Chapter 4: You Had to Be There
On October 8, 1993, Ted Danson walked into the Friars Club Roast of Whoopi Goldberg in full blackface.
Not metaphorical blackface. Burnt cork, white lips, the entire nineteenth-century theatrical vocabulary of racial caricature, the imagery that had been used for nearly a century to reduce Black Americans to a comic type. He wore it through a forty-minute set that included the N-word more than twelve times, explicit jokes about the sexual relationship he and Goldberg were in, watermelon jokes, and jokes that referenced the history of American racism in the most direct possible terms. Approximately three thousand people were in the room.
Roger Ebert was there. He was reviewing the event, which was his job. He watched audience members hide their faces in their hands. His subsequent review described the evening as the end of the Friars Club roast tradition — that what had happened in that room had been too much, had gone somewhere the tradition could not survive having gone.
Montel Williams was there. He turned his back to the stage. When the evening ended, he went home and sent the Friars Club a telegram resigning his membership. The telegram read: "I was confused as to whether or not I was at a Friars event or at a rally for the KKK and Aryan Nation."
Beverly Johnson was there. She said afterward: "If you can't see the humor at a place where there's supposed to be over-the-line jokes, then there's something really wrong."
Whoopi Goldberg was laughing. She had co-written portions of the material.
Everyone in this account was in the same room. They were not in the same circle. And the gap between those two facts — same room, different circles — is not a scandal requiring adjudication. It is a mechanism requiring examination. Because what happened that night in New York is structurally identical to what kept a team of surgeons functional in a freezing Korean tent in 1951, identical to what a council of fourth-century Egyptian monks enacted with a leaking jug, identical to what two clans in Mande-speaking West Africa have been enacting at market crossings for the better part of a thousand years. The mechanism is the same. The content is maximally different. And understanding the mechanism is the only way to understand why both of those things can be simultaneously true.
Don't decide yet who was right. The chapter's question isn't who was right. The chapter's question is: what, exactly, is a circle?
Robin Dunbar's team at Oxford figured out one way to measure this by putting people's arms in things that hurt.
The experiment sounds like the kind of hazing that produces lawsuits, but it is in fact rigorous neuroscience. Participants submerged a forearm into a sleeve chilled to negative sixteen degrees Celsius, held it there as long as they could manage, then watched either a comedy video or a nature documentary in small groups, and held it there again. The comedy-watchers could hold out significantly longer the second time. The documentary-watchers could not. Pain tolerance is a well-established proxy for endorphin release — the body's own opiate system — and the implication is straightforward: group laughter triggers that system in a way that is measurable, replicable, and distinct from the effect of simply being in a good mood. Dunbar's team ran the experiment a second time with blood-pressure cuffs inflated to 260-280 mmHg, which produces a different kind of insistent discomfort. Same result. They ran it a third time at live comedy and drama shows, strapping performers and audience members into a ski-wall exercise — legs bent ninety degrees, back flat against the wall — before and after the show. Comedy audiences showed elevated pain thresholds; drama audiences did not. And here is the finding that matters most: positive affect alone — smiling, feeling good, the general warm glow of an enjoyable evening without laughing — did not produce the effect. It was specifically the laughing. Not the mood. The physical act of convulsing, which takes over the diaphragm and makes breathing temporarily impossible and does something to the body that no amount of merely feeling pleased can replicate.
Dunbar's evolutionary interpretation is that laughter replaced mutual grooming as the primary social bonding mechanism once human groups grew too large for one-on-one physical contact to do the job. A chimpanzee can groom one other chimp at a time. A comedian — or, in the ancestral environment, a storyteller, a clown, a trickster — can trigger the same endorphin cascade in a hundred people simultaneously. The scalability is the point. Laughter is social grooming operating at broadcast range.
