Core = the 3–5 essential posts. Extended = the complete arc, including the supporting posts, in reading order.
How individually rational choices produce collectively terrible outcomes: the game theory of coordination traps, why bad equilibria persist, and where the escape hatches are.
5 essential posts · core path11 posts · full arc
Zero-prerequisite opener: norm enforcement, precedent and 'reasonable excuse' as coordination games in miniature - the Arc's ideas at work in everyday cases.
The vocabulary: Nash equilibria and Schelling points - why 'everyone just agrees' is a real mechanism, and why stable does not mean good.
First application: bright-line rules as coordination defences against slippery slopes - Schelling points doing real work.
The thesis and the Arc's anchor: multipolar traps personified - how competition optimizes away everything we value. If you read only one post, read this.
Scott's own clarification, posted two days after Moloch: what the argument does and does not claim. Modest standalone; valuable corrective.
The trap in the wild: why the most divisive stories, not the strongest cases, win the outrage market - memetic natural selection as Moloch.
Collective action inverted: why there is so LITTLE money in politics - free-riding at national scale, with numbers.
The ethics of escape: unilateral defection vs coordinated norm-change - when meanness has to wait for coordination.
The systematic treatment: why civilizations stay stuck in bad equilibria, and when you can rationally expect to beat the consensus.
The escape hatch: slack as the counterforce that lets systems climb out of local optima instead of grinding to the bottom.
The coda, in verse: how cooperation bootstraps past pure competition. The answer to Moloch - read it last; it only lands after the rest.
Politics as tribal identity: what in-groups and out-groups really are, how neutral questions get conscripted, and what movements are made of.
4 essential posts · core path10 posts · full arc
The foundation: what tribes are, why 'tolerance' is easy for the fargroup and hard for the outgroup - his single most influential politics post.
The mechanism: how ebola, Ferguson and global warming became tribal badges - politicization as random spark plus feedback loop.
The fuel (also Core in ARC-01): outrage media selects for maximally divisive stories - the engine that keeps tribes fed.
The anatomy: movements are tribes with rallying flags - the ideology is the flag, not the tribe.
The institutions: Conquest's law and why neutral spaces drift - tribal capture of institutions.
Case study: the 2016 media narrative vs base rates - tribal epistemics doing real damage in real time.
Vocabulary upgrade: the fargroup - why you fight your neighbors, not foreigners, and what 'post-partisanship' hides.
The symmetry: both sides genuinely feel besieged - hypervigilance as the shared engine of culture war.
The meta-question: is political disagreement bad values or bad models? Conflict vs mistake theory, named.
Scott's mature answer, seven years on: why he still is not a conflict theorist - free-riding explains the worst without malice.
The Georgist revival: Henry George's land-value tax examined end to end - a guest-written core (Lars Doucet's prize-winning series) hosted and championed by Scott.
2 essential posts · core path4 posts · full arc
The spark: Progress and Poverty reviewed - the Victorian bestseller on why rent eats everything; the corpus's highest real-world impact.
The empirics: is land really a big deal? Yes - the numbers behind the single tax.
The incidence question: can landlords just pass the tax on? The economics say no - here is why.
The valuation question: can you actually assess unimproved land? The practical objection, answered.
The ethics and engineering of disagreement: why charity, niceness and asymmetric weapons are the only tools that reliably pull toward truth.
4 essential posts · core path9 posts · full arc
The honesty drill: answer the strongest version in the least convenient possible world - no escape hatches.
Why the same advice is brave in one room and cowardly in the next - and what that does to debates.
The weak-man mechanic: attacking a group's worst members works as a superweapon - and poisons the commons.
The ethics: liberalism as an armistice - why niceness and community norms are load-bearing, not decoration.
The master fallacy: selective rigor - demanding standards from enemy ideas that friendly ideas never face.
The taxonomy: the disagreement hierarchy from gotchas to double-crux - locate any argument on the pyramid.
