Jerusalem is a city where coexistence isn't an abstract ideal — it's the Tuesday morning bus. Jewish and Arab neighbors share streets, markets, municipal services, the same checkout lines at the grocery store. And then a security incident happens and suddenly the same street that was just a street becomes a fault line. Everyone knows which side they're on, everyone knows which side everyone else is on, and the whole thing goes quiet for a few days. The coexistence was real and it was also paper-thin. Daniel's been thinking about this — not the policy architecture, but what's actually happening inside individual people who manage to coexist, and what's missing in the ones who don't.
He sent in a whole thing. His argument is we've been looking at coexistence backwards — treating it as a policy problem when it's really a personal characteristic. He wants to know whether tolerance is something you can teach or whether it's baked into personality. Does the capacity for tolerance vary widely across individuals in the same society? And if you try to impose coexistence on an intolerant population through top-down programs, are you doomed from the start? So today we're going bottom-up — not at peace treaties or municipal budgets, but at the individual psychology of tolerance, what education can and can't do, and what it actually takes to build a coexisting society from the ground up.
The standard framing of coexistence is almost entirely top-down. Political agreements — the Oslo Accords, power-sharing arrangements in municipal councils, budgets for shared public spaces. Jerusalem has all of these. You've got mixed neighborhoods like Abu Tor and the German Colony, joint municipal committees, dozens of coexistence NGOs. And yet the city still sees regular breakdowns. Which suggests something deeper is missing.
Daniel's core insight — and I think he's right about this — is that coexistence isn't primarily a sociological or political question. It's a question about individuals. Whether they possess the personal capacity for tolerance, whether that capacity varies across a population like any other trait, and whether it can be deliberately developed. This reframes the entire problem space from institutional design to human psychology.
And it's a uncomfortable reframe, because if tolerance is partly dispositional — if some people are just better at it than others — then a lot of the policy work we've been doing is operating on the wrong layer. You can design the most elegant power-sharing agreement in history, and if the population doesn't have the individual capacity to live it, the agreement is words on paper.
So the first question we need to answer is whether tolerance is something you can actually teach, or whether it's more like a personality trait you're born with.
This is where the personality psychology gets really interesting. The Big Five personality model — Openness to Experience, Conscientiousness, Extraversion, Agreeableness, Neuroticism — is the most robust framework we have for describing stable individual differences. And the trait that consistently predicts tolerance across cultures and contexts is Openness to Experience. People high in openness are more comfortable with ambiguity, with difference, with unfamiliar cultures and practices. They seek out novelty rather than retreating from it.
So the tolerant person is basically the one who hears a language they don't understand at the next table and thinks "interesting" rather than "threat."
No, wait, that's the word you told me not to use. The point stands. High openness correlates with lower prejudice toward out-groups, greater support for multiculturalism, more willingness to engage with people whose values differ from your own. But here's where it gets counterintuitive. Agreeableness also correlates with tolerance — but mainly toward in-group members. Highly agreeable people are warm and cooperative with people they already identify with. That same agreeableness doesn't necessarily extend to out-groups. You can be the nicest person in your community and still deeply suspicious of outsiders.
That's a specific kind of trap. The person who's generous and kind to everyone they know, and uses that as evidence they're not prejudiced, while being completely closed to anyone outside the circle.
And Conscientiousness has its own complicated relationship with tolerance. High conscientiousness — being orderly, dutiful, rule-following — sometimes correlates with rigidity toward groups perceived as norm-violating. If your whole psychological orientation is toward order and rules, and you encounter a group whose practices seem to violate those rules, the reaction isn't curiosity, it's disapproval. The conscientious person isn't necessarily hostile — they just see deviation from norms as a problem to be corrected.
Which means the personality profile of the naturally tolerant person is fairly specific. High openness, moderate to low conscientiousness in the rigidity dimension, and an agreeableness that extends beyond the in-group. That's not everyone.
It's not. And this brings us to the heritability question. Twin studies — comparing identical twins who share all their genes with fraternal twins who share about half — estimate that Openness to Experience is roughly forty to sixty percent heritable. That's a significant genetic component. It doesn't mean openness is fixed — heritability isn't destiny — but it means education and environment are working with a substrate that varies from person to person. Some people start with a dispositional advantage for tolerance. Some start with a headwind.
That's the part of this conversation that makes people uncomfortable. We want tolerance to be purely a matter of values and choices, something anyone can develop with the right education. And to some degree it is. But the starting points aren't equal.