Six years after Dunbar's arm-freezing study, a Finnish team made the same argument in a PET scanner. They recruited twelve men — close friends, ages twenty to thirty-two — and gave them a radiotracer that competes with the body's own endorphins for mu-opioid receptor binding in the brain. Half watched a thirty-minute comedy playlist with their friends in the scanner. The other half watched alone. Social laughter produced measurable changes in receptor binding in the thalamus, the caudate nucleus, the putamen, the insula, the anterior cingulate cortex, and the frontal cortex — the regions associated with social bonding, reward, emotional regulation, and the integration of sensory experience with meaning. More laughing produced more binding change, in a relationship that was direct and linear. More friends in the room produced more laughing. The researchers were watching friendship happen at the molecular level. This is the event that you have been calling "a good time," their data said, and it is chemistry, and it is specifically the chemistry of being together.
Rod Martin at the University of Western Ontario spent years developing what became the Humor Styles Questionnaire, a tool for measuring individual differences in how people use humor. Validated now in more than twenty-eight countries and across dozens of languages, it identifies four styles: affiliative humor (making others laugh, easing tension, reinforcing connection), self-enhancing humor (finding things funny even alone, as a coping mechanism), aggressive humor (ridiculing, sarcasm, the joke as weapon), and self-defeating humor (being the butt of your own jokes to gain social approval). The affiliative subscale — "I laugh and joke a lot with my friends," "I enjoy making people laugh" — predicts positive affect, self-esteem, life satisfaction, intimacy, and perceived social support. The aggressive subscale predicts hostility, neuroticism, and anxiety. They are negatively correlated: people who score high on affiliative humor tend to score low on aggressive humor. And the finding that surprised even Martin: there is no gender difference in affiliative humor scores. Men score higher on aggressive humor. Men score higher on self-defeating humor. But the impulse to bond through laughter, to reach for another person with a joke, to make someone else's day lighter as a primary social mode — this appears to be equally distributed between men and women. Which suggests it is, in some sense, the default. The background system. The baseline from which the other styles diverge.
Laura Kurtz and Sara Algoe brought seventy-one couples into a laboratory at the University of North Carolina and asked them to have a four-minute conversation about how they had met. They recorded everything — audio and video — and then had coders go through 1,399 laugh events, frame by frame, coding each one for whether it was shared (both partners laughing at the same moment) or unshared (one partner laughing while the other did not). Of those 1,399 events, 256 were simultaneous, with an average overlap of 1.49 seconds. Shared laughter predicted closeness. Shared laughter predicted social support. Unshared laughter — one partner laughing while the other did not — predicted lower closeness. Not zero effect. Negative effect. An asymmetric laugh is a small rejection. The body keeps score.
What the experiment is measuring, in behavior, is the same event the Finnish PET scanner was measuring in chemistry: two systems synchronizing. The laugh is the synchronization made audible. The shared laugh is the bond made real. And the bond is not metaphorical — it is the opioid system releasing, measurable in brain chemistry and in the secondary effects of that release (pain tolerance, blood pressure, capacity for sustained discomfort), measurable in behavioral proxies like closeness and perceived support.
Now put three thousand people in a room, all carrying different histories, all synchronized with different communities, and see what happens when someone makes a joke that references a history only some of them share.
The circle is not the room. The circle is never the room. The circle is the set of people whose nervous systems are primed by shared experience to synchronize on the same reference point. You can be in the same room and in completely different circles. You can be in different countries and in the same circle. What the Friars Club room on October 8, 1993 contained was not three thousand people — it contained several overlapping circles that happened to share a physical space, and a joke that landed differently depending on which circle was receiving it.
In Mali, Burkina Faso, the Gambia, and most of the Mande-speaking world, certain surnames carry an ancient and specific obligation. A Camara and a Kouyaté, meeting at a market, are expected to begin a ritual of insults: accusations of gluttony, disgraceful ancestry, legendary laziness, livestock that are an embarrassment to anyone with a functioning sense of shame. In any other context, these would be fighting words. Between a Camara and a Kouyaté, they are the expression of a bond that predates European contact by centuries.
The practice is called sanankuya. Its origins are disputed in the details, but its age is not: the tradition was reportedly codified as a formal civic institution in the Kurukan Fuga, the oral constitution of the Mali Empire, by Sundiata Keita around 1236 CE. The Mandinka, Fula, Wolof, Jola, Serahule, Serer, and Bambara peoples are all enmeshed in overlapping joking-kinship networks. The banter tends to focus on food — accusations of enormous appetites, of ancestors who ate their way through other families' harvests — though it extends to lineage, marriage practices, and the relative merits of various kinds of livestock. One documented clan pairing, Fofanna and Jaiteh, traces its joking bond to a founding story in which the ancestor Fofanna cut meat from his own leg to feed his teacher Jaiteh during a long and hungry journey. The obligation of mutual mockery encodes an obligation of mutual support. The insult is the proof of the bond.
