and MoneyGram International Inc., have stopped doing business in Afghanistan. The United States has frozen money in the Afghan central bank to keep the Taliban from accessing it. People fear that it’s only a matter of time before airports close and borders shut down. Already there are reports of the Taliban using force to crush protests.

As of today, some families who have won the lottery in 2021 may still be able to go to Islamabad or other embassies in neighboring countries if the State Department transfers there. Many fear that in a few days, the Taliban might shut down borders completely and persecute those who tried to leave to the United States.

Since Najma, the health care worker whose father won the diversity visa lottery, heard the news of the Taliban taking over her country, she has been having nightmares. All the progress she’d made over the course of the year processing the trauma of the terrorist attack she survived has been erased, she said. 

When I talked to her on the phone last week, she broke down. “It’s very important for us to leave this country,” she told me. “It’s been a few days since the Taliban has taken over Kabul and I have been having the same feelings as I had on the day of the terrorist attack.”

The American nonprofit that Najma now works for instructed her and her colleagues to destroy all proof of identification for their safety, but the organization didn’t have the resources to extradite her.  

“I don’t know what will happen to us if the telephone or internet gets disconnected,” Najma said. “There is no guarantee.”

Najma hasn’t been able to sleep for the last few days. She worries about her future, and she worries about the future of her younger sisters, who might not be able to go to school. She worries too much, she said: “I am feeling very hopeless. I am young, but if you live in Afghanistan, you become old.”

Morrison, the immigration attorney, has been urging the State Department to intervene quickly and issue visas to the Afghan lottery winners. He’s also urging judges to intervene on the lawsuits. “To these families, it's now or never, if the judge is going to issue an order," he said. "Let's hope he issues that now.”

This article has been updated.

The Delta variant has been causing a surge in cases, hospitalizations, and deaths across the country, and public health experts have observed that areas with high vaccination rates are better protected than areas where few people are vaccinated.

Today, the 5 most vaccinated states (14M people) had

580 people in hospital, 12 deaths

In the 5 least vaccinated states (16M people)?

6,600 hospitalized, 104 deaths

Per capita , least vaccinated states have

10X hospitalizations and 7X deaths

So yeah, vaccines are working

— Ashish K. Jha, MD, MPH (@ashishkjha) August 10, 2021

The reason behind the recent uptick in vaccinations are many: Former anti-vaxxers who have become seriously ill with the virus are now urging people to get vaccinated. Vaccine mandates are becoming more common: Hospitals, government agencies, and large private companies such as McDonalds are requiring vaccinations for employees. New York City has started asking people to show proof of vaccination to dine indoors, go to gyms and entertainment venues.

As experts point out, kids have a higher a chance of being protected when adults around them are vaccinated.

Actually, while vaccination rates for kids under 12 are identical (0%) across states,

Infection numbers are not

In Massachusetts, about 250 kids under 10 years of age infected last week

In Florida, it was about 9,000

In Louisiana, it was about 1600

3/5

— Ashish K. Jha, MD, MPH (@ashishkjha) August 3, 2021

If you're one of the many Americans who recently got vaccinated, we want to hear your story. You can share it here. 

For a large chunk of last year, my then 3-year-old daughter and my husband did what a lot of parents with children did during the pandemic: They wore masks and went out for bike rides in our neighborhood in the evenings.

One evening, my daughter and my husband took an extraordinarily long time to get back. Out of curiosity, I stepped out to see what was up. Our apartment complex had more than 100 condominiums all lined up next to each other. Redwood and Cherry Blossom trees surrounded it, and there was a lot of common space for kids to run and bike around. I walked three times around the area before I called my husband. They were hanging out in a corner that I’d never been to. Under a canopy of trees, my daughter was playing “camp camp” with a neighbor’s daughter. They were collecting twigs and sticks and piling them up in one place. My daughter refused to leave. She’d found her best friend.

Since that day, every evening after we wrapped up work, we’d take my daughter to play outside with her best friend. Soon she found other best friends in the complex. Bike rides were only a ruse. She’d step off the bike the moment she saw any of the kids and start playing. After a day of screens and Zooms, she was eager to see someone her age. I could sense her emotional health improving. She felt lighter and happier after these spontaneous playdates.

This was all happening in April last year when so much about the virus was still unknown. We didn’t know that the virus was less likely to spread through surfaces. We didn’t know that kids were less likely to get seriously infected. We didn’t know that the virus was less contagious outdoors. Still, as parents, we took the risk to let our kid hang outdoors around other mask-wearing kids. We suspended our disbelief about the virus for a few hours every evening. Initially, our conversations with the other parents were awkward and shrouded in anxiety, but still, as our children played, we talked. We shared notes about what each of us was doing to get through the day. We shared homeschooling resources and information about preschool Zoom classes that had popped up. We consoled each other about our little kids learning to “mute” and “unmute” when they barely knew what a computer was. I deluded myself into thinking I was doing this for my daughter, but in the process, I was talking with adults, real people, parents who were struggling to get through the day as I did.

After a few weeks of leaning on each other for support, one of my neighbors said that our community had reminded her of her own childhood in India. I agreed with her. I remember stepping outside my apartment to play with lots of kids in common spaces in our complex as my mom made dinner. Mumbai is an overcrowded city, but one of its virtues is there is no dearth of people around you. And like every big city, it can be harsh and alienating, but to survive it, you need to find pockets of humanity. Remembering my upbringing reminded me of how much my mom invested in building a community around her. Finding people who were accessible was much more important than finding like-minded people with whom you’d like to have a beer. These don’t always have to be at odds, but when you optimize for close geographical proximity, it’s rarer to find someone whose world views exactly match yours.

Before the pandemic, I was commuting three hours to work in San Francisco. My everyday routine consisted of dropping my daughter at a preschool in the morning and in the afternoon dashing through traffic stops and delayed BART rides to make it to the 6:30 p.m. deadline to pick her up. I was completely exhausted by the time we hit the parking lot of our home. There was no bandwidth for taking her out to a park on a weekday. She’d also had her fill of friends and playtime at her daycare.

It wasn’t as if I completely ignored our neighbors before. I knew the names of their dogs and collected their packages when they were away. But during the pandemic, our neighbors became our friends. When the pandemic hit, the lockdown tied us to our geography. Our community is diverse, with people from different age groups, races, and socioeconomic backgrounds. Our next-door neighbors were a family from India. Next to them, some graduate students teamed up together to share an apartment. Next to them, two single moms were raising their children and sharing an apartment. Before the moms moved in, an old disabled man lived with his sister-in-law and brother-in-law from El Salvador, who were taking care of him. In the row of apartments opposite to ours, was an old couple who’d lived there for more than 20 years. Often our mail and deliveries would get mixed up with theirs because our apartment numbers were too similar. When we’d dutifully return the packages; we’d share a thought or two.

Many times during these brief interactions, I got a sense that their beliefs differed from ours. A neighbor commented about how lucky we were that we lived next door to a police station. I’d quietly disagreed but never engaged. Every once in a while, I’d receive the cold shoulder from a group of moms who stayed home with their kids. They’d taunt me about not having the time to invite them for dinner or attend their weekly lunch potlucks.

When the pandemic hit, suddenly, everything that separated us—from our sociopolitical beliefs to our hobbies—became smaller. We were bound together by the universal experience of the pandemic.

The pandemic gave us a way to work through many invisible fault lines. I was better able to relate to moms who stayed home all day taking care of their kids. I was able to even more appreciate the labor that went into child-rearing through the day. Stay-at-home moms in our community were aghast at how much worse it must be for me to hop on office Zoom calls with wailing kids in the background. Even though I couldn’t attend their Tuesday lunch potlucks, I was at least able to meet up with them for a few hours in the evening.

Even if I end up moving away from this apartment complex and even after the pandemic ends, I am going to spend time building a sense of community around me, to find support as I raise kids. This past year has taught me that children playing with other children is the best form of childcare, and neighbors are a low-cost solution to finding support. A community has always been humanity’s answer to survival. I would never agree with some of their viewpoints, and they’d probably never come around to seeing things the way I did, but we all acknowledged that our kids needed a community. We’d found a common language to ground us and build trust. My hope is that our commonalities may open up ways for us to work through other differences and learn from each other.