Right. And that doesn't mean we throw up our hands. It means we need to understand what actually works to shift people along that spectrum, and what doesn't. Which brings us to the most influential framework in the psychology of prejudice reduction — Gordon Allport's Contact Hypothesis, published in 1954.
The Nature of Prejudice.
That's the one. Allport proposed that contact between groups can reduce prejudice, but only under four specific conditions. Equal status between the groups in the contact situation. Common goals — they're working toward something together. Intergroup cooperation — they're collaborating, not competing. And institutional support — the authorities and norms of the setting explicitly back the contact.
Four conditions. And if any of them are missing?
If they're not met, contact can actually increase tension. Putting groups together under conditions of inequality, competition, or without institutional backing doesn't reduce prejudice — it reinforces it. You just give people more ammunition for their existing stereotypes. The person who had a bad interaction now has a story, and the story becomes evidence.
So the naive version of coexistence — just put everyone in the same room and let them figure it out — is worse than doing nothing.
In many cases, yes. And this is where the Hand in Hand school network in Jerusalem becomes such an important test case. Founded in nineteen ninety-eight, it now operates six bilingual, integrated Jewish-Arab schools across Israel. Children learn together in both Hebrew and Arabic, with co-principals from each community. Every classroom has two teachers, one Jewish and one Arab. The entire structure is designed around Allport's conditions — equal status, common goals, cooperation, institutional backing.
And it works, at least while the kids are in it. Research on Hand in Hand graduates shows they develop higher cross-group empathy and more nuanced political views than peers in segregated schools. They form genuine friendships across the divide.
They do. But — and this is the part that keeps coming up in the research — the effects weaken once students leave the protected school environment and re-enter segregated adult society. They face military service, which is a profoundly different experience for Jewish and Arab citizens. They enter separate social networks, separate universities to some degree, completely different media ecosystems. The school was a greenhouse, and the outside climate is... not that.
The inoculation effect. Sustained contact during formative years builds tolerance, but it requires constant reinforcement. You can't do it once and expect it to hold.
That's exactly the finding from the broader literature. A meta-analysis of intergroup contact studies found that the duration and institutional support of contact programs were the strongest predictors of lasting attitude change. One-off programs — summer camps, dialogue workshops, weekend retreats — show minimal long-term impact. You get a burst of goodwill that fades within months. The attitudes snap back.
Which is a pretty damning indictment of how most coexistence funding is spent. The weekend workshop is cheaper and easier than running a bilingual school for twelve years, so that's what gets funded. And then everyone's surprised when it doesn't stick.
The Hand in Hand model is expensive. It requires trained bilingual teachers, dual administration, curriculum development that represents both narratives. And even with all of that, the effects are partial and fragile once the institutional container is removed. That's not an argument against doing it — it's an argument for understanding what it can and can't accomplish on its own.
So if tolerance can be cultivated but has real limits, what happens when you try to do the opposite? When education and leadership actively promote intolerance?
This is where the asymmetry problem becomes really stark. Tolerance-building education takes years of sustained contact and institutional reinforcement, and the results are partial. Intolerance education can be rapid and self-reinforcing because it taps into evolved psychological mechanisms — in-group bias, threat detection, coalitional thinking. The brain is already wired to be suspicious of outsiders. You don't have to teach that. You just have to give it a target.
The stickiness of intolerance is higher.
Much higher. And we have case studies. Rwanda's pre-genocide education system, from the nineteen seventies through nineteen ninety-four, formally categorized Hutu and Tutsi as distinct racial groups in school curricula. Complete with pseudo-scientific measurements — nose width, skull shape, height. This wasn't casual prejudice. This was state pedagogy, delivered by trained teachers, tested on exams, repeated year after year. Students learned to see their neighbors as biologically different categories of human being.
And it took less than a generation.
Less than a generation to entrench categories that had been far more fluid before colonial and post-colonial intervention. Nazi Germany's biology textbooks taught racial hierarchy as science, with mathematical problems calculating the cost of caring for disabled people versus the cost of euthanizing them. When you put prejudice into the curriculum — when you make it something children are graded on — you're leveraging the full authority of the educational system to hardwire intolerance.
There's a grim efficiency to it. The same mechanisms that make education powerful for good make it powerful for the opposite. Structure, repetition, authority, testing — all of it works just as well in reverse.
And this creates the grafting problem Daniel's asking about. Can you graft coexistence onto an intolerant society? Evidence from post-conflict settings suggests that top-down coexistence programs imposed on populations that haven't undergone individual-level change produce shallow compliance, not genuine tolerance.
Say more about shallow compliance.
