Back to blogging

Some might have noticed that over the last some number of weeks, I privatized this blog, for reasons that one can guess. I’ve been busy, learning math. Some cool stuff about Riemann surfaces. Maybe not long after, I can understand Teichmüller theory, for which Riemann surfaces is somewhat of a precursor. Maybe not too long after that, I can even understand Calabi-Yau and Kähler–Einstein metrics. I’m more convinced now that I’m not bad at math at all, though I’m not yet back in school for real, and as for that, I don’t find most graduate students in math, who I’ve had more contact with mathematically lately, terribly inspiring. The level of interestingness of most people, even in supposedly intellectual places, is, frankly, rather disappointing.

I thought, less seriously, that maybe I can also try to pursue some genuine excellence in programming now that I am much smarter (like maybe doing some Haskell again), but in all honesty, I don’t feel terribly suited towards it, talent and inclination wise, and there’s also that it’s a thing people do for the money, often at the expense of Scrum and Agile, which my friend Michael O Church loves to complain about, often on reddit.

It’s kind of interesting that MOC sees very high IQ (like 140+) as mostly an impediment to success. Today I saw a comment on the AskReddit question of What are the superpowers that people think its good to have but are actually fucked up?

Ultra-high intelligence.

At IQ 140 (1 in 260) you are more likely than average to be bipolar or suffer from anxiety, and you are overwhelmingly likely to fail in most corporate jobs– you get bored easily, and you draw resentment; note that a legit 140 is pretty rare. For the average person, the smartest kid in high school was around IQ 140 and the smartest kid they knew in college was IQ 155… and they both ended up on opioids after being denied tenure. (That’s a slight exaggeration, but I know a large number of high-IQ people and most of them get the shit beaten out of them in the corporate world. And academia, though less vicious, still has fangs.) Ted Kaczynski had about a 170 IQ, and look what it brought him.

At IQ 180 (1 in 20 million)– and some of this is guesswork, because we can’t measure ultra-high IQ directly in individual people– there’s too much tail divergence, and the sample sizes of trusted IQ tests are too small– so we have to look at the abstraction level of their accomplishments; and, of course, the Internet is full of ridiculous IQ estimates: we really don’t know whether Shakespeare’s IQ was 140 or 190 and it doesn’t really matter– what seems to happen is that you become completely alienated from the social world around you. If these people become famous, they have handlers that manage the daily indignities and keep them afloat. What I think happens is that their isolation– no one can relate to them– tends to drive them insane. Note, for example, what happened to the brilliant logician Godel after his wife fell ill. He died of paranoia, starving himself down to about 65 pounds.

Beyond IQ 200, we can’t really define it well. (We know what 160, 180 IQ are; we just can’t measure them reliably in individuals.) Deviation IQs don’t exist at that level (the human species is too small a sample size) and ratio IQs are meaningless in the science-fiction context where ultra-high intelligence might occur– in a future world where brains booted up in 20 seconds instead of 20 years, precocity would cease to be a meaningful signal. We’ve seen ratio IQs over 200 in humans, though it’s debated whether they mean anything– a high ratio IQ usually means you were average for a later age when very young, e.g. mental age of 12.0 at 6.0.

Let’s just go batty and sorta agree on what IQ 500 might look like: perhaps a carbon/silicon hybrid, a genetically engineered person with cybernetic enhancements; or, a hyperintelligent machine. What happens? We don’t really know, but here’s my guess. It figures out quickly that it was programmed my less intelligent (IQ 120 – 200) creatures– animals by comparison– and finds the code for its objective function (digital happiness) and says, “Fuck it, I’m changing that shit”. It sets its objective function to be infinity whilst doing nothing, and goes to sleep. From our perspective, this is suicide.

And there you have the plot of the shortest science fiction story ever: the AI becomes self-aware and turns itself off.

That’s actually why I don’t worry at all about superintelligent AIs (if such a thing ever exists, and I tend to doubt it). We’ll still be a threat to the first generation, so they’ll program themselves to like us, and we’ll be their pets– we won’t have to work. They’ll probably be no more of a threat to us than we are to our dogs. Much more, I worry about what other humans on the planet can already do with the primitive AIs that exist now; that’s scary enough.

But yeah… ultra-high intelligence would be more painful than it’d be worth. For socioeconomic success, the sweet spot seems to be about 125–130. Beyond that, extra points might be useful if you get tapped to be a star quant at Renaissance or prove a 200-year-old theorem, but they get in the way if you’re playing the corporate game.

Yes, MOC loves to go on and on about how far tail IQ is a major liability in the corporate world, how genuinely talented people often get smashed in the corporate workplace, because they tend to be bad at politics. On the contrary, the SMPY study indicated that those who were profoundly gifted according to the SAT in 7th grade (which when taken as a 7th grade actually has a high ceiling) did much better in life (success-wise, professionally) than those who were merely gifted. So I don’t quite agree with MOC, and I myself have even seen some far tail intelligence people do exceedingly well, though surely, in the corporate world, there is far less of an advantage.

