Jacob Sweet | Vox

The real science behind the billionaire pursuit of immortality

2024-05-09T14:31:44-04:00

Jonathan An tries to ignore the hype about new life-extension treatments, but it’s caught up to him anyway.

He has heard the gospel of the longevity influencers, including that one multimillionaire who has been on a media campaign for months claiming that the 111 pills he takes each day will help him live forever. An, an assistant professor of oral sciences at the University of Washington, doesn’t buy it. But he recently found himself inadvertently ensnared by the fervor around anti-aging — thanks to his mice.

An has studied mice suffering from periodontal disease, a bacterial-induced inflammatory infection of the gums that can lead to tooth loss. Mice (and more than 60 percent of human adults over 65) have to deal with this uncomfortable oral illness — and they don’t have much choice but to cope. When people’s teeth fall out, dentists like An replace them. But he would rather not have to remove so many.

While studying for his doctorate in dentistry at the University of Washington, An pursued a joint PhD to research preventive dental measures. He experimented with giving mice chow infused with the drug rapamycin each day to see if it would improve their oral health.

It worked. Mice treated for eight weeks with the drug — traditionally used to help prevent organ-transplant rejection — not only experienced delayed symptoms of periodontal disease, but saw regrowth of their tooth-supporting jaw bones.

This year, An is planning to test rapamycin in humans. If it has the same effect in adults as it did in mice, people might eventually be able to pick up a drug at the pharmacy that helps them avoid unwanted trips to the dentist’s office.

Better dental health would be a pleasant effect, but that’s not why An’s research drew an unusual amount of attention. Because the drug An chose to test was rapamycin, the longevity field took notice. In separate lab experiments over the past decade, rapamycin has been found to extend the lifespan of yeast, nematodes, fruit flies, and mice. It has helped mice delay or reverse immunity decline, muscle decline, cognitive decline, and cancer growth.

This string of successes for rapamycin, which belongs to a class of drugs that stifle one biological pathway for cell growth, has caught the eyes of renowned longevity researchers. It’s also attracted the attention of wealthy lifehackers and the clinics, supplement companies, and biotech investors who — out of true belief, opportunism, or a combination — stand to make money from people seeking an elixir for longer life.

Since An’s study was published in 2020, longevity clinics from across the country have asked him how they can incorporate rapamycin into their practices. Some scientists consider rapamycin a strong candidate for life-extension purposes both because it has helped lab species live longer and because it has already been approved as an immunosuppressant in humans. Today, doctors can and do prescribe rapamycin for off-label use — including for longevity.

An wants to believe that these clinics — part of a fledgling longevity industry that includes between 50 and 800 providers across the US, according to the Wall Street Journal — are genuinely trying to improve their clients’ health. But he suspects that may not always be the case.

He tells the longevity crowd what he does know, which is less exciting than they might hope. When it comes to human health, “I don’t know what rapamycin does,” he said. “But I always tell them to make sure to have a dentist on hand because some of the side effects are oral-related.”

Other companies want him to help with their own studies, the results of which they plan to keep private. An says no. “I’m a dentist,” An said. “Not a salesperson.”

A longer, healthier life is one of the easiest products in the world to sell. According to a Deloitte report, the 50 biggest longevity companies raised more than $1 billion in venture capital funding as of 2020 — a number that the company said would rise “due to the growing conviction that the longevity market could outstrip the existing health care market.” Altos Labs, a “rejuvenation” biotech whose investors include Jeff Bezos, announced in 2022 that it had raised $3 billion in funding.

An astronomer’s discovery of a neutron star has much less commercial potential and therefore generates much less interest than a researcher’s discovery that the micronutrient resveratrol helps yeast live longer — even if it’s likely that neither ultimately affects human lifespan. The attention paid to billionaire-funded research risks obscuring whether the longevity field is genuinely on the verge of a breakthrough or whether a clinic is just saying that to promote their experimental blood transfusion.

In reality, longevity research is advancing — but slowly. Clinical trials are moving forward on select uses for longevity drugs, younger researchers are taking the field more seriously, and private organizations are pledging significant support to research: The Saudi-based Hevolution Foundation has promised up to $1 billion in funding annually for biotech startups and academic researchers.

But while there likely remain many promising treatment candidates that have yet to be identified, they would take decades to reach clinical trials. Even academics who are bullish on the promise of longevity research fear that, for all the fanfare, the field has become too fixated on a few drugs and lifestyle adjustments that have been under investigation for years, while neglecting the basic research that could reveal novel pathways to slow down human aging.