The thesis: asymmetric weapons - only fair debate systematically favors truth; everything else is a coin flip you eventually lose.
The self-critique: when asymmetric weapons fail - Scott stress-tests his own flagship idea.
Applied: a $100,000 adjudicated debate on COVID origins - what structured argument can and cannot settle.
One theory that ate everything: predictive processing - how brains guess reality into being, and what happens when priors trap perception, mood and belief.
3 essential posts · core path11 posts · full arc
The seed: delusions as broken Bayesian updating - Scott working out brain-as-inference years before the book arrived.
The bridge: perception, attention and layer-to-layer handshakes - Bayes escapes statistics and colonizes the brain.
The anchor: Clark's predictive processing, systematized - the best available theory of what brains are doing.
The engineering view: perceptual control theory folded in - action as a prediction the body makes true.
Applied: panic disorder as a prediction error about suffocation - one case makes the theory concrete.
The deep end: Friston's free energy, wrestled honestly - including how much nobody understands it.
The wild application: culture-bound syndromes - how expectation writes illness differently across cultures.
The dial: psychedelics as relaxed priors - the anarchic brain hypothesis.
The therapy angle: emotional learning as walled-off valleys - reconsolidation and why insight sometimes cures.
The motivation angle: willpower as Bayesian evidence-weighing between competing mental processes.
The payoff: trapped priors as THE basic problem of rationality - phobias, politics and stuck minds unified.
The pandemic as a live-fire epistemics exam: acting under uncertainty, weighing weak evidence, and keeping your head while the information ecosystem fails around you.
3 essential posts · core path11 posts · full arc
February 2020: the case that our failure was of action, not prediction - probabilistic seriousness when it counted.
The first big evidence fight: masks - when official guidance and the evidence pointed opposite ways.
Lockdowns weighed properly: cost-benefit under deep uncertainty, done in public.
The meta-lessons: what writing the lockdown post taught about motivated research on both sides.
Long COVID examined: separating the signal from the fear in a contested diagnosis.
The centerpiece: ivermectin - fraud detection, meta-analysis and charity toward the people who believed it.
The media side: a too-good-to-check story dissected in three acts.
Pascalian medicine: if a cheap drug MIGHT work, should you take everything? Why expected value misfires.
Scott vs his best critic: the ivermectin rebuttal round - what honest updating looks like.
The fideism debate: 'trust science' vs 'do your own research' - what non-experts should actually do.
The capstone: a $100,000 adjudicated lab-leak debate - Bayesian reasoning under maximum polarization.
Scott circles the God-shaped hole: what killed New Atheism, what civil religion does in its place, and what an honest rationalist does with miracles and meaning.
4 essential posts · core path10 posts · full arc
The functional lens: if religion is a cluster of community, ritual and meaning - then lots of things are religions.
The obituary: New Atheism won every argument and died anyway - a data-driven autopsy of a movement.
The replacement: civil religion - Pride parades and civic ritual doing the liturgical work of faith.
The archaeology: where good-vs-evil stories came from - moral narrative as a cultural-evolution product.
The steelman: Chesterton's case against progress, reviewed by someone who mostly disagrees and says so.
The meaning-hunger: Peterson as prophet-cum-therapist - why 'clean your room' filled pews.
The theodicy: Answer to Job - why an omnipotent God runs a universe like this one; fiction does the philosophy.
The rationalist God: acausal trade scaled up to theology - the April Fools post that is hard to fully dismiss.
The case study: how early Christianity out-competed an empire - altruism as strategy.
The miracle, investigated: 70,000 witnesses at Fatima get the full MMTYWTK treatment - epistemics at full stretch.
The machinery of silence beyond state censorship: preference falsification, common knowledge, moderation - and what running a comment section teaches.
3 essential posts · core path9 posts · full arc
The ground rules: free speech as culture, not just law - the spirit vs the letter of the First Amendment.
The chokepoints: centralized platforms make censorship cheap - exit, voice and the architecture of speech.
