About 13 years ago, well before I became a parent, I had a conversation with my aunt. She was the kind of aunt a young person could talk to: hilariously frank, slow to judge, and not easily scandalized. We were seated in her rumpus room, me on the couch and her on the floor, as one of her four children (she now has five) toddled back and forth. The topic turned to motherhood. “I’m not sure I like kids,” I said. If she was offended, she didn’t show it. In fact, she seemed to get what I was saying. “Yeah,” she replied, as she looked at her son, “I don’t think I used to like kids either. But I like my own kids.”

Neither of us meant any harm by our bluntness. My aunt, I’m sure, was attempting to be reassuring, and I was just trying to make sense of my ambivalence. In adolescence and early adulthood, I wasn’t someone whom anyone described as being “good with kids.” When a family friend or relative was looking for a babysitter, it wasn’t unheard of for them to ask my younger sister before they asked me. Little kids didn’t usually gravitate toward me, and when they did, I found feigning interest in whatever game they wanted to play a bit laborious. Our interactions often felt nerve-racking or forced, and I wasn’t sure what to make of this; I sensed—or perhaps just assumed—that most women felt otherwise.

Of course, people frequently use reductive language when talking about children: They “like,” “do not like,” or even “hate” kids. Sometimes, particularly in fringier corners of the internet, people appear to mean exactly what they say: They don’t like children as a class of human. But most of the time, I think people are attempting to express more complex emotions in language that feels intuitive. For example, they might be using “I don’t like kids” as shorthand for why they don’t want to become a parent—or regret becoming one. I’ve heard people speak this way to explain why they’d rather not hold a child, or even use the phrasing as a compliment: “I don’t usually like babies,” a young man once told me, “but yours is pretty cool.”

Probably more than anything, people say “don’t like” to express irritation over the disturbances that inevitably occur when children occupy public space: the whining, the shrieking, the knocking-over of things. In those situations, even people who rush to kids’ defense can end up leaning on language that focuses on likability. Children are lovely, they might say, and if you can’t see that, then something is wrong with you.

Stepping back, though—doesn’t something about this feel weird? When you talk about kids in terms of “like” or “don’t like,” you’re basically treating them as objects, the same way you’d talk about cars or handbags or a specific brand of Scotch. But kids aren’t commodities that we accessorize our life with. They’re humans.

In general, I don’t think it’s terribly useful to micromanage the way people speak. But over time, I’ve become convinced that we do need to scrutinize the language many people use to talk about kids, because it reflects and reinforces a view of children as somehow “other”—a view that gets in the way of conversations we ought to be having about children’s place in society and who is responsible for them.

More people than not (I hope) understand that it’s wrong to write off entire categories of humans based on superficial characteristics such as height, weight, skin color, and age. If I were to hear someone say they “don’t like old people,” I wouldn’t hesitate to call them out on it. Yet people talk about children that way all the time. Such broad-based, categorical phrasing effectively functions as a linguistic sleight of hand, allowing people to implicitly dismiss kids as a matter worthy of their concern. If kids are commodities, then responsibility for them falls on the owner and the owner alone. If kids are commodities, then it’s reasonable for me to feel violated when a child who isn’t “mine” throws a tantrum anywhere near my personal space.

I don’t think it’s wrong to be frustrated when a baby cries in the seat behind you on a plane, or when a toddler talks more loudly than social norms would consider polite. Kids do have a tendency to disrupt the tranquility of public life. Yet I believe that as a society, we genuinely need to discuss how adults—parents and nonparents—should engage with and accommodate children, and having that conversation becomes more difficult when people stake out black-and-white positions on kids’ likability.

[Read: An ode to crying babies]

This is a point that most people seem to understand in other circumstances. For instance, whether someone ought to help a blind person cross a busy road has essentially nothing to do with whether you like blind people. What any of us owe to our fellow humans, with all their different capacities and at various stages of life, is a matter of morals—the social contract we share—and not of preference. The goal here, in focusing on language, is not to shame anyone or to make people self-conscious about their use of words. It’s to open up discussion in a way that reduces the likelihood of endlessly speaking past one another.

As a person who spends quite a bit of time writing about the challenges of modern parenting, I want to talk with other people about, say, their hesitation to raise children. In my view, the interests of parents and the child-free are intimately bound together; we each, in our own way, resent the attitude that parenthood is something to be taken for granted. As a parent, I’d like American policy makers to stop taking my domestic labor as a given, to start appreciating the work that mothers and fathers do to raise decent members of society, and to pair that appreciation with more material support. I also get the sense that a lot of child-free people—in particular, child-free women—are bothered by those who believe that parenthood is a default condition, and who suggest, as our new vice president once did, that people who aren’t raising kids “don’t really have a direct stake” in what happens in our nation. But as soon as someone who is ambivalent about children declares that they “don’t like” kids, a wedge is driven between parents and nonparents. We’re no longer on the same team.

[Read: Cultural shifts alone won’t persuade people to have kids]

This goes for all of the other pressing concerns about child-rearing that Americans ought to be discussing. Is the country’s threadbare family-policy framework, with its nonexistent paid parental leave and meager funding for child care or other financial support, adequately addressing the needs of children? (No? Then let’s talk about it.) Do we owe it to kids to take their needs into consideration when we’re setting workplace policy? Is the way we’ve divvied up our public resources—with the country spending far more on the elderly than on the young—truly just? Parent or not, whether you “like kids” or not, decisions about policy at some point wind up affecting all of us. And discussing these concerns would be easier if we could dispense with the “don’t like” language and strive to use words that reflect children’s humanity.

I won’t try to offer readers a set of scripts to use in place of our more objectifying terminology. But I would like to propose an experiment: If you find yourself moved to say you don’t like kids, swap in another group of people and see whether that feels like an acceptable thing to say. If it doesn’t, consider thinking in more nuanced terms about the idea you’re trying to express—terms that make clear you’re talking, with all due respect, about your fellow humans. Most likely, doing so will help your position sound a lot more reasonable. And it will certainly improve your odds of being heard.

In 2012, Colorado and Washington State legalized the commercial production and sale of cannabis for nonmedical use, and since then 22 other U.S. states have followed. The shift was viewed in many quarters as benign and overdue—involving an organic, even medicinal, intoxicant with no serious drawbacks. Advocates promised safe and accurately labeled products, reduced addiction to opioids, smaller prison populations, surging tax revenue, and a socially responsible industry that prioritized people over profits. But all of those promises have turned out to be overstated or simply wrong.  

Legalization has raised cannabis consumption dramatically, and also altered patterns of use. In the 1990s and early 2000s, most consumers smoked the drug and did so only occasionally or semi-regularly—say, on weekends with friends. Some people used more regularly, of course: In 2000, 2.5 million Americans reported daily or near-daily cannabis use. But by 2022, that had grown sevenfold to 17.7 million. Remarkably, that’s more than the 14.7 million who reported using alcohol that often. Today, more than 40 percent of Americans who use cannabis take it daily or near-daily, and these users consume perhaps 80 percent of all the cannabis sold in the U.S.

The drug’s potency has also risen sharply. Until the year 2000, the average potency of seized cannabis never exceeded 5 percent THC, the principal intoxicant in the plant. Today, smokeable buds, or flower, sold in licensed stores usually exceed 20 percent THC. Vapes, dabs, and shatter—all of which are forms of drug delivery that commercialization spread—are more potent still.

[Malcolm Ferguson: Marijuana is too strong now]

More frequent use of more potent products has led to a staggering rise in the typical consumer’s average weekly dose of THC. Back in the 1980s and ’90s, when potency averaged about 4 percent, someone consuming one 0.4-gram joint each weekend night—and none on weekdays—was averaging roughly 32 milligrams of THC a week. Average daily users today are consuming about 1.6 grams of high-potency flower a day, or its equivalent in other forms. That works out to more than 2,000 milligrams of THC a week—or about 70 times as much.

