8 ms·

This sounds ridiculous. In the writer's own words, "As a mathematician who studied at Berkeley, Harvard and Princeton, I’ve known geniuses". Maybe that's what s

This sounds ridiculous. In the writer's own words, "As a mathematician who studied at Berkeley, Harvard and Princeton, I’ve known geniuses". Maybe that's what she means by "geniuses", but usually the word tends to mean "who studied at Berkeley, Harvard and Princeton". In terms of IQ, I think mathematicians are averaging something like +2 SD (>= 130). They are certainly "geniuses" (or "gifted"). But they themselves don't think they are geniuses, because by that word they mean +4 SD (>= 160) people.

tacomonstrous 9y ago
Genius has nothing to do with IQ. Genius is a level of creative thought that appears magically simple once you behold it, but yet was beyond people up to that point. The things that IQ tests for are almost orthogonal to this. Sure, there are probably geniuses with high IQs, but I'd be very surprised if high IQ has any predictive ability to detect genius.
- sanxiyn 9y ago
  > I'd be very surprised if high IQ has any predictive ability to detect genius. This is the most ridiculous thing I have read for a while.
  - f055 9y ago
    Why? High IQ means you are very good at solving the IQ test problems, not that you are intelligent or genius. I don't think these problems represent anything else, or if there are examples of high IQ people that made actual, genius contribution to humanity. Do you?
    - sanxiyn 9y ago
      Are you familiar with modern intelligence research? IQ test is reliable (get the same score when you test multiple times) and valid (correlate with things we actually care about, not just being good at solving IQ test problems). Intelligence: All That Matters by Stuart Ritchie is a good summary of modern intelligence research. https://www.amazon.com/Intelligence-That-Matters-Stuart-Ritchie/dp/1444791877 https://www.amazon.com/Intelligence-That-Matters-Stuart-Ritc... To quote: "The scientific evidence is clear: IQ tests are extraordinarily useful. IQ scores are related to a huge variety of important life outcomes like educational success, income, and even life expectancy, and biological studies have shown they are genetically influenced and linked to measures of the brain. Studies of intelligence and IQ are regularly published in the world's top scientific journals."
      - lainon 9y ago
        http://www.pnas.org/content/109/2/425.full.pdf http://www.pnas.org/content/109/2/425.full.pdf (Schooling in adolescence raises IQ scores)
        sanxiyn 9y ago
        Summary: "The quasiexperimental results suggest that the reform increased the average IQ score for Norwegian men by a statistically significant 0.6 IQ points." This paper shows two things, although only one is surprising enough for publication. The surprising part is that "schooling in adolescence raises IQ scores", which is the title of the paper. Another is that IQ is so reliable that increase of 0.6 IQ points is statistically significant.
      - paulpauper 9y ago
        agree...IQ correlates with a lot. It's not just a measure of "how well someone does on IQ tests".
      - DamonHD 9y ago
        My mum, after getting her PhD (and having taught herself through A-levels because her Yorkshire school stopped at O-levels, and got into Oxford) decided to take her first-ever IQ test. She rated mentally subnormal. Astonished, she thought about what it was trying to test and took another test: now genius. None of us think that either was right...
        hinkley 9y ago
        Up until I was eight they thought I might be deficient. So they poked and prodded and eventually some people came to school and took me to the library and gave me an IQ test. Then they came back and gave me another one. And I believe another one after that, by which point I was a little disgruntled. Didn't I already take one of these? Did you lose my answers? Turns out that the test can be tuned to a range. If you take the wrong one it loses some accuracy at the low and high end of the range, so they had to retest me. They had assumed I might be dim and gave me the wrong one. The real problem was that I was so insufferably bored all the time that I wouldn't engage. But I liked puzzles and the test had a bunch I'd never seen before, so by the last one I was taking it in the spirit it was given. My guess is in two parts. Either your mother took the wrong one(s), or took a bogus one, or you're thinking of elements of a successful adult that the test can't measure, like common sense and social graces. In any group of peers at that level there are bound to be people who can certainly do the work, possibly better than you, but who are painfully, even cringingly, bad at certain other life skills. It gets a little uncomfortable to acknowledge these people as geniuses. And there are people who seem not to be all that, but on occasion surprise the hell out of you by coming through in a clutch.
    - Gauc2 9y ago
      My impression is that IQ test roughly measures g [0], which does correlate with intelligence.Note that this is not my research area by any stretch. [0] https://en.wikipedia.org/wiki/G_factor_(psychometrics) https://en.wikipedia.org/wiki/G_factor_(psychometrics)
