For a long time, I went to bed and imagined falling into a black hole.

I was young. My parents had bought me a thick blue book called something like 1001 Facts About Space. One page explained that inside every black hole, as matter spiraled past the fluttering edges of rent time and space, there was a boundary called the event horizon. At that point, the force of gravity became so strong that nothing, not even light, could escape its grasp.

We don’t know what lies beyond the event horizon. We can’t know, unless we travel into a black hole ourselves — and even then, we wouldn’t be able to tell the people we left behind what we saw. Observers on the outside of the event horizon cannot receive signals from the other side.

I liked the other thousand facts, but I didn’t like the uncertainty of this one. Surely, matter couldn’t just fall, forever, into nowhere. At night, I would close my eyes and slip into the abyss; I would turn around to face the vanishing universe and watch the stars blur into a bright blue eye. And then it was dark, which didn’t make sense at all. Beyond the event horizon, there must be somewhere to go.

Subscribe now

So I solved the puzzle. I drew a red crayon diagram of a man falling into a black hole, crossing the event horizon, and emerging into another place. I don’t remember what I thought this other place was: a different location in spacetime on the other side of a wormhole, the same black hole that the man had entered in the first place, or a new universe entirely. But I do remember showing the diagram to my barber, with whom I was very much in love, and feeling my heart race when she told my mom that I was a little Einstein.

My parents helped me mail the diagram to NASA. They never responded — for a few years, I would periodically check their website for news about black holes, certain that they had stolen my theory and would soon pass it off as their own.

Of course, to think that you can do what a century’s worth of astrophysicists could not, you must be wonderfully arrogant. Which I was; most children are, and this naiveté is probably the most important feature of their genius, the reason that they are at times capable of intellectual insights beyond the reach of adults.

By middle school, however, my baseline pretension had ballooned into a seething tower of hubris, expanded and sustained by the adult praise and dominant early grades that cripple many a gifted child.

At some point, I did poorly on a few math tests. I decided I was not a genius. So I stopped thinking about space. There were girls to impress and friendships to maintain; the stars were not meant for puny minds like mine.

Sometimes I feel like my entire life has been spent trying to turn back into the seven-year-old who thought he could tell NASA the truth.

The egg and the rock

Which is, in part, why I have lately found myself fascinated by , a writer-turned-cosmologist hailing from England and Ireland. Some of you may know Julian’s Substack, : there, he writes about his theory of why the universe works the way it does, as well as the empirical predictions that follow from the theory. Others will know him, indirectly, through his “End Poem,” the verse dialogue that plays at the end of Minecraft. It is probably the most-read poem of all time.

I spent this summer writing for Smithsonian magazine. I am proud of everything I published, but even before starting at the outlet, I knew I wanted to write an article that would encourage more people to pay attention to Julian. That article is out today. In it, I wanted to get readers thinking not just about his theory— though that’s the crux — but Julian himself, as a thinker. Unlike most working scientists, Julian has preserved the spirit of a child falling into black holes in his dreams. More people should live and think like him.

Julian has no academic credentials. Prior to his cosmological research, he was a novelist, poet, musician, and children’s book author.

But about a decade ago, he became obsessed with the striking fact that our universe is becoming more complex over time. Everything started as a ball of hot gas. The hot gas built out stars and galaxies, which in turn generated the planets and heavy elements necessary to sustain life, including life that is smart enough to figure out that everything started as a ball of hot gas. As Julian put it in one of our interviews: “That’s a very strange thing for hot gas to do.”

The laws of physics also seem weirdly hospitable to complexity: for instance, if certain physical constants changed even slightly, atoms wouldn’t be able to form molecules. Cosmologists generally treat this as the product of chance.

But Gough wondered whether it might instead be the product of evolution—when we observe complexity elsewhere in nature, evolution is usually the explanation. As Julian put it: “Why wouldn’t ‘evolution’ apply to universes when it applies to everything else?” He theorizes that the universe is complexifying because, like any naturally selected entity, it evolved to develop into a form that would maximize its reproductive potential.

Yes, this implies that universes can make babies. But Julian was actually not the first person to suspect that this was possible. John Wheeler and Bryce DeWitt hypothesized that the singularities within black holes — the hypothetical place where matter is crushed into a point of infinite density — unfold into Big Bangs, which also begin with singularities. It’s quite intuitive: after all, black holes are the only objects we know of that make singularities, and universes are the only objects we know of that seem to emerge from them.

Later, Lee Smolin made the obvious leap — if Wheeler and DeWitt were right, then universes could undergo a kind of natural selection. Specifically, they would optimize over time to make as many black holes as possible, just as bacteria and rabbits and people have been optimized to make as many babies as possible.

Early universes would not have been very fertile, because their physics would not yet have been fine-tuned. But so long as they could make a couple of black holes that would create baby universes with slightly different physics, natural selection would kick in. As the generations ticked by, universes would rush toward virility, and eventually reach a near-optimal point.

What might that near-optimal point look like? Presumably, something like our universe, which is estimated to contain 40 quintillion black holes. If Wheeler, DeWitt, and Smolin are right, we must lie pretty far down the evolutionary line.

The implications of so-called ‘cosmological evolution’ get even weirder. The complexification of the universe eventually settles on life. Why? Perhaps, as various futurists and philosophers have suggested, because life needs energy to survive, and certain life-forms — like humans — are smart enough to find more and more efficient ways to acquire it. The most efficient way of all is not coal, oil, solar energy, or even nuclear reactors. It is an artificial black hole.

So if intelligent life can climb the energetic chain and not destroy itself in the process — so far, so good — it’ll eventually become the ultimate reproductive organ for the universe. We will make black holes to fuel our civilization, and in doing so, we will make many babies for our universe.

None of these ideas originate with Gough. He knew his revelations were hand-me-downs. But he also noticed that cosmological evolution had basically vanished from mainstream science. No one was talking about Lee Smolin. The main problem, it seemed, was that the theory did not make easily testable predictions — and without the prospect of falsification, there is only so much you can do with a great scientific idea.

But Gough was thinking about these things three decades after Smolin: times had changed, and telescopes had gotten bigger. By 2022, the James Webb Space Telescope was preparing to send its first batch of data back to Earth. Unlike any telescope before it, Webb could see into the first billion years after the formation of the universe. If cosmological evolution made any predictions about what Webb should observe, falsification was back on the table. Gough’s first major contribution was to make that set of predictions.

If universes evolved, Gough reasoned, their infancy might preserve the simpler reproductive mechanisms that their ancestors must have used. The form he settled on was direct collapse: the formation of supermassive black holes out of the uniform clouds of hot gas that dominated space just after the Big Bang.

Working cosmologists had already theorized that direct collapse was possible, albeit rare. Gough predicted that Webb would see it happening regularly and extremely early: giant black holes would form through direct collapse after only a couple hundred million years, and spiral galaxies would form around them.

None of these ideas required any grand leaps of genius: treat universes as evolved organisms, and these predictions all easily follow.

What they did require, however, was a grand leap of audacity. Gough had never published an academic paper. How humiliating for someone like him, a scientific nobody, to publish his predictions and end up entirely wrong. Novelist, poet, musician, crackpot — an unfortunate addition to his Wikipedia page.

Four days before JWST was scheduled to release its first batch of data, he wrote a post on Substack detailing his predictions. He was terrified. But he did it anyway.

Wake up

As you probably guessed, this story has a happy ending — I wouldn’t have been able to write about it for Smithsonian if it didn’t.

Gough’s predictions were right. Webb has found spiral galaxies forming around black holes in the first few hundred million years of the universe, including one that must have begun forming only one hundred million years after the Big Bang. In a 2025 paper published in the Astrophysical Journal, a group of scientists described these discoveries as a “conundrum [that is] part of the larger challenge to understand the stunning prevalence of massive structures and galaxies in the first few 100 million years after the Big Bang.”

Because of his public success, he has since received support from private grants like Emergent Ventures and O’Shaughnessy Ventures, and he has been endorsed by many renowned scientists. He has continued to develop his models of early universe formation, which I encourage you to read about in his words. I remain shocked that, so far, only one other mainstream outlet has written about his work.

I’m hoping that my article will encourage other institutions to take cosmological evolution seriously as the explanation for these discoveries, and in particular, other cosmologists. It is a beautiful idea. It accords with everything we know about nature. If it is true, it bears serious implications for the future of human (and alien) civilization, and maybe even the nature of morality.

But if Gough had been wrong, I still would’ve wanted to write this piece.

As I’ve discussed elsewhere, the essence of great science is contained in the curiosity, grandiosity, and naïveté that characterize the minds of children. Yet the academy crushes these qualities out of you: it rewards methods over ideas, discourages research that is important enough to potentially result in total failure, and coaxes you toward a kind of incrementalism that rapidly collapses into mere social games.

Gough is the antidote to this trend of cynical craftsmanship. He has protected the child-like spirit that all scientists—hell, all humans—should aspire to keep. If he was wrong, then fine. At least he would’ve been wrong about something worth being wrong about. I dream of a future science full of people like Gough, working rigorously to place themselves in positions that bear both the possibility of great insight and the risk of total embarrassment. Science is a strong link problem, and strong links are forged in the fires of likely humiliation.

Where will we find those willing to stick needles in their eyes, stare directly at the sun, give all of themselves to the chance of finding something true? Perhaps not in the academy, not anymore. But it is good to know that they still exist.

The personal irony — that despite abandoning my early love for astrophysics, I have now circled back to black holes — is not lost on me. They still capture my imagination; there is something about them, perhaps the scope or the circularity, that makes me think of the human self. I could probably end here with a quip about event horizons.

But I’d rather finish by talking about the End Poem. When I beat Minecraft for the first time, I was a teenager. I had stopped reading books and thinking about science. The End Poem, I remember well, was one of the first pieces of literature that made me feel a sense of the mystery again. “Respawn in the long dream,” Julian wrote. “Wake up.”

After one of our interviews, I had the opportunity to thank Julian for writing that poem. I also pointed out that it seems to converge on the same idea as his theory of the universe— we living things are the universe, unfolding itself into eternity. His son had wandered onto the screen; Julian was dressed in entirely neon green. He smiled and said that when he wrote the poem, he did, in fact, feel like something was speaking through him. “The universe talking to itself,” as he wrote in the poem. “Reading its own code.”

They say don’t meet your heroes. I say, pick better ones.

I love Lionel Messi. My brother and I grew up with him as a sort of demigod flitting across our televisions, video game consoles, recess scrimmages, and daydreams. I also love reading essays about Lionel Messi: it’s nice to see your feelings reflected in someone else’s mind, especially if they’re expressed in words better than the ones you would’ve chosen.

Subscribe now

Messi is playing in the World Cup final on Sunday — it is likely the last major game of his career. So, feeling nostalgic, I was excited to see this piece published in the New York Times.

A Guest Essay written by a sports writer and poet about the legacy of Lionel Messi? Shove it directly into my eyes. What does Rowan Ricardo Phillips want to tell me about La Pulga?

Huh. His style seems so familiar . . .

If you’re skimming, the metaphor almost makes sense, maybe? But you’d expect something more coherent from an author whose work has been longlisted for the National Book Award.

Ah. There it is.

Surely, you see now what the editors of the most important newspaper in the world missed. Here’s the score from Pangram for the first half of the article. If you’re unfamiliar, Pangram is the only reliable way to check if something is written by AI. It preferentially assumes that words were written by humans. False positives are a bigger problem than false negatives, according to its philosophy. So if Pangram says that a piece is written by or with AI, then it almost definitely was.

But you don’t really need Pangram to tell what Phillips is doing. His essay abounds with tripartite structures, fuzzy metaphors about quietness and concealment, and big paragraph breaks that set up quips so blasé you might as well just douse your phone screen in tapioca.

Maybe I’m overestimating how obvious the cues are if you don’t spend lots of time reading words generated through linear algebra. But come on — who writes like THIS about Messi’s sudden retirement from the national team?

How moving. No interest in writing about the Argentines who burned Messi’s jersey in the streets, his empty stare as he announced his retirement to the world, or the subsequent nationwide exhortations for him to return (including one from his future teammate, a then-15-year-old Enzo Fernández)? Evidently not. He said he was finished . . . and then . . .

Bravo, Rowan.

In principle, I’m actually not opposed to AI-generated journalism. Major outlets publish lots of articles that ChatGPT could write already: for instance, the standard “New study says X, and it means Y” content that passes for most of modern science writing. I don’t care whether humans take charge of articles that mainly exist to transport information — the job of the author, in these cases, is spiritually akin to the job of an old-timey telephone operator. In the next ten years, perhaps much sooner, even the esteemed Gray Lady will automate those segments of her production.

But as Sasha Chapin has discussed, information transfer is not the only reason for us to write, or even the most important reason! Long-form text is not an efficient way to move information from one brain to another: that is not the medium’s comparative advantage. Instead, text is near-optimal for mind-merging. There is no other medium equally good (for now) at simulating someone else’s experiences in your head. We write to share our souls; we read to receive the souls of others.

Of course, I often do read for the sake of transferring information. But increasingly, I rely on AIs to generate the text — you can ask follow-up questions, ask for different analogies, create questions to test your knowledge, and so on. In that writerly province, they are already as good as or better than most human beings.

So if I clicked on an article announcing the results of a new scientific study and then realized it was written by ChatGPT, I wouldn’t care. I want all societal functions executed as well as possible. If AI executes some of them better, great. I can live with that, just as I can live with the fact that we no longer have telephone operators.

But I didn’t click on this essay about Lionel Messi so I could learn something new. I know everything about his career — there is simply no statistic, factoid, or anecdote the NYT could publish about him that I haven’t already memorized and debated with my brother. I clicked on the essay because I wanted to experience Messi through the eyes of another human being, one who (being a poet and award-winning author) should be uniquely equipped to scaffold that simulation. Even when we reach the point where AI writing can bring us to tears, an essay about the shared joy of watching great athletes will still need a human author.

The entire purpose of such an article is to bring a few members of Homo sapiens together. This is a domain where AI shouldn’t trespass — not because it can’t write well enough (it can’t yet, but it will), but because the function of the writing literally cannot be fulfilled by artificial intelligence. Nevertheless, Phillips persisted.

“Thank you for your beautiful essay,” says one commenter. “Watching Messi play was like watching poetry in motion. And reading your words put that poetry in motion on paper. Thank you for sharing.” I imagine an old man tapping out his gratitude with two fingers. Maybe he sends the article to his grandchildren. Maybe the grandchildren, better versed in this new world, explain to him that he is communing with no one at all. The old man feels embarrassed. He tries to remember when it became so hard to tell real from fake, or when that even became a meaningful question.

Do the editors of the New York Times disagree with this analysis? I doubt it. They probably just haven’t figured out how to make sure they aren’t being deceived.

Which is a bit surprising, given this whole debacle, in which essays published in the paper’s “Modern Love” column (pieces about relationships, feelings, betrayals, and revelations — you know, human experiences) were clearly written by AI.

Not to mention the Times’ various articles and wacky games crusading against AI-written slop:

Perhaps the New York Times would like to avoid these situations in the future. I’ve heard that the editors take themselves pretty seriously. So, in case any of them happen to see this — here’s one simple way you can make sure you’re not scamming your readers into merging with matrix multiplication. It takes about ten seconds.

In 2019, before ChatGPT, Phillips wrote an essay about Messi called “They Think They Know You, Lionel Messi.” I read it in the Paris Review. It was beautiful. It made me feel like I knew him.

In the spring of 2024, Dr. Martha Weiss checked her mailbox at Georgetown University and found a large envelope postmarked from Japan.1 Inside was a handwritten letter. Here is what it said:

To Martha Weiss,

Hello, nice to meet you! My name is Jo Nagai. I’m from Japan. I live in Kobe, Japan. I’m in the second grade at Ibuki Elementary school. When I found your research on the internet, I was so delighted!!

Subscribe now

Weiss is an entomologist. Her most famous study, coauthored with Doug Blackiston and Elena Casey, showed that moths keep memories they form as caterpillars. This was remarkable, considering that when caterpillars wriggle into their chrysalis and metamorphose, their brain dissolves into goo.

The paper inspires most of Weiss’ fan mail; on this count, Nagai’s letter was no exception. But he wasn’t writing as a fan. Instead, he wanted scientific advice. He was also an entomologist, and he was going to replicate her experiment with butterflies. As he wrote:

I’ve studied swallowtail butterflies for three years . . . I’ve always thought that my butterflies could remember me, even after their metamorphosis, because they always flutter around me when I try to let them go into nature. Sadly, some say that’s impossible and ridiculous.

I have some questions to you. Have you ever experimented with swallowtail butterflies? I want to try to find if a swallowtail butterfly could remember what it learned as a caterpillar.

Weiss was delighted by Nagai’s proposal. In her reply, she encouraged him to continue his research and promised to advise him however she could.

Of course, she did not expect him to actually carry out the PhD-level experiment he was proposing. To test the persistence of memory in moths, she and Blackiston had trained caterpillars to associate the scent of ethyl acetate with electric shocks, then examined whether they were still afraid of the chemical after metamorphosis. Weiss suggested Nagai test whether his butterflies could remember colors instead.

Nagai’s response arrived a couple weeks later — he thanked Weiss for her time and insisted that he would replicate her experiment properly.

They corresponded for the rest of the summer. Near the end, Nagai sent her a 33-page manuscript documenting his results. He had used lavender oil instead of ethyl acetate, and he had adapted a muscle therapy device to provide electric shocks. His results showed that butterflies, like moths, kept the memories they form as caterpillars.

There is a long history of children contributing to mathematics and theoretical physics: my favorite is Évariste Galois, who laid the foundations for group theory at 17 and then died in a duel at 20. But off the top of my head, I can’t think of a single child or teenager — with the exception of Nagai — who has advanced our understanding of biology, psychology, or any other empirical science.

Why not?

Maybe when you’re older

One common explanation is that children are underdeveloped in ways that make them bad at empirical sciences. Maybe their executive function isn’t powerful enough to plan out long-term projects. Maybe they’re not good at synthesizing lots of information. Maybe they lack the metacognitive awareness to realize when their reasoning is flawed.

I’m open to this account of why children don’t write great novels. But I don’t think it’s a good account of why they never publish biology papers in Nature. Those same cognitive limitations seem like they would make mathematics difficult too. And yet, every couple of months, Quanta publishes an article documenting some teenage mathematical breakthrough that makes me want to throw my brain in the garbage disposal.

A better argument is that in mathematics and some branches of physics, it is easy to identify a problem, work on it, and know if you’ve found the answer. This doesn’t mean “easy” in the sense of “anyone can do it” — rather, it means that problems are clearly stated and widely identified, and there are procedures in place that can tell you definitively whether you’ve solved them.

Take Daniel Larsen: he is one of the aforementioned prodigies described by Quanta. In his senior year of high school, Larsen published a proof resolving a decades-old conjecture about a rare class of numbers.

So, good for Larsen. But note two things. First, he did not invent the problem he solved. Like most open problems in mathematics, it had been sitting there for decades, clearly stated, waiting for someone clever enough to crack it. Second, his solution required no experiments to verify that it was correct, because a mathematical proof is inherently either right or wrong.

In this sense, mathematics is “well-structured”: it is clear where you should start, what kinds of rules to follow, and whether your ideas are correct. According to some popular arguments, well-structured fields are perfect for young prodigies, because the only barrier to entry is raw ability. The rules of the domain are clear enough that a child can simply absorb them and get to work.

Fields like biology are instead arguably “ill-structured.” There are open problems, but you can’t actually work on them without figuring out a more specific and actionable version. Maybe you want to solve the origin of viruses — but “the origin of viruses” is something you figure out over the course of an entire research program, not a few months of desk work.

Nor do your experiments, however elegant they are, always tell you whether you’ve actually answered a question. The accuracy of a mathematical proof is not debatable. But as anyone working in the life sciences will tell you — and especially, shudders, in the social sciences — researchers can and do argue about how to interpret empirical results until the coffin door finally shuts them up. So if you want your results to be taken seriously, you need to build rigorous procedures and analyze your data like someone tall enough to look through a microscope.

I think this is a neat explanation of why we don’t have empirical prodigies. I also think it is probably wrong.

Smart kids play with worms

As framed, asking and answering questions in an ill-structured field seems like a skill you need to build over time. On this account, doing great biology is like writing novels — kids can’t do it well, because they haven’t built the relevant muscles.

If this is true, however, then we should also be able to imagine the following scenario. Jo Nagai — still ten years old — receives a position at Georgetown University as an independent researcher. He is free to work on whatever he wants; the university gives him some money; and he can collaborate with more experienced scientists and technicians on campus. Basically, he’s a pretty well-off PhD candidate. But even with these resources, his research fails, because children can’t perform well in ill-structured fields.

Would you bet against Nagai? I wouldn’t, and not just because he did great work without any of the above. Rather, I think the parts of empirical sciences that close them off to child prodigies — namely, the fuzziness of questions and answers — are not actually hard for young brains to handle. The barriers are not cognitive limits: they are resource limitations wearing a fake mustache and glasses. With money, facilities, and the right advice, ill-structured fields become structured just fine for the likes of children.

Obviously, empirical questions in these fields do not work as neatly as in math: empirical scientists can’t literally scan a list of actionable open puzzles and pick one. But I still doubt that it’s tough for a child to come up with a good project. Often, great researchers ask questions like little kids.

My favorite example of a scientist with this trait is Michael Levin. He’s a developmental biologist at Tufts University; many people think he’ll win the Nobel Prize someday.

Here are a few of the questions Levin has worked on over the years:

• If you put eyes on the tail of a tadpole, can it see through them?

• How does an embryo know which part of it will become the left side?

• What happens if you take cells from a frog embryo that are destined to become skin, plop them in a petri dish, and let them figure out what to do next?

• When you cut off a flatworm’s tail, it will correctly regenerate a tail. Can you make it regenerate a second head instead?

How sure are you that a classroom of second-graders, given the chance to watch a flatworm regenerate its tail, wouldn’t also ask whether you can make it grow another head? I’d bet on those kids coming up with the question over a bunch of grown-ups. If they did, it would be big news — having answered this puzzle, Levin may well end up with a free trip to Sweden.

Thinking like a second-grader is the most underrated element of genius. Unfortunately, academic adulthood tends to stamp out these parts of your soul. A child-like brain can drag you to embarrassing places, and feeling embarrassed is scary; that’s probably why so few working scientists seem to maintain the spark. But those who keep that fire alive sometimes see the basic questions that people have been too scared or unimaginative to ask.

If adult scientists can succeed by thinking like second-graders, why not the second-graders themselves? In fact, extensive evidence from experimental psychology suggests that children might be even better than adults at solving problems that have weird answers.

