Hes definitely higher IQ than people think. Pretty much all of these people are high IQ.
Even Kanyes mom was a college professor who chaired the English dept at Chicago State University. To be super successful in the music industry in the way Kanye has been requires a lot of higher order thinking. It wouldn’t surprise me in the slightest if he was in the top 2 percent or so of black IQ, like 120+.
Interesting survey though, I wonder which ones were most accurate. Coincidentally I know for a fact my dads iq is 127 (he’s a VP of global manufacturing & logistics at a well known company, for reference).
Lol these are horrible. Way, way, *way* too low across the board. Based on numerous surveys, the average IQ of people in the broader rationalist sphere is in the vincinity of 130. All of these people are noticeably smarter than the average commentator. They have acquired their fame through mastery of technical subjects and winning fair-and-square in the Marketplace of Ideas. The average IQ here should be 145 or maybe even as high as 150.
I might write up a more detailed comment for how to better estimate IQ (as well as my own opinion on the IQ of some of the people here). But for now , here's an intuition pump for why IQ 145 should the baseline.
Google says that average high school in the United States has around 500 people. So the smartest person at a given high school will be in the 99.8th percentile in intellegence. If you convert that to a z-score, you get a score of 2.88 (that is, 2.88 standard deviations above the mean). You can then convert the z-score to IQ by multiplying by 15 and adding the product to 100 which gets you an IQ of 143.
So the average smartest person in a given high school has an IQ of 143. Eyeballing the graph, it seems that the average estimate for Charles Murray (the greatest social scientist of the last 50 years) is about 135. Do you guys really think that if you go to an average high school, you will find someone (let alone multiple people) smarter than Charles Murray? Are you guys insane?
My take on this is that people vastly underestimate how smart many popular writers are, and how much that contributes to their output.
130 IQ doesn't get you the kind of output on technical subjects that many of these guys put out. Say you what you want about Noah Smith, but my lower bound for him would be 140 (that should tell you something about the hardware Emil is probably working with).
In terms of raw intellect of people who are currently alive, it's hard not to see Robin Hanson and Gwern as gross underestimates. Gwern's writing reminds of me a bit of Chris Langan's (albeit much less insane), so it wouldn't surprise me if he was close to 200.
Gwern is plainly smarter than ASC, but ASC has him beat in trait psychoticism (maybe) and the type of unenviable phenotype that makes someone desperate for approval & hence apt to produce a lot.
Isn’t this all just garbage in-garbage out because most people are clueless about real IQ scores and have never taken a properly administered and normed test? Pretty sure most people think it’s linear like school grades and just assign “a bit more” IQ to some they think is “a bit more” smart.
I doubt anything above 130 iq would help much with substack writing, unless your substack was about theoretical physics or whatever. So it doesn’t surprise me that there are a lot of estimates around there or slightly above. 130 puts you in the top 2% of the population.
As I wasn't smart enough to know who all these people even were let alone how smart they are, I slunk back into my midwittery to read about the Dunning-Kruger effect on Wiki...again.
These are way too low. Newton comes in at 160, which is +4SDs. Over 10,000 people in the United States are over 160. There are not 10,000 people in America with more intellectual horsepower than Isaac Newton. There’s a strong case to be made that there are 0 in history. A more accurate estimation would have exceeded the limit of 200.
In some cases, we have public statements of their IQ, and the estimates are way lower. Aella has been tested at 140, Hanania has suggested he’s around 145, and in this thread you can find people providing better estimates of Yud and Yarvin.
I’m surprised at how bad people were at this given the community’s obsession with IQ. Even numbers like 160 aren’t that rare, and this is a selection of some of the era’s (and history’s) foremost intellects.
Greg Cochran says he was in the top half percent of scholastic achievement tests like SAT, GRE etc. back in the day. He also said he got over 700 SAT Math without prep. There are some reasonable estimates about IQ-old SAT/GRE conversions and this puts him above 145 easily.
Scott Alexander probably has an uneven cognitive profile due to lower correlation of sub-factors among high ability people. Guessing ASC has near 3std for verbal ability and lower (1.5-2sd) quant ability. If I'd have to take a shot I'd say his FSIQ is 134-135.