  - walterstucco 9y ago
    I'm quite sure Van Gogh's IQ wasn't that high...
    - lgas 9y ago
      Why?
      - walterstucco 9y ago
        because... history? he was a Genius, but his IQ was probably average, according to what we know about him
        dang 9y ago
        Hard to say. He was nearly as good a writer as a painter and presumably wasn't even trying at that craft. He would be remembered for his letters even if he never painted a thing, and they were purely accidental because Theo wasn't always in town. But you still have a point.
        lgas 9y ago
        I don't know the history, that's what I was asking. I googled before I asked but everything I could find seemed relatively split with many sources implicitly or explicitly referring to him as a genius.
    - dando 9y ago
      And who said Van Gogh is a genius?
      - walterstucco 9y ago
        basically everybody
- Splatter 9y ago
  > Genius has nothing to do with IQ. Genius is a level of creative thought that appears magically simple once you behold it, but yet was beyond people up to that point. Not sure why the downvotes. While discounting the relationship of genius with IQ is specious in light of their point, this person offers a possible nugget of truth on an alternative understanding of genius.
- freyr 9y ago
  In one sense, I think you're correct: the term "genius" should be reserved for people who produce extraordinary output in some field -- whether that's math or science, literature, music, visual art, etc. Even if we restrict ourselves to scientific geniuses, the idea that everyone who gets above a certain score on an IQ test should be labelled a genius seems, frankly, small-minded. Not everyone with an IQ >= X should be labelled a genius because not everyone with a very high IQ has the other qualities required to produce genius-level work. But in many fields, such as mathematics, high IQ seems to be a necessary but not sufficient condition for genius. This makes sense -- doing well in mathematics courses requires the same logical and analytical skills required to score high on an IQ test. In that sense, it obviously has some predictive value.
  - tacomonstrous 9y ago
    I agree with everything you say, but if >140 IQ accounts for the 99.5 percentile and true genius (in my estimation) is closer to 1 in 100000, then, even if every genius is within that IQ percentile, the predictive value is still essentially worthless.
    - freyr 9y ago
      Using those numbers, a random person has a 0.001% chance of being a genius. After they score >140 on an IQ test, this improves to a 0.2% chance. That's a big increase, but it isn't high enough to predict that an individual is a genius with any degree of certainty. But, we can use the IQ test to predict who isn't a genius, or to make large scale predictions. For example, if a state has 100,000 students entering middle school per year, there's approximately one genius in that group. If you can afford an intensive math program for 7,500 students, the IQ test would let you predict that any potential geniuses get enrolled with high probability. That raises issues of fairness, discrimination, etc., but that's another topic.
paulpauper 9y ago
mathematicians probably have a much higher standard for what constitutes 'genius' than do regular people. For a layperson, a genius may be someone who understands elliptic integrals, but to a mathematicians such a concept is rather ordinary.
- joatmon-snoo 9y ago
  My senior year of undergrad, there were 4 of us living together who were math majors. There were so many things that we laughed or argued about that when non-math major friends were over, they sort of just stared at us because nothing we were saying made sense to them, even though it made perfect sense to us. Part of it, I think, is because of the sheer knowledge threshold: much of higher level maths has its own specialized language, so when you start discussing isomorphisms between Z_p and geometric structures, and you don't know what those words mean, your eyes can't help but glaze over. So I don't know if it's so much that there's a higher standard of 'genius' so much as it is that it's a different standard, because much of maths is so far removed from the knowledge base of the layperson.
  - sndean 9y ago
    > Part of it, I think, is because of the sheer knowledge threshold: much of higher level maths has its own specialized language Not sure if that's specific to math though... Certainly at a PhD-level there's going to be a specialized language. But I'm trying to imagine undergrads of different majors having a conversation and seeing how far outside language of a layperson they'd be. Maybe most STEM majors would be incomprehensible? Math and physics may be the furthest outside. If you heard a bunch of chemistry students talking, your eyes may glaze over. Biology may be not too far outside. Once you get into psychology/sociology you might understand most of it. I have a STEM PhD, so it's a difficult exercise. Essentially, "what would my mom understand?"
    - chestervonwinch 9y ago