Colonial-era French administrators and British political officers encountered these exchanges in market squares across West Africa and had to be repeatedly corrected. A French official records in his dispatches two men in what he describes as a heated argument — accusations, retorts, escalating volume — until his interpreter leans over to explain that this is friendship. That these two men are demonstrating their bond. The identical exchange — the same words, the same volume, the same gestures — reads as social bonding or imminent violence depending entirely on whether you carry the tradition that makes it legible. The circle is invisible to people outside it. From outside, you see two men insulting each other in a market. From inside, you see eight centuries of civic architecture.
Researchers have observed sanankuya functioning during politically tense moments in recent decades — election periods in the Gambia, in which joking scripts were deployed across clan lines to defuse tension that might otherwise have become something harder to defuse. The ritual insult as a pressure valve. But it only relieves pressure for the people who are inside the tradition. For someone outside it, encountering the same exchange, it raises the pressure.
There is something worth lingering on in the founding story of the Fofanna-Jaiteh bond: the ancestor who cut meat from his own leg to feed his companion on a long journey. The mockery this story generates — the ritual accusations of gluttony, the jokes about enormous appetites — is not random transgression selected for maximum shock. It inverts the original act. The founding act was self-sacrifice for the other's hunger; the ritual joke is an accusation of consuming too much for oneself. The insult is the inside-out version of the love. To call your joking-kinship partner a glutton, in a tradition whose origin story is about feeding a starving companion with your own flesh, is to say something about the depth of the obligation that straight sentiment could not carry. The joke carries more than the sentiment because it arrives through the inversion, the risk, the fact that from outside it sounds like an insult and from inside it sounds like everything the tradition has been carrying for eight centuries.
This is what the mechanism is always doing at its most sophisticated. The play signals that surround the insult — the tradition that makes the insult legible as affection — are not decoration. They are the substance. The same words that would be a fight from one mouth are a bond from another, and the difference is made entirely by the invisible infrastructure of shared history and shared understanding that neither party needs to name because both already carry it.
Sometime in the fourth or fifth century, in the Egyptian desert, a council of monks assembled to judge one of their number for a sin. These were the Desert Fathers — the Christian contemplatives who had retreated to the Egyptian wilderness to pursue their practice under conditions of extreme austerity. The tradition was serious about judgment and serious about sin and serious about the importance of communal accountability.
Abba Moses arrived carrying a jug with a hole in it. Water was leaking out behind him as he walked.
When the other elders asked what this meant, Moses said: "My sins run out behind me and I do not see them, and today I am coming to judge the errors of another."
The council disbanded without passing judgment.
The story appears in the Apophthegmata Patrum — the collected sayings of the Desert Fathers, one of the foundational documents of Christian monasticism — and it is simultaneously funny, devastating, and practically invisible to anyone who doesn't carry the tradition that makes it operative. The absurdity of the leaking jug is funny in the simple sense. The dissolution of a formal judicial proceeding with a single visual metaphor is devastating in the theological sense. But neither works without the lock the key opens. To understand why the council disbanded — why this gesture was sufficient to dissolve the entire proceeding, why not a single elder objected or suggested that Moses was being evasive — you need to carry the tradition's specific understanding of what it costs, and means, to judge another person when your own failures are continuous and unexamined. The joke is simultaneously a teaching, a rebuke, and an act of mercy. It is a door. The question is whether you have the key.
In Ukraine, approximately 1,400 years later, a rabbi laughed three times at his Sabbath table for no apparent reason. His students were confused. He offered no explanation. After nightfall he told them to harness the horses and drive through the dark. They arrived at dawn at the home of an elderly bookbinder and his wife, who were dancing and singing together in the small hours, celebrating their love and the particular pleasures of their shared life. The rabbi told them that Heaven had rejoiced in their joy. He explained his three laughs to his students retrospectively — each had corresponded, across the distance, to a moment of this old couple's selfless, uncalculated gladness.