Consider the blicket machine. Designed by the Berkeley psychologist Alison Gopnik and her colleagues, the machine lights up and plays music when clay shapes called “blickets” are placed on top of it. In one seminal experiment, Gopnik’s lab asked preschoolers and college students to figure out which arrangement of blickets would turn on the eponymous machine.

You might assume that college students would trounce preschoolers in any problem-solving task. And sometimes they did: if activating the machine required a specific blicket, the college students were better at finding the right one. But if activating the machine required a strange combination of blickets, the preschoolers trounced their wizened peers. The kids were better at coming up with unusual ideas.

Ten years later, the data supporting that thesis extends across dozens of studies. Not that this is surprising if you’ve spent much time around children.

What happens if you’re willing to entertain ideas that other people don’t even imagine in the first place? In science, something like Jo Nagai’s subsequent study — after proving that caterpillars keep their memories through metamorphosis, he wrote to Weiss and told her that he wanted to see if they would pass those memories onto their offspring. Weiss said she had never even considered that possibility.

Nagai ran the study. He was right: the offspring of his caterpillars displayed the same learned behaviors as their parents.

So, fine — children can come up with good empirical ideas. But can they answer them without supervision? For some ill-structured fields, the idea is arguably less important than the execution. Everyone has a great idea for a movie. That doesn’t make them Steven Spielberg.

In the empirical sciences, though, I don’t think it matters whether kids can design, run, and analyze the experiments entirely on their own. In my department, many graduate students (and even some professors) don’t know how to do advanced statistics. That’s fine, because we have a department statistician — he’ll meet you anywhere on campus and help you figure out how to process your data. Recently, I wondered if a bacterium could learn Pavlovian associations. I don’t know any of the techniques needed to figure that out. So I emailed an expert, and he’s helping me do the experiment. Science, like all human enterprise, relies on a division of labor.

If you have to call upon those divisions, are you incapable of doing good science? Obviously, no. My brother runs a very successful company. He doesn’t personally know how to manufacture their products — that job falls to the people who run his factories — but no one would say that his lack of methodological knowledge means he can’t build a business.

If you’re agentic and can find the right people, methods are cheap. Ideas are the important part.

Scientific progress goes boink

But if ideas are the important part, and children are great at coming up with ideas, the question remains: why aren’t there more kids like Nagai?

Because methods are cheap, but not free. If you’re ten, and you want to run a study, you need some or all of the following: money to buy equipment, advice from adults who can advise you on experimental design, access to papers locked behind paywalls, analytical support so you can actually do something with your results, and an avenue to release your work to the public.

A determined kid like Nagai can find their way around these institutional and financial obstacles. But it certainly takes a level of agency, discipline, and probably parental support that not all children possess. This is an unnecessary waste of interesting questions — the empirical sciences are in fact ill-structured for kids, but their ills seem pretty arbitrary. We could certainly use more weird ideas. How many Nagais have we already missed?

In a perfect world, we would set up a governmental fund to support the good works of the youth. You may have noticed that we don't live there. Even in our imperfect world, though, there are solutions. Programs like Emergent Ventures and 1517 Fund already offer microgrants and connections to individuals — including children — working on radical projects. The managers of these programs want young people to apply. In an interview last year, Danielle Strachman, the founder of 1517 Fund, said: “I want to be funding 11-year-olds. Because some of them are going to be ready.”

Still, these programs are small, hard to find, and there are not nearly enough of them to go around. So we should make more, far more, and advertise them outside the standard coastal circles. If we can specify them to young scientists, all the better — I’d love to see a microgrant that gives kids a stipend to run an experiment, then pairs them with a trained scientist who can help them carry it out.

For every Jo Nagai, there are ten thousand kids who can’t finagle their way into the good graces of Georgetown. We can fix that: the barrier between a complete scientific project and zilch is often small. With a few hundred dollars, you can buy twenty butterflies and a prodigy.

Note: This post is adapted from a preprint of an academic paper on psilocybin use in Lesotho. Portions of the text below are copied directly from the paper. You can read it here. It also contains the references for any empirical/statistical claims made in this post.

If you’re interested in psychedelics, you’ve probably encountered a book, podcast, article, or academic paper that makes a claim like this one: for thousands of years, most or all human cultures have used psychedelic substances for many religious and medicinal purposes.

This might sound uncontroversial. Hallucinogenic plants and fungi grow all over the world. These days, you can’t make it through the week without hearing a news story about the magical powers of psilocybin, iboga, or another psychedelic du jour. Of course people use them! Why wouldn’t they eat the things that one-shot their depression and make them feel like they’re talking to gods?

I agree — why wouldn’t they? But there is a big difference between a claim that is actually true — as in, verified by data — and a claim that simply feels like it must be true.

Much to the chagrin of podcasters and academics alike, the claim that psychedelic use is global and ancient falls in the latter camp. It feels true. In reality, we only have good evidence of psychedelic use in a small subset of societies. Virtually all of them come from a few scattered regions in the Americas, with a few minor exceptions. When it comes to rest of the globe, we simply do not yet have direct evidence of people using psychedelics for anything.

Classical psychedelics — those that mainly work upon serotonin 5-HT2A receptors in the brain, like psilocybin — are especially rare in human cultures. These are the psychedelics that appear to have lots of clinical utility. Yet outside the Americas, there is zero direct evidence of classical psychedelic use.

Which is a bit odd — psilocybin mushrooms grow on every inhabited continents, and we discover new species every year. Mainly because they’re so common — though also because of psilocybin’s massive cultural cachet — scholars have spent several decades speculating about all the other cultures that surely, probably, maybe consumed (or continue to consume) these fungi.

Next year, we’ll hit the 70th anniversary of Gordon Wasson’s article in LIFE Magazine, in which he revealed the use of psilocybin in Mexico among the Mazatec Indians. His story spurred our modern obsession with psychedelics, a billion-dollar industry, and myriad claims about their role in human culture and evolution.

Here’s a list of every place where, since that article, ethnographers have found evidence of indigenous psilocybin use: Mexico, among the Mazatec Indians. End of list.

Until now.

Beginnings

Last spring, I was wrapping up the first year of my PhD in Evolutionary Anthropology. I needed to find somewhere to do fieldwork and a subject to study while I was there.

Over the prior months, I had spent a lot of time thinking about religious experience — specifically, I was interested in the techniques that people develop through cultural evolution to induce these experiences and make them as immersive as possible. So I decided to work in a community where people were using interesting and novel techniques to structure these experiences: Sibundoy, Colombia. Known for its ubiquitous ayahuasca rituals, often scaffolded by music, Sibundoy seemed like a great place to study how two especially powerful techniques — music and psychedelics — intersected to structure religious experiences.

Was this an especially novel subject? No. Did I have an actual research question? No. Did I have a good plan for answering the central question that I had not yet selected? Definitely no. But the first fieldwork session is usually a pilot — the budding anthropologist travels to a site, gets to know the community, then finds interesting stuff to study as they go.

I bought my tickets. As of late May 2025, I was prepared to learn about ayahuasca in Sibundoy.

Then my advisor, Manvir Singh, sent me a paper about mushrooms in Africa. The authors described two new species of Psilocybe mushroom — per the genus, this is the group of mushrooms that contains psilocybin. Both species were native to southern Africa. On its own, that was intriguing.

But the real kicker came in a short paragraph titled “Traditional Use” that focused on a species native to Lesotho, Psilocybe maluti. Here is the excerpted paragraph:

The indigenous name for P. maluti is given as koae-ea-lekhoaba, and the psychoactive mushroom is used to induce a trance-like state. A brew is made by collecting large quantities of P. maluti and steeping the mushrooms in warm water, along with Boophone disticha (L.f.) Herb., a strong hallucinogenic plant, known locally as seipone or leshoma . . . The Psilocybe and Boophone extract, referred to as seipone sa koae-ea-lekhoaba, is consumed by the patient, who is then placed in front of a reflective surface and relays the hallucinations/visions seen in the reflection to the healers who interpret these as answers to the patient’s spiritual questions.

As the authors note: “This appears to be the only recorded firsthand report of hallucinogenic mushrooms being used traditionally in Africa and the first mention of hallucinogenic mushroom use in Sub-Saharan Africa.” Actually, this was an understatement. If true, it would have constituted the only direct evidence of classical psychedelic use outside the Americas.

Naturally, the report sparked a small media storm, much of which heavily exaggerated the claims. On the Joe Rogan Experience, Paul Stamets claimed that the local healers sang songs and told stories about P. maluti.1 Hamilton Morris published a 90-minute interview with one of the co-authors on the paper. Various science writers and Substackers wrote in breathless terms about the ‘ancient traditions’ that had now been ‘rediscovered.’

But as Manvir pointed out in a Guardian article on psychedelic use around the world: “Big if true.”

The report was suggestive, but it wasn’t scientifically robust. How many healers said they use the mushroom, and how many said that they don’t? Were they interviewed independently? What kinds of interview procedures were used to make sure they had hadn’t misidentified the mushroom or confabulated their answers? If the mushrooms were, in fact, being used, was their usage restricted to the mirror ritual?

Etcetera. On almost every count, whether the mushrooms were definitively being used — and if so, how — was an open question. And it was all much more interesting than whatever I had planned to do in Sibundoy.

I scrapped my tickets to Colombia, emailed the first author of the paper — Breyten van Der Merwe, now a close friend and collaborator — and asked who had conducted the initial interviews with the healers. He put me in touch with a woman named Mamosebetsi Sethathi: Betsy for short.

Betsy is utterly unique. She is a trained traditional healer from Lesotho, fluent in four languages, and a Master’s student at the University of Exeter. Given her wealth of traditional and scholarly knowledge, as well as her connections to healers across Lesotho and South Africa, I am convinced that there is literally no one else in the region so prepared to carry out scientific research on Basotho healing practices.

Fortunately for me, Betsy was keen to work together. In July 2025, I traveled to Lesotho, and we spent six weeks talking to as many Basotho healers as possible. I returned a few months later for another four weeks of research.

Leh-soo-too, not Leh-soe-thoe

Before getting to the results of those interviews, I’ll provide some ethnographic context on traditional healing in Lesotho.

First, a terminological note: a healer is technically referred to as lethuela in Sesotho, but it is more common for people to use the term sangoma, derived from Zulu. From here, I’ll use sangoma to refer to the general category of traditional Basotho healers.

Despite the widespread influence of Christianity throughout Lesotho, traditional healing remains central to Basotho life. There is roughly 1 sangoma per 500 people in Lesotho compared to only 1 doctor per 40,000 people. Approximately 40% of Basotho people in Maseru, the urban capital, report consulting with traditional healers; a similar percentage of pregnant Basotho women have reported such consultations during their pregnancy.

Most or all sangomas acquire their profession and abilities by communicating with and exerting control over ancestral spirits. The journey into healing begins with a calling, in which ancestral spirits demand that an individual begin initiation. They assert this demand by striking the victim with a sickness that manifests through physical and mental pathologies, including bodily pains, vivid dreams, and auditory hallucinations. To alleviate the sickness, the victim must find a gobela — a spiritual teacher — to initiate them as a sangoma. Refusing this call leads to the worsening of the sickness and ultimately death.

The training process varies in length. Some initiates report training for several months, while others report training for a year or more. The details of training are similarly variable. However, it is common for initiates to live in the home of their gobela throughout the entirety of their initiation. During this time, they generally must abstain from sex and obey the orders of their gobela completely. In some cases, initiates must also refrain from contact with direct sunlight; if they need to go outside, they must cover themselves completely and wear a straw hat. At three-hour intervals, initiates pray, drum, and dance. At the conclusion of this process, they perform a public graduation, in which they demonstrate their capacities to summon ancestral spirits, diagnose sickness, find lost objects, and perform other abilities acquired throughout initiation.

Another central part of initiation is the acquisition of an herbalistic toolkit. Upon graduation, sangomas usually possess extensive ethnobotanical knowledge that they will continue to expand for the rest of their lives. To some degree, this knowledge is acquired through direct teaching from a gobela. But to prompt transmission of how to use a particular plant, initiates typically must see the plant in a vision or dream and describe it to their gobela; at that point, the gobela knows that the ancestral spirits wish for the initiate to learn how to use the plant.

Many plants have been reported in use by Basotho sangomas. One comprehensive survey identified 303 distinct plant species used for ethnomedicinal purposes. Some are clearly psychoactive — Boophone disticha (L.f.) Herb., most obviously, is common and well-known for its hallucinogenic effects. The usage of others suggests possible psychoactive effects – for instance, many different plants are consumed to enhance the memory and visualization capacities of initiates – though most of the repertoire has not yet been directly tested for such effects.

But most surveys have ignored mushrooms entirely. Even mushrooms known to be used widely among Basotho healers, like the Podaxis Desv. mushroom, are typically not mentioned in these surveys.

Thinning the Veil

It was with this ethnographic and ethnomycological context in mind that Betsy and I approached the interviews. Our questions for the sangomas were straightforward. Did they know about the mushroom? We confirmed their knowledge using detailed pictures and a series of confirmatory questions. If they were familiar — did they or others use it? If so, how, and what were the effects?

Below is a map of where we did the interviews (Figure 1). We focused on four locations that either had a high concentration of Basotho healers or were especially close to the documented range of P. maluti.

In other Substack posts, I’ve written about some of my specific experiences during these two fieldwork sessions. I attended all-night dances and witnessed people entering trance for the first time. I boofed a hallucinogenic brew and spent eleven hours being attacked by demons born from a shadowy wall. I saw rock art painted ten thousand years ago: I felt the horror of entropy tearing at the human soul, only to find it replaced by a sense of the Sisyphean beauty of our fruitless struggle to defeat it.

In this post, though I mostly want to focus on the results. Betsy and I have now completed the first in-depth ethnographic investigation of psilocybin use among Basotho healers.

In total, we interviewed 26 healers. 15 correctly identified P. maluti; 10 failed to identify it; one person was excluded from the sample due to suspicions of falsified information. Of those that identified P. maluti, 7 healers independently reported both personal use and use by others, 5 reported only use by others, and 1 reported only personal use. Two reported no personal use or use by others.

We also interviewed 8 non-healers. Six were adolescent boys charged with grazing cattle, sheep, and goats, commonly referred to as herd-boys (n = 6). Two were neither herd-boys nor healers. Six of these non-healers were independently familiar with P. maluti. None mentioned personal use, while 3 mentioned use by others.

Interviewees reported uses in four categories: initiation, healing, recreation, and magic (Figure 2). Our interviews did not indicate that healers were aware of ritualistic, medicinal, or recreational psilocybin mushroom use in other parts of the world, though this possibility cannot be entirely ruled out.

Below, I’ll go into some depth about how interviewees described the role of P. maluti in each use category.

Initiation

Across healers from southern African Bantu-speaking groups, it is common for initiates to consume a preparation comprising various herbs, creepers, stems, and bark mixed together to produce a foamy brew. Xhosa and Zulu healers refer to the preparation as ubulawu; in our investigation, most healers used the term sethoto (Figure 3). In some cases, initiates consume sethoto in the mornings before vomiting it back out. Foam consumption typically occurs periodically throughout the day, before sleeping, and at intervals throughout the night.

There is no singular recipe for sethoto — different healers use different plants in their preparations. Among Basotho healers, this variation is related to the spirits that an initiate is attempting to access; depending on the pertinent spirits, different recipes may be called for. However, the usage of the mixture is clearly based around the induction of psychoactive effects. The healers who discussed the use of sethoto, as well as prior ethnographic investigations, report that consuming and purging sethoto boosts the vividness and memorability of dreams and visions. Healers commonly frame this as a matter of thinning the veil between themselves and their ancestral spirits, which allows the spirits to communicate with them more fluidly.

Five healers independently reported the incorporation of P. maluti into the sethoto brew (Figure 4). Four had either consumed sethoto that contained the mushroom during their own initiation or had prescribed it to their initiates; one had not used or prescribed it but was aware of other healers who had. In all cases, the reported purpose was roughly the same: sethoto (with P. maluti included) helped enhance the vividness of dreams and visions for initiates. More generally, the mixture helped initiates break through “blockages” in their training.

Healing

Seven healers mentioned use of P. maluti for treating physical, mental, and spiritual problems. I’ll note that it’s hard to distinguish between strictly mental ailments and ailments that are understood as spiritual in nature. Still, our initial interviews indicated the use of P. maluti for problems comparable to specific mental illnesses without any explicit spiritual framing. So I refer to those problems using Western clinical terms, such as “addiction” and “epilepsy”. However, further investigation may reveal a more culturally bound basis for these conditions.

Regarding physical problems, two healers reported powdering the mushroom and applying it to wounds or incisions. A third healer reported conducting the same process but as a form of protection against curses rather than a tool for physical healing (as discussed in a later section).

Four healers reported using the mushroom for alleviating mental problems that appeared to lack a spiritual framing. Two healers described preparations of P. maluti for treating addiction. One said that she created a snuff out of the mushroom to cure a patient’s addiction to methamphetamine. The other also created a snuff out of P. maluti and three other unidentified ingredients. Patients suffering from addiction were to snuff the powder twice a day for seven days, and family members would report back to him as to the patient’s progress.

A third healer described mixing the mushrooms into water alongside an unidentified white plant. Patients drink the preparation directly to treat epilepsy and asthma. A fourth healer reported mixing the mushrooms into an infusion or porridge, which patients then ingested repeatedly for a month.

More explicitly spiritual problems were another common category of application. Two healers described using P. maluti to cure “food poisoning”: as the result of a curse being cast upon them, a victim dreams that they have eaten poisoned food, which makes them sick upon awakening.

One healer also mentioned having patients consume the mushroom as a brew or porridge, usually over multiple sessions. During these sessions, patients would stare into a mirror and access their “higher selves” to discover the nature of their ailment. The healer described a feeling of heaviness and overwhelming emotion, followed by eventual relief. Other subjective effects included time dilation, exhaustion, and a sense of one’s body shrinking (Figure 5).

Notably, the mirror ritual described above sounds a lot like the ritual described in the original P. maluti paper, in which patients consume a psychoactive brew comprised of B. disticha — a hallucinogenic bulb — and stare into a mirror, where they receive visions sent from their ancestors.

The original reports from 2024 suggested that healers incorporate P. maluti into this brew. None of our interviewees explicitly reported this use case; however, it is probable that the healer mentioned above mixes in B. disticha as well. The fact that we did not explicitly document further use of P. maluti in this particular ritual does not necessarily mean that the initial report was incorrect. However, it suggests that this psychoactive application of the mushroom is at least rarer than has been suggested, especially in comparison to its use in sethoto as described above.

Recreation

Five subjects mentioned recreational consumption of P. maluti by herd-boys. One healer, who did not include the mushroom in his healing practice, recalled powdering the mushroom into a snuff as an adolescent and consuming it with his peers. Subjective effects included enhanced colors, a boost of energy, and seeing “crystals” erupt out of the ground.

Two other individuals — one healer and one herd boy — similarly mentioned people consuming the mushroom as a snuff. A non-healer also said that, during her youth, she and other girls would prepare the mushrooms as a powder and roll the powder into cigarettes, which they then provided to herd-boys. The herd-boys would smoke the mushroom cigarettes. However, she never observed any intoxicating effects; psilocybin would be unlikely to maintain any psychoactive effects at the temperatures found in a burning cigarette.

Interestingly, several other herd-boys recognized the mushroom but were unfamiliar with its effects. Others claimed to not recognize the mushroom at all. We suspect that recreational use may be more common than these negative reports would suggest, perhaps due to stigma around recreational consumption: one healer said the mushrooms were “like a drug” and “should be used carefully.”

In one suggestive case, a herd boy recognized the mushroom and, without prompting, claimed that another individual regularly snuffed a powder made from them. However, upon being interviewed, the mentioned individual claimed that he had neither used nor heard of anyone using the mushrooms for anything.

Magical Protection

Alongside physical application and consumption of P. maluti, five subjects mentioned using the mushroom as a protective charm. Applications included applying the mushroom to wounds in a powdered form, in an unprocessed form as a protective charm, or bathing with it. Uses included protection or the removal of curses and charms, particularly those that take the form of lightning. This latter belief may stem from the common folk belief that mushrooms are likely to sprout where lightning has struck.

The past and the present and the future

Basotho people know about the mushrooms and they use them for many different purposes. Their use of P. maluti constitutes the first clear evidence of classical psychedelic use outside the Americas.

One obvious question is how long this has been going on. Did they start using psilocybin in the 90s? If so, the case study becomes much less interesting. Or have they been using psilocybin for three thousand years?

We can’t know for sure. Speculatively, though, here are some reasons to think that their use of psilocybin did not start with psychedelic tourism or colonial influence.

First, P. maluti looks nothing like P. cubensis, the common strain of Psilocybe mushroom grown and consumed in the West. Even if tourists had brought P. cubensis to the Lesotho highlands, it is unlikely that its proliferation among local peoples would have led to recreational, spiritual, and medicinal interest in P. maluti given their visual dissimilarity. Furthermore, there is no evidence of P. cubensis growing or being cultivated in the Lesotho highlands for personal or local consumption, nor is there any apparent awareness among healers of psychoactive mushrooms other than P. maluti.

Second, interviews with and references to usage by older individuals suggest that some reported use cases of P. maluti nearly or entirely predate Wasson’s 1957 article in LIFE, which spurred the so-called “golden age of psychedelics.” Two unrelated individuals — both of whom we estimate to be in their mid-to-late 60s — reported participating in or witnessing recreational use during their adolescence. This would place their reports around the 1970s. One healer also stated that he learned about the mushrooms from his grandmother, who, according to his recollections, was initiated with P. maluti around 1955.

Finally, many ethnobotanists believe that southern African groups have consumed psychoactive substances for centuries, if not millennia. Particularly strong evidence stems from ethnographic and archaeological investigations of Boophone disticha use among Khoisan peoples^. So it is at least possible that psychoactive fungi also formed part of this pre-colonial repertoire for groups, like the Basotho, that interacted regularly with the (probably) hallucinogen-using Khoisan.

Ultimately, only archaeometric methods can definitively tell us if the Basotho have been using psilocybin for many generations — we’d need to analyze residues from an animal horn, snuff box, or some other artifact and find traces of P. maluti. My friend Breyten and I are working on that. Until we find something cool, speculation will have to suffice.

With that out of the way — what does all this mean for anthropology, cognitive science, and psychedelic medicine?

To date, our academic understanding of traditional psychedelic use has been informed almost entirely by observations of Indigenous American groups. Research on peyote use among the Huichol, virola snuffs among the Yanomami, ayahuasca-like brews among the Desana, and psilocybin mushroom consumption among the Mazatec and historical Mesoamerican groups, among other cases, has suggested common tendencies in Indigenous psychedelic use. These include:

• large ritualized doses;

• administration in structured ceremonial contexts by individuals with special training;

• a central cultural role for the psychedelic substance.

