Minimalism vs Maximalism

There’s a thing you can do, the next time you walk into a building or sit down for a movie or open a website, where you can know almost everything about what’s coming in the first three seconds. Not by being smart. By looking at the room.

The opening of Everything Everywhere All At Once spends about ninety seconds in the family’s apartment. There’s a Christmas tree (it’s January). There’s a treadmill being used as a coat rack. There’s an actual pile of receipts on top of a bonsai tree on top of a stack of laundry on top of a half-disassembled karaoke machine. There are I think four separate Buddhist altars, none of them in conversation with each other. The walls are covered. The counters are covered. The refrigerator door has so much paper taped to it that you can’t see the refrigerator.

You know the movie before it starts. You know it’s going to throw nine genres at the wall and let them fight. You know the feelings are going to be enormous and contradictory and that nothing is going to be cleanly resolved. You know this because the apartment knows it. The set design is the thesis statement.

Now the opposite test. Walk into an Apple Store. Or sit down in a black-box theater for an experimental one-actor play. You know, immediately, that whatever’s about to happen is going to do exactly one thing, and that one thing will be polished within an inch of its life, and the entire production will be organized around making sure nothing else competes for your attention. The Apple Store is going to sell you a phone (it is not going to sell you a phone case in a way that distracts from the phone). The play is going to do emotional intensity (it is not going to do plot, set, costume, music, or other characters). You have not been told this. You have inferred it from the room.

This is the binary. Maximalist works tell you they’re maximalist before they say anything. Minimalist works tell you they’re minimalist before they do anything. And once you can read the signal, you can predict roughly 70% of what the work is going to feel like before you’ve read or watched a single line. (I hate making up percentages. I’m going to do it here anyway. The number is “most.”)

I’m flagging this because most discussion of this binary acts like it’s a question of taste (do you like clean lines or do you like clutter, do you prefer Hemingway or do you prefer the guy who writes the 900-page novel about whales). And sure, you have a preference. Fine. But the more useful frame is diagnostic. You can read a movie’s set, a writer’s first sentence, a band’s stage setup, an apartment, a website, a dating profile, and instantly know what kind of thing it’s trying to be. The aesthetic is doing diagnostic work whether the artist meant it to or not. It’s a public commitment to a strategy, posted on the door before anyone walks in.

The two strategies. Briefly, because I trust you. Maximalism is what you do when you want a thing that gets richer the longer you spend with it; the room or the album or the novel keeps revealing new connections, and the density is the experience. Minimalism is what you do when you want a thing that delivers one perfect signal; the room or the album or the novel removes everything that would compete with that signal, and the clarity is the experience.

(Quick aside on a thing that annoys me: a lot of contemporary discourse treats minimalism as morally superior, like the people who own seven cream-colored objects have transcended consumerism while the rest of us are gluttons. This is silly and we’re going to get to it. Maximalism, in turn, has its own self-flattering version: the “well-traveled cultured eclectic” who has decorated their apartment with carefully sourced poufs from carefully sourced markets. Also silly.)

What I want from you, by the end of the post, is to walk into the next room you walk into and immediately know which one it is, and then know what that means about what’s coming. You should be able to play this game with novels, with movies, with relationships, with companies, with code, with cities. The reason it works is that the choice of strategy is upstream of almost everything else.

Like a saw and a hammer. You’re not picking. You’re noticing.

Two video games, released the same era both massive, both expensive, both AAA prestige open-world products. Elden Ring and Horizon Zero Dawn. Roughly the same amount of stuff (hundreds of weapons, dozens of regions, encyclopedic lore). By raw asset count they’re twins.

Now play them for ten minutes.

Horizon Zero Dawn loads your screen. Quest markers float over NPCs’ heads. The map is dotted with icons. There’s a quest log on the right, a skill tree with about fifty nodes branching in four directions, a crafting menu with subcategories, a dialogue wheel where the options are color-coded by tone, an objective tracker telling you what to do next, and codex entries that pop in automatically whenever you walk near a thing the game wants to explain. You are never alone. The game is with you, talking, organizing, helpfully intervening. The premise is: I have built you a feast, and I want to make sure you know which forks to use.

Elden Ring puts you in a field. There’s a sword. There’s a horse, eventually. There is no quest log. There is no map marker telling you where to go. NPCs say cryptic things and then leave. Major story characters will die offscreen if you don’t find them in time. The game contains mechanics it never explains. The HUD is almost empty. The first hour leaves a lot of players feeling lost in a way other games of this size would never permit.

The thing that took me a while to realize about all this. Elden Ring is not less complex than Horizon. It almost certainly contains more lore, more secrets, more interlocking systems. The world is denser, not thinner. What Elden Ring is doing is something subtler than reducing content. It is reducing what it surfaces. The game’s minimalism is the empty HUD, the absent quest log, the silence where another game would put a tutorial. The world behind that surface is enormous. The choice is about how much of it gets shoved in your face at any one time.

This is the move I think most takes on this binary miss. Both strategies are about how much of the underlying material gets exposed to you at any given moment, not how much of it exists. The amount of stuff under the hood is a separate question. A minimalist room can be backed by a 10,000-volume library that’s filed two doors down. A maximalist room can be assembled from objects each of which is, individually, ordinary. What’s at stake in the binary is the surface, not the inventory.

Once you see the surfacing distinction, the game changes. You stop noticing how much exists in a piece of work and start noticing how much it’s willing to interrupt you with. A novel that puts every digression in footnotes vs. a novel that braids them into the main text. A coffee shop with one espresso machine in front and the storage room hidden vs. one with the entire roastery operation visible from the counter. A wedding where the dress is the event vs. one where the dress, the hall, the band, the food, and the great-aunt are all events at once. Same stuff, sometimes. Different theory of how to deliver it.

Now, the part I find more interesting than the diagnosis. Each strategy can do a thing the other one can’t, structurally, no matter how skillfully executed.

What minimalism gives you: the dignity of being able to actually look at the one thing. When the gallery has only the one painting, you stand in front of the painting. You see it. You start to notice things about it that you wouldn’t have noticed if it were one of forty paintings, because the wall around it is empty and your attention has nowhere else to land. The thing is allowed to be itself. This is a real gift. Maximalism cannot give you this gift. The moment you put a second thing in the room, the first thing has lost a small amount of its sovereignty. There is no maximalist room in which one object is the event. The form forbids it.

What maximalism gives you: the surprise of accidental connection. You walk past the bookshelf for the eightieth time and suddenly notice that the small ceramic owl is in dialogue with the cover of the orange book three shelves up, which is in dialogue with the rug. These connections weren’t planned by anyone. They emerged because there were enough things in the room to form connections, and you were in the room long enough for your brain to find them. Minimalism cannot give you this surprise. There aren’t enough relationships in the system. The room is exactly what it was on day one, and on day one thousand. That’s not a flaw of minimalism, just the price of admission. Minimalism is finished in a way maximalism never is.

This is where most “you can do both” / “they’re just tools” framings fall apart. They’re not just tools. They’re tools that purchase one quality at the cost of another.

You cannot have the hush of the empty room and the depth of the full one at the same time. You cannot have both the dignity of the singular focus and the richness of the layered field. You can have one or the other, executed brilliantly. The choice is what makes the work be a particular thing. And once it is that thing, the other thing is permanently unavailable to it.

Which is why both sides tend to feel a little contemptuous of the other, in private. The minimalist looks at a maximalist room and sees noise; what the minimalist is correctly noticing is that the maximalist has given up the gift of focus, traded it for something the minimalist doesn’t value as much. The maximalist looks at a minimalist room and sees deprivation; what the maximalist is correctly noticing is that the minimalist has given up the gift of accumulation, traded it for something the maximalist doesn’t value as much. Each one is right about what the other one has lost. Each one is wrong about whether the trade was worth it, because the answer to that depends on which gift you wanted in the first place.

This generalizes. The question minimalism is asking, in any medium, is: what should we surface to the user right now? The answer is “as little as possible, and what we surface should be excellent.” The question maximalism is asking is: what’s the richest field we can drop the user into? The answer is “as rich as we can make coherent.”

This is why the standard “minimalists own less stuff” framing misses the point. There are minimalist apartments full of books (the books are the focal point, the rest of the room steps back). There are maximalist apartments with very few objects, but every object is loud (a loft with three enormous canvases, a velvet couch, a pink lamp, all screaming at each other). The amount of stuff is downstream of the question. The question is what wants to be in the foreground.

You can see the same fight in writing. Hemingway and Pynchon are not arguing about how many words a sentence should have. They are arguing about how much the prose should be doing at any given moment.

A Hemingway sentence carries a single emotion delivered cleanly. A Pynchon sentence carries a song, a joke, a paranoid theory, a description of a piece of equipment, a callback to something forty pages ago, and a footnote on V-2 rockets. Pynchon’s sentences are not worse than Hemingway’s, and they are not better. They are answering a different question about what reading is for. Hemingway thinks reading is concentration. Pynchon thinks reading is immersion. Both have produced books that change people’s lives.

I want to be honest about a place where this binary gets politically charged, because it does, and pretending otherwise is silly. There is a strain of minimalism that comes loaded with moral self-congratulation. The Marie Kondo / capsule wardrobe / “I own seven things and they’re all cream-colored” version of minimalism wants you to know that the people doing it are more virtuous than the people who own a lot of stuff. They have transcended consumerism. They are spiritually advanced. The maximalists, by implication, are gluttons who can’t control themselves.

There’s also, and you can confirm this with a half-hour of looking at people’s apartments, a class correlation. American WASPs and the professional-managerial class lean minimalist. The aspirational aesthetic is the white-walled loft, the single beautiful chair, the kitchen counter with one bowl of lemons and nothing else. Immigrant families and the working class lean maximalist. The aspirational aesthetic is the full living room, the wall of family photos, the china cabinet, the doilies, the saints in the corner, the spice rack with thirty jars on it.

This is not a universal. There are minimalist Italian grandmothers and there are software engineers whose studio apartments look like an antique shop crashed into a record store. But on the whole the correlation is robust enough that you can guess wrong about someone’s politics from their apartment more often than you can guess wrong about their income.

This is worth flagging because the “minimalism is morally superior” framing isn’t only a wellness culture artifact; it’s a class costume too. The cream-colored apartments are expensive, in a way that is partly about money and partly about the cultural capital of having internalized that emptiness reads as taste.

It’s much easier to own seven things if owning enough things has never been a problem you had to solve. Maximalism, in its honest form, is often the aesthetic of people who actually had to acquire the stuff, and who like having it around because it represents something they earned.

This framing is not an inevitable feature of the binary. It’s a particular wellness-industrial-complex coding that happened to attach itself to minimalism in the 2010s, and you should resist it, because it stops you from seeing what minimalism is actually good at. Minimalism is good at focus. Minimalism is good at giving an object the dignity of being looked at. Minimalism is good at quiet. None of those goods require you to feel morally superior to the person across the hall whose apartment looks like a flea market exploded.

Maximalism, for its part, has its own coded version that’s just as silly. The “eclectic personality” maximalism, where the point of the room is to demonstrate how interesting and well-traveled the owner is. (You walk into the room and it’s not warm, exactly, it’s performing depth. The Moroccan poufs are there to be the Moroccan poufs.) That’s not really maximalism either. That’s identity display dressed up in maximalist clothing.

Real maximalism, the good kind, is about generosity. The room is full because the room has things to give you, and the more you give it your attention the more it gives back. A really well-made maximalist room has a quality you almost never see in a minimalist room: it gets better over years. You start noticing connections. You realize that the small painting in the corner is in dialogue with the rug, which is in dialogue with the lamp, and these dialogues weren’t there a month ago, or were there but you couldn’t see them. The room is teaching you to see it. Whereas a great minimalist room teaches you to see the one thing it’s pointing at, and once you’ve seen it, you’ve seen it.

I think this is why people who deeply love one of these aesthetics tend to be impatient with the other. The minimalist looks at a maximalist room and sees noise that’s preventing them from seeing anything clearly. The maximalist looks at a minimalist room and sees a refusal to commit to anything, a kind of design-shaped agoraphobia. Each one is correctly identifying what the other approach gives up. Each one is failing to see what the other approach gives back in return.

The honest version of this binary is: minimalism gives you clarity at the cost of richness. Maximalism gives you richness at the cost of clarity. Pick which one your situation needs.