For now, the three best ways to extend your life remain boring: eating a healthy diet, exercising regularly, and sleeping well. We aren’t going to add decades to human life any time soon; living to 150 or 200 remains in the realm of science fiction. But in decades to come, advancements in the science of aging may still lead to therapeutic breakthroughs that lengthen human healthspan — the period of life spent in good health. Perhaps a few more people will become centenarians, but the real success would be having more years when you can live well.

How longevity went mainstream in academia

Matt Kaeberlein, a longevity researcher at the University of Washington, remembers a time when few in academia took the study of aging — much less the idea of longevity — seriously.

“When I came into the field as a graduate student in 1998, there was nobody who went to graduate school to study aging,” he said. “The perception among the broader scientific community was that it was mostly snake oil and crap. There’s still a lot of snake oil and crap, but it is more accepted now than it used to be.”

The field began gaining wider recognition in 1993 when Cynthia Kenyon, a pioneer in aging research who now works at the Alphabet-owned life sciences company Calico Labs, discovered that mutating a single gene of the roundworm Caenorhabditis elegans doubled its lifespan. Other scientists soon figured out why. Gary Ruvkun, a professor of genetics at Harvard Medical School, and his colleagues found that the altered gene regulated an insulin-signaling pathway similar to one in humans that might play a role in slowing cell growth and metabolism. Researchers like Andrzej Bartke found similar mechanisms in mice, which have been the subject of much of the relevant research so far.

“One of the key things that’s happened is that the evidence that you can actually slow down and interfere with the aging process in mammals … has become so overwhelming that only the willfully blind can ignore it,” Richard A. Miller, who leads the University of Michigan’s Paul Glenn Center for Biology of Aging Research, told me.

In the last two decades, scientists have performed hundreds of lab experiments — mostly on animals — on drugs like rapamycin, canagliflozin, acarbose, empagliflozin, metformin, and on interventions like calorie restriction in diets and removal of nondividing senescent cells. Instead of testing the effects of these treatments on specific illnesses, many of these studies test whether certain interventions slow down animals’ aging processes and help them live longer.

The expansion of longevity research has unearthed some potentially useful information about which biological mechanisms control aging and how to alter them. In mice and other species, changing a single pathway has the power to extend life by significant margins, raising hopes that if humans respond similarly, certain drugs could extend human lives by years.

“We just have a better understanding of what those pathways are,” said Tom Rando, director of the UCLA Broad Stem Cell Research Center, “even if we don’t have a complete understanding of why they work and why they extend lifespan.”

Though most experiments with potential longevity drugs and other interventions like blood transfusions are still being tested on lab animals, two dozen candidate drugs have moved to clinical trials with human patients. Daniel Promislow, a University of Washington professor of medicine and pathology, told me that when he got into the field three decades ago, researchers talked hopefully about early developments someday making it to the lab. “Fast forward 25, 30 years, and many of these lab-based discoveries are now at the heart of a large number of clinical trials,” he said.

The clinical trials could allow researchers to produce evidence for interventions — besides diet, exercise, and sleep — that might help people live longer. Coleen T. Murphy, professor of molecular biology at Princeton, wrote in her 2023 book How We Age that, “What drugs can I take to live longer?” is becoming an increasingly tangible goal.

“A few years ago I might have chuckled at the naivety of this question,” she wrote, “but now it’s not so crazy to think that we will be able to take some sort of medicine to extend our healthy lifespans in the foreseeable future.”

The horizon for this future is still far off. Most researchers I spoke to didn’t believe that humans were going to experience a rapid increase in life expectancy any time soon — or maybe ever. They believed progress would instead be made in healthspan, helping people stay healthier for longer and avoiding long periods of physical and cognitive decline as they get older.

Such results probably won’t lead to someone living an extra decade. But they could make old age less burdensome. That would matter enormously for individuals, who could enjoy more years in good health, and society, by potentially reducing the high costs of late-in-life medical care.

“I can’t fathom saying, ‘Yeah, we’re going to try to extend someone’s lifespan by nine years,’” An told me. “There’s really no way to do that.”