The story comes from the Golden Mountain collection of tales about Israel ben Eliezer, the Ba'al Shem Tov, the founder of Hasidic Judaism. It is funny in the way all good Hasidic stories are funny: the humor of gap, of violated expectation, of the solemn suddenly requiring a midnight carriage ride to a bookbinder's house. An elderly rabbi giggling inexplicably at a dinner table. Pressing his confused students into a vehicle at night. Driving several hours to watch an old couple dance. This is strange behavior, and strange behavior, correctly timed, is the engine of comedy. But the story only works at full depth — the depth at which a student hearing it might feel genuinely undone — if you carry the Hasidic theology of simcha, which is the teaching that joy is not a reward for righteous living but a form of it, that the old bookbinder and his wife dancing together in their kitchen at four in the morning is not merely charming but is a kind of holiness, and that perceiving it is itself a spiritual act. Without that framework, the midnight road trip is inexplicable and charming. With it, the three laughs are the sound of Heaven noticing something.
A comparative study published in 2019 drew the explicit structural parallel between the Apophthegmata Patrum and the Ba'al Shem Tov stories: both traditions, 1,400 years apart and from completely different civilizations with no institutional connection, use humor as communal formation. The story demonstrates whether you belong by whether you understand it. The laugh is the initiation. Understanding equals membership. The mechanism is identical. The traditions are unrelated. The convergence is not coincidence; it is what happens when human beings in any tradition need to bind communities with a transmission that can only be received from inside.
The Nasreddin Hodja stories offer one more variation worth noting. Across the Turkish, Persian, Arabic, and Central Asian traditions in which the wandering fool-sage Nasreddin (or Nasrudin) appears, these stories were classified by Western scholars in the nineteenth and early twentieth centuries as folk absurdism — charming peasant nonsense from an oral tradition that hadn't developed the sophistication to know it was being silly. The Idries Shah Foundation, drawing on Sufi scholarship, offers a different account: the stories were devised as teaching vehicles for Sufi practitioners, each of them operating at multiple levels simultaneously. The story of Nasreddin fishing in a bucket is a joke about foolishness from the outside; from inside the tradition, it is a teaching about the confusion of form with substance. The story about being paid for the smell of soup with the sound of coins — nonsense from the outside, an argument about reciprocity and the limits of exchange from the inside. The laughter of the inside hearer and the laughter of the outside hearer are both real laughter. They are not the same laugh. The inside hearer laughs twice: once at the surface absurdity, once at the recognition of what the absurdity is pointing at. The double laugh is initiation. To have been initiated is to have access to the depth. The story is the door; the tradition is the key; the laugh at full depth is what the door opens onto.
Will Somers, court fool to Henry VIII, told the king directly that he had "so many frauditers, so many conveyers and so many deceivers to get up your money, that they get all to themselves." Playing on the words auditors, surveyors, and receivers — substituting, in each case, an accusation of theft. Thomas Cromwell apparently appreciated that Somers drew the king's attention to waste and extravagance in the royal household by means of jokes. Whether Cromwell found the jokes funny or merely useful is unclear from the record. The jokes worked either way.
The jester's license operated inside a framework that everyone in the room understood at the institutional level: this was the role that could speak what no other role could speak. The same words from a courtier's mouth would have been treason. From the fool's mouth, they were comedy, and their comedic nature was, paradoxically, what gave them force. A formal accusation of corruption goes into the machinery of official response — denial, counter-accusation, the grinding of administrative process. A joke in the hearing of the king, in front of the court, goes directly to the king's ear in the one moment he is permitted to hear it without having to respond administratively. The joke gets past the defenses. It reaches something.
Then there is the milk incident. Somers threw milk at a court figure — the records are thin on the identity of the target, but the event is documented — at a court dinner, apparently as a way of settling a dispute that could not be settled by any of the formal mechanisms available to him. This is not impulsive clowning. This is a specific application of the jester's only available power: the license to perform with his body what other bodies could not perform. The target apparently never returned to court. Whether from shame, from injury, or from the calculation that anyone who could do that to them in public and survive was a person with more protection than they wanted to contest — the record doesn't say. But the outcome is the outcome of power used well: the problem was resolved.
