Is the US a ruthless cognitive meritocracy that reliably promotes outlier talent? VB Knives defended that claim in a Twitter argument against Living Room Enjoyer that got my attention. ^[1] Knives argued that if you have a 150 IQ, you'll be a National Merit Scholar, which "at a minimum" gets you a free ride at a state flagship university, from which you can proceed to law school, med school, etc. Enjoyer shot back: I'm a Merit Scholar, where's my free ride? Knives asked Grok, Elon Musk's AI; Grok recommended the University of Alabama, ranked #169.

How elite is elite?

About 1.3 million high school juniors take the PSAT each year. Around 16,000 become Semifinalists (top 1.2%), of whom about 95% become Finalists. Of those 15,000 Finalists, only about 6,930 receive any NMSC-administered scholarship at all. The best-known category is a one-time $2,500 payment; most other awards are corporate- or college-sponsored.

The prospect of a free ride comes from a handful of schools that use National Merit status as a recruiting tool. The University of Alabama (the example Grok cited in the thread) offers Finalists a package covering tuition for up to five years, housing, a $4,000/year stipend, and a $2,000 research allowance. Florida State offers a full out-of-state tuition waiver. UT Dallas covers tuition and fees for up to eight semesters plus housing. A few dozen other schools do similar things. ^[2] The common requirement for college-sponsored awards is that you designate the school as your "first choice" through NMSC's system. ^[3]

To estimate where these schools sit in the overall population, I calculated cumulative freshman enrollment by US News 2026 rank as a fraction of the 4.2 million surviving members of the relevant birth cohort. I used two methods: individual freshman class sizes from institutional fact sheets and Common Data Sets for the top 50 and key schools, and total undergraduate enrollment ÷ 4 as a cross-check. I then spot-checked averages for ranks 51-169 against published figures. ^[4]

The results:

• University of Florida (#30, full cost of attendance via the state Benacquisto Scholarship for Florida-resident NM Scholars at any eligible Florida public or independent institution): about 2.0% of the age cohort attends a school ranked this well or better. 98th percentile.

• Florida State (#51, best out-of-state full tuition waiver I found at a reasonably well-ranked school): about 4.5%. 96th percentile.

• UT Dallas (#110, canonical NM full-ride school): about 10%. 90th percentile.

• University of Alabama (#169, Grok's pick): about 16%. 84th percentile.

The remaining 84% of the cohort either attends a school ranked below #169, attends a liberal arts college or regional university outside this ranking frame, attends a two-year institution, or doesn't enroll at all. All top-50 schools combined enroll about 4.4% of each year's birth cohort. ^[5]

If the system were a ruthless cognitive meritocracy, the NM Semifinalist (one of 16,000 selected from a 4.2 million cohort — the top 0.4% by count) would find themselves surrounded in college mainly by people like themselves, the way students at Australia's selective-entry public high schools (James Ruse, Melbourne High, Mac.Robertson) or at the University of Melbourne are surrounded by cognitive peers selected primarily by exam. ^[6] Instead, our Semifinalist is grafted into a hybridized class that mixes high-g outliers with the broader 84th-96th-percentile population: the children of comfortable professionals, kids with good-but-not-extraordinary test scores and polished applications. The NM Finalist at FSU or Alabama is not entering a community of intellectual peers. They're being absorbed into a much larger social stratum defined less by cognitive ability than by a mix of family resources, geographic convenience, and institutional compliance.

Suppose a bright youth is ruthlessly optimizing for the meritocratic track. They don't even have to be an extreme outlier with an IQ of 150; they could have an IQ anywhere above 130, the 98th percentile. ^[7] They score in the top 1.2% of PSAT test-takers, becoming a National Merit Semifinalist. They become a Finalist. They designate one of the nine Florida NMSC college-sponsor institutions as their first choice, which makes them eligible for a college-sponsored scholarship (sponsor schools typically offer these to all or most designating Finalists, converting them to Scholars; University of Florida at #30 is the obvious pick for a Florida resident, since the state's Benacquisto Scholarship then covers full cost of attendance at any eligible Florida institution; if they are not fortunate enough to already reside in Florida, they can attend FSU at #51). ^[8] They attend for free. They graduate with a strong GPA. They score well on the LSAT or MCAT. They get into an elite professional school such as a T14 law school or equivalent. ^[9]

How strong is this sort? Sarah Constantin's taxonomy of extreme elites distinguishes "One-Percenters" (groups of a few million: engineers, programmers, people with IQ over 130), "Aristocrats" (hundreds of thousands: doctors, lawyers, Ivy alumni), and "Elites" (tens of thousands: Google engineers, AIME qualifiers). The total number of people entering some recognizably elite professional or graduate track each year is probably 20,000-40,000, or about 0.5-1.0% of the age cohort — Constantin's "Elite" tier. ^[10]

This bright youth, if they also happen to be more enterprising than neurotic, on good enough terms with their school principal to get an endorsement, and have noticed the opportunity, can reach a credentialed position roughly commensurate with their ability rank. The financial barriers at each stage are low. Undergrad is free if you play the NM game. Professional school is debt-financed against high expected returns; federal lending covers it without requiring family money.