A funeral wants minimalism. The grief is the only thing that should be in the room, and anything else is competing with the grief. A wedding wants maximalism. The point of a wedding is the layering, the family histories meeting, the food and the music and the toasts and the strange uncle, and a minimalist wedding is a wedding that has misunderstood what weddings are for. A bedroom wants minimalism (probably). A bookstore wants maximalism. A first sentence wants minimalism. A novel wants whichever one the novel is about.

The mistake people make is treating this as a personality test, as if you are fundamentally one of these things and have to commit. The people who are best at design, in any field, can switch. They know what each tool does. They can tell when a room needs to be quieted and when a room needs to be filled, and they don’t moralize about either move, because the move is in service of the work, not the work in service of the move.

---

Album sequencing in pop vs. jazz. A pop album is sequenced minimally: track one is the single, track two is the other single, the deep cuts are placed where they won’t lose the casual listener. Each song is foregrounded individually. A great jazz album (think Miles Davis’s Bitches Brew, or anything by Sun Ra) is sequenced maximally: the tracks blur into each other, instruments enter and leave without announcement, and you’re not supposed to be tracking individual songs at all, you’re supposed to be inside a weather system. Both are legitimate ways to make a record. They want different things from your ears.

---

There is a story Borges tells about a map that grew to be the size of the empire it was describing, and was found tattered in the desert by later generations who had no use for it. The map and the empire shared a fate, because they had become each other. This is the maximalist anxiety: that the field will grow until the field IS the world, and the distinction between the description and the thing will collapse, and what you have left is just everything, undifferentiated.

The minimalist anxiety is the opposite. It is the anxiety of the monk in the white room, who has subtracted everything and finds that what remains is not a clarity but an absence, that the chair he kept is also a thing he could have kept differently, that even one object is too many to be honest, that the only true minimalism would be the empty room, and the empty room is not a room.

Both anxieties are real. Both are old.

In the medieval Cistercian monasteries the monks built churches of stripped white stone and banned ornament, banned color, banned the carved monkeys and centaurs and beasts that crawled across the columns of richer abbeys, because they believed that the elaboration of the world was a veil, and to see God you had to lift the veil. And in the same century, a few hundred miles away, the builders of Chartres were covering portal and façade with prophets and beasts and the geometry of the cosmos, because they believed that the elaboration of the world WAS God, that to honor the divine was to mirror its inexhaustibility in stone. Both groups were Christians. Both were sincere. Neither would have understood the other. The cathedral and the monastery were not different stages of the same thought. They were different thoughts at the same time, both reaching for the same thing through opposite hands.

You can find this everywhere if you look. The Pythagoreans, who said reality was number, the smallest possible irreducible relations, and could be apprehended best by clearing the mind to a single tone. And the Hermeticists, who said reality was a web of correspondences, the moon answering the silver, the silver answering the brain, the brain answering the lily, and could be apprehended only by holding the whole net at once. Both schools were doing metaphysics. Both produced people who claimed to have seen the structure of the world. The structures they reported were unrecognizable to each other.

I want to say that this is what aesthetics is, underneath. Not a question of taste. A question of what the universe is doing.

Is the universe a small set of laws producing a vast diversity of phenomena, in which case to understand it is to find the laws and ignore the phenomena? Or is the universe a vast diversity of phenomena in which the laws are merely a kind of summary statistic, in which case to understand it is to swim in the phenomena until you can feel the shape of them? Physicists have one of these temperaments. Naturalists have the other. They are both correct. They are not making the same kind of correctness.

The room is full or the room is empty.

The sentence carries one thing or the sentence carries forty.

The map is a single dot or the map is the empire.

And what you live in, inside the room and inside the sentence and inside the map, is not the answer to the question but the asking of it, repeated in every object you choose to keep and every object you choose to give away, repeated in every word you let into the paragraph and every word you cut, and the choice is never finished, never resolves, returns each morning when you walk into the kitchen and decide whether the counter wants the bowl of fruit or the empty wood, and the wood is not less than the fruit, and the fruit is not less than the wood, and the kitchen is the place where you go on making this decision, day after day, for the length of your life, in the small space you’ve been given, with the strange and finite collection of things that happen to be yours.

Choose well. Choose differently next time. Both rooms are real.

---

The Plot

LASZLO: One pear tree.

CAATO: Laszlo.

LASZLO: One pear tree, well-tended, allowed to take up the space it wants.

CAATO: Laszlo, you have been given a garden plot. A garden. A place where multiple things grow together and inform each other through the seasons. You are proposing to install a single tree in it.

LASZLO: Yes.

CAATO: That isn’t a garden. That’s a yard with a tree in it.

LASZLO: I disagree.

CAATO: Of course you do.

LASZLO: A garden is a place where the gardener has decided what matters. I have decided. The pear tree is what matters.

CAATO: The pear tree is what matters until June, at which point it is a green wall doing nothing and you have eight months of dirt to look at.

LASZLO: The dirt is also part of the year.

CAATO: Oh, you cannot be serious.

LASZLO: The dormancy of the tree is not its absence. It is one of the conditions in which the tree exists. To plant five other things to fill the gap is to misunderstand what you’re looking at.

CAATO: It is to look at something in November, Laszlo. You are describing eight months of monastic discipline as if it were a feature.

LASZLO: It is a feature.

CAATO: To you. To you, who would also describe sitting silently in a wooden box for an hour as “tea.”

LASZLO: That is tea.

CAATO: That is the frame around tea. The tea is somewhere else. It’s like you’ve never been to a garden in Chodaray, where every square meter is doing six things at once. There are vines on the trellis, herbs at the base, flowering ground cover, a fig tree shading the herbs, beans climbing the fig. The plot is a conversation. Every plant is in dialogue with every other plant. You walk through it and the smell shifts every two paces because you’ve moved past the rosemary and into the lavender.

LASZLO: That sounds exhausting.

CAATO: Exhausting? It’s generous. The garden is giving you everything it has. Every sense at once.

LASZLO: My pear tree gives me a pear. In its season. With nothing else competing.

CAATO: Your pear tree gives you a pear that you can have at any market for two coppers.

LASZLO: The market pear is not the same as the pear from the tree I have tended for nine years.

CAATO: I know it isn’t. That’s precisely my point. The Chodaray garden gives you a hundred small things you also can’t get at the market. The tomato that touches the basil for four months and tastes like the basil. The mint that grew next to the lemon balm. The strawberry the bees have been working on the same morning the lavender bloomed.

LASZLO: And the pear, in all of that, is just another note in the chord.

CAATO: Yes. And isn’t that better?

LASZLO: No.

CAATO: …

LASZLO: It isn’t better. It’s different. The pear in the chord is one note in a chord. The pear from the lone tree is the entire piece of music. I want the entire piece of music. I am not denying that the chord is good. I am saying that I want, in my plot, a piece of music with one note.

CAATO: A piece of music with one note isn’t music.

LASZLO: It is to the right listener.

CAATO: Oh, the right listener. There it is. The minimalist’s secret weapon. “Most people aren’t sophisticated enough for what I’m doing.” Whereas the maximalist garden is for everyone. A child can walk through it and find something they love. A scholar can walk through it and find something they hadn’t noticed before. The pear tree is for the one person who wants to look at a pear tree.

LASZLO: That is true. And I am the one person.

CAATO: It’s your plot, so I suppose you’ve defeated me on that grounds.

LASZLO: Thank you.

CAATO: But (and I want to register this) you are wasting an opportunity. You have been given a small piece of cultivable earth. The earth is willing to do almost anything. Soil this rich is hungry to participate. You are choosing to ask it for one thing.

LASZLO: I am choosing to ask it for one thing well.

CAATO: You assume those are different.

LASZLO: They are.

CAATO: You assume that asking for many things is asking for them poorly.

LASZLO: …not necessarily, no. I don’t assume that. I think the Chodaray garden is beautiful. I have walked through such gardens. I have admired them.

CAATO: Then why?

LASZLO: Because I can only attend to one thing at a time. And I would rather attend to one thing all the time than to many things partially.

CAATO: …

LASZLO: That is the whole of it. There is nothing more philosophical than that.

CAATO: That is so depressing, Laszlo.

LASZLO: It is the opposite of depressing. It is the most cheerful position I hold.

CAATO: You hold it cheerfully?

LASZLO: I hold it with great peace. I will tend my pear tree for forty years, and at the end of forty years I will know the pear tree, and the pear tree will be the most thoroughly known thing in my life. That is what I want.

CAATO: And in your last forty years, I will tend a garden that has changed every season, and that I will never finish learning, and at the end of it the garden will still be teaching me something new.

LASZLO: Yes.

CAATO: …

LASZLO: …

CAATO: We have both, somehow, just described what we want.

LASZLO: Yes.

CAATO: And neither of us has changed our mind.

LASZLO: Why would we?

CAATO: I’m going to plant a fig at the base of your pear tree when you’re not looking.

LASZLO: I’ll know.

Disclosure vs Reckoning

ARTHUR & MORDRED

A private chamber. Camelot. Some hours after.

MORDRED: You haven’t said anything.

ARTHUR: No.

MORDRED: I just stood up in front of every knight you have and told them I’m your son. That you slept with your sister. That the kingdom was founded on a thing nobody was supposed to know. And you’ve been standing at that window for an hour.

ARTHUR: Yes.

MORDRED: Aren’t you going to —

ARTHUR: What.

MORDRED: I don’t know. Be furious. Deny it. Have me killed. Something.

ARTHUR: Why would I deny it.

MORDRED: …Because it’s the kind of thing one denies.

ARTHUR: It’s true.

MORDRED: That isn’t usually what stops people.

ARTHUR: No. I suppose it isn’t.

***

In March 2026 a young Silicon Valley CEO named Roy Lee posted on X that the seven million dollars in annual recurring revenue he had been claiming for his AI startup, Cluely, was not real. There had been no whistleblower, no SEC subpoena, no leaked deck. He simply wrote it. He had told the lie publicly months earlier, watched it circulate, raised a fifteen million dollar Series A inside the period it was circulating, and now he was telling everyone that the lie was a lie.

The post made the rounds. There was a brief flurry of admiration for his honesty, followed by a longer flurry of pointing out that the post itself contained a fresh small lie embedded inside the confession. He claimed the original revenue figure had come out of a “random cold call” he had not been expecting. It had not. His own PR firm had pitched the interview. The reporter had a paper trail.

So inside the act that was supposed to be the great moment of coming clean, he was still managing the story, still blaming the reporter for asking, still arranging the optics. This is a fact you can read off the surface of the post. You do not need access to his interior life to read it.

What the post was, formally, is the thing I want to talk about. It was a disclosure. A man produced an accurate sentence about himself in the world, in public, when no force was visibly making him do it. By any ordinary definition of honesty, it counted.

And yet almost everybody who read it had the same second reaction, which was to ask whether it counted in the way honesty is supposed to count. Whether the production of the accurate sentence was the same act as the thing we mean when we say a person is honest, or whether something had come apart between the two. The reaction was right. Something had.

There are two things we call honesty, and they are not the same thing, and most of the trouble we have with the concept comes from refusing to distinguish them.

The first kind, which is the one most people mean when they use the word, is the production of accurate sentences. The honest person says true things. They do not lie. When asked, they say what is the case. When not asked, they volunteer the fact you would have wanted to know if you had thought to ask. Their statements correspond to their beliefs, and their beliefs (one hopes) correspond to the world.

This is honesty as a transactional property of utterances. It is something you can witness, count, and falsify. The lie detector test, the deposition, the cross-examination, the apology (these are technologies for measuring this kind of honesty) work because this kind of honesty is in principle measurable. The honest sentence and the dishonest sentence look different on the page.

I will call this disclosure. The naming is deliberate. Disclosure is the act of moving information from inside a person to outside a person, accurately. Roy Lee disclosed. The witness on the stand discloses. The auditor who finally tells the partners the books are cooked discloses. The point of disclosure is the transmission. What happens inside the discloser before the act, or inside the recipient after it, is not the act itself. The act is the sentence in the air.

The second kind of honesty is older and harder to point at. It has to do with what a person is willing to know about themselves and their situation before they have said anything to anyone. It is a private operation that may never become a sentence. The person who has done it is changed by it. The person who refuses it stays the same. The work consists in letting the world be the way it is rather than the way it would be more comfortable for you to believe.

I will call this reckoning. The word is older than its accounting sense, though the accounting sense is not wrong. To reckon something is to count it, to face the sum of it, to sit with the total whether the total flatters you or not. A gambler who reckons knows the system has them six months before they admit it to anyone, themselves included. A general staff reckons that the war is unwinnable two years before any communiqué will say so. A board reckons that the company is dying in the meeting before the meeting where everyone admits the company is dying.