The numbers are shocking, and yet this is what happens when frequency, potency, and quantity all rise in tandem. For some consumers, high potency itself encourages more frequent use by delivering a stronger effect.

Medical science can’t yet clarify the effects of long-term use of 300-plus milligrams of THC a day, because this consumption pattern is new. Most controlled studies work with short-term exposure to smaller doses, often in the 20-to-50 milligram range, and observational studies that followed users for years were examining a drug—low-strength, infrequently used cannabis—that barely exists anymore.

But high-frequency use of high-potency marijuana raises a range of concerns. For one thing, there is little question that cannabis intoxication can impair cognitive functions including concentration and memory formation. That was not a big worry when most people used only on weekends. Daily use, however, means using on work and school days. The drug also impairs perception and motor control; the availability of strong, legal marijuana has been followed by increases in automobile crashes and emergency-room visits.

And over the long term, although some people can handle a wake-and-bake lifestyle, just like some alcoholic people are functional, there are likely millions of users for whom couch lock impedes career advancement, academic success, or meeting responsibilities to family.

On surveys, 63 percent of high-frequency users report enough cognitive, emotional, employment, and social problems as a result of using the drug to be coded as meeting the criteria for a cannabis-use disorder (a condition defined by being unable to fully control drug-use behavior despite its negative consequences). For technical reasons, we think that figure overstates the problem, but there is no doubt that the problem exists: 17 percent of high-frequency users report wanting cannabis so badly that there are times they can’t think of anything else. Chronic use may lead to other health problems as well. Most notably, evidence is mounting that frequent use of high-strength products raises the risk of serious mental illnesses such as schizophrenia.  

To be clear, these risks and harms do not remotely add up to a “cannabis crisis” in the same way that we speak of an opioid crisis or a meth crisis—calamities marked by widespread premature mortality and shattered families. Many people who enjoy cannabis have no trouble managing their use. They can now buy it cheaply and without stigma, in a variety of forms. And everyone can be relieved that adolescents’ cannabis use has stayed roughly where it was since legalization began. By one standard measure, use rose just 3 percent among 12-to-17-year-olds from 2012 to 2022.

But use has soared for adults (up 155 percent), especially for those 35 and older (up 300 percent), and the increase cannot be characterized as entirely benign. Many assumptions made about what would follow legalization seem naive in retrospect.

Those assumptions extended beyond the nature of the relationship between pot users and the drug, and how it might change. Soaring cannabis use would still have been a win, from public-health and crime-control perspectives, if it had resulted in less use of even more dangerous drugs. But it hasn’t. Predictions that cannabis legalization would reduce consumption of alcohol, a drug much more strongly associated with physical aggression, were not realized—reductions observed in some groups or contexts were offset by increases in others.  

Based on weak scientific evidence, many advocates likewise promised that legal cannabis would lead people to use fewer opioids. (Weedmaps—an online review site for pot—put up billboards all over the country promising reductions in opioid overdose, for instance.) Yet those early findings were reversed as more data became available, and recent reviews suggest that legalization is more likely to increase than reduce opioid-death rates. This should not be too surprising: Although the old “gateway drug” arguments of the 1970s and ’80s overstated the risk of merely trying marijuana, the commercialization of cannabis has clearly expanded high-frequency use, and dependence on any drug can increase the likelihood of using and developing dependence on other drugs.

[Read: Marijuana’s health effects are about to get a whole lot clearer]

Some promised criminal-justice benefits have also proved illusory, in part because advocates exaggerated the extent to which marijuana use entangled people in the criminal-justice system. “Discriminatory enforcement of marijuana laws is one reason that black and Latino Americans make up two-thirds of the U.S. prison population,” the progressive Center for American Progress noted in 2018, in a report advocating national legalization. But even before legalization, very few people were in prison for pot possession alone. There were a lot of pot-smoking burglars and robbers behind bars, but only about 2 percent of inmates were in prison solely for marijuana offenses, and most of those were traffickers or their employees.

That there had been too many marijuana-possession arrests is undoubtedly true. And legalization has cut them sharply, leaving mostly only arrests of underage users and of residual illegal suppliers. But even here, the case for outright legalization of supply was oversold: States that merely decriminalized marijuana possession saw declines almost as large. In California, for example, converting marijuana possession from a misdemeanor to a civil infraction reduced possession arrests by 86 percent in just 12 months. Subsequent legalization had only a modest incremental effect.

Allowing commercial supply—as opposed to merely decriminalizing possession—has produced other unintended consequences, though these consequences could easily have been anticipated because businesses typically follow the laws of economics.    

Large producers run by MBAs have adopted industrial agricultural practices that are brutally efficient, dramatically outcompeting the artisanal production that many advocates foresaw. Before legalization, much high-quality cannabis was grown in small indoor facilities; one 2006 Dutch study of 77 illegal grows reported an average size smaller than 200 square feet. Now an average-size commercial grow might operate on 10,000 to 20,000 square feet, and an industry magazine lists one producer (Copperstate Farms) as operating almost 2,000,000 square feet of greenhouse grow space; mixed-mode growers are even larger.

Commercial production has driven down prices, and so the cannabis tax windfall touted by many supporters of legalization has also been underwhelming. In California, cannabis excise and sales taxes peaked in 2021; by the first quarter of 2023, they were reported as accounting for only 0.2 percent of total state tax collections. Not all taxes due even get collected; in 2023, for instance, 15 percent of the state’s cannabis firms defaulted on taxes they owed.  

Falling prices have thinned profit margins, adding to the commercial imperative to expand the market and attract new customers. Hence the proliferation of edibles and other products that are more accessible to nonsmokers. The industry is targeting women—who historically used cannabis less than men did—as a growth demographic, just as the cigarette and alcohol industries had before. From 2012 to 2022, high-frequency use grew strongly for men (up 137 percent), but exploded among women (up 300 percent).

Many commercial cannabis providers have proved difficult to regulate. Initially, regulatory enforcement efforts tended to be modest, and that was an error. Misleading labels are commonplace in the cannabis industry today, and some producers use unapproved pesticides or exploitative labor arrangements.  

The 2018 Farm Bill created further opportunities for bad behavior. The bill was supposed to legalize nonintoxicating uses of the cannabis plant, such as growing fiber for clothes or seed for food and oil. Unfortunately, loopholes let unscrupulous actors sell intoxicating products completely outside of most states’ regulatory systems. The Farm Bill permits the production and sale of “hemp”—defined as any cannabis product containing less than 0.3 percent of delta-9 THC, the primary THC variant in cannabis. But edibles, being relatively heavy, can contain a lot of delta-9 THC and still, by weight, remain under the 0.3 percent threshold. What’s more, the marijuana plant contains nonintoxicating cannabinoids that can be chemically transformed into intoxicating cousins such as delta-8 THC. The resulting array of products, which can appeal to youth, may have no labeling requirements (depending on what state they’re being sold in) and no protection against unfamiliar and potentially dangerous synthetic by-products. They may not have been tested for pesticides either.

Unsurprisingly, hemp producers who do not follow product-safety rules have in many cases been outcompeting those state-licensed cannabis companies that try to follow the regulations, contributing to high cannabis-business failure rates and less reliable products for consumers.

These ills and others—the sprouting of cannabis shops on seemingly every block in some city neighborhoods, the smell of pot that greets many riders of public transportation—have not gone unnoticed by the American people. The election in November underscored the degree of disappointment with the results of marijuana legalization. Though Nebraska did become the 39th state to approve the drug for medical purposes, North and South Dakotans voted down ballot initiatives to legalize recreational use. Floridians did the same—despite $150 million in campaign spending by the industry and an endorsement from Donald Trump.