After an afternoon of play at a park in Oakland, as we were driving back to Fremont, my four-year-old daughter announced that she needed to pee.

We took the nearest exit and parked in a strip mall with a Starbucks. Restrooms were closed. We hit up Quiznos next. No restrooms there, either. Seeing us dash door-to-door, a man asked us if we were looking for restrooms and pointed to a gas station at an adjacent parking lot. “I have kids so I understand the urgency of the situation,” he said.

It would have been a bit much to walk, so we got into our car, strapped our seat belts, and went to the gas station. The guy at the counter said the restrooms at this gas station had been shut since the pandemic started last year.

I had never been madder at another parent.

We asked the guy at the gas station if he knew of a place where our daughter could pee. He pointed to Wendy’s, a two-minute car ride away. We got into our car again. Wendy’s was closed. We saw a McDonald’s nearby. My daughter and I ran inside. We found a clean restroom at last. What a relief.

My biggest beef with McDonald’s in the past has been that it had few vegetarian food choices. But today, I felt grateful, and we celebrated our day without accidents with french fries and some coffee. As my daughter dipped her fries in ketchup, I explained the importance of peeing at home before we leave for someplace.

And yet, when we were on our way to United States Citizenship and Immigration Services today for an important appointment, our daughter said again, “Mom, I need to pee.” It was a tense moment in our car. We were almost at the office, so we went to a gas station nearby. Again, no luck. Restrooms were closed. I knocked on the Dunkin’ Donuts next door. The restroom was out of order.

I could have scurried to a strip mall next door to try Starbucks, Chipotle, or Black Bear Diner, but I didn’t. I won’t tell you what happened next. Did it happen behind a bush? Or did her pants come to her rescue? Did she borrow her baby brother’s diapers for this one time? Did a kind stranger guide us to the right doors this time around? You’ll just have to guess.

Increasingly, Republican and Democratic lawmakers in both chambers—especially those vilified by Trump and his allies—have started hiring security consultants and bodyguards, upgrading their home security systems with cameras, and, in some cases, employing firms that specialize in fortifying residences with reinforced doors, bulletproof glass, and other high-end protective features. An analysis of campaign finance records by Mother Jones found that in the three months after the Capitol attack, security spending jumped 176 percent from the same period last year. Such spending is up 233 percent from the first quarter of 2019.

Prior to the January 6 attack, three-term Rep. Liz Cheney (R-Wyo.) had never spent campaign funds on security. But in the first three months of this year—as she publicly denounced Trump’s role in fueling the insurrection, a transgression for which she was deposed from her leadership post—she paid $58,000 for protective measures, hiring a security consulting firm that specializes in executive protection and retaining three former Secret Service agents who previously served on her father’s vice presidential detail.

While Cheney has continued to speak out against Trump, she said recently that many of her GOP colleagues are too fearful for their safety to risk riling up Trump’s conspiracy-crazed supporters by going against the former president.

“I have had a number of members say to me, we would have voted to impeach, but we were concerned about our security,” she told David Axelrod on an episode of his podcast. “I think that in some ways people have sort of glossed over that, but I think that’s a very important point to pause and contemplate, that you have members of the United States House of Representatives for whom, you know, security—their personal security or their family security, their concerns about that affected the way that they felt they could vote. That’s a really significant thing to say about the current state of our politics.”

Like Cheney, Sen. Pat Toomey (R-Pa.) rebuked Trump for his election fraud lies. After the insurrection, he called on Trump to resign and was one of seven Senate Republicans to vote for Trump’s impeachment in February. The cost of opposing Trump’s bid to overturn the election has been steep: During the first quarter of 2021, his campaign racked up a whopping security tab of nearly $70,000. His campaign’s first payment came on February 8, the day before Trump’s Senate impeachment trial commenced. Among other expenditures, Toomey paid $39,000 to T&M USA, a New York City–based private security and intelligence firm. His campaign filings also list a $7,300 payment to Fortified Estate, a Texas outfit that bills itself as the “leading company for bespoke, high-security hardening of residential and commercial structures” and specializes in installing panic rooms, bulletproof doors, and blast windows.

The campaign operation of Sen. Mitt Romney (R-Utah), who voted to convict Trump during his first and second impeachment trials, paid more than $43,000 to Ambolt Security Group, a Utah-based firm run by former law enforcement officers that offers services including close protection, “safe room consulting and design,” “residential escape plans,” and threat assessments. “Political uncertainty, rioting in US streets, attacks on governance are contributing to a climate of fear within our communities, the business sector, and most troubling our homes and families,” the company notes on its website. (Security is not a subject that members of Congress seem eager to discuss. Mother Jones reached out to 10 lawmakers, including Cheney, Romney, and Toomey. None responded.)

The GOP’s most lavish spenders on security are two of its biggest promoters of the Big Lie. Ted Cruz’s campaign dropped more than $74,000 on a security firm in early 2021, the most of any GOP lawmaker. (In February, when Cruz jetted to Cancun in the wake of a deadly winter storm that left much of his state without power, a reporter dropped by Cruz’s home and found one of the senator’s guards caring for the family rescue dog, Snowflake.) Cruz, who offered to argue a lawsuit seeking to overturn the results of the presidential election before the Supreme Court, first began spending heavily on security in 2020, making payments to a company that installs security cameras and safety glass. Sen. Josh Hawley (R-Mo.), whose memorable raised-fist salute to Trump supporters outside the Capitol on the morning of the attack became an iconic image of that dark day and his role in helping to inspire it, hired security consultants, to the tune of at least $44,000, following the insurrection.

On the other side of the aisle, the two biggest spenders on security are newly elected Sens. Raphael Warnock of Georgia and Mark Kelly of Arizona. Both hail from states that were (and remain) a focus of election fraud conspiracies propagated by Trump and his backers. Warnock, Georgia’s first Black senator, has spent more on security than any Senate candidate in history. Since last fall, his campaign has paid more than $245,000 to Executive Protection Agency, a security service, with $136,000 of that spending coming during the first quarter of 2021. (His fellow Georgia senator Jon Ossoff spent nearly $50,000 on protection between January and April.) The campaign of Arizona’s Kelly—whose wife, former representative Gabby Giffords, was nearly assassinated during a constituent event in 2011—has doled out $130,000 for security so far this year.

Some Democrats—especially high-profile progressives and people of color—have long voiced security concerns. Among them is one of the right’s favorite targets, Rep. Alexandria Ocasio-Cortez (D-N.Y.), who has said she feared for her life during the Capitol insurrection: "I did not know if I was going to make it to the end of that day alive.” AOC’s campaign spent $47,000 on security, including payments to Three Bridges NY, a New York company that provides personal protection to celebrities, between January and April—more than she spent during the entirety of last year. Fellow “squad” member Rep. Cori Bush (D-Mo.), who complained publicly about being harassed by the QAnon-promoting Rep. Marjorie Taylor Greene, paid $35,000 during this period.

In total, congressional campaigns and political parties spent an estimated $2.6 million on security in the first three months of 2021. During the comparable timeframe in 2019, at the start of the last Congress, Federal Election Commission filings show campaigns and political parties made $902,000 in security expenditures. In 2017, parties and campaigns together spent $2 million over the course of the entire year—less than the security costs racked up in the three months after January 6.

Members of Congress who can show they face a genuine threat can request protection from the House or Senate Sergeant-at-Arms offices, but the resources are limited, with only members of congressional leadership receiving full-time security details (Cheney was recently assigned protection by the Capitol Police). Most members of Congress have usually traveled and interacted with their constituents with only a minimal level of security—if any. Their relative accessibility has been one of the features of our congressional system. But the possible threats that lawmakers face were on display in viral videos that circulated this winter showing irate Trump supporters confronting members of Congress including Sens. Romney and Lindsay Graham (who was labeled a “traitor” for mildly rebuking Trump for stoking the insurrection). In footage of the Graham episode, a Trump supporter wearing a QAnon T-shirt declares ominously, “One day they will not be able to walk down the street. It is today.”

The attack on the Capitol has revived an old debate over to what extent political candidates can dip into their campaign coffers to fund security expenses. The FEC has rules preventing the use of campaign funds for personal expenses and has generally applied a strict interpretation of that. For years, it did not allow candidates to spend campaign funds on home security systems and upgrades under the theory that such expenditures would also increase the values of their homes.