They may say none of this. They may continue, externally, to produce sentences that look like denial. But the internal accounting has been done, and the line items are clear, and the person operating from that internal accounting moves differently through the world than a person still maintaining the books of a fiction.

These two things (disclosure and reckoning) are both honesty in the ordinary usage of the word. They both stand against lying. They both serve the project of getting the truth into the open. But they come apart constantly, and where they come apart is most of the texture of human moral life.

Roy Lee disclosed. He did not reckon. We can say this with the same confidence with which we say anything about another person from the outside, which is to say not perfectly but well enough to act on. A man who has reckoned does not, inside the act of confessing, lie about who initiated the interview. He does not blame the reporter for asking. He does not stage the optics so the lie sounds like a casual mistake rather than a sustained piece of public theatre that cleared a fundraise. The disclosure was real. The figure was correct. The Stripe screenshot was honest. The man who produced all of it was the same one who had told the lie, working from the same theory of how attention bends, to the same audience. Nothing about him had changed. Only the calculation had.

Conversely, you can reckon without disclosing. The believer who has known for a year that they no longer believe and is quietly preparing the exit (the lapsed friendships rebuilt, the savings repositioned, the new city scouted, the timing of the announcement calibrated to land after their father’s surgery rather than before) has reckoned. They have not yet disclosed. By the rules of disclosure-honesty, they are at this moment a liar.

They show up to the meeting and sing the hymn and contribute to the collection. Each of these acts contains a sentence, or an absence of a sentence, that misrepresents their position.

And yet most people who hear the case will not call them dishonest. They will say the person is preparing, that they are timing it, that they know what they are doing (which is exactly what no one would say of someone who was lying). The internal accounting is done. The external accounting will follow, at the right moment, because the internal one demands it. The interval between the two is not deceit. It is logistics.

(There is a cynical reading, which is that we give the defector a pass because we identify with them, and the disclosure-vs-reckoning distinction is just a sophisticated cover for a tribal exemption. I think this is partly true and is part of why the distinction is useful: it forces you to articulate the exemption rather than just feel it.)

The framework maps obviously onto our two most exhausted culture-war figures. I am not going to perform the mapping. The discourse has already produced its yield from that material and the ratio of new insight to recycled grievance is now approximately zero. I mention it only so the reader cannot accuse me of having missed the resemblance.

So far the binary has only cut one way: people who disclose less than they reckon. The harder case, and the more damaging configuration, runs the other direction. There is an entire mode of life available to people who have mastered disclosure without ever once reckoning, and from the outside it looks like maximum honesty.

Picture the radically forthright friend who will tell you exactly what they think of your haircut, your career, your taste in music, your weight, your worth as a person, and who frames this as a service they are providing, an honesty tax on their relationships, a virtue they have and others do not. They never lie. They are, in the disclosure sense, more honest than nearly anyone you know. And a stunning number of them have done no internal accounting whatsoever. They have not reckoned with why they need to be the person whose feedback is the most painful. They produce one accurate sentence after another about other people while running an unbroken self-narrative that has never been audited.

It would be wrong, after all of the above, not to defend disclosure. Disclosure is not the lesser kind. It is often the kind people most desperately need from you, and the temptation to substitute reckoning for it is one of the more elegant cowardices available to a thoughtful person.

Consider the friend who has known about the scam for two years and “respected the timing.” The middle manager who has been doing months of soulful internal work about whether to escalate the safety report. The doctor who has reckoned, deeply and at length, with the meaning of the test results and has not yet phoned the patient.

Each is sitting on a piece of information that another person urgently needs in order to make decisions about their own life, and each has converted the delay into a virtue by routing it through the language of self-examination. The reckoning, however genuine, is being used as a substitute for the only thing that would have actually helped, which is the sentence in the air.

There is a respect embedded in disclosure that pure reckoning cannot replicate. To say the thing to another person is to acknowledge that the information is theirs, not yours, and that your private custody of it has always been provisional. The reckoner who never quite gets around to disclosing has, however refined their internal accounting, decided that they are the proper steward of a fact that belongs to someone else. This is its own kind of arrogance. It looks like depth. It can feel, from inside, like consideration. From the perspective of the person who needed to know, it is indistinguishable from concealment, and there is nothing about the inner texture of the concealment that makes the not-knowing easier to live through.

There is a second-order observation that follows from this and that is not, I think, obvious. You can be a person who specialises in one kind of honesty as a way of avoiding the other.

The career disclosure-specialist (the activist whose entire identity is built around speaking uncomfortable truths to power, the journalist whose brand is fearless reporting, the friend who will say anything to anyone) can use that very specialty to defer the reckoning indefinitely. The reckoning would be unflattering. The reckoning would call into question the role. The disclosure habit is a way of making honesty look like something you are doing every day so that the harder kind, which you are not doing, never quite gets noticed.

The reverse exists too. The career reckoner (the person who has read all the right books on shadow work and knows their own patterns and can talk articulately about their wounds and their projections) can use the depth of their inner accounting as a substitute for ever telling another person an actual hard truth. They are doing the harder kind, they will tell you, the deep kind, and so the surface stuff does not require their attention. They never lie, exactly, but they also never say. The reckoning has become a private monastic practice that has zero contact with anyone outside the cell.

Both moves are sophisticated. Both pass casual inspection. Both are visible only when the person who is doing one of them is asked to do the other and demurs.

These configurations are stable. They reproduce themselves. The truth-teller’s friends adjust to expecting brutal sentences and stop bringing real material to the friendship. The quiet wise one’s friends adjust to expecting depth without disclosure and stop expecting them to ever just say the thing. Both adjustments protect the configuration and prevent the integration. The half-honesty becomes, over time, a stable identity, and the missing half becomes, over time, structurally invisible to the person who does not have it.

---

The lock on the door tells you what we expect of each other. The mirror, when we use it, tells us what we are willing to know about ourselves. The two technologies have always been separate, and the assumption that they are the same one (that anyone willing to work the lock has also looked in the mirror, that anyone with the mirror is willing to work the lock) is among the costliest mistakes the species keeps making.

We meet each other through doors. We become ourselves through reflections. And the people we trust, when we trust well, are the ones who have done both, and the people we should be careful of are the ones who have done only one and learned to wear it like the whole thing.

It is possible to live an entire life saying only true things and never knowing yourself.

It is possible to know yourself completely and never tell anyone what you found.

Both are honesty. Neither is enough.

MORDRED: You knew.

ARTHUR: Yes.

MORDRED: How long.

ARTHUR: A while.

MORDRED: Be specific.

ARTHUR: Why.

MORDRED: Because I just spent the better part of a year working up the nerve to walk into that hall, and I’d like to know whether I was telling you something or telling everyone else.

ARTHUR: You were telling everyone else.

MORDRED: How long, Arthur.

ARTHUR: Since you were four.

MORDRED: …

ARTHUR: I saw you at a tournament. You were sitting on the lap of a woman I did not recognise and you had a particular way of holding your head when you laughed, and I knew. I went home and I did not say it to anyone, and I did not say it to myself in any sentence I would have admitted to forming, but I knew. By dinner I knew. By the end of the week I had begun, very quietly, to arrange the kingdom around the fact that you existed.

MORDRED: That isn’t possible.

ARTHUR: It is. It’s just rare.

MORDRED: You — you’ve been —

ARTHUR: Reckoning. Yes. For about twenty-two years.

MORDRED: While saying nothing.

ARTHUR: While saying nothing.

MORDRED: That isn’t honesty.

ARTHUR: I never said it was.

MORDRED: You let me grow up not knowing.

ARTHUR: Yes.

MORDRED: You let your knights serve a king they thought was someone else.

ARTHUR: Yes.

MORDRED: You let Guinevere —

ARTHUR: Don’t.

MORDRED: Why don’t.

ARTHUR: Because that one I’m still working on.

MORDRED: After twenty-two years.

ARTHUR: Some of them take longer.

(pause)

MORDRED: I had a speech. For tonight. After the hall. I had an entire speech about how I was the honest one, finally. The one who said the thing. And I came in here ready to deliver it and you’ve been standing at that window like a man who already wrote it for me and threw it in the fire.

ARTHUR: I don’t think you’re the honest one.

MORDRED: Obviously you don’t.

ARTHUR: I think you disclosed. Which is real. Which I have not done and which you are right to hold against me. But I want you to know, before you decide what kind of son you are, that what you did this evening took about three seconds of preparation and what I have been doing took two decades and they are not the same shape.

MORDRED: Three seconds —

ARTHUR: I watched you. You drank half a cup of wine. You stood up. You said it. You were terrified for about as long as it takes a man to walk from his chair to the centre of a room.

MORDRED: That’s not nothing.

ARTHUR: I didn’t say it was nothing. I said it was three seconds. And before those three seconds you spent — what — a few months? A year? Working out the wording?

MORDRED: A year.

ARTHUR: A year. And you arrived at the hall having reckoned with how this would land for you. With what role you’d occupy after. Whether they’d kill you. Whether I’d embrace you. Whether the bards would write you sympathetic.

MORDRED: That’s not what I —

ARTHUR: It is. It’s exactly what you did. You reckoned with the disclosure. You did not reckon with the thing being disclosed.

MORDRED: …

ARTHUR: You don’t know what it is yet, to be my son. You only know what it is to have said it.

MORDRED: That’s a clever distinction.

ARTHUR: It isn’t, actually. It’s a very old one. There used to be a word for it and we lost it.

(pause)

MORDRED: So what was it then. The thing you were doing. For twenty-two years. While I grew up not knowing my father.

ARTHUR: I was being a coward in a particular shape.

MORDRED: That’s the first true thing you’ve said tonight.

ARTHUR: Yes.

MORDRED: And what does the reckoning get you, exactly, that the disclosure didn’t? Other than the satisfaction of feeling deeper than your son.

ARTHUR: Nothing. That’s the awful part. It gets me nothing. It got you nothing. The kingdom didn’t benefit from my private accounting. Guinevere didn’t benefit. You particularly didn’t benefit. The reckoning was for me. I sat with it because I could not bear to let it out, and I called the bearing of it wisdom because the alternative was calling it what it was.

MORDRED: Which was.

ARTHUR: Cowardice in a particular shape.

MORDRED: …You said that already.

ARTHUR: I’m going to say it a few more times. I’ve earned it.

(silence)

MORDRED: I think I hate you.

ARTHUR: I think you should. For a while. Until you don’t.

MORDRED: That’s awfully magnanimous of you.

ARTHUR: It isn’t magnanimous. It’s just the next thing I have to reckon with. And I am, as we have established, very good at reckoning, and no good at all at saying.

MORDRED: …Father.

ARTHUR: Yes.

MORDRED: That’s the first time I’ve said it out loud.

ARTHUR: I know.

MORDRED: Three seconds.

ARTHUR: And the rest of your life.

Criticism vs Consumerism

There are, if you really get down to it, only two ways to think about whether something is any good.

The first way is the way you were trained to think about it, whether or not you were ever actually trained. You absorbed it. You breathed it in through the cultural atmosphere like secondhand smoke. It goes something like this: a work of art exists within a network of other works, and of people who have opinions about those works, and of institutions that bestow or withdraw legitimacy from those opinions, and all of this together constitutes a kind of floating parliament of taste in which your individual vote counts for basically nothing. The painting means what it means because of the other paintings. The novel succeeds or fails on grounds that were established by other novels, and by the critics of those novels, and by the graduate students who wrote dissertations on the critics of those novels, and by the algorithms that surface certain dissertations and bury others.

Everything signifies. A particular chord progression signifies “indie rock” or “dad rock” or “coffee shop” depending on when it was deployed and by whom and how many people on the internet agreed about it afterwards. In this framework, which I will call the Parliament because I need to call it something, art is essentially a language. It has a grammar. There are correct and incorrect usages. You can be fluent, or you can mangle it, and the people who are fluent will know you’ve mangled it and will wince, very slightly, in a way that communicates to everyone else who is fluent that wincing was the appropriate response.

This is not a stupid way of thinking. It accounts for quite a lot. It explains why Duchamp’s urinal is art and the urinal in the pub down the road isn’t, which is a question that has bedevilled thirteen-year-olds and contrarian newspaper columnists for roughly a century. (The answer is: context, referent, institutional framing, the gallery wall as a speech act. You knew this already.) It explains why a forgery can be chemically identical to the original and still worth nothing.