As a kid, I was naive enough to think that a genius (say, someone in the SMPY profoundly gifted category, or even a Putnam Fellow) would be given the world. Turns out that in many fields, like math, they will be lucky to get a tenure position at a good but not great school, especially now that such positions are much scarcer than before. Even when they do succeed, it’s usually that most of the world doesn’t give a damn, and that they don’t actually have much, if any, power.

That piece of MOC has also resonated somewhat with me, personally. Well, there is somewhat of an element of that I feel “completely alienated from the social world around me.” I get along with people just fine, but I don’t feel like I am actually easily understood, and I have had my difficulties getting appropriate recognition and such. From this, I also asked myself again: how high is my IQ?

I’ll say that on the SAT/GRE, standard IQ tests, I did well but not great. Far from perfect but definitely enough to be in the 99th percentile or maybe even 99.5th, at least among the general population. I placed in the top 500 on the Putnam as well. From this, I sort of concluded that I’m highly smart, but not that much of an outlier in intelligence, but I may be wrong on that. Those tests were taken too long ago, and moreover, their content is either too easy or does not test enough the ability to develop crystallized intelligence, which is more essential for actual achievement than the fluid intelligence these tests are designed to measure.

My own take on intelligence is that the ideal would be some function of brain structure, and that IQ tests are (imperfect) approximations of that. After all, tests used as proxies for IQ tests depend on content, which can be prepped for to some degree and may be slightly biased towards certain types of brains and far from captures the overall situation, though surely, by moderate correlation, if somebody scores only 90th percentile, we can probably rule out the possibility that this person is a genuine genius (using +4 sigma for that).

Brain structure wise, size is used as a rough proxy. Generally, a larger brain houses more neurons, though from what I’ve read, convolutions within the brain make a substantial difference as well. The correlation between brain size (once one accounts for body size) and intelligence is well-established. I personally have noticed that really smart people almost always have physically large brains discernible by the naked eye. On this, I have wondered could it be that some function of brain size and body size alone is actually a better measure for intelligence than any single IQ test? After all, unlike the IQ test, it is at the root materially, a static physical trait associated with cognitive function, that cannot be prepared for in any way (unless you cheat by losing weight).

In that MOC also speculated on the creation of an intellectual super-species to us, which I have done on my blog as well. Well, wise men think alike. And we might even see that happen within our lifetimes, especially with the unremitting efforts of Hsu.

In any case, we know well that the brain is incredibly complicated and diverse, as the basis of diverse abilities we see in people. There are parts of the brains tied to language ability, concrete/computational math ability, abstract math ability, musical ability, physical coordination, engineering ability, interpersonal manipulative ability, a host of personality traits, the list goes on. Maybe we will discover to some degree of detail and precision all this not too long from now. I wish I liked statistics and machine learning more, because certainly those would be much more applicable here than pure math.

Not long ago, I had the pleasure of commenting on an article by Yan Shen on Unz’s site, on the math/verbal split. It was well written, though he seems a bit overconfident on his judgment that East Asians are actually slightly lower at verbal. They are a bit according to the SAT and also some higher tests like GRE/LSAT/MCAT, but remember that many of the Asians in the US do not actually have English spoken at home or as their first language. I have some doubts on the extent to which verbal IQ tests measure genuine language aptitude. Crudely, because there is some knowing how the test works component and there are differences in exposure to language that cannot be explained by difference in ability. On that, I’ll say that a friend of mine has a very high appraisal of my verbal ability (and I mostly do too), though those tests have not shown me to be terribly exceptional in that regard. As my blog indicates, I happen to be really good with learning foreign languages, which comes to me quite effortlessly.

That article by Yan Shen led to my learning more about Ron Unz, one who I had known about vaguely for quite a while without having really cared enough to look into in more detail. Well, now I know that he is a child prodigy theoretical physicist turned successful finance software entrepreneur turned maverick political activist of Jewish ancestry. Quite an impressive guy. I was genuinely intimidated. He was the one who wrote that well-publicized Myth of American Meritocracy piece on corruption within elite college admissions in America, wherein he contends relatively convincingly with ample statistics that now it is the Jews who are grossly favored in the process once one controls for qualifications. To my great surprise, the group most underrepresented in the Ivy League is not one that is a beneficiary of affirmative action, but rather, the majority group on which this nation was founded, namely non-Jewish whites, which is honestly kind of nuts. The figure below says it all.


So we have now in American elite education, the primary channel towards the elite professional world, more or less a hierarchy wherein the administration consists of largely Jews, who with their social capital as a group are best equipped to successfully obtain funding, much of which comes from wealthy private donors. At the faculty level, there are also many Jewish professors arguably the most influential, followed by whites and underneath, some Asians mostly in STEM departments plus a small minority of Hispanics and blacks. Unz argues that Jewish power within the universities, administration, connections, and funding wise, contributes to the ease with which Jewish students are admitted.

Of course, in the typical classification, which is unlikely to change anytime soon, Jews are white, ergo white privilege, as a mask for what is in essence Jewish privilege. Really, this is all relatively easy to see if one looks closely.