But in 2011, following the shooting that severely injured Gabby Giffords, the FEC began to ease its rules when it granted her campaign’s request to use contributions to install a security system at her home. Since then, campaign spending on residential security systems has risen from about $7,100 in 2013 to $143,000 last year. Campaigns have spent $82,000 on security systems in the first three months of 2021 alone.

In 2017, in the wake of the shooting at a congressional softball practice that hospitalized Rep. Steve Scalise (R-La.), then–House Sergeant-at-Arms Paul Irving, citing numerous requests for protection from members of Congress, asked the FEC for more clarification on acceptable security expenses. According to Irving, threats against members of Congress were escalating—in 2016, his office investigated 902 threats. In just the first six months of 2017, it had already looked into 950. Irving testified that his office didn’t have the ability to fully investigate all threats against lawmakers, much less protect them from unknown dangers they might face. More recent numbers suggest the problem has worsened. 

In response to Irving's request, the FEC concluded that members of Congress, whether or not they were the target of a threat, could use campaign funds “to pay for the reasonable costs associated with installing (or upgrading) and monitoring a security system at Members’ residences.” The agency issued no further guidance in the years that followed. Then came January 6.

Days after the Capitol siege, the National Republican Senatorial Committee and the National Republican Congressional Committee appealed to the FEC to formally authorize the use of campaign funds for security details—something lawmakers had already started paying for in the absence of any concrete guidance from the commission on the subject.

In late March, the FEC finally ruled that members of Congress “may use campaign funds to pay for bona fide, legitimate, professional personal security personnel to protect themselves and their immediate families due to threats arising from their status as officeholders, when federal agents are not protecting the Members or the Members’ families.” 

But that decision sparked a debate of its own. Who did the FEC consider a “legitimate” security professional and what would their roles be? That is, would they protect members of Congress only from threats—or also from potentially uncomfortable encounters with citizens voicing their opinions?

As FEC Commissioner Ellen Weintraub put it during the commission’s hearing in March, “I want to ensure that the people who would be in a position to perhaps block members of the public’s access to their elected leaders would be doing it in a way that is sensitive to the First Amendment concerns.”

Lawyers for the NRSC and the NRCC argued that drawing up language to police the credentials of the guards would slow down the process of getting protection in place for members of Congress who needed it. And the FEC ultimately imposed no strict limitations on who could be paid with campaign funds to perform security. In a brief, the Democratic Senatorial Campaign Committee and the Democratic Congressional Campaign Committee—the NRSC's and the NRCC’s Democratic counterparts—contended that in doing so the FEC was "opening the door to the improper use of campaign funds to compensate fringe militia groups under the guise of legitimate security expense." The groups noted that some Republicans have known links to far-right organizations like the Three Percenters, Proud Boys, and Oath Keepers, whose members participated in the Capitol attack. (In at least one instance, members of the Three Percenters militia group provided security at a campaign event for then-candidate Marjorie Taylor Greene, although they don’t appear to have been paid.) What was stopping Republican lawmakers from employing guards associated with the very same groups involved in the assault on the Capitol?

Nick Steen, a retired Secret Service agent and former supervisor of the agency’s presidential protection division, says it’s clear that members of Congress require more protection. He also notes that it’s not practical to have law enforcement guarding every member of Congress and candidate.

“I don’t know that it is right, cost effective, or efficient to add protection of every congressional or senatorial candidate to the growing list of federal law enforcement responsibilities,” he says. But he adds that leaving security up to the campaigns raises its own set of questions. “If a model for private security is developing for that gap, then a lot of work will need to be done to set parameters and scope for those private security folks. Cost is just one aspect of it,” he says.

The questions of how much security is appropriate and who pays may be coming to a head. In May, the Capitol Police disclosed that already this year its case load had risen 107 percent from 2020 and said it had just 30 officers in its threat assessment division to respond to over 9,000 cases. (By comparison, the Secret Service employs 100 threat assessment staffers who handle 8,000 cases.) Meanwhile, Trump and his allies have stoked conspiracy theories, pinned to quixotic election audits in Arizona and other states, that he could be reinstated to office this summer, concocting a new version of the Big Lie for supporters to rally—and perhaps, plot—around. Rep. Peter Meijer, a first-term Republican from Michigan who voted to impeach Trump and recently decried members of his party for “salivating for civil war,” told Rolling Stone he foresees more political violence ahead: “I don’t put it beyond the realm of assassinations.”

Nearly two weeks ago, Ashu arrived in India, and rushed to his father’s bedside to say goodbye. His father passed away a few days later. Ashu was consumed by grief, yet he was also eager to get back to Neha and their nine- and 15-year-old daughters in the United States. But there’s a problem: He can’t. President Biden’s ban on travel from India began Tuesday night, so Ashu is effectively stranded.

Even before Ashu left, Neha said she knew in her heart that her husband may have bought a one way ticket to India.

“We both knew that this can get really, really bad for us,” she said. “Immigrant workers here in the United States of America, we’re all going through this dilemma—to take care of our sick parents in India, or to stay here just to make sure that our visa stamping doesn’t expire, so that we don’t lose what we’ve built with all the hard work that we’ve done.”

Technically, Ashu should be able to come home. President Biden’s travel ban, instituted amid skyrocketing cases of COVID-19 in India, isn’t supposed to include US citizens, permanent residents, international students on visas, or Indians on non-immigrant work visas such as H-1B who have children born in the United States.

Ashu is in that last category, but a bureaucratic hurdle stands in his way: H-1B workers, whenever they leave the United States, have to get their visas stamped at a US consulate in their home country before they can return. But right now, all the consulates in India are closed because of the pandemic. “I heard about the travel exemption, but it doesn’t do anything for us,” said Neha, who is an advocate for skilled visa workers and an outreach manager at a law firm.

What about those who need to get a visa stamped but cannot get it as the consulates are closed? DOS should restart visa stamping within US like they did till 2004. Makes no sense to make someone travel to to get a visa when he is living in for 10+ years.

— Anirban Das (@anirb_das) May 5, 2021

Neha and Ashu aren’t the only ones whose family has been separated by the travel ban. While it’s hard to estimate the actual number of Indian H-1B visa holders stuck in India, my reporting turned up dozens of stories. Earlier this week, I joined a Telegram group on Indian visa holders with 20,000 members and asked how many were in this situation. More than 200 people replied to my message on the same day, though Indian advocates say the number is likely to be in the higher hundreds or thousands.

On Telegram, the heart-wrenching stories just kept coming: a father separated from his toddler son. Children on dependent visas of H-1B holders unable to get their visas stamped and return to school. Pregnant women who couldn’t get their visa on time and had to deliver in India in the middle of a pandemic. A fiancé who flew for his wedding and is stuck. Many left the United States to take care of sick parents during the pandemic. My phone is continuously pinging with replies such as these (responses very lightly edited for clarity):

“Both me and my husband are stuck here. My husband is going to lose his job with a reputed employer as they don’t allow working from India.”

“My wife and US born kid are stuck. Cannot get stamping.”

“We’ve been stuck in India for the last six months. I got my visa stamped, but my son who is not a citizen and is on a dependent visa had his appointment cancelled. He’s very upset and depressed.”

“I traveled while pregnant as my mom and dad were both diagnosed with COVID and were admitted to hospital. My son needs an H4 visa which he can’t get because the consulates are closed. Husband is also here and can’t work remotely for very long. Our future has become uncertain.”

“I am stuck in India and two-year-old son and wife are alone in San Francisco.”

“We came since my father passed away, but our appointments were canceled last week We can travel with a US born kid, but how can we travel without a valid visa? My elder son studies there and his education has also gone for a toss.”

“We are stuck in India because of stamping and our company can’t hold the job for long…we could lose our jobs. We have mortgages.”

According to a 2020 report from the United States Citizenship and Immigration Services, there are around 500,000 workers on H-1B visas in the United States Most of them work in the technology sector, though there are also many doctors, pharmacists, consultants, and marketing professionals. Around 70 percent of these visas are held by Indians.

Because of a Trump administration policy, many H-1B workers were already in India before the ban went into effect. In 2020, during the last year of his presidency, Trump ordered a temporary suspension on all new H-1B visas. For many months last year, US consulates in India were also closed due to the pandemic. Both of these things together made it difficult for H-1B visa holders to travel back and forth to India.