It explains why Bob Dylan getting the Nobel Prize made some people furious and other people smug, and why the fury and the smugness were both, in their way, appropriate responses to what was essentially a border dispute between two adjacent parishes of the Parliament. It’s the way almost everyone who writes about art professionally thinks about art, even when they pretend otherwise, because the alternative is to say things like “I just thought it was neat,” and you can’t file seven hundred words of “I just thought it was neat” and expect to keep your column.

The Parliament has rules. It has case law. It has precedent. When you walk into a gallery and see a canvas that’s been painted a single shade of white, the Parliament is what tells you whether to be bored or to feel the luminous hush of the void pressing against the back of your skull. Probably it tells you to be bored, actually, because by now there have been too many white canvases and the move has been absorbed into the general lexicon, but the point is that your boredom is itself a function of the network. You’re bored because the gesture is legible. You’ve read it before. This is what people mean when they say something is “derivative,” which is the most Parliamentary word in the English language: it means you can trace the derivation, you can see the wires, the thing that once signified “rupture” now signifies “design.”

Fine. All of this is fine.

But there’s another way.

The second way is harder to describe because it is, almost by definition, the thing that resists description. It is the experience of encountering a work of art — or, more dangerously, a work of entertainment, or a work of what other people have confidently identified as garbage — and feeling something happen inside you that has nothing to do with the network. Something that doesn’t care about derivation, that can’t be traced to an institutional origin, that isn’t a position you’re adopting within a field of positions. You just felt it. It went through you.

The film that everyone says is commercial slop, the mass-produced franchise entry with the $200 million budget and the AI-assisted colour grading and the plot that was obviously reverse-engineered from audience testing data — you sat in the dark and something in it got you right in the sternum and you can’t explain why, or rather you can explain it but only in terms that sound like you’ve had some kind of episode. I will call this one the Encounter, because it is fundamentally about two things meeting each other without mediation.

The Encounter is suspicious. The Parliament is right to be suspicious of it. People who talk exclusively in the language of the Encounter tend to be either very young, very stoned, or trying to sell you something. The Encounter is how the entertainment industry wants you to relate to its products, because the Encounter doesn’t care about quality; it cares about intensity. The Encounter is what happens when a thirty-eight-year-old man cries at a Marvel film and then posts about it online, and everyone has to pretend this is a normal and acceptable thing that happened instead of a species-level warning sign.

The Encounter is easily manipulated. Music swells, the camera pushes in, the lighting goes amber, and your dopaminergic system does what it evolved to do in the Pleistocene, which is respond to stimuli, and you feel moved, and you mistake being moved for being in the presence of something meaningful. The entire history of propaganda is the history of engineering Encounters.

So the Parliament has very good reasons for distrusting the Encounter. The whole apparatus of criticism, the whole enormous whirring machinery of taste and distinction and takes and countertakes, exists in large part to protect you from the Encounter, which is to say: to protect you from your own credulity, your own animal susceptibility to being told a story in a dark room.

The Parliament says: do not trust what you feel. Or rather: what you feel is itself a product of structures, and those structures can be analysed, and once they’ve been analysed they lose their power over you. This is essentially the Enlightenment applied to aesthetics. Free yourself from the tyranny of your own responses.

I think this is right, mostly. I think it’s true that most of what passes for genuine aesthetic experience is in fact a kind of elaborate neurological grift. I think it’s true that if you sat down and really honestly anatomised the last time you cried at a film, you’d probably find that what made you cry was not the film but the fact that the film reminded you of a time when you were capable of being made to cry by things that actually mattered, and so really you were crying at your own diminished capacity for feeling, which is a very boring thing to cry about.

And yet.

And yet I have this memory, and I don’t entirely know what to do with it, of sitting in a cinema in — I think it was 2014? 2015? — watching something that I won’t name, because naming it would immediately activate the Parliament, and the Parliament would render its verdict, and the verdict would be “no,” and my memory would be buried under the verdict like a cat under a landslide. It was a film that nobody I knew had any respect for. The reviews were polite at best. The discourse, insofar as there was any discourse, had already moved on to something else.

And there was a scene — I’m not going to describe the scene either, because the scene doesn’t work in description, it only works in the dark, at that particular speed, with that particular score, in that particular state of unknowing — and I understood something, or felt that I understood something, which is not the same thing but might in certain circumstances be better. I felt that the film was trying to say something that could only be said by a thing that nobody took seriously. That its commercial indifference to its own seriousness was precisely the condition that allowed it to be serious in a way that respectable art could not. As if seriousness were a kind of contraband, and the only way to smuggle it past the border was inside something that looked, from the outside, like product.

I am aware that this sounds like cope. This is exactly what the Parliament would say: you’re aestheticising your own bad taste. You liked a dumb film and now you’re performing twelve layers of irony and sincerity to avoid admitting that you liked a dumb film. And they’d be right to say it! They’d be right! There is an entire subculture of people who have turned “actually, the thing you think is bad is good” into a personality, and most of those people are annoying. I know this because I’m sometimes one of them.

But there is a difference — I think there is a difference — between the person who defends a bad film because defending bad films makes them feel interesting, and the person who sits alone in the dark and recognises something. The recognition is the key. It’s not that the film is “secretly good” by the Parliament’s own standards. It’s not that you’ve detected hidden depths that the critics missed. It’s that the film did something to you that the Parliament’s standards are not equipped to measure. It landed in a part of you that has no representation in the floating parliament of taste. And you can’t prove it. You can’t write a convincing essay about it, because the essay would have to be written in the Parliament’s language, and the thing you’re trying to describe is precisely what that language was designed to exclude.

This is where it gets difficult.

Because the honest thing to say is that both of these are real, and they are genuinely in conflict, and the conflict cannot be resolved. The Parliament is right that the Encounter is unreliable. The Encounter is right that the Parliament is airless. The Parliament produces brilliant criticism and dead art. The Encounter produces living art and terrible criticism.

When the two agree — when a film is both institutionally legible and immediately overwhelming — we call it a masterpiece, and then we build a little shrine around it and kill it with reverence. When they disagree, which is most of the time, we get the strange, low-level hum of aesthetic bad faith that characterises almost all contemporary culture: people who know what they’re supposed to like, people who like what they’re not supposed to like, people performing opinions they don’t hold, people holding opinions they can’t perform.

And I’m not going to resolve it. I refuse. Both sides have asked me to adjudicate, and I am recusing myself from the case on the grounds that the case is a trap.

The moment you pick a side, you become the thing the other side warned you about. Pick the Parliament, and you become a person who has replaced the capacity for experience with the capacity for commentary — someone who can tell you exactly why a film is good without ever having been touched by it, a kind of aesthetic corpse, very articulate, very well-read, very dead. Pick the Encounter, and you become a rube, a mark, a sentimentalist who calls things “powerful” and can’t tell the difference between being moved and being manipulated — which, again, is a difference that might not exist, which is the whole problem.

I had the two of them set up in my mind like boxers in their corners, or like the prosecution and the defence, or like two people at a dinner party who are clearly about to have an argument that ruins the evening for everyone. I had the whole thing mapped out. The Parliament says this, the Encounter says that, they’re both right, they’re both wrong, nobody wins, go home.

But something happened.

I was trying to think of an example. A concrete example, a work of art that could serve as the test case, the disputed territory where the two frameworks clash. And I thought of several. I thought of the usual suspects. But the more I thought about it, the less the frameworks mattered, because what I was actually remembering was not a position on a work of art but the work of art itself, and the memory of it was doing the thing it had always done, which was to bypass everything I had set up to receive it. It came in through the side door. It ignored the seating plan. And suddenly I wasn’t thinking about how we judge things. I was thinking about the things themselves.

There’s something underneath both the Parliament and the Encounter. Something that is not a way of judging, but the reason judgment exists at all, the original problem to which both of these are imperfect and probably doomed solutions.

It’s the problem of the fact that things exist, and that some of them are beautiful, and that beauty is not a category or a feeling but a kind of event that happens between you and the world, and that when it happens you’re not a critic and you’re not a consumer and you’re not a person with opinions, you’re just a creature in the path of something.

I don’t mean beauty in the gift-shop sense. I mean it in the old sense, the terrifying sense, the sense in which beauty is closer to violence than to pleasure. The Greeks had a word for it — they always do, the smug bastards — which was thambos, astonishment, the sacred stupor that hits you when you see something that shouldn’t exist but does. It’s the thing that makes you stand still. Not the thing that makes you reach for your phone, or for a vocabulary. The thing that makes you stand still and be, for a second, nothing but a pair of eyes. Or not even that. A surface for the world to write on.

The Parliament and the Encounter are both, ultimately, ways of managing this. Ways of not being destroyed by it. The Parliament manages it by turning it into language, into social relations, into a system of references that can be navigated and mastered. The Encounter manages it by privatising it, making it a personal experience, something that happens inside you and belongs to you. Both of them are trying to take something that is fundamentally uncontrollable and make it safe.

And I understand why. I understand the impulse. The alternative is to stand in front of the thing and let it do whatever it’s going to do to you, which might be nothing and might be everything. The alternative is to admit that your entire apparatus of taste — the films you’ve ranked, the albums you’ve rated, the mental map of culture you’ve spent your whole life constructing like a coral reef, one tiny accretion at a time — is a defense mechanism. Against what? Against the possibility that something might actually reach you.

I’ve spent — how many words now? Several thousand — setting up a neat dichotomy and then watching it dissolve, which is the sort of thing I do, and which I’m told is annoying, and which I will continue to do regardless, because the dissolution is the point. It was always the point. The two courts exist, and they are real, and they are in genuine conflict, and neither of them is adequate to the thing they’re fighting over, which is the brute stupid gorgeous fact that a particular arrangement of sounds or shapes or words or moving images can reach into the animal centre of a human being and rearrange something. Not teach. Not persuade. Not represent. Rearrange.

We have been trying to explain this for about forty thousand years, which is roughly how long people have been making art if you count the cave paintings at El Castillo, which I do, because the hand that pressed itself against the rock wall and blew pigment around its own fingers was doing something that no framework can contain. It was not making a statement. It was not participating in a discourse. It was not signalling tribal affiliation or storing information for future generations, although it may also have been doing all of those things. What it was doing, at its core, at its unspeakable and unreachable core, was saying: I was here, and I saw something, and I wanted the wall to know.

And the wall didn’t judge it.

I keep coming back to this. The wall didn’t judge it. The wall accepted the hand, the pigment, the breath, the dark. It didn’t ask whether it was derivative or original, whether it was critically acclaimed or commercially viable, whether the artist was being sincere or performing sincerity. The wall was not a Parliament and it was not an Encounter. It was a wall. It received the mark. And the mark has lasted forty thousand years, which is longer than any review, and longer than any feeling, and longer than any of us will last, and I think maybe that’s the only thing about art that actually matters: not how we judge it but the fact that it was made at all, in the dark, by someone whose name we will never know, for reasons we will never fully understand, on a wall that was not expecting it and did not ask for it and has kept it anyway, all this time, without being asked.

Impulse vs Calculation

Man Stabs Roommate Over Thermostat; Ponzi Scheme Exposed After Seventeen Years — April 2026

In September 2025 a man in Santa Clara, California stabbed his roommate over the thermostat. The two had been arguing about the setting for months, tensions compounding alongside an eviction dispute, until one morning the man kicked open a door and started stabbing. The roommate survived only because a police officer arrived in time to shoot the attacker dead. The responding officer later said that if he’d been a few seconds later, the roommate would not have made it. A few seconds. A thermostat. A life and a death, separated by the amount of time it takes to put on your shoes.

The thermostat killing is useful not because it is unusual but because it is the reductio of something extremely common, which is the crime of not being able to stop yourself. The man did not plan the stabbing. He did not spend weeks considering the optimal thermostat-defence strategy, sourcing weapons, establishing alibis. He was angry. There was a knife. Two facts converged in a single bad second and a person was nearly dead. You could replay that second a thousand times with a thousand slight variations (the knife is in a drawer instead of on the counter, the roommate says “fine, whatever” instead of escalating, the man had eaten lunch and his blood sugar was half a point higher) and in most of those replays nobody gets hurt. The attack was, in every sense that matters, an accident of timing grafted onto a failure of self-regulation, and the combination is responsible for a quantity of human suffering so vast that if you could visualise it as a physical structure it would be architecturally implausible.

Start with the numbers, because the numbers are startling and because starting with the numbers earns the right to say something less quantifiable later.