The jester's power existed entirely in the license of the role. Outside the role, nothing. Which is the thing about roles: they are entirely real until they aren't, and the moment they aren't, they were never real at all.
Dacher Keltner and colleagues, studying fraternity pledging in 1998, documented something that turns out to explain a great deal about how the circle draws itself in real time.
The setting is a fraternity house. Established brothers are teasing new pledges. This is an old tradition and a documented one. What Keltner's team documented — with specific attention to the behavioral content of the teasing — was the difference between teasing that bonded and teasing that didn't. High-status brothers teasing pledges used less redressive action and greater face threat: fewer softeners, fewer explicit signals that the tease was playful rather than hostile, fewer of the vocal and gestural markers that signal "this is a game." Low-status pledges teasing upward used more cushioning: singsong voice, elongated vowels, exaggerated facial expression, formulaic utterances (oh sure you did), explicit hyperbole, embedded laughter. The content of the teases was sometimes identical. The delivery was what determined whether it bonded or cut.
The play signals — the singsong, the elongation, the exaggeration, the signals that say this is performance rather than attack — are the brackets that make the same words into play rather than aggression. Remove the brackets and you have cruelty in joke form. Keep them and you have what the sanankuya partners do in the market square: the same content transformed by the frame that surrounds it.
Campos and Keltner extended this finding cross-culturally in 2007, studying teasing in individualist and collectivist cultural contexts. Collectivist cultures needed less redressive cushioning for a tease to read as warm — the baseline assumption of shared belonging was doing work that the verbal cushions were doing explicitly in individualist contexts. The threshold for what signals "this is play" varied across cultures. The underlying mechanism — play signal plus shared understanding equals bond — did not.
The Pueblo Koshare and the Lakota Heyoka are worth a brief mention here, because they demonstrate that play signals can operate at the scale of an entire ceremonial tradition. The Koshare are sacred clowns of the Keres-speaking Pueblo peoples — their function in ceremony is explicitly transgressive: lewd, scatological, rule-inverting. They mock those in power. They subject audience members to public ridicule. They eat things that should not be eaten. This is not chaos; it is licensed transgression — a specific social function that requires initiation, carries obligations, and serves the community's need to have someone authorized to say what cannot otherwise be said. The Lakota Heyoka operate similarly through inversion: riding horses backwards, saying goodbye when arriving, laughing at bad news. The signal in both cases is the tradition itself: the community knows the role, knows the license, and the humor operates within a framework of shared understanding that makes the transgression legible as social function rather than mere offense.
When outsiders — colonial administrators, missionaries, government officials — encountered these ceremonies, they saw the transgression without the tradition. They saw offense without the play signal. The tradition was the invisible play signal, and without access to it, the behavior read as incomprehensible or scandalous rather than as humor performing its ancient function. This is the insider/outsider dynamic operating at the level of an entire ceremonial system: the same behavior, the same gestures, the same content — but without the shared framework, the laugh has nowhere to land.
Thai and colleagues, in 2019, turned this into a direct experiment. The same joke, word for word, was told to participants by a member of the group being targeted in the joke or by an outsider. Gay jokes told by gay tellers versus straight tellers. Asian-themed jokes told by Asian tellers versus non-Asian tellers. The same joke was rated as funnier, less offensive, and more socially acceptable when told by a member of the targeted group. In-group membership is a license that unlocks the laugh. The content is held constant. The license is the variable.
And here is the precise principle that makes the Friars Club on October 8, 1993 legible: what the content contains is less important than who is holding it.
In Korea in 1951 and 1952, a surgical unit processed 1,936 admissions and recorded eleven deaths. The survival rate was 99.4 percent. This is remarkable medicine by any era's standards, conducted in a canvas tent in sub-zero temperatures by surgeons who were operating on conscripted teenagers at an average of ninety surgeries per week and who had no way of being elsewhere.