The Trump-era ban expired at the end of March, and the situation was supposed to improve. The Covid situation in India improved for a little bit earlier this year, and consulates opened for a short while and resumed processing visas, but with this new travel ban and an escalating pandemic, those who have been stuck since last year have no way of getting back anytime soon.

Another complicating factor is the absurdly long waitlist for green cards: over to 50 years, according to estimates from the CATO institute. Every year, thousands of H-1B holders qualify for green cards, yet there’s a cap on how many green cards can be issued to each nationality. As of May 2018, nearly 600,000 workers and their families were waiting for employment-based green cards. More than 90 percent are Indians, who receive about 10,000 employment-based green cards a year. Meanwhile, they have to rely on visa renewals and stamping of the visa at consulates at regular intervals.

“Without a green card, there is always an invisible sword hanging over my head,” said Neha, who has been stuck in the green card backlog since 2012.

Immigration advocates believe the travel ban isn’t worth the stress it’s causing families. Cyrus Mehta, a New York based attorney who specializes in immigration law, said that the ban isn’t very useful in stopping the spread of the virus since citizens, permanent residents, and others are still allowed to travel. But the ban disproportionately affects those who are working on temporary visas such as the H-1B visa.

“Representing H1B visa holders, I know what they’ve gone through,” he said. “Each time they go to India, there’s a ban imposed on them. In the Trump administration, they were subject to bans. Then they wait patiently. They’ve now scheduled a visa appointment. And their appointment for later this week has been cancelled because of this latest COVID ban on India. So they’ve got a double whammy.”

Mehta suggests stricter controls and protocols such as rigorous testing, quarantining, and vaccination requirements as a better approach to controlling the spread of the virus.

Neha has a full-time job, two kids and is recovering from surgery. For the past few weeks, Neha has spent all her free time calling consulates, politicians and speaking with journalists to find a way to get her husband back into the country.

“I haven’t slept a wink,” she said. “I am not a religious person, but these days I am praying to god everyday.”

Everything was going according to plan—we’d started looking for flight tickets. About six weeks ago, my dad got his first dose of Covishield, the Indian version of the Astrazeneca vaccine, in Mumbai, India. COVID cases and deaths had started slowing down, and my conversations with my dad revolved around how India, due to its population density, hot weather, or some other unknown factor, had successfully dodged the worst of the virus.

India’s health minister, Harsh Vardhan, announced early March in a medical press conference in the country’s capital, “We are in the end game of the COVID-19 pandemic in India.” He assured us that India had an oversupply of vaccines and was ready to distribute them to the rest of the world.

Other family members and neighbors in our apartment complex in Mumbai had already received their doses. Although my father was anxious to meet his grandson, he dilly-dallied getting his first shot. He wanted to wait and watch the side effects of the vaccine and wrap up some of his board meetings before he could take time off.

Then, a few weeks ago, everything changed. COVID cases began to creep up in India, then they skyrocketed within a span of days. My dad got his second shot this week, under very different circumstances. For days, he’d wake up in the morning and visit four different vaccine centers scattered across the city, only to hear that they had run out of doses. Finally, on Monday, he queued up at the center early in the morning and waited for four hours along with what he estimates to be thousands of people.

India now has the highest number of reported coronavirus cases in a single day worldwide since the beginning of the pandemic. On Tuesday, India reported more than 300,000 cases and more than 2,000 deaths. These numbers are an official count—experts say the real death toll could be five to 30 times greater. Hospitals have run out of beds, and people have died on their doorsteps. Cities are making makeshift mass graves to accommodate the death toll from the virus. So many parents have died during this wave that the government has created a special hotline for children orphaned due to the virus to provide resources.

With crumbling infrastructure and a failing healthcare system, people have turned to social media platforms to find ICU beds and oxygen tanks. Actors, activists and other high profile celebrities are using the platform to connect people to hospitals, deliver food and provide information about oxygen. India’s soccer team captain Sunil Chhetri even allowed others to use his Twitter handle to amplify voices of those in need.

Hey, everyone. As promised, I'll be handing my Twitter account to some of the real-life heroes over the coming weeks, to try and help them get their messages of support and relief to a wider audience.

— Sunil Chhetri (@chetrisunil11) April 29, 2021

URGENTSOSOXYGENDELHI

Lives of 175 patients at stake in #VIMHANS NAYATI Hospital, Lajpat Nagar, New Delhi!

Less than 4 hours of Oxygen supply is left. Still awaiting fresh supplies from INOX

HELP! @CMODelhi @PMOIndia @msisodia@drharshvardhan@SatyendarJain pic.twitter.com/gB84pdDlBB

— Shoaib Daniyal (@ShoaibDaniyal) April 24, 2021

A friend of mine said that every family WhatsApp group has turned into a COVID support group to exchange information about cases in the neighborhood, distribute food for those in need, and access information about tests and hospital beds. Germany, France, the United Kingdom and the United States pledged aid in the form of oxygen cylinders and ventilators. After celebrities, researchers and vaccine makers complained of vaccine hogging, the US finally agreed to lift the ban on exporting some raw materials required for making the vaccines.

Dear @potus, @JoeBiden – the US has something like a vaccine glut. India is in dire straits. Please overturn this export ban ASAP. https://t.co/tOSFGRbN6O

— Salman Rushdie (@SalmanRushdie) April 24, 2021

Respected @POTUS, if we are to truly unite in beating this virus, on behalf of the vaccine industry outside the U.S., I humbly request you to lift the embargo of raw material exports out of the U.S. so that vaccine production can ramp up. Your administration has the details.

— Adar Poonawalla (@adarpoonawalla) April 16, 2021

Technology companies with Indian-origin executives, such as Google and Microsoft, pledged financial support.  Under pressure, the United States agreed to donate unused doses of the AstraZeneca vaccine to India as vaccine shortages surge in the country.

The details of why the situation in India got so bad so fast are a little fuzzy. The surge in large part may have been fueled by relaxing measures too soon. Everything in the country opened up in February after vaccinations arrived: there were large, unmasked election rallies, mass gatherings for religious festivals, cricket matches in stadiums filled with spectators. India has fully vaccinated only 2 percent of its population. We know that the new B.1.617 variant is potentially more transmissible and lethal, and we’re not sure how well the available vaccines work against it.

The part that I am still trying to wrap my mind around is how unpredictable this virus has been. It feels like just yesterday that we started making plans for my dad and my partner’s parents to visit. Now, it would be almost impossible for them to get here: On Friday, the Biden administration announced that the United States will severely limit travel from India starting early next week. 

When I talked to my dad last night, he told me that the Indian government has been paralyzed, but people are turning to each other to help. It reminded us of a time 15 years ago when Mumbai was heavily flooded. All Mumbai residents—”Mumbaikars” in Hindi—remember exactly where they were and what they did on that day. And even before there was social media, residents helped each other out. Gymnasiums were converted into makeshift canteens that fed hundreds. People opened their homes to strangers who were stranded and offered them food and a cell phone so they could tell their loved ones that they were doing okay. My dad called us using one of those borrowed cell phones.

Media and politicians applauded the city’s resilience, and I bought that story back then. But now I wonder, what other choice did the people have? Helping each other out in the face of failing infrastructure is the only shot a poor country has at survival.

It feels like my heart is split and my mind is processing two different realities. As more and more people get vaccinated in the United States, I feel palpable hope that we’re inching closer to a post-pandemic world. But I also see how my future for this year looks so starkly different from my family’s life in India. My dad has hardly left his apartment building for weeks now. His everyday act of bravery is double masking to take a walk to keep his sanity intact or buy groceries so he can stay alive. Despite these stark differences in our realities, we both desperately want the same thing. We both hope that in a few months, somehow, we’ll be able to meet and he’ll be able to hold his grandchildren in his arms.

As the state races to vaccinate residents and reopen its economy and schools, it appears that, like the beginning of the pandemic, Black communities across generations are getting infected and dying of COVID at higher rates than their white peers, according to a Mother Jones analysis of state data. Likely because older residents received the early share of vaccinations, this surge has unevenly afflicted younger residents—young Black men and women in particular.

Last March, Black Michigan residents, who made up 14 percent of the state’s population, accounted for 41 percent of deaths. A year later, as of April 12, Black Michiganders account for 22 percent of deaths.