This pause in what had seemed an inexorable movement toward wider—and eventually national—legalization is healthy. Leaping all the way from prohibition to the enthusiastic embrace of a for-profit, freewheeling, corporate cannabis industry has clearly created downsides and excesses that legalization advocates did not initially imagine (or, in some cases, admit). States still considering legalization—and those that may be reconsidering how legalization has worked out for them so far—would be wise to instead explore the ample middle ground, or what the late drug-policy expert Mark Kleiman called a “grudging toleration” of legal use and supply. Even a society that otherwise embraces free-market capitalism should be open to middle paths for addiction-inducing intoxicants, which are not ordinary commodities.  

[Read: I don’t want to smell you get high]

What might grudging toleration look like in practice? In addition to eliminating the Farm Bill loopholes that have contributed to a Wild West environment in many places, we would offer four specific suggestions.   

1. Restrain the power of large-scale producers.

The cannabis supply chain spans growers, manufacturers who process and package the plant material, and retailers. Regulation is needed for farmers (concerning which pesticides are allowed, for instance) and retailers (testing compliance with laws blocking sale to minors, for example)—but the bigger challenges involve the manufacturers who produce the concentrated products, control the brands, and dominate marketing and advertising. Two remedies can help.

First, there is no reason to allow for-profit corporations to participate in product manufacturing. For-profit businesses are fabulously efficient at developing new products and driving up consumption. That’s fine when the product is cornflakes or canola, but not when it involves addictive drugs. Cannabis is an addiction-inducing intoxicant for which rapidly expanding consumption has significant costs.  

Instead, legalization could restrict cannabis-product manufacturing to nonprofits or public-benefit corporations. Reliance on nonprofits is a norm in some other industries providing goods or services that in one way or another involve issues beyond pure commerce. Most hospitals and universities, for instance, are either nonprofit or government-owned. In the cannabis industry, these organizations could be chartered to undercut illegal supply by producing to meet existing demand, without promoting greater consumption.

Second, especially in places where for-profit manufacturers are still permitted, major manufacturers should be barred from owning, operating, or controlling either farms or retail outlets. Similar restrictions were part of many states’ plans when alcohol prohibition was repealed, and they might limit big corporations’ power—including their lobbying power. Likewise, they should be barred from merging with tobacco and alcohol companies.

2. Curtail high-potency products.

For many, the purpose of legalization was to replace the illegal market with legal, regulated supply. But legalization has also changed the market, bringing in a slew of more potent products. Drug-reform advocates sometimes invoke the so-called iron law of prohibition, which claims that prohibition begets more potent forms (because, being more compact, they are more easily hidden). But with cannabis, the opposite happened: It was legalization that spread higher-potency forms of the drug.  

Whether cheap, higher-potency products necessarily exacerbate health harms is much debated. But history is replete with examples of inexpensive, high-potency forms of drugs creating new problems, from the British Gin Craze of the first half of the 18th century in London (during which consumption increased eightfold to about one gallon per person per year) to the current fentanyl epidemic, which has killed more Americans than heroin ever did.

Some fear that banning higher-potency products will create or greatly expand illegal markets, but modern societies often ban certain forms of a product without creating big illegal markets—as long as other forms remain legal. For instance, throughout much of the 20th century, many countries banned the sale of absinthe, but there was no big illegal market for absinthe because other liquors were available. Likewise, today’s bans on caffeinated alcoholic drinks and flavored cigarettes are mostly honored.

Quebec already essentially bans dabs, butane hash oil, and other high-potency products, and it has considerably less cannabis use than other provinces of Canada.    

The U.S. should likewise ban such products, and maybe also the synthesis of artificial cannabinoids. And for products that stay within the potency limit, a further safeguard could be taxing more potent products at a higher rate, just as is done with alcoholic beverages.

3. Leave room for small-scale producers and other small businesses.

The primary challenge to public health does not come from the many small artisanal producers of marijuana, or from retailers. The greater problem is Big Marijuana. Applying the same rules to all parties burdens hobbyists and boutique producers while letting corporations run amok.

Most states have enacted cottage-food-production laws that exempt small-scale producers of craft-food products (baked goods, pickles, honey, etc.) from the strict scrutiny that is appropriate for agribusiness and food conglomerates. Cannabis policy could make similar distinctions.  

Small growers might be exempted from certain regulations when selling only flower—and also prohibited from selling refined or dangerous products, just as cottage-food producers are usually banned from selling meats or goods that need to be refrigerated.

4. Get public safety and public health off the sidelines.

In order to limit the damage done by legal addictive products, society needs effective public-health regulation. And in order to thrive, licensed legal industries need government enforcement against illegal suppliers. Both of these necessities have been lacking.

Neither regulators nor police have attacked illegal production, promotion, and sale with sufficient vigor, perhaps because any enforcement involving marijuana has become entangled—at least in the minds of many progressives—with concerns about the carceral state, or anti-police sentiment more generally.

[Jane C. Hu: Almost no one is happy with legal weed]

But when the legal risk of, say, operating an unlicensed weed shop drops to near-zero, the illegal industry grows and the legal industry suffers, undermining the legalization regime and tax revenue at the same time. Enforcement agencies, those who oversee them, and the activist community need to shake off the misperception that enforcing the rules of a legal industry is a revival of the War on Drugs.

Similarly, public-health departments must start informing the public more vigorously about the health risks of cannabis, just as they do those of tobacco, alcohol, and gambling. Thus far, they have generally failed to do this, perhaps because of misplaced fears of reenacting hysterias of prior eras. Today, a big, legal industry is selling a product with established health risks, and public health needs to embrace its traditional role as an advocate for health over profit.  

These and other reforms would better balance the trade-offs between profit and public interest. Naturally, the industry will fight them, but this should only increase urgency. Naive and self-serving advocates shaped (and, via the initiative process, sometimes wrote) many state-level legalization bills, with results that should trouble us. Legalization should be redesigned where it already exists, and efforts to expand it to other states or nationally should learn from the mistakes of the recent past.

Shortly after midnight, a few hours after the horrifying collision between an airplane and a helicopter at Reagan National Airport, President Donald Trump felt the time was right for a shocked nation to hear his insights into the tragedy. “It is a CLEAR NIGHT, the lights on the plane were blazing, why didn’t the helicopter go up or down, or turn,” he wrote on Truth Social. “Why didn’t the control tower tell the helicopter what to do instead of asking if they saw the plane. This is a bad situation that looks like it should have been prevented. NOT GOOD!!!”

While you might question the sophistication of his analysis, Trump was correct about both the physics of the collision (namely, that it could have been avoided if the helicopter had gone either up or down) and the moral valence of the mass casualty event (bad, not good).

But, by midday today, without the benefit of any important conclusions about the cause of the crash, Trump adopted a different perspective. “We do not know what led to this crash, but we have some very strong opinions and ideas,” he told reporters in a rambling press conference. His strong opinion was that the cause was a “diversity push” in the Federal Aviation Administration’s hiring process.

Lest that comment be dismissed as the half-formed musings of a president reacting in real time to a developing event, a few hours later Trump doubled down. In a live broadcast from the Oval Office, he signed an executive order that, in the words of an off-camera Vice President J. D. Vance, pinned responsibility for the crash on “the Biden administration’s DEI and woke policies.”

The purpose of Trump’s wild finger-pointing appears to be twofold: first, to avoid taking any blame for a disaster; and second, to exploit the tragedy while it is in the public’s mind, using it to advance the notion that his administration is replacing favoritism toward minorities with pure, race-blind merit. “As you said in your inaugural, it is color-blind and merit-based,” said Defense Secretary Pete Hegseth, joining Trump at the press conference. “The era of DEI is gone at the Defense Department, and we need the best and brightest, whether it’s in our air-traffic control or whether it’s in our generals or whether it’s throughout government.”