Dr. H. Richard Hornberger was one of those surgeons. He was drafted into the Army at roughly thirty years old and posted to the 8055th Mobile Army Surgical Hospital. The unit called their tent The Swamp. Hornberger later wrote about the experience — under the pen name Richard Hooker, in the 1968 novel MASH: A Novel About Three Army Doctors — and his son described it with the kind of precision that only comes from having watched someone try to write around the thing they could not quite bring themselves to write directly: "a humorous account of his work, with serious parts thrown in about the awful kind of work it was."
The humor of the MASH unit — pranks, dark jokes, elaborate insubordination structured as comedy, the collective maintenance of an identity that was defiantly not the identity the Army was assigning them — reads, from outside, as immaturity or callousness. As people who didn't take seriously what they were doing. This reading is wrong in a specific way: it confuses the form with the content. The surgeons took their work extraordinarily seriously. The survival rate is the evidence. What they could not afford to take seriously, under the weight of what they were doing every night, was the performance of appropriate gravity. The gravity would have broken them. The humor was not the absence of seriousness. It was the container that made the seriousness survivable.
The MASH unit's humor was insider humor in its most elemental form: made by people who were inside an experience together, for each other, about the specific, unreproducible texture of being inside that experience. Its jokes referred to things only the people who were there could know — the smell, the cold, the specific faces of the specific patients, the way the work felt at three in the morning when you had already been operating for six hours and there were more patients. From outside, those jokes read as dark, as inappropriate, as signs of moral numbness. From inside, they read as the only available evidence that the people making them were still fully present, still noticing, still themselves.
The Swamp was the circle. The jokes were the mechanism that kept the circle closed against the pressure of what was outside it. This is what affiliative humor does when the stakes are high enough to make the function visible: it does not entertain. It protects.
Now go back to New York.
Whoopi Goldberg held the license. She had co-written portions of the material. She was the roastee and the author and the laughing presence on the dais, and her authorization was exactly the kind of license the tradition requires: in-group membership, shared history, the person who has standing to grant permission and is doing so. Beverly Johnson's response — "if you can't see the humor here, something is really wrong" — reflects the insider understanding precisely. From inside the circle that Goldberg drew, the performance was a love letter. An extreme love letter. A love letter that could only be written in a tradition that authorizes extreme expression as the measure of extreme affection, by a person who understood what she was authorizing and why she could authorize it.
Montel Williams was not inside that circle. His white wife, who cried at jokes about mixed-race children, was not inside that circle. The material reached into a racial history that they carried in their bodies — a history that Goldberg and Williams and the Friars Club tradition all shared, but that Williams and his wife inhabited differently, in ways that Goldberg's authorization could not override. The license can be granted within a relationship. It does not automatically extend to every person in the physical room. And some material — material that references not just a shared relationship but an entire history of violence and dehumanization — reaches outside any individual relationship into territory that no individual's authorization can fully contain.
The analytical move that matters here: both readings were correct. This is not a diplomatic both-sides construction designed to avoid taking a position. It is a description of how the mechanism works. Goldberg's reading was grounded in a real tradition, a real relationship, a real license. Williams's reading was grounded in a real history, real wounds, a real experience of the material that Goldberg's authorization did not touch. The mechanism ran on one track. It broke on another. Both tracks were real, and the mechanism that produced them was the same mechanism.
Then there is Sammy Davis Jr., NBC, April 25, 1975.
Don Rickles and Dean Martin and Nipsey Russell making jokes about slavery, lynching, back-of-the-bus seating, Davis's jewelry, his conversion to Judaism. Martin opened with a joke about an NBC peacock wearing an afro. Davis laughed the hardest at everything. The reviews noted that despite the content, nothing about the evening felt mean-spirited or cruel. At the conclusion, Martin, Bishop, and Rickles kissed Davis. He said: "The day they don't make fun of you, that means they don't give a damn about you."
This is the mechanism's self-description. This is also its operating manual.
Rickles and Martin had performed beside Davis on Rat Pack stages when Las Vegas hotels were still turning Davis away at the front entrance. He had been turned away at the entrance, walked around to the service door, performed on stages that had his name on the marquee, and those specific men were on the stages with him when it was happening. The jokes they made at the roast referred to a shared history that was not merely "the history of American racism" in the abstract — it was the specific, granular, lived history of having navigated that racism together, on particular stages in particular cities on particular nights. When Davis laughed hardest, it was the laugh of someone hearing their own experience described accurately by people who were there for it. The license was not granted because Davis was Black and therefore authorized any racial humor. The license was granted because these specific men shared a history that was also Davis's history, and the jokes were made from inside that shared history, not aimed at it from outside.