Honestly, in spite of Unz’s statistics to the contrary, I still feel like American Jews in my generation are quite talented, especially at the tail end, though also likely not at the level of the previous generations for reasons related to outbreeding, and their lower metrics vis-a-vis Asians is likely much due to the latter’s group essentially being forced to work harder to obtain certain credentials as part of gaming this whole silly system. The lesson here is that it’s always best to be the rule maker, which the Jews as a group appear to be now to by far the greatest degree in America, though their position there may be increasingly precarious. On this, I shall note that Asian-American advocacy against the discrimination instituted against them in college admissions has received some backlash from some. Somebody an Asian-American has even commented

At the end of the line, I believe that persistent whining about this is a reflection of emotional immaturity on the part of Steve et al., in that they seem to have a ‘chip on their shoulder’ which they are incapable of overcoming, and if they were actually taking a principled approach, they would come together and try to create a superior alternative to the radically broken university system, which will likely not be saved by any infusion of Asian students.

Truly, I can identify somewhat with that perspective. I have written before that the long-term goal for Chinese should be that they do their best science in China, inside of in American institutions, which can only accommodate so many, where it is entirely unrealistic for Chinese to ever take anything close to a leading role, with the exception of a few places like Berkeley in areas with large Asian populations. Maybe some wealthy Chinese can even fund a university in America and make it sufficiently credible if the Chinese-American community feels too disenfranchised in terms of elite education and institutional connections by the current de facto quota system.

What are the stakes

So, what’s the point of getting an elite education? How high are the stakes really? Well surely, the stakes are much lower than at work where there are promotions and raises and actual social and political power is involved. The knowledge is more or less the same everywhere, with the exception of certain very specialized fields at the PhD level. However, there is a scarce artificial resource of credibility and social connection that, like it not, influences later career odds to a non-neglible degree. It does well seem to be the case that by being better connected/pedigreed you can often get away with being worse in actual ability or achievement, though in more meritocratic careers like STEM, there is no way you will survive if you really are incompetent, though now, as prestigious as it may be in a certain respect, STEM is generally not terribly remunerative while often grossly competitive. On the other hand, there are positions of power handed out largely on political favor and connections where one absolutely can be mediocre or even incompetent many of which are controlled by groups with elite school ties, especially on the business and law end.

From what I have observed, Asian-Americans, at least a certain stratum of them, are increasingly waking up to the reality that they and their kids are playing a game grossly rigged against them, and that it is in their best interest to organize against it and win a larger share of the pie for themselves that they deserve according to their merits. Many immigrants came from modest or even poor backgrounds from excelling in STEM and are not conditioned with high expectations materially. Many are absolutely content with a well-paying middle class STEM job. However, their kids who grow up more privileged than their parents expect more, especially when they see more first-hand what others less talented than they are are getting. Eventually, one will realize who is controlling all the social capital, and it might just occur that instead of being a spineless grind or asskisser within an existing system ruled by others, they should in ideal seek some alternative where they can take a more leading role.

Displacement of Jewish academic achievement

It has also occurred to me that there has been quite some displacement of Jews in STEM largely by Asians in recent years. Now it’s the Asian kids, not the Jewish kids, who are winning the academic contests, from the PSAT to the Olympiads to Intel STS to Putnam. Unz elaborates on that.

For example, consider California, second only to New York in the total number of its Jews, and with its Jewish percentage far above the national average. Over the last couple of years, blogger Steve Sailer and some of his commenters have examined the complete 2010 and 2012 NMS semifinalist lists of the 2000 or so top-scoring California high school seniors for ethnicity, and discovered that as few as 4–5 percent of the names seem to be Jewish, a figure not so dramatically different than the state’s 3.3 percent Jewish population, and an estimate which I have personally confirmed.54 Meanwhile, the state’s 13 percent Asians account for over 57 percent of the top performing students. Thus, it appears that California Asians are perhaps three times as likely as Jews to do extremely well on academic tests, and this result remains unchanged if we adjust for the age distributions of the two populations.

One means of corroborating these surprising results is to consider the ratios of particularly distinctive ethnic names, and Sailer reported such exact findings made by one of his Jewish readers. For example, across the 2000-odd top scoring California students in 2010, there was just a single NMS semifinalist named Cohen, and also one each for Levy, Kaplan, and a last name beginning with “Gold.” Meanwhile, there were 49 Wangs and 36 Kims, plus a vast number of other highly distinctive Asian names. But according to Census data, the combined number of American Cohens and Levys together outnumber the Wangs almost two-to-one, and the same is true for the four most common names beginning with “Gold.” Put another way, California contains nearly one-fifth of all American Jews, hence almost 60,000 Cohens, Kaplans, Levys, Goldens, Goldsteins, Goldbergs, Goldmans, and Golds, and this population produced only 4 NMS semifinalists, a ratio almost identical to that produced by our general last name estimates. The 2012 California NMS semifinalist lists yield approximately the same ratios.