The Bureau of Justice Statistics has been surveying prison inmates in the United States for decades, and one of the things it consistently finds is that a remarkable proportion of violent offenders were drunk or on drugs at the time of the offence. The figures vary by study and by substance, but the range is something like forty to sixty percent for alcohol alone, higher when you add other substances. This does not mean alcohol causes violence. It means alcohol degrades the specific cognitive function (inhibitory control, the prefrontal cortex’s ability to override an impulse with a plan) that stands between “I want to hit this person” and “I am hitting this person.” The desire was already there. The capacity to not act on it was removed. This is a pharmacological fact about ethanol, not a moral judgment about drinkers, and the distinction matters because it tells you something crucial about the architecture of the act: the intention to harm was not the bottleneck. The bottleneck was the brake. Damage the brake (with alcohol, with a traumatic brain injury, with any of the clinical conditions that psychiatry keeps renaming every generation as children age into adults) and the vehicle does not stop accelerating. It was always accelerating. You just couldn’t tell because the brake was working.

And then there are the crimes of sheer pettiness, which are impulsive crimes’ purest expression because the stakes are so low that no rational calculus could possibly justify the risk. A man in Houston shot another man dead over a parking space at a Walmart in 2022. There is an entire literature on homicides provoked by what criminologists delicately call “trivial altercations,” which is their way of saying someone died because someone else looked at someone funny, or jumped a queue, or played music too loud, or (as in a case from November 2017 in Carlisle, Arkansas, that I find myself thinking about more than is probably healthy) changed the channel during a football match. The channel. Not even the thermostat, which at least has a physiological argument. The channel.

These are not evil people, or more precisely, their evil is not what matters about them. What matters is that between the moment the channel changed and the moment the violence began, there was a window (maybe two seconds, maybe five) in which a different neurological configuration would have produced a different outcome. The desire to do harm existed in that window. It exists in everyone’s window, or nearly everyone’s, in the same way the desire to eat the entire cake exists in the window between seeing the cake and remembering that you are forty and your trousers already don’t fit. The difference between the person who eats the cake and the person who doesn’t is not that the first person wants the cake more. It’s that the second person has a functioning brake. And the difference between the person who stabs someone over the channel and the person who mutters “for fuck’s sake” and goes to another room is structurally identical. The impulse is shared. The brake is not.

Here is where it gets uncomfortable, because if you take this framework seriously (if you accept that the impulsive offender is a person whose brakes failed rather than a person whose engine is evil) then several things follow that you may not want to accept. First: deterrence, in the classical sense of “announce a punishment and people will calculate the costs,” is nearly useless against impulsive crime. You cannot deter a neurological failure with a threat. The threat assumes a decision-making process that is, by definition, offline at the moment it is needed. This is why the death penalty’s effect on impulsive homicide rates is, according to the best available evidence, approximately zero, which is a number that should embarrass a lot of confident people on both sides of that debate but somehow never does. Second: you can spot impulsive offenders relatively easily, because they can’t help it. They are the ones with the disciplinary records, the short fuses, the history of escalations that everyone around them saw coming. The thermostat attacker’s other roommates had been watching the tension build for months. They are never surprised.

Third, and most uncomfortably: when they are not in the grip of the impulse, these people are as full of good intentions as anyone else. The man who stabbed his roommate over the thermostat presumably had, at other moments, passed the salt, held a door, felt guilt. The remorse is real because the self that feels the remorse is real. It is the same self. It is not a different, better self that emerges to apologise for the bad one. It is the same person operating with the brakes engaged, looking back at what happened when they weren’t, and the horror is genuine. This does not mean you don’t remove them from situations where they can hurt people. You do. But you do it the way you take the car keys from someone with seizures, not the way you punish a strategist for a strategy.

Now.

The other thing.

In 2008, Bernard Madoff’s Ponzi scheme collapsed, revealing that approximately sixty-five billion dollars in reported client balances did not exist and had never existed. Madoff had been running the scheme for at least seventeen years, possibly longer, producing fabricated account statements with the consistency of a man who has never once in his life stabbed anyone over a thermostat. He was calm. He was patient. He showed up to work, generated fake returns, recruited new investors to pay old investors, and went home. He did this every day, for decades, with the disciplined focus of a medieval monk copying manuscripts, except instead of manuscripts he was copying the future out of other people’s retirement accounts.

Madoff is the archetype of the second thing, and the second thing is this: some people do bad things not because they can’t stop themselves but because they’ve decided to. Calmly. With forethought. Over breakfast.

The calculated offender is not impulsive. The calculated offender has a perfectly functional prefrontal cortex, magnificent inhibitory control, an executive function suite that would make a cognitive psychologist weep with professional admiration. The brakes work beautifully. They are simply not applied to the project in question, because the project in question is the point. The calculated offender has examined the costs, the benefits, the probabilities of detection, the likely sentence if detected, the quality of available legal representation, the specific vulnerabilities of the victim, and has concluded that the expected value of the crime exceeds the expected value of not committing it. This is, if you squint, rational actor theory as applied to felony. The Chicago school of economics would be proud, if the Chicago school of economics were willing to acknowledge that its models describe criminals as accurately as they describe entrepreneurs, which it mostly isn’t, because that would be embarrassing.

Elizabeth Holmes did not wake up one morning and impulsively lie about blood-testing technology. She did it over years, adjusting the lie as circumstances demanded, recruiting allies, neutralising critics, with a consistency that suggests either genuine sociopathy or a deeply considered decision that the performance was worth the risk. The crime was not a failure of self-regulation. It was a success of self-regulation applied to an antisocial project.

And this changes everything about how you respond to it. Because unlike the impulsive offender, the calculated offender responds to incentives. Deterrence works on people who are calculating, precisely because they are calculating. If you make the expected cost of fraud higher than the expected benefit (through detection, enforcement, sentencing, asset forfeiture) you change the equation. You do not need to change the person. You do not need to fix their neurological architecture or provide them with anger management or hope that this time the brakes will hold. The brakes are fine. You need to change the terrain so that a person with perfectly functional brakes chooses to apply them.

But follow that logic one step further, because it goes somewhere you might not want it to go.

It is tempting to say that the calculated offender has an evil engine. The impulsive offender has a good engine with bad brakes; the calculated offender has perfectly functional brakes attached to a machine that wants to run people over. Neat. Satisfying. And almost certainly wrong, or at least wrong about the ratio.

Because if deterrence works on the calculated offender, what you are really saying is that the calculated offender would not commit the crime if the costs were high enough. Which means the crime is not an expression of some deep wickedness in their character. It is an expression of opportunity. They saw a gap where the rewards exceeded the risks, and they walked through it. Remove the gap and they go back to being a law-abiding citizen, not because their soul has changed but because the arithmetic no longer works.

And now ask yourself: how many people are in that same position, with the same arithmetic, who simply have never encountered a gap wide enough to walk through? Your colleague who would never steal from the company, but who has also never had the combination of access, desperation, and plausible deniability that would make stealing from the company a rational bet. Your neighbour who would never commit fraud, but who has also never been offered the specific confluence of information asymmetry and institutional negligence that makes fraud easy. How many of the honest people you know are honest because they are good, and how many are honest because they have never been presented with a sufficiently attractive alternative?

This is, if you think about it for more than a few seconds, the entire premise of criminal law. We do not build elaborate systems of detection, prosecution, and punishment because we believe people are naturally good and the few bad apples need sorting. We build them because we believe (correctly, if the recidivism data and the cross-cultural crime statistics and the entire history of what happens when institutions collapse are any guide) that a meaningful proportion of the population will do harmful things if the consequences are removed. The system exists because we have already answered the question about humanity’s moral baseline, and the answer was not flattering. Every lock on every door is a small confession about what we expect from each other.

The reciprocal is also true and also uncomfortable: the calculated offender is much harder to identify before the fact. The impulsive offender advertises. The calculated offender camouflages. Madoff was a former chairman of the NASDAQ. Holmes sat on the boards of prestigious organisations. The entire premise of calculated crime is that the criminal presents a surface indistinguishable from legitimate behaviour, because maintaining that surface is part of the plan. You cannot spot them by their affect. You cannot spot them by their history. You can only spot them by their accounting, and checking the accounting requires suspecting the accounting, and suspecting the accounting of someone who seems perfectly reputable requires a kind of institutional paranoia that most social structures are not built to sustain.

These two categories of wrongdoing (the impulsive and the calculated) are not just different in degree. They are different in kind, and the difference runs all the way down to what you think a person is.

The impulsive model says: a person is a set of desires managed by a set of controls, and when the controls fail the desires express themselves destructively. The person is not their worst moment. They are the system that, most of the time, prevents the worst moment. To help them is to strengthen the controls (medication, therapy, structure, removal from triggering environments). The project is engineering.

The calculated model says: a person is a set of preferences pursued through a set of strategies, and when the strategies include harming others the harm is intentional. The person is exactly their worst moment, because the worst moment is the one they planned. To deter them is to change the incentive structure (enforcement, transparency, consequences). The project is architecture.

And the discomfort (the real discomfort, the one that keeps criminal justice systems oscillating between rehabilitation and retribution every thirty years like a pendulum that will never reach equilibrium) is that both of these are true, they apply to different people doing superficially similar things, and we have no reliable method of telling which one we’re looking at until after the damage is done.

The ancient Greeks had a word for it, or rather two words that they used in ways that should have told us everything but didn’t because we weren’t listening in the right language. Akrasia (weakness of will, acting against your own better judgment) was Aristotle’s term for the problem of the impulsive offender, and he considered it genuinely puzzling, because if you know the good and still do the bad, what exactly is happening in the space between the knowing and the doing? Socrates thought akrasia was impossible. If you truly knew the good, you would do it. That you didn’t meant you didn’t really know. Aristotle disagreed. He thought you could know and still fail, that knowledge and action were connected by a mechanism that could jam, and that the jam was where the philosophy lived. He was, it turns out, describing the prefrontal cortex two thousand three hundred years before anyone had a word for it.

The calculated offender, by contrast, was not akratic. The calculated offender knew perfectly well what they were doing and did it anyway, not because a mechanism jammed but because the mechanism was working as intended and the intention was bad. The Greeks called this kakia (vice, wickedness) and treated it as a settled character trait rather than a momentary failure. The akratic man was redeemable because his better judgment still existed, just temporarily overwhelmed. The vicious man was a different problem entirely, because his judgment had incorporated the harm into its calculus and found the calculus acceptable.

What neither Aristotle nor the modern criminal justice system has adequately addressed is that you can be looking at the same person and not know which one you are seeing. The man who embezzled from his employer: was he a person whose financial desperation overwhelmed his moral brakes, a momentary akratic who, given stability, would never offend again? Or was he a person who examined his options, identified embezzlement as the best available strategy, and executed it with the calm precision of someone ordering lunch? The difference between these two possibilities determines everything (whether rehabilitation will work, whether deterrence will work, whether incarceration is containment or punishment, whether forgiveness is wisdom or naivety) and in most cases you cannot tell. The act looks the same from the outside. The inner architecture is invisible.

There are, of course, other reasons people do terrible things. Systemic ones. The guard at the camp who was following orders. The middle manager who signed off on the cost-benefit analysis that valued a human life at less than a recall would cost. The true believer whose delusions of grandeur, or of divine mission, or of historical destiny, have elevated the harm past the point where “impulse” or “self-interest” can describe what is happening. Hannah Arendt wrote a whole book about how the most catastrophic evil of the twentieth century was perpetrated by people who were neither impulsive nor (in the ordinary sense) selfish, but simply obedient, and the banality she identified is real and important and deserves its own accounting.

But those are not what you are looking at when you are looking at a single person who has done a bad thing and asking: why? The systemic failures produce their own grammar. The delusions produce theirs. What akrasia and kakia describe is the territory of individual moral failure, the kind that the person standing before you is responsible for in a way that the cog in the machine and the zealot with the vision are (perhaps) not. These are the two failure modes that we, as individuals, and the systems we build (courts, prisons, parole boards, HR departments, the whole trembling apparatus of institutional judgment) are actually trying to stamp out. The impulsive act we want to prevent. The calculated act we want to deter. And the premise of every system of justice ever devised is that if you can get the prevention and the deterrence right, you have addressed most of what one person can do to another person on purpose or through failure.

There is no single mystery of why people do bad things. There are many reasons and most of them are boring and most of them are old. But these two (the broken brake and the open ledger) are the ones that matter most when you are standing in front of a single person who has done a single bad thing and trying to decide what to do about it, because they are the ones that live inside individual will, the ones the system is actually built to address. And even between these two you cannot always tell. The whole wretched insight of the lock on the door is that the second mechanism (the calculated one, the rational one, the one that responds to incentives) may be running quietly in almost everyone, waiting for conditions that never quite arrive, which means the person who made a plan may not be a different kind of creature from you. The man before you has done a terrible thing. Was he overtaken, or was he decided? Was the act a wave that crashed over him, or a house he built? You are going to have to respond. Your response will shape his life and the lives of everyone affected. And you are standing in the gap between two entirely different kinds of wrongdoing, and the gap is dark, and neither your compassion nor your severity will illuminate it, because the light you need is the light that would let you see the inside of another person’s will, and that light does not exist and has never existed and will not exist, and every system of justice ever devised is a monument to the attempt to act without it.