The contrast with Danson clarifies something the Davis roast alone does not. Goldberg could grant Danson a license. She could not grant him a history. And the material referred to a history — the history of blackface as a theatrical form for dehumanizing Black people, the history of the specific material Danson was performing — that extended beyond any individual license, into the bodies of people in the room who carried it in ways neither Danson nor Goldberg could fully account for.
Belonging is not binary. The circle is not simply inside and outside. It has a radius, and some material has an amplitude larger than any circle of insiders can contain.
Eighteen days after September 11, 2001, Gilbert Gottfried told a joke about a passenger airplane needing to make a stop at the Empire State Building.
This was at the Friars Club Roast of Hugh Hefner. The Friars Club roast is an institution designed to have no limit on what can be said. If there is anywhere the joke should have landed, it was that room, those people, that tradition.
The room booed him.
Gottfried understood, in the way that experienced performers understand the specific quality of a room's silence, that he had hit the actual limit — not the institutional limit, not the stated limit, but the nerve-ending limit, the place where the wound was too recent and too present to be reached by any frame. He pivoted. He told The Aristocrats — a joke that exists almost exclusively as an in-group signal among professional comedians, never told to civilian audiences, the setup and the extended middle and the punchline all oriented entirely inward toward a tradition rather than outward toward any reference in the world. He ran it for twelve minutes. The first six were flat. The last six, by Kimmel's account, had people crying from laughter.
Both jokes were dark. The Aristocrats is objectively more transgressive in content. Same comedian, same audience, same room, fifteen minutes apart. One produced silence and booing. One produced twelve minutes ending in tears of laughter.
The only variable was whether the joke gestured at a wound the room still carried or whether it pointed inward, toward a tradition the room was inside. The September 11 attacks were eighteen days old and belonged to everyone in the room in a way that no collective ritual had yet processed — the wound was present, uncontained, too large for any comedic framing to reach without touching it first. The Aristocrats referred to nothing outside itself. It was the most insider joke available. Gottfried didn't find a safer joke. He found a more enclosed one.
The affiliative apparatus requires not just shared membership but shared readiness. The wound has to be processed enough for the frame to work. When it hasn't, the mechanism that would otherwise produce synchronization produces rupture — the same room, the same comedian, the same impulse, and silence instead of laughter. The circle can only hold what has been digested together.
NASA has been studying team humor in isolated environments for longer than you might expect, because they discovered that humor predicts survival in ways that other variables don't. In Antarctic winter-over studies — teams of researchers spending months in stations at the South Pole, with no possibility of evacuation and no other humans within several hundred miles — teams that developed what researchers called a "clown role" maintained significantly better cohesion through the long dark winter than teams that did not. The clown role is not about which team member was the funniest person. It is a social function: one member whose purpose, acknowledged or not, is to generate affiliative humor, relieve tension, keep the group's collective nervous system from freezing into isolation. The function cannot be assigned by management; it has to emerge organically from the group. Teams where it didn't emerge showed measurable cohesion deterioration over the course of the winter. The cold was the same for everyone. The isolation was the same. The presence or absence of someone whose job was to make the group laugh together was what differed, and the difference was measurable.
The NASA HERA study, tracking four-person crews isolated for forty-five days in a simulated spacecraft at Johnson Space Center, found the same distinction running at the level of individual humor events: positive humor directed at the group as a whole tracked to cohesion increases; negative humor directed at specific individual teammates tracked to cohesion decreases. The content of the jokes was not the variable. The direction was. Humor that drew a circle around everyone versus humor that drew a circle around everyone except one person: the opioids either fired together or they produced the molecular opposite of bonding. Affiliative humor predicts team survival in closed systems. Aggressive humor predicts team dissolution.
This is what the endorphin studies were always going to arrive at, if followed far enough. Not amusement. Not entertainment. Survival of the group.