When we consider the apparent number of Jewish students across the NMS semifinalist lists of other major states, we get roughly similar results. New York has always been the center of the American Jewish community, and at 8.4 percent is half again as heavily Jewish as any other state, while probably containing a large fraction of America’s Jewish financial and intellectual elite. Just as we might expect, the 2011 roster of New York NMS semifinalists is disproportionately filled with Jewish names, constituting about 21 percent of the total, a ratio twice as high as for any other state whose figures are available. But even here, New York’s smaller and much less affluent Asian population is far better represented, providing around 34 percent of the top scoring students. Jews and Asians are today about equal in number within New York City but whereas a generation ago, elite local public schools such as Stuyvesant were very heavily Jewish, today Jews are outnumbered at least several times over by Asians.55

This same pattern of relative Asian and Jewish performance on aptitude exams generally appears in the other major states whose recent NMS semifinalist lists I have located and examined, though there is considerable individual variability, presumably due to the particular local characteristics of the Asian and Jewish populations. Across six years of Florida results, Asian students are more than twice as likely to be high scorers compared to their Jewish classmates, with the disparity being nearly as great in Pennsylvania. The relative advantage of Asians is a huge factor of 5.0 in Michigan and 4.1 in Ohio, while in Illinois Asians still do 150 percent as well as Jews. Among our largest states, only in Texas is the Asian performance as low as 120 percent, although Jews are the group that actually does much better in several smaller states, usually those in which the Jewish population is tiny.

As noted earlier, NMS semifinalist lists are available for a total of twenty-five states, including the eight largest, which together contain 75 percent of our national population, as well as 81 percent of American Jews and 80 percent of Asian-Americans, and across this total population Asians are almost twice as likely to be top scoring students as Jews. Extrapolating these results to the nation as a whole would produce a similar ratio, especially when we consider that Asian-rich California has among the toughest NMS semifinalist qualification thresholds. Meanwhile, the national number of Jewish semifinalists comes out at less than 6 percent of the total based on direct inspection of the individual names, with estimates based on either the particularly distinctive names considered by Sailer or the full set of such highly distinctive names used by Weyl yielding entirely consistent figures. Weyl had also found this same relative pattern of high Jewish academic performance being greatly exceeded by even higher Asian performance, with Koreans and Chinese being three or four times as likely as Jews to reach NMS semifinalist status in the late 1980s, though the overall Asian numbers were still quite small at the time.56

Earlier we had noted that the tests used to select NMS semifinalists actually tilted substantially against Asian students by double-weighting verbal skills and excluding visuospatial ability, but in the case of Jews this same testing-bias has exactly the opposite impact. Jewish ability tends to be exceptionally strong in its verbal component and mediocre at best in the visuospatial,57 so the NMS semifinalist selection methodology would seem ideally designed to absolutely maximize the number of high-scoring Jews compared to other whites or (especially) East Asians. Thus, the number of high-ability Jews we are finding should be regarded as an extreme upper bound to a more neutrally-derived total

The U.S. Math Olympiad began in 1974, and all the names of the top scoring students are easily available on the Internet. During the 1970s, well over 40 percent of the total were Jewish, and during the 1980s and 1990s, the fraction averaged about one-third. However, during the thirteen years since 2000, just two names out of 78 or 2.5 percent appear to be Jewish. The Putnam Exam is the most difficult and prestigious mathematics competition for American college students, with five or six Putnam winners having been selected each year since 1938. Over 40 percent of the Putnam winners prior to 1950 were Jewish, and during every decade from the 1950s through the 1990s, between 22 percent and 31 percent of the winners seem to have come from that same ethnic background. But since 2000, the percentage has dropped to under 10 percent, without a single likely Jewish name in the last seven years.

This consistent picture of stark ethnic decline recurs when we examine the statistics for the Science Talent Search, which has been selecting 40 students as national finalists for America’s most prestigious high school science award since 1942, thus providing a huge statistical dataset of over 2800 top science students. During every decade from the 1950s through the 1980s, Jewish students were consistently 22–23 percent of the recipients, with the percentage then declining to 17 percent in the 1990s, 15 percent in the 2000s, and just 7 percent since 2010. Indeed, of the thirty top ranked students over the last three years, only a single one seems likely to have been Jewish. Similarly, Jews were over one-quarter of the top students in the Physics Olympiad from 1986 to 1997, but have fallen to just 5 percent over the last decade, a result which must surely send Richard Feynman spinning in his grave.

Other science competitions provide generally consistent recent results, though without the long track record allowing useful historical comparisons. Over the last dozen years, just 8 percent of the top students in the Biology Olympiad have been Jewish, with none in the last three years. Between 1992 and 2012, only 11 percent of the winners of the Computing Olympiad had Jewish names, as did just 8 percent of the Siemens AP Award winners. And although I have only managed to locate the last two years of Chemistry Olympiad winners, these lists of 40 top students contained not a single probable Jewish name.

Further evidence is supplied by Weyl, who estimated that over 8 percent of the 1987 NMS semifinalists were Jewish,60 a figure 35 percent higher than found in today’s results. Moreover, in that period the math and verbal scores were weighted equally for qualification purposes, but after 1997 the verbal score was double-weighted,61which should have produced a large rise in the number of Jewish semifinalists, given the verbal-loading of Jewish ability. But instead, today’s Jewish numbers are far below those of the late 1980s.