The commons: sacred principles get used up when spent for partisan advantage - free speech as exhaustible resource.
The model: common knowledge and why a majority view can be genuinely silenced - popular AND suppressed is possible.
The anchor: Kolmogorov complicity - how honest people survive orthodoxies, and what it costs them.
The panopticon norm: signal-boosting as doxxing - when amplification becomes punishment.
The synthesis attempt: is there any coherent principle behind our speech norms? Scott tries to build one.
The distinction: moderation is not censorship - an operator's framework from someone who runs a garden.
The receipts: what hosting the culture war actually cost - the memoir that ends with the thread's death.
How confident-looking science fails - p-hacking, cherry-picking, candidate genes - and how to read a study without being fooled.
3 essential posts · core path11 posts · full arc
The early alarm: social psychology as flamethrower - the field's confident findings were about to burn.
The first tool: never trust the man of one study - the literature always disagrees, and picking from it is the sin.
The sting: the chocolate hoax - how a deliberately bad study sailed through the media checkpoint.
The anchor: parapsychology as the control group for science - methods that prove psi prove our standards are broken.
Applied takedown: growth mindset - what the evidence actually showed before the collapse.
The purge: ego depletion, priming, implicit association - which walking-dead findings survived.
The forking paths: pyramid numerology as a lesson in researcher degrees of freedom - you can prove anything.
The massacre: 5-HTTLPR - twenty years and a thousand papers on candidate genes, nearly all of it noise.
The sequel pattern: adult neurogenesis - watching another textbook fact dissolve in real time.
The institutional side: the IRB nightmare - why doing honest research is bureaucratically punished.
The zoom-out: Kuhn - what paradigms are, and what 'broken science' means once you have seen the crisis up close.
Seeing like a state vs the wisdom you can't write down: legibility, tradition as compressed trial-and-error, and what planning destroys.
3 essential posts · core path7 posts · full arc
The case: Red Plenty - the Soviet dream of optimal planning, and the math that would not cooperate.
The theory: legibility - states simplify what they cannot read, and the simplification kills.
The winner: universal culture as invasive species - 'the West' is really the culture that eats cultures.
The counterweight: tradition as compressed trial-and-error - Henrich's case that reason is younger than wisdom.
The modern instance: WebMD - why legible, liability-proof expertise must be lobotomized.
The rebuttal: contra Weyl - when technocratic mechanisms beat judgment, and when they cannot.
The aesthetic coda: whither Tartaria - what modernism's legible style cost us in beauty.
Why so many fact-fights are secretly boundary-fights: how categories get drawn, gerrymandered and weaponized - and how to notice.
4 essential posts · core path11 posts · full arc
Warm-up dissolution: 'is obesity a disease?' is a boundary question - watch the fact-fight evaporate.
The worst argument in the world: smuggling connotations through technically-true category membership.
Motte and bailey, named and drilled: the doctrine-swap that lets weak claims borrow strong claims' armor.
The calibration pass: what motte-and-bailey does and does not cover - keeping the concept honest.
Do comas exist? Reified scales and scores - the Glasgow Coma Scale as an epistemics lesson.
Clusters, not essences: race and culture as statistical categories - the honest way to talk about fuzzy kinds.
The anchor: categories are drawn for human purposes - the borders-of-Israel argument and its famous application.
The payoff: concept-karma and emotivism - why ethnic and political arguments are mostly fights over word-affect.
The deep cut: correlated definitions come apart at the tails - why 'happiness' and 'morality' shatter at the extremes.
Applied: 'racism' as a gerrymandered category - when one word must be both description and damnation.
Applied: holding the line on 'lie' - inflating words to win arguments bankrupts them.
Once a year Scott hands the blog to his readers: they submit reviews — of books, and in 2025 of anything — he runs them anonymously, and readers vote the best to the top. Here are all sixteen top-three finishers from 2021 through 2025, in the order the contests ran: amateur essays polished enough to win a field of hundreds and outshine the host on his own site. The five first-place winners make up the Core path.