Behind the hype, longevity research is moving — but slowly

In a way, some of the biggest improvements to human lifespans have already been made. Initiatives in public health — water sanitation, vaccination campaigns, sewage systems — have added decades to the average person’s life over the past few centuries. Since 1900, the average lifespan of a newborn has more than doubled worldwide — from 32 years old to 71 years old.

But the very fact that humans already live far longer than a lab animal is part of the reason that longevity research is so slow and difficult. For experimental purposes, laboratory mice live less than three years. Researchers have tested rapamycin in both young and old mice at a range of doses and then waited for them to die. Doing the same in humans would be far more expensive and take much longer. It’s also not strictly legal. The Food and Drug Administration doesn’t classify aging as a disease, which means that clinical trials can’t set out solely to test how much longer an intervention keeps someone alive. Instead, researchers must study age-related indicators like cardiovascular function and cognitive impairment instead of “aging” itself.

To compensate, longevity researchers are looking for other ways to measure aging that don’t require a patient’s death. They have identified several biomarkers that could serve as surrogate endpoints, but none have reached a scientific consensus. These include “aging clocks,” predictive models that purport to measure biological age or the age of specific biological organs; Bryan Johnson, the multimillionaire tech founder who calls himself a “professional rejuvenation athlete,” touts such data as proof that he has reversed his aging.

These tests are ostensibly based on the research of Steve Horvath, a former professor at UCLA who now works at Altos Labs. He has used age-related DNA methylation to determine biological age. Though most researchers I spoke to expressed cautious optimism about the potential of Horvath’s findings, they were skeptical of the extant consumer tests.

“We’re not really sure if the age we tell you is accurate and if it’s going to be the same tomorrow and whether it has any value,” said Tony Wyss-Coray, a Stanford professor of neurology who has found that elderly mice given the blood of younger mice see improvements in brain function. “And of course, no company wants to tell you that, but that’s just a fact.”

Most longevity researchers think about their research environment the same way: The flashiest stories are usually pretty removed from the actual state of the field. A drug that just helped mice live 50 percent longer is unlikely to do exactly the same for humans, no matter what a press release implies. Human bodies are much better at repairing their DNA than mice are, which makes them less susceptible to diseases like cancer. Plus, studies that would definitively prove a certain intervention would aid human life would take decades, and experts believe they could struggle to demonstrate their effectiveness to the FDA.

“You’ll rarely find a scientist funded by the [National Institutes of Health] who’s doing work in the biology of aging who would claim that their research could or will allow people to live to 140,” Rando told me. “It’s really coalesced around the idea that our main successes will be in reducing the burden of disease.”

It reflects a realism among the real experts. In longevity, there is not going to be a moment when a chrysalis bursts and a butterfly flies out, Miller said, a sudden leap forward in people’s life expectancy. “It’s more like the evolution of land plants. Gradually, they creep up over the beach, and then onto the meadow and then into the meadows. This is sort of creeping through the scientific community — too slowly.”

According to many researchers, part of the reason for the relatively slow progress in longevity treatments is lack of funding in the field. For all the flashy announcements about companies like Calico and Altos Labs, academic researchers struggle to find financial support. The National Institute on Aging, the NIH division that funds research on the aging process, projects that it will spend about 9 percent of its budget on the biology of aging in 2024 and just under 60 percent on neuroscience-specific research. (The NIA’s total projected budget in 2024 is about $4.4 billion of the NIH’s $47.1 billion.) Promislow and Kaeberlein, who co-run a long-term study on biological and environmental factors that could contribute to aging in dogs, are currently fighting to keep their project alive with their NIH funding expected to end in June.

“I think there’s an assumption by a lot of people that there’s a ton of money in aging research,” Murphy told me. “If you’re an academic trying to get funding from the NIH, it’s actually not true.”

The lack of funding also draws university researchers out of their scholarly institutions and to companies like Calico and Altos Labs. “The idea of working with very smart people with lots of resources, all that’s really attractive,” Miller told me.

But that drift to the private sector could actually slow down aging research, already a sluggish endeavor, even more in the long run. The field is trending toward investor-driven research, while the basic research studies necessary for the next generation of possible interventions languish because they depend on public or philanthropic funding.