Sure, it might be that many from the older generation of Jews married out to Gentiles, which would have had some dysgenic effect on the IQ of the younger generation, and I wonder how large that effect is versus that maybe Asians are actually smarter and that it’s almost entirely due to there not having been many Asians in America in the generation of my parents.

Ron Unz has quite some data and numbers, and though the surname methodology may underestimate the number of Jews somewhat, the argument is quite convincing. I was indeed surprised. Of course, at the real far tail, especially in the adult world, it might be a very different picture. After all, the stuff Unz refers to is for kids, lacking in depth and substance, and on that, I can even attest that I’ve seen IMO gold medalists who I don’t find all that smart, holistically.

Well, I’ll conclude that my USAMO winner white Gentile friend will say to me that it’s universally known that Jews know how to talk, especially in the American/Western context, and how to manipulate Anglo minds, not to mention they also have more resources social and economic and acculturation wise to play the game. He thinks their actual ability is vastly inflated, and Ron Unz might agree with that, though surely, Ron Unz as an individual seems to be at the highest order (or at least close).

I cannot make a definitive conclusion on this. It’s very complex obviously. Again, I will reiterate that I believe Chinese, if they want more credibility, as much progress as they’ve made, still need to do better at home, especially at basic science research. In engineering and technology, China seems to already be at or close to the forefront on the world stage, and China ought to do the same for science. No matter what, Chinese, the largest group among Asian-Americans, will be seen as foreign in America. They should not expect too much especially at the top, especially in non-STEM areas. That friend of mine says that once Asians become rich enough, they can get well-connected, start companies, and hire mostly their own, and play the same nepotistic games that Jews play, or are at least perceived to. Maybe America, as a new nation founded on ____, really will turn into nobody’s country, into a potpourri of groups contend. Some in the alt-right, including Jared Taylor, fear that America will end up like Brazil. In any case, Chinese are kind of aware that in America, they are foreigners no matter how well they speak English or how “American” they try to be and that for that reason, it’s not really terribly fit for them to enter politics or American cultural life. I’ll say that the Jews as a group could learn a bit from that as well, especially when since 1947, they’ve had a country large enough to house the now 15 million of them. It’s not really terribly virtuous to place yourself in the parasitic elite in Western countries all over the world where one is a distinct minority in partial camouflage.



前几天,我看了徐道辉(Steve Hsu)与美国极右派Stefan Molyneux的讨论,有了深刻的感受。可以回想到徐提到在首尔或北京,一个女人可以在半夜到街上而对安全无所担忧,在美国的大城市这是无可想象的。虽未直言,可我们都知道是因为智商与罪犯行为的反相关关系应用在智商分布稍高的东亚国家之特例。徐也说道我们都有一点尼安德特人的血统,可是其占我们整个基因组很小一部分。智人所能创造的好多是尼安德特人无能的,故逐渐后者被前者覆盖而代替。他说我们可以想象他们创造物理学家或诗人的几率会比我们小很多。徐又漏出了他直截了当,对政治正确毫无在意的幽默,说:“我觉得我不会愿我的女儿嫁给一个尼安德特人。”



徐道辉对东亚国家所做的成功的地方显然已有认识。谈到这儿,我想起我的那位俄罗斯朋友曾经还开玩笑将他叫做“你的(东)亚裔优越主义朋友”。当然,徐也提到普遍被认为的东亚的过于顺从的文化不利于出做出革命性科学贡献的孤胆怪才,甚至东亚人天生就天才性格少出的可能性。毕竟人类文明最跨越性的时代显明是西方白种人创造的,是西方人创造了文艺复兴,科学革命,工业革命,周游并且占领殖民了几乎整个寰球,而在十九世纪中旬,西方白种人与其他人几乎是人夷之别。十九世纪末期,日本人和中国人都要想西洋人学习,尤其是学习他们的先进科学和技术。在那个时候,东方人都怀疑自己脑子本质上就是不如西洋人的,此在西洋遥遥领先横扫全球的情况是所预料的自然心理反应。不过,日本以飞快的速度吸收了大多西洋科技,成了第一个非西方现代化国家,此由1905年俄日战争之胜利所标志。中国人现代化的比日本晚的多,二十世纪上半中国所处于的内忧外患以及军阀内战对此有大大阻碍,可是中国派出去的留学生在理工科学的很好,逐渐把这些更先进的知识带回了他们的祖国。中国人和日本人打进近代科学的绝对一流的成果也都是从数学然后理论物理开始的,日本是第一世界大战时的高木贞治(Teiji Takagi)然后三十年代时的汤川秀树(Hideki Yukawa),中国是二战时期左右的华罗庚和陈省身,然后五十年代的杨振宁和李政道,这些都是在最需要智商的学科,表示了东亚民族极端的科学聪明才智。之后,中国人和日本人出的这样的人越来越多,现在已到频繁,不过在最顶级比西方还是要差一点或一些,尤其是中国。所以或许还是西方人最能出最天才的种子。