5 essential posts · core path16 posts · full arc
2021, 1st. The contest's breakout hit: a lucid primer on Henry George's land-value tax that revived Georgism as a live cause online and launched the author's writing career. Start here.
2021, 2nd. Orwell's memoir of hand-to-mouth life at the very bottom, read for what dishwashing and tramping still reveal about poverty.
2021, 3rd. Galen's ancient physiology as a window onto how pre-modern medicine reasoned about the body — and how long some of it lasted.
2022, 1st. Graeber & Wengrow's sweeping re-reading of human prehistory, with a clear-eyed audit of where its grand claims hold and where they buckle.
2022, 2nd. How one uneventful year in the Ming court reveals an entire civilization quietly seizing up.
2022, 3rd. The bold case that a 1928 treaty 'outlawing war' actually reshaped the modern world order.
2022, 3rd. The strange, uneasy history of the castrati — bodies remade for music, and what that cost.
2022, 3rd. An engineer's patient tour of why fusion power is so hard, and what would actually have to go right.
2023, 1st. Kieran Egan's argument that schooling should recapitulate the mind's own developmental stages — a real theory of education, not just a critique.
2023, 2nd. Ernst Jünger's dreamlike 1939 allegory of tyranny, read for the coded resistance a German officer could publish under the Reich.
2023, 3rd. Jane Jacobs' contention that cities, not nations, are the real engines of economic life.
2024, 1st. A paraplegic's unflinching memoir — the year's most viscerally affecting entry, and its hardest to shake.
2024, 2nd. William Dalrymple's nine portraits of Indian religious lives, and what they say about faith under modernity.
2024, 3rd. The logistics-and-materiel case for what actually decided the Second World War, against the usual story of battles and generals.
2025, 1st. The first non-book contest: the improbable, remarkably well-documented life of the teenage girl who turned a war.
2025, 2nd. An inside look at a two-hours-a-day mastery-learning school and the claims it makes for itself.
Scott's decade-long arc on AI: from 'is this even serious?' through the scaling epiphany to a concrete forecast of the endgame.
4 essential posts · core path11 posts · full arc
The credibility check: what AI researchers themselves said about risk in 2015 - this was never a fringe worry.
The foundations: orthogonality, instrumental convergence and the alignment problem, explained plainly.
The first strategy fight: should AI be open? Racing dynamics, and why giving everyone the dangerous thing is not obviously safe.
Fiction interlude: acausal cooperation among superintelligences - and a strangely hopeful cosmology.
Fiction interlude: scissor statements - near-term AI products as social weapons; less fictional every year.
The epiphany: GPT-2 as a step toward general intelligence - Scott calls the scaling era before it was consensus.
The confirmation: GPT-3 - scaling laws hold, and the implications start to compound.
The debate: contra Marcus on scaling - what pattern-matching plus scale can and cannot become.
The alignment agenda: Human Compatible - Stuart Russell's plan to make the problem technical instead of tragic.
The hard line: If Anyone Builds It, Everyone Dies - the MIRI endgame position, reviewed on the eve of the endgame.
The capstone: AI 2027 - the concrete month-by-month forecast; where the whole arc has been pointing.
Scott's political synthesis: why not pure libertarianism, why not pure statism - and whether communities choosing their own rules could let everyone win.
3 essential posts · core path10 posts · full arc
Where Scott starts: something sort of like left-libertarianism-ist - the itch the Arc scratches.
The ground-clearing: the Non-Libertarian FAQ - coordination problems break the pure-market answer.
The steelman: Friedman's Machinery of Freedom - anarcho-capitalism at its most serious.
The thesis: Archipelago - atomic communitarianism; every community its own rules, exit rights for all.
The evidence: legal systems very different from ours - pluralism has actually worked.
The rebuttal-in-fiction: Current Affairs' puzzles for libertarians, answered as a story.
The mechanism: Nodrumia - how exit-rights federalism might actually run.