Drugs like rapamycin have already taken decades to enter clinical trials, but it’s possible that none of the current leading longevity candidates work. Researchers don’t even agree on which of the current drugs and interventions is the most promising: Miller, for example, told me he thinks that rapamycin is “the wrong drug” and that more funding should go to canagliflozin, which has increased median survival age in male mice by 14 percent and for which human side effects are better known due to its use in treating type 2 diabetes since 2013. Still, he doesn’t think it’s easy, “from our limited amount of knowledge, to be confident as to whether rapamycin, or canagliflozin, or any other promising drug would produce major benefits in people with acceptably low side effects.” Most aging-related biotechnology companies use investor money to test aging interventions already proven in mice. Few are conducting the basic research to find new possible pathways for future therapies.

The more aging-related pathways scientists can find, the more possible targets for longevity drugs they would have. Each discovery opens the possibility for new interventions. Kaeberlein said that though the field has expanded in terms of the number of studies on certain drugs and mechanistic pathways, it’s also become in a sense more narrow.

“We think, ‘This is how the system works. So we’re going to test these parts of the model,’ instead of the more exploratory science that was being done when I was a graduate student, which was, ‘We have no frickin’ clue how the system works. Let’s go do some unbiased screens to figure out what’s happening here,’” he said.

Longevity researchers may be playing in a tiny corner of the sandbox, investigating just a few pathways while ignoring other possibilities. Scientists blame such myopia for the long gap between breakthroughs. The most consistently effective intervention for extending animal lifespan has been known for decades: restricting the number of calories they eat.

“I think that shift in mentality has led to more incremental results and fewer big, exciting, new discoveries,” said Kaeberlein, “and I think, personally, that’s why nobody has done better than rapamycin in 15 years and no one has done better than caloric restriction in 50 years.”

There’s also the possibility that drugs that have worked consistently across different species will work for some humans but not others. “The vast majority of studies in our field are done in one genetically identical strain of mouse,” Rando said. “It’s sort of like running a clinical trial in humans and only using identical twins. … Even if something could work, it’s likely to work in a subset of the population and not in everybody.”

Oddly, even the most brazen of the (non-expert) anti-aging boosters have uninspiring perceptions of the current state of longevity research. I was surprised when Bryan Johnson explained to me that, despite having a team of doctors who track the age of his organs and feed him a daily canister of pills, his choices weren’t really made based on today’s advancements in health and wellness.

He instead puts his faith in the continued evolution of artificial intelligence capabilities, which has advanced greatly over the past few years. He sees AI continuing to develop at an exponential rate — and longevity research eventually progressing at a more rapid speed than human researchers could hope to replicate.

“It’s an observation that we are baby steps away from super intelligence,” Johnson told me, “and it’s improving at a speed that we can’t imagine.”

It’s that, he hopes, that will bring about eternal life. The mice studies are less relevant.

A more realistic future for the longevity field

Immortality is enticing, but it’s not coming anytime soon. Neither is living to 150. Some people — hopefully more than now — will live to 100, but they will still be the exception. The way longevity research might push the field forward could look very similar to the treatments we already have. For people with a high risk of cardiovascular disease, statins are a sort of longevity drug. For those dealing with certain cancers, chemotherapy can be considered a longevity treatment.

The future of longevity likely looks more like the world where we discover that rapamycin — a drug that can extend the lives of mice and help humans accept a new organ — can also treat elderly patients for periodontal disease. It could mean that people take a blood sugar-regulating drug like canagliflozin and suffer from fewer heart attacks and cancers.

“I don’t really care about life extension because there’s no way to measure it,” An said. “It’s really about your health.”

Even in slow motion, the field keeps advancing. Murphy told me she was excited to see trial results from the longevity company Unity Biotechnology back in 2020. The drug UBX0101, which interacts with a tumor-suppressing pathway, cleared a phase 1 clinical trial.

When it moved to phase 2, though, it failed to achieve its aim of helping patients with osteoarthritis of the knee. A success could have been a promising sign for treatments to get rid of non-dividing senescent cells. But even a failure was valuable. It might not have been the result that anyone wanted, but it was a result, and it was public.

“That’s progress for our field,” she told me. “This is moving forward.”

Nobody knows what the point of homework is

2023-02-23T06:04:34-05:00

As the Covid-19 pandemic began and students logged into their remote classrooms, all work, in effect, became homework. But whether or not students could complete it at home varied. For some, schoolwork became public-library work or McDonald’s-parking-lot work.

Luis Torres, the principal of PS 55, a predominantly low-income community elementary school in the south Bronx, told me that his school secured Chromebooks for students early in the pandemic only to learn that some lived in shelters that blocked wifi for security reasons. Others, who lived in housing projects with poor internet reception, did their schoolwork in laundromats.