Researchers and popularizers alike tend to emphasize the use of these substances for healing, divination, and other spiritual activities. Ethnographic observations in the Americas also suggest that it was historically rare for patients to consume psychedelics. Instead, ritual specialists generally took substances on behalf of clients, usually for the purpose of spiritual services or diagnosis.

The uses we documented among the Basotho diverge from these broad patterns. We found no reliable evidence of the consumption of large doses of P. maluti. Uses were diverse, including not only spiritual healing but also recreation and magic. The Basotho also do not seem to afford P. maluti greater significance or centrality than other plants or mushrooms, instead treating them as one spiritual and medical tool among many others in their herbalist toolkit. We also found considerable evidence that patients themselves consume the mushrooms.

These uses of psilocybin indicate that the varieties of psychedelic use likely extend well beyond prevailing models — to paraphrase Aldous Huxley, the model we’ve observed so far is not the only one there is!

Basotho use also suggests some interesting routes for psychedelic medicine. Healers usually apply P. maluti alongside other psychoactive plants for both initiation and healing purposes. Are there notable stacking effects from combining these substances together? Maybe, maybe not. These recipes at least indicate that looking at how psilocybin interacts with alkaloids from these other plants might be productive.

Another big question is how and why psilocybin seems to affect non-waking states during initiation. Most research on psychedelics and consciousness has tended to focus on how substances alter waking states of consciousness, but several Basotho respondents emphasized the effects of psychoactive brews on dreams. I suspect that a research program on psychedelic oneirogens — psychedelic substances that alter the structure of sleep and dreams — would lead to some interesting results!

Above all, it now looks like some cases of Indigenous psychedelic use outside the Americas have been overlooked. Psilocybin mushrooms grow on six out of the seven continent. New species are discovered every year. The fact that it’s taken this long to find another society that uses them means that they probably aren’t used that much.

But it’s plausible — as evidenced by the Basotho — that some such uses may not have been investigated or recorded. I hope that this work inspires more exploratory fieldwork in understudied regions of the world: as cultural diversity continues to vanish over the coming decades, it’ll be crucial to identify and document these psychedelic practices before they disappear forever.

Godspeed

I could spend another essay thanking the people and institutions who helped make this research happen. Without Betsy Sethathi, in particular, the project would have been impossible to carry out; without Breyten van Der Merwe, it never would have begun, nor would have the many exciting projects we’ve initiated since. I am proud of this study, but I am even more grateful to now call them both collaborators and friends.

But thank-yous make for poor reading. I want to end with what can be, I hope, a note of inspiration for other budding scientists.

I will not feign total humility — I know that this study is very big news for the psychedelic world. Many central figures in that world have told me as much. How did I, in particular, get lucky enough to lead it?

Thinking back on the process, I don’t think it was a matter of intelligence: the interview process was straightforward and focused on questions that anyone would think to ask, and anyone could have seen that the answers to those questions could potentially be super interesting. I also don’t think it had much to do with charisma: my interviewees were almost always friendly and open to speaking with me, especially given my collaboration with Betsy. I think that any other sufficiently smart and personable anthropologist could have achieved the same results if they were dropped into the same project.

If there was something to this particular success other than chance, I think it was the following: over the last three years, both before and during my PhD, I have been fortunate to work with two mentors who urge me to always work on the most important thing that I possibly can. You get one chance to make an impact on the world; it would be foolish to spend that chance on trivial questions. So when Manvir sent me that paper about psilocybin in Lesotho, I did not hesitate. I ate the cost of my flights to Colombia and purchased flights to Maseru instead.

Increasingly, I think this mindset — more than intelligence, charisma, or any other seemingly important trait — is by far the most important contributor to impact in any domain. It is now cliche to say you can just do things. But it’s true! You can just do things! And if you always aim for the most important things, and then do them, you might find yourself in some interesting places.

The world is not settled. Everything is left to be known. I’ll close with some words from Luke Glowacki, a mentor and friend, who wrote the following in a card when I started my PhD.

We are explorers dreaming of what Magellan saw as he rounded the corner of those straits into the unknown where nothing had been charted. My advice to a young explorer — pick the most significant questions that match your ambitions. Don’t be afraid to swing as hard as you can. You may need to create methods to do so. Work to the bone — no substitute for this. And pick close colleagues and collaborators who inspire you.

Godspeed.

A few years ago, I had a big argument with one of my relatives. She had suggested that cultural differences between social groups might partially explain other differences between them — for instance, mental health outcomes and wealth. This is obviously true. Why are Jews so disproportionately successful compared to other religious groups? Presumably, something useful trickled down from five thousand years of being the “people of the book.”

If she had used Jewish success as an example, we probably wouldn’t have had much to argue about. But instead, she used crime rates among African-Americans. There is no reason to reject this argument without seeing where it goes: if we can attribute at least some portion of a social group’s success to cultural differences, then it is not prima facie wrong to attribute at least some portion of their problems to those same differences.

Alas, I was an English major. So I committed a mortal sin of argumentation — I allowed myself to follow a chain of preexisting assumptions about what her point implied. She said this about crime rates and cultural differences —— which means that she thinks structural barriers don’t matter —— which means that she doesn’t believe in systemic oppression —— which means that she thinks the plight of African-American people is entirely their own fault and that we owe them no care or support.

If I had just asked her to elaborate, instead of following these associations toward an inaccurate interpretation of her argument, then the next three hours of useless yelling probably could have been avoided.

I call this cascading: I’ll define it as “the process of converting someone’s defensible point into an indefensible point based on preexisting assumptions about their views.”1 I want to get better at avoiding it, and I want other people to get better at avoiding it too.

Unfortunately, it is difficult to avoid: cognition loves a cascade, because cascading is a very efficient way to extract useful models of the world. Look at the pictures below. You see faces. Why? There aren’t any faces here. But that doesn’t matter. Your brain sees shapes arranged in the way that facial features tend to be arranged, so it cascades toward the conclusion that this array of pickles, cones, purses, and kayaks all possess faces.

For vision, cascading is just fine — you can’t rid yourself of the face illusion, but you’re smart enough to know that you shouldn’t feel bad about eating the pickles just because they’re apparently smiling.

For argumentation, however, it is a mortal sin. In my experience — both as a sinner and a victim of the sin — rhetorical cascading tends to shut down conversations before they can start. It limits our ability to learn from each other, maintain mutual respect, and arrive at more objective views of ourselves and the world around us.

So how do you stop yourself from cascading during debates? The crucial step, I think, is to notice what it feels like when you’re doing it.2 Here is a non-exhaustive diagnostic checklist — any one of the symptoms should signal that you’re probably about to cascade.

1. You’re preparing to reply before the other person has finished their current point.

2. You’re primarily thinking of rebuttals rather than questions.

3. Perhaps it is, in fact, time for a rebuttal — but you find yourself thinking mainly of buzzwords rather than arguments.

4. You do not feel like you could repeat back what the other person just said in a way that they would agree with.

If you’re in a debate and you’re displaying some or all of these symptoms, then you’re almost certainly about to ruin the conversation by jumping down a rhetorical waterfall.

Admittedly, I was mostly motivated to write about this by a few recent conversations where I’ve felt like I was the target of a cascade. It is disturbing to watch someone take a point you’ve thought carefully about and twist it into a completely different (and patently nonsensical) cliche. I could see the river rushing over the cliff, and there was nothing I could do to stop it.

Or perhaps there was! Cascades beget cascades — I suspect that in these conversations, I allowed the waterfalls of my interlocutors to shatter my own poorly constructed dam. Deciding not to initiate cascades is a virtue; getting smacked in the face by one and still managing to hold your ground is next to godliness.

So I urge you to be better than me. Yes, some people believe awful and ridiculous things. But before you jump to that conclusion in the course of a conversation, it’s best to hold back the flood and ask a few questions first.

“If I had an hour to solve a problem and my life depended on the solution, I would spend the first 55 minutes determining the proper question to ask… for once I know the proper question, I could solve the problem in less than five minutes.” — Albert Einstein

Last week, Claude — the large language model developed by Anthropic — made me go slightly insane. I had seen on social media that some scientists were now deploying it to conduct literature reviews, build computational models, run data analyses, and in some cases package everything into a publishable scientific paper. They suggested that large language models were now research assistants possessed of infinite patience, speed, knowledge, and potential. Scientists could automate most of their day-to-day labor.1

Big if true, I figured. That would mean the end of science as we know it.2

Subscribe now

So I started messing around with Claude. I told it to write a detailed literature review on shared principles in perception and memory. I told it to find out if precipitation variance affects the technological development of human societies. I told it to build a Python script that could extract acoustic features from MP3 music files. I wrote up a rough sketch for a cognitive psychology experiment, then told it to code the experiment into an HTML prototype.

And it worked. Every single request was granted. Ultimately, I fed Claude a proposal for a minor research project I had been working on and asked the machine to conduct it from start to finish: experimental design, analysis, literature review, figure generation, paper writing. After a couple of hours, I had an entire scientific paper. For obvious reasons, I don’t intend to submit it, but I suspect that it would be published.

So, science is over. Right? I should drop out of my PhD program and apply to McKinsey.

Well, maybe. Or maybe science has just begun.

55 minutes

For the past century or so, scientific success has been measured in papers. No matter what kind of research you’re doing, the final product is a published article that people can read, assess, and build from in the future. Prestige, jobs, and grants all largely ride on your publication record. Scientific institutions are very conservative — so this is unlikely to change, despite the grand predictions made by some commentators.

So when we think about how artificial intelligence could transform science, we should think about how artificial intelligence could transform the production of scientific papers. Which parts of the game have changed?

We can roughly identify two kinds of labor that go into a paper: empirical labor and theoretical labor. Empirical labor means creating a methodology to answer a question, following the methods, then analyzing the results. Per the above epigraph, it is the work that Einstein wanted to spend five minutes doing. Theoretical labor means contextualizing and identifying a question, exploring the import of the answer, and synthesizing everything into the broader literature. It is the work that Einstein wanted to spend 55 minutes doing.

To some extent, artificial intelligence eases the burden of theoretical labor. For instance, it is very good at reviewing prior work on some subject of interest. Rather than spending weeks clicking through Google Scholar, you can now ask Claude to give you a report on everything pertinent to your research question, and it will do a decent job.

But for now, at least, artificial intelligence is still pretty mediocre at two key theoretical tasks. First, it is bad at picking important research questions. By “important,” I mean a question for which the answer fundamentally contributes to the expansion of human knowledge. Second, it is bad at using existing research to build or contribute to scientific theories — that is, the frameworks we derive from our scientific labors to help us understand the world.

Arguably, these are the two most important tasks in all of scientific practice. We’re trying to build pyramids of knowledge, not scattered piles of rocks. If we’re not spending our scientific time trying to advance theories, then we should throw away our lab coats (or notebooks, in my case) and find something less trivial to do.

For reasons that I don’t understand, though, artificial intelligence is not very good at laying out the blueprint for a pyramid, or even identifying where the next brick should be placed. Maybe it’s hard to figure out what kind of knowledge is meaningful to humans if you don’t have a human body. Regardless, theoretical labor remains our province alone.

Empirical labor, on the other hand, is kaput. Artificial intelligence can do almost all of it instantly, and it will only improve with more computational power.

In a paper, the outcomes of empirical labor fall into two sections: Methods and Results. A standard Results section takes quantitative data and reports statistical analyses from it. For example, an author might calculate differences between the performance of two groups in an experiment and declare that the gap is statistically significant.

A large language model can handle that easily. Anthropic co-founder Jack Clark says that the company will likely rely on Claude to produce 99% of its code by the end of this year. If artificial intelligence is handy enough to write the actual code used by artificial intelligence companies to improve their multi-billion dollar product, then presumably, it’s good enough to write a data analysis pipeline in Python.

The Methods are more varied and complicated. Obviously, Claude can summarizes the techniques that were applied to answer the inquiry driving the paper. But the interesting question is not whether a large language model can, say, summarize your methods from an outline — it can — but whether it can actually carry out those methods on its own or even independently come up with the protocol.

Increasingly, though, this also looks like a possibility. Alberto Acerbi recently published a preprint of an AI-written paper that generated three computational models to formalize one of his theories. Paul Novosad has similarly noted that artificial intelligence can easily replicate a vast swath of the research conducted in economics. Literature on so-called “self-driving labs” also suggests that even basic physics and chemistry experiments can be identified and carried out by artificial intelligence.

As for the various exceptions one might want to raise — you still need research assistants to dissect the mice! — always keep in mind that we’re at the very beginning of the agentic growth curve. What we’re seeing right now is the worst that these agents will ever be at science. Presumably, if you told certain ENIAC3 technicians that high-powered computers would someday fit in your pocket, they would have laughed in your face.

In effect, artificial intelligence has eradicated (or will soon eradicate) virtually all the time and effort that goes into empirical labor. Data collection for some fields, of course, remains firewalled from this turbo booster. Anthropologists (lucky me!) will not see Claude traveling to the Kalahari and shacking up with the Ju|‘hoansi. But for many disciplines, methods are now cheap or, in some cases, free. Once you have your data — which again, large language models can increasingly fetch for you — you can now find out the answer to your scientific question within days or hours, when once the pursuit might have taken years.

But artificial intelligence has still not changed much about the time and effort that goes into theoretical labor. Theory is still expensive. No matter how you prompt it, the current iteration of Claude is not going to discover a new paradigm for psychology or suggest the next great experiment.

But theoretical labor was always the important part. Artificial intelligence doesn’t cut down from Einstein’s 55 minutes of theory; instead, it drops his five minutes of methods down to almost zero.

Okay, cool. 55 minutes is still close to an hour. Are we really going to get this worked up over an ~8% boost in efficiency?

The artisan is dead; long live the artisan

My PhD advisor likes to ask a scary question. If you pitch him a research idea, or talk about an ongoing project, there is a good chance he’ll nod slowly, lean back dangerously far in his chair, and say: “Right, word, okay, for sure. What are the stakes for that?”

By “stakes,” he means the proof that there is real theoretical labor behind your research — that answering the research question would reshape our scientific theories, or expand them, or generate a new one entirely, or at least resolve some obviously interesting unknown in the world. Insofar as scientists should spend 55 minutes out of their gun-to-head hour identifying a question that has real stakes, they should have no trouble replying.

Yet many capable, hard-working, and very intelligent scientists regularly seem stumped by his question. And I think that if he went around and asked every working scientist in the world to explain the stakes of their research, the majority would not know how to reply. That’s because the scientific world is dominated by one kind of researcher: the artisan.

An artisan is a master of their craft. They know all the best techniques, and they love to use them. Sometimes, they even invent new ones. They possess the soul of a blacksmith, or carpenter, or cobbler. They relish the act of production for its own sake. Like all great artisans, the actual outcome is rarely the point. Cobblers don’t think about all the shoes they make, or how they’ve advanced the theory of shoe-making. They just make the goddamn shoes.

Science is full of artisans. I can’t imagine a statistic that would verify my claim — perhaps the boutique literature arguing that most research is totally useless — so I’ll ask you to take my word for it.

At present, most working researchers possess the knack for methodology and analysis; they are great at empirical labor. But their theoretical labor tends to fall by the wayside, because it’s hard to be great at both. Hence why no one can tell my advisor why their research matters.

But if theoretical labor is the important part, why don’t we instead have a scientific world maniacally focused on the stakes above all else?

As always, the answer is the incentive: publish or perish. If you want to succeed in science, you must publish papers. It’s better if they’re good or even groundbreaking. But something is better than nothing. And right now, empirical labor — not theoretical labor — is the main barrier to publication. Data analysis is slow. Iterating experiments is hard. If you can’t do those things right, you can’t do them at all.

You can, however, pick an unimportant research question for which to carry out those tasks. That bit is easy, albeit unadvisable. So those skilled in empirical labor, despite their lack of theoretical skill, can still regularly publish and thus receive jobs, grants, and prestige. And because artisans are usually the ones judging other artisans, poor theoretical labor can easily go unnoticed.

Their counterparts, however — those skilled in theoretical labor, but not empirical labor — just won’t cut it. It isn’t enough to pick a good question. If you want a paper, you need to pour the tears and screams into working out the solution. Those unwilling or unable to do so will not usually find their way to the faculty lounge.

The incentive to publish widely, paired with the difficulties of empirical labor, makes artisanhood the most viable strategy for success. The success of artisans then elevates more artisans into prestigious positions, because artisans (like anyone else) are unlikely to disdain their own path to greatness. Hence we find ourselves in a scientific world full of people who opt to spend 5 minutes on the question and 55 minutes on the answer.

With all that preface, I can finally explain why artificial intelligence spells the end of science as we know it: it lets anyone work like an artisan. With Opus or Codex on your side, it isn’t hard to publish decent papers. I could easily pump out sixty papers in a year if my theoretical standards were low. That isn’t a flex. You could do it too — in fact, a large language model isn’t even half-bad at identifying a workable question.

I imagine some researchers will get in on that action and publish at a ridiculous rate. But that won’t last long, because the benefits of artisanmaxxing are about to vanish. If everyone’s an artisan, no one is. Empirical labor no longer distinguishes you from anyone else. Large language models did not just shave 5 minutes from the hour of work. They dropped almost the entire shebang.

The artisan is dead; long live the artisan.

So what comes next?

The era of the seer

Empirical labor is now easy. But theoretical labor is still hard. And despite our recent history of artisanal dominance, grand visions of science have always centered on the theory. We worship the people who dared to see further; but in practice, it is hard to make your way onto the shoulders of the giants if your back hurts from hunching over R Studio.

No longer. Theoretical labor is the only game in town. That means scientific success will accrue to a new group of researchers: the seers.

A seer understands that the purpose of science is the pursuit of knowledge, not self-actualization4. So they pour their labors into picking out questions that will produce useful knowledge and building theoretical frameworks that refine and guide their efforts. They always know the stakes. Seers refuse to work on anything that lacks a clear theoretical purpose.

To be clear, the most influential scientists have always been seers, at least partially. The problem is that any successful seer also needed to pair their theoretical labor with intense empirical efforts; they needed the soul of a seer in the body of an artisan.

This brings to mind a famous quote from Stephen Jay Gould: “I am, somehow, less interested in the weight and convolutions of Einstein’s brain than in the near certainty that people of equal talent have lived and died in cotton fields and sweatshops.” I am similarly interested in the near certainty that people with lots of great research questions have lived and died as consultants, because they just couldn’t bear to spend two years writing up an experiment in Javascript.

With the advent of artificial intelligence, though, the only distinction that remains between scientists is that of seerdom: who can ask the right questions, and who can figure out how to put everything together? Finally, the seer soul is free to roam.

By moving into an era in which empirical labor is near-instantaneous, we also move to an era in which theoretical labor is the main difference between scientists in the race for prestige. Suddenly, the stakes of scientific research cannot be ignored or brushed aside. Stakes are all that remain of the scientific enterprise; though of course, they were all that the enterprise ever should have been.

Or does it explode?

But institutions do not necessarily evolve at the same pace as technologies. As better writers have pointed out — special shout-out to Adam Mastroianni, whose work made me want to start a Substack in the first place — the current scientific publishing system is stupid. Peer review massively slows down the publication process, while simultaneously failing to make papers better and adding further unpaid labor to the lives of scientists. To compensate for the months or years it takes to see papers in print, scientists get to pay exorbitant publishing fees to make sure people can actually, you know, read their work. Just as musicians pay their label to let them release an album. Right? Is that not how it works?

Most importantly, the existence of the Internet (check it out, super useful) raises the question of why we need scientific journals at all. It seems that we could speed up the scientific process, avoid publishing fees, abolish the burden of peer review, and improve science in a million other ways by simply publishing papers as blog posts or uploading them onto free host sites.

Oh, right. We tried that. Most researchers don’t even get their papers from journals anymore; they download preprints uploaded by the authors or pirate them from SciHub. And yet, here I am, still trying to optimize my paper for Nature.

Our ongoing failure to fix scientific publishing suggests that the mere emergence of useful technologies does not mean that institutions will adjust to support them. The scientific ecosystem is built for the 20th century. Even the Internet couldn’t make it budge.

What then of artificial intelligence? The number of scientific papers seeking publication is about to skyrocket. Peer reviewers are delegating their work to large language models. Researchers may soon automate most or all of their labor, which raises the question of what ‘authorship’ should even mean in the age of Claude. These changes are seismic — I can’t imagine reforms to the current system that would allow it to withstand the quake.

So I won’t propose any. Instead, I’ll make a bold prediction. The automation of science will eventually prove too much for scientific journals to handle; they cannot, nor should they, arbitrate between ten million papers each year. The system will collapse. We should start thinking about what to build from the rubble.

I, for one, welcome our robot Einsteins

In sum, then: the automation of science, and specifically the automation of empirical labor, will set science free, both from the demands of artisanhood and the absurdities of scientific publishing.

But serious questions remain. What happens if artificial intelligence becomes great at theory, too? Really, what I’ve described here is not true automation, but instead the distribution of empirical resources to a much broader subset of scientific thinkers. For the near future, this seems like the most plausible outcome. But it seems possible that we might someday get true automation — that artificial intelligence may take over the entirety of Einstein’s hour.

Whether that possibility should scare us or not depends on what we think science is for. The goal of literature, for example, is arguably to facilitate the experience of another human mind. Automating literature with large language models would therefore defeat the point of the entire enterprise.

By comparison, the goal of firefighting is to put out fires. If large language models were better at putting out fires than people, we would hopefully endorse the automation of firefighting. It doesn’t matter who puts out the fires; what matters is that the fires get put out as quickly, efficiently, and safely as possible.

Some people probably think science is more like literature than firefighting. If that’s your perspective — and I don’t disdain you for it — then the prospect of automated science is horrifying.

But much as I empathize with this romantic view, I disagree. I think the point of science is to expand our understanding of the world. If God descended from heaven and offered us answers to every burning scientific question, no labor required, I would hope that we’d take him up on it.

It doesn’t matter who does the expanding. Science was never about the scientists.5

4

Borrowing this lovely phrase from the following essay:

Popular by Design

Academics Need to Wake Up on AI

This piece is inspired by a wave of recent AI-related writing from people I respect: Dan Williams, Alex Imas, Ben Ansell, Tibor Rutar, scott cunningham, Kevin Munger, Hollis Robbins, Claude (yes!) Blattman, Kevin Bryan, Andy Hall, Kelsey Piper, Sean Westwood, and many others. So here, I’m continuing the tradition of writing the takes that are upsetting but needed…

Read more

6 months ago · 114 likes · 31 comments · Alexander Kustov

5

Regarding this (possibly inevitable) shift in scientific practice, it’s worth reading this recent piece from Asimov Press on the ‘legibility problem.”