The latest outbreak is especially acute among children and young adults. A Mother Jones analysis found that in Michigan, residents between 20 and 39 years old account for the highest case rates statewide, and people under 19 years old have contracted COVID at a rate higher than any other time during the pandemic.

Younger Black residents are also dying at disproportionately higher rates than white residents, taking years of life from an entire generation of Black Michigan residents. In the last year, Black Michiganders between 20 and 39 years old account for around 17 percent of the state’s Black residents but 44 percent of deaths. It reflects a startling national trend that Black and Latino Americans throughout the pandemic have died of COVID at younger ages than their white peers.

Because of the surge, in mid-April, Governor Whitmer called on schools to pause in-person learning and youth sports and pleaded with residents to avoid eating indoors at restaurants and bars. Her power to impose stricter restrictions was limited, thanks to a state Supreme Court ruling last October curtailing her executive authority.

On NBC’s Meet the Press this week, Whitmer conceded that she lacked the “exact same tools” she had a year earlier after being sued by Republicans. “At the end of the day this is going to come down to whether or not everyone does their part,” Whitmer said on Sunday.

But for some cities, the race for vaccinations and personal responsibility plea only goes so far. Dr. Lawrence Reynolds, a pediatrician in Flint who has been chief health adviser to Flint mayor Sheldon Neeley since last March, points to the state’s dedication to “home rule,” the notion that every school district and city government operates on its own, even as the state offers resources and direction. Reynolds, who sits on the Greater Flint Coronavirus Taskforce on Racial Inequities, saw Flint’s high coronavirus burden as an example of how home rule can perpetuate segregation and the unequal allocation of resources. That, along with Republican resistance against Whitmer, creates a “recipe for moving backward and not responding appropriately” to the pandemic, he added.

Flint is in Genessee County, which has a daily case rate of 89.2 per 100,000 people—the second-highest rate of all metropolitan areas in the United States over the last two weeks, according to the New York Times. During the pandemic’s early days, Black Flint residents, who make up 53 percent of the city’s population, were nearly four times more likely to get infected by COVID than their white peers and six times more likely to die, according to county data provided to Mother Jones.

By the end of last summer, through a concerted effort to eliminate testing barriers, expand community testing sites, impose a moratorium on water shutoffs, and leverage pre-existing networks of community groups, activists, church leaders, public health officials, philanthropists, and business leaders for aid, among other initiatives, the city managed to close that gap in infections and deaths between white and Black residents. So did the rest of the state: An interim report from the state’s Coronavirus Task Force on Racial Disparities noted that in March and April 2020, Black residents averaged 176 new confirmed cases for every million residents per day—a stark departure from 39 per day for white residents.

By September and October, the trend nearly flipped: Black residents accounted for 59 cases per million while white residents made up 130 per million. Nationally, although disparities in COVID infections and deaths among communities have narrowed over time, Black, Latino, Native American, and Pacific Islander people are still dying of COVID at at least twice the rate of white people when adjusted for age, according to APM Research Lab. 

But sadly, much of that progress has been undermined in the most recent surge. As COVID sweeps through cities shaped by pre-existing inequities and political division, it takes advantage of structurally unequal systems that disproportionately afflict communities of color, many of whom are so-called essential workers who risk exposure in ways their peers don’t. The trajectory in Flint and the surrounding county is no different. Even the vaccine rollout has unfolded unequally. Though Black residents account for 14 percent of the state’s population, just 8 percent of all vaccinated people in Michigan so far are Black as of April 12. White people, who make up 80 percent of the population, account for 79 percent of vaccinated Michiganders.*

Debra Furr-Holden, a Flint epidemiologist and professor of Public Health at Michigan State University, told Mother Jones that the state’s decision to reopen in February, just as the B.1.1.7 variant collided with the state’s hopes for returning to normalcy, was a “big mistake.”

“It’s a really unfortunate state of affairs—despite people’s desire to return to business as usual, and what many are calling the new normal, we weren’t ready,” she says. “We didn’t give enough time for vaccinations to outpace spread and the impact of variants and communities. So now we are where we are. I’m pretty sure we ruined summer.”

This article has been updated to correct a mischaracterization of the racial breakdown of fully vaccinated Michigan residents.

We’d rejoice and dance when the assistant played the right number, which happened about half the time. My daughter was going to a Bollywood dance class and we’d finally found a use for the device that my husband had won at a tech conference.

Often, however, it would mishear our requests and play something else. My daughter and I would look at each other and chuckle, like the only people in the room who got a joke. We’d roll our eyes and bond over our assistant’s incompetence. These moments turned out to be funny and special, and secretly, I enjoyed the role reversal of having an assistant who sounded like a stereotypical American. When would that happen in real life?

Yet several days into our routine, I noticed something strange. My daughter and I were contorting our mouths to pronounce the names of Bollywood songs with an American accent. I don’t know if our exaggerated Midwestern accents improved Google Home’s hit rate or if we were doing it unconsciously so we felt like we were being understood. Either way, the gadget that had entered our house as a helper had turned into an intruder. Not just an intruder that could listen to our private conversations, but an intruder that was telling us how we should speak our own language in our own home. I’d been wrong about our reversed power dynamic.

My hunch was confirmed when I spoke with Halcyon Lawrence, an assistant professor of technical communication and information design at Towson University who studies user accessibility and design for voice recognition systems such as Google Assistant, Amazon’s Alexa, and Apple’s Siri. “Your daughter is being disciplined by Google Home. You are being disciplined,” she told me. This artificial intelligence–­powered machine, she explained, either understands users’ accents based on its programming or it doesn’t. If it misunderstands something, it just assumes it knows what it’s hearing and powers through its mistake. It’s essentially a one-way feedback loop where humans must change their behavior to make the machine run more smoothly.

“If I am going to use this technology then I must assimilate. I must code-switch,” Lawrence said. “I find there is something inherently violent about that, because it is no different than the kind of language discipline that we faced when we were colonized.” Like most postcolonial English speakers, I float in an in-between land of languages. I speak four Indian languages and I speak English fluently. Yet my accent and dialect are seen not as marks of erudition or class like British accents, but as punchlines that reinforce stereotypes. (Think Apu from The Simpsons.)

Lawrence’s own experiences being misinterpreted because of her Trinidadian accent inspired her to study how voice recognition systems embed “accent bias.” Linguistic studies have found that nonnative English speakers and people who don’t have standard American accents, particularly immigrants from non-­European countries, are penalized in the job and housing markets because they are perceived to be less intelligent and less competent. Vice President Kamala Harris has described how people would assume that her mom, who had a PhD in nutrition and endocrinology, was unintelligent because of her Indian accent.

Switching your personal digital assistant’s accent won’t affect its ability to understand yours. But it will tell you more about who it thinks it’s talking to. The most popular services—Google Assistant, Alexa, Siri, and Microsoft’s Cortana—can speak in a range of languages, dialects, and accents, with notable exceptions. Alexa has only one version of a standard American accent. Siri has American, British, Irish, Indian, Australian, and South African accents. Google Assistant speaks with one of eight color-coded American-accented presets as well as “British Racing Green” and “Sydney Harbour Blue.” None of the assistants, however, offer any regional or ethnic American accents or African American Vernacular English, a dialect with its own accent and unique grammatical features. Google Home, however, does have a limited-time “cameo” voice appearance by com­edian Issa Rae (who was preceded by John Legend). Alexa features Samuel L. Jackson.

But these celebrity voices aren’t really about functionality—Jackson can offer his opinion on snakes but isn’t programmed to help you with your shopping. Miriam Sweeney, an associate professor at the University of Alabama’s school of communication who studies voice assistants, told me that the virtual Jackson and Rae are further examples of technology companies using Black voices to entertain white consumers while ignoring Black consumers. A recent study by computer scientists and linguists at Stanford University found that all the major speech recognition systems routinely misunderstood users who speak African American Vernacular English at almost twice the rate of their white counterparts.

