Two people, one cookie left on the plate. Watch what happens. Nobody says "I prefer cookies." They say "that's MY cookie." The preference has quietly turned into a property claim, and the property claim is one bad afternoon away from a fistfight. Selvalingam tells it flat: the last cookie "becomes a question of genocide" faster than anyone wants to admit.1 Not because cookies matter. Because the moment my wanting becomes a fact about the world rather than a fact about me, anyone who wants differently is now wrong — and wrong people, eventually, get hurt.
That's the whole engine of this page. There's a sentence I do when I drink chai. I say I like this chai. And I think I'm describing the chai. I'm not. I'm describing me. "If I like this chai," Selvalingam says, "that says something about ME, not the chai."1 The chai is just hot leaf-water. My liking lives in my nervous system, my memory, my morning. The error is small and almost invisible: I take a thing that's true for me and I file it as true period. And once it's filed there, your disagreement isn't a difference in taste. It's an attack on reality. "When you don't share my opinion," he says, "sooner or later I might kill you. That's the history of religion and politics."1
The claim is narrow and exact: most human conflict runs on a single confusion — treating a subjective preference as an objective truth. Liking is a verb that happens inside a person. Likeable is an adjective people pretend belongs to the object. The slide from the first to the second is where the trouble starts.1
Selvalingam roots this in the Śaiva view of where reality lives. If everything you experience is happening in your own awareness — if the chai's "deliciousness" is an event in consciousness, not a property baked into the leaf — then dressing up your preference as a cosmic fact is a category mistake about the structure of experience itself.1 The world doesn't hand you "good" and "bad" pre-labeled. You label it, then forget you did, then defend the label as though it were the thing.
This is why he reaches straight for religion and politics. Those are the two domains where private preference dresses up as absolute truth most aggressively — my god, my nation, my way of life — and where the body count proves how serious people get once they've forgotten the preference was theirs to begin with.1
Here's the move, step by step, so you can catch yourself doing it. First there's a sensation — pleasant, yours, no problem. Then there's a verdict: this is good. The verdict still feels personal. Then comes the smuggle: the "for me" drops off the end of the sentence. This is good becomes this IS good — full stop, no observer attached. Now it's floating free, looking like a fact about the universe. And a fact about the universe is something you're entitled to enforce.
Selvalingam's tell that the content never really mattered is Mao. Set the young against the old, he points out, and you get exactly the same ferocity you'd get from race or money or religion.1 Swap the variable — age instead of skin, class instead of creed — and the machine runs identically. That's the diagnostic. If you can plug anything into the slot and still get war, then the slot was never about the thing in the slot. It was about the human need to have a side, "something that makes us feel like we have to fight."1 The cookie, the chai, the flag — interchangeable handles on the same underlying contraction.
This page hands the vault a clean mechanism for the origin of intolerance, and it slots under the theodicy cluster without repeating it. The existing theodicy pages explain why the Śaiva view sees no evil in the cosmos — only ignorance. This one supplies the ignorance: the specific cognitive slip (preference-mistaken-for-truth) that generates the human cruelty those pages then dissolve. It also gives Pratyabhijñā Recognition View a moral payoff — recognizing that your verdicts are yours is the same recognition, scaled down to a teacup. And it feeds the Śiva-portrait pages a reason Śiva is "outside" all camps: he doesn't run the preference-as-truth smuggle, so he has no side to defend.
Take the case Selvalingam picks, because it's the one that strips the disguise off. We tell ourselves conflict is about something — about race, about religion, about money, about the deep incompatibility of two peoples. Mao's manoeuvre exposes that as a story we tell after the fact.1
He didn't need race. He didn't need religion. He took one population and split it along age — young against old — and produced the same denunciations, the same purges, the same readiness to destroy a neighbor that any "real" grievance produces. The variable was arbitrary. The ferocity was not. That asymmetry is the whole lesson: if you can generate maximum hatred from a difference as bloodless as how many birthdays someone has had, then the hatred was never caused by the difference. The difference was just the peg you hung it on.1
Now run it back through the cookie. The cookie "becomes a question of genocide" not because cookies are precious but because the cookie is a peg — a place to plant the flag of mine, my side, my claim is the right one.1 Mao and the cookie are the same case at different scales. One is two friends at a table; the other is a nation. Both run on a preference that forgot it was a preference and went looking for someone to enforce itself against.
And here's the turn that makes it Śaiva rather than merely cynical: the diagnosis is liberating. If the fight was never really about race or age or cookies — if it was always the misfiled preference looking for a target — then the way out isn't winning the fight. It's catching the misfile. Notice that "I like this" is about you. The war loses its fuel the instant the preference is returned to its owner.1
You're at dinner. Someone says the food is too spicy — not "too spicy for me," just too spicy, as a fact about the food. You feel the small reflex to correct them, because to you it's perfect. Stop there. That reflex is the whole thing in miniature. You're about to defend a property of your own tongue as though it were a property of the dish.