Asimov Press

The Legibility Problem

The last World Computer Chess Championship was held in October 2024. It ended because, in the words of its organizers, “top programs are unbeatable by humans; making them stronger has no real research value.” Its mission, after half a century of effort, was complete…

Read more

6 months ago · 53 likes · 6 comments · Asimov Press

For man will be blotted out, the blithe earth die, the brave sun
Die blind and blacken to the heart:
Yet stones have stood for a thousand years, and pained thoughts found
The honey of peace in old poems.

Subscribe now

Breyten and I waited beneath a water tank at the base of the mountain. Our film crew — we were, somehow, being filmed for a documentary about psychedelic use around the world, in this case Zimbabwe — had gone ahead to scout a rock art site called N’gomakurira, guided by a quiet but vastly knowledgeable archaeologist named Stephen.

Whether the site was sufficiently filmable remained to be seen. But its potential value and power was well-known to the locals: nearby, the owners of the land on which the site rested were whispering about how to maximize their profits from our obvious interest in the paintings.

“Oy!” We looked up. Our fixer, Dan, was holding up his phone and gesturing toward the mountain. “They’ve said to meet them. Just follow the white arrows.”

We tossed our bags in the car and headed up the steep trail, passing on the way under two skittish baboons that had scurried into a tree. Breyten, ever the sub-three-hour marathoner, practically sprinted up the hill, stopping only to scan for white arrows. I lagged behind and wondered if this path had been even slightly difficult for the San foragers who had inadvertently led me onto it.

Small gray mushrooms peeked over the adjacent foliage. Occasionally, pieces of elephant dung greeted us on the switchbacks, though their progenitors were nowhere to be seen.

Eventually, I caught up to Breyten at the top of the ridge. We had overshot the trail. After doubling back, we found a series of white arrows leading into a massive ravine, winding and jagged, the seeming remnants of some leviathan carving a trail through the earth.

In many cultures, people tell stories about a race of giants that preceded our own — monster slayers, figures closer to gods than men, gazing amusedly from the abyss of the past at our present puny doings. Places like this one help you understand why such stories felt like sensible explanations.

Following the arrows, we soon lost the way again. Breyten went looking further down the ravine for the crew, and I followed a slight creek trickling in the other direction.

“Eli, hey! I found them!” He had made the right bet. I clambered through the bushes and over the boulders and discovered Breyten staring wide-eyed at the towering blood orange walls of the canyon, accompanied by our producer, who held one finger over her lips and said in a near-frantic whisper: “It’s incredible, just so stunning, we don’t have the cameras but we wanted to make sure you saw it.”

A few dozen meters away, the crew was holding what little filming equipment they’d brought next to a section of the wall. I joined Breyten and stared into the granite.

There is a certain expression of joy, which most are only fortunate enough to wear once or twice in their life: wherein a smile expands to contain all eternity, draws inevitably toward its limit and, seeing no further path forward, spreads its burgeoning roots into one’s eyes and ears and forehead and chest and knees, the only plausible reaction to something that viscerally transcends the boundaries of parsible emotion.

Seeing the figures laid out on the wall, I was buckled by that desperate grin. Ten thousand years ago, someone had stood where I stood and left a message in the stone. And here I was, ten thousand years later, confronted with visions of the dead.

Of course, these thoughts were at first only implicit. The paintings stretched on and on, and the only reasonable course of action was to see as much as I could. I walked carefully along the walls, attempting to tiptoe for no real reason at all.

In single-file, the first painters had left massive outlines of elephants. Within those outlines, later artists — one can tell, based on the slightly newer coats of ochre — designated that elephant herd as the base of a palimpsest and drew zebras, rhinos, kudu, even an aardvark. A man leveled his reddened spear; nearby, a woman drove into the stone void with her digging stick. Two figures wrestled next to a giant, whose stomach ballooned with the spiritual potency called n/um, a so-called boiling energy that rose among San shamans from their stomach into their neck before exploding through their head, catapulting them into a supernatural world of half-animal deities and cruel illness-causing demons.

Most striking — for me, at least — was the dance. A man was bent over in the ritual posture of the great dance of the San, spewing blood from his nose and holding either a rattle or a mushroom, depending on the stories you want to tell about the world. Just above him, two women sat and watched, almost conversational in their posture. They would have helped to form the outer ring of the circle, where children and the less spiritually potent adults would chant and clap to guide the healers into trance.

Not far away, a hunter ground a plant into submission; next to him, a collaborator doused arrowheads in the resultant powder. In January, archaeologists reported traces of a toxic plant on San arrowheads dated to 60,000 years ago. This was very possibly a representation of the production process.

The crew did not really need our help. Breyten, Stephen and I wandered down the ravine, marveling at each new silhouette. For a while, the excitement of discovery carried me away.

But eventually, I’d seen it all. The paintings deteriorated and trailed into mere granite. I doubled back toward the dance and looked more closely at the two sitting women. Among their many artistic merits, the San depicted motion marvelously: I was reminded of drawings Kafka left in his notebooks, a few sheer lines carving high-dimensional motion into empty space, bodily equivalents to Mona Lisa’s ever-roving eyes.

I craned my neck toward the upper edges of the canyon, then turned around and stared across the ravine. My breath seemed to echo into the drought-weary waters; bird cries ricocheted lazily against the walls.

I returned to the women and thought of an image penned by Richard Dawkins. Imagine, he said, your mother holding her grandmother’s hand, and your grandmother holding your great-grandmother’s hand, and your great-grandmother holding your great-great grandmother’s hand, and so on in an endless line. Somewhere, the lines of every living person converge on one woman — mitochondrial Eve, the ultimate mother to us all.

And then the excitement fell away, and I could think only of how much had been lost. Slender artists had roamed this path, declaring in fat and ochre and blood the fact of their existence, perhaps returning every now and then to refine their handiwork. They had died, eventually, buried in caves or killed in battles or occasionally left to hyenas. But their children knew the old ways. They came back and added a rhinoceros, a flower, a portrait of ecstasy.

Then, in a few hundred years, they lost everything. The descendants of the artists now scramble for life in reservations. The remnants of their world gasp for air in the pages of textbooks and ethnographies, the speculations of podcasters, the half-interested shrugs of those who care little for what has been lost and scan the remainders on Instagram.

Here lay the last testaments of the great ones; nothing besides remains; and though this asphyxiation was accelerated by colonial horrors, there is no real counterfactual. The fact is that all would have been lost, in time. Everything always is. Even giants collapse beneath the faceless weight of forever.

I felt intactile hands on my shoulders. Breath rattled the wind. From the other side of entropy, the women in the walls considered my stare, and they found it wanting. The walls of the ravine drew away from each other, slowly but surely. I saw them crumble beneath a million cuts.

Today, I could hear the painters murmuring across the expanse of time. Tomorrow or tomorrow or tomorrow, some strange beast would stand in my place, long after I too had melted into air, and perhaps look toward where the ghosts, finally laid to rest, had once clung to the edges of the world.

But then again — tomorrow is never today. At that moment, if only that one, a new group of artists in search of something true was paying homage to their forebears. I looked around and saw that the cameras had gone away. Everyone was simply standing, now well beyond words, listening to a language with no name.

“You know,” Stephen said, “we still haven’t figured out how to paint something that will last this long.”

The ultimate truth remained; but the ghosts had not ceased roaming, not yet, and how remarkable that such things as ghosts could live on beyond their time. What a feat these painters had accomplished; how strange it was, several thousand years on, for them to be anything at all.

The sun was setting. Our director wanted to collect some footage of the pink light over the mountains. After taking a few more pictures, we went on our way. The white arrows guided us back over the ridge, down the hill, past the baboons and into the parking lot, where two children were waiting with plates of mangoes. They were perfectly ripe.

On the ride home, I considered our usual view of ghosts: tormented things, unhappily chained to life, begging and pleading to sleep. I found it wanting. There is no greater achievement, in the end, than to leave something of yourself behind.

In the last few months, the governments of Australia, France, Denmark, India, England, Norway, Spain, and the United States all instituted, introduced, or began to debate laws that would ban teenagers from accessing social media due to its alleged effects on the mental health of young people. Simultaneously, public outlets began to realize that most scholars don’t think social media has virtually any negative health impacts at all. Too little, maybe too late.1

Subscribe now

But like the politicians behind this wave of legislation, let’s temporarily ignore the evidence. In fact, I want to make up some data from scratch.

For instance: did you know that daily social media use increases the likelihood a child will commit suicide by 12-18%? Or that teenagers are far more likely to visit the ER for psychiatric problems if they have an Instagram account? Or that a child’s amount of social media use, past a certain threshold, correlates exponentially with poorer sleep, lower reported wellbeing, and more severe mental health symptoms?

If that was all true for social media— and again, none of it is — you and I both would agree that people under 16 or so should not have access to platforms like Instagram or Snapchat. Imagine allowing your child to enter any system that would make them 12-18% more likely to kill themselves. That would be insane. You wouldn’t let your kid anywhere near that system, and the public would protest until it was eliminated once for all.

Great. So let’s get rid of school.

I Don’t Like Mondays

Yes, there’s the obvious twist — all the data I just listed is true for the effects of school. The modern education system is probably the single biggest threat to the mental health of children. At the very least, the evidence for its negative effects is unambiguous: the same cannot be said for social media.

The first statistic is the most damning. In March 2020, COVID was born, and students were removed from school. Generally, we view COVID as an absolute scourge upon the mental health of young people. But when children stopped going to school, something interesting happened. Their suicide rates plummetted and remained low throughout the summer. In the fall — when most schools returned to in-person instruction — they started killing themselves again.

You can see the trend in the figure below. Basically, the line shows suicide rates relative to the number of months before and after the resumption of in-person school. The line down the middle indicates that change. Panel A shows suicide rates for teenagers aged 12-18, while Panel B shows suicide rates for adults aged 19-25.

For adults aged 19-25, the resumption of in-person school had no effect on suicide rates. For teenagers, though, it’s a different story.

This data on the effects of COVID — or rather, the effects of escaping and then returning to school — is incredibly telling. The reported shifts appear specifically for adolescents who returned to school, not for their peers who remained at home during the same period. It is causation, not correlation.

But the correlations are interesting too. For decades, researchers have noted the calendar effects on child suicide and mental health visits. From 1990-2019, suicide rates among young people have always dropped precipitously during the summers and spiked again in September. Adults show no such trend.

For English students, stress-related presentations in emergency rooms also rise during school periods and drop during holidays. This is true for both girls and boys across four years of data. Below is a figure showing rates from 2017-2018.

Beyond these clinical statistics, there’s also the simple fact that kids say they find school more stressful than pretty much anything else in their life. A study of 2,000 English secondary school students found that roughly half had skipped class due to anxiety. A 2018 survey from Pew Research similarly found that 88% reported academic pressure as a source of distress; comparatively, only 28% felt distressed by pressure to fit in socially.

In short, school sucks so much that it reliably makes students want to hide at home, visit the ER, and take their own lives. The data has been completely clear on these points for years.

Thanks, Obama

However, the negative mental health effects of school do not seem to have stagnated over the last few years. Instead, they’ve gotten worse. For instance only 43% of teenagers cited school as a significant source of stress in 2009. By 2013, that percentage had nearly doubled, with school identified as by far the largest source of stress in the lives of adolescents. As of 2024, stress from academic pressure remained dominant at around 70%.

During the period when that initial jump in school-related stress occurred, suicide rates for children aged 15-19 suddenly spiked.

Of course, the shocking drop-off in youth mental health around 2012 is a well-known phenomenon: it is usually interpreted as the result of newly widespread social media use and smartphone ownership among young people. This is the basic argument Jonathan Haidt famously makes in The Anxious Generation. In the early 2010s, children suddenly became much less happy. What else could explain this shift other than the influence of social media?

While Haidt’s theory is plausible, the aforementioned survey data makes me very suspicious. If social media was the culprit, we should expect that teenagers started feeling more stressed about (for instance) their appearance or social status.

But they didn’t; mostly, they reported feeling more stressed about school. Intriguingly, this effect was potentially sharper for adolescent females: the exact group that, according to Haidt, suffered the most from the advent of social media.

However, to believe that increases in school-related stress largely account for the onset of the current youth mental health crisis, we would want to identify a seismic change in the schooling system around 2012 that could have plausibly driven the trend. And there was, in fact, such a change: the implementation of Common Core.2

Common Core was a set of educational standards meant to nationalize and clarify the skills that K-12 students should acquire over the course of their schooling. The new standards were notoriously difficult and cognitively rigorous: I recall my mother, at the time a fifth-grade teacher, complaining about the impossible word problems laid out in her new math textbooks.

One consequence of Common Core was that children had to spend much more time doing homework and much less time socializing with their friends. Data from the Pew Research Center shows these time allocation differences between 2003-2006 and 2014-2017. (Apologies for the blurry figure).

Notably, the increased number of students who reported school as a source of angst were primarily stressed about their homework loads. A survey of parents in 2013 also found that 63% said their child experienced “a lot of stress” due to the amount of homework they were assigned. And most importantly, recent studies suggest that the effect of homework load on mental health is non-linear — once the amount passes a certain threshold, the negative impacts suddenly skyrocket.

Am I willing to say that Common Core, rather than social media, was the singular force underlying the heightened destruction of youthful minds? No, I’m not. As Tyler Cowen pointed out in a conversation with Haidt, reducing these mood shifts to one cause or another is a bit like reducing a hurricane to the flapping wings of one particular butterfly.3

Still, it is striking that most discussions of the 2012 mystery tend to ignore what the kids themselves said about it. The children don’t think social media added much stress to their lives; they think the shift has much more to do with the increased rigor of school. Perhaps we should take their opinions seriously.

Extraordinary policies require extraordinary evidence

Okay, fine — school totally blows. But surely, social media can’t be entirely benign. Right?

Maybe, maybe not. Public assurances to the contrary, we don’t really know either way. The main problem is that most studies on the effects of social media fail to distinguish how users are engaging with their platforms. On Snapchat, you can make group chats with your friends and send pictures throughout the day; you can also receive nude pictures from random sex predators. Both forms of engagement would show up in a study as “screen time” or “time spent on Snapchat” — presumably, the latter is bad for mental health, and the former is potentially good. Whether one is substantially more common than the other is an unanswered empirical question.

The data we do have, however, is extremely murky. If you squint carefully, the studies summarized in The Anxious Generation do suggest that social media is somewhat harmful for teenage girls who are already at risk of mental health problems.

But on the other hand, a major longitudinal study found that using social media or playing video games more frequently had zero impact on the mental health of adolescents. A prior survey drawing from eight years of data had found that daily social media access had no effect on depressive symptoms. Data from Pew also shows that teens are more likely to report positive than negative experiences from their social media use.

Who’s right? Usually, it comes down to comically subtle arguments about survey methods. But one would think that if social media was truly so harmful — if, as Haidt has argued, it is the planet-sized dragon that flapped its wings and started a depressive hurricane — then there would be both minimal ambiguity and a strong academic consensus around his view. Notably, there is no such scholarly confusion about the effects of school.

Nonetheless, my gut (like yours) says that social media is not entirely neutral. My guess is that one particular form of engagement — doomscrolling through an addictive algorithmic feed — is bad for young brains. If studies could distinguish that activity from acceptable or even beneficial forms of engagement — for instance, forming group chats with your friends — then I think the statistics would be stronger. The benefits of socialization through these platforms is probably disguising the harms of consuming algorithmic slop.

Importantly, many platforms now emphasize content feeds over social interaction. That change is bad, and as some kids themselves have suggested, governments should probably move to regulate it.

Even if this somewhat generous intuition is right, though, the step of banning children from social media entirely is extraordinary and uncalled for. As far as we can tell, there is no clear evidence that usage is generally bad for mental health.

What we do know — or what we should know, at least — is that unimpeachable logic should back up any policy that limits the autonomy of a large and vulnerable population. That is clearly not the case for the present bans on adolescent social media access.

If the same fervor switched over to revamping the school system, however, that justification would exist. Scientists, parents, teachers, and children all agree that modern education is horrifically stressful. If we’re searching for a mental health bogeyman, then we adults should follow our own advice: look up from the phones and start paying attention to the classroom.

Play impactful games, win impactful prizes

So if the harms of schooling are so obvious — and particularly, so much more obvious than the harms of social media — why isn’t anyone talking about it?

To wax anthropological for a moment: when we identify an important but opaque problem that we want to solve, we tend to search for a single obvious and comprehensible solution. Factory jobs are gone? Simple: deport all the immigrants who took them. My cows got sick? Simple: figure out who cursed them and reverse the magic spell. Kids aren’t happy anymore? Simple: stop them from doing the things that seem different from the things we did throughout our adolescence, during which we, of course, were entirely pleased with our lives.

Simple solutions are easy to think about and apply. But many important problems aren’t simple, which means that the simple solutions are often wrong. Social media, like fear of witchcraft and immigration, is yet another all too obvious answer to a much more complicated question: how did our children become so sad?

School is also not the only right answer; as always, the truth is multi-faceted. But the data shows that it is one especially important facet.

Yet we ignore the melancholic elephant in the room, because you can’t pass a bill to ban anyone under 16 from going to school. Nor should you, probably. Making school less awful requires careful research and debate. As evidenced by the ongoing panic over social media, that is not a specialty of politicians, the commentariat or, for that matter, most of the public.

It is, however, a speciality of scientists. And fortunately for us, scientists have identified several changes to school that are about as simple as banning social media, which also come with the added bonus of addressing an actual problem.

Change #1: Start school later

Excessive screen use (I’ll admit it), early school start times, and heavy homework loads contribute to greater sleep deprivation among students. Obviously, sleep deprivation is bad for mental health.

To address it, one easy step is to push back school start times. Many correlational studies note that starting the school day later is associated with students sleeping longer, feeling better, and getting better grades. This is an easy fix that would be widely welcomed by students.

Change #2: Give students less homework

Homework is the greatest source of academic stress faced by students today. As mentioned above, increases in homework load harm mental health at a non-linear rate. This is a good reason to cut homework loads down, since doing so might provide a similarly non-linear boost to the health of young people.

While there is some evidence that spending more time on homework leads to higher levels of academic achievement, we should probably prioritize the reduction of massively negative impacts on wellbeing rather than the increase of small positive impacts on grades and test scores. State policies should set caps on the amount of time students are expected to spend on homework each night, with the acceptable amount increasing gradually across later grade levels.

Change #3 Increase recess and lunch time

Students now receive substantially less time to eat and take breaks during the school day than they did in the 1980s. For instance, after the implementation of No Child Left Behind, 20% of schools reduced recess time by an average of fifty minutes per week. In England, school break times have been utterly slashed since 1996 to allow more time for “learning.” All this, despite the fact that longer recesses and more physical activity both produce happier, healthier, and more academically succesful students.

Many states have mandatory minimums on recess time, mainly for elementary school students: 30 minutes in California, 40 minutes in Arkansas, etc. These numbers are abysmally low, especially when it comes to young children. I always find it slightly tragic when I hear kids say their favorite part of the day is recess. We should indulge those kids, and all the evidence supporting their preferences, by raising the recess minimums and enforcing those rules with all the same vigor that has been applied to the distribution of Yondr pouches.

If you don’t eat yer meat, you can’t have any pudding

Like all good interventions, these small changes would probably have outsized effects. Children would be happier, less stressed, and less likely to kill themselves during school time. That’s great. If these modest proposals received half the attention of the latest moral panic, young people would be far better off.

But schooling is a system, and ultimately, it is the system itself that is ruining the minds of young people. Well before 2012, children were still committing suicide in the fall. Fundamentally, school is not what kids have evolved to do: they’re supposed to play freely with their friends, not spend their childhoods performing for adults under constant supervision.

Yet this is our brave new world. For the last fifty years, adults have opted to put their children in a panopticon. Social media bans are yet another brick in the wall.

I hope we won’t let the greatest threat to the mental health of our children go unaddressed. Nor do I know how to fix it: to truly remedy the system, we’ll need a much larger reckoning. But we can, at least, start bringing it down the same way we built it up: one brick at a time. And to decide which bricks to remove first from the prison, we should maybe consider the opinions of the people we’ve locked inside.4

The future is already here; it’s just not evenly distributed. — William Gibson

Two days ago, OpenAI announced that they’ll soon introduce ads to free and lower-tiered subscriptions to ChatGPT. Allowing advertising will give them a bigger revenue stream, which they can use to build toward a profitable business structure, improve their product, and expand global access to their platform.

Unsurprisingly, everyone’s mad. Social media apps and search engines famously rely on a similar revenue model, but the data available to advertisers through OpenAI is much more refined and personal. Several dozen articles note that data from user conversations will probably be used for targeted advertising — all the time you spend pouring your heart out to LLMs might lead to lots of ads for Better Help. Of course, hypocrisy throws fuel on the fire. In 2024, Sam Altman said the potential combination of ads and AI was “unsettling.”

Subscribe now

Personally, I’d also prefer a world where ChatGPT doesn’t try to sell me Athletic Greens after I ask it for a workout plan. Still, I can’t help thinking of this conversation between Louis C.K. and Conan O’Brien.

I recommend watching it, because it’s very funny. But the title is self-explanatory. We live at a higher standard than medieval kings, teleport around the world in a day, fit the entire Library of Alexandria on small SD cards — and now, we have FREE access to an intelligent entity that can patiently and capably deal with mundane tasks, translate our words into any language, and explain virtually everything under the sun in whatever way we see fit.

And yet, no one is happy. We don’t just expect access to a polymathic and infinitely forbearing assistant; we apparently expect it at zero personal cost to ourselves.

Undoubtedly, part of the anger here stems from the fact that some people don’t want it at all. A quasi-religious fear of AI has swept through liberal spheres, which is to say, the spheres inhabited by people like me who write op-eds and blog posts on this topic. In these circles, admitting that you find AI useful for certain tasks — that it is, perhaps, worth something — is social poison.

Unless you’re underprivileged in some way, that is; then they’re obligated to hear you out. I have seen this happen a couple times — amidst an angry conversation about the horrors of AI, a Spanish-speaker (for instance) volunteers that ChatGPT enables him to translate his writing into English and thus share his work with the overwhelmingly English-dominated modern world. Everyone else gets quiet. “Right, yeah,” they admit. “That makes sense.”

But those use cases aren’t exceptions; the vast majority of the global population has been saddled by the womb lottery with linguistic, economic, racial and other barriers to success. For them, AI is a godsend.

Consider the graph below, which shows the degree to which people from different parts of the globe feel excited or nervous about the impact of AI.

For some countries, the majority of respondents said they were not excited about products and services using AI — they fall on the left side of the vertical dotted line. Guess who the pessimists are? The United States, Canada, France, Belgium, Japan; countries that are mostly English-speaking, economically dominant, or (usually) a mix of both.