      Jargon certainly isn't unique to any field. In my view, jargon repackages conceptual or procedural knowledge into shorter phrases for the purpose of more efficient communication. The less familiar another person is with a particular set of jargon, the more it must be "un-packaged" into its fundamental ideas. Of course, jargon can be nested, so this process may potentially be tedious. On the other hand, this jargon nesting problem can be avoided by approximating ideas with ones that the other party is assumed to be familiar with. An expert knows the jargon, the ideas behind the jargon, and many of the first- and second-order approximations of the ideas.
csa 9y ago
> Maybe that's what she means by "geniuses", but usually the word tends to mean "who studied at Berkeley, Harvard and Princeton". You made an excellent point here. I hope that more people who don't go to elite schools realize that many (most?) people who do are distinctly not geniuses either on the surface (e.g., if you talk to them) or in actuality (i.e., 140+ IQ).
gjm11 9y ago
"Genius" is like "rich": as actually used, it's implicitly a comparative term. When someone talks about geniuses they usually mean "people much smarter than me" (either in general intelligence, in so far as that's meaningful, or in a particular area like mathematics or music or languages). I think this sort of usage is so deeply ingrained that there's not much point in trying to counter it by adopting a fixed formal definition (e.g., someone elsewhere in the thread said something like "genius means IQ >= 140, get over it"). Instead, either pick a definition and say what definition you're using or use it informally and live with the fact that sometimes you'll be misunderstood. In this particular case, it seems reasonably clear that what she means by "geniuses" is "people with the sort of ability the big names in mathematics generally have": her point is that mathematical progress depends not only on the Gausses and Thurstons and Ramanujans but also, just as much, on the ordinary mathematicians who, yes, are generally also extremely bright but are very conscious of the gulf between them and the famously-brilliant ones. (Is it absurd to say "just as much"? No, because there are a lot more ordinary mathematicians than once-in-a-generation ones.)
fullquoteplease 9y ago
> This sounds ridiculous. It sounds less so if you read the next sentence, and the one after that. As a mathematician who studied at Berkeley, Harvard and Princeton, I’ve known geniuses. I got to hang out with Andrew Wiles, who is credited with solving Fermat’s Last Theorem, and I met Grigori Perelman, who solved the Poincare Conjecture. They’re great guys, but they didn’t do it on their own. For each certified genius, there are at least a hundred other great people who helped achieve such outstanding results.
beefield 9y ago
I think here are two things in play: First the other end of the Dunning Kruger effect where competent people underestimate their competence rank. Second, only good mathematicians[1] can recognize the massive gap between good mathematician and a real top class mathematical genius. For the mere mortals they both are so far that they look alike. [1] Mathematician can be replaced with any other talent/competence. Actually I realized this phenomena first in sports. The world's ultimate top and a very good amateur look perfectly as good to a complete beginner. But when you get to the very good amateur level, you start to realize the massive amount of work (and maybe talent) that makes the difference between you and the ultimate top.
avip 9y ago
Someone has to build the confusion matrix for the geniuses "> 130 IQ" classifier. I mean, if you have a definition for "a genius" that likely ends up excluding Bach, Escher, Darwin - then something's fishy about your definition. In terms of communicating with the others, that's akin to having a definition for "great basketball player" to which Jordan doesn't fit.
- Retric 9y ago
  Darwin was something of a plodder who get's a lot of great press. He did not really come up with the idea of evolution, just came up with more support, a convincing argument, and most importantly communicated this to a wide range of people. https://en.wikipedia.org/wiki/History_of_evolutionary_thought https://en.wikipedia.org/wiki/History_of_evolutionary_though... It's kind of the great man argument taken to the extreme.
photoJ 9y ago
Remember in a stadium of 80,000 people there are 320 people who have an IQ of >=140.
- orbifold 9y ago
  Only true if the distribution of people going to a stadium is not biased to below average intelligent people, or biased in some other way, which I suspect might be the case.
  - SamReidHughes 9y ago
    Above average, considering the cost of a ticket.
YeGoblynQueenne 9y ago
She names specific people who she met at Berkeley as geniuses: As a mathematician who studied at Berkeley, Harvard and Princeton, I’ve known geniuses. I got to hang out with Andrew Wiles, who is credited with solving Fermat’s Last Theorem, and I met Grigori Perelman, who solved the Poincare Conjecture. I don't think she means that everyone she met at Berkeley was a genius.