So you do the small move. Silently, you re-attach the missing words. They find it too spicy. I find it perfect. Two preferences, two nervous systems, no contradiction — nothing in the world needs to be adjudicated, because there was never a fact in dispute, only two tastes pretending to be one.
Now take it up a level, where it costs something. A friend says a thing about politics, about God, about how people should live, and you feel the heat rise — the you're wrong about reality heat. Catch it at the cookie. Ask, before you answer: is this a fact, or is this my chai? Almost always it's the chai. The practice isn't agreeing with them. It's noticing that your disagreement is a difference in preference, not a war over truth — and letting the heat drain out before it goes looking for a target. Do this for a week and you'll feel how much of your daily friction was the smuggle running unnoticed.
You've lost it when "I like" has vanished from your speech and only "this is good / this is bad" remains. That missing "for me" is the leak. Listen for it.
You've lost it when disagreement feels like insult — when someone preferring a different food, faith, or party lands as a personal affront rather than information about their nervous system. The affront is the preference defending its disguise.
You've lost it when you catch yourself needing the other person to be wrong, not just different. Difference is two preferences. Wrongness is one preference that has crowned itself the truth and now requires the other to bow. The clearest red flag of all: you notice you'd fight over the cookie — and a quiet voice tells you it isn't about the cookie, and you tell that voice to be quiet.
The whole argument rides on a metaphysical commitment — that experience-qualities live in awareness, not in objects — which is the Śaiva nondual view, not a neutral observation everyone shares.1 A realist about value would push back: some preferences track real goods, and not every moral claim is a chai-preference in disguise. Selvalingam's frame risks flattening that — if all value-claims are subjective smuggles, it's not obvious how he keeps "religious violence is bad" from being just his own chai. He gestures at this by grounding the ethic in recognition rather than in a competing absolute, but the tension is live and worth holding open.
This is a single practitioner transcript, auto-captioned, every claim paraphrased — it carries an experiential warrant, not a scholarly one. It sits alongside the vault's scholarly Śaiva sources (Wallis, Samuel, White), who supply the textual grounding this lived account leans on but never cites. Read it as a teacher's working model, pressure-tested against the metaphysics elsewhere, not as a sourced philosophical proof.
Why would a teaching about chai connect to social psychology and to a healing model of shame? Because all three are circling the same hinge: the moment a person's private state gets mistaken for a public fact, and what that mistake then licenses you to do to other people.
Start with Us-vs-Them Psychology: The Minimal Group Paradigm. The lab finding is brutal and exact: split people by a coin-flip, a meaningless tag, a "you're a dot-overestimator" — and they instantly favor their own and shortchange the others. No history, no grievance, no real stake. That is Mao's move proven in a controlled room. Selvalingam reaches the same floor by intuition — that the conflict is never about the variable, only the having-of-a-side — and the minimal-group experiments hand him the data he's working without. Put them together and you get something neither states alone: the cookie isn't an analogy for tribalism, it's the same neural reflex caught at its smallest possible size. The lab shows the side-taking is reflexive; Selvalingam shows the exit — that the reflex runs on a misfiled preference, and re-filing it is the off-switch the lab never tests for. The behavioral page tells you the trap is automatic; the Śaiva page tells you it's seen-through-able.
Now Shame as a Survival System. Shame is the felt cost of being wrong in the eyes of the group — the body's alarm that you've fallen out of belonging. That alarm is exactly what powers the smuggle. If my preference is just mine, I'm exposed; I could be wrong, odd, outside. But if my preference is the truth, then I'm not the weird one — you are, and the shame transfers to you. Dressing a taste up as a fact is, underneath, a maneuver to dodge shame by exporting it. That's the link the religion-and-politics line needs: people don't enforce their preferences because they're certain, they enforce them because uncertainty feels like the threat of exile, and converting "I like" into "this is good" makes the threat go away. The insight that falls out of holding the two pages together: religious violence isn't a surplus of conviction, it's a defense against shame — and the Śaiva move of returning the verdict to its owner only works for someone who can tolerate being the one with the odd taste without feeling annihilated for it. Recognition, here, requires having somewhere to stand that the group's approval can't reach.
The Sharpest Implication. The thing you call your morality may be, in large part, your chai — preferences you've crowned as truths because the alternative (sitting with the fact that you simply like this and not that, with no cosmic backing) is unbearable. If that's right, then your most cherished certainties about how people should live are not your nobility. They're your nervous system, armed.
Generative Questions.