On the other side of the line fall countries that are excited about AI. Overwhelmingly, this group consists of countries in the Global South, like Peru, Malaysia, and Brazil. How strange. What do those misguided saps find so intriguing? Don’t they know about the horrors of plagiaristic undergrads?

Maybe not. Or maybe they don’t care, because AI can help them navigate the 50% of web content that is available only in English, speed up mundane and inefficient features of the informal labor that predominates in poor countries, ease the cognitive load required to do tasks like programming and writing (especially in a foreign language), and help them figure out how to apply to college.

Obviously, the individuals I have in mind won’t usually have the resources to pay for a subscription to ChatGPT. If we want them to have access to this life-changing tool — as we should, compassionate believers in equity that we are — then we should be perfectly fine with companies like OpenAI finding some way to fund the global distribution and maintenance of their gadget.

This, I think, is the biggest problem with elite perspectives on AI. It is a global technology, and we are not representative of the globe. Consequently, we tend to underestimate its benefits (since we don’t need them) to everyone other than ourselves.

Does that mean we should waive aside all concerns about targeted advertising, cheating students, AI slop and all the rest? Certainly not. But this present round of anger, like so much else in the anti-AI discourse, is not only unfounded but cruel. OpenAI needs revenue if they want people to keep accessing their technology for free. That access is indubitably good. Targeted ads are annoying and intrusive, but a world where anyone can write a resume in English is worth the price.1

If you’re reading this, writing resumes in English probably isn’t an issue for you — the one language to rule them all was your birthright. You’re also probably well-educated, conveniently placed in the West, and comfortable with the tacit knowledge required to move through modern industry. Congratulations: you don’t need any help from ChatGPT. Perhaps I can introduce you to several billion people who do.

When I was younger, I spent a lot of time arguing about art in my friend’s hot tub. We would tiptoe through his parents’ backyard and ease into the tub — his dad was not fond of nighttime noise — then hit a cannabis vape pen until our faces were numb. We would then discuss Superman, weightlifting, our friends, our girlfriends, Osama bin Laden, etcetera for between four and seven hours. Roughly once per session, we debated whether there was such a thing as objectively “better” or “worse” works of art. I took the positive view.

Subscribe now

My friend is very smart, open-minded, and amenable to disagreements. But not everyone is like my friend. I have had this same argument with other people, and in some cases, the people I’ve talked to about it have gotten very mad. Usually, that’s because they watch a lot of reality TV or mainly read romantasy books. Even before I provide the standards for my aesthetic hierarchy, they know that my pyramid won’t rank their pleasures kindly.

Now, the view that all art is equal in quality is pervasive in our culture: a philosopher would refer to this worldview as “aesthetic relativism.” Beyond widespread endorsement, though, aesthetic relativism also seems to have become something of a liberal virtue signal. Anyone who disagrees with it is seen as somewhere between rude and fascistic.

But as you might have guessed, I remain (more or less) an aesthetic absolutist. I think we can come up with defensible principles that distinguish some works of art as better or worse than others, just as we can come up with defensible principles that distinguish some moral choices as better or worse than others. Joni Mitchell writes better songs than Sabrina Carpenter; The Graduate is a better movie than Fast and Furious; Pride and Prejudice is a better book than A Court of Thorns and Roses.

Here, I want to explain one principle that I think clarifies why all of the above inequalities are true. In short, great works of art — in comparison to bad works of art — can induce both a greater breadth and depth of subjective responses in their audiences. I call this principle generativity.

But let’s ease into that argument. I’ll start with an analogy from a much less controversial topic: the nature of human intelligence.1

‘Twas brillig, and the slithy toves

Picture the smartest person you’ve ever met. Maybe a former professor comes to mind. Maybe you had your self-esteem rocked by a peer in one of your classes who could intuit an elegant and accurate answer to every question posed by the instructor. Maybe you had a passing conversation with someone brilliant at a party.

Whoever that person is, think about them carefully. What exactly made them seem so smart? What felt different about the way their mind works, compared to your cousin or a random stranger?

I’ll tell you who I’m thinking of, and why: Dr. Colin Milburn, a professor of English and Science & Technology Studies at UC Davis. I first met Colin as a freshman, when I took his class on science fiction literature. To this day, I’ve never seen a lecturer who equalled his elegance and coherence. My strongest memory from the course is him reciting the Jabberwocky poem from memory to close out a class on “Mimsy Were the Borogoves.”

But stunned as I was by his lectures, it was not until three years later — when I asked Colin to advise me on my honors thesis — that the singularity of his brain really hit me.

During our first conversation about the thesis, I ventured dozens of references from the books I was reading for my project and several vague connections between them. He somehow had read all the books, even though they had little to do with his field of expertise.

The kicker, though, was how he quickly and enthusiastically revealed the useful core in every one of my half-baked ideas, or if no such core existed, invented one from scratch. It simply did not matter what we were talking about: he could generate deep and important insights on every subject I raised. He was the conversational equivalent of Neo at the end of The Matrix.

I bet you had a similar standard in mind when you picked your smart person. A great mind can take almost any solvable problem and find an exhaustive solution.

Intelligence, in this sense, is a combination of analytical breadth and depth. No one thinks true intelligence is just extreme competence in a single area or superficial competence in several; otherwise, Googling “smartest people who’ve ever lived” would turn up the biographies of autistic savants and podcasters. Sure, some geniuses are not polymaths — the great physicist Paul Dirac, for instance, was maniacally focused on physics alone. But many others are: Einstein, Aristotle, Polanyi. And for the exceptions, there is always the sense that they could probably solve other kinds of problems if they were motivated enough to try.

Let’s summarize. The domain of intelligent activity is problem-solving; exerting intelligence means, most fundamentally, solving problems. Based on the above arguments, high intelligence is not being able to solve one kind of problem really well or being able to solve many kinds of problems poorly. Intelligence is the generalized capacity for solving many kinds of problems well. Someone jumped into your head when I asked you to picture a smart person: the reason you picked them, probably, is the degree to which they displayed this ability.

Okay, fine. What does any of that have to do with art?

Heaven in a wild flower, eternity in an hour

In the previous section, I said that the domain of intelligent activity is problem-solving. By identifying that domain, we came to a defensible conclusion about what makes someone more or less intelligent.

Similarly, I think we can come up with a standard for aesthetic quality by identifying the domain of aesthetic activity: the purposeful generation of subjective responses in an audience. Art is our preeminent experiential technology.

By calling art a “technology”, I mean that art is designed to serve a function. Bikes and cars function to get us from place to place. Phones and emails function to help us communicate at long distances. Music, books, movies, and other aesthetic mediums help us experience worlds, people, and feelings that we otherwise probably or definitely could not access.

This is a pretty uncontroversial take; you find similar arguments at the heart of most theories on the nature of aesthetic activity.

Consider the ancient Indian perspective on art as the work of rasa, which translates from Sanskrit to something like “essence.” Indian aesthetes believed that the central aim of any musical, narrative, or dramaturgical work was to trigger a particular emotional state in its most purified form. Artists drew upon the eight rasas — love, laughter, fury, compassion, and so on — to structure their works; most famously, musicians composed ragas that corresponded one-to-one with particular rasas.

Placing emotion at the core of aesthetics is common. Aristotle held a similar view. More recently, though, philosophers have emphasized other experiential capacities latent in aesthetic activity. One especially interesting framework — focused on video games — was presented by C Thi Nguyen in Games: Agency as Art.2 He argues that games are uniquely adept at helping us experience other kinds of being. When I play Dishonored, I feel what it would be like to be an assassin; when I play Minecraft, I feel what it would be like to be an entity that can fluidly reshape the world.

But I digress. It is hard — perhaps impossible — to conclude that the domain of aesthetic activity is anything other than experience. Where disagreements exist, they usually focus on which kinds of experiences art is meant to induce, or which mediums best serve different varieties of experience. As far as I can tell, there is very little contention about the broader point.

The domain of intelligent activity is problem-solving. Consequently, I said that greater intelligence means greater capacity for solving many kinds of problems well. I’ll draw a similar conclusion about art. The domain of aesthetic activity is the induction of experience. Consequently, aesthetic quality stems from the capacity to induce many kinds of experiences well. Again, I’ll refer to this capacity as generativity.

To pump our intuitions a bit, let’s consider two songs: based on their relative generativity, I think one of them is better than the other. Please listen to them closely.

The first song is called “33 ‘God’’” — it’s by Bon Iver.

The second song is called “PART OF ME (HARDSTYLE SPED UP)” — it’s a remix of a Katy Perry song produced by SICK LEGEND.

Based on the principle I’ve outlined — and based on my own intuition — I think the first song is better than the second. Why?

Speaking from experience, “33 ‘God’” is a song that can make you feel an utterly piercing sense of nostalgia. I used to listen to it while driving back to college after visits home; in my mind’s eye, the flickering piano and cicada-like synths flashed grainy memories of movie nights with my dad and games of tag on my elementary school blacktop. But the song can also scaffold sublimity — I have played it while standing atop mountains — or soothe irreversible losses. It is infinitely rich: listening to it repeatedly, you come to hear the universe in a single song.

Speaking again from experience, “PART OF ME (HARDSTYLE SPED UP)” can do exactly one thing well — it makes you feel manically energetic. I’ve found that useful. I play it if I’m lifting heavy or falling asleep while driving. But there is no there there: what you can experience with this song is probably limited to what I’ve described, though it does that one job well.

Bon Iver’s song is generative; SICK LEGEND’s less so. That makes Bon Iver a better artist.

I started with an obvious comparison. I would be shocked if anyone, based on their gut reactions, disagrees that Bon Iver is superior in this case. But I suspect that some readers might like my other comparisons less. By the standard of generativity, I think most fairy smut books (I’m assuming — I haven’t read them) are artistically deficient. Same goes for Marvel movies, most of Drake’s music, and Rupi Kaur’s poetry.

We could keep going through examples until our brains explode. But it’s really not that hard to figure out whether a work of art is generative or not. If you imagine that going back repeatedly to the song, movie, book, or whatever it may be could yield distinctly new experiences every time, then that work of art is generative. If you cannot imagine the work containing that flexible richness — as I cannot for songs by SICK LEGEND — then the work is not generative. By my standard, at least, that makes it quasi-objectively bad.

To be or

What does generativity — and more broadly, accepting some kind of aesthetic absolutism — mean for how we should structure our artistic lives? Nowadays, this is a really important question. We spend huge swaths of our time listening to music and engaging with narrative media. To a very real extent, art is no longer an imitation of life, but the stuff of life itself.

So, the answer depends on what you want from your life. That’s a personal choice: some people just want to feel comfortable. The full richness of the world is not for them. They don’t care to travel and argue and gaze into the swirling Aleph. For those in that group, I take the cliche view: like what you like, and don’t let people like me judge you for it.

But if your rasa of life is wandering as widely and deeply through the jungle of experience as you can, then you should think carefully about your art. Of course, I won’t ruthlessly demand that you cut anything less than Tolstoy from your library; I love listening to SICK LEGEND at the gym, and I can’t wait for his next Sabrina Carpenter remix. But I will not tell you to like what you like, because liking is not the point of living — at least, not if you’re as pretentious as me.

I’ll close with a quote from the great literary critic Harold Bloom. An interviewer once asked him: why read?

Because you will be haunted by great visions: of Ishmael, escaped alone to tell us; of Oedipa Mass, cradling the old derelict in her arms; of Invisible Man, preparing to come up again; like Jonah, out of the whale’s belly. All of them, on some of the higher frequencies, speak to and for you.”

According to Youber — Uber’s knockoff of Spotify Wrapped — my rideshare archetype is the ‘Go-Go.’ Here is a picture of the Go-Go.1

Subscribe now

Do not be fooled by the title of the picture. I am the Go-Go. You are not the Go-Go. ‘No one moves like me,’ aside from the other 1,303,289 Uber users who were assigned to the category of Go-Go.

Spotify introduced its Wrapped feature in 2016: at the end of each calendar year, the app generates a summary of your top artists and genres and songs and, more recently, characterizations of your listening habits as archetypes, or locations on the globe, or whatever. It is absolutely magnetic. The social media director for McDonalds called it a “masterclass in fan advocacy.”

So naturally, every company copied Spotify, even those whose products do not lend themselves to an interesting year-in-review. Here’s an incomplete list for 2025: Uber, Goodreads, Duolingo, Reddit, Twitch, Youtube, Nike Run Club, Netflix, Reclaim, Hevy, Flow Master, Strava, Ryanair and Merriam Webster. I look forward to Meta’s eventual release of ‘Life Wrapped’ for owners of their Ray Ban AI Glasses.

Presumably, companies don’t introduce these features unless they’re pretty sure customers will enjoy it. It seems like they’re right, at least some of them: I am the kind of person who writes critical think-pieces about these matters, but when I saw the Wrapped button pop up on Spotify, I hit that thing like the cocaine lever in a Skinner box.

Why? What is so interesting to us about ourselves?

I’m Not Like the Other Boys

I’m going to give you an unfinished sentence, and you need to think of three ways to complete it. Ready? I am _____. Come up with three answers before reading the next paragraph.

I might say curious, kind, and a scientist. You probably live in an industrialized Western nation. So I bet you, like me, came up with a set of adjectives, career descriptors, or other phrases that capture your personal characteristics.

Fine, okay. But what does living in an industrialized Western nation have to do with your responses?

You might be familiar with the most famous acronym in social science: WEIRD. People use it to describe the core sociological traits of populations that live in Western, Educated, Industrialized, Rich and Democratic countries. These sociological traits encourage a suite of psychological traits that are otherwise both globally and historically rare. One of these features is heightened individualism — we WEIRDos are unusually self-obsessed.

The “I am ____” game provides a classic demonstration: people from WEIRD places tend to fill-in-the-blank with personal characteristics. But people from non-WEIRD places supply answers like “I am Benji’s brother” or “I am Wendy’s son.” They’re more likely to define themselves in terms of their relationships to other people.

Further support for this result comes from the associations between the wealth of a country and the values expressed by its citizenry. The figure below, drawn from the World Values Survey, captures this correlation. Economic development is clearly bound to ‘secular’ and ‘self-expressive’ values: religious freedom, economic liberty, the right to divorce and so on.

How do we explain these differences? If you want a full answer, I’d recommend a book called The WEIRDest People in the World, where Joe Henrich lays out a full theory of the origins of WEIRDness. But I can give you the Cliff Notes: individualism is very useful in societies where your subsistence depends on the ability to form associations with strangers.

Such is the economic structure of WEIRD nations. In these places, you must cultivate traits that make you, the singular entity that bears your name, desirable to economic and romantic partners. No one cares who your daddy is; they want to know who you are. Conformity makes you replaceable. If you’re just like everybody else, then tough luck — that means you’re nobody at all.

Autoerotic Individuation

So individuality is a means to various evolutionary ends. But something interesting has happened to that arrangement: individuality is also now an end in itself. Bear with me. I want to convince you that we now pursue uniqueness via roughly the same logic that leads us to masturbation.

Sex often produces babies, and evolution loves some good baby-making. So we’ve evolved to derive pleasure from sex, because we do pleasurable things more often. But as a result, we can also maximize the signal in ways that don’t lead to the useful outcome: for instance, through masturbation.

In WEIRD societies, similarly, a sense of uniqueness is a good signal that you’re doing something right. We must try to distinguish ourselves in the market, and we can use signals of our individuality to note whether we’re succeeding. But the pleasures associated with that signal can be hijacked. We can be tricked into chasing individuality for its own sake.

And that is exactly what has happened. Consider the only group of people that literally everyone hates: NPCs. This acronym stands for ‘non-player character.’ Traditionally, it refers to entities in video games that are not guided by the agency of the player. They repeat voice lines, walk in endless loops, and otherwise act in stereotyped ways that indicate their lack of autonomy.

Now, the word has escaped gamer lingo and become a popular insult. We use it to disdain someone whose actions seem predictable based on some mental model, like a political stereotype. An NPC lacks individuality; they act like iterations of an archetype, rather than agents in and of themselves.

Despite our many divides, we all seem to agree that there is nothing worse than an NPC. People throw this insult from every political direction. Zoomers use it to insult Boomers, and Boomers would use it to insult Zoomers if they knew what it meant: when they criticize the youth for being brainwashed by TikTok, they basically mean the same thing. In 2023, three boys yelled ‘NPC’ at a man from across the street. The man chased the kids into a Dollar Tree, then stabbed and killed one of them with a five-inch knife.

But as much as we hate conformists, we equally love their opposites. We make quirky wallflowers the heroes and heroines in every modern coming-of-age movie. We treat strange sleeping habits and social awkwardness as signs of entrepreneurial genius. We adore and lust after celebrities like Robert Pattison, Timothee Chalamet, and Lakeith Stanfield in part because they act like freaks in their interviews.

And that’s without mentioning all the ways we fetishize our personal sense of uniqueness. Personality tests, mental health self-diagnoses, the development of ‘personal aesthetics,’ the learned sycophancy of ChatGPT and the vast glut of self-help literature can all be understood as super-stimuli for our individualistic instincts. Individuality is our kink, and we just can’t stop gooning.

Happy Angels, Happy Ants

Thus Wrapped is born. Cocaine lever, corn syrup: a Fleshlight to sate our limitless lust for singularity. Spotify Wrapped says “Taste like yours can’t be defined.” Youber says “No one moves like you.” They know exactly what their customers want.

And the customers want it— they really, really do. This year, Spotify provided customers with a “Listener Age.” Mine was 22, which seems embarrassingly ordinary. I keep justifying the result to people by saying that I play too much pop music at the gym: if only Wrapped knew the real me! Then my interlocutors tell me that their results were also wrong, because they play too much lo-fi while studying. We bask together in the glow of our shared need for eccentricity. What else is a WEIRDo to do?

But ironically, the release of Wrapped can also encourage the opposite conclusion: that we are not special at all. Everybody I know posts their top songs on Instagram, sends me a screenshot of their most listened-to album, and comes up with the same shitty explanation for why they listened to so much Drake. And I do it too; I just can’t help myself. Observing and riding this wave of excitement, I find myself feeling less and less like an individual, and more and more like an NPC.

Which is, of course, the truth. The social strictures of WEIRDness demand individuality, but our evolved psychology wants no part of it — though we are told to distinguish ourselves from the crowd, we cannot stop ourselves from surrendering to its motions. We are 10% angel and 90% ant, impossibly reaching toward divinity.

Wrapped understands this fact; it knows that alongside our cultural desire for uniqueness, a deeper part of us wants to ensure the contours of our conformity. Spotify assigns me an archetype, then tells me about the other 100 million users who fell into the same camp. “All your listening,” it says, “made you part of something bigger.” In 2023, I and every other young leftist on Earth were categorized into the listenership of Burlington, Vermont. There was something lovely about that. I felt my entire generation alongside me, soundtracking our confusion to Big Thief.

Every December, we rush to Wrapped to trace the shape of our solitary journeys. Then we find that all the searching has led us right back to each other. Ants transcend themselves in their masses; angels kneel together before God. Like our species, the power of Wrapped is two-fold, Janus-faced, dialectic: a singularity formed from the division, culture and cognition, that lines our human hearts.

So, to the other 1,303,289 Go-Gos out there — it was a pleasure to have Gone-Gone with you this year.

Roughly twice a week, the Savanna Pumé — a tribe located in southwestern Venezuela — hold an all-night dance. An anthropologist named Vincenzo Petrullo described the early stages as follows:

The song was first a murmur and then a shout, but maintained its steady rhythm. When the moon brightened the tropical night with its colorless light, sharpening the many shadows . . . the Yaruros forgot all about the stranger in their midst. As though on a prearranged signal, they arose and began to dance . . . they danced all night, and it was not until the sun rose above the hills of Guiana that they stopped to seek an hour or two of sleep.1

Subscribe now

On eHRAF, a database of ethnographic records from hundreds of human societies, this dance is described as a means for “communicating with the spirits of dead relatives.” A serious affair, then: an example of capital-R Religion, replete with snuffs and shivering and desperate outcries to the gods. Like so many other societies, the Pumé engage their beliefs through ecstasy.

Anthropologists like to explain the ubiquity of these practices in all kinds of complicated ways. We exhaust ourselves to signal the depths of our conviction to others. Our irrational minds lead us to invent ways of controlling the uncontrollable. Dancing and drumming and chanting is a form of self-deception: the resultant chaos feels like a legitimate sign of divinity.

All of these explanations probably capture bits of the truth. But there is also a simpler account: the gods are a great excuse for a party.

When we look only to familiar practices from Abrahamic religions, like Christianity or Islam, this possibility is easy to dismiss. Going to church may feel rewarding — but my god, is there anything more boring than sitting in a pew for three hours?

But the absence of joy in these doctrines is an exception2. In many religions, ritual takes the form of festivity.

Last summer, I conducted my first round of fieldwork in southern Africa. My collaborator and I started out in Johannesburg, interviewing traditional healers in the townships — and during one of these interviews, we were invited to attend an initiation ceremony.

I couldn’t believe my good luck. A sacred event! Terrified to screw it up, I asked my collaborator what I should bring, how I should act, whether I was allowed to use my camera and notebook.

She shrugged. “You’ll be fine. Let’s just pick up some beer on the way.”

She was right. When we arrived at the house where the ceremony would take place, the presiding healer warmly greeted our 12-pack and placed it in a growing stack of alcohol. Then she went on her way.

I wandered the premises, took notes and periodically explained to a teenager that I did not know Jaden Smith. Every ten minutes or so, someone would sound a piercing whistle, and a group of initiates — waiting in a cordoned-off room — would roar loudly in reply.

The ceremony happened in a very big tent. Three adolescent boys dragged a mix of animal-hide drums and metal tubs into one corner and began to pound an utterly deafening polyrhythm. The initiates emerged. They danced before an audience — I was placed in the front row, due to my (undesired) honorary status — and followed the drums. As the beat slowed, their stomping slowed; as it quickened, so did their feet.

The dancers had reached the zenith of their training. They fell into trance and summoned ancestral spirits. I felt distinctly like there were, in fact, presences in the tent that had not been there before.

But as the night progressed, the barrier between audience and initiates broke down. Boys traded their places at the drums and danced impossibly fast for the cheering crowd. Old women rose and hit a careful two-step. At one point, an initiate swung himself wildly around a pole, and everyone whooped like they had just seen a trick from Cirque du Soleil.

Around 3 AM, I left the tent and joined several men by a fire that had been built in the courtyard. They were all very drunk. They called for me to take pictures of them: “Shoot, umlungu3, shoot!” One of the men cornered me and demanded that I finish a 40 oz bottle of beer with him. He reminded me of the strangers that would grab me at house parties during college, proclaim their undying friendship and promptly vanish into the night.