tacomonstrous 9y ago

Genius has nothing to do with IQ. Genius is a level of creative thought that appears magically simple once you behold it, but yet was beyond people up to that point. The things that IQ tests for are almost orthogonal to this. Sure, there are probably geniuses with high IQs, but I'd be very surprised if high IQ has any predictive ability to detect genius.

sanxiyn 9y ago
> I'd be very surprised if high IQ has any predictive ability to detect genius. This is the most ridiculous thing I have read for a while.
- f055 9y ago
  Why? High IQ means you are very good at solving the IQ test problems, not that you are intelligent or genius. I don't think these problems represent anything else, or if there are examples of high IQ people that made actual, genius contribution to humanity. Do you?
  - sanxiyn 9y ago
    Are you familiar with modern intelligence research? IQ test is reliable (get the same score when you test multiple times) and valid (correlate with things we actually care about, not just being good at solving IQ test problems). Intelligence: All That Matters by Stuart Ritchie is a good summary of modern intelligence research. https://www.amazon.com/Intelligence-That-Matters-Stuart-Ritchie/dp/1444791877 https://www.amazon.com/Intelligence-That-Matters-Stuart-Ritc... To quote: "The scientific evidence is clear: IQ tests are extraordinarily useful. IQ scores are related to a huge variety of important life outcomes like educational success, income, and even life expectancy, and biological studies have shown they are genetically influenced and linked to measures of the brain. Studies of intelligence and IQ are regularly published in the world's top scientific journals."
    - lainon 9y ago
      http://www.pnas.org/content/109/2/425.full.pdf http://www.pnas.org/content/109/2/425.full.pdf (Schooling in adolescence raises IQ scores)
      - sanxiyn 9y ago
        Summary: "The quasiexperimental results suggest that the reform increased the average IQ score for Norwegian men by a statistically significant 0.6 IQ points." This paper shows two things, although only one is surprising enough for publication. The surprising part is that "schooling in adolescence raises IQ scores", which is the title of the paper. Another is that IQ is so reliable that increase of 0.6 IQ points is statistically significant.
    - paulpauper 9y ago
      agree...IQ correlates with a lot. It's not just a measure of "how well someone does on IQ tests".
    - DamonHD 9y ago
      My mum, after getting her PhD (and having taught herself through A-levels because her Yorkshire school stopped at O-levels, and got into Oxford) decided to take her first-ever IQ test. She rated mentally subnormal. Astonished, she thought about what it was trying to test and took another test: now genius. None of us think that either was right...
      - hinkley 9y ago
        Up until I was eight they thought I might be deficient. So they poked and prodded and eventually some people came to school and took me to the library and gave me an IQ test. Then they came back and gave me another one. And I believe another one after that, by which point I was a little disgruntled. Didn't I already take one of these? Did you lose my answers? Turns out that the test can be tuned to a range. If you take the wrong one it loses some accuracy at the low and high end of the range, so they had to retest me. They had assumed I might be dim and gave me the wrong one. The real problem was that I was so insufferably bored all the time that I wouldn't engage. But I liked puzzles and the test had a bunch I'd never seen before, so by the last one I was taking it in the spirit it was given. My guess is in two parts. Either your mother took the wrong one(s), or took a bogus one, or you're thinking of elements of a successful adult that the test can't measure, like common sense and social graces. In any group of peers at that level there are bound to be people who can certainly do the work, possibly better than you, but who are painfully, even cringingly, bad at certain other life skills. It gets a little uncomfortable to acknowledge these people as geniuses. And there are people who seem not to be all that, but on occasion surprise the hell out of you by coming through in a clutch.
  - Gauc2 9y ago
    My impression is that IQ test roughly measures g [0], which does correlate with intelligence.Note that this is not my research area by any stretch. [0] https://en.wikipedia.org/wiki/G_factor_(psychometrics) https://en.wikipedia.org/wiki/G_factor_(psychometrics)
- walterstucco 9y ago
  I'm quite sure Van Gogh's IQ wasn't that high...
  - lgas 9y ago
    Why?
    - walterstucco 9y ago
      because... history? he was a Genius, but his IQ was probably average, according to what we know about him
      - dang 9y ago
        Hard to say. He was nearly as good a writer as a painter and presumably wasn't even trying at that craft. He would be remembered for his letters even if he never painted a thing, and they were purely accidental because Theo wasn't always in town. But you still have a point.
      - lgas 9y ago
        I don't know the history, that's what I was asking. I googled before I asked but everything I could find seemed relatively split with many sources implicitly or explicitly referring to him as a genius.
  - dando 9y ago
    And who said Van Gogh is a genius?
    - walterstucco 9y ago
      basically everybody
Splatter 9y ago
> Genius has nothing to do with IQ. Genius is a level of creative thought that appears magically simple once you behold it, but yet was beyond people up to that point. Not sure why the downvotes. While discounting the relationship of genius with IQ is specious in light of their point, this person offers a possible nugget of truth on an alternative understanding of genius.
freyr 9y ago
In one sense, I think you're correct: the term "genius" should be reserved for people who produce extraordinary output in some field -- whether that's math or science, literature, music, visual art, etc. Even if we restrict ourselves to scientific geniuses, the idea that everyone who gets above a certain score on an IQ test should be labelled a genius seems, frankly, small-minded. Not everyone with an IQ >= X should be labelled a genius because not everyone with a very high IQ has the other qualities required to produce genius-level work. But in many fields, such as mathematics, high IQ seems to be a necessary but not sufficient condition for genius. This makes sense -- doing well in mathematics courses requires the same logical and analytical skills required to score high on an IQ test. In that sense, it obviously has some predictive value.
- tacomonstrous 9y ago
  I agree with everything you say, but if >140 IQ accounts for the 99.5 percentile and true genius (in my estimation) is closer to 1 in 100000, then, even if every genius is within that IQ percentile, the predictive value is still essentially worthless.
  - freyr 9y ago
    Using those numbers, a random person has a 0.001% chance of being a genius. After they score >140 on an IQ test, this improves to a 0.2% chance. That's a big increase, but it isn't high enough to predict that an individual is a genius with any degree of certainty. But, we can use the IQ test to predict who isn't a genius, or to make large scale predictions. For example, if a state has 100,000 students entering middle school per year, there's approximately one genius in that group. If you can afford an intensive math program for 7,500 students, the IQ test would let you predict that any potential geniuses get enrolled with high probability. That raises issues of fairness, discrimination, etc., but that's another topic.