This was rich coming from a man who might be the least qualified secretary of defense in American history—a Cabinet official whose professional qualifications include mismanaging two small lobbying organizations and whose alleged history of drinking and mistreatment of women led his own sister-in-law to urge the Senate to reject his nomination, as it very nearly did.

[Jonathan Chait: Donald Trump’s most dangerous Cabinet pick]

And Hegseth is hardly an outlier. Trump has already done more to abandon the ideal of meritocracy than perhaps any presidential administration since the Progressive Era. He is going to war against the civil-service system, which was established more than a century ago to ensure that federal jobs go to qualified civil servants, rather than as rewards for party hacks, as had been the case previously. Trump, who believes that nonpartisan civil servants constitute a “deep state” conspiracy against him, would rather lose their expertise than risk it being deployed in ways that thwart his personal ambitions.

He has gone even further in this direction in selecting his Cabinet. Every president tends to fill such roles with supporters, but Trump has elevated loyalty to an almost comical degree. Not only must Trump’s Cabinet officials have supported him in the election, but they must endorse, or at least refuse to contradict, his infamously false claim to have won the 2020 election. The driving logic behind many of his most high-profile Cabinet picks appears to be a desire to find individuals who will stand behind the president if and when he violates norms, laws, or basic decency.

That is how Hegseth, despite his miserable record of management experience, was elevated to run the Pentagon. It is how Kash Patel, the author of a ridiculous children’s book portraying himself as a wizard and Trump as a king, was nominated to run the FBI. And it is how Robert F. Kennedy Jr., who has waged a pseudoscientific war against vaccines and appears to not know basic facts about Medicare and Medicaid, was tapped to run the federal department that oversees those programs.

One problem with discussing Trump’s opinions on fast-moving matters like the plane crash is that, in the absence of a completed investigation, it’s impossible to say for sure what did cause the disaster. Investigators haven’t even determined which errors were made, let alone why they occurred. It is possible that the entire fault rests with the helicopter pilot, as Trump himself suggested the night of the crash.

It’s true that the federal civil service has many problems, not least the extreme bureaucratic hurdles that stand in the way of both hiring qualified candidates and firing low-performing employees. It’s true, too, that the FAA has been sued over a clumsy program to boost minority representation. That effort arose out of an understandable desire to broaden the overwhelmingly white hiring pipeline for air-traffic controllers, but is alleged to have included perverse hiring criteria that unfairly filtered out qualified applicants.

There is no evidence yet that the FAA, let alone its hiring practices, had any responsibility for the crash. But to the extent that Trump thinks the underlying issue is an insufficient focus on merit, his moves to purge the government of non-Trumpist civil servants is all but guaranteed to make the problem worse. When you are not only selecting for loyalty, but defining that loyalty to mean “affirming morally odious values and factually absurd premises,” you are reducing your hiring pool to the shallowest part.

[David A. Graham: Blind partisanship does not actually help Trump]

And to be sure, when loyalty itself is the job requirement, this makes a certain kind of sense. La Cosa Nostra does not recruit its members very widely, because, as with Trump, its fear of betrayal outweighs its interest in hiring and promoting the most skilled racketeers and leg-breakers. When you are trying to run a government along Mafia hiring and promotion principles, you are necessarily forfeiting expertise and intelligence.

If Trump has his way, over the next four years, the political composition of the people engaged in directing air traffic, testing food for safety, preventing terrorism, and other vital public functions will change dramatically. The ones who have a serious problem with January 6 will be gone, replaced by people who are willing to repeat Trump’s lies—if they are replaced at all. You can justify that process as the president’s prerogative to shape the executive branch. What you can’t call it is an elevation of merit.

In a Massachusetts cellar in 1873, Lydia Pinkham first brewed the elixir that would make her famous. The dirt-brown liquid, made from herbs including black cohosh and pleurisy root, contained somewhere between 18 and 22 percent alcohol—meant as a preservative, of course. Within a couple of years, Pinkham was selling her tonic at $1 a bottle to treat “women’s weaknesses.” Got the blues? How about inflammation, falling of the womb, or painful menstruation? Lydia E. Pinkham’s Vegetable Compound was the solution. Pinkham’s matronly smile, printed on labels and advertisements, became as well known as Mona Lisa’s.

Lydia E. Pinkham’s Vegetable Compound was one of thousands of popular and lucrative patent medicines—health concoctions dreamed up by chemists, housewives, and entrepreneurs—that took the United States by storm in the 19th and early 20th centuries. These products promised to treat virtually any ailment and didn’t have to reveal their recipes. Many contained alcohol, cocaine, morphine, or other active ingredients that ranged from dubious to dangerous. Dr. Guild’s Green Mountain Asthmatic Compound was available in cigarette form and included the poisonous plant belladonna. Early versions of Wampole’s Vaginal Cones, sold as a vaginal antiseptic and deodorizer, contained picric acid, a toxic compound used as an explosive during World War I. Patent-medicine advertisements were unavoidable; by the 1870s, 25 percent of all advertising was for patent medicines.

After the Pure Food and Drug Act was passed in 1906, the newly created Food and Drug Administration cracked down on miracle elixirs. But one American industry is still keeping the spirit of patent medicine alive: dietary supplements. In the U.S., vitamins, botanicals, and other supplements are minimally regulated. Some can improve people’s health or address specific conditions, but many, like the medicines of old, contain untested or dangerous ingredients. Nevertheless, three-quarters of Americans take at least one. Some take far more. Robert F. Kennedy Jr., the longtime conspiracy theorist and anti-vaccine activist who’s awaiting Senate confirmation to run the Department of Health and Human Services, has said he takes a “fistful” of vitamins each day. Kennedy has in recent years championed dietary supplements and decried their “suppression” by the FDA—an agency he would oversee as health secretary. Now he’s poised to bring America’s ever-growing supplement enthusiasm to the White House and supercharge the patent-medicine revival.  

The newly created FDA eventually required all pharmaceutical drugs—substances intended for use in the diagnosis, cure, mitigation, treatment, or prevention of disease—to be demonstrably safe and effective before they could be sold. But dietary supplements, as we call them now, were never subject to that degree of scrutiny. Vitamins were sold with little interference until the “megadosing” trend of the late 1970s and ’80s, which began after the chemist Linus Pauling started claiming that large amounts of vitamin C could stave off cancer and other diseases. The FDA announced its intention to regulate vitamins, but the public (and the supplement industry) revolted. Mel Gibson starred in a television ad in which he was arrested at home for having a bottle of Vitamin C, and more than 2.5 million people participated in a “Save Our Supplements” letter-writing campaign. Congress stepped in, passing the 1994 Dietary Supplement Health and Education Act, which officially exempted dietary supplements from the regulations that medications are subject to.

Since then, the FDA has generally not been responsible for any premarket review of dietary supplements, and manufacturers have not usually had to reveal their ingredients. “It’s basically an honor system where manufacturers need to declare that their products are safe,” says S. Bryn Austin, a social epidemiologist and behavioral scientist at the Harvard T. H. Chan School of Public Health. The agency will get involved only if something goes wrong after the supplement starts being sold. As long as they disclose that the FDA hasn’t evaluated their claims, and that those claims don’t involve disease, supplement makers can say that their product will do anything to the structure or function of the body. You can say that a supplement improves cognition, for example, but not that it treats ADHD. These claims don’t have to be supported with any evidence in humans, animals, or petri dishes.

In 1994, the dietary-supplement industry was valued at $4 billion. By 2020, it had ballooned to $40 billion. Patent-medicine creators once toured their products in traveling medicine shows and made trading cards that people collected, exchanged, and pasted into scrapbooks; today, supplement companies sponsor popular podcasts, Instagram stories are overrun with supplement ads, and influencers make millions selling their own branded supplements. The combination of modern wellness culture with lax regulations has left Americans with 19th-century-like problems: Pieter Cohen, an associate professor of medicine at Cambridge Health Alliance, has found a methamphetamine analogue in a workout supplement, and omberacetam, a Russian drug for traumatic brain injuries and mood disorders, in a product marketed to help with memory.