We left the next morning. The party would continue without us for another day and night.

Now, the people living in townships outside of Johannesburg are obviously not traditional foragers. They are thoroughly industrialized; some attendees of the ceremony scrolled Facebook during the climactic goat sacrifice.

Yet across human societies, events like the one I’ve just described are remarkably common. My advisor, Manvir Singh, has written beautifully about the celebratory aspects of such ceremonies among the Mentawai, a hunter-gatherer group living on a tropical island off the western coast of Sumatra. Here’s how he describes a healing session:

On paper, this was a religious ritual. Yet more than anything, it felt like a party. Mangga’s ankles looked like they were decaying, but the attitude was neither solemn nor serious, but amusing and playful. Full from pig and sago flour and sweet coffee, we laughed and chatted during dance breaks. It felt less like Mangga had called these shamans to cure him and more like he was hosting an all-nighter for his younger relatives.”

Trance dances among the San are similarly festive. Here’s a description from Elizabeth Marshall Thomas:

The women sit in a circle around the fire with their babies on their backs and sing the medicine songs in several parts with falsetto voices and clap their hands in a sharp, staccato rhythm. Behind them the men dance one behind the other and circle around slowly taking short, pounding steps in counterpoint to the rhythms of the singing and clapping. This is accompanied by the sharp, high clatter of rattles – made from dry cocoons strung together with sinew cords – that are tied to their legs.4

In some sense, this process is rather grave. Through the dancing and clapping and singing, healers channel boiling energy through their stomachs, transition into a state called ‘half-death’ and use their power to remove ghosts from the bodies of the sick.

Yet the dance circle also resembles something less sacred: the spontaneous groups one might find forming in a rave. Less skilled performers bob at the edge of these circles; the bold step inside and perform grand feats of bodily imagination, much to the delight of their audience. The great ones seem to lose their soul in the crowd.

So much of what we so carefully call religion is, perhaps, just a means for collective joy.

Of course, it’s not all for fun. In Lesotho, healers told me that falling into trance is frightening and sometimes painful. One described the terror she first experienced upon surrendering her body to a spirit; she felt that her soul had been trapped on an unknown island, and she wondered if her volition would ever return. The pleasure of these ceremonies is often paired with immaculate pain.

But I wonder if we sometimes take these things more seriously than we should. A trance dance is seemingly alien: all the talk of possession strikes us as fundamentally strange. Yet we too love to lose ourselves in a spinning crowd. Call it trance, or half-death, or ecstasy. Beneath the sound and fury of our science, there is always that selfsame fire.5

Republished from the November 2029 issue of Wired, with permission.1

Eliud looked a little bit dizzy. He confessed as much before we started our third virtual interview — I was in San Francisco, and he was in Nairobi. “The new work is more tiring,” he explained. “You stop trusting your eyes.”

Subscribe now

In the last ten years, venture capital has steadily transformed Kenya into a so-called “Silicon Savannah,” spurred by the country’s vast supply of unemployed but well-educated workers. Controversially, many of the interloping firms — in particular, OpenAI — have hired Kenyans en masse to serve as ‘humans in the loop’ for AI training.

Their job is to label the contents of visual, textual, auditory and other forms of data to help neural networks extract the relevant patterns. An employee might look at several hundred pictures of food, for instance, and note all the pictures that display soup.

The work is low-paying and exhausting. For struggling workers, however — and for companies that want to develop state-of-the-art AI — there is no alternative.

But recently, centers like the one that employs Eliud have been retrofitted for a new kind of training. Eliud still sits down at his computer every day. But the text files are gone. Instead, he puts a net of electrodes on his head. For the next ten hours, the electrodes stimulate his brain for periods ranging from three to thirty seconds. When they stop, he tells the computer what he experienced.

Eliud was disturbed by this development. He acquired my email address from a friend. I had interviewed her previously for a story about the Kenyan government’s use of machine vision algorithms to identify and arrest youth activists.

“Can you tell me what happened today?” I asked.

“Redness,” he said. “Just a wall of color. Then a very painful pinch in my shoulder.” He rubbed his neck. “Some music, later. But simple music, not good. One note going on and on and on. Oh, and I smelled trees during the music. Usually it’s only one thing. But sometimes there are more.”

OpenAI, it seems, is trying to build an experience machine.

In 1974, Robert Nozick imagined a machine “that would give you any experience that you desire.” His vision was hypothetical; he used it to demonstrate the pitfalls of hedonism. But in principle, it is no longer hard to design an experience machine.

First, recall that we can induce different conscious experiences by stimulating the brain with electrical currents. Wilder Penfield, a legendary neurosurgeon, is usually credited with this discovery. Seventy years ago, he placed electrodes directly onto the temporal cortex of an epileptic patient. She was fourteen years old; history knows her as J.V. When Penfield stimulated one location, J.V. said: “I imagine I hear a lot of people shouting at me.” In another: “I saw someone coming toward me, as though he was going to hit me.”

So a functional experience machine would work by tracking which patterns of electrical stimulation produce particular experiences in the brain. If a machine possesses that map, and a tool that could safely apply it, a user would be able to trigger experience on demand.

In practice, though, this blueprint faces two difficulties. One is mechanical: the machine would need detailed control over the brain to cover the vast varieties of possible experience. But this is arguably just a matter of building better and more particular tools — a hard problem, but not an impossible one.

The (seemingly) impossible problem is scientific: how, exactly, are we supposed to precisely map the varieties of experience onto different parts of the brain? To figure out the sounds made by strings on an instrument you’ve never played before, you can just hit the strings and listen. But what if the instrument has 100 billion strings?

During the winter of 2027, Dmitriev Levin started sleeping in his office. I met him on the Charles Esplanade a few months later, when the Boston cherry blossoms had just begun to bloom. His gait was short and careful. He stared at the ground when he spoke.

“A bad divorce,” he explained. “I brought an old futon out of storage.”

Levin is a renowned neuroscientist — his lab has been responsible for many of our most fundamental insights into the mechanisms that underly consciousness. Around the time he brought out his futon, OpenAI had successfully established new data centers in Texas, Mexico, and Ohio as part of their ‘Stargate’ project. Simultaneously, the wave of interest in artificial intelligence had finally begun to seep into his field, not only philosophically but methodologically.

“I never cared if it was conscious,” Levin told me. “Too speculative for debate, yes? But I am always looking for new empirical techniques. That seemed more interesting.”

Levin was especially interested in a new approach from cell biology: prompt-to-intervention models. Several prominent biologists had been arguing for years that one could command a group of cells without guiding the movements of every individual cell. They analogized the idea to the power of a programming language; if you can use Python, you don’t need to write directly in the much more confusing machine code that underlies it. Ideally, we could tell the cells in an amputee’s stump to make a new arm using high-level instructions, and they would figure out how to do it themselves at the low level.

Michael Levin, a distant cousin of Dmitriev, had referred to this hypothetical bio-Python as the “anatomical compiler.” Dmitriev said he and Michael had never spoken; nonetheless, he found the idea inspiring. And in August of 2025, Michael and his colleagues released a paper that would inadvertently lead Dmitriev to a blueprint for the experience machine.

The paper detailed a new technique for cellular control. These models allowed users to command a cell culture through natural language commands, like “cluster” or “scatter.” A neural network would then try to translate the command into low-level instructions for a simulated cell culture.

Over many attempts, the network would learn how to accurately map commands onto behavioral outputs for the culture. Consequently, the final model could take a command like “make the cells form a tight cluster in the center”; with no further intervention from a user, the model would translate that into cellular movement. Hence the name: a prompt leads to an intervention, with a neural network playing the role of middleman.

The paper didn’t get very much attention. But Dmitriev, leaning against the lightless windows of his office, read it several times over. His cousin could use artificial intelligence to tell groups of cells how to behave, even if he didn’t know how his commands were translated by the network into practice. Why couldn’t Dmitriev do the same thing with groups of neurons?

Good artists borrow; great artists steal. “I am shameless,” Dmitriev admitted. “But Mike didn’t mind. He emailed me to say he liked the paper.”

In an article published a couple months later, Dmitriev proposed a P2I model for conscious experience. He suggested that a neural network could be trained on data about the subjective effects of stimulating the brain in different ways. Take a subject, electrify their brain, and ask them to describe what they felt. Then try a new stimulus pattern, and ask them to describe the experience again. And again, and again, and again.

Eventually, if the neural network is paired with a sufficiently advanced controller, it should be able to convert natural language into patterns of neural stimulation. You could tell it “Show me a pink tree,” and using an internal map between natural language and patterns of electrical stimulus, it would galvanize the brain to do exactly that.

Of course, neither the user nor the designer would ever get to see the map. Dmitriev’s proposal, brought to life, would bring his field no closer to understanding the neural correlates of consciousness. But that didn’t restrict the potential for such a machine to provide detailed control over conscious experience. Programmers don’t write in machine code; pianists don’t play the cochlea; to use the experience machine, you wouldn’t need to know how it works.

I asked Dmitriev if he had thought about whether someone might actually implement his idea. He considered it, briefly, then shrugged. “No,” he said. “Not really. It just seemed true. So I did it.”

So he did. There was little response from the press; the concept was too speculative to make many waves. By the time Eliud contacted me with his story, Dmitriev had moved back into his apartment.

When Eliud first tried the experience machine, he thought he had gone insane.

“It looked like a kofia,” he said. “But made of a very fine material. Like chainmail from Lord of the Rings. The supervisor wouldn’t answer my questions. He just puts it on my head and leaves. Then I open the computer, see the console, the net gets warm, and I hear a little ‘pop’ in my ear, over and over and over. And I think, oh no, tar chera. I’m done.”

“Tar chera?”

“Crazy. Wires ripped in the head.”

Nozick described the experience machine to make a political point. Developing a version of the machine that would actually fit his vision, though, requires lots of trial-and-error. The data-heavy demands of neural networks, and the complexity of the problem they’re attempting to solve, inflate this problem exponentially.

To build an experience machine, OpenAI will need thousands of people like Eliud: workers who, by dint of the womb lottery, have few alternatives to any wage on offer, even if the price is their sanity.

“Sometimes I get very confused,” Eliud said. “I came home from work, yesterday, and I remembered a necklace my grandmother had. It was a pendant, a very beautiful blue pendant. And I remembered the kind of blue it was, then realized it was not really that kind of blue.”

Eliud tapped his temple with his index finger. “I was picturing a color the machine had shown me. The things get in your memories, I think. They get in your dreams.”

For now, the experience machine mostly generates “atomic qualia,” to borrow a phrase from Levin. In the philosophy of mind, qualia are the raw feelings of consciousness: the soursweet tang of a strawberry, the sound of a chord, the particular redness of a sunset. By extension, atomic qualia are the most basic features of experience — the atoms of subjective life. The taste of a strawberry might comprise the atomic qualia of sweetness, sourness, and so forth.

Not everyone agrees that there are such things as atomic qualia. But as a shorthand, it usefully describes the kinds of experiences that Eliud and his colleagues have reported. Fundamental forms of consciousness, colors and sounds and smells: and occasionally, a more complex mixture, a molecular qualia.

Almost undoubtedly, though, the end goal for OpenAI is a machine that can do exactly what Nozick described — generate any conceivable experience using the contours of natural language. We can tell a neural network to reorganize a colony of cells, write computer code, and create paintings. If the project achieves its aim, we will also be able to tell them to change our minds.

It is hard to say whether this aim is within reach, or whether it is possible at all. Dmitriev suspects that it is closer than one might think. “All you need is data and power,” he said. “For solving these kinds of problems, it is that simple.”

“Will it tell us anything about consciousness?” I ask.

“Ha. That will be much more complicated. But I think most consumers can do without the understanding. They’ll be happy enough with their porn.”

Dmitriev, like Nozick, is pessimistic about the effects of the experience machine. He worries that it will accelerate us further into an age of lonely hedonism. This is a common view among intellectuals — barring a few notable exceptions, the post-Nozick tradition has preferred to argue against the worth of false experience.

Eliud wasn’t so sure. When I asked him whether he thought an experience machine would be good or bad for the world, he shrugged. “I don’t know. Maybe it will distract us from real problems. But we already have to distract ourselves so much, all the time. Because life is so hard, you see? Shouldn’t people get to be happy?”

“Has the machine made you happy yet?”

Eliud looked away from the camera for a moment. “Mm. A few times, yes.”

“Can you tell me about one of those times?”

“Well. There was something else I saw today. I did not say it, because it sounds so strange. But I saw a woman’s face.”

“I see,” I replied. “Did you recognize her?”

“No, no. She could not be from here.”

“And seeing her made you happy?”

Eliud smiled. “Yes. She had the most beautiful eyes I have ever seen.”

I was interviewed about this piece on NPR Weekend Edition. Here’s a link to the interview.

I don’t remember the first time I held a machete, because I’ve never held one. Most members of the BaYaka — a group of nomadic hunter-gatherers found in the Congolese rainforests — probably don’t remember either, but for different reasons. Early memory has its limits. Among the BaYaka, picking up a machete is developmentally akin to language, walking, and chewing solid food.

Subscribe now

So goes a BaYaka childhood. Children wander the forests in packs. They climb saplings and bathe in the rivers. They conduct day-long fishing trips: their parents glance toward them as they organize themselves, then let the kids go on their way.

A few months ago, an anthropologist named Gül Deniz Salalı documented these dynamics in a brilliant documentary (linked below), which I cannot recommend highly enough. To our eyes, their lives are utterly strange. We shouldn’t forget how lucky we are to live in a time where we can see such wonders from the comfort of a chair.

But the BaYaka childhood isn’t a novelty. As I’ll discuss shortly, it is probably the norm for our species. And that means something has gone terribly wrong in the West.

Consider some statistics on the American childhood, drawn from children aged 8-12:

• 45% have not walked in a different aisle than their parents at a store;

• 56% have not talked with a neighbor without their parents;

• 61% have not made plans with friends without adults helping them;

• 62% have not walked/biked somewhere (a store, park, school) without an adult;

• 63% have not built a structure outside (for example, a fort or treehouse);

• 71% have not used a sharp knife;

Meanwhile, 31% of 8-12 year olds have spoken with large language models. 23% have talked to strangers online, while only 44% have physically spoken to a neighbor without their parents. 50% have seen pornography by the time they turn 13.

In physical space, Western children are almost comically sheltered. But in digital space, they’re entirely beyond our command; and increasingly, that’s where children spend most of their time. You don’t need me to tell you about the dire consequences of that shift.

Why do our children spend more time in Fortnite than forests? Usually, we blame the change on tech companies. They make their platforms as addicting as possible, and the youth simply can’t resist — once a toddler locks eyes with an iPad, game over.

I want to suggest an alternative: digital space is the only place left where children can grow up without us1. For most of our evolutionary history, childhood wasn’t an adult affair. Independent worlds and peer cultures were the crux of development, as they still are among the BaYaka; kids spent their time together, largely beyond the prying eyes of grown-ups.

But in the West, the grown-ups have paved over the forests and creeks where children would have once hidden. They have exposed the secret places. So the children seek out a world of their own, as they have for millennia, if not longer. They find a proverbial forest to wander. They don’t know what we know: this forest has eyes and teeth.

Bridges to Terabithia

In most human societies, children have spent much of their time exploring and playing within independent peer cultures. This term reflects two important features of human childhood. First, the groups consist entirely of other children. Second, they are functionally and culturally distinct from adult society; they exist alongside but apart from the world of adults.2

The evidence for this pattern is rich and widespread. During his research among the Trobriand Islanders, for instance, Bronislaw Malinowski described the “small republic” of children that “acts very much as is its own members determine, standing often in a sort of collective opposition to its elders.”

Margaret Mead reported similar patterns in Samoa. One group of young girls formed a cohort “which played continually together and maintained a fairly coherent hostility towards outsiders.” Of their activities, she writes:

On moonlight nights they scoured the villages alternately attacking or fleeing from the gangs of small boys, peeking through drawn shutters, catching land crabs, ambushing wandering lovers, or sneaking up to watch a birth or a miscarriage in some distant house . . . They were veritable groups of little outlaws escaping from the exactions of routine tasks. (Coming of Age in Samoa, p. 62).

These peer cultures don’t like to be seen — many ethnographers have noticed how playgroups prefer to segregate themselves from adults. Among the Mbuti, another Central African foraging group, Colin Turnbull observed children spending most of their time in a bopi, a playground set away from the main camp:

The water was fairly shallow there, and all day the long the children splashed and wallowed about to their hart’s content . . . Infants watched with envy as the older children swung wildly about, climbing high up on the vine strands and performing all sorts of acrobatics. (The Forest People, p.128).

Even in industrial societies, this love of solitude shows up. In the 1950s, Iona and Peter Opie documented the behaviors of post-war British children. They noticed that the children liked to roam through bomb sites, where they would build fires and play hide-and-seek.

It is possible to see these interests playing out in ancient societies, too. A paper by Ella Assaf and colleagues documented handprints, footprints and paintings left in Paleolithic caves, all probably produced by children. Through involvement in ritual activities, they propose, “Upper Paleolithic children were able to actively shape their own reality as individuals, as well as the reality of their community and its well-being.”

Why do independent peer cultures emerge so reliably in our species? Anthropologists tend to view childhood as a period during which we over imitate and copy our way to cultural competency. But if that’s true, it seems hard to account for the fact that kids really, really want to get away from adults. They would rather spend time with each other. If childhood is all about efficient cultural transmission, children should be hanging on our every word.

As of yet, I don’t think there’s a good explanation. Maybe independent peer cultures expose children to a wider variety of information. Maybe they provide safe spaces for the mimicry of adult activities. Maybe adults just aren’t very much fun. As Antoine de Saint-Exupéry writes in The Little Prince: “All grown-ups were children once, but only a few of them remember it.”

All of these accounts are probably a little bit true. The important point is that kids want to spend time together, in their own space, away from the tiresome grown-ups.

Man is born free, and everywhere he is in chains

Which is relatively easy, if you’re living in a giant forest with parents who don’t pay much attention to your activities. But our children now find themselves in a strange situation — they have nowhere to hide, and even if they did, we might not let them go there in the first place.

Over the past few decades, childhood mobility in the West has dropped precipitously. You might think that the change has something to do with the emergence of the Internet. But longitudinal data suggests otherwise. Check out these statistics from a decades-long survey on the mobility of English children:

At least in England, independence isn’t an Internet problem. Drop-offs in English childhood mobility have been ongoing since the 1970s. In 1971, 80% of seven and eight-year-old children went to school unaccompanied by an adult, and 55% of children under ten-years-old were allowed to travel alone to places other than school within walking distance. By 1990, those numbers had dropped to 9% and 30%, respectively.

We see similar patterns in other places. Here’s data from Sweden, for instance.

In the United States, similarly, there was a drop-off from 42% (1969) to 16% (2001) in the number of children who walked or biked to school alone.

So it doesn’t seem like the collapse of independent mobility is a phone issue. The truth is much more complicated.

One element is parental attitudes: according to responses from a survey by Play England, many parents fear “stranger danger” or judgement from neighbors if they let their kids play unsupervised outdoors.

Adult employment patterns and lifestyle changes have also been slowly trending toward car-dependency, which means that kids often end up living far away from their friends. If children want people to play with, the most efficient solution is for their parents to drive them to an organized sport or other structured activity.

In the Play England survey, though, parents were most afraid that their kids would get hit by a car. Sadly, this isn’t an unreasonable fear. All the forests are covered in concrete. What would we make of a city-bound parent who let their toddler roam the streets without an adult nearby?

Unsurprisingly, these changes have been very bad for children. In 2013, UNICEF tracked childhood well-being against independent mobility, and their results reveal a stark correlation.

A mix of shifting parental attitudes, car-dependency, and urbanization have led to an unprecedented situation in the history of human childhood. Children don’t have peer cultures, because they have nowhere to play and no one to play with. Even if those factors were ameliorated, we might not let them out anyway. Too bad for them — they can’t hide from us anymore.

Or can they?

A world without us

The kids won’t get off their damn phones. Children aged 6-14 average nearly three hours per day of screen time, not counting schoolwork. 50% of teenagers average over four hours of screen time daily. And so on — you know the statistics.

Amidst all the horror stories, though, people often ignore the fact that the kids don’t like it either. For example: most teens, especially girls, say they spend too much time on their smartphones. 72% of 8 to 12-year-olds say they would rather “spend most of there time together doing things in-person, without screens.” The kids are not alright, but they aren’t dumb — they understand that something is wrong with the technological world we’ve handed them.

So why don’t they just stop?

Usually, we blame the corporations. And there are lots of fingers to point. It is now well-known, I think, that many platforms design their applications like slot machines to maximize our attention, and that they purposefully aim for the more malleable minds of children.

But data on the varieties of screen usage suggests another explanation. Social media and gaming are now arguably the two most common digital activities for children: notably, and unlike television, both allow children to engage with distinct virtual communities that adults don’t notice or understand.

Tellingly, kids really want those communities to exist. 45% of American 8 to 12-year-olds say they would prefer to “participate in an activity with their friends in person that’s not organized by adults.” 61% wish they had more time to “play with friends in person without adults.” And again, most of them wish they could spend less time on screens. It seems like what they want is to wander together in a forest.

But they can’t. So they boot up Fortnite or TikTok instead.

By retreating to digital space, children have found an open frontier that lies beyond the interest or comprehension of their parents. We don’t know how to play Roblox, or what “6-7” means. These worlds are so impenetrable that The Guardian writes explanatory articles to keep us at least slightly up-to-date.

But the kids know, and they know that the other kids know, and they know that we don’t know shit. Children have always looked for a world without us. With the advent of the Internet, they have found one lying in the rubble.

Learning to walk in the (e-)forest

Of course, the problem is that social media platforms and video games are not Congolese rainforests — though at least there are no leopards. These digital spaces make our children depressed and anxious and insecure. They expose them to pornography. They drive them into frightening political rabbit holes. The children need somewhere to play, but this can’t possibly be the right place for it.

And yet, it’s hard to picture an alternative. Parents should worry less about stranger danger: fine. But everything is still covered in concrete, and everyone is still moving to cities. Do we want to tear down the skyscrapers and unpave the roads? Should we demand a forest for our children?

The truth is that we’re not going back. My children will not grow up like the BaYaka, and there is nothing I can do about it.

But that doesn’t change the fact that kids need their independent peer cultures. If we can’t provide physical spaces for them to form, then we must accept that they will often form in digital spaces instead. So if we’re unhappy with the digital spaces on offer — if we think there are too many figurative leopards in those forests — then we should make something better.

What does ‘something better’ look like? Well, probably a platform that preserves the aspects of digital space that kids find freeing, but without the aspects that make those spaces dangerous and addictive.