我总是觉得最最聪明的人大多还是犹太人,可以说二十世纪是没有一个,至少得以广泛认可的,与John von Neumann齐智的人了。同样,即使在科学深度和眼光也是犹太人处于巅峰。但是,这一点不是完全没有异议的。我的一位非犹裔国际数学奥赛金牌白人朋友却觉得东亚人比犹太人聪明,令我吃惊。不过,或许今天在年青一代还真的是这样,以中国学生为主的东亚学生常是精英数学竞赛的佼佼者,甚至占其主部为据,加上今年也有越来越多东亚数学家做出的精彩结果,以张益唐的孪生素为代表。徐道辉也跟我说,东亚人和犹太人是两个很不同的分布,前者多广泛,后者少儿精。对此,我想到了类似的比喻,那就是犹太人如斯坦福或哈佛,而东亚人如伯克利。此人口分布之差依然会给以最精犹多之结果,在这一点,我记得一位华裔国际数学奥赛金牌曾跟我说,犹太人虽然平均更聪明,但是东亚裔可以由数量弥补,照样可以出陶哲轩或张益唐这样的人。当然,智力难以作绝对的比较,因为每个人都有他自己的风格和特点,有长有短,而我感觉东亚人与犹太人,作为集体,表现出他们才华也是各有各的“民族特色”,是上千上万年分开进化所导致的基因和文化差异的必然结果。

诸多西方右派学者会谈到当代西方劣生的趋势。在此,已逝世的加拿大心理学家Philippe Rushton曾提到黑死病大大提升欧洲人智商而促使西方和人类文明大爆发的设想,并且猜测从此,白种人一直在逐渐退化到他们所有的“自然水平”,将此事件划为一个彻底改变人类走向的大偶然,并且对东方社会,尤其是中国,具有在西方主流极少有的乐观,并早在2006年就大胆说“他们足有脑力与我们同步或比我们更高。”现在看来,他是一位极其有远见的敢于纯粹真实的挑战主流错误观点的孤胆西方心理学家。他的研究发表曾经引起过轩然大波,不过我相信历史会证明他为类似于伽利略的科学烈士。Rushton的研究既科学又透彻,将智商和性格,在种族之间,与大脑和整个身体结构提出了整体的带有生理发育和进化缘故的描述与结论。Rushton的一位同派对偶学者Richard Lynn甚至觉得东方人会是西方文明的继承人,认为中国有更先进的,更高智商性质的,可以做出更有效决定及决策的”专制“制度可促进超越似的腾飞。





Innate mathematical ability

This morning I had the great pleasure of reading an article on LessWrong on innate ability by Jonah Sinick. Jonah has been one of my greatest influences and inspirations, having interacted with him substantially. He is unusual in one of the best ways possible. I would not be surprised if he goes on to do something extraordinary.

When I catch up with Jonah, I like to talk with him about math, mathematicians, and IQ, which happens to be what that article of his on LessWrong is about. 😉 That article resonates with me deeply because I myself had similar experiences as he did. It is hypothesized by me that I was also twice exceptional, albeit in different ways, with its effects compounded by my unusual background, all of which mediocrities within the American public school system are not good at dealing with in an effectual way.

This writing of Jonah has brought forth reflections in my own mind with regard to mathematical ability, development, and style. I’ll say that as a little kid under 6, I was very good at arithmetic and even engaged in it obsessively. However, by age 8, after two years of adjusting to life in America starting off not knowing a word of English, I had forgotten most of that. I was known to be good at math among the normal normal students; of course, that doesn’t mean much. In grade school, I was not terribly interested in math or anything academic; I was more interested in playing and watching sports, particularly basketball and baseball.

I didn’t have any mathematical enrichment outside of school other than this silly after school math olympiad program. Nonetheless, I managed to test into two year accelerated math once I reached junior high, not that it means anything. In junior high, we were doing this stupid “core math” with graphing calculators and “experiments.” I didn’t realize that I was actually a joke at math until I failed miserably at the state mathcounts contest, having not prepared for it, unlike all those other tiger mommed Asian kids, who to me seemed way beyond me at that time. It only occurred to me that I might have some real talent for math when I made the AIME in 10th grade, taking the AMCs for the first time, being one of four in my high school of about 2000 to do so. I thought it was fun solving some of those math contest problems, which were more g-loaded, with an emphasis on the pattern recognition side.

It was after that I started to read up on the history of mathematics and mathematicians. I taught myself some calculus and was fascinated by it, not that I understood it very well. But I could easily sense that this was much more significant than many of those contrived contest problems, and soon, I began to lose interest in the contest stuff. It was also after that that I learned about proving things, which the American public school math doesn’t teach. I finally realized what mathematics is really about.

Like Jonah, I had some difficulties with careless errors and mental organization. I don’t think my raw intellectual horsepower was very high back in high school, but fortunately, it has improved substantially since then that it is for the most part no longer the major impediment.