In Bosnia, after the Dayton Accords, peace education programs taught multiculturalism in schools. Researchers found that students learned the right vocabulary — they could recite the principles of coexistence, they knew what they were supposed to say. But in anonymous surveys, the same students expressed the same ethnic prejudices. The attitudes hadn't moved. What moved was their ability to perform tolerance when someone was watching.
They learned the script.
They learned the script. Northern Ireland's integrated schools show similar patterns. Graduates have more cross-group friends than peers from segregated schools, which is a real outcome. But they don't necessarily show reduced prejudice toward the other community as a whole. The friendship is cordoned off — "he's one of the good ones" — without generalizing to the group.
Which is a partial win that's also a failure. The individual relationship is real. The structural attitude is untouched.
Jerusalem is a case study in this grafting failure. Municipal coexistence programs exist. Mixed neighborhoods have joint community initiatives. Hand in Hand runs intensive, well-designed programs. But political leadership at both the municipal and national levels often sends contradictory signals. When politicians use sectarian rhetoric during election campaigns or after security incidents, it directly undermines the educational work. The same students who learn coexistence in school hear their leaders frame the other community as a demographic threat.
You can't out-teach the political environment. If the curriculum says "we are equal citizens with shared futures" and the evening news says "they are a fifth column," the news wins. It's more emotionally salient, it's repeated more often, and it comes with the authority of people who hold actual power.
There's a personality distribution problem underneath all of this. In any society, tolerance varies across a bell curve. A small percentage of people are naturally high in openness and will seek out coexistence regardless of environment — these are the people who join the dialogue groups, who send their kids to integrated schools, who read the other community's newspapers. A small percentage are extremely low in openness and will resist any tolerance education — the predispositions are too strong, the environment reinforces them, and no program is going to reach them.
And the majority are in the middle.
The majority are in the middle. Persuadable but susceptible to the prevailing political and educational winds. These are the people who can go either way, and which way they go depends on what signals they're getting from institutions, media, leadership, and their social environment. The question isn't "can we make everyone tolerant?" It's "can we shift the middle of the curve?"
That's a much more tractable question. And it reframes the whole enterprise. You're not trying to convert the hardliners. You're trying to create an environment where the persuadable middle finds tolerance easier and more rewarding than intolerance.
And that requires sustained, multi-generational effort across education, media, and political leadership simultaneously. You can't fix one and leave the others broken. If the schools teach coexistence but the media broadcasts sectarianism, the media wins. If the media promotes tolerance but political leaders campaign on ethnic fear, the politicians win. The system has to be aligned.
Which is why the grafting problem is so hard. You're trying to impose a tolerant superstructure on a population whose median attitude is somewhere between indifferent and hostile. The institutions say one thing, the social environment says another, and the institutions lose every time because they're shallower.
I want to push on one part of that. I don't think the institutions always lose. I think they lose in the short term. Hand in Hand has been operating for almost thirty years. Its graduates are now adults with children of their own. Some of them are sending their kids to the same schools. That's a multi-generational effect starting to compound. It's slow, it's small relative to the size of the society, but it's real.
So the question is whether you can scale it before the political environment tears it apart.
Or whether the political environment is itself partly downstream of enough people moving along the curve. This is the chicken-and-egg problem. You need tolerant political leadership to create the conditions for tolerance education, but you need tolerance education to produce a population that demands tolerant political leadership.
Which is why this is a generational project, not an electoral cycle project. The time horizons don't match. A politician thinking about the next election has no incentive to invest in something that pays off in twenty years. The educational system can think in those time horizons, but only if it's protected from political interference.
And that protection is itself a political question. You see the loop.
I do.
Hilbert: Nineteen ninety-seven.
Hilbert: I worked a summer program called Kids for Peace. Two weeks, Jewish and Arab teenagers, West Jerusalem and East Jerusalem. Activities, facilitated conversations, the whole model. I was one of the facilitators.
Hilbert: I spent two summers watching kids become genuine friends over a fortnight, and then watching those friendships evaporate by December. They went back to schools that taught entirely different narratives about the same city. The same street names. The same historical events described in completely incompatible language. The camp was a greenhouse — beautiful, controlled, and completely disconnected from the real climate outside.
Hilbert: There was a boy from Beit Safafa. Last day, he told me he'd learned that Jews weren't monsters. And I thought — that's the bar? That's the victory condition? That a child learned that another group of children aren't literal monsters? We were aiming so low we didn't even realize how low we were aiming.
Hilbert: The program measured success by individual attitude change. Pre-camp survey, post-camp survey. The numbers went up. Everyone felt good. Nobody measured what happened three months later. Nobody wanted to.