Games like Roblox present some interesting ideas. Children absolutely love Roblox; according to the game’s parent corporation, their monthly player base includes half of all American children under the age of 16. They love it because it’s multiplayer, exploratory, and extremely open-ended. Independent cultures and systems of governance emerge in the course of gameplay. They’re sufficiently complex that some of my colleagues are now using Roblox as a playground for studying collective behavior in humans.

But there are adults in Roblox, hidden behind avatars. Their presence ensures that kids will end up seeing disturbing content that they don’t have the tools to understand. Roblox is also highly Vegasified, featuring all the usual exploitative tactics of loot boxes, season upgrades and cosmetic passes. So it is nowhere near the ideal; all things considered, children would probably be better off if we banned them from playing.

Still, I can picture something like Roblox that might also sustain peer cultures. When I remember the happiest parts of my childhood, everything centers on secret places: pillow forts, hidden corners in parks, soccer fields after dark. But I also remember Minecraft. My friends and I would set up a Skype call, start a world and spend hours exploring and building.

When we spent time together in physical space, we were nearly always being supervised. Not so in Minecraft. I spent thousands of hours mastering the rules of that procedurally generated world; my parents didn’t even know what ‘procedurally generated’ meant. So, dystopian as it may sound, Minecraft servers were the closest thing I had to sprawling Congolese rainforests.

Perhaps that isn’t so bad. I wish the children of today had a forest. But they don’t. They’re making do with what history has handed them.

We can complain about their screen time, lament the anxious generation, scoff at how ‘unnatural’ this brave new world has become. Simultaneously, though, we should do our best to understand why kids are behaving this way. There’s no point in whining about the impulses endowed to them by several hundred thousand years of evolution. Don’t hate the player; hate the game. And if you really hate the game, make a better one.

Please read this footnote first.1

Take this as a baseline: factory farms are hellscapes, and the suffering of the several billion sentient beings we have placed inside them is unimaginably horrific. If you’re reading this article, you probably already knew that. If you’re surprised to hear it, then you should read this essay by Bentham’s Bulldog.

Subscribe now

Moving on — I ate three slices of pepperoni pizza last weekend. In this post, I want to convince you that this small act means I probably would’ve been a Nazi, and that anyone who buys and eats factory-farmed meat falls into the same moral boat.

Why2 ? Bear with me. Consider the story of Charlie and Dennis.

Orgy-porgy, Ford and fun

On a chilly night, you walk up a narrow staircase and enter a 2nd-floor apartment overlooking the hills of San Francisco. You take off your shoes (the host is very clean), grab a drink, and strike up a conversation with Dennis. He compliments your jacket and poses thoughtful questions.

During this conversation, you ask Dennis about his hobbies. He tells you that he loves to spend time with his family. Every Sunday, he makes a batch of bacon for his wife and their two kids. It’s their weekend tradition. How cute!

Eventually, you excuse yourself to grab another drink. You think to yourself: Dennis is a decent fellow.

Then you start talking to Charlie. He also seems nice: he compliments your shoes and listens carefully to everything you say.

You ask about his hobbies. He says: “Oh, I like to pay a kennel in Oakland to take dogs from the pound, cut off their snouts, tails and testicles with no anesthetic, stuff them in tiny pens where they can’t turn around or move for 99% of their lives, and feed them kibble laced with feces. Once they’re old enough, the kennel beats them so they’ll get into a truck, takes them to a slaughterhouse, and slits their throats. Then every Sunday, I pick up the meat and cook it for my wife. I like it crispy!”

Horrified, you back away from Charlie and tell the host what he said. She gasps and kicks him out of the party. You think to yourself: Charlie is a fucking monster. He might be the worst person I have ever met in my life. You reconvene with Dennis and wonder how such awful people can exist in the modern world.

As you already figured out: everything that the hypothetical kennel does to those dogs is exactly what virtually all factory farms do to pigs3. The pepperoni on my pizza, and the bacon that Dennis cooks every weekend, is sourced directly from a nightmare. In material terms, Charlie and Dennis and I are committing identical crimes.

But only Charlie gets kicked out of the party. Why?

I hope you find this question as disconcerting as I do. Intuitively, I don’t think of myself as a sadistic psychopath. Nor do I think that the vast majority of Americans are sadistic psychopaths.

Yet if you accept three basic premises — (A) Charlie is a monster, (B) pigs are just as capable of suffering as dogs, and (C) Dennis is paying for exactly the same service as Charlie — then it seems very difficult to avoid the conclusion that Dennis is morally equivalent to Charlie, and therefore that anyone who behaves like Dennis is also equivalent to Charlie.

In other words: without a good answer to this question, I must conclude that pretty much everyone I know, myself included, is a monster.

Four Dumb Retorts

Here are four ways to fail at escaping this repugnant conclusion.

1. Pigs are not morally equivalent to dogs. Therefore, their suffering is irrelevant, just as the potential suffering of any minimally or non-sentient being is irrelevant.

The only principled way this could be true is if pigs were significantly less intelligent than dogs, since we tend to treat intelligence as a proxy for the capacity to suffer. But pigs are at least as smart as dogs. And regardless, intelligence is a really bad proxy for moral worth. Eating healthy adult humans is bad, but no one thinks it’s any better to eat a baby or someone with mental disabilities.

In sum, this response is bullshit. How about Door #2?

2. Torturing and eating pigs is an unavoidable part of life, while torturing and eating dogs is not. Therefore, we cannot judge someone for doing it, just as we cannot judge them for killing poisonous spiders in their bathroom.

Clearly, this is also not true. In this particular case, the alternative isn’t veganism, or even vegetarianism — it’s not eating pigs. Given that a few billion people manage without it by dint of religious scripture, eating factory-farmed pigs is clearly avoidable. But Dennis is doing it anyway.

3. The benefits outweigh the costs.

Nope. Pig meat isn’t very healthy, and there are no pathologies that require someone to consume it. Here are the main benefits of eating pigs: they taste good, and it makes some social events easier (at least in the United States). Here are the main costs: you’re paying to wreak unspeakable horrors on an animal that is just as conscious as your golden retriever.

This retort is logically equivalent to a child molester who says: “I know it’s bad, but having sex with children is the best!” And of course, you wouldn’t feel any better about Charlie if he explained the delectable wonders of dog meat and talked about how it’s the centerpiece of his weekly barbecues with the boys.

4. Dennis doesn’t know anything about factory farms, so you can’t blame him for purchasing meat from them.

I mean, maybe? If Dennis, somehow, really has no idea what happens on a factory farm, then maybe he’s less culpable than someone who does.

But for most people — particularly the kinds of people likely to read this post — that level of ignorance is exceedingly rare. Not everyone is an expert, but I suspect that pretty much everyone knows, at least vaguely, that factory farms make animals suffer very, very badly. I also assume that the average 19th-century Southerner didn’t know exactly what happened on slave plantations, even though they knew about slavery. How do you feel about the 99% of them that weren’t virulent abolitionists?

So none of these responses help us to avoid thinking that everyone we love is a psychopath. In my view, there is only one option left. But you’re not going to like it.

The Moral Law of Social Baselines

There is exactly one way to think that someone who pays to torture and consume pigs is not as bad someone who pays to torture and consume dogs. The former is a social baseline, and that means we can’t judge anyone who participates in it. In other words: if a practice is the norm in a society, then an average level of engagement in that practice is never worthy of moral condemnation.

One implication of this claim is pretty standard fare in left-leaning circles: we call it moral relativism. Infanticide, genital mutilation and ritualized murder — among other abominations — are common across many indigenous cultures. Confronted with these practices, we reply: well, that’s normal within their culture. We might not like certain cultural norms, but we can’t make moral assessments about people who participate in them, because the norms are (by definition) normative.

The next step is obvious, but less popular. When considering practices in Western cultures — slavery in the United States, for example — we can only conclude that an average level of participation was acceptable. I’m not talking about the cinematic overseers who whipped and raped slaves on a whim; presumably, they’re beyond the social baseline. I’m talking about a standard American who purchased cotton shirts and nodded when people said, “Africans are less intelligent, so slavery is good for them.”

Most people were of this sort. So unless you want to conclude that, until recently, the entire country was comprised of psychopaths, you cannot condemn them for going along with the times.

Of course, you see where I’m going here. Maybe you want to say that factory farming is not as bad as slavery, or the Holocaust, or other standard atrocities held up for questions about whether we would’ve done any better.

I think you’re wrong; I think factory farming is as bad, and probably worse, than anything humans have ever done to each other in the course of our bloody history. But if you dispute my claim, fine. You should probably read the aforementioned essay by , which documents exactly what happens to the animals we consume with gusto.

Even then, we might disagree on the exact moral awfulness of this industry. But how much less atrocious does factory farming have to be before the argument falls apart? If factory farming is morally equivalent to, say, .25 Holocausts: does materially supporting it really make me much better than an average, run-of-the-mill Nazi, who just as passively ceded to Auschwitz?

weren’t there better ones than I

So I think that eating animal products in 2025 is a good signal that you would’ve been a Nazi. Not Mengele himself, by any means — but also not part of any meaningful resistance. Just a regular member of the party, attending rallies and wearing armbands and avoiding Jewish-owned shops as the regime and social mores demanded.

Of course, this is kind of just a statistical fact. Most Germans were Nazis, not Righteous Gentiles, and the German genes (much to their chagrin) of 1939 are no different than yours or mine. So most living people would’ve been Nazis. Yet that claim, incorrectly and illogically, still allows the majority to think they would’ve been in the minority.

But if you’re like me — if you agree that factory farms contain horrors beyond our comprehension, yet still casually support it with your dime, protected by the veneer of the social baseline — then why on Earth should you think that, as a citizen of Salem or Mississippi or Nazi Germany, you in particular would have magically teleported beyond the moral average that you’re occupying right this very second?

In fact, our society makes the path to a just life (in this domain) infinitely easier than it was in Germany. To not be an average Nazi in the eyes of history, you had to overcome the possibility of imprisonment and execution. To not be an average supporter of factory farms, you have to overcome the possibility of less enjoyable food and an occasional awkward dinner. If you’re too scared to come up with a way to tell your in-laws why you’re not eating bacon at breakfast, then you definitely would have been too scared to hide Anne Frank in your attic.

Lest I come across as holier-than-thou: I am, by some standards, worse than most. I learned about factory farms several years ago, and for a while, I stopped eating meat. But then I decided I liked it too much, and started eating it again. What do you make of the Nazi who drops a few capsules of Zyklon B, decides to find a new job because he can’t take the screams, then rejoins a few months later because he couldn’t get a better payday?

Fortunately, I believe my own argument. I believe that I’m not a psychopath, and that I have not surrounded myself with a host of grinning monsters. I’m just a regular guy, like all the other regular guys who have shrugged at all the regular-guy awful things thrust upon them by their culture. It is simply too ridiculous to think otherwise.

And don’t worry, friend. If you’re like me, know that I don’t fault you for it either. How could I? Hypocrites must stick together.

History is often less kind to us, though. Someday, when the torture chambers have been emptied and the last broken bones have sunk beneath the surface of the soil, people will wonder, rhetorically: “Would I have eaten bacon on Sundays?”

Like us, they will applaud themselves on their enlightenment. They will curse the brutality of their peers. Their neighbor voted wrong. They’ll say he would have eaten turkey on Thanksgiving, just as some of us say that our political opponents, unlike us, would’ve raised their arms at Nuremberg.

But you and I will be listening from the depths of hell. We will know the truth.4

Note: I’m covering only the most salient point from my recent article — there are others that I also think are important! I would encourage you to read the article in full, though some will be followed up in future papers.

My advisor and I just published a commentary on Disbelief, a recent book by Dr. Will Gervais. It is open-access (as all articles should be); you can read it here.

Subscribe now

The book tries to explain why most humans believe in supernatural entities. Given that there is no apparent evidence for these entities, and given that our brains have evolved to make accurate inferences about what’s going on around us, this consistent inaccuracy is an important puzzle.

More adroitly, though, Gervais considers the modern phenomenon of atheism: many Westerners don’t seem to believe in those entities. This is another worthy puzzle, considering how common such beliefs are across all human cultures — what has happened in the West to trigger this shift?

Gervais impressively synthesizes several decades of research on the psychology of religious belief to develop his account. If you want a great survey of how scientists think about religion, I recommend reading the book!

However, our commentary is (respectfully) critical of this emerging consensus on the nature of religious beliefs. There are several reasons why, but I want to highlight one in particular. To understand the origins of supernatural belief, you need to consider the role of supernatural experience.1

i’m not religious BUT . . .

A central tenet of Disbelief, and the literature it pulls from, is that people adopt religious beliefs because they’re surrounded by other people who hold those beliefs. If you grow up with Christian parents, go to church, join Bible study groups and so forth, you end up believing in an all-powerful God. If you don’t do those things, you are unlikely to hold those beliefs.2

Okay — that all sounds reasonable. But the implication here is that people who grow up outside these communities or leave them entirely will usually become atheists. Without social influence, people are unlikely to hold religious beliefs. The supernatural is never a solitary affair.

This seems to be true for beliefs like the following: There is an all-powerful God who controls the fate of the universe.

But there are other kinds of supernatural beliefs that are less often discussed in the cognitive science of religion. For instance: The soul persists after death. Or: Plants, rocks and rivers are conscious. Do those also depend on social influence?

Apparently not. Here’s a figure we included in the commentary. It tracks rates of supernatural belief and practice across people who describe themselves, respectively, as ‘spiritual-but-not-religious’ (aka SBNR, in blue) ‘religious’ (in red), or ‘neither spiritual nor religious’ (in yellow).3

For three supernatural beliefs, SBNRs match or exceed religious people in their rates of endorsement. Yet SBNRS rarely believe in a Biblical god, pray, or involve themselves in either religious or spiritual communities.

What does this tell us? Well, it suggests that some supernatural beliefs are disconnected from social influence. For reasons other than peer pressure, people often believe in souls, spirits, and ancestor communication.

I heard all things in the heaven and in the earth

Why might that be? One strong possibility, I think, is that human minds — regardless of religious affiliation — are really prone to having supernatural experiences. By supernatural experiences, I mean experiences that seem to provide sensory evidence toward the existence of disembodied entities: for instance, auditory hallucinations or sleep paralysis.

More survey data is needed to get a firm sense of just how ubiquitous these experiences are, especially among SBNRs and so-called atheists. But there are some suggestive recent results from a survey by Ronald Fischer and colleagues. The figure below shows results from a massive survey of Brazilians, in which respondents indicated whether they’d experienced things like deja vu, hallucinatory voices and faces, a sudden sense of meaning, and so forth.4

Just to toss out a few numbers: 44% of respondents said they had seen a hallucinatory face. 80% had experienced deja vu. 42% had sensed a disembodied presence in their vicinity. And these are not responses from churchgoers. They’re from the general population of Brazil.

Maybe Brazil is an especially spiritual country. But if anything close to these numbers is generally true for humans, then we’ve vastly underestimated the role of supernatural experience in driving the growth of SBNR beliefs within Western nations. Social learning can’t explain why, despite ongoing secularization, SBNR beliefs have persisted and grown in these places. Baseline rates of supernatural experience, however, potentially can.

But the implications here extend well beyond one peculiar Western phenomenon. They pertain to the emergence and evolution of religion itself.

Consider a study published by Peoples et al. (2016), which uses cross-cultural data from foraging societies to reconstruct the cultural evolution of religious traits in early humans. Their analysis identified animism as the most basic feature of religion. After animism, beliefs in an afterlife tend to emerge, then shamanism, then ancestor worship.5 ‘High gods’ didn’t follow from any of these beliefs. They seemed to constitute parts of a different supernatural system.

What would we call someone who does not believe in a God, but does believe rivers are conscious and souls live on after death? In the United States, we would call them ‘SBNR.’

So most pre-agricultural humans would probably register on modern surveys as SBNRs. So-called ‘spiritual’ beliefs arose naturally from our evolved psychology. So-called ‘religious’ beliefs, on the other hand, could only emerge through particular forms of cultural evolution — perhaps the social structures that Gervais identifies as the scaffolds for maintaining membership in Western religions.6

The apparent alignment between SBNR beliefs and beliefs in foraging societies suggests a fascinating possibility; the emergence of religion itself in early humans may have depended on supernatural experiences.

All paradigms are mortal, social learning is a paradigm

This is not a very popular view. Most current evolutionary theories explain the emergence of a religion through its adaptive benefits — for instance, how beliefs in a punitive God can reinforce cooperation.

Yet there was a time before this one, when our understanding of supernatural belief was deeply entangled with the study of supernatural experience: see the work of one William James for proof.7

Gervais’ book is outstanding, and the literature he draws from is crucial. But I wonder if Disbelief will come to resemble something like B.F. Skinner’s Verbal Behavior: a proud declaration of scientific consensus, made on a shoreline from which the water is quickly retreating, while a circumspect few see the coming wave and rush to higher ground.8

So here’s my inexpert advice, if you want to contribute to the evolutionary study of religion anytime in the next twenty years. Don’t drown.

Running up to the launch of GPT-5, Sam Altman claimed that his company’s new model would display “PhD-level intelligence.” Lots of people, particularly PhD students, were annoyed by his declaration. They replied that there is no innate measure of intelligence; that getting a PhD requires not computational power, but hard work and curiosity; that if a system can’t even label a map, it doesn’t matter if it can write a grammatically-sound dissertation in fifteen minutes or less.

Not to mention that GPT-5 mostly failed to correct the hallucinations, basic fallacies, and other errors displayed by prior iterations. PhD-level intelligence? Start with fifth grade first!

Subscribe now

And so on. I don’t particularly care about the validity of “PhD-level intelligence” as a metric for the performance of large language models. Altman used the term as a marketing ploy, not as a quantifiable metric that GPT-5 had actually hit.

But I was struck by all the confusion over what, in principle, PhD-level intelligence could even mean: that is, what kinds of abilities are ostensibly cultivated during a doctoral degree that a layman does not possess.

Throughout the first year of my PhD program, I’ve wondered this quite a lot. What am I learning, exactly, for 8-10 hours a day? Maybe some new techniques — basic programming, statistics, whatever — but I could have learned all of that as an undergraduate. Am I gaining unique expertise? By the time I’m done, I will definitely know way more about certain scientific subfields than most people on Earth. But you can also cultivate expertise by working, say, as a labor lawyer or a chef.

Nor is any of this a matter of raw intelligence, such that a PhD simply acknowledges preexisting ability, like a more prestigious version of Mensa. I know lots of smart graduate students, but I can also think of several people smarter than me who chose career paths that I am probably not capable of following.

So if the difference between “Eli Elster” and “Dr. Eli Elster, PhD,” is not technical skill, or expertise, or intelligence . . . what is it? Theoretically, what is PhD-level intelligence supposed to mean — or in other words, what is supposed to be different about someone who has acquired a PhD, versus someone who has not?

Here’s my answer. A well-designed PhD program teaches you how to ask good questions. Trivial as this may sound, no ability is more central to scientific progress than asking good questions. And doing it is much, much harder than you think.

Why is that important?

Last October, I went for a walk with my advisor, Manvir, to talk about some ideas. I rambled for ten minutes about some potential study on the evolution of storytelling; honestly, it wasn’t worth remembering. Manvir listened quietly, nodding and staring at his flip-flops as we looped aimlessly around Davis.

When I finally petered off, he said: “Okay, okay. Word. But why is that important?”

You may have heard that there is no such thing as a stupid question. In principle, this is true, because unadulterated curiosity is beautiful. In practice — that is, if you’re trying to become a scientist — it is false. Rather than ‘stupid,’ though, it is more accurate (and kinder) to say that some questions are either not worth pursuing or not pursuable in the first place.

For instance, consider a question like: “Do more open-minded people prefer snowboarding over skiing?” You could totally answer that question! Just ask people to take the OCEAN personality test, then answer questions about their snow sport preferences. Cool. Now do various statistical tests to see if there are any correlations between open-mindedness and snowboarding, and voila, you have an answer to your scientific question.

Great! But why do we care about that result?

Intuitively, we don’t. The fact that open-mindedness is correlated with snowboarding over skiing, or vice versa, is a fun fact that you see on social media and forget about four seconds later. You could publish the result and no one would ever cite it or notice it or think about it ever again.

Here’s another question: why do humans make art? Unlike the first question, this seems worth pursuing — aesthetic practice is a strange and central feature of our existence, and everyone would like to know why we do it.

But there’s a problem: this question is not specific enough. It is hard, perhaps impossible, to come up with a way to systematically answer it in its current form. So this too is not a good scientific query.

From these two examples, we can (by way of omission) identify a few characteristics of good scientific questions:

Pursuability

• Can we identify a method capable of producing an answer that would seem to answer the question?

Generativity

• Does answering the question reveal new research directions, either by applying new methods or opening new gaps in the literature?

Fundamentality

• Does an answer to the question alter or improve our understanding of many other scientific domains?

The relationship between snowboarding/skiing and open-mindedness is pursuable, but not generative or fundamental — the answer doesn’t seem to matter for other fields, and it doesn’t suggest obvious next steps for research. Conversely, modeling the evolution of art is generative and fundamental, but not really pursuable per se — the answer doesn’t seem to follow from an actionable method, because the question is too big to handle.

So our two examples are both bad questions, but for very different reasons. One is too big, and one is too small.

We must instead try to pose Goldilocks questions. A Goldilocks question, of course, is one that is just the right size. It isn’t so big that you can’t answer it, but not so small that it isn’t worth answering in the first place.

Notably, once you’ve posed such a question correctly, the answer often falls right out of the formulation. Not literally — you’ll still need to collect the data, apply the methods, and so on. But that’s often just a matter of doing lots of grunt-level work: hence why research assistants are regularly hired to assist or take over the process.

As Einstein put it: “If I had an hour to solve a problem and my life depended on the solution, I would spend the first 55 minutes determining the proper question to ask… for once I know the proper question, I could solve the problem in less than five minutes.”

Two paths to finding the porridge

Coming up with Goldilocks questions might seem like a means to an end, rather than an end itself. It isn’t. My sense is that doctoral students often spend years developing the ability to ask these kinds of questions, if they manage to do it at all.

Sometimes they’ve gotten there before starting their program, which makes it easy for them to complete their dissertation without hiccups. Good for those people (he said, resentfully). But this seems to be rare. Usually, students must follow one of two paths before they can figure out how to ask the right kinds of questions.

Small → big → just right

The first path is small → big → just right. Consider a typical student who follows this path. Let’s call him Dennis.

Before starting his doctoral program, Dennis worked in a lab where he was tasked with carrying out a pretty niche task: say, coding responses to surveys about attitudes toward violence. Methodologically, he knows what’s up. But his role in the lab was oriented more toward assisting senior scientists rather than developing his own projects. He also never read very widely outside his field.