Google told me that fairness is one of its “core AI principles” and that the company seeks to make its digital assistants accessible to as many people as possible. “From day one, we’ve strived to build an inclusive product that can be helpful for all users and equally serve them,” Beth Tsai, the director of Google Assistant policy, told me. Much as it doesn’t specify the gender of its assistants’ voices, Google seeks to make its default American voice raceless. “Labeling a voice—even if it’s recorded by a Black actor—as a ‘Black voice’ defines what a Black voice sounds like…Similar to gender, voices for people of different races are really diverse. We’d be doing a disservice and be leaning into stereotypes if we applied those labels,” Tsai said.

Amazon told me something similar. “Alexa’s understanding of different languages, dialects, and accents is of the utmost importance to Amazon and our customers,” said a spokesperson for Amazon. “Alexa has been designed to work well for everyone, and our speech recognition models work with many different dialects and variations in speech. We continuously improve our models in order to accurately recognize variations in speech.”

Even as the tech companies boast that their voice recognition products are accessible to a wide range of users, they add that the technology is still difficult to develop, which partly explains why they haven’t introduced more accent options. Even as they say African Americans are an important consumer market, some in Silicon Valley argue that market dynamics dictate which accent and language options are available. These arguments aren’t entirely convincing. All four digital assistants offer Italian, which is spoken as a first language by 63 million people. (Siri even comes in Finnish, which has about 5 million native speakers.) Yet only a couple assistants offer Swahili, Telugu, and Marathi—languages with nearly 100 million speakers each.

It’s nothing new for companies to exclude consumers of color, says Safiya Umoja Noble, an associate professor of information studies at UCLA and the author of Algorithms of Oppression, an investigation into how search engines reinforce racial biases. For years, consumers of color have heard that “they are not a market that matters the way high-end luxury, middle-class, and affluent consumers matter,” she says. “The whole history of advertising in the United States has been about prioritizing people who don’t have accents, as if there is some type of neutral space of language, which of course we know is absurd.”

Another likely explanation for the digital assistants’ limitations is that they reflect their creators’ blind spots. An analysis that I did for Reveal of 177 large US technology companies found that in 2016, 73 percent of their executives and senior managers were white, 21 percent were Asian (including South Asian), 3 percent were Latino, and 1.4 percent were Black. (In December, a prominent Black AI ethics researcher claimed Google fired her after she co-wrote a draft paper about the risks of text recognition systems reinforcing racial and gender biases.)

However, both Noble and Sweeney think it may be a good thing that voice recognition devices aren’t trained to recognize many accents of marginalized groups, effectively stymieing the devices’ primary function, which is collecting users’ personal data. Sweeney tells her students to throw away their smart speakers, and Noble refuses to buckle under pressure from her 9-year-old to buy one. When we use these technologies, she said, “we teach our kids that our voice isn’t the normative voice, that they have to be something else in order to engage, in order to participate, in order to find themselves.”

Lawrence notes that when voice recognition technology misinterprets people of color, it’s not simply inconvenient but sometimes harmful. A ProPublica investigation into “aggression detectors” that are being installed in schools found that they incorrectly identified kids’ voices as aggressive even when they were saying completely innocuous things. On the other hand, she noted, if Black people don’t use digital assistants, they will continue to be overlooked and misunderstood by tech companies, which will keep baking biases into their algorithms. “For how long will we be spared not being recognized? In my mind, it’s a no-win situation,” Lawrence said.

Within my home, I found a small, twisted way to win. I was excited to learn that Google Home has a setting called English (Indian), which speaks Indian-accented English. I also selected the Hindi language option, though I quickly became frustrated because it only understood a formal Hindi that is totally unlike the version we speak at home. For the first time in my life, I had to look up the Hindi word for “movie.” The English–Indian assistant didn’t understand me any better than the default setting, but at least it didn’t make me feel like I had to imitate an American accent.

Talking to someone—or something—that sounded like us did improve our experience, even if it didn’t completely reflect our multilingual reality. I noticed that my daughter wasn’t code-switching her English anymore. And every so often, we’d google phrases to speak in pure Hindi with our assistant. And we’d look at each other and laugh at how abnormal our nonconversational Hindi sounded.

It was clear that we didn’t have a perfect substitute for the polished product aimed at “American” speakers. When my daughter would ask the American-accented assistant to tell her a story, it would regale her with fairy tales like “Cinderella” or “Hansel and Gretel.” Yet when she asked the Hindi assistant for a story, it revealed a hole in its programming—not a serious functional issue, but a revealing sign of the lack of imagination that had gone into it. It replied, “Once upon a time there was a king and once upon a time there was a queen. They both slept and that’s the end of the story.”

Among the first results on a cursory Google search is usually GreatSchools, a nonprofit site that ranks public schools nationwide and feeds that information to real estate sites such as Redfin, Zillow, and Realtor.com. GreatSchools is easy to find, but its ratings correlate closely with students’ racial or economic backgrounds. “The schools that get high marks and diverse schools tend to not be the same ones,” Tringo says of the exams GreatSchools uses to produce its score. She recalls going to an eighth grade graduation for a client’s kid at Winter Hill Community Innovation School in Somerville, just outside of Boston, and marveling at the friendly atmosphere among the middle schoolers and the diversity of the graduating class. “[It] was just amazing,” she says. “But you wouldn’t get that from looking at rankings.”

Winter Hill has one of the most diverse student bodies in Somerville: Half of its students are Latino and 13 percent are Black. Yet with a rating of 4 on a scale of 1 to 10, GreatSchools deems Winter Hill “below average.” Its students’ poor test scores drag its rating down, despite evidence that they’re making as much academic progress as most students in Massachusetts. Buyers looking for homes near highly rated schools might skip right over the Winter Hill neighborhood and look closer to one of Somerville’s other schools or in another town altogether.

There’s evidence that GreatSchools’ ratings are exacerbating racial segregation, not just within school systems but in the communities around them. “What makes GreatSchools popular is partly that they’re linked to real estate sites, which is partly what makes them dangerous,” says Sean Reardon, an education professor at Stanford University who studies poverty and inequality. “They start to overtly link people’s residential choices to what seems to be a measure of school quality. While that makes lots of sense if it’s a high-quality metric of school quality, if it’s more of a measure of socio­economic composition of schools, then it runs the risk of creating incentives for more socioeconomic segregation.”

If you don’t have kids or haven’t been looking for a house recently, you may not have heard of GreatSchools, but its influence is profound. Its ratings are incorporated into popular real estate websites, which pay GreatSchools to use its data. It is supported by deep-pocketed philanthropies, especially ones with ties to the charter school movement. GreatSchools says its site attracts 45 million people a year; by comparison, there are about 40 million households with children in the United States.

Due to the pandemic and the reenergized movement to fight for racial justice, the inequities between school districts have been thrown into sharp relief: Some can easily afford to implement distance learning while others lack the resources to effectively teach online. Institutions of all sorts are being pushed to reexamine their roles in reinforcing systemic racial divides, and GreatSchools, which had already been reevaluating aspects of its school-rating metrics, appears to be doing the same. “Racial segregation and racist policies are woven into our education system and everyone involved has to do better to create more equitable outcomes for all students, GreatSchools included,” CEO Jon Deane told me in an email. But Deane added that he still believed GreatSchools can be an important tool for parents. “We feel strongly that it’s a parent’s civil right to have access to information about their schools.”

Bill Jackson got the idea for Great­Schools in the early 1990s, when San Francisco’s school superintendent wanted to know if a new restructuring program was working. He dispatched the twentysomething Jackson, who had a few years of teaching experience and a public affairs fellowship, to investigate. Jackson concluded that the rosy reports that principals were feeding back to the superintendent were mostly untrue. “This experience led me to believe that parents and the public—not to mention school administrators—needed more and better information about school quality,” Jackson later recalled.

In 1998, after working for a tech startup and a voter information website, Jackson launched GreatSchools. The nonprofit expanded rapidly after President George W. Bush signed No Child Left Behind four years later, requiring states to implement standardized testing and publish the results. GreatSchools quickly went from evaluating schools in two Bay Area counties to becoming a de facto grading system for all public schools. In 2011, as it added an unprecedented number of schools to its database, it brought in $10.8 million. (In 2018, the most recent year for which data is available, its revenues were $7.3 million.)

GreatSchools now maintains more than 100,000 school profiles, one for nearly every public school in the country. Its data, pulled from states and the federal government, includes test scores, teacher-to-­student ratios, discipline rates, and students’ racial backgrounds. The site also includes user reviews from parents and community members, though their input doesn’t affect a school’s overall rating. GreatSchools lets schools add information to their profiles, including schedules, uniform policies, and what kinds of classes and extracurricular activities they offer.