paulpauper 9y ago

mathematicians probably have a much higher standard for what constitutes 'genius' than do regular people. For a layperson, a genius may be someone who understands elliptic integrals, but to a mathematicians such a concept is rather ordinary.

joatmon-snoo 9y ago
My senior year of undergrad, there were 4 of us living together who were math majors. There were so many things that we laughed or argued about that when non-math major friends were over, they sort of just stared at us because nothing we were saying made sense to them, even though it made perfect sense to us. Part of it, I think, is because of the sheer knowledge threshold: much of higher level maths has its own specialized language, so when you start discussing isomorphisms between Z_p and geometric structures, and you don't know what those words mean, your eyes can't help but glaze over. So I don't know if it's so much that there's a higher standard of 'genius' so much as it is that it's a different standard, because much of maths is so far removed from the knowledge base of the layperson.
- sndean 9y ago
  > Part of it, I think, is because of the sheer knowledge threshold: much of higher level maths has its own specialized language Not sure if that's specific to math though... Certainly at a PhD-level there's going to be a specialized language. But I'm trying to imagine undergrads of different majors having a conversation and seeing how far outside language of a layperson they'd be. Maybe most STEM majors would be incomprehensible? Math and physics may be the furthest outside. If you heard a bunch of chemistry students talking, your eyes may glaze over. Biology may be not too far outside. Once you get into psychology/sociology you might understand most of it. I have a STEM PhD, so it's a difficult exercise. Essentially, "what would my mom understand?"
  - chestervonwinch 9y ago
    Jargon certainly isn't unique to any field. In my view, jargon repackages conceptual or procedural knowledge into shorter phrases for the purpose of more efficient communication. The less familiar another person is with a particular set of jargon, the more it must be "un-packaged" into its fundamental ideas. Of course, jargon can be nested, so this process may potentially be tedious. On the other hand, this jargon nesting problem can be avoided by approximating ideas with ones that the other party is assumed to be familiar with. An expert knows the jargon, the ideas behind the jargon, and many of the first- and second-order approximations of the ideas.