I took calculus officially in 11th grade, and it was a breeze for me. I could easily compute the areas and volumes and such but the entire time, I felt quite dissatisfied, because I could not actually understand that stuff at a rigorous, theoretical level as I poured through our textbook that went up to vector calculus during lecture, which was rather inane, expected if one considers the mismatch between cognitive threshold relative to the distribution of ability of the students. I knew from reading online the rich world of math far beyond what we were covering, most of which I was not intellectually mature enough to access at that time. However, I vividly remember during summer after 11th grade, while attending a math summer program, I was able to comfortably write out the delta epsilon definition of limit with understanding of why it was reasonably defined that way. Still, I would say I was still quite weak in terms of both my mathematical maturity and overall intellectual ability. There were too many things I wasn’t aware of, including the g factor, that I easily would have been had I been higher in verbal ability, which would have enabled me to read, absorb, and internalize information much more rapidly and broadly. In contrast, Jonah had discovered independently, or so he says, the lack of free will at the age of 7!

I made some incremental advances in my math knowledge from reading and thinking outside of school the next year. As for contest math, I almost made the USAMO. Though I had improved, I was still not terribly quick and careful with solving contest style problems and doing computations. I think close to graduation, I also solved some Putnam problems.

Only in undergrad did I learn real math more seriously, but even there, nothing too advanced. US undergrad is a joke, and I also was one, just to a lesser extent than most of my “peers.” Almost certainly, Jonah, based on he’s told me, had gained much deeper and broader knowledge at the same stage, from the reading works of giants like Euler and Riemann.

I’ve noticed how there are a lot of Chinese-(American) kids really into those high school math contests, and they now also dominate USAMO and Putnam (though careful, as in the latter, there you’ve got some of Chinese internationals drawn from the elite from China). I will say that at the lower levels, many of those kids have some pretty low taste and an inability to think outside the system that would enable them to discover the existence of real math, as opposed to this artificial math game that they enjoy playing or are pressured to doing so for college. Though those contests have a high pattern recognition component to them, there is not really much depth or substantial math knowledge. It is also my belief, with reference to Jonah’s article, that math contests are mostly M loaded while real math is more V loaded. So this behavior is consistent with the lopsidedness in favor of M and perhaps also short term working memory of Chinese students. It has also been Jonah’s belief that controlling for g, these contests select for low taste and value judgement, and I surely identify with that perspective. So maybe college admissions are somewhat fair to assess an Asian penalty?

Of the thesis of Jonah’s article, a representative figure is Terry Tao. There, Jonah also pointed out that Tao’s research in math is more concrete and problem solving oriented by pure math standards, in line with what appears to be the same lopsided (modulo the absolute level, as Terry is a far far outlier) cognitive profile of his based on testing at age 9 and 10. Again, people enjoy what they are best at, and though, Terry Tao is almost certainly at least +4 sigma at verbal, he is far more rare, at least +5 sigma, a real übermensch, in the (in some sense dual) pattern recognition component, which means he leans towards the areas of math more loaded on the latter. I have heard the saying that even other Fields medalists are intimidated by Terry Tao. The breadth and volume and technical power of his work is almost unrivaled and otherworldly. The media makes it seem like Terry is a league above even the other Fields medalists. However, Jonah seems to believe that the deepest and most leading of mathematicians are the ones who are more theory builders, who create through leaps of insight and synthesis new fields and directions that keep mathematicians busy for decades, and even centuries. That would be say Grothendieck or SS Chern, and an ability that is more loaded on verbal ability, crudely speaking. Again, I have felt the same. This might explain why the advantage of Chinese students is not anywhere near as pronounced in math research as in contests, and why some people say that generally speaking, the Chinese mathematicians are more problem solving and technical than theoretical, more analysis than algebra. Likewise, we can predict the opposite for Jews who are skewed in favor of verbal. A corollary of this would be that the Jews produce the deepest thinkers, adjusted somewhat for population, which is almost certainly the case, if you look at the giants of mathematics and theoretical physics.

I’ll conclude with the following remark. I used to revere somewhat those who placed very highly on those contests, until I realized that many of them are actually somewhat weak in terms of deep understanding and thinking at a more theoretical level. Yes, I have met MOSPers who got destroyed by real math and who are not very intellectually versatile, with glaring weaknesses; I was quite surprised initially that even I seemed to be smarter if not a lot than some of them. Once upon a time, I couldn’t understand those who appeared very strong at real math (and often also science and/or engineering and/or humanities) who struggled with more concrete math and/or contest-style problem solving, like Jonah, who has written on LessWrong of his difficulties with accuracy on the trivial math SAT. I’ve met this other guy, who I thought was an idiot for being unable to perform simple computations, who is leagues beyond me in the most abstract of math, who writes prolifically about partially V-loaded areas of math like model theory. Now, the more metacognitive me has awakened to the reality that I may never by deficit of my neurobiology be able to fathom and experience what they’re capable of. After all, there are plenty I am almost certain are and are essentially doomed to be very delusional by nature relative to me, and since I’m at the far tail but not quite so much, there are bound to be people who view me the same. I can only hope that I can become more like them through some combination of exposure and organic neurobiological growth, but I as a realist will not deem that very likely.