Last year, Kennedy accused the FDA of suppressing vitamins and other alternative health products that fall into the dietary-supplement category. But “there is no truth about the FDA being at war on supplements over the last several decades,” Cohen told me. “In fact, they have taken an extremely passive, inactive approach.” Experts have repeatedly argued that the FDA needs more authority to investigate and act on supplements, not less. And yet, Kennedy continues to champion the industry. He told the podcaster Lex Fridman that he takes so many vitamins, “I couldn’t even remember them all.” Kennedy has vocally opposed additives in food and conflicts of interest in the pharmaceutical industry, but has failed to mention the dangerous additives in dietary supplements and the profits to be made in the supplement market. (Neither Kennedy nor a representative from the MAHA PAC responded to a request for comment.)

In an already permissive environment, Kennedy’s confirmation could signal to supplement manufacturers that anything goes, Cohen said. If the little regulation that the FDA is responsible for now—surveilling supplements after they’re on the market—lapses, more adulterated and mislabeled supplements could line store shelves. And Americans might well pour even more of our money into the industry, egged on by the wellness influencer charged with protecting our health and loudly warning that most of our food and drug supply is harmful. Kennedy might even try to get in on the supplement rush himself. Yesterday, The Washington Post reported that, according to documents filed to the U.S. Patent and Trademark Office, Kennedy applied to trademark MAHA last year, which would allow him to sell, among other things, MAHA-branded supplements and vitamins. (He transferred ownership of the application to an LLC in December. Kennedy’s team did not respond to the Post.)

A truly unleashed supplement industry would have plenty of tools at its disposal with which to seduce customers. Austin studies dietary supplements that make claims related to weight loss, muscle building, “cleansing,” and detoxing, many of which are marketed to not just adults, but teenagers too. “Those types of products, in particular, play on people’s insecurities,” she told me. They also purport to ease common forms of bodily or mental distress that can’t be quickly addressed by traditional medical care. Reducing stress is hard, but ordering the latest cortisol-reducing gummy on TikTok Shop is easy. Your doctor can’t force vegetables into your diet, but a monthly subscription of powdered greens can.

Judy Z. Segal, a professor emerita at the University of British Columbia who has analyzed patent-medicine trading cards from the 19th and 20th centuries, told me that supplement-marketing strategies “have not changed that much since the patent-medicine era.” Patent medicines appealed to ambient, relatable complaints; one ad for Burdock’s Blood Bitters asserted that there were “thousands of females in America who suffer untold miseries from chronic diseases common to their sex.” And the makers of patent medicine, like many modern supplement companies, used friendly spokespeople and customer testimonials while positioning their products as preventive care; according to one ad for Hartshorn’s Sarsaparilla, “The first deviation from perfect health should receive attention.”

In 1905, the muckraker Samuel Hopkins Adams lamented that “gullible America” was so eager to “swallow huge quantities of alcohol, an appalling amount of opiates and narcotics, a wide assortment of varied drugs ranging from powerful and dangerous heart depressants to insidious liver stimulants; and, far in excess of all other ingredients, undiluted fraud.” Compounds and elixirs go by different names now—nootropics, detoxes, adaptogens—but if Adams walked down any supplement aisle or browsed Amazon, he’d still find plenty of cure-alls. He could even pick up a bottle of Lydia E. Pinkham’s Herbal Supplement, which is sold as an aid for menstruation and menopause. Pinkham’s face smiles at buyers from the label, though its advertised benefits are now accompanied by a tiny disclaimer: “This statement has not been evaluated by the FDA.”

This is Atlantic Intelligence, a newsletter in which our writers help you wrap your mind around artificial intelligence and a new machine age. Sign up here.

Earlier this week, almost overnight, the American tech industry entered a full-on panic. The latest version of DeepSeek, an AI model from a Chinese start-up of the same name, appeared to equal OpenAI’s most advanced program, o1. On Monday, DeepSeek overtook ChatGPT as the No. 1 free app on Apple’s mobile-app store in the United States.

So far, China has lagged the U.S. in the AI race. DeepSeek suggests that the country has gained significant ground: The chatbot was built more quickly and with less money than analogous models in the U.S., and also appears to use less computing power. Software developers using DeepSeek pay roughly 95 percent less per word than they do with OpenAI’s top model. One prominent AI executive wrote that DeepSeek was a “wake up call for America.” Because DeepSeek appears to be cheaper and more efficient than similarly capable American AI models, the tech industry’s enormous investments in computer chips and data centers have been thrown into doubt—so much that the top AI chipmaker, Nvidia, lost $600 billion in market value on Monday, the largest single-day drop in U.S. history. Sam Altman, the CEO of OpenAI, said that it was “invigorating to have a new competitor” and that, in response, the company would move up some new software announcements. (Yesterday morning, OpenAI said that it is investigating whether DeepSeek used ChatGPT outputs to train its own model.)

But many prominent American researchers and tech executives celebrated DeepSeek, as well. That’s because “the most notable feature of DeepSeek may be not that it is Chinese, but that it is relatively open,” I wrote on Monday. Whereas the top American AI labs at OpenAI, Google, and Anthropic have kept their technology top-secret, DeepSeek published an in-depth technical report and is allowing anybody to download and modify the program’s code. “Being democratic—in the sense of vesting power in software developers and users—is precisely what has made DeepSeek a success,” I wrote. Start-ups and researchers love this relative transparency. In theory, competitors can use DeepSeek’s code and research to rapidly catch up to OpenAI with far fewer resources—you might not need colossal data centers to get to the front of the AI race. (The Atlantic recently entered into a corporate partnership with OpenAI.) However, there’s substantial uncertainty about just how much cheaper DeepSeek was to build, based on reports about the start-up’s hardware acquisitions and uncertainty about how the model was trained.

Meanwhile, for national-security hawks, the fear is that an open-source program that won’t answer questions about the Tiananmen Square protests could become a global technological touchpoint. DeepSeek could face similar privacy concerns as TikTok: Already, the U.S. Navy has banned its use, citing security concerns.

Any predictions, for now, are highly speculative. The global AI race is far from over, and forthcoming products from Silicon Valley could leap ahead once again. At the very least, U.S. tech companies may have to reconsider whether the best way to build AI is by keeping their models a secret.

Illustration by The Atlantic. Source: Getty.

China’s DeepSeek Surprise

By Matteo Wong

One week ago, a new and formidable challenger for OpenAI’s throne emerged. A Chinese AI start-up, DeepSeek, launched a model that appeared to match the most powerful version of ChatGPT but, at least according to its creator, was a fraction of the cost to build. The program, called DeepSeek-R1, has incited plenty of concern: Ultrapowerful Chinese AI models are exactly what many leaders of American AI companies feared when they, and more recently President Donald Trump, have sounded alarms about a technological race between the United States and the People’s Republic of China. This is a “wake up call for America,” Alexandr Wang, the CEO of Scale AI, commented on social media.

But at the same time, many Americans—including much of the tech industry—appear to be lauding this Chinese AI. As of this morning, DeepSeek had overtaken ChatGPT as the top free application on Apple’s mobile-app store in the United States. Researchers, executives, and investors have been heaping on praise. The new DeepSeek model “is one of the most amazing and impressive breakthroughs I’ve ever seen,” the venture capitalist Marc Andreessen, an outspoken supporter of Trump, wrote on X. The program shows “the power of open research,” Yann LeCun, Meta’s chief AI scientist, wrote online.

Read the full article.