According to a 2021 Pew survey, 25 percent of lower-income parents said their children, at some point, were unable to complete their schoolwork because they couldn’t access a computer at home; that number for upper-income parents was 2 percent.

The issues with remote learning in March 2020 were new. But they highlighted a divide that had been there all along in another form: homework. And even long after schools have resumed in-person classes, the pandemic’s effects on homework have lingered.

Over the past three years, in response to concerns about equity, schools across the country, including in Sacramento, Los Angeles, San Diego, and Clark County, Nevada, made permanent changes to their homework policies that restricted how much homework could be given and how it could be graded after in-person learning resumed.

Three years into the pandemic, as districts and teachers reckon with Covid-era overhauls of teaching and learning, schools are still reconsidering the purpose and place of homework. Whether relaxing homework expectations helps level the playing field between students or harms them by decreasing rigor is a divisive issue without conclusive evidence on either side, echoing other debates in education like the elimination of standardized test scores from some colleges’ admissions processes.

I first began to wonder if the homework abolition movement made sense after speaking with teachers in some Massachusetts public schools, who argued that rather than help disadvantaged kids, stringent homework restrictions communicated an attitude of low expectations. One, an English teacher, said she felt the school had “just given up” on trying to get the students to do work; another argued that restrictions that prohibit teachers from assigning take-home work that doesn’t begin in class made it difficult to get through the foreign-language curriculum. Teachers in other districts have raised formal concerns about homework abolition’s ability to close gaps among students rather than widening them.

Many education experts share this view. Harris Cooper, a professor emeritus of psychology at Duke who has studied homework efficacy, likened homework abolition to “playing to the lowest common denominator.”

But as I learned after talking to a variety of stakeholders — from homework researchers to policymakers to parents of schoolchildren — whether to abolish homework probably isn’t the right question. More important is what kind of work students are sent home with and where they can complete it. Chances are, if schools think more deeply about giving constructive work, time spent on homework will come down regardless.

There’s no consensus on whether homework works

The rise of the no-homework movement during the Covid-19 pandemic tapped into long-running disagreements over homework’s impact on students. The purpose and effectiveness of homework have been disputed for well over a century. In 1901, for instance, California banned homework for students up to age 15, and limited it for older students, over concerns that it endangered children’s mental and physical health. The newest iteration of the anti-homework argument contends that the current practice punishes students who lack support and rewards those with more resources, reinforcing the “myth of meritocracy.”

But there is still no research consensus on homework’s effectiveness; no one can seem to agree on what the right metrics are. Much of the debate relies on anecdotes, intuition, or speculation.

Researchers disagree even on how much research exists on the value of homework. Kathleen Budge, the co-author of Turning High-Poverty Schools Into High-Performing Schools and a professor at Boise State, told me that homework “has been greatly researched.” Denise Pope, a Stanford lecturer and leader of the education nonprofit Challenge Success, said, “It’s not a highly researched area because of some of the methodological problems.”

Experts who are more sympathetic to take-home assignments generally support the “10-minute rule,” a framework that estimates the ideal amount of homework on any given night by multiplying the student’s grade by 10 minutes. (A ninth grader, for example, would have about 90 minutes of work a night.) Homework proponents argue that while it is difficult to design randomized control studies to test homework’s effectiveness, the vast majority of existing studies show a strong positive correlation between homework and high academic achievement for middle and high school students. Prominent critics of homework argue that these correlational studies are unreliable and point to studies that suggest a neutral or negative effect on student performance. Both agree there is little to no evidence for homework’s effectiveness at an elementary school level, though proponents often argue that it builds constructive habits for the future.

For anyone who remembers homework assignments from both good and bad teachers, this fundamental disagreement might not be surprising. Some homework is pointless and frustrating to complete. Every week during my senior year of high school, I had to analyze a poem for English and decorate it with images found on Google; my most distinct memory from that class is receiving a demoralizing 25-point deduction because I failed to present my analysis on a poster board. Other assignments really do help students learn: After making an adapted version of Chairman Mao’s Little Red Book for a ninth grade history project, I was inspired to check out from the library and read a biography of the Chinese ruler.