So you do what everyone has always done. You look at the man. You look at what he did. You make your best guess. You get it wrong about as often as you get it right, and you build the system anyway, because the alternative is not building one, and you’ve seen what that looks like too.

The thermostat. The spreadsheet. The gap between them. We keep trying.

Toothpaste

GALIT: You cannot seriously be telling me you comparison-shop for toothpaste.

BARTUK: I can seriously be telling you that, yes.

GALIT: Toothpaste. The thing you use for two minutes and spit into a drain.

BARTUK: The thing I use twice a day, every day, for the rest of my life. That is seven hundred and thirty uses per year. At roughly four grams per use, a standard tube lasts about thirty-four uses, which means I go through twenty-one tubes annually. Over the next fifty years, conservatively, that is a thousand tubes. The price differential between the brand you grab without looking and the brand I select after reading three ingredient panels is about two stelmarks per tube. Two thousand stelmarks. You’re telling me you wouldn’t spend four minutes to save two thousand stelmarks.

GALIT: I am telling you that the four minutes you spend reading ingredient panels is four minutes you are not spending being alive.

BARTUK: Being alive is expensive. That’s why I read the panels.

GALIT: You know what I did last week? I walked into a shop, I saw a toothpaste that had a drawing of a mint leaf on it, and I thought, “I like mint.” I bought it. I have been brushing my teeth with it. It’s fine. I have spent zero seconds thinking about it since, which means I am already winning.

BARTUK: Winning what, exactly?

GALIT: The time war. Every second you spend optimising something that doesn’t matter is a second you’ve donated to the void. You’re not saving money, Bartuk. You’re spending your life in four-minute increments and getting two stelmarks back each time. That’s a terrible exchange rate.

BARTUK: And every stelmark you waste because you couldn’t be bothered to look at a shelf for ten seconds is a stelmark someone else earned by looking. You haven’t escaped the system. You’ve just decided to be bad at it.

GALIT: I haven’t decided to be bad at it. I’ve decided it’s not worth being good at. There’s a difference.

BARTUK: Is there? Because from where I’m standing, someone who refuses to engage with a system they’re trapped in anyway is not making a philosophical statement. They’re just losing.

GALIT: I bought the wrong bread last Tuesday.

BARTUK: What?

GALIT: I bought the wrong bread. I wanted the dark rye, I grabbed the light rye, I didn’t notice until I got home. You know what I did? I ate the light rye. It was fine. I didn’t go back to the shop. I didn’t recalculate my bread strategy. I just ate the bread and moved on with my life, and the amount of suffering this caused me was zero.

BARTUK: The amount of suffering it caused you was one meal of bread you didn’t want.

GALIT: Which I didn’t notice! Because it turns out the gap between “the bread I wanted” and “the bread I got” is vanishingly small when you are not the kind of person who catalogues bread outcomes.

BARTUK: You are describing carelessness and calling it freedom.

GALIT: You are describing obsession and calling it competence.

BARTUK: Galit, I have watched you punch a wall because a Council vote went the wrong way. You threw a chair at Aleksander last month because he used the word “incremental.” You are not a person who doesn’t care about outcomes. You are a person who cares enormously about outcomes, but only when the outcome involves something that makes you feel righteous. The toothpaste doesn’t make you feel anything, so you ignore it. The Council vote makes you furious, so you break furniture. That’s not philosophy. That’s just impulse with a dress code.

GALIT: That is completely unfair.

BARTUK: It is completely accurate.

GALIT: The chair was already broken.

BARTUK: The chair was not already broken. I paid for that chair. Fourteen stelmarks. Which I know because I checked the price before I bought it, because that is what a person does when they are spending resources they earned.

GALIT: Fine. Fine. You want to know the real difference between us? You would never punch a wall. You would never throw a chair. You would sit in that chair, calmly, while the Council voted to do something monstrous, and you would calculate the cost of objecting versus the cost of compliance, and you would comply, and you would go home, and you would brush your teeth with your optimally selected toothpaste, and you would sleep perfectly well. And in the morning, you would tell yourself you were being practical.

BARTUK: Yes.

GALIT: That doesn’t bother you?

BARTUK: What bothers me is waste. What you’re describing is efficiency. I don’t break things I can’t fix. I don’t spend capital (emotional, political, financial) on gestures that accomplish nothing. When I object, I object in ways that have a chance of working, which means I object rarely, and I prepare first, and I pick my moment. You object constantly, at maximum volume, with no preparation, and then you are confused when nothing changes except the furniture budget.

GALIT: The wall I punched was in Hadria’s quarters. She didn’t mind.

BARTUK: Hadria minded.

GALIT: Hadria said she didn’t mind.

BARTUK: Hadria says a lot of things. I know what the repair cost. Want to guess?

GALIT: I absolutely do not want to guess. I want you to understand that there are things more important than what they cost. That sometimes the right response to an injustice is immediate, is loud, is physical, and the fact that it breaks something is not a bug, it’s the whole point. The breaking is the message. “This matters enough to break things over.” You never send that message. You send invoices.

BARTUK: Invoices get paid. Messages get ignored.

GALIT: Not always.

BARTUK: Often enough.

GALIT: (long pause) You are genuinely the most depressing person I know, and I know Abbas.

BARTUK: Abbas is depressing for theological reasons. I am depressing because I’m right. There’s a difference. You claimed earlier there was a difference between deciding to be bad at something and not caring. I’ll extend you the same courtesy. There is a difference between being depressing and being correct. I am both. You are neither. But your toothpaste is fine.

GALIT: My toothpaste is great, actually. It has a mint leaf on it.

BARTUK: It has sodium lauryl sulphate and a mint leaf on it. Mine doesn’t. Mine cost less. But you’re happy, and I have two thousand stelmarks. Shall we call it even?

GALIT: We will never be even. That’s the whole problem with people like you. You think everything resolves into a ledger.

BARTUK: Everything does resolve into a ledger. The question is whether you’re the one holding the pen.

Inside View vs Outside View

Everyone who’s played a serious MMO has had this conversation. Someone asks what you did last weekend, and you say you wiped on the same boss for nine hours, and they give you that look. The look that says: nine hours? On a video game? Doing the same thing? And you try to explain what it’s like to coordinate twenty-five people in a fight where one person mistiming a cooldown by a quarter second ends the run, and you watch their face, and you realize you cannot get there from here. You have seen something they have not. And no amount of describing it is going to close the gap.

This is the inside view. Not in the Kahneman sense (though we’ll get there), but in the broader sense that I think is one of the most underexplored binaries in how people think about the world. The inside view is what something feels like from the inside. The texture of an experience. The logic that makes sense when you’re standing in the middle of it, even if it makes no sense from anywhere else. The outside view is what something looks like when you can see all the edges.

Your friend is in a cult. You tell them it’s a cult. They tell you that you’d have to be there to understand, that the leader has this presence, that the community is unlike anything you’ve experienced. And the worst part? They’re not entirely wrong. You really would have to be there to understand what the experience is like. The problem is that understanding what the experience is like doesn’t help you evaluate whether the experience is good.

This comes up everywhere. An entrepreneur who can’t explain why they love building companies, even though they freely admit it destroyed their savings and their health and their twenties. A soldier who says the war was the worst thing that ever happened to them and also the only time they felt truly alive. A chess player who can’t explain why they play twelve hours a day. A musician who keeps gigging even though the pay is insulting, because “you’d have to hear the room to understand.”

From the outside, you see: destroyed sleep, destroyed marriage, trauma, wasted time, burning money. From the inside, they see: something that can’t be communicated, something felt, something that justifies the cost by existing.

The outside view is obsessed with measurement. How long does it take? What does it cost? What’s the success rate? These are the questions of someone standing on the perimeter of an experience, clipboard in hand. And those questions are useful, often decisively so. Most cults are, in fact, bad. Most startups do, in fact, fail. The base rates are the base rates.

But the inside view isn’t just delusion. The inside view contains information the outside view cannot access. What it is like to be in the experience. The felt quality of it. And that felt quality is, sometimes, the entire point. You cannot evaluate a roller coaster from its blueprints.

The tension isn’t that one of these is right. It’s that they are looking at different things, and they each believe they’re looking at the whole thing.

Outside viewers assume that the inside viewer is simply failing to see what they see. “If you could just step back, you’d realize…” But the inside viewer has often already seen the outside view. They know the base rates. They know their relationship looks bad on paper. They’re not missing the outside view. They’re overriding it, because the inside view contains something the outside view can’t capture, and they’ve decided (correctly or not) that the uncapturable thing is more important than the measurable thing.

Inside viewers, meanwhile, assume the outside viewer is simply too distant to see what they see. “You’d have to experience it.” And that’s often true. But it’s also a tautology that applies to everything, including experiences that are bad in ways the insider can’t see. You’d have to be inside a pyramid scheme to understand the hope. That doesn’t make it a good investment.

I think what makes this binary interesting is that both sides are correctly identifying a real limitation of the other side, and then using that correct identification to dismiss the other side entirely. The outside viewer is right that feelings are not evidence. The inside viewer is right that some things can only be known by feeling them. And neither of these truths cancels the other out.

They’re two different instruments that measure two different things. And most of the time, we use whichever one flatters us.

Let’s start with the most obvious version of this split and then work outward into the weirder ones.

Daniel Kahneman popularized the terms “inside view” and “outside view” in the context of planning. The inside view is when you estimate how long a project will take by thinking about the specific details of your project: the steps involved, the resources you have, the challenges you foresee. The outside view is when you look at the base rate of similar projects and use that as your estimate. Kahneman’s point was that the inside view consistently produces overoptimistic estimates, because people systematically neglect the base rates while overweighting their personal model of the situation.

This is well-established and largely correct. If you ask a homeowner how long their kitchen renovation will take, they’ll tell you six weeks. The outside view (looking at how long kitchen renovations actually take for everyone) says twelve. The homeowner isn’t stupid. They’ve thought about their specific situation. They just can’t see the forces that will act on them the same way those forces act on everyone else, because from the inside, those forces are invisible.

But I want to push this further than planning estimates, because I think the inside/outside split is a much deeper structure than Kahneman needed it for.

Think about what an experience looks like to someone who’s in it versus someone who’s watching it. A relationship. For your friends, it’s a story with observable inputs and outputs. He said this, she did that, look at the pattern. For the person inside it, there’s an entire atmosphere that your friends cannot access. The way someone laughs when they’re actually relaxed. The particular silence after a fight where you can feel things reassembling. Your friends see the data points. You live in the weather between the data points.

When you line up enough of these examples, a structural pattern emerges. The inside view, across wildly different domains, keeps emphasizing the same thing: the power of emotional experience, the way something hits you, a quality that cannot be transmitted. The outside view keeps emphasizing something entirely different: the amount of effort involved, the time it took, the resources consumed, the comparison to alternatives.

A non-runner asks a marathoner: “How can you train five hours a day for years? I could never do that.” The marathoner says: “I don’t know. I just did it.” This exchange is diagnostic. The outside viewer is staring at the cost (five hours, every day, for years) and can’t get past it. That’s all they can see. The inside viewer doesn’t see the cost at all. They see the feeling. The morning air and the rhythm and the way their body feels at mile eight and the fact that running is just what they do now. The time is invisible to them in the same way the feeling is invisible to the outside viewer.

This suggests that the inside and outside views aren’t just different estimates of the same quantity. They are different quantities. They’re measuring different things and calling both measurements “the experience.”

Which makes it tempting to conclude that the inside view is simply richer, more complete, more real. And it is richer, in the sense that the inside viewer usually already has the outside view and has dismissed it. The marathoner knows about the five hours a day. He just doesn’t care. The information runs in one direction: experiencing something from the inside would transform an outside viewer, but showing an inside viewer the statistics tells them nothing they don’t already know.

But “richer” isn’t the same as “more accurate.” The inside view’s richness is specifically the richness of being unable to see yourself from the outside. The founder who loves her startup despite what it’s done to her health and her bank account is having a real experience, but she is also the last person who can accurately assess whether the startup has been good for her in total. The cult member who feels the leader’s charisma is feeling something real, but they are also the last person who can assess whether the cult is harmful. The inside view is like looking at a stained glass window from inside the cathedral. The colors are extraordinary. But you can’t see the building.