What to Read Next

The GPT era is already ending: “The release of o1, in particular, has provided the clearest glimpse yet at what sort of synthetic ‘intelligence’ the start-up and companies following its lead believe they are building,” I wrote in December. The new AI panic: “The obsession with frontier models has now collided with mounting panic about China, fully intertwining ideas for the models’ regulation with national-security concerns,” Karen Hao wrote in 2023.

P.S.

After several major tech executives announced their support for Donald Trump, many liberal internet users are now alleging that they are being censored on certain social-media platforms. “To some, this pattern was as unmistakable as it was malicious,” my colleague Kaitlyn Tiffany writes. “Social media was turning against Democrats.” And they are panicking.

— Matteo

Donald Trump likes to tell his supporters that he’s a fighter, a fearless champion who always has their back. Such guarantees, however, apparently do not apply to people who worked for him when they’re threatened by foreign assassins. Former Chairman of the Joint Chiefs of Staff Mark Milley, former National Security Adviser John Bolton, former Secretary of State Mike Pompeo, and the former Pompeo aide Brian Hook have all been targeted by Trump for political retribution. They are also being targeted by the Iranians, but the regime in Tehran has marked them all for death.

The president may be spoiling for a fight with career bureaucrats and “woke” professors, but when it comes to Iranian assassins, he is willing to walk away from men who carried out his orders. Milley, Bolton, Pompeo, and Hook all served in Trump’s first administration—he appointed them to their posts—and they were part of the Trump national-security team when the United States killed Iranian General Qassem Soleimani in a strike in January 2020. In 2022, an Iranian national was arrested and charged with trying to arrange Bolton’s murder, and American intelligence believes that other officials—including Trump himself—have been targeted by Iran because of their involvement in killing Soleimani.

The Biden administration briefed the incoming Trump administration on these threats and on the security details it had authorized to protect Bolton and others. Last week, Trump removed the details protecting Bolton, Pompeo, and Hook; yesterday, Secretary of Defense Pete Hegseth removed the guards around Milley and announced that he would be investigating Milley for undermining the chain of command during Trump’s first term. Trump also revoked the security clearances held by all four men.

[From the November 2023 issue: The patriot]

The revocation of security clearances is petty, but it harms the administration more than it does any of these men. Retaining a clearance helps former federal employees find work in the consulting world, and it is typical to hold on to them after leaving government service. (I was offered the opportunity to keep mine when I left the Naval War College.) But at more senior levels, clearances allow people in government to get advice from former leaders. Some of these people could have been of significant help to Trump’s staff during a crisis, although Trump himself is unlikely to care about that possibility.

Removing the security details, however, could have deadly consequences. The Iranians seem determined to seek revenge for the killing of Soleimani, and sooner or later, they might succeed. (“The Iranians are not good but they’re very enthusiastic,” a former Pentagon official said in October. “And of course, they’ve only got to get lucky once.”) And the Iranians aren’t the only threat out there; the Russians have no compunctions about attacking people in their home country, often using gruesome methods.

Trump takes such threats very seriously where he is concerned. When Biden officials alerted Trump to the danger from Iran, Trump asked for more security from the U.S. government, and during his campaign, according to The New York Times, he even asked that military assets be assigned to protect him, something usually provided only to sitting presidents.

Lesser mortals, however, must fend for themselves: Trump and Hegseth not only took away the security details of these former policy makers but did so with significant publicity, almost as if to broadcast to America’s enemies that anyone who wanted to settle scores with these officials would get no trouble from the current White House. (Trump also canceled protection for 84-year-old Anthony Fauci, the former director of the National Institute of Allergy and Infectious Diseases, who has been the target of multiple threats from other Americans.) Trump despises critics such as Bolton and Milley, and it is unsurprising that he has no obvious issue subjecting them to physical danger. But even some Republicans —who should be used to this kind of vengefulness from the leader of their party—have been shocked, and are trying to get Trump to reverse course. They are particularly concerned about Pompeo and Hook, loyalists whose lives have been placed in jeopardy for sins that are known only to the president.

[Read: Trump can’t escape the laws of political gravity]

In another time, Americans would rally to protect their own from the agents of one of their most dedicated enemies. Today, most citizens seem either unaware or unperturbed that the president of the United States is exposing his own former staff to immense risks. Nevertheless, it should be said clearly and without equivocation: President Trump will bear direct responsibility for any harm that could come to these people from foreign actors.

This is far more than Trump’s usual pettiness. He has always considered the oath of federal service to be little more than an oath of loyalty to him, and he has always been willing to threaten his opponents. (In 2018, he apparently considered handing Michael McFaul, the former U.S. ambassador to Russia, over to Moscow, a move that provoked a level of outrage that seems quaint today.) Trump’s message in this second term is that friends and subordinates are literally disposable if they cross him: He will not only humiliate and fire them, but he will also subject them to actual physical danger.

This escalation of Trump’s vindictiveness should serve as a very personal warning to anyone willing to work for him in his second term. Senior officials at the Pentagon, the State Department, the CIA, the National Security Council, and other organizations are routinely asked to go head-to-head with representatives of some of the most dangerous nations on the planet, and to contribute to operations against those regimes. In the past, such officials could do so knowing that their own government would do everything it could to keep them—and their family—safe from foreign agents. As one of Bolton’s former deputies, Charles Kupperman, told the Times: “Trump’s national security team must provide guidance based on their assessment of what needs to be done to protect America without regard to their personal security.”

Good luck with that. No one who works in defense or national-security affairs can assume that, when Trump orders them to cross America’s many enemies in the world, he will protect them from foreign vengeance. Trump has now made clear that he will abandon people who have taken risks in the service of the United States—even those who were following his own orders—if they happen to displease him. (Or, in the case of Pompeo and Hook, for no apparent reason at all.) Hegseth, for his part, may have no real idea what he’s done, and may merely be courting favor from a boss who has elevated him far beyond his abilities. But Trump knows better; he is himself the survivor of an assassination attempt, and no level of security was enough when he thought the Iranians were gunning for him.

People still considering whether to serve Trump can have no illusions about what awaits them. True leaders take responsibility for their team. Trump is no such leader; he will, on a whim, place other Americans in danger and then, as he famously put it in his previous term, take no responsibility at all.

A new chapter is about to begin in Guantanamo's long history as a site of American international law violations. 

The first time I recall my mother mentioning Amway, we were in the car late at night, coming back from a meeting at her boss’s house. Ten years old, I’d gone upstairs to play and missed the whole point of the whiteboard sitting on an easel downstairs. My mother, however, had been rapt. Riding home with my brother and stepfather, she seemed almost to glow, as if she were throwing off sparks in the darkness.

The name Amway, she told me, was short for the “American Way.” We could sign up and buy products we already needed for the house, then sign up friends and neighbors to buy things, too. We would get rich by earning a little bit from everything they sold.

It was 1978. I didn’t realize that this was one of those moments, like Waterloo or Watergate, after which nothing would be the same. Amway—or, as we soon began to call it, the business—would become the load-bearing beam of my mother’s existence for the next four decades.

The business as then practiced in our West Virginia river town had its own culture. I found myself plunged into religious nationalism, anti-communist obsessions, denunciation of the very idea of public schools, and the worship of money. Across my lifetime, versions of these ideas would be marketed again and again to working-class Americans. Amway leaders would help elect presidents. Familiar characters from my childhood—the Amway celebrity Doug Wead, members of the DeVos family, which co-founded the company—would reappear in Republican administrations. In many ways, Amway adherents embraced a fusion of conspiratorial thinking and populism that would remain a central thread of America’s political story, prefiguring the Trump era.

But for many years, I had no context for what had swallowed my family. I had no way to understand how I’d managed to lose my mother.