For homework opponents, the first example is more likely to resonate. “We’re all familiar with the negative effects of homework: stress, exhaustion, family conflict, less time for other activities, diminished interest in learning,” Alfie Kohn, author of The Homework Myth, which challenges common justifications for homework, told me in an email. “And these effects may be most pronounced among low-income students.” Kohn believes that schools should make permanent any moratoria implemented during the pandemic, arguing that there are no positives at all to outweigh homework’s downsides. Recent studies, he argues, show the benefits may not even materialize during high school.

In the Marlborough Public Schools, a suburban district 45 minutes west of Boston, school policy committee chair Katherine Hennessy described getting kids to complete their homework during remote education as “a challenge, to say the least.” Teachers found that students who spent all day on their computers didn’t want to spend more time online when the day was over. So, for a few months, the school relaxed the usual practice and teachers slashed the quantity of nightly homework.

Online learning made the preexisting divides between students more apparent, she said. Many students, even during normal circumstances, lacked resources to keep them on track and focused on completing take-home assignments. Though Marlborough Schools is more affluent than PS 55, Hennessy said many students had parents whose work schedules left them unable to provide homework help in the evenings. The experience tracked with a common divide in the country between children of different socioeconomic backgrounds.

So in October 2021, months after the homework reduction began, the Marlborough committee made a change to the district’s policy. While teachers could still give homework, the assignments had to begin as classwork. And though teachers could acknowledge homework completion in a student’s participation grade, they couldn’t count homework as its own grading category. “Rigorous learning in the classroom does not mean that that classwork must be assigned every night,” the policy stated. “Extensions of class work is not to be used to teach new content or as a form of punishment.”

Canceling homework might not do anything for the achievement gap

The critiques of homework are valid as far as they go, but at a certain point, arguments against homework can defy the commonsense idea that to retain what they’re learning, students need to practice it.

“Doesn’t a kid become a better reader if he reads more? Doesn’t a kid learn his math facts better if he practices them?” said Cathy Vatterott, an education researcher and professor emeritus at the University of Missouri-St. Louis. After decades of research, she said it’s still hard to isolate the value of homework, but that doesn’t mean it should be abandoned.

Blanket vilification of homework can also conflate the unique challenges facing disadvantaged students as compared to affluent ones, which could have different solutions. “The kids in the low-income schools are being hurt because they’re being graded, unfairly, on time they just don’t have to do this stuff,” Pope told me. “And they’re still being held accountable for turning in assignments, whether they’re meaningful or not.” On the other side, “Palo Alto kids” — students in Silicon Valley’s stereotypically pressure-cooker public schools — “are just bombarded and overloaded and trying to stay above water.”

Merely getting rid of homework doesn’t solve either problem. The United States already has the second-highest disparity among OECD (the Organisation for Economic Co-operation and Development) nations between time spent on homework by students of high and low socioeconomic status — a difference of more than three hours, said Janine Bempechat, clinical professor at Boston University and author of No More Mindless Homework.

When she interviewed teachers in Boston-area schools that had cut homework before the pandemic, Bempechat told me, “What they saw immediately was parents who could afford it immediately enrolled their children in the Russian School of Mathematics,” a math-enrichment program whose tuition ranges from $140 to about $400 a month. Getting rid of homework “does nothing for equity; it increases the opportunity gap between wealthier and less wealthy families,” she said. “That solution troubles me because it’s no solution at all.”

A group of teachers at Wakefield High School in Arlington, Virginia, made the same point after the school district proposed an overhaul of its homework policies, including removing penalties for missing homework deadlines, allowing unlimited retakes, and prohibiting grading of homework.

“Given the emphasis on equity in today’s education systems,” they wrote in a letter to the school board, “we believe that some of the proposed changes will actually have a detrimental impact towards achieving this goal. Families that have means could still provide challenging and engaging academic experiences for their children and will continue to do so, especially if their children are not experiencing expected rigor in the classroom.” At a school where more than a third of students are low-income, the teachers argued, the policies would prompt students “to expect the least of themselves in terms of effort, results, and responsibility.”

Not all homework is created equal

Despite their opposing sides in the homework wars, most of the researchers I spoke to made a lot of the same points. Both Bempechat and Pope were quick to bring up how parents and schools confuse rigor with workload, treating the volume of assignments as a proxy for quality of learning. Bempechat, who is known for defending homework, has written extensively about how plenty of it lacks clear purpose, requires the purchasing of unnecessary supplies, and takes longer than it needs to. Likewise, when Pope instructs graduate-level classes on curriculum, she asks her students to think about the larger purpose they’re trying to achieve with homework: If they can get the job done in the classroom, there’s no point in sending home more work.