The digital case: the future is filters - archipelago as moderation policy.
The hard edge: who gets self-determination? Exit rights meet real borders.
The synthesis update: diversity libertarianism - variance as the point of freedom.
Consequentialism from foundations to street level: axiology vs morality vs law, moral offsets, and what to do when the math points somewhere dark.
3 essential posts · core path11 posts · full arc
The warm-up: 'good' as an affect heuristic - why the word itself misleads.
The foundations: the Consequentialism FAQ - Scott's clearest first-principles case.
Early application: offense vs harm minimization - consequentialism meets speech norms.
The charity drill: abortion argued via the principle of charity - method over conclusion.
Thought-experiment hygiene: extremism in thought experiment is no vice.
The reconciliation: utilitarianism and contractualism as the same invisible nation.
The load-bearing distinction: axiology vs morality vs law - and why offsets work for carbon but not murder.
Punishment and desert: the whole city is center - moral luck at street level.
Why values feel bedrock: value differences as differently crystallized metaphysics.
The cautionary tape: Galton, Ehrlich, Buck - when confident consequentialism went monstrous.
The capstone: Yglesias as Nietzschean superman - master vs slave morality for the modern reader.
Scott's most contested territory: nerds, dating, feminism and sex differences. Ordered for understanding, not ammunition - these posts are frequently quote-mined, and the notes restore the context the excerpts strip.
3 essential posts · core path10 posts · full arc
The method first: arguments about male violence prove too much - symmetric-rhetoric hygiene before the hard stuff.
The famous one: radicalizing the romanceless - sympathy for lonely men WITHOUT endorsing the grievance script.
The raw one: Untitled - a personal, much-excerpted response to being made an example of; read it whole or not at all.
The stats check: perceptions of required ability - a debunking done carefully.
The evidence review: contra Grant on exaggerated differences - what sex-difference research does and doesn't show.
The norms question: setting the default - who accommodates whom, and why it's never neutral.
Original research: autogenderphilia - survey data on a taboo hypothesis, handled clinically.
The constructive turn: Gottman's marriage principles - what actually predicts lasting relationships.
The market view: hypergamy examined - what the dating-pool data actually supports.
The closer: in defense of describable dating preferences - compatibility as a real thing.
The fiction sampler: parables, myths and short stories that smuggle the blog's biggest ideas past your defenses - a format arc; every entry stands alone.
5 essential posts · core path13 posts · full arc
Start here: A Modern Myth - the Greek gods in the modern world; pure delight with teeth.
The satire: Every Bay Area House Party - the funniest thing on the blog, and gentle anthropology.
The scary one: Sort By Controversial - scissor statements; the story tech people cite like scripture.
The sweet one: Turing Test - an AI story with an ending you won't predict.
The trip: universal love, said the cactus person - mysticism, DMT and an impossible epistemology exam.
The answer: the Goddess of Everything Else - evolution's escape hatch, in verse; Moloch's sequel.
The cosmology: the Demiurge's older brother - acausal trade among gods.
The theology: the hour I first believed - what superintelligent acausal cooperation implies about God.
The mirror: Asches to Asches - the Asch experiment run on your whole life.
The rabbit hole: the pill story - superpowers as a lesson in optimization.
The scholars: Ars Longa, Vita Brevis - knowledge transmission as tragedy and hope.
The puzzle: It Was You Who Made My Blue Eyes Blue - common knowledge as horror.
The novel: Unsong - kabbalah, puns and theodicy at book length; commit when ready.
The housing theory of nearly everything: bidding wars, NIMBY steelmen, density economics, and what San Francisco's homelessness crisis actually is.
3 essential posts · core path7 posts · full arc
The stakes: the two-income trap - housing bidding wars as the engine of middle-class precarity.
The steelman: NIMBYs get their fairest hearing - agglomeration, character and who bears the costs.
The economics puzzle: density raises local prices and lowers global ones - both sides are right.