But any nuance is overwhelmed by the single score featured prominently at the top of every profile. A rating of 1 to 4 indicates a school is “below average”; 5 or 6 signals it is “average”; and 7 to 10 means it is “above average.” Those scores are generated using complex formulas that rely heavily on standardized tests. Yet education researchers and advocates worry about evaluating schools based on their test scores. The tests, they say, measure surprisingly little of what students learn; have debatable abilities to predict kids’ future success; and produce results so highly correlated to students’ racial and economic backgrounds that they are, essentially, demography in disguise.

Since 2017, GreatSchools has tried to make its scores take into account whether students across demographic groups are falling behind their peers both within the same school and in the rest of their state. In August, GreatSchools rolled out a major overhaul of its system in California and Michigan, which produced new ratings that give less weight to raw test scores and more importance to equity and year-to-year improvements on test scores. GreatSchools plans to expand that approach nationwide. (Three days after this story was published online, GreatSchools announced that it was implementing the new ratings across the country.)*

But any radical changes to Great­Schools’ ratings would require states and the federal government to provide additional data for all schools in a consistent fashion. The appeal of Great­Schools ratings, after all, is that they provide a uniform way to compare schools across the country.

For now, GreatSchools’ heavy reliance on test scores—and other measures that are highly correlated with race, like graduation rates and Advanced Placement test performance—means homebuyers looking at its ratings don’t have to harbor any racial animus to steer clear of neighborhoods with sizable Black or Latino populations. They just have to use GreatSchools’ filters on real estate websites to skip over listings in areas with poorly rated schools. Since homebuyers who can afford to move into areas with highly rated schools are largely white and Asian, the scores could reinforce the separation of neighborhoods along racial lines.

In 2018, two business professors, Sharique Hasan of Duke University and Anuj Kumar of the University of Florida, published a study that examined what happened to neighborhoods in 19 states and Washington, DC, after GreatSchools ratings were introduced. Their analysis took into account local real estate trends and recognized that homes near schools with better test scores were already priced higher before GreatSchools rated them. When the researchers updated their findings this spring by adding seven more states, they found that the gap between average home prices near schools with better ratings increased by more than $16,300 after three years compared to those near average-rated schools. And those areas close to highly rated schools attracted more white and Asian residents. Near lower-rated schools, the gap in property values also decreased by roughly $16,000 over three years, and white and Asian residents left. “Broader access to information increased segregation because high-income families could more readily leverage school ratings to move to neighborhoods with better schools,” Hasan and Kumar wrote. “In this case, knowledge was indeed power, but only for the powerful.”

While residential segregation can lead to school segregation, the inverse is also true. A major reason for that is a 1974 Supreme Court ruling that school desegregation efforts cannot stretch across district lines, except in the most egregious cases. But even within districts, boundary lines that dictate which schools kids attend can skew neighborhood demographics. For decades, parents have used the racial breakdown of schools as a proxy for assessing their quality. Ann Owens, a sociology professor at the University of Southern California, found that households with children live in more segregated areas than those without children. “As long as neighborhoods are demarcated by school district boundaries limiting enrollment options, parents will take these boundaries into account when making residential choices, which may contribute to segregation,” she wrote.

Last year, journalists Matt Barnum and Gabrielle LaMarr LeMee of the education news site Chalkbeat concluded that GreatSchools’ ratings “effectively penalize” schools serving higher proportions of low-income, Black, and Latino students. The Chalkbeat writers noted that GreatSchools’ own data showed that many of these schools were doing a good job helping students learn math and English. Yet they still had a tough time getting above-average ratings on GreatSchools. “The result,” the reporters wrote, “is a ubiquitous, privately run school ratings system that is steering people toward whiter, more affluent schools.”

GreatSchools says its ratings are not the final word on schools’ quality, but a starting point for parents to investigate further. But despite requests from Kumar, it won’t provide data on how many users scroll past the top number to read more nuanced information. Kumar says his study’s results fly in the face of hopes that equal access to information will promote equality, a premise that remains at the heart of GreatSchools’ mission. “I personally feel their intentions are noble,” he says. “But what we are seeing is, there are often unintended consequences.” Even if all parents had the same access to information, he says, they do not all have the same ability to put that information to use.

Deane, GreatSchools’ CEO, told me that the conclusions reached by Chalkbeat and Hasan and Kumar “overlook, or at best minimize,” the systemic racism that has “steered our society for decades.” “It’s very important to look at underlying issues and policies that segregate communities and create opportunity gaps, including school boundary designations, property taxes and the relationship to school funding, discipline and safety policies, among others,” Deane wrote in an email. “If we are to address the fundamental inequalities that still exist in America’s schools, we must respect the right of all parents to access critical information about their schools. This is GreatSchools’ mission.”

GreatSchools management has been aware of these issues for years. Jackson, who left the organization in 2016, says that as early as the mid-2000s, he and others at GreatSchools had worried that assigning school ratings could exacerbate residential segregation. “People are absolutely using [GreatSchools ratings] to find other people like them and to sort economically and sometimes racially,” he says. But he recalls that GreatSchools’ management concluded that self-sorting wouldn’t go away even if GreatSchools did. People would find the same information elsewhere, either from competitors or government websites.

It appears that internal conversations about these issues were limited. The group’s board never formally discussed them, Jackson says. Education writer Leanna Landsmann, who was on the board for 14 years and also served as the board chair, told me flatly, “I don’t remember any discussions ever about GreatSchools resegregating neighborhoods. Ever. I would have remembered that. I came out of the civil rights era.” Eric Hanushek, a prominent education researcher at Stanford University and former GreatSchools board member, says he remembers conversations about what kinds of families would use the site, but not about how ratings could affect residential segregation. (Several board members declined to comment or did not respond to requests for comment.)

The way GreatSchools hopes people will use its rankings has shifted over the years. In 2011, the organization stated that “informed school choice” was one of its top missions. At the time, Jackson told me, charter schools and school choice had bipartisan support, and President Barack Obama’s administration promoted them as well. Now that charter schools have lost favor among Democratic leaders—largely because of concerns that they siphon resources from traditional schools and undermine teachers’ unions—GreatSchools has shied away from explicitly endorsing them.

Nevertheless, GreatSchools’ biggest donors read like a Who’s Who of the school choice movement: the Walton Family Foundation, the Bill & Melinda Gates Foundation, Bloomberg Philanthropies, and Arnold Ventures. When I asked representatives of these organizations whether they thought GreatSchools helped boost the case for charter schools, they did not answer directly. “We are longtime supporters of the organization,” says Marc Sternberg, the Walton Family Foundation’s K–12 education program director, “because of our belief that knowledge is power, and the more clear, accessible information parents have about school options, the better.”

The Walton foundation has given GreatSchools more than $23 million since 2009. In 2011, the foundation’s $4.8 million grant made up almost half of GreatSchools’ revenue. The Gates Foundation, which has given $9.7 million to GreatSchools (most of it more than a decade ago), pointed out that its most recent grant supported an effort to track how high school graduates fare once they get to college. “Our investments in GreatSchools’ rating systems,” senior adviser Bill Tucker says, “are grounded in our belief that parents deserve and value information about how their schools are performing.”

Along with representatives from tech and finance firms, people with private and charter school backgrounds still feature prominently on GreatSchools’ board of directors. Deane, who became CEO in early 2019, has a long background in charter schools. He briefly worked on school issues at the Gates Foundation and spent three years as the director of education for the Chan Zuckerberg Initiative, which has funded charter school programs. Earlier in his career, Deane worked as a math teacher and administrator in Bay Area charter schools. But he says he sends his children to a public school in San Francisco (it gets an 8 on GreatSchools). The organization, Deane says, is “agnostic” about what kind of school parents send their children to.

“Schools are not commodities that can be shopped for like cereal,” says Jack Schneider, an assistant professor of education at the University of Massachusetts, Lowell, who has become one of Great­Schools’ fiercest critics. (The Chan Zuckerberg Initiative has funded some of his research, too.) In 2017, he wrote a piece for the Washington Post castigating GreatSchools and the real estate websites that use its ratings. His daughter’s school in Somerville, he noted, received a 6 from Great­Schools while schools in nearby towns scored much better. “These towns, no doubt, have excellent schools,” he wrote. “But are they better than ours? There’s little evidence of that. The main difference, it seems, is that they are whiter and more affluent.”