But the attribution of "ruthless meritocracy" to the system is misplaced. If the entrant ruthlessly exploits the meritocratic elements of the system, they can reach a social rank commensurate with their g without taking the sorts of risks an enterprising young man would have needed to take in, say, medieval Europe, to reach a similar rank.

Think of it this way. There is a system in this country where you take a test in high school, and if you score really well, the system sends you a letter saying congratulations, you are very smart, and then — this is the important part — does nothing. It's like when your dentist sends you a reminder postcard. Except the postcard says "You are a genius" and the next step is "Figure it out yourself, genius."

Caroline Hoxby and Christopher Avery's "The Missing 'One-Offs'" showed that high-ability, low-income students, particularly outside metro areas, systematically undermatch. ^[11] The binding constraint isn't financial access (elite schools often have generous need-based aid) but the kind of strategic knowledge about how to navigate the system that is distributed by class, geography, and school counselor quality, not by cognitive ability. This is concentrated not at the 99.5th percentile, where the system's detection apparatus engages, but around the 90th, where it barely notices you.

What meritocracy was for

The word "meritocracy" entered English from Michael Young's 1958 dystopian satire The Rise of the Meritocracy, but the concept is older. Napoleon's "la carrière ouverte aux talents" is usually cited as its modern origin. Revolutionary France was fighting simultaneous wars against multiple European coalitions while suppressing internal insurrections, having just guillotined much of its hereditary officer class. The old-regime army had numbered roughly 150,000-200,000; the levée en masse, the first modern mass conscription, produced over 750,000 men under arms by 1794.

Promotion on the basis of merit was a desperate response to the problem that the people who had been running the army on the basis of birth were dead or in exile, the army was five times larger than it used to be, and the external performance pressure (lose and the Republic falls) was existential. Before the Revolution, 90% of officers had been aristocrats; by 1794, only 3% were. Even then, Napoleon's own practice drifted back toward class selection: his lycées were few and expensive, his officer corps increasingly recruited from "good families," and Rafe Blaufarb's The French Army, 1750-1820: Careers, Talent, Merit concludes that the Revolutionary meritocracy "institutionalized a concept of merit that combined talent with social rank and family ties." Stendhal's novel The Red and the Black (1830) depicts the result: Julien Sorel, a carpenter's son who idolizes Napoleon and dreams of rising by military merit, finds that after Waterloo the army (the red) no longer offers a path for men of his class. Only the clergy (the black) does. He advances through the seminary, through seduction and the careful management of appearances and relationships (one is reminded of Dale Carnegie's How to Win Friends and Influence People, or Castiglione's The Book of the Courtier).

The American system is not under that kind of pressure, but it does have procedures for promotion partly on the basis of test scores. Poor and unconnected high-g types can use these procedures to climb from a high PSAT score to a top professional school degree while spending very little money, if they know the tricks and are willing to spend a decade on it. The result is something like a hybridized mandarinate: a credentialed administrative class selected partly by examination and partly by family resources and social compliance, whose training consists largely of learning to operate the system that selected them. This system affords somewhat more meritocratic upward mobility than the Ottoman or Austro-Hungarian Empire. ^[12]

Not all American elites come through this pipeline; the truly wealthy bypass it, and some technical fields still select on demonstrated ability. But the professional-managerial class that staffs law firms, consulting firms, hospitals, and bureaucracies is produced by it, and the cognitive outliers who enter it do not find themselves concentrated among peers the way they would in a meritocratic sorting. They find themselves dispersed through a much larger population selected on a mix of test scores, grades, compliance, and parental resources.

The compliance pipeline

When I think about where I'm dependent on the applied intelligence of others — airline pilots, nurses, plumbers, power grid operators, and probably still most people who work at Amazon regardless of job title or school — I'm mostly thinking about people whose competence is tested against reality on a daily basis, not people whose competence was certified once by an admissions committee and never checked again. The credentialing pipeline at its highest levels selects for something different from what keeps planes flying in the air or patients alive in the ICU: it selects for the ability to keep producing legible outputs for evaluators, not the ability to notice when something is wrong.