Semi-relatedly, one of the more interesting pieces of data I saw was that the average IQ of an Oxford math PhD is 128 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5008436/), so I guess you don't have to be exceptionally gifted to be very good at math. So "[X] is very good at math, s/he must be 140+" (which is an heuristic lots of people use) might be misleading and biased the results. idk.
gwern, Scott Alexander, and Aella all strike out as people whose IQ scores would dramatically underestimate how smart or prolific they are.
I would expect them to underperform tests that MIT students or Physics PhD students would perform *well* on, even if they are *on net* more impressive than MIT students or Physics PhD students
Also there's no way in hell Eliezer has a lower IQ than Scott Alexander (even though Scott Alexander is plenty smart)
Despite being one of the most important intellectuals of the 20th century, Edward O. Wilson famously scored in the 120s on an IQ test. I'm not surprised. The cognitive archetypes of Scott Alexander and gwern and AellaGirl are probably similar to EO Wilson (gwern once mentioned he was "too lazy" to learn linear algebra)
Firstly, ancient thinkers aren't likely to be much smarter than everyone else in the modern day; they were big fish, yes, but in a much smaller pond. Just looking at, for example, Razib, I find it very hard to believe Raz is less intelligent than Kant, Socrates, Plato, Aristotle, Nietzsche, and Diogenes of Sinope. Is Razib *really* almost two standard deviations below Newton. The same reasoning works for you, Emil.
Secondly, equating creativity with trait psychoticism is terrible; I know Eysenck loved Psychoticism, but his tools for measuring it were abysmal, and even modern correlations between P and creativity are not high. Openness to Experience plays a bigger role in creative genius than susceptibility to mental illness.
Thirdly Scott Alexander is (well, was) totally rad, but not because his IQ is amazing or because he's borderline psychotic. What Scott does have is phenomenal verbal ability, and a good social network. It never hurts to have a little help from your friends!
I think you could safely lift the cutoff from 70 to 100 and still keep all potentially accurate guesses. Probably even 110. Would be cool to see a table for the median guesses per lower cutoff increment.
Interestingly enough, Andrew Tate’s father was a chess grandmaster: https://en.m.wikipedia.org/wiki/Emory_Tate
Hes definitely higher IQ than people think. Pretty much all of these people are high IQ.
Even Kanyes mom was a college professor who chaired the English dept at Chicago State University. To be super successful in the music industry in the way Kanye has been requires a lot of higher order thinking. It wouldn’t surprise me in the slightest if he was in the top 2 percent or so of black IQ, like 120+.
Interesting survey though, I wonder which ones were most accurate. Coincidentally I know for a fact my dads iq is 127 (he’s a VP of global manufacturing & logistics at a well known company, for reference).
Curtis Yarvin's IQ is public at 167.
Eliezer Yudkowsky's can be pretty well estimated at 165 based on his public "Midwest Talent Search" results:
https://web.archive.org/web/20010309014808/http://sysopmind.com/eliezer.html#timeline_the
Lol these are horrible. Way, way, *way* too low across the board. Based on numerous surveys, the average IQ of people in the broader rationalist sphere is in the vincinity of 130. All of these people are noticeably smarter than the average commentator. They have acquired their fame through mastery of technical subjects and winning fair-and-square in the Marketplace of Ideas. The average IQ here should be 145 or maybe even as high as 150.
I might write up a more detailed comment for how to better estimate IQ (as well as my own opinion on the IQ of some of the people here). But for now , here's an intuition pump for why IQ 145 should the baseline.
Google says that average high school in the United States has around 500 people. So the smartest person at a given high school will be in the 99.8th percentile in intellegence. If you convert that to a z-score, you get a score of 2.88 (that is, 2.88 standard deviations above the mean). You can then convert the z-score to IQ by multiplying by 15 and adding the product to 100 which gets you an IQ of 143.
So the average smartest person in a given high school has an IQ of 143. Eyeballing the graph, it seems that the average estimate for Charles Murray (the greatest social scientist of the last 50 years) is about 135. Do you guys really think that if you go to an average high school, you will find someone (let alone multiple people) smarter than Charles Murray? Are you guys insane?
Both Fuentes and Spencer are vastly underestimated probably due to them being controversial in the community.
My take on this is that people vastly underestimate how smart many popular writers are, and how much that contributes to their output.
130 IQ doesn't get you the kind of output on technical subjects that many of these guys put out. Say you what you want about Noah Smith, but my lower bound for him would be 140 (that should tell you something about the hardware Emil is probably working with).