Because if you measured it, you'd have to admit the program wasn't working.
Hilbert: The program worked fine. For two weeks. The problem was everything outside the program.
This is exactly the finding from the -analyses. Short-term contact without structural reinforcement produces short-term effects. The friendships were real. The context that dissolved them was also real.
Hilbert: I remember the staff meetings. We'd debrief after sessions and everyone would talk about how powerful the experience was, how much the kids had grown. And I'd think — they're going home on Sunday. They're going back to schools where the other side is the enemy, where the maps don't show the same borders, where the history textbooks describe different wars. We gave them two weeks of a different world and then sent them back to the real one. What did we think was going to happen?
You thought the experience would inoculate them.
Hilbert: We thought a lot of things.
The inoculation metaphor is wrong, isn't it? An inoculation gives you a weakened dose of something so your immune system can learn to fight it. What you were giving them was the opposite — a dose of a world that didn't exist anywhere else in their lives.
Hilbert: A dose of a world we wished existed.
And the kids knew it.
Hilbert: The kids always know it. They're not stupid. They knew the camp was a bubble. They enjoyed the bubble. They made real friends in the bubble. And they also knew the bubble would pop. Some of them exchanged phone numbers. A few stayed in touch for a while. Most didn't.
Hilbert: I still think about that boy from Beit Safafa. He was maybe fourteen. Smart kid, funny. The kind of kid who'd be fine in any environment you put him in. And the best we could offer him was two weeks where he learned the other side wasn't monsters. Two weeks. Out of a childhood.
Hilbert: My brother-in-law works in educational assessment. He says the problem with these programs isn't the design, it's the theory of change. The theory is that changing individual attitudes will eventually change the society. But the society is organized to prevent exactly that. The segregation, the separate school systems, the different languages, the different media — all of it is structured to maintain the separation. A two-week camp doesn't compete with a lifetime of structure.
The structure is the curriculum.
Hilbert: The structure is everything.
So where does that leave us? If the greenhouse approach doesn't work and the grafting approach doesn't work, what's left?
I don't think the conclusion is that nothing works. I think the conclusion is that isolated interventions don't work. The Hand in Hand model is more intensive — it's not two weeks, it's twelve years. And even that shows attenuation effects when students leave. But the attenuation is partial, not total. Graduates carry something with them. The question is whether enough of them carrying enough of something over enough generations changes the distribution.
Which brings us back to the bell curve. If the persuadable middle is where the battle is won or lost, and if shifting that middle requires simultaneous intervention across education, media, political leadership, and institutional design — sustained over generations — then what does a effective coexistence strategy look like?
It looks like Hand in Hand plus a media environment that doesn't actively undermine it plus political leadership that treats coexistence as a national priority rather than a slogan. And all of those things have to be true at the same time, for a long time. The failure mode isn't that any one of them is impossible — it's that they're never all true at once.
Daniel's prompt reframes coexistence from a political problem to a human one. And the uncomfortable implication is that some people are simply more capable of coexistence than others. That's not a moral judgment — it's a description of variance in a psychological trait. The question for societies is whether to work with that reality by designing systems that support the persuadable middle, or to try to change the underlying distribution through intensive, multi-generational investment.
I think you have to do both. You design systems that make tolerance the path of least resistance for the middle of the curve — that's the institutional work. And you invest in the intensive educational interventions that actually shift people along the curve over time — that's the generational work. Neither one works without the other.
You accept that some percentage of the population is never going to move, and you design around them rather than trying to convert them. The goal isn't universal tolerance. The goal is a society where the tolerant middle is large enough and supported enough to set the norms.
That's a more modest goal than what most coexistence rhetoric promises. It's also more honest.
If you take one thing from this, it's that coexistence isn't a policy you implement — it's a disposition you cultivate, and it varies across a population like any other human characteristic. The institutional design matters, but it only works if enough individuals have the personal capacity to live inside it.
That capacity can be built — slowly, partially, at significant cost — but only when education, media, and political leadership are all pulling in the same direction. The moment any one of them pulls the other way, the whole thing starts to unravel.
The open question we're left with is whether any society has actually achieved that alignment for long enough to see the generational shift. I'm not sure the test case exists yet.
Hand in Hand has been running for almost thirty years. Ask me again in another thirty.
This has been My Weird Prompts, with production by Hilbert Flumingtop. If you want to dig deeper into the research on any of this — the personality psychology, the contact hypothesis, the case studies — the show notes are at my weird prompts dot com. We'll be back soon.