The Boston area is a great illustration of how an overreliance on test-based evaluations can profoundly disrupt communities, whether it results in the state shutting down “underperforming” schools that serve Black and Latino students, or wealthy residents driving up suburban real estate prices as they look for the “right” school attendance zone.

School ratings are only one factor behind segregation in the Boston region. White racial animosity is so ingrained in the area that Stony Brook University professor of Africana Studies Zebulon Miletsky calls it “the Deep North.” That hostility exploded in the mid-1970s, after federal Judge W. Arthur Garrity ordered about 17,000 Boston students to change schools in an effort to desegregate them. White parents clashed with police in front of TV cameras as the integration plan was carried out. The “anti-busing” protests had long-­lasting effects: They weakened the resolve of Northern politicians to carry out the types of desegregation efforts Southern cities had gone through decades earlier, and they led to white flight from Boston to its suburbs.

The desegregation efforts ultimately did improve race relations in Boston, Miletsky says, but disputes about racial integration have flared up again. In 2014, Boston replaced the desegregation-era plan with one that prioritizes allowing children to go to nearby schools. Because of these changes, researchers have found that students in predominantly Black and Latino neighborhoods south of downtown now have fewer top-tier schools to choose from, face more competition to get into those schools, are less likely to attend them, and have to travel farther to get to them. The researchers pointed out that no students in the predominantly Black neighborhood of Mattapan lived within 1.5 miles of any of the district’s best-rated schools. Not a single one.

Ruby Reyes, the director of the Boston Education Justice Alliance, says test scores, which GreatSchools relies on for its assessments, set off a chain reaction that schools struggle to recover from. The cycle is exacerbated by gentrification that has pushed lower-income residents out of their neighborhoods or the city completely. At the same time, charter schools have poached students, particularly in predominantly Black and Latino neighborhoods, further draining public schools short of funds. The result is a “culture of fear and scarcity” in schools with low test scores. “It’s this vicious cycle of [schools] losing kids in enrollment and then their budgets get cut, and so services are lost,” Reyes says. “Then test scores go down and it becomes a ‘bad school,’ and nobody wants to send their kids to bad schools.”

Residents’ fears of sending kids to “bad” schools appear to be fueling “a growing mismatch between the demographics of kids who attend Boston’s K–12 public schools and the city overall,” the Boston Foundation reported in January. “The families who leave Boston when their kids approach kindergarten are predominantly middle and high income. Today, almost 8 in 10 students remaining in Boston’s public schools are low income…and almost 9 in 10 are students of color.”

Ultimately, Schneider says, families just want an easy way to assess the true quality of schools, and the simplified score GreatSchools offers doesn’t give much insight. “It’s not GreatSchools’ fault that the data wouldn’t answer that question for this family,” he says. “But if their theory of change is we should provide good information to people so that they can make decisions about where to send their kids, then you ought to ensure that you have the information that people need.” “Schools are not uniformly good or bad. Schools have different strengths and weaknesses.”

To find a better answer, Schneider partnered with eight school districts in the state to create an alternative. The Massachusetts Consortium for Innovative Education Assessment (MCIEA), which started in Somerville in 2016, now includes a range of districts, including Boston, wealthy suburbs, and outlying towns. It is developing its own evaluation system, for both schools and individual students, that is meant to replace standardized testing.

Most school rating systems don’t ask people what they actually want out of their schools, Schneider says. So that’s where he started. When he and other researchers talked to teachers, parents, administrators, local leaders, and students, they heard a list of concerns that aren’t reflected in GreatSchools’ ratings or the data that governments collect. For example, people said they wanted qualified teachers who connect with students and stick around long enough to be part of the community. They wanted rigorous curricula. They expected kids to feel safe so they can concentrate on learning. And they wanted a place where students could be well-rounded and engaged citizens.

One striking difference between Schneider’s school profiles and those of GreatSchools is that there’s no overall score for each school. “The fact that most states present [an overall] rating of schools is just absolutely evidence of malpractice,” he says.

But how do you gauge that without relying on standardized tests? I saw the alternative in action in an eighth grade science classroom in Milford early this year before schools were closed. The teacher broke her class into teams to assess the contents of four beakers containing clear liquid. The students had all sorts of ideas for how to do the test, or the “performance evaluation,” as MCIEA calls it. Some weighed the beakers to calculate the liquids’ density and compared their results to values they found on the internet. Several smelled the contents and noticed an odor suspiciously like vinegar. Two teams were at the back of the room, boiling the liquids over a burner. If there was any salt residue once the liquid evaporated, one of the students explained, that would probably mean it was salt water. Another kid asked for baking soda, presumably to see if it would foam up when poured in.

The students were devising their own experiments—and seemed to be thoroughly engaged even as they were being evaluated for their problem-solving and teamwork. Teachers in several MCIEA districts told me about the projects they’d been able to create thanks to the freedom they were given. A high school social studies teacher in Attleboro described a project that explored the controversy over whether Harriet Tubman should replace Andrew Jackson on the $20 bill. “I can remember a kid ranting about the Specie Circular,” the teacher told me excitedly. Sally Eosefow, a former engineer who now teaches math, rattled off several challenges she presented to her students. In one, precalculus students had to find a plane wreck on a map given just two angle measurements, which requires them to use sines and cosines. In another, kids chart their own pulse and blood pressure readings based on sinusoidal functions.

MCIEA’s long-term goal is to show that these methods produce more valuable results than the existing state exams, so that school districts could replace the tests completely. Yet first both the state and federal education departments would have to agree that MCIEA’s methods could produce the same results as standardized tests. Schneider says that would defeat the purpose, since he sees standardized tests as flawed to begin with. The consortium is “waiting for the political winds to turn” so it can replace standardized tests with its evaluation tools in the districts it works with.

If it succeeds, it could offer parents and educators a very different way of looking at school performance. “If we built better data systems that aligned with the things that people care about and that treated schools and communities fairly,” Schneider says, “maybe we would be able to unravel and unstitch this national narrative we have that good schools are a rare commodity that parents need to compete over and that you can’t just send your kid to the neighborhood public school.”

GreatSchools’ partners are showing signs of taking the problem of racial segregation and school ratings more seriously. Redfin recently changed how it displays Great­Schools ratings. Glenn Kelman, Redfin’s CEO, tells me that putting the ratings on home profiles has been a “real point of contention here at the company.”

Kelman says his employees worry about the potential to encourage white flight and exacerbate residential segregation. Plus, as a company that employs real estate agents, it has to worry about their running afoul of fair housing laws. “The challenge that GreatSchools and Redfin have is that the consumer does want a really simple answer, and we want to provide context,” Kelman says. “People want to know which one is good and which one is bad. When you try to say it’s more complicated than that, you sometimes lose your grip on their attention.”

Kelman says an intern confronted him two years ago about how reductive the GreatSchools ratings could be, so he contacted GreatSchools about providing more details on Redfin’s website. GreatSchools, Kelman says, jumped at the opportunity. Now users can click on a school listed on a house profile and see the different components of GreatSchools’ ratings as well as residents’ reviews of the school. In its first year, the new feature didn’t lead to any changes in consumer behavior, according to Redfin. (Only 10 to 15 percent of people who click on a home profile scroll down far enough to see school ratings in the first place, the company says. It could not provide data on how many users filtered their search results by school ratings, but said few people use the feature.)

Still, Kelman says he’s glad his site included the new information. “I think tech, more broadly than Redfin, is just waking up to the idea that websites matter, mobile applications matter. We have to be careful about the information that we provide,” he says.

“Being in this seat a long time and trying to influence where people ought to want to live, what kind of home they ought to want for environmental reasons or for reasons of social justice, I just feel like I’ve come up against the most deep-seated desires people have,” Kelman tells me. “They call us when they just had their first baby and their world has narrowed to those two or three people—one or two parents and the child. Trying to get them to think about the rest of an American city is hard.”

This story was published with the support of a fellowship from Columbia University’s Ira A. Lipman Center for Journalism and Civil and Human Rights.

This story has been updated to reflect GreatSchools change to its rating system.