In terms of raw intellect of people who are currently alive, it's hard not to see Robin Hanson and Gwern as gross underestimates. Gwern's writing reminds of me a bit of Chris Langan's (albeit much less insane), so it wouldn't surprise me if he was close to 200.
Gwern is plainly smarter than ASC, but ASC has him beat in trait psychoticism (maybe) and the type of unenviable phenotype that makes someone desperate for approval & hence apt to produce a lot.
Isn’t this all just garbage in-garbage out because most people are clueless about real IQ scores and have never taken a properly administered and normed test? Pretty sure most people think it’s linear like school grades and just assign “a bit more” IQ to some they think is “a bit more” smart.
I doubt anything above 130 iq would help much with substack writing, unless your substack was about theoretical physics or whatever. So it doesn’t surprise me that there are a lot of estimates around there or slightly above. 130 puts you in the top 2% of the population.
As I wasn't smart enough to know who all these people even were let alone how smart they are, I slunk back into my midwittery to read about the Dunning-Kruger effect on Wiki...again.
n = n -1
These are way too low. Newton comes in at 160, which is +4SDs. Over 10,000 people in the United States are over 160. There are not 10,000 people in America with more intellectual horsepower than Isaac Newton. There’s a strong case to be made that there are 0 in history. A more accurate estimation would have exceeded the limit of 200.
In some cases, we have public statements of their IQ, and the estimates are way lower. Aella has been tested at 140, Hanania has suggested he’s around 145, and in this thread you can find people providing better estimates of Yud and Yarvin.
I’m surprised at how bad people were at this given the community’s obsession with IQ. Even numbers like 160 aren’t that rare, and this is a selection of some of the era’s (and history’s) foremost intellects.
Greg Cochran says he was in the top half percent of scholastic achievement tests like SAT, GRE etc. back in the day. He also said he got over 700 SAT Math without prep. There are some reasonable estimates about IQ-old SAT/GRE conversions and this puts him above 145 easily.
Scott Alexander probably has an uneven cognitive profile due to lower correlation of sub-factors among high ability people. Guessing ASC has near 3std for verbal ability and lower (1.5-2sd) quant ability. If I'd have to take a shot I'd say his FSIQ is 134-135.
Semi-relatedly, one of the more interesting pieces of data I saw was that the average IQ of an Oxford math PhD is 128 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5008436/), so I guess you don't have to be exceptionally gifted to be very good at math. So "[X] is very good at math, s/he must be 140+" (which is an heuristic lots of people use) might be misleading and biased the results. idk.
v. surprised yarvin is not higher! dude is defnly a genius imo.
It seems meng hu and cremieux have great math ability. I think they are 130+
gwern, Scott Alexander, and Aella all strike out as people whose IQ scores would dramatically underestimate how smart or prolific they are.
I would expect them to underperform tests that MIT students or Physics PhD students would perform *well* on, even if they are *on net* more impressive than MIT students or Physics PhD students
Also there's no way in hell Eliezer has a lower IQ than Scott Alexander (even though Scott Alexander is plenty smart)
Despite being one of the most important intellectuals of the 20th century, Edward O. Wilson famously scored in the 120s on an IQ test. I'm not surprised. The cognitive archetypes of Scott Alexander and gwern and AellaGirl are probably similar to EO Wilson (gwern once mentioned he was "too lazy" to learn linear algebra)
There's no Dean Simonton here
Three comments here:
Firstly, ancient thinkers aren't likely to be much smarter than everyone else in the modern day; they were big fish, yes, but in a much smaller pond. Just looking at, for example, Razib, I find it very hard to believe Raz is less intelligent than Kant, Socrates, Plato, Aristotle, Nietzsche, and Diogenes of Sinope. Is Razib *really* almost two standard deviations below Newton. The same reasoning works for you, Emil.
Secondly, equating creativity with trait psychoticism is terrible; I know Eysenck loved Psychoticism, but his tools for measuring it were abysmal, and even modern correlations between P and creativity are not high. Openness to Experience plays a bigger role in creative genius than susceptibility to mental illness.
Thirdly Scott Alexander is (well, was) totally rad, but not because his IQ is amazing or because he's borderline psychotic. What Scott does have is phenomenal verbal ability, and a good social network. It never hurts to have a little help from your friends!
I think you could safely lift the cutoff from 70 to 100 and still keep all potentially accurate guesses. Probably even 110. Would be cool to see a table for the median guesses per lower cutoff increment.