When the United States has faced problems where deploying high-g people mattered, problems with external performance pressure comparable to Revolutionary France's, it built specific institutions to do it. The Manhattan Project built the bomb. The MIT Radiation Laboratory recruited physics professors from across the country and had them building radar systems that worked; the slogan was "the physicists built the bomb, but radar won the war." Los Alamos hired Feynman at twenty-four. During the wars, incompetents in positions of authority were exposed and replaced, and the people who survived on merit found themselves in charge. After the wars, some of those people built institutions that preserved wartime selection principles in peacetime: Bell Labs had the people who invented the transistor, information theory, Unix, the laser, and the solar cell, all under one roof in Murray Hill, New Jersey. DARPA funded the original internet. These were not credentialing pipelines. They hired people who had demonstrated the ability to solve hard problems, gave them hard problems, and got out of the way.

The modern professional-managerial pipeline is not that. It recruits intelligence into a compliance meritocracy ^[13] , the sorts of loyalties that generate a class of 24/7 vibes-performing PMC apparatchiks instead of the hypercompetent manager-leaders who won the World Wars. Robert Jackall's Moral Mazes documented this transformation in corporate management: the replacement of technical competence and independent judgment with an anxious attunement to what the people above you want to hear.

Michael O. Church analyzes American class as three parallel ladders: Labor, defined by work, reputation, and personal accountability; Gentry, defined by education, professional identity, collegiality, and cultural influence; and Elite, defined by control of large enterprises that insulate them from accountability. This gives us more adequate terms to describe what the credentialing pipeline does. It does not move people up the Labor ladder; it teleports them from Labor to middle Gentry with an option to audition for Elite status, ^[14] adopting the values and anxieties of a social infrastructure that, past a certain point, checks not for credentials or competence but for whether you're part of the vibe.

There are excuses of a sort, offered for the non-test elements of admission and advancement: GPA maintenance, extracurriculars, "holistic" criteria, the "leadership" requirement that selects for a specific flavor of self-promoting sociability. Some say that they screen out mere test-taking machines, but in favor of what?

A medical student I know was working in a hospital where stated policy diverged consistently from implemented practice on the ward. ^[15] She then had to take an ethics exam that posed a scenario drawn from exactly that gap. She knew the "correct" answer was to say she'd follow the written policy. She gave it. She knew it was a lie. The evaluators knew the ward didn't work that way. The function of the test was not to assess ethical reasoning. Or rather, the function of the test was to assess ethical reasoning, and suppress it. It was to identify that she could and would identify which audience she was performing for and produce the expected output, without ever forcing the contradiction into common knowledge. ^[16]

The NMSC application requires a principal's endorsement, "leadership," and social legibility. Paul Graham's "Why Nerds Are Unpopular" reports that teenagers whose attention is on the object-level world rather than on social navigation are a marginalized minority; popularity is a full-time job and they won't pay the full cost. (I think even having an identifiable minority category like that was a Gen X thing; nowadays "nerd" just means "member of a fandom.") Graham is a little unhelpful, though, because he relegates normality to the background as an unexamined default from which nerds deviate. Ayn Rand's "The Comprachicos" (1970) problematizes it: ^[17] the Progressive education system replaces cognitive training with "social adjustment" starting at age three, producing the pack-reading baseline Graham takes as given. John Taylor Gatto's The Open History of American Education develops a compatible thesis from the same historical material, tracing the design of compulsory schooling as a vehicle for social engineering from its Prussian origins through its American adoption. His Prussian-model genealogy is well-documented in mainstream education history, as the annotated text at the link attempts to show. An academically validated analogue of this argument is Bowles and Gintis's Schooling in Capitalist America (1976), which documents what they call the "correspondence principle": the internal social structure of schools (hierarchy, fragmented tasks, extrinsic rewards, obedience to authority) mirrors the internal structure of the capitalist workplace, so that students arrive already habituated to the relationships of domination their jobs will require.

So an otherwise marginal high-g teenager applying for a National Merit scholarship is being asked to simultaneously win the acceptance of an anticognitive pack, ruthlessly instrumentalize the system, go along with being piecemeal instrumentalized and conditioned to follow orders, and then somehow emerge with enough independent orientation left over to do something worthwhile once they've won their way into the elite.

I was a National Merit Semifinalist. I was in the 5% who never became Finalists. I didn't complete the application. The whole thing seemed like it was asking me to perform a kind of narcissism: to collect figurehead titles in high school clubs and then boast about them, as though being elected secretary of the Latin club contributed to my community or meant people respected and relied on me for my sound judgment, work ethic, and initiative. It would have been different if any of it had been organized around meeting some real need. Instead the process seemed to be asking me to pretend that my value was constituted by the approval of the same institutional structure that was conspicuously wasting everyone's time.