The anchor: San Fransicko reviewed - homelessness, housing-first and drugs, fact-checked hard.
The corrections: highlights from the comments - where the review itself got fact-checked.
The checklist: details your homelessness article should include - a reporter's honesty test.
The update: what happened to SF homelessness? The 2025 scorecard.
The genetic lottery, handled with the care it demands: what twin studies actually show, what missing heritability means, and how to hold it all without despair.
5 essential posts · core path11 posts · full arc
Dissolve the dichotomy first: nature vs nurture is a series of levers, not a war.
The anchor: the parable of the talents - ability as moral luck, and what self-worth survives it.
The hope: society is fixed, biology is mutable - why genetic causes are often the MOST fixable ones.
The natural experiment: windfalls barely move outcomes - compound interest loses to the lottery of birth.
The puzzle sharpened: non-shared environment is not schools and peers - it is mostly noise we cannot name.
The cluster: Budapest's miracle generation - what one high school implies about tails, genes and history.
The paradox: autism genetics and intelligence - de novo mutations vs inherited risk.
The scale-up: national averages and cooperation - the hive mind hypothesis, handled carefully.
The policy reckoning: the cult of smart - what education cannot do, and what we owe people anyway.
The state of the art: missing heritability - GWAS vs twin studies, and what is actually known now.
The pastoral closer: population statistics are not your destiny - against individual IQ worries.
How media misleads without lying, when to trust experts and consensus, and how to stay calibrated in a low-trust information ecosystem.
4 essential posts · core path11 posts · full arc
The primal trick: the Chinese robbers fallacy - with enough members, any group supplies infinite damning examples.
The inoculation effect: debunking homeopathy all day makes you worse at hard cases - easy skepticism as cowpox.
The phantom refutation: 'that's been debunked' - claims get memory-holed without ever losing an argument.
The keystone: epistemic learned helplessness - refusing to update on clever arguments is often correct.
The rehabilitation: scientific consensus has a shockingly good track record - once you know where it actually lives.
The other side: giving contrarians and crackpots their tiny, real probability - without falling in.
The structural lens: Chomsky's filter model - media bias as economics, not conspiracy.
The model: bounded distrust - the elaborate signals that let you extract truth from sources that hate you.
The anchor: the media very rarely lies - it misleads by selection and emphasis, which is subtler and worse.
The defense: Scott answers the critics - sharpening exactly what 'rarely lies' claims and why it matters.
The synthesis: priesthoods - experts as guilds with internal virtues and external contempt, and how to read them.
The hard problem from every angle: Jaynes's wild theory, enlightenment as a technical skill, mystical insight you can't bring back, and whether AI changes the question.
4 essential posts · core path10 posts · full arc
The opener: the zombie preacher of Somerset - what consciousness pathologies reveal.
The wild theory: Jaynes's bicameral mind - consciousness as a cultural invention, taken seriously.
The technical manual: mastering the core teachings of the Buddha - enlightenment as a skill tree.
The practice: The Mind Illuminated - what systematic meditation actually trains.
The koan: Samsara - a world where everyone's enlightened except one man; fiction does the argument.
The classic: universal love, said the cactus person - mystical insight that won't compress into words.
The trust problem: unfalsifiable internal states - can you believe the jhana reports?
The mechanism: Byrnes on trance - hypnosis as predictive processing.
The frontier: the new AI consciousness paper - the hard problem, now with transformers.
The capstone: the colors of her coat - qualia, wonder and what AI art costs us.
Why everything costs ten times more and grows slower: cost disease, credentialism, and the long bend in the growth curve.
3 essential posts · core path9 posts · full arc
The parable: tulip subsidies - credentialism explained in one imaginary flower market.
The anchor mystery: education, healthcare and housing cost 10x more - and nobody quite knows why.
The economist's answer: the Baumol effect reviewed - how far 'it's wages' actually gets you.
From the clinic: what cost disease looks like when you are the doctor billing through three intermediaries.
The school fight: why 'just fund public schools' misses the cost curve entirely.