At its best, pandemic-era teaching facilitated that last approach. Honolulu-based teacher Christina Torres Cawdery told me that, early in the pandemic, she often had a cohort of kids in her classroom for four hours straight, as her school tried to avoid too much commingling. She couldn’t lecture for four hours, so she gave the students plenty of time to complete independent and project-based work. At the end of most school days, she didn’t feel the need to send them home with more to do.

A similar limited-homework philosophy worked at a public middle school in Chelsea, Massachusetts. A couple of teachers there turned as much class as possible into an opportunity for small-group practice, allowing kids to work on problems that traditionally would be assigned for homework, Jessica Flick, a math coach who leads department meetings at the school, told me. It was inspired by a philosophy pioneered by Simon Fraser University professor Peter Liljedahl, whose influential book Building Thinking Classrooms in Mathematics reframes homework as “check-your-understanding questions” rather than as compulsory work. Last year, Flick found that the two eighth grade classes whose teachers adopted this strategy performed the best on state tests, and this year, she has encouraged other teachers to implement it.

Teachers know that plenty of homework is tedious and unproductive. Jeannemarie Dawson De Quiroz, who has taught for more than 20 years in low-income Boston and Los Angeles pilot and charter schools, says that in her first years on the job she frequently assigned “drill and kill” tasks and questions that she now feels unfairly stumped students. She said designing good homework wasn’t part of her teaching programs, nor was it meaningfully discussed in professional development. With more experience, she turned as much class time as she could into practice time and limited what she sent home.

“The thing about homework that’s sticky is that not all homework is created equal,” says Jill Harrison Berg, a former teacher and the author of Uprooting Instructional Inequity. “Some homework is a genuine waste of time and requires lots of resources for no good reason. And other homework is really useful.”

Cutting homework has to be part of a larger strategy

The takeaways are clear: Schools can make cuts to homework, but those cuts should be part of a strategy to improve the quality of education for all students. If the point of homework was to provide more practice, districts should think about how students can make it up during class — or offer time during or after school for students to seek help from teachers. If it was to move the curriculum along, it’s worth considering whether strategies like Liljedahl’s can get more done in less time.

Some of the best thinking around effective assignments comes from those most critical of the current practice. Denise Pope proposes that, before assigning homework, teachers should consider whether students understand the purpose of the work and whether they can do it without help. If teachers think it’s something that can’t be done in class, they should be mindful of how much time it should take and the feedback they should provide. It’s questions like these that De Quiroz considered before reducing the volume of work she sent home.

More than a year after the new homework policy began in Marlborough, Hennessy still hears from parents who incorrectly “think homework isn’t happening” despite repeated assurances that kids still can receive work. She thinks part of the reason is that education has changed over the years. “I think what we’re trying to do is establish that homework may be an element of educating students,” she told me. “But it may not be what parents think of as what they grew up with. … It’s going to need to adapt, per the teaching and the curriculum, and how it’s being delivered in each classroom.”

For the policy to work, faculty, parents, and students will all have to buy into a shared vision of what school ought to look like. The district is working on it — in November, it hosted and uploaded to YouTube a round-table discussion on homework between district administrators — but considering the sustained confusion, the path ahead seems difficult.

When I asked Luis Torres about whether he thought homework serves a useful part in PS 55’s curriculum, he said yes, of course it was — despite the effort and money it takes to keep the school open after hours to help them do it. “The children need the opportunity to practice,” he said. “If you don’t give them opportunities to practice what they learn, they’re going to forget.” But Torres doesn’t care if the work is done at home. The school stays open until around 6 pm on weekdays, even during breaks. Tutors through New York City’s Department of Youth and Community Development programs help kids with work after school so they don’t need to take it with them.

As schools weigh the purpose of homework in an unequal world, it’s tempting to dispose of a practice that presents real, practical problems to students across the country. But getting rid of homework is unlikely to do much good on its own. Before cutting it, it’s worth thinking about what good assignments are meant to do in the first place. It’s crucial that students from all socioeconomic backgrounds tackle complex quantitative problems and hone their reading and writing skills. It’s less important that the work comes home with them.

Jacob Sweet is a freelance writer in Somerville, Massachusetts. He is a frequent contributor to the New Yorker, among other publications.