csa 9y ago

> Maybe that's what she means by "geniuses", but usually the word tends to mean "who studied at Berkeley, Harvard and Princeton". You made an excellent point here. I hope that more people who don't go to elite schools realize that many (most?) people who do are distinctly not geniuses either on the surface (e.g., if you talk to them) or in actuality (i.e., 140+ IQ).

gjm11 9y ago

"Genius" is like "rich": as actually used, it's implicitly a comparative term. When someone talks about geniuses they usually mean "people much smarter than me" (either in general intelligence, in so far as that's meaningful, or in a particular area like mathematics or music or languages). I think this sort of usage is so deeply ingrained that there's not much point in trying to counter it by adopting a fixed formal definition (e.g., someone elsewhere in the thread said something like "genius means IQ >= 140, get over it"). Instead, either pick a definition and say what definition you're using or use it informally and live with the fact that sometimes you'll be misunderstood. In this particular case, it seems reasonably clear that what she means by "geniuses" is "people with the sort of ability the big names in mathematics generally have": her point is that mathematical progress depends not only on the Gausses and Thurstons and Ramanujans but also, just as much, on the ordinary mathematicians who, yes, are generally also extremely bright but are very conscious of the gulf between them and the famously-brilliant ones. (Is it absurd to say "just as much"? No, because there are a lot more ordinary mathematicians than once-in-a-generation ones.)

fullquoteplease 9y ago

> This sounds ridiculous. It sounds less so if you read the next sentence, and the one after that. As a mathematician who studied at Berkeley, Harvard and Princeton, I’ve known geniuses. I got to hang out with Andrew Wiles, who is credited with solving Fermat’s Last Theorem, and I met Grigori Perelman, who solved the Poincare Conjecture. They’re great guys, but they didn’t do it on their own. For each certified genius, there are at least a hundred other great people who helped achieve such outstanding results.

beefield 9y ago

I think here are two things in play: First the other end of the Dunning Kruger effect where competent people underestimate their competence rank. Second, only good mathematicians[1] can recognize the massive gap between good mathematician and a real top class mathematical genius. For the mere mortals they both are so far that they look alike. [1] Mathematician can be replaced with any other talent/competence. Actually I realized this phenomena first in sports. The world's ultimate top and a very good amateur look perfectly as good to a complete beginner. But when you get to the very good amateur level, you start to realize the massive amount of work (and maybe talent) that makes the difference between you and the ultimate top.

avip 9y ago

Someone has to build the confusion matrix for the geniuses "> 130 IQ" classifier. I mean, if you have a definition for "a genius" that likely ends up excluding Bach, Escher, Darwin - then something's fishy about your definition. In terms of communicating with the others, that's akin to having a definition for "great basketball player" to which Jordan doesn't fit.

Retric 9y ago
Darwin was something of a plodder who get's a lot of great press. He did not really come up with the idea of evolution, just came up with more support, a convincing argument, and most importantly communicated this to a wide range of people. https://en.wikipedia.org/wiki/History_of_evolutionary_thought https://en.wikipedia.org/wiki/History_of_evolutionary_though... It's kind of the great man argument taken to the extreme.

photoJ 9y ago

Remember in a stadium of 80,000 people there are 320 people who have an IQ of >=140.

orbifold 9y ago
Only true if the distribution of people going to a stadium is not biased to below average intelligent people, or biased in some other way, which I suspect might be the case.
- SamReidHughes 9y ago
  Above average, considering the cost of a ticket.

YeGoblynQueenne 9y ago

She names specific people who she met at Berkeley as geniuses: As a mathematician who studied at Berkeley, Harvard and Princeton, I’ve known geniuses. I got to hang out with Andrew Wiles, who is credited with solving Fermat’s Last Theorem, and I met Grigori Perelman, who solved the Poincare Conjecture. I don't think she means that everyone she met at Berkeley was a genius.