So upon arriving in graduate school, Dennis can come up with lots of small questions — he knows how to copy the methodological structure of projects he’s worked on before. He tells his advisor he wants to study the relationship between personality traits and hobbies. But then his advisor asks him: Hey Dennis, why are these questions important?

Dennis realizes he isn’t sure. At night, he starts to wonder whether he possesses the basic curiosity one needs to come up with interesting questions, or if he’s just a research robot unblinkingly following a track that he, in fact, doesn’t care all that much about.

If all goes well, though, Dennis will start to read widely. He’ll have lots of conversations with his mentors and his peers. Ultimately, he’ll identify the basic interests that inclined him toward his field in the first place. He never really cared about personality trait/hobby dynamics; what interests him, he realizes, is the way that personality traits change over our lifespan!

From that point, he can apply his methodological knowledge to narrow his well-identified interests into Goldilocks questions. For instance: can open-mindedness continue to shift as we age, or is it concretized after some point in development?

Big → small → just right

The second path is big → small → just right, embodied by a student named Kat.

As an undergraduate, Kat bounced between many different majors, read anything and everything, and collected an intriguing but disparate CV. Her research experience was limited. When she applied to graduate programs, her application contained several grand areas of interest and few concrete ideas.

Upon starting her program, however, Kat realizes that she doesn’t really know how to create projects within those areas of interest. She simply has the interests themselves. She likes books about storytelling, and music, and psychedelics1 — but she doesn’t really know how to make a publishable study about any of those things. Interests are not the same as questions. And without questions to answer, she will never be a scientist.

Despair! Kat wonders about her career choice. Maybe she should’ve been a journalist instead.

But she doesn’t quit. She starts to keep a big list of ideas, and starts reading papers that are not small (never!) but that explore her interests in ways that still strike her as novel and ambitious. Most of her ideas are bad. Over time, though, she learns to copy the way that her intellectual forebears have asked questions, and her ideas get better. She also talks to her peers — mostly people who are following the small → big → just right path2 — and uses their framings to guide her own.

She shares these ideas with her advisor, and her advisor agrees that most of them won’t really work. But that’s okay, because a few of them definitely will. She hears a Hemingway quote one day: I write 99 pages of shit for every one page of gold.

Teach the children well

These are both legitimate paths to learning how to ask Goldilocks questions. The truth, though, is that many PhD students probably never figure it out.

In some cases, that might be the fault of their work ethic or intelligence. But usually, it’s because their advisor doesn’t understand what it is that their students are supposed to be learning. They push their advisees to pursue projects they didn’t come up with on their own, or suggest that they focus on teaching, or only make time to discuss action items. In short, they emphasize everything but the cultivation of Goldilocks questions.

This is bad practice, and it has predictably bad results. A big → small → just right student can start to feel like they’ll never focus enough to publish a paper. So they quit their program, or switch to asking small questions out of desperation. A small → big → just right student can stick with it more easily by publishing on all manner of tiny and useless questions; but that means they too cannot become a worthwhile scientist, because no one is teaching them how to be one.

I suspect that I’ve gotten lucky — my advisor is uniquely focused on helping me figure out how to ask Goldilocks questions, and I think I have benefited tremendously from it. 3

But his focus seems to be exceedingly rare. Mentors tend to think of themselves as technical advisors, or administrators, or CEOs, or anything other than intellectual models for their students. And their students suffer as a result, because even with the trappings of productive mentorship — meetings, answered emails, edits on papers — they are not learning the only skill that plausibly differentiates PhD-holders from everyone else.

Learning how to ask Goldilocks questions has always been important. But I should note, too, that the advent of generative AI is making it more important than ever. GPT-5 doesn’t possess PhD-level intelligence. Not because it hallucinates or can’t label a map; because it can’t ask a good question to save its nonexistent life.

But it can do statistics, and program, and make figures, and review literature, and do basically everything else that a PhD program ostensibly teaches you how to do. And five years from now, it will do all of those things better than me or any other scientist in history.

So a graduate student in the modern age must cultivate the one ability that has always distinguished scientists from everyone else, and now distinguishes scientists from GPT-5. They must learn how to ask Goldilocks questions.

To all the advisors out there, then — you must teach them. Take your students for a walk.

Note: This is a post I sent to my friends and family while conducting fieldwork in Lesotho. Due to popular demand from them, I’m publishing it, with a few edits from when I originally wrote it. I look forward to addressing the concerns of my future employers.

Okay, let’s be clear: did I want to take seipone through an enema? No, I did not. I did not want to use an enema to take a hallucinogenic drug1. But I am an anthropologist, and I make sacrifices for my research.

Subscribe now

Let’s start from the beginning. Seipone — also known as leshoma, or Boophone disticha — is a plant used ubiquitously across Southern Africa as, among other purposes, a tool for divination. People ingest it and stare into a mirror, dark wall, or cloth, and visions induced by the plant indicate the sources of their misfortune: usually, someone who has bewitched them. A healer then interprets the visions to help them parse out an interpretation.

I thought it would be interesting to try the ritual myself. Betsy (my main collaborator, and a Sotho traditional healer) and I had been interviewing healers and their patients about their experiences with seipone. Perhaps doing it would help me better understand their experiences, since it’s famously hard to describe a drug-induced state to someone who’s never done that drug. So I asked Betsy to find someone trustworthy who could administer the ritual to me.

She replied about two hours later to let me know that we had an appointment set. On 7/19, we drove to Roma (not that one, I’m in Lesotho), where the healer was based. On the way, though, we picked up a woman named Gogo and her son, Robby (names changed for anonymity). Robby was 22, did not seem to have yet shaved his mid-pubescent mustache, and had just accidentally fathered a child. His mother was not pleased by this. She wanted him to do the seipone ritual to figure out why, exactly, the spirits had placed this burden upon him.

Robby did not talk very much on the way to Roma. I think he was playing Clash of Clans on his phone.

A note on my mental state, at this point. Betsy and I had spent the entire previous night at a ceremony in the hills of Mafateng — which required some insane offroading to reach — and the ceremony lasted until sunrise. There was prayer, dancing, singing, goat meat (though no sacrifice), crowds of children clustering around me, a car ride to get more beer at a local shop where the manager, perhaps thinking that he was not speaking aloud, said “Ah, a white man means money.” I’ll write about it another time. The takeaway is that I was completely exhausted, had not eaten for 20 hours (one must fast before seipone), and was primarily functioning off a strange South African energy drink. An optimal approach to meeting my ancestors in a wall, I think.

After a long and bumpy ride, we drove up a terraced hill and parked in front of a series of three clay huts, designed in the usual Basotho way — like straw farmer hats placed on top of very large, cylindrical heads. The healer was leading a group of four people into one of the huts, presumably to administer seipone. We sat down on a row of wooden benches to wait. Occasionally, children would emerge from the nearby shrubs carrying piles of firewood. Cows and donkeys grazed in nearby yards, while dogs and a set of gray kittens wandered the roads in search of food.

I had not yet said more than a few words to Robby. So I asked him how he was feeling. This was a mistake — Robby was feeling good, because his last experience with seipone had been absolutely awful, which meant that this one had to be better. His father had (for unclear reasons) prepared a dose for Robby when he was about 17, then left him alone with his nanny. Robby drank the concoction and walked outside to look at the sky. He saw a dragon. Then he was unconscious for three days. His nanny later told him that he had spent an entire weekend in his room, intermittently groaning, yelling, and sleeping. Robby remembers none of these events.

When you’re about to try a psychedelic, you do not want to hear that someone has been pushed into a fugue state by that same psychedelic. It was good to hear that the dose had apparently been far too large and prepared by someone with literally no experience — a bit like asking your accountant to take his best shot at giving you morphine through an IV. Still, I spent the next hour wondering if I was making a terrible mistake, walking around the yard and intermittently staring at a large cross that had been built on a nearby boulder. But I was reassured upon seeing the original four patients emerging from the healer’s hut, evidently unscathed by the drug.

I was less reassured by Betsy: she was speaking to the healer, and her face suggested a mix of confusion and amusement. She walked over to me and said: “So, it’s only 180 rand.”

“Oh, great!” I said. “That’s fantastic.”

“Yes,” she said. “But the standard procedure is an enema.”

“What?”

“Yes.”

I was not sure what to say. Down the hill, Robby looked like he was about to vomit.

“He can’t let us drink it, like literally everyone else we’ve met?”

“I know, I know,” she said. “They’d have to brew it from scratch. We’d be here all day. Try anything once, yeah? Brave anthropologist?”

She was kidding, but a set of images crossed my mind — personal heroes, placing themselves in odd and perhaps humiliating positions to better explore their subjects. Richard Schultes, the legendary ethnobotanist, spent an entire day rolling around in an Amazonian hut to understand the effects of datura. My own advisor, Manvir, had imbibed yopo in Colombia and promptly vomited all over himself, his best friend and his wife. Who was I to refuse a minor medical procedure that would both circumvent the usual nausea that comes from drinking seipone and make for a much funnier story?

“Okay, okay,” I said. “Enema it is.”

“Have fun!” Betsy said. I handed her the keys to our car — she and Gogo were going to get hamburgers and sit by the river. Meanwhile, Robby and I followed the healer — carrying a small red tube, one end of which gleamed with Vaseline — into his hut.

“I think this is a bonding experience, Robby,” I said to the soon-to-be first person to watch me boof a plant.

“Bro, I so did not sign up for this,” he said.

The healer’s hut was about fifteen feet across. The walls were painted with light-blue stripes and polka dots, and several bags of animal feed littered one side of the room. The other side was dominated by a blue cloth hanging from the ceiling. In front of it sat two woven mats, and a bench; to either side were a series of multi-colored cloths, as well as a table littered with what appeared to be monkey hands.

The healer did not speak English, so Robby translated his instructions. We would stare at the cloth — nowhere else, or we would not learn what we needed to learn — and our ancestors would appear. We could then direct the ancestors to show us visions that would answer our questions. If we became overwhelmed, we could simply ask not to see what we were currently seeing, and the ancestors would politely change the channel, so to speak.

Then he held up the enema and bent over one of the benches. Robby translated: “He said we should put it in, like, three inches. But make sure it’s centered.” The healer spoke again. “And you have to bend over like him.”

“What does that mean,” I asked. “Centered?”

More Sotho from the healer. Robby: “Like, in the middle. I don’t really know, man. Fuck.”

At this point, I had completely ceded to the experience, and volunteered to go first. The healer smiled — a little too gleefully, I thought — and handed me the enema. Several steps followed from this point, such that I was no longer holding the enema in my hands. I’ll spare those particular details.

But not these, because they’re too funny to leave out. Apparently, I am bad at inserting enemas: the liquid was not flowing. The healer spoke. Robby: “He says go back and center it.”

“Meaning, out?”

“I guess, yeah.” Robby was facing the opposite wall and looking at his feet.

I adjusted. The healer spoke. Robby: “He said go back and center, again.”

I adjusted. The healer spoke. Robby translated. I adjusted. The healer spoke. Robby translated. My legs were beginning to shake from holding the sacred position.

Finally, something clicked, etcetera. Operation completed, I placed the enema in a bowl of soapy water. “I promise it’s not that bad, Robby,” I said. Robby was looking exceptionally pale. “It’s like being slightly constipated.”

“Bro. Are you sure?”

I left the hut so Robby could take his turn. I looked at some flowers and slowed my breath; I focused on my intention with doing this ritual, and not the fact that I had just mainlined a vaguely poisonous liquid into my rectum. As instructed, I waited for a few minutes before going to the toilet. Upon entrance, though, I realized I had no idea whether I was supposed to — let’s say “void — just a little bit of the liquid, some of it, most or all of it. I took my best guess and contributed to the pit toilet that had been constructed specifically for people with seipone in their intestines; the healer had a separate space for the usual suspects. Robby emerged from the hut at the same time that I emerged from the toilet, by which point the familiar buzzing and far-too-close sounds of birds had begun to set in.

I reentered the hut, followed closely by Robby, and we sat down on the bench to wait. Before he left, the healer suggested we pray, and handed us what appeared to be an anti-emetic plant as well as a bucket for vomit. I don’t really know how to pray on cue. Robby, though, began to whisper to himself in Sotho, so I looked into the cloth and stated my question to the ancestors.

With the door shut, the room was very dark. Shadows roamed over the cloth, forming slight eccentricities in the light that could, even to the sober eye, transform into complex shapes. Our eyes were not sober. The shadows stretched and congealed into parallel lines, then a tangled mess of branches and leaves. A circle swelled on one branch, bearing arms and legs and needles on its back. The silhouette of a porcupine — it jittered darkly through the canopy, searching for something unknown, before settling down and shifting into a man in a woven hammock, swaying gently and watching the sky.

An ancestor, perhaps? I had no idea what Robby was seeing. I again whispered my question. The man gasped and tumbled from the branch, and the whole scene fell to pieces. Curious about how else my perception had been affected, I closed my eyes and saw nothing save a bright, distant beam of light.

Returning to the cloth, a cityscape was slowly blooming. Each time I moved my eyes, though, the hallucination would break; and if I held my eyes steady, a blur would descend over the world, leading me to what felt like the edge of sleep. Entities began to emerge at speed: faces contorting as if passed through a thousand funhouse mirrors, placed upon bodies regaled in Hindu vestments. They leered at me. They smiled, and they seemed to have sharp teeth.

Then they stepped out of the cloth, and before me stood a shimmering series of gods, ancestors, demons — I don’t know what to call them, really. There was no sense of meaning or joy to their presence; to those who have tried classic psychedelics, I can relay no shared feeling of the world’s unknowable beauty. Instead, the world felt exceedingly alive, every slight nuance in vision emerging into alien being; but these beings stalked me like novel prey, as if trying to discover a weakness. Somewhat afraid, now, I stated my question more loudly. The entities laughed and displayed a blank tab from Google Chrome: for the first time, a vision in complete color.

From here, the experience disintegrated. I remember very little from the next hour, save occasional glimpses of bodies contorting out of the cloth — or rather, from what had been the cloth and now seemed to be more like a doorway. My most salient memory is of the table strewn with gibbon hands. In the periphery of my vision, it merged into a Hanuman-like royal figure, comprised of spare parts from some poor clan of monkeys. It reached for me repeatedly; I could banish it only by turning my gaze directly toward it, at which point it would revert to the standard form. At some point, Betsy and Gogo returned to ask how it was going. Robby and I did not reply.

Was Robby enjoying himself? Was he finding the experience meaningful? I have no idea.

Eventually, I regained my memory (apparently) and turned away from the cloth. Robby was standing there with his hands in his pockets, as though he had not been tripping at all. From what I recall, he seemed to be trying to strike up a genuine conversation.

“So, what are you hoping to learn from the healers you’re talking to?”

“Robby,” I replied. “There is a pig-face man lying behind you.”

We — or just I — must have come down enough by then to be ushered from the hut and into another room. I am told that the healer prayed, but did not have time for a consultation. So Gogo drove us home. Out the window, I watched full-fledged people rupture from the earth, then melt away into air. The visions had ended by the time we had arrived at my guest house, fortunately. I called my girlfriend and went to sleep.

So went my experience with seipone. Was it enjoyable? No: in fact, it was deeply unpleasant. Did I learn anything about myself? No: I can attest to no personal revelations, no renewed love for the world and all the good people in it, nothing that has endeared me to psychedelics in the past. Would I do it a second time? Probably not: and I understand why very few of our informants report having tried it more than once.

Ultimately, though — and strange as it sounds — this was participant observation, conducted for my research. I now understand what people mean when they discuss their visions, the strangeness and paranoia beneath them, and the necessity for external interpretation. Like most divination methods, seipone is a Rorschach test: an ambiguous set of data lies before a trained individual who, whether through true intuition or charlatanism or something in between, can provide some order to the chaos to a patient in need of that structure.

But unlike tarot or the I Ching, the data falls from the skull of the person searching for meaning within it: the ink blot splatters from the patient’s open mouth. It is perhaps especially compelling to many, many people across Southern Africa — despite being so harsh on the mind — because the difficulty of the experience makes one feel as though the visions are wrenched from an entirely separate world.

Also, healers give me props now when I tell them I tried seipone. And props are the beating heart of anthropological fieldwork. For props, I will apparently do just about anything.2

Note: I’ve realized that I have lots of ideas that seem fun to write about, but that probably don’t warrant full essays. Hence, “Eli’s Extras.” These pieces will emerge sporadically, often, or pretty much never. Hard to say: and if I knew, that would take the fun out of it anyway!

Usually, comparisons between up-and-comers in a profession and the legends that preceded them are unearned and inaccurate. Occasionally, they’re earned, but still mostly inaccurate. Only rarely is the comparison both earned and accurate. Such is the case for Jesse Welles, a brilliant singer-songwriter from Arkansas who is, to my mind, already one of the great lyricists of the past few decades.

Subscribe now

You may have seen Jesse on social media, standing in a cornfield with his guitar and singing about Palestine or United Health. His popularity has skyrocketed because of these biting protest songs, some of which he releases within hours of a major world event. They are very good. They are also the obvious reason that people keep calling him ‘Dylanesque.’ He’s a young poet with a guitar singing folk music about politics. Just like Dylan, right?

I mean . . . kind of? But there have been lots of young poets with guitars who sang folk music about politics. Pete Seeger, Joan Baez, Peter Paul and Mary, Neil Young, Billy Bragg. The list goes on. Sure, Dylan is arguably more famous than all of them put together. Yet it isn’t just casuals from Reddit giving Jesse the label of Dylanesque. Music critics and reviewers apply it incessantly, including to his songs and albums that have few or no political overtones.

No, ‘Dylanesque’ doesn’t just mean ‘political.’ No one meaningfully compares Dylan and Rage Against the Machine. There’s a literary quality at work. Just listen to a song like “Middle.” If you’ve listened to much Dylan, you’ll hear it — but what exactly is ‘it’?

I love Bob Dylan, and I love Jesse Welles, and beyond both, I love thinking about the nature of ‘esques’ — what it is that distinguishes style in the work of someone like Joan Didion, Jorge Luis Borges, Franz Kafka, whoever. So: let’s talk about being Dylanesque.

The joker and the thief

Consider the following excerpts from two seminal lyricists, united in genius but entirely different in style. Excerpt #1:

Yesterday
All my troubles seemed so far away
Now, it looks as though they’re here to stay
Oh, I believe in yesterday 

This is the first verse of “Yesterday,” by Paul McCartney, widely considered one of the greatest songs of all time.

Now, Excerpt #2:

“There must be some way out of here”
Said the joker to the thief
“There’s too much confusion
I can’t get no relief
Businessmen, they drink my wine
Plowmen dig my earth
None of them along the line
Know what any of it is worth.

This is the first verse of “All Along the Watchtower,” by Bob Dylan, though the cover by Jimi Hendrix is better known (and just, you know, better).

Both songs are brilliant. For what it’s worth, Rolling Stone usually ranks them in their Top 50, with “Yesterday” slightly higher. They’re also, I think, both decent representations of the (admittedly diverse) catalogues of both McCartney and Dylan — or at least, of the kinds of lyrics people have in mind when they say ‘McCartneyesque’ or ‘Dylanesque.’

However, these verses are very, very different. There are, of course, lots of interesting comparisons one could make between McCartney and Dylan. But I want to focus on a big one. The meaning of “Yesterday” is abundantly clear. “All Along the Watchtower,” not so much. Over the years, I have listened to it hundreds of times; when a friend recently asked me what it was about, I realized I literally had no idea.

McCartney is supremely good at capturing a very complex feeling or situation — for “Yesterday,” lost love and regret over the mistakes that led to it — in a searingly direct and coherent set of lyrics. The meaning of his words is viscerally clear; it is a mainline to your heart. The song makes you think about heartbreak, and you know that’s exactly what McCartney had in mind.

Dylan, by contrast, is opaque. You hear him sing, and it works, and you can tell he’s really good at writing — but you mostly have no idea what exactly he’s talking about. He conjures an almost archetypal world. His lyrics aren’t surreal, per se. Instead, their meaning seems to sit just beyond the edge of your understanding; you feel as if something is there, waiting to be apprehended, and so you can’t help reaching for it.

But the meaning of a song like “All Along the Watchtower” is a horizon line. You can walk toward it as long as you want, and you may find intense joy in the movement — but you’ll never reach the end, because ultimately, there is no end to reach.

Now, back to Welles. What makes him Dylanesque? Exactly the same thing that made Dylan ‘Dylanesque.’ His lyrics gesture at some grand intent, yet try as you might, you can never quite settle upon it.

For example, here are the first lines from his song “Rocket Man”:

I heard the altar call of the rocket man
As he pleaded on the mount
Then were we like those who dreamed
And the tired tiger jumped through hoops of fire
One last time for the crowd
Then were we like those who dream

You can see the image he’s building; you feel as though it’s laden with purpose, as if you’re listening to a sermon rather than a song. But what is that purpose? What is the nature of the sermon? You can’t really tell.

A couple more examples. Here’s from “Middle”:

The road warrior Nazareth patriot
Sighing Zion in excess
While the Lord and God of faces
Donned a youthful chain and dress

And another from “Saint Steve Irwin”:

Sail me out of this hell, Virgil
They poisoned the well
And I can't drink here now
Saint Steve Irwin, I'm a child
And I'm certain them are wild crocodilеs
Won't you choke them out?
Everything I'd еver seen
Was piled up into a dream
I was afraid I would forget
Now I remember everything

Like Dylan, Welles’ lyrics are full of superattractors for the interpretive mind: mythical references, Biblical imagery, poetic forms and the occasional line that seems to provide a burst of McCartneyesque meaning.

This latter device is especially interesting to me. In “Rocket Man,” the chorus includes more transparent tattoo-worthy aphorisms, buried within the opacity: “Time is just a thing you live through, baby.”

Dylan would respect the game. “My Back Pages,” for instance, is virtually all obscure. And then, from the mist: “I was so much older then / I’m younger than that now.”

Sighing Zion in excess

To be clear, I’m not saying that Jesse Welles is bullshitting you. Dylan might’ve been, but I don’t think Welles has anything like the same arrogance and self-grandiosity. I should also note that much of Welles’ catalogue is more direct, especially his political work, and that his clear influences extend well beyond Dylan: John Prine, Ty Segall, and Willie Nelson are a few obvious candidates.

Nonetheless, Welles is employing many of the same lyrical techniques that define ‘Dylanesque’ songwriting — the same signposts of infinite interpretability that made Dylan, but not McCartney, a candidate for the Nobel Prize in Literature.

Will Welles go that far? Probably not. But, who knows? The answer is blowin’ in the wind.