I was unhappy in high school. I didn't write treatises like this about it, but when anyone asked I was articulate enough to say that it didn't seem like I was being prepared to be of real use to anyone. I was visibly unhappy enough that my mother sent me to a psychiatrist. The psychiatrist prescribed low-dose Lexapro, gave my mother a photocopied magazine article about Temple Grandin, and told her that while I didn't technically have Asperger's, this would help her understand me better. I was the back-row thinker the system had correctly detected and was now asking to perform enthusiastic consent.

I was not typical of the people Hoxby's research describes. I grew up in an upper-middle-class Jewish family in Stamford, Connecticut; some of my high school classmates ended up at Harvard or Princeton; my family was close with a legacy Yale family from the time I was in preschool; I've shared meals with billionaires and my former housemates are people the New York Times and New Yorker write about. I'm not the kind of person who's supposed to get lost by the system. I went to an alternative primary school for kids whose parents didn't want to put them through the ordinary schooling meat grinder, and even there, at least one classmate who stayed on track went Ivy League and medical school.

The cases common enough to show up as a statistical signal look different from mine: Hoxby and Avery found 25,000-35,000 high-achieving, low-income students (top 10% of test-takers, A- GPA or above) who don't apply to any selective college, even though selective schools would often cost them less than the community colleges they actually attend, because of financial aid. These students come from districts too small to support selective public high schools, aren't surrounded by other high achievers, and have never met a teacher or older student who went to a selective college. Seventy percent of high-achieving students who do apply to selective schools come from just fifteen urban areas. Steve Sailer's gloss on this is that the system works at the 99.5th percentile and fails at the 90th. But my case suggests something worse: even at the top of the top, even with every demographic advantage, the system isn't looking for people who won't perform enthusiastic consent. It sent me an invitation to audition, but when I wouldn't dance, it silently moved on.

What does the pipeline produce? A person who successfully navigates it spends roughly a decade in continuous performance mode: maintaining GPA, prepping for standardized tests, performing in professional school, billing hours. They end up credentialed and well-compensated. And that's it.

My son's fifteen-month well-child visit was at a community health clinic in New Haven staffed by Yale-affiliated physicians. The pediatrician who saw him, a Yale resident still in training, came in, did what I understood to be the standard exam (stethoscope, abdomen, genitals), and then, without context, asked me what he was eating. I described his diet. When I mentioned that he sometimes ate processed snacks like Goldfish crackers, she told me to cut back on those. She didn't say why. I asked. She said it was his weight. I said, "I know he's big, but he's also tall." She said she was going by his height-weight ratio. That's when I remembered: they hadn't measured his height that day. The ratio the pediatrician was using to tell me to feed my toddler less had a fictitious denominator. She hadn't noticed. She hadn't looked at the child on the exam table and thought does this kid look fat? She had looked at a number on a screen and dispensed advice. Once the nurse came back and measured his height, nobody had anything more to say about my supposedly obese toddler. I'd had the sense for a while that there was no channel for information from an ordinary doctor's visit to flow upward into the system and cause heuristics to get reexamined. Unless, of course, it's not an ordinary visit at all, but one performed under the aegis of a study design and blessed by an IRB. But it's easy to second-guess that kind of impression, until you run into clear, unambiguous evidence of gross inattention. ^[18]

A physician still in the credentialing pipeline — college GPA, MCAT, medical school, now residency — has had her cognitive ability verified at every stage. But the system that verified it now gives her eight-minute visits, metrics-driven reviews, and a workflow that treats the chart as the patient. She is not being asked to notice things. She is being asked to comply with a documentation protocol. The intelligence the system selected for is not being used; it is being actively suppressed in order to process more units.

Elite lawyers mostly work to fight elite lawyers. The professional-managerial class functions as a self-perpetuating job-creation scheme, its members selected not for the ability to solve problems but for the disposition to manage them in ways that preserve the positional advantages of the class they've been recruited into. The financial-political system that produces demand for this is a debtor-aristocracy in which capital is not allocated on the basis of productive capacity, but rather members of the job-creation class act as gatekeepers with preferential access to low interest rates.

The modern credentialing pipeline imposes exactly this tax. The hoops are not individually onerous. But the disposition required to keep jumping them, the learned vigilance about what the next evaluator wants, is structurally antagonistic to the kind of cognition that produces work commensurate with 150-IQ capacity. Paul Graham identified this pattern in "The Lesson to Unlearn" and then built Y Combinator to select for exactly the hoop-jumping disposition he was criticizing. The thing being selected for is anxious responsiveness to authority, not the capacity to sit with a hard problem for months.

Isaac Newton had a Cambridge fellowship that was basically a sinecure. He sat in his room thinking about Galileo and Kepler and the motions of the celestial bodies, and then (the story goes) an apple fell on his head and he saw a way to unify terrestrial and celestial mechanics.