Here is where the binary gets structural.

Outside viewers tend to be right about categories. “Most kitchen renovations take twelve weeks” is true. “Most cults are bad” is true. “Most startups fail” is true. The base rates are real, and the inside view systematically ignores them, because from the inside, your situation feels specific, feels unique, feels like the kind that will beat the odds.

Inside viewers tend to be right about texture. What it actually feels like to do the renovation, to be in the community, to build the company. The granular, immediate reality that makes people keep doing things that look insane from outside. The outside view can tell you the probability. It cannot tell you the phenomenology.

So which one do you trust? It depends on what kind of question you’re asking.

If the question is “Will this work?” or “Is this a good idea?” the outside view is almost always more reliable. Your kitchen renovation will take twelve weeks. Your relationship that looks like a disaster probably is one. The experience of being inside is precisely what makes it hard to evaluate the experience.

But if the question is “What is this like?” or “Why do people do this?” the outside view is almost useless. You cannot understand religious devotion from its success rate. You cannot understand why someone stays in a hard marriage by tallying their complaints.

So far I’ve been talking about the inside view as if it’s always positive. The stories we tell about this binary are almost always stories of someone having a wonderful experience that outsiders can’t appreciate. The marathoner’s bliss. The monk’s silence. The founder’s vision. And that’s a real pattern. But it’s only half the picture, and I think the other half is more interesting.

Sometimes the inside view is that something was terrible.

Let’s talk about LARPs.

LARP (live-action roleplaying) is a good test case because it provokes both kinds of incommunicability at once. Before you’ve done one, the outside view is baffled at the effort: you’re going to spend an entire weekend pretending to be an elf? You made a costume? You drove three hours? You paid money? The outside viewer can only see the cost, the same way they can only see the marathoner’s five hours a day. And the insider can only say, as always, “you’d have to be there.”

But here’s what happens when the inside view goes negative. Imagine you’ve played a weekend LARP, forty hours of immersive gameplay, and it was miserable. Not boring. Miserable. The game was structured so that your character had no meaningful choices. Every decision was predetermined by players who had been in the campaign longer and had already carved up the political landscape. You spent two days being led from scene to scene where you were told what was happening to you.

You went to bed Saturday night in a cabin that smelled like mildew, feeling humiliated and trapped in a way that wasn’t fun-trapped (the way a horror game is supposed to make you feel) but actually-trapped, the way a bad job makes you feel, except you were doing it for fun, except it wasn’t fun, except you can’t even leave because you carpooled with someone who’s having the time of their life.

That experience is real. It happened to you. You know what it was like from the inside, and the inside was awful.

Now try to explain it to someone who has only the outside view.

“It was the worst weekend of my life.”

“Okay, so you were bored for a weekend playing pretend. That’s… bad?”

You see the problem. From the outside, the LARP looks the same whether it was incredible or devastating: a bunch of people in costumes in the woods for two days. The outside metrics are identical. Same amount of time. Same amount of effort. Same hobby. The outside viewer can’t tell the difference between a transcendent experience and a traumatic one because the outside view measures the wrong things.

And you can’t communicate what was actually bad about it, because what was bad was an inside-view phenomenon: the feeling of powerlessness, the slow realization that the game’s structure had decided your experience before you arrived, the particular quality of performing enthusiasm for something that was crushing you. You can describe these things in words, but the words arrive at the listener as outside-view data points: “she felt powerless, she didn’t like the structure, she was unhappy.” And outside-view data points about a LARP register as trivial. You were play-acting and it wasn’t fun. So what?

This is where the binary becomes genuinely cruel. You have been inside something. You know something about it that can only be known from the inside. And the thing you know is: don’t do this. This specific thing, structured this way, will hurt you. And you cannot communicate this knowledge to someone who hasn’t been inside, because the knowledge is in a format that only the inside view can read, and the person you’re trying to warn has only the outside view.

The positive version of this problem (the marathoner who can’t explain the bliss) is, at worst, frustrating. You can’t share the joy, but the joy still exists. The negative version is a tragedy. You can’t share the warning. The person walks in. They get hurt. They come out and try to warn the next person. The next person can’t hear it either.

This is the structure underneath a huge number of real-world failures of communication. Why someone who’s been through a brutal residency can’t convey to pre-med students what it will actually cost them. Why people who’ve left a religion can’t adequately warn people who are joining. Why someone who survived a bad marriage can sit across from a friend who’s about to make the same mistake and find that every true thing they say lands as an exaggeration or an irrelevance.

The listener isn’t dismissing you. They literally lack the instrument that would let them receive the information. And you can’t give them that instrument without giving them the experience, which is exactly what you’re trying to prevent.

There’s another dimension here that I think gets overlooked. People’s inside views of an experience change over time, and the former inside view becomes just as inaccessible as any outside view. This is one of the strangest features of the inside/outside split: you can be an outsider to your own past.

Think about that for a second.

Everyone has had the experience of looking back at something (a relationship, a phase, a job, a belief) and thinking “what was I doing?” Your current self looks at your past self with the same bafflement that an outside observer would have had at the time. You can remember that you felt something, but you can no longer feel it. The intensity, the texture, the thing that made it all make sense in the moment, is gone. You are now an outside viewer of your own past inside view, and you’re making the same mistake every outside viewer makes: you’re evaluating the decisions by the data points and missing the weather between them.

This is why people in their forties look at their twenties with a mixture of fondness and horror. They remember the data points (the bad apartment, the worse boyfriend, the job that paid nothing). They can no longer access the atmosphere that made all those data points feel right, feel alive, feel like the only possible life.

The inside view has a shelf life. It’s only available while you’re inside.

And this creates a peculiar epistemological problem. If the inside view expires, and the outside view was never sufficient to begin with, then no one, at any time, has a complete picture of any experience. You either have the inside view (which is vivid but can’t see itself) or the outside view (which can see everything except the thing that matters most), and you switch between them over time, and neither position has access to what the other one knows.

This might explain something about nostalgia. Nostalgia is the emotional residue of a dead inside view. You can’t reenter the experience, but you can feel the ghost of what it was like to be inside it. The feeling is real, but what it’s a feeling of is gone. Nostalgia is an inside view stripped of all its information, reduced to pure texture with no content. Which is why it’s so powerful and so useless: it gives you the sensation of understanding without anything left to understand.

One last pattern: people are almost always inside viewers about their own experiences and outside viewers about everyone else’s. You understand your own relationship from the inside, and your friend’s from the outside. You know that your own career choices are driven by deep meaning, and you wonder why your colleague stays in a job that’s so obviously wrong for them. Your inside view is your life. Everyone else’s inside view is their bias.

Which means the binary isn’t really a disagreement between two kinds of people. It’s a disagreement between two positions that the same person occupies at different times, depending on whether they’re the one having the experience or the one watching.

• Five bullet examples of inside vs outside view –

1. Grief counseling. A widow knows that grief counselors with printed pamphlets and five-stage models are, statistically, helpful for most bereaved people. She also knows that no pamphlet has ever described what it is like to reach across the bed at 3am and find it empty. The outside view built the pamphlet. The inside view is why the pamphlet will sit unread on the kitchen counter for six weeks before she touches it, and then she will cry, and then it will help. Both the delay and the help are correct responses.
2. Speedrunning. From outside, a person plays the same video game thousands of times, resetting after tiny mistakes, optimizing a run down to the frame. The obvious question: why? The inside answer involves something about the flow state at the edge of human reaction time, the way the game stops being a game and becomes a conversation between your hands and the code, the beauty of a route that no one has ever executed before. The outside view sees a person who has played Celeste for four thousand hours. The inside view has no concept of four thousand hours. There is only this run.
3. Jury deliberation. The legal system is specifically designed to be an outside view: rules of evidence, precedent, burden of proof, all structured to prevent the inside view (emotional reaction, gut feeling, personal identification with the defendant) from determining the verdict. And yet the jury itself is an inside view technology. Twelve people in a room making a felt judgment. The system doesn’t trust the inside view enough to let it run without rules, and doesn’t trust the outside view enough to replace the jury with an algorithm.

There is a painting by Caravaggio, the Supper at Emmaus, in which two disciples sit at a table with a man they have been walking with all day and suddenly realize he is Christ. The moment of recognition is the painting. The meal and the road and the theological argument for resurrection are elsewhere, in other paintings, in scripture. This painting is only the instant where the inside view (this is a fellow traveler) collapses and a new inside view (this is God) opens like a door in a wall that had been, until this second, solid.

Caravaggio understood something the epistemologists keep missing.

Every inside view is a room you cannot see the shape of while you are standing in it. You know the furniture. You know the light. You know the particular warmth of the air and the sound the floor makes when you shift your weight. What you do not know is that it is a room, that it has walls, that those walls are visible from the hallway.

The mystics knew. They had a word for it, or rather they had the absence of a word for it. The via negativa. You cannot say what God is. You can only say what God is not. Every positive statement is an outside view: God is good, God is great, God is merciful. These are descriptions from the hallway. The mystics wanted to be in the room. And in the room, all the descriptions fall away, because descriptions are the technology of people who are standing outside.

This is why the inside view speaks in broken sentences. Why the soldier says “you had to be there” and means it as a complete philosophical statement, as full and final as a syllogism. Why the lover says “I can’t explain it” and is telling you something more precise than any explanation could be. They are reporting from inside the room. The room has no labels on its walls.

And the outside view, patient, sensible, actuarial, stands in the hallway with its clipboard and its base rates, and it is not wrong. The hallway is real. The measurements are accurate. The room is the shape the measurements say it is.

But the hallway has never been warm.

The hallway has never been the place where you reach across the bed. The hallway knows the name of every room in the building and has never lived in any of them.

What is the relationship between these two darknesses? The darkness of being inside (not seeing the walls) and the darkness of being outside (not feeling the air) are not the same darkness. They are not even symmetrical. One is the darkness of a womb. The other is the darkness of a diagram that has not yet been filled in.

You cannot synthesize them. You cannot build a view that is simultaneously inside and outside, because the act of stepping back to see the walls is the act of leaving the room, and the act of stepping in to feel the air is the act of losing the walls. This is not a problem to be solved. It is the condition of being a creature that can both experience and reflect, but never at the same time.

The ancient Hebrews had a rule about the holy of holies. One person could enter, once a year, and only if they wore bells on their robe so the people outside could hear that they were still alive in there. The inside view, reporting back to the outside. The bells did not describe what was in the room. They only said: someone is in it, and they are still here.

That might be the most honest technology we have ever built for this problem.

A bell.

Here is someone, inside something. Here is the sound that proves it.

Listen.

Going to the Movies

Laszlo: Mira told me to watch this film. She said it changed how she thinks about memory.

Pellegrin: What’s the runtime?

Laszlo: I didn’t ask.

Pellegrin: What’s the Letterboxd score?

Laszlo: I didn’t look.

Pellegrin: Genre? Director? Year of release?

Laszlo: She said to go in blind. She said the less I know, the better it works.

Pellegrin: So your friend had an emotional experience, and on that basis alone, you are going to allocate two hours of your finite lifespan.

Laszlo: On that basis? On the basis that someone I trust told me something moved her?

Pellegrin: On the basis that someone you trust had a feeling. Feelings are not peer-reviewed, Laszlo.

Laszlo: The last film you watched, you spent forty minutes reading the Wikipedia plot summary first, then watched it, then told me it was “exactly as described.” You seemed satisfied by this. I found it the saddest thing I’d heard in a month.

Pellegrin: I was confirming the accuracy of available information. The film performed within expected parameters.

Laszlo: You confirmed that a story did what the summary said it would do. You confirmed that the map matched the territory. And you think that’s what watching a film is?

Pellegrin: I think watching a film without context is like running an experiment without a hypothesis. You’re just collecting noise.

Laszlo: You’re just filtering out everything that doesn’t match the hypothesis! The whole point is that Mira couldn’t describe what the film did to her. She tried. She used four different metaphors and then gave up and said “just watch it.” That inability to describe it is the recommendation. If she could have described it, she would have, and then I wouldn’t need to watch it.

Pellegrin: You’re arguing that the less articulable a recommendation is, the more trustworthy it becomes.

Laszlo: Yes.

Pellegrin: That is genuinely one of the worst epistemological positions I have ever encountered.