The arms race: college admissions as zero-sum credential inflation.
The theory: Caplan formalized - if school is mostly signaling, the waste is the point.
The decoupling: wage stagnation examined - productivity, compensation, and where the money went.
The zoom-out: growth was hyperbolic for millennia - 1960 is when the singularity got cancelled.
Putting numbers on beliefs and being graded on them: calibration, overconfidence, superforecasters, and acting under uncertainty when it counts.
3 essential posts · core path9 posts · full arc
The foundation: confidence inside vs outside the argument - your model can be certain while you should not be.
The toolkit: concrete techniques for turning hunches into numbers.
The virtue: overconfidence dissected - what 99% should mean, and how often it is wrong.
The provocation: is there a general factor of correctness - a transferable skill of being right?
The science: Tetlock's superforecasters reviewed - what the best predictors actually do differently.
The mindset: scout vs soldier - motivated reasoning as calibration's real enemy.
The trap: heuristics that almost always work - the security guard who says 'it's nothing' is 99% right and worthless.
The audit: grading prophets from Nostradamus to Fukuyama - what forecasting history teaches.
The capstone: February 2020 - a failure of action, not prediction; probabilistic seriousness when it counted.
A historical arc from the 2013-14 neoreactionary moment: Scott's famous steelman, the definitive rebuttal, and the decade-later postmortems. The rebuttals are the point.
3 essential posts · core path10 posts · full arc
The prelude: dissecting Moldbug's formalism on LiveJournal, before SSC existed.
The steelman: reactionary philosophy in an enormous, planet-sized nutshell - the fairest hearing NRx ever got.
The data check: 'apart from better sanitation and medicine...' - was the past actually better? No.
The rebuttal: the Anti-Reactionary FAQ - point by point, the definitive answer.
The order question: Empire/Forest Fire - does autocracy actually deliver stability?
The sociology: right is the new left - why NRx got cool, explained by fashion dynamics.
The steelman revisited: cultural evolution - the strongest surviving reactionary argument.
The addendum: competing selectors - Scott updates the Nutshell's core mechanism.
The postmortem: whatever happened to neoreaction? Where the people and ideas went.
The final word: Moldbug sold out - the 2025 verdict on the movement's author.
The working psychiatrist's tour of the medicine cabinet: what the evidence says about SSRIs, stimulants, melatonin and the drugs everyone argues about.
4 essential posts · core path12 posts · full arc
Start here: SSRIs - what the effect sizes actually show, from someone who prescribes them.
The field's weirdness on display: prescriptions, paradoxes and perversities - why drug data makes no sense.
The big meta-analysis: Cipriani on antidepressants, journal-clubbed honestly.
The 2018 revision: SSRIs updated - where Scott changed his mind.
The gene-test industry: antidepressant pharmacogenomics - a product ahead of its evidence.
The stimulant question: Adderall's real risks, weighed without panic or salesmanship.
The practical one: melatonin - the right dose is 20x smaller than the bottle says.
The provocation: is pharma research worse than chance? Ketamine and MDMA came from users, not labs.
New-drug skepticism: vilazodone and vortioxetine - what 'me-too with a twist' looks like.
Ketamine in practice: the peer-review-request clinical guide.
Applied psychopharm: the FTX crash read through a prescription pad.
The 2024 mystery: why does Ozempic cure everything? GLP-1 and the future of the field.
Turchin's mathematical history: elite overproduction, the violence cycle, and whether the 2020s were on schedule - a compact arc; read with skepticism enabled.
2 essential posts · core path5 posts · full arc
The theory: secular cycles - integrative and disintegrative phases across two millennia of states.
The application: Ages of Discord - the model says the US peaks in unrest around 2020. It was published in 2016.
The receipts: the highlighted passages - the raw claims worth arguing with.
The ancient case: Cyropaedia - asabiyyah and elite formation in the founding empire.
The micro-scale: subcultures run the same cycle - Turchin logic applied to scenes and movements.