Albert Einstein was working as a patent clerk in Bern, spending his free mental bandwidth on a specific technical problem: Newtonian gravity assumes instantaneous action at a distance, but special relativity prohibits faster-than-light causal influence, so gravity as Newton described it can't be right. He'd been turning this over for two years when one day, sitting in his office chair, he was startled by the thought that a person falling freely would not feel their own weight. He later called it "the happiest thought of my life." It became the equivalence principle: gravity and acceleration are locally indistinguishable, which means gravity isn't a force propagating through space but a property of the geometry of spacetime itself. The insight took eight more years to work out mathematically, but the flash came while he was sitting in a chair, not navigating a credentialing pipeline.

Spinoza ground lenses, and lived cheaply, which gave him plenty of time to think about foundational problems profoundly enough that the next generation of great philosophers all understood themselves to be living in his world.

The pattern is the same in each case: a mind that had been dwelling on a hard problem for a long time, in conditions of sufficient slack to let the pieces rearrange themselves.

What's the big deal? Spinoza, Newton, and Einstein had their political realities to accommodate; we have ours. Why would we think that kids these days don't stand a chance?

Newton ran the Royal Mint, prosecuted counterfeiters, and sat in Parliament. But Isaac Barrow had talent-scouted him at Cambridge, championed his work, and resigned the Lucasian Chair so Newton could have it at twenty-six. The revolutionary work was already done. The politics came after.

Einstein had a harder road: quit high school at fifteen, failed an entrance exam, graduated with an undistinguished record, spent two years unable to find academic work, and got the patent clerk job through a friend's father. That was all before the 1905 papers. Afterward came fleeing Germany, helping other scientists get out, and lobbying Roosevelt to build the bomb. Despite all this, Einstein seems to have believed that things had deteriorated badly between his youth and 1954, when he told The Reporter that if he were a young man again, he would not try to become a scientist: "I would rather choose to be a plumber or a peddler in the hope to find that modest degree of independence still available under present circumstances."

Half a century later, Peter Higgs was no Einstein, but he did identify the mechanism by which subatomic particles acquire mass. Shortly after receiving his Nobel Prize, he said in 2013 that he would not get an academic job today because he would not be considered "productive" enough: "It's difficult to imagine how I would ever have enough peace and quiet in the present sort of climate to do what I did in 1964." These are not outsiders complaining about a system they failed to enter. These are people who won an earlier generation's meritocracy and are telling you the current one wouldn't have let them play.

Gottfried Wilhelm Leibniz was brilliant like Newton and Einstein, but he was an anxious striver. He co-invented calculus, and his notation (dy/dx, the integral sign) proved so superior to Newton's that continental mathematicians using Leibniz's system pulled ahead of the British for most of the 18th century; the British eventually had to abandon Newton's notation and adopt Leibniz's. He identified the conservation of kinetic energy (his "vis viva"), anticipating what became a cornerstone of physics. He did important work in formal logic that wouldn't be appreciated until the 19th century. He conceived of binary arithmetic, the characteristica universalis (a formal language for all reasoning), and the idea that space and time are relational rather than absolute, which Einstein would vindicate two centuries later. But the gap between what Leibniz did and what he could have done is visible.

The Internet Encyclopedia of Philosophy sums up the gap honestly: Leibniz's mathematics made a significant difference in 18th-century European science; his contributions as engineer and logician, however, were relatively quickly forgotten and had to be re-invented elsewhere. Leibniz himself was aware of the problem. He wrote: "If I were relieved of my historical tasks I would set myself to establishing the elements of general philosophy and natural theology, which comprise what is most important in that philosophy for both theory and practice." He spent decades compiling a genealogical history of the House of Brunswick because that's what his patron required. He never finished it.

There was a wedge forcing open that gap between Leibniz's extreme potential, and his merely great achievements: patron management. Leibniz had to constantly attend to his relationships within the Hanoverian court, pursue the priority dispute with Newton, seek appointments, and produce work-for-hire genealogy. The tax on his attention was not that any single task was strenuous, but that none of them could be safely ignored. Newton could ignore the world for months; Leibniz could not.

Matthew Stewart's The Courtier and the Heretic argues that Leibniz's metaphysics was substantially motivated by the need to preserve the metaphysical privilege for old-regime power relations that Spinoza's monism eroded. Where Spinoza's one-substance ontology amounted to a radical leveling (no part of nature is closer to God than any other), Leibniz's otherwise unmotivated commitment to metaphysical pluralism (the monads, the pre-established harmony, the "best of all possible worlds") preserved a hierarchical cosmos in which some beings are genuinely higher than others, which is the kind of metaphysics a court philosopher serving a Holy Roman elector needs. Einstein implied Spinoza powered his thought; Leibniz's metaphysics embedded the social commitments of his patronage, and those commitments likely fractured and enervated his. ^[19] That Leibniz felt the need to oppose Spinoza may have denied him the class of insight available to Einstein. The patron didn't just take Leibniz's time. It shaped what he was willing to think.