The NASA material is interesting not just because it confirms the opioid studies and the couples research in an applied setting, but because of what it reveals about the clown role's emergence. It cannot be assigned. Management cannot appoint a team member to be the person who makes everyone laugh. The function has to arise organically because it only works if it is genuine — if the humor is real, if the person filling the role is actually finding things funny and communicating that, rather than performing funniness as a task. The difference between Duchenne laughter and non-Duchenne laughter applies to the clown as much as to the audience. A designated clown is not a clown. A person who notices the genuinely absurd and says so, and in doing so invites the rest of the group to notice it too — that is the clown role. It cannot be manufactured because the thing that makes it work is that it isn't manufactured.
This is why the sanankuya tradition had to be codified in an oral constitution rather than just practiced spontaneously: the bonds it maintains are real, but the specific form — which surnames mock which surnames, what the founding stories are, what the appropriate content of the ritual insults is — requires transmission, requires teaching, requires that the tradition be deliberately kept alive. The laugh itself cannot be assigned. The tradition that makes the laugh available can be. The Kurukan Fuga did not invent the bond between Fofanna and Jaiteh; it preserved the story that makes the bond legible, so that eight centuries later, a Fofanna and a Jaiteh meeting at a market still have access to the full weight of what their surnames carry.
Here is what "you had to be there" actually means.
Not that you had to be in the physical location. Not even that you had to be present for the specific event. It means that the reference points to something that was only accessible to people who shared the experience — the months in The Swamp, the road trip to the bookbinder's house, the years performing in Las Vegas hotels that wouldn't let you sleep there, the tradition your surname carries about a clan ancestor who cut meat from his own leg on a long journey. The joke draws a circle around the people who were there, in whatever sense of "there" the tradition requires. And the laugh is the recognition: we were there together. We are still there together. We know what this is.
Every human being has stood inside a circle when the joke landed and felt the thing the Finnish PET scanner was measuring — the synchronization, the warmth, the sense of being found. Every human being has also stood outside a circle when the joke landed and felt the asymmetry, the small rejection, the fact of not getting it. The double recognition — inside and outside, both — is the universal. Not the content of any specific in-group. Not any particular tradition. The structure of belonging-through-laughter itself, which requires an outside before it can have an inside.
"You had to be there" is not a limitation of in-group humor. It is the entire point. The experience of being brought inside — which requires, first, having been outside — is something every human being has had. The circle is always being drawn. You have always been on both sides of it. That double knowledge is what makes the laugh, when it arrives, feel like coming home.
Consider what the Irish wake was doing in those centuries when the Church was repeatedly trying to suppress it. The Synod of Tuam in 1614 condemned the "obscene songs and suggestive games" practiced at wakes. The Synod of 1660 repeated the condemnation. Protestant reformers saw the combination of vigil for the dead with drinking and games and organized humor as evidence of Catholic moral corruption — incomprehensible, blasphemous behavior at a religious event. From outside the tradition, this is not an unreasonable reading: you are watching people laugh and drink in the presence of a corpse.
From inside the tradition, the logic is transparent. The community gathers at the worst thing that can happen to it — the loss of one of its own — and maintains itself through the worst of it by doing what communities do when they are together: telling stories, making jokes, singing, playing games. The humor is not disrespect to the dead. It is respect for the living. It keeps the community intact through the rupture. The grief is real; the laughter is real; they are not opposites. They are the two things a community needs at the same time, and the wake is the institution that holds both.
The Church's repeated attempts to suppress the merry wake failed because the practice was meeting a need that the Church's prescribed alternative was not meeting. The Mass for the dead is a beautiful ceremony. It is also, in the specific register of community survival through grief, less useful than two hundred people staying up all night together telling stories about the person who died, and laughing together, and maintaining the circle that the death had threatened to break.
Beverly Johnson was inside the circle in that Friars Club room in 1993. Montel Williams was outside it. Roger Ebert was reporting from its edge. Whoopi Goldberg, laughing, was the one who had drawn it.
All of them were doing exactly what humans do. The mechanism is the same. What varies is who it includes, and where the circle stops.
That asymmetry is not a flaw in the apparatus. It is its most human quality. Every circle has an inside and an outside. The outside is how you know what you're offering when you let someone in.