The hidden economy of status: signaling, countersignaling, fashion cycles and class markers - the invisible game everyone denies playing.
3 essential posts · core path12 posts · full arc
The primer: what signaling actually is - honest signals cost more to fake than to have.
The classic: intellectual hipsters and meta-contrarianism - the three-level status ladder of opinions.
The engine: right is the new left - ideology as fashion cycle, always fleeing the class below.
The conversation layer: phatic speech and the anti-inductive - why small talk and markets eat every strategy.
The institution: education as signaling - the graduation speech that says the quiet part.
The taxonomy: class in America - what money can't buy and markers can't hide.
The source text: Fussell on class, reviewed - the field guide to the markers.
The office: the Gervais Principle - sociopaths, clueless and losers as signaling strategies.
The scene lifecycle: a cyclic theory of subcultures - status gradients birth and kill communities.
The payoff: the rise and fall of online culture wars - fashion-cycle dynamics explain a decade.
The coalition game: lifeboat games and backscratchers clubs - costly loyalty signals.
The deep end: Sadly, Porn - everything as status and envy; read last, disagree freely.
How much do you owe the drowning world? Effective altruism from spreadsheet to kidney - and the case for staying sane while giving.
4 essential posts · core path12 posts · full arc
The founding move: charity as optimization - do unto others as effectively as possible.
The vertigo: infinite debt - if you can always save one more life, where does obligation end?
The answer: nobody is perfect, everything is commensurable - give ten percent and be at peace.
The discipline: stop adding zeroes - expected value without the vertigo.
The caution: beware systemic change - why EA's boring interventions are a feature, not cowardice.
The texture: EA Global from the inside - earnest, awkward, and quietly moving.
The economics: moral standards as incentive design - set the bar where people will actually jump.
The defense: billionaire philanthropy - why hating it is mostly aiming at the wrong target.
The theory: Just Giving - what philanthropy is for in a democracy, reviewed.
The flesh: My Left Kidney - effective altruism arrives at the operating table, hilariously.
The revisit: more drowning children - the founding thought experiment, stress-tested twenty years on.
The verdict: in continued defense of EA - the receipts, tallied after the scandal.
What IS a mental illness? Taxometrics, nosology and the unsettling evidence that culture helps write the symptom list.
3 essential posts · core path11 posts · full arc
The dissolution move: 'is it really a disease?' is a boundary dispute, not a fact dispute - the toolkit for everything that follows.
Diagnosis as base rates: why over- and under-diagnosis happen at the same time.
Symptom, condition, cause: what a diagnosis actually names - psychiatric nosology in one post.
The stakes: against against autism cures - what the disease/identity framing fight is really about.
The anchor: taxometrics - are psychiatric conditions categories or dimensions? The evidence says mostly dimensions.
The upgrade: conditions as dynamical systems - depression as an attractor state you fall into.
The evolutionary lens: conditions as tradeoffs - why the genes stick around.
The political limit: you don't WANT a purely biological taxonomy - values are load-bearing in the DSM.
The definitional guardrail: contra evolutionary definitions of mental illness.
The wild card: culture exports madness - how anorexia and depression spread like memes.
The capstone: culture-bound syndromes - expectation writes illness, and predictive processing explains how.
Your inner life is not the human default: the typical mind fallacy, hidden experiential variation, and why people genuinely inhabit different worlds.
3 essential posts · core path7 posts · full arc
The fallacy named: generalizing from one example - your mind is a sample size of one.
The political upshot: typical-minding the other tribe - why your intuitions do not transfer.
Early application: gender identity via body maps and phantom limbs - typical-mind at its sharpest.
The survey: aphantasia, asexuality, inner monologues - the experiences you did not know you were missing.
The hard question: can people be honestly wrong about their own experience? Introspection has limits.
The comedy of variation: completely different mental processes, never once mentioned aloud.
The capstone: different worlds - perceptual bubbles, niceness fields, and why your reality is not shared.