The credentialing pipeline can be thought of as a gigantic robot patron, operating at hundreds of thousands of times the throughput of the House of Brunswick, ceaselessly converting creative intelligence into loyalty. It does not just take a decade of your time. It shapes what you are willing to think. The system does not try to make use of intelligence. It does not identify 90th-percentile people and pull them toward positions of moderate responsibility and influence. It does not give 99th-percentile people hard problems and room to work on them. VB Knives's framing, that the system works if you just know how to use it, makes the solution look like an information problem: just tell smart kids about National Merit scholarships and LSAT prep. Solving the information problem gets more 150-IQ kids into large law firms. Is that really our problem? Not enough clever elite lawyers?

---

The stone which the builders rejected is become the head stone of the corner. — Psalm 118:22

1. The thread started with Living Room Enjoyer (@Emptier_America): "Conservatives above the age of 40 genuinely believe that we live in a 'ruthless cognitive meritocracy,' and that intelligent people who don't flourish are lazy or otherwise defective. In fact, 150 IQ White kids from Montana are priced out of elite schools if not outright denied." </br>VB Knives (@Empty_America) responded: "If you have 150 IQ, you are going to be a national merit scholar, which at a minimum leads to attending a state flagship research university for free, from there you would have scores to attend a good Law School, Med School, etc. But many of you wouldn't know this..." </br>Living Room Enjoyer replied: "I am a National Merit Scholar. Where is my free ride to a state flagship university?" </br>VB Knives then asked Grok to help, and Grok cited the University of Alabama package. </br>Living Room Enjoyer quoted: "Thanks, I'll take Grok's advice and attend the University of Alabama (ranked #169 nationally) because I'm a Highly Valued Genius. Does this constitute 'ruthless cognitive meritocracy'? Is that the best you have to offer your best and brightest?" </br>𝔐𝔽𝓩 (@mean_field_zane) chimed in: "Actually, the scholarship only happens if you're low income, and it's only $2500." </br>VB Knives had also previously tweeted (Jan 8, 2025): "The relatively small number of 'DEI' slots serves as a distraction from the fact that the USA is a rather ruthless cognitive meritocracy at every level. We are tested, sorted, and tracked from childhood with G-loaded tests. 8/9 PSAT, PSAT 10, NMSQT, SAT, LSAT, GRE, GMAT." </br>Separately, Steve Sailer responded: "The research by this black lady economist at Stanford whose name I forgot is not that there are many overlooked white boys at the 99.5th National Merit Scholar percentile, but that a lot of red state white boys around the 90th percentile get badly overlooked by the system." He then followed up: "Professor Caroline Hoxby." ↩︎

2. These numbers are from NMSC's published materials for the 2026 competition and the schools' own scholarship pages (University of Alabama, UT Dallas, University of Tulsa, among others). On the non-transferability of college-sponsored awards: Florida's Benacquisto FAQ states that "once the National Merit Scholarship Corporation has mailed out an offer, the student will not be able to change the institution designation." ↩︎

3. This isn't quite the same as binding early admission (you don't forfeit a deposit if you change your mind), but it's exclusive: NMSC only notifies one school per student, and if that school makes you an offer and NMSC announces it, you can't transfer the sponsorship. So it functions as a soft commitment that forecloses other NMSC college-sponsored awards. ↩︎

4. The denominator is the surviving members of the birth cohort. US births in 2006-2007 averaged 4.29 million (CDC vital statistics). Cumulative mortality from birth to age 18 (infant mortality (0.56%, CDC WONDER) plus childhood and adolescent deaths (0.5-0.7%, derived from CDC mortality tables)) reduces this to approximately 4.2 million. The WICHE projection of 3.9 million annual high school graduates is about 7% lower. As a cross-check: the NCES status dropout rate for 16-24 year olds was 5.3% in 2022 (about 2.1 million people in a 9-year age band). Applied to a single-year cohort of 4.2 million, that implies 220,000 dropouts; adding these to the 3.9 million graduates gives 4.12 million, leaving an 80,000 residual attributable to GED recipients still in progress, emigration, and institutionalization. Using graduates rather than the birth cohort as the denominator would make the system look slightly more intensive than it is — every dropout is someone the system didn't reach. For the enrollment numerator: I looked up individual first-time-in-college (FTIC) enrollment from institutional fact sheets, Common Data Sets, and IPEDS Fall Enrollment reports for the top 50 and key schools, and cross-checked using total undergraduate enrollment ÷ 4. The first method gives a top-50 total of 185,000 freshmen (4.4% of the 4.2M cohort); the second gives 213,000 (5.1%). The gap is mostly transfer students inflating undergraduate totals at large publics. For ranks 51-169, spot-checking published FTIC enrollment (Penn State 9,500; CU Boulder 7,400; Binghamton 4,050; Baylor 3,550; UT Dallas 4,900) gives an average higher than 3,000-3,500, because the distribution is right-skewed by large publics; 4,000-4,500 is more defensible. The underlying data is available as a spreadsheet. ↩︎