Amway products began to appear around the house. We changed our laundry detergent to SA-8 and swapped our toothpaste for Glister. I rode with my mother to upline distributors’ houses to pick up the boxes that had been shipped from headquarters in Michigan. My mother and stepfather sponsored people into the business, who in turn came to our house to pick up their own orders: makeup, hair spray, a liquid soap you could use to clean anything, a portable medicine case of expensive daily vitamins called Nutrilite Double X.

My stepfather, who ran a local charity, began to introduce himself as a businessman. My mother was even more smitten with the beautiful future that Amway offered. Everywhere we went—the mall, state parks, grocery stores—she’d ask people whether they could use a little more money each month. “I’d love to set up a time to talk to you about an exciting business opportunity.” The words should have seemed suspect. Yet people almost always gave her their number. Her confidence and professionalism were reassuring, and her enthusiasm was electric, even, at first, to me. “What would you do with $1 million?” she’d ask, spinning me around the kitchen.

My mother and stepfather stayed out late on weeknights and weekends, bringing new recruits to see “the plan.” They paid to go to meetings and rallies. I had no idea at the time that these events were hosted not by corporate Amway but by high-level distributors, who were technically independent business operators. We bought books and cassette tapes by the Amway personalities Doug Wead and Dexter Yager, with titles such as Tales of the Super Rich and Becoming Rich: Eleven Principles of Material and Spiritual Success. Wead had been an evangelical minister before gaining a higher profile with Amway. Yager had sold cars and Utica Club beer before becoming one of a handful of top distributors. Their wives wrote a book together. We bought that, too.

We ended up collecting more “motivational tools” than cleaning supplies. A few people sold soap or makeup to their friends at parties, Mary Kay–style. But for us, the business mostly meant recruiting people to sign up and buy products they would use themselves, while earning points toward advancing to the next level and higher bonuses.

We became students of success, advised to set goals of a bigger house and more expensive cars, as if wishing alone could make it happen. But by this point, whatever cash we had was spent on Amway. I had a pair of bell-bottom jeans with three bright satin stripes sewn diagonally across one knee. They were the only pants I owned.

One weekend during the summer of 1980, we packed jars of peanut butter, loaves of bread, and fruit into our car, then drove 300 miles east for a rally at the Washington, D.C., Hilton. On the road, my mother and I imagined what we would do when we reached the Diamond level of the business, when true wealth would arrive.

After we checked in, my brother and I were left to our own devices, running the halls and playing in the elevators. I read a pamphlet about how John Lennon’s “Imagine” threatened America as a Christian nation, which introduced me to the (dangerous) phrase secular humanism. I listened as leading Amway distributors denounced public schools for brainwashing children.

In the hotel ballroom, distributors sang along to songs like “Rut Job Blues,” about how stupid it was to work a regular job: “I feel so D-U-M-B / I’ve got a J-O-B.” Cheers went up at any mention of Ronald Reagan, who had embraced Amway for years—and would soon be president. (A few years earlier he’d told a crowd of Amway distributors that “for me to come here and talk to you about free enterprise is like saving souls in heaven.”)

We went to more rallies—in Pittsburgh, Cleveland, and other faltering Rust Belt cities where people were laid off and looking for hope. We ate up testimonials to God’s grace and to his desire that everyone should become as rich as possible. High-ranking distributors encouraged low-level distributors like us to Drop that stinkin’ thinkin’ and Fake it till you make it.

At one rally, my brother and I ran into Doug Wead’s son, who was about our age. After walking around the hotel, the three of us sat in our room and talked. I said how great it would be when our mother and stepfather became Diamonds, so we would be rich, too.

He told me I had it all wrong. His dad didn’t make serious money through Amway products. Most of what he earned came from writing books and recording talks. That was how people got rich in Amway—selling motivational books and tapes to distributors like my parents. Didn’t I know?

He spoke honestly, without malice, and the words rattled around in my brain for the rest of the trip. I picked at the upholstery on the seat of the car on the ride home. We would never be rich. There was no other plan. We were doomed.

What was it about Amway that so captured a bright, extroverted woman like my mother? Abandoned as a child when her own mother ran off to become a nightclub singer, she’d been raised by her grandparents. She graduated high school with a journalism scholarship to college, but met my father that summer and never left town. She became a stringer for the local paper, later working as a lunchtime anchor and interviewer for our local television station. When I was a preschooler, she took night classes and earned a bachelor’s degree in social work. By the time she discovered Amway, my mother had divorced and remarried. My stepfather had a more fundamentalist view of religion than I had been raised with—a view that dovetailed with many Amway leaders’ emphasis on biblical literalism and wives submitting to their husbands.

My mother couldn’t imagine life without a husband. More crucially, she believed herself destined for something extraordinary. But how could someone achieve greatness in Parkersburg, West Virginia? Amway promised to deliver what nothing else in our town could—or at least to give her a community that would pretend along with her.

For some Americans, joining the business might have been harmless. For us, it was not. Soon my mother and stepfather had no other job. Their bad decisions ricocheted in the echo chamber of Amway culture, where they were encouraged to dedicate themselves more deeply. Surely, any day now, we would make it. Within three years, we were living in a filthy house without electricity, eating food out of a cooler that we kept filled with ice. Then we were evicted, and my mother and stepfather declared bankruptcy. Ordinary people might have thought twice about sticking with Amway. But by that point, we had left the small dreams of ordinary people behind.

A few months later, we climbed in a van headed to New York to stay at another Hilton. It was New Year’s Eve. My parents went to see the Rockettes and to hear the same speakers they’d cheered on in other cities, singing songs, giving glory to God, and talking about his vision for America.

When I was a teenager and my mother was in her early 40s, she stopped talking to me about Amway. She filed for divorce from my stepfather and started a graduate-school program in behavioral psychology in hopes of becoming a therapist.

Despite being more than a decade older than her classmates, she was well liked and a good student. My brother and I had already escaped to college, thanks to cobbled-together loans, grants, and multiple part-time jobs. I didn’t talk to either of them often, because in 1988, long-distance phone calls were expensive. But my mother called one day to chat.

“Going crazy isn’t like being hit by a car,” she said in the middle of our conversation. “People make a small but conscious decision to give up. At some point, it’s easier than living in reality.”

She was deep in clinical work with the mentally ill at the time; I assumed she was drawing on that experience. Still, the line stayed with me. In recent years, I’ve wondered whether she was talking about herself, and whether there might have been some way to intervene that I didn’t see. Because, just two years later, in the last semester of her Ph.D. program, my mother decided to quit and marry a third husband, one who would do Amway with her.

Only much later would I hear stories about distributors like us who had declared bankruptcy and begin to understand how common our experience was. A 1980 study of tax returns conducted by Wisconsin’s attorney general showed that the top 1 percent of Amway distributors in that state had lost, on average, $900 in the business. In 1994, Dexter Yager and Amway faced a class-action lawsuit claiming that they had fraudulently misrepresented how much distributors were likely to earn and illegally pressured people to buy books and tapes. The case was settled with Amway promising compensation and changes that would require distributors to make clear that motivational tools were optional and didn’t guarantee success. The FTC had determined in 1979 that Amway was not a pyramid scheme, but the company continued to face allegations to the contrary. In 2010 it settled another class-action suit alleging that it operated a pyramid scheme. The company did not admit to guilt but did agree to pay plaintiffs $56 million, in the form of cash and Amway products.

In the years that followed, my mother and I would sometimes talk about real life—a birth, a death, a grandchild—and flashes of who she used to be would shine through. But she also shared long lists of people the Clintons had supposedly murdered, and continued to insist on Amway’s tremendous potential. She always sounded a little embarrassed by the things she said, as if she understood that they were hard to believe. I think she wanted me to see that she knew that the most cultlike aspects of the business were over the top, that she hadn’t been taken in entirely, that she wasn’t some kind of fool. But it didn’t matter. In the end, Amway owned her as fully as if she’d believed every word. Despite interventions my brother and I attempted, despite the money she continued to lose year after year, our mother never gave up on the business.