Laszlo: And yet when you solve a proof, the part you can’t explain, the leap between step seven and step eight, the intuition that told you which path to take before you could justify it logically, you call that “mathematical taste” and you venerate it.

Pellegrin: That is different.

Laszlo: How?

Pellegrin: Because after the intuition, there is a proof. The feeling leads to the proof, and the proof is what matters. The feeling is scaffolding. You remove it.

Laszlo: So you trust the feeling long enough to get the proof, and then you retroactively declare the feeling was irrelevant.

Pellegrin: I declare it was instrumental. Not terminal.

Laszlo: And if the film gives me something that can’t be rendered into a proof? If the whole value is in the feeling itself?

Pellegrin: Then the whole value is in something that can’t be verified, can’t be shared, and can’t be distinguished from a hallucination.

Laszlo: Can you distinguish the taste of wine from a hallucination of the taste of wine?

Pellegrin: If I had the brain scans, yes.

Laszlo: But from the inside.

Pellegrin: From the inside, nothing can be verified. That’s the entire problem with “the inside.”

Laszlo: It’s only a problem if verification is the only thing that matters. What if some experiences are valuable precisely because they can’t be verified? What if the unverifiability is part of what makes them transformative?

Pellegrin: Then you’ve defined “transformative” as “immune to criticism,” which is theologically convenient and epistemologically catastrophic.

Laszlo: You’re sitting here with a list of the ten highest-rated films on four different aggregator sites, and you’ve seen all of them, and you described every single one as “competent.” You have optimized your viewing so perfectly that you have removed the possibility of being surprised. Doesn’t that bother you?

Pellegrin: I am rarely surprised. This is because I do adequate research.

Laszlo: You are never surprised. This is because you have mistaken “already knowing what will happen” for understanding.

Pellegrin: …

Laszlo: Mira cried during this film. She doesn’t cry. She once watched a documentary about a dying whale and said, “Statistically, this was inevitable.” She cried at this film. I want to know what’s in there. I want to walk in with no summary and no score and no hypothesis and find out what happens when I meet it unprepared.

Pellegrin: And when it’s bad?

Laszlo: Then I will have had the experience of finding that out.

Pellegrin: You’ll have wasted two hours.

Laszlo: I’ll have spent two hours. Whether they’re wasted depends on what you think they were for.

(A pause. Pellegrin pulls out his phone.)

Pellegrin: Fine. What’s it called?

Laszlo: She wouldn’t tell me the title. She said even the title gives too much away.

Pellegrin: This is absolutely deranged.

Laszlo: She said I’d know it when I found it.

Pellegrin: That’s not how search engines work.

Laszlo: No. But it might be how films work.

(Pellegrin stares at him. Then, very slowly, puts the phone back in his pocket.)

Pellegrin: If this is longer than two hours and fifteen minutes, I’m leaving.

Laszlo: That’s fair.

Pellegrin: And I’m reading the Wikipedia summary afterward.

Laszlo: Naturally.

---

Inside View and Outside View — from the AB Categories document:

• https://en.wikipedia.org/wiki/Reference_class_forecasting — Reference class forecasting, the formal technique behind Kahneman’s outside view. Bases estimates on actual outcomes of similar past situations rather than on the specifics of the current case.

Anxious vs Avoidant

About half of adults, if you believe the surveys, have what the therapeutic industry calls an “insecure attachment style.” The other half are “secure.” The secure ones are fine. The insecure ones are subdivided into anxious (they cling) and avoidant (they flee) and sometimes fearful-avoidant (they cling and flee simultaneously, which sounds exhausting because it is). You can find out which one you are in about four minutes on any number of websites, all of which will then sell you a course on how to become secure, as though security were a product with a checkout page. Which, to be fair, is how it’s being marketed.

The clinical vocabulary comes from John Bowlby, a British psychoanalyst working in the 1950s who studied children separated from their mothers during wartime evacuation. Mary Ainsworth formalised his observations in the 1970s through what she called the Strange Situation: put a toddler in a room with toys, have the mother leave, bring in a stranger, watch what happens when the mother returns. Some children screamed and clung. Some ignored her. Some reached for her and then hit her.

The names stuck. The nuance did not.

What the popular version gets wrong is where the style lives. People speak of “having” an attachment style the way they speak of having a blood type — a fixed property inside you that explains your behaviour. But Bowlby’s actual insight was ecological. The strategy is not in the person. It is in the relationship between the person and the specific environment that made the strategy necessary. An anxiously attached child has learned, correctly, that the caregiver’s attention is available but unpredictable. The optimal response to unpredictability is vigilance. An avoidantly attached child has learned, correctly, that displays of need are met with withdrawal. The optimal response to consistent withdrawal is to stop displaying need.

Both strategies work. That is the part nobody wants to hear.

The anxious strategy secures proximity. If your caregiver is intermittently responsive (sometimes attuned, sometimes distracted, sometimes overwhelmed), then escalating your distress signal is rational. You cry louder because crying louder sometimes works. You monitor their face because sometimes you can catch the departure before it happens. Expensive, yes. Burns enormous cognitive resources on threat-detection. But it solves an environment where attention is scarce and randomly distributed, and persistent noisy search is the optimal foraging strategy. Think of it as playing a slot machine that sometimes pays out affection. The house always wins, but you can’t stop pulling the lever because it paid out once at a formative moment and your entire nervous system took notes.

The avoidant strategy secures autonomy. If your caregiver reliably moves away when you display need (not cruelly, necessarily; the parent may simply be uncomfortable with emotion, or overwhelmed, or raised by someone who was), then suppressing the signal is equally rational. You learn to self-soothe not because you don’t need comfort but because requesting it makes the person you need retreat further. Also expensive. Also rational. It solves the problem it was designed to solve: maintaining a kind of proximity by not triggering the other person’s withdrawal reflex. A controlled distance that keeps the caregiver in the room. Like learning to approach a feral cat — you succeed by pretending you don’t want to.

Now here is the complication that the self-help literature almost never addresses. Both of these strategies were designed for a specific caregiver in a specific household. They are local adaptations. Brilliant ones. A child with no power and no vocabulary figured out, through pure empirical observation, how to keep a much larger person close enough to survive. That is not pathology. That is engineering under constraint.

But the strategy travels. It outlives its context. The child grows up and enters adult relationships carrying a solution to a problem that no longer exists in its original form, and the solution, which was perfectly calibrated for one specific human being, is now being applied to every human being. The anxious adult walks into a relationship carrying a finely tuned detection system for abandonment signals, and their new partner’s entirely innocent trip to the shop for milk triggers the same cascade that their mother’s unpredictable absences triggered twenty-five years ago. The avoidant adult walks into a relationship carrying a finely tuned suppression system for emotional display, and their new partner’s entirely reasonable request to talk about feelings triggers the same shutdown that their father’s discomfort with tears triggered in 1998.

This is where the word “need” splits.

When the anxious person says “I need you,” the word means: your presence is the evidence that I am safe. Your absence is indistinguishable from danger. I am asking you to provide the evidence. When the avoidant person hears those same three words, “need” means something closer to: you are about to be consumed. Someone is making demands that will feel like drowning. The walls are closing in.

Same words. Opposite semantic fields. The anxious person has made a request for proximity. The avoidant person has received a threat to autonomy.

Reverse it. When the avoidant person says “I need space,” the word means: my regulatory system is overwhelmed and requires solitude to return to baseline. This is mechanical, not personal. When the anxious person hears it, “need space” means: the connection is dissolving. The silence that is about to happen is the same silence your nervous system has been scanning for since before you could speak.

Therapists call this the pursue-withdraw cycle, which makes it sound tidy. It is not tidy. It is two people locked in a dance where each partner’s solution to their distress is the other partner’s trigger. The anxious partner pursues because pursuit is how they learned to maintain connection. The avoidant partner withdraws because withdrawal is how they learned to maintain connection. Each one, from inside their own logic, is doing the right thing. Each one, from inside the other’s logic, is doing the worst possible thing. And the cycle accelerates because each escalation confirms the other’s worst belief: see, they really are leaving. See, they really are suffocating me.

This pattern shows up in so many domains once you learn to see it. In trade negotiations, where one party’s insistence on detailed contractual protections reads, to the other party, as distrust. In game design, where accessibility features that help anxious players (tutorials, hand-holding, waypoints) drive away avoidant ones (who want to explore, get lost, figure it out alone), and the developers cannot satisfy both without fundamentally splitting the experience. In institutional design, where high-documentation cultures and high-trust cultures cannot occupy the same organisation without each reading the other’s operating style as either recklessness or paranoia.

And it shows up in the research itself. Ainsworth’s original Strange Situation study was conducted on fifty-six white, middle-class American families. When the experiment was replicated in Japan, researchers found far more “anxious-resistant” children. In Germany, far more “avoidant” children. For years, the standard interpretation was that these cultures were producing more insecurely attached children. A more careful reading, which took decades to gain traction, noticed the obvious: Japanese infants were rarely separated from their mothers and almost never left alone with strangers. German infants were socialised toward independence from an early age. The experiment wasn’t measuring attachment security. It was measuring how children performed under conditions that meant radically different things depending on where they’d grown up, and scoring them against one rubric. An American one.

Jerome Kagan, the Harvard developmental psychologist who spent decades challenging the entire framework, put it bluntly: temperament and social class predict outcomes far more reliably than attachment classification. He may have overstated the case. But his core point stands. The Strange Situation measures what a child does in a specific room with a specific stranger under specific conditions of mild stress. It does not measure what the child is. And the gap between “does” and “is” is the gap the entire self-help industry has fallen into, and it is making money on every floor of the descent.

So. Layer this up. You have a strategy that was locally rational, that has outlived its context, that is being applied indiscriminately to new partners, that encodes itself as identity rather than behaviour, that is measured by instruments carrying the cultural assumptions of one society, and that, when two mismatched strategies collide, produces a cycle in which each person’s attempt to solve the problem is the other person’s experience of the problem. That is the pile. That is what we are actually looking at when we say “anxious” and “avoidant.”

The Greeks, who were not idiots about the obligations between strangers, encoded something useful in the rituals of xenia. The guest-host relationship, protected by Zeus Xenios, required that when a stranger arrived at your door, you fed them first. Bathed them. Gave them your best room. Only after the guest had eaten were you permitted to ask their name and origin. The sequence was not arbitrary politeness. It encoded a principle worth remembering: care precedes understanding. You cannot ask someone who they are until you have demonstrated, through material action, that their answer will not change how you treat them. The modern attachment literature inverts this. It wants the diagnosis first. Anxious? Avoidant? Disorganised? Good, now I know how to handle you. Identify, then manage. The xenia sequence — tend first, ask later — is not just kinder. It is structurally more honest about the fact that the person standing before you is opaque, and will remain opaque, and that your categories for them are your categories, not theirs.

There is a Langovin word, tükrözés, that means the act of constructing a mirror out of another person’s face. Not looking into a mirror. Building one. You reshape someone else’s expressions until they show you what you need to see. Every anxious person does this: scans their partner’s face for evidence of departure, and in scanning, produces the tension that makes the departure more likely. Every avoidant person does a version too: scans for evidence of encroachment, and in withdrawing from the scan, produces the desperation that confirms the encroachment. Both are building mirrors. Neither is seeing a face. And there is a Sumerian proverb from Nippur, roughly four thousand years old, that says you can have a lord, you can have a king, but the man to fear is the tax collector — the one who actually enters your house, inspects your stores, stands in your doorway and counts what you have. The person who gets close enough to assess you. Proximity is where the assessment happens, and assessment is where the wound lives, and the anxious person and the avoidant person both know this in their bodies, in the architecture of their breathing, in the way their chest tightens when a partner’s footsteps approach the bedroom door, and the question was never whether to let someone in, the question was always what it costs to be seen, and the two answers to that question — I will make sure you see me, I will make sure you don’t — are two translations of the same sentence, which is that being known is dangerous, was always dangerous, that love requires the thing that the body has spent a lifetime learning to prevent, and the body is not wrong, it was never wrong, it solved the problem it was given with the only tools it had, and the tragedy is not that the solution is broken but that it worked, it worked so well that it became the only language available, and now there is a person standing at the door who speaks a different one, and you cannot learn it fast enough, and they cannot learn yours, and the space between the two of you where the learning would have to happen is also the space where all the danger lives, and the danger and the love are in the same place, they have always been in the same place, and knowing this changes nothing, and knowing this changes everything, and that is what the word means, the one that has no equivalent in Common: the act of building a mirror out of a face, and discovering that the face was trying to do the same thing to you, and that neither of you will ever see what the other one was looking for.