5. Over a hundred schools participate in NMSC's college-sponsored scholarship program. FSU at #51 is the best-ranked school I found that publishes explicit out-of-state NM full tuition coverage. USC (#29) offers NM Finalists $20,000/year against tuition of $73,260. Some schools in the #50-100 range offer partial awards or competitive NM scholarships. ↩︎

6. Australia concentrates gifted students into selective-entry public high schools via competitive exam (in New South Wales and Victoria, the two largest states). Its flagship universities (Melbourne, Sydney, ANU) select primarily on the ATAR (Australian Tertiary Admission Rank), a standardized percentile rank derived from final secondary-school exam performance. The University of Melbourne's admissions page states: "Applicants are selected according to academic merit, in the form of the ATAR"; a personal statement is not required. Small alternative pathways exist for equity (Access Melbourne), Indigenous students, elite athletes, and mature-age applicants, but the standard channel is exam-rank-driven — in contrast to the US system, where elite schools predominantly center "holistic" criteria whose history traces to early 20th-century efforts to limit Jewish enrollment (documented in Jerome Karabel's The Chosen: The Hidden History of Admission and Exclusion at Harvard, Yale, and Princeton, Houghton Mifflin, 2005; winner of the ASA Distinguished Scholarly Book Award) and where the litigation record in SFFA v. Harvard (2023) documented systematically lower "personal ratings" for Asian American applicants, a disparity the district court found "not fully and satisfactorily explained." My impression is that the Australian selective-school pipeline fed a disproportionate number of early Effective Altruism figures into the same intellectual milieu: Toby Ord, Katja Grace, and Rob Wiblin all came through the University of Melbourne. ↩︎

7. About 1.3 million juniors take the PSAT each year out of a 4.2 million age cohort, so PSAT-takers are roughly the top 70% of the cohort by self-selection (the bottom 30% don't take college-prep tests). NM Semifinalists are the top 1% of PSAT-takers per state. The top 1% of the top 70% is approximately the top 0.7% of the full population. On a normal distribution (mean 100, SD 15), the top 0.7% corresponds to roughly IQ 135-137. The actual threshold is somewhat uncertain because the degree of selection into PSAT-taking varies — if PSAT-takers are more representative than "top 70% by ability" implies, the IQ equivalent drops a few points. IQ 130-135 is a likely range. SAT-family tests are strongly g-loaded: Frey and Detterman's "Scholastic Assessment or g?" (Psychological Science, 2004) found a correlation of .82 between SAT scores and g extracted from the ASVAB. ↩︎

8. Other schools offer automatic full-tuition-or-better NM packages, but they cluster in the #100-250 range: Oklahoma (#110), UT Dallas (#110), Texas Tech, University of Houston, and various smaller schools. FSU at #51 is the clear outlier in quality among schools offering this deal to out-of-state students. Choosing anywhere else roughly halves the apparent selectivity of the institution on your transcript, which may or may not matter to graduate admissions committees. ↩︎

9. The LSAT has a mean of about 152 and SD of about 10 (LSAC percentile tables). LSAT-takers are a positively selected population — college graduates planning to apply to law school — so the same percentile on the LSAT corresponds to a higher general-population percentile than on an IQ test. Mensa accepts LSAT scores at the 95th percentile of test-takers as equivalent to their 98th-percentile-of-population IQ threshold (IQ 130). On current LSAC tables, the 95th percentile corresponds to an LSAT of roughly 170. An LSAT of 170 is approximately the median at the least selective T14 law schools — Georgetown, UCLA, and UT Austin all report 25th-percentile LSATs of 166 and medians of 171-172 in their 2025 ABA 509 disclosures (Georgetown, UCLA, UT Austin). So someone at IQ 130 is solidly competitive for T14 admission. Preparation typically adds several points beyond baseline. ↩︎

10. The "T14" law schools (legal industry jargon for the 14 schools that have historically occupied the top of the US News law school ranking) enroll roughly 6,000-7,000 new JD students per year. Top-20 medical schools add 3,500 matriculants. Elite professional tracks also include top MBA programs, top PhD programs, and some non-credentialed paths like founding successful companies. These overlap in the population they draw from, so the 20,000-40,000 total is a rough estimate, not a sum. ↩︎