Illustration by Anthony Gerace

When I tell people how I grew up, I get a few different reactions. Sometimes I meet people who thought about joining Amway, and are relieved they never signed up. Sometimes they’re surprised that Amway still exists—they thought it disappeared decades ago. Most barely know what it is. And why should they? They themselves might never fall for such a hustle. But whether they know it or not, Amway has deeply influenced American politics for decades.

Amway supported Reagan’s candidacy in the 1980s. In the ’90s, a co-founder of the business, Rich DeVos, gave the GOP what was believed to be the largest-ever-recorded individual political donation. Less than a decade after I first listened to him on Amway tapes, Doug Wead became Vice President George H. W. Bush’s liaison to right-wing Christians. The Bush-era term compassionate conservatism may have been an Amway invention—Wead is said to have coined it. Dexter Yager, who had paid Reagan and Bush to speak at his events, reportedly mass-distributed voicemails pushing support for Republican candidates and accusing Bill Clinton of trying to “force the emergence of deviant lifestyles, of a socialist agenda.”

I grew up hearing rumors about the satanic influences motivating Procter & Gamble, which Amway considered a business competitor—stories that led to another lawsuit and required distributors to pay $19 million in damages. Amway didn’t invent the art of communal delusion via disinformation—the John Birch Society had already perfected it in the 1960s. The Birchers’ influence was in decline by the time we joined the business, but Amway’s culture helped carry their unhinged style into the digital era.

In 2021, Doug Wead died. At the time, he was under federal indictment—not for anything related to Amway, but for allegedly funneling Russian money into Donald Trump’s 2016 campaign. In Trump’s first administration, he nominated Betsy DeVos as secretary of education. An advocate for school choice and religious education, she is married to Rich DeVos’s son, Dick, who was president of Amway himself in the 1990s, and whose family still co-owns the company. She said she’d be open to returning to the post, “with the goal of phasing out the Department of Education.” The rallies leading up to Trump’s latest election, with their euphoric resentments and tent-revival energy, recalled nothing so much as a 1980s Amway function.

My mother had fallen so deep into the delusional communities of Amway and religious extremism that I took a while to realize she was developing dementia. Her Alzheimer’s manifested in part as paranoid psychosis. Over time, as her memory failed and her sense of her own importance ballooned, she exchanged my actual childhood for one in which we’d been staggeringly wealthy. She had once been engaged to Trump, she told me. When a court-appointed attorney came to assess her legal competence, my mother threatened to have Trump fire her. For months, my mother believed she was working as Trump’s campaign director for Ohio and Michigan. They had met through Amway, of course.

It’s hard to leave a delusion behind. In the run-up to the 2024 elections, I noticed the ways in which Trump’s political followers likewise struggled to abandon him. Some prominent Trump supporters may see him as a means to wealth or power. Others find meaning and community—or even vindication—in accepting the lies he tells. Maybe, eventually, when they see what his second administration delivers, some voters will peel away.

That’s what happened with Amway. The company is still a multibillion-dollar, global enterprise, though its domestic profile is now so much smaller that it has a page on its own website answering the question: “Does Amway still exist?” In the end, more people left than stayed. Those who came to their senses or were unable to sustain the delusion eventually quit. But things can get bleak in the middle.

My mother was an outlier. As the illness devoured her mind, she stopped recognizing her friends. But she still remembered the business. At the beginning of 2020, just three weeks before the pandemic began, I brought her to live with me and my brother in Virginia. She set off the fire alarm and constantly announced that the belongings she’d misplaced had been stolen. But the hardest part was her insistence that we all inhabit her imaginary world—one where she lives in grievance and terror, a place of invented enemies.

When I cleaned out her old house for her, I found storage shelves in the basement filled with Amway binders, makeup tutorials, old catalogs, and hundreds of motivational CDs and cassettes. Like some ritual to release the dead, I emptied the binders one by one. I filled a dozen Hefty bags, and then more. When the outdoor bins could no longer contain the trash, I stacked the rest on the ground by the curb: relics that would help no one, souvenirs of a lost life.

Until just moments before an American Airlines regional plane and an Army helicopter collided over the Potomac River last night, nothing in particular seemed amiss. Conditions were clear, Sean Duffy, the new secretary of transportation, noted in a press conference this morning. The passenger jet, coming from Wichita, Kansas, was about to arrive at Ronald Reagan Washington National Airport—one in a succession of airliners landing about two minutes apart. The Black Hawk helicopter was on a training mission from Virginia’s Fort Belvoir. Both aircraft were in a “standard flight pattern,” Duffy said. Referring to the crowded and shared air space around D.C., he added, “This was not unusual.”

And that may turn out to be the problem. The precise immediate cause of the crash—which killed all 64 passengers and crew members aboard the airliner and all three people in the helicopter—will not become clear until investigators fully analyze recordings of air-traffic-control communications and the plane’s black box. But the accident follows a long string of alarming near collisions at airports across the country—a pattern suggesting that the aviation-safety systems upon which human life depends are under enormous strain.  

[Conor Friedersdorf: Don’t politicize aviation safety]

In 2023, the Federal Aviation Administration identified 19 “serious runway incursions,” the most in almost a decade. The causes of these events are varied: air-traffic-control staffing shortages, pilot inexperience, demand for air travel, outdated technology. The increase in near misses led the FAA to create a safety review team and issue a rare industrywide “safety call to action” demanding greater vigilance throughout the community. These incidents do not appear to have prompted any major changes in safety practices either nationally or in the Washington area. Last year, the number of serious incursions declined, making the issue seem less urgent.

Reagan National’s tight footprint and three intersecting runways, along with the presence of military and other government operations nearby, make the air space surrounding the facility relatively tricky for pilots to navigate. As the popular open-source intelligence account @OSINTtechnical noted on X once footage of the accident and its aftermath began spreading on social media, “For many in the DC-area flying community, the crash tonight wasn’t a matter of if, but when.” (This morning, Defense Secretary Pete Hegseth announced an investigation of the Army helicopter’s role in the incident.)

In 2013, an airliner and a military helicopter flying at the same altitude near Reagan National came within 950 feet of each other. Last May, a Boston-bound jet traveling 100 miles an hour on the runway had to abort its takeoff because another plane had been cleared to land on an intersecting runway. Even so, the FAA added additional flight slots to Reagan National last year, over the objections of local politicians who worried about congestion and overburdening capacity.

[Read: Inside the busy, stressful world of air traffic control]

The crash near Reagan National was the first major aviation disaster involving a U.S. airline since 2009—long enough that nearly a generation of Americans are experiencing this crash as their first. Such incidents have become so rare that Americans come to assume that safety precautions automatically work.

Safety systems are vulnerable to a phenomenon known in the disaster-management world as the “near-miss fallacy”—an inability to interpret and act upon the warnings embedded in situations where catastrophe is only narrowly avoided. Paradoxically, people may come to see such events as signs that the system is working. In her groundbreaking research on NASA after the space shuttle Challenger exploded in 1986, the American sociologist Diane Vaughan faulted the agency for its “normalization of deviance.” The direct culprits in the spacecraft’s fate were faulty booster-rocket parts known as “O-rings.” Vaughn noted that shuttle missions had been experiencing problems with the parts for years, but NASA had downplayed their importance. Engineers were able to normalize O-ring incidents and other safety issues because none had caused significant harm—until one did.

The immediate cause of the crash over the Potomac may turn out to be a single tragic mistake. But this deadly tragedy occurred within a broader context. For some time, our aviation system has been ignoring warning signs and normalizing deviance. Good luck can last only so long, and it ran out last night.

An army helicopter crashed into an American Airlines plane carrying 64 people including staff.