11. Caroline M. Hoxby and Christopher Avery, "The Missing 'One-Offs': The Hidden Supply of High-Achieving, Low-Income Students," NBER Working Paper 18586 (2012); published in Brookings Papers on Economic Activity 2013(1), pp. 1-65. "High-achieving" is defined as 90th percentile on ACT/SAT comprehensive plus A- or above GPA, roughly 4% of US high school students. The 25,000-35,000 estimate of income-typical high achievers is from their analysis of College Board and ACT microdata. The 70% geographic concentration figure is from their comparison of achievement-typical vs. income-typical application patterns (Table 6 and surrounding discussion). Their central finding is that these students don't apply to selective schools not because of cost (selective schools would often be cheaper after financial aid) but because of information and application behavior: they come from districts too small to support selective public high schools, aren't in a critical mass of fellow high achievers, and are unlikely to encounter a teacher or schoolmate who attended a selective college. Thanks again to Steve Sailer for flagging this in a related thread. ↩︎

12. Both empires were substantially aristocratic, with family and class dominating access to positions of responsibility. The notable exception was the Ottoman devshirme, which sent recruiters to Christian villages, identified promising boys, and trained them intensively for positions of real power. The regulations stipulated one boy from every forty Christian households, roughly 3,000 annually at the system's peak; between the 15th and 17th centuries, some 200,000-300,000 boys passed through it. This was far more intensive than anything in the American system — the affected Christian population was perhaps 8-10 million, making the per-capita intake an order of magnitude higher than our 5,000 NM full-ride recipients from a population of 330 million. And the devshirme was actively searching for talent in communities that had no reason to volunteer it, whereas ours waits for the entrant to find the system. But it was a forcible brain-drain from a conquered subject population, not a within-group promotion ladder: Muslim subjects were excluded, the boys were converted to Islam and severed from their families, and the talent was deployed in service of a ruling group that the source communities had no part in. ↩︎

13. Bryan Caplan argues in The Case Against Education (2018) that roughly 80% of the economic return to schooling is signaling rather than human capital, and that the key signals are conscientiousness and conformity: an employer who wants workers at the 90th percentile of conformity needs to set the educational bar high enough that 89% of people give up despite the rewards. Robin Hanson's "School Is To Submit" makes the sharper claim: prestigious schools exist to habituate children into accepting workplace domination, by disguising it as copying high-status behavior. ↩︎

14. Church describes four rungs within each ladder. G4 (transitional Gentry — first-generation college students, community college transfers) is "how you get onto this ladder if you weren't born there." The pipeline deposits most graduates at G3 (established professionals — doctors, lawyers, mid-career engineers). The crossing to E4 (entry-level Elite — junior investment bankers, first-year biglaw associates) requires a further transition that Church associates with the G2 level (high Gentry — people with creative control of their work, respected institutional positions, or cultural visibility). ↩︎

15. A traveling nurse I know told me that it's standard practice to chart the time you were supposed to administer a medication or perform a test, not the time you actually did. ↩︎

16. The structure here is the same one Václav Havel identified in The Power of the Powerless (1978): the greengrocer who displays the sign "Workers of the world, unite!" is not expressing a belief but demonstrating that he will say what is expected of him. The ethics exam, the charting practice, and the NM application's "leadership" requirement all function the same way: they force participants to remember themselves as, and appear to others as, people who will tell or at least play along with any lie on command, which ruins their credibility as independent agents. For more, see Calvinism as a Theory of Recovered High-Trust Agency and Civil Law and Political Drama. ↩︎

17. Originally published in The Objectivist in 1970; collected in Rand's The New Left: The Anti-Industrial Revolution (1971), later reissued as Return of the Primitive: The Anti-Industrial Revolution (1999). ↩︎

18. I sent a written complaint to the clinic's Chief Medical Officer, who investigated and called me back within a few days. He told me they'd done a systemic analysis of height-weight measurements and found my case was probably a fluke, not a pattern. He also said they'd spoken with the physician involved. When I asked what he thought the right adjustment would be, he said something about how in his own medical training it had been emphasized that you listen to the patient and they'll tell you the diagnosis. In my and my partner's experience, younger physicians seem to limit touch-based and visual examination to what's specifically called for by a checklist. I think many of them are afraid to do more, lest they commit some kind of boundary violation. ↩︎

19. Einstein explicitly identified with Spinoza's monism ("I believe in Spinoza's God, who reveals himself in the harmony of all that exists") and spent the last decades of his life pursuing a unified field theory. The connection between "the universe is one substance" and "I want one set of equations for everything" is not subtle; plausibly Spinoza's vision of a unified rational cosmos made Einstein more willing to look for a geometric resolution to the gravity problem, one in which gravity is a property of spacetime itself rather than a force propagating through it. That Spinoza's day job was grinding lenses, not navigating a patronage network, may be reflected in the quality of his metaphysics. ↩︎