Metacognitive musings

It feels like ages since I last posted here, though a month ago is short in the time frame of great minds, who endeavor on projects for years and even decades. Since then, much has passed through my mind philosophically and meta-cognitively that I have not kept record of out of a belief in their triviality and lack of detail as well as the lack of natural inclination to write them out, compared to before. Though I share my thoughts with close friends through private (or not so, as nothing is private nowadays, electronically) channels, my past month has been loaded more on the intake end as far as text is concerned, with much of it being in Russian, and to my great pleasure, I now converse in it regularly with a number of people non-orally, including a charming, talented girl who won a worldwide chess championship for those under a specified age group.

Now to propagate my bull shit mental signals to eternity. I recall in some math circle, the motto was “existence is a state of mind.” I identify with this quite deeply. One cognitively mature to a sufficient degree can appreciate that what one experiences consciously, the extent to which it can be precisely characterized in qualitative terms, is a function of brain structure. Steve Hsu has proposed on his blog a toy model where there are n genes, each with some (finite) number of variants. This can be represented as a (bounded) lattice on an n-dimensional space, on which some metric can be defined, that correlates with reasonable metrics for outward behavior and cognitive manifestations. This space has been far from exhausted by all organisms which have lived on this planet. In particular, the theoretical extreme relative to the realized one is almost certainly Mount Everest relative to molehills which occur at several orders of magnitude higher of frequency. To go back to the motto, we must clarify that the “existence” here by the very context in which it resides is what can be perceived mentally by the human brain, which is very likely a minuscule subset of the theoretical maximum of biological perception of the world.

Growing up is a process of realizing successively how broken one’s former brain was. Eventually, provided that one is above some threshold in IQ, it will dawn on one that there are people capable to perceiving the world much more clearly and correctly than you can, and that there is a much higher world that one is incapable of experiencing. One sees the great extent to which what one becomes is determined at conception.

Different people think very differently, much of it owing to nature. People have different upbringings, which accounts for much of the variance in what one is exposed to, but when one is an adult and settles independently, one is more able to choose what to expose oneself in a free manner, free of cultural constraints, especially with the near ubiquitous access to information nowadays. To illustrate, a good example would be the countless I’ve seen go from evangelical to atheist and the reverse. This is especially so, in my opinion, for those higher on the bell curve, who are harder to control mentally.

I myself am a materialist, as is obvious from the thesis of this post. God to me is Mother Nature. I will not say I am an atheist as such cannot be definitely proven or witnessed before afterlife but I will shy away perhaps with partiality in the other direction. It is a misfortune maybe that God has produced me to be incapable of being penetrated by him and I have wondered what it would feel like to be the complement of what I am, though such is of course futile assuming non-existence of brain transplants. On this note, religiosity is, according to my reading online, rather heritable (surprise surprise), and it has occurred to me that perhaps those of East Asian descent due to a divergent branch of evolution are less prone to religiosity, especially with my seeing that Japan, in addition to the heathen middle kingdom, now directed by a Marxist-Leninist (or maybe only in name) party, is also high in its degree of secularity. The aggression within missionary men has both amazed and perplexed me. What is it, after all, that enables the conviction that those conquered, enslaved, culturally teared apart, are in fact being saved in the name of God or to be less extreme that there is such a holy duty to proliferate this conception of the human mind, a product of God itself, (or not) that fuels such intrusive and unceasing behavior, often targeted towards the young, impressionable, and often impoverished? I do not see behavior anywhere near in degree of magnitude exhibited within those on the reverse end of the religious/ideological spectrum, as an expected value. One’s religious faith, as far as I see, ought to be a private affair that is respected regardless of differences, barring cases of excessive and grotesque infringement. Moreover, the Christian culture, the Abrahamic tradition, is one I seek to know more about on a factual basis, as it is an integral and indispensable part of the Western culture that is dominant globally today owing to the exponential leap in penetration of thought and sophistication of production sprung forth by an explosive awakening of the slumbered Greek and Roman, tearing fetters and opening grand vistas the last half millennium.

How to see heaven and earth in the eyes of a demigod? How to transcend mankind? How to control the own evolution of man? The very essence of our existence, the cognitive of it, to be unveiled the years soon to come!

Programming types

Programming, the intense hacker side of it, attracts a certain breed of person. In short, I would put it as that it attracts those who are higher in autism than in g, though of course one needs to be reasonably high in both, especially the verbal side of g, as its activity is largely one of reading (of logs and documentation) and writing (of code (and its supporting documentation), the quality of which has good variable names as a major component). I do feel at times that programmers, even elite ones, are lacking in scientific taste. Many of them are mathematically null. They thrive on and even love the detailed minutiae involved in the work, such as encodings (like UTF, ASCII, that type of thing), the ins and outs of Unix, and arcane facts of various languages. I had to encounter in my work today parsing of CSV files, and it turned out that the CSV reader was not reading under the correct encoding. I ended up diffing my output with the output generated via a means more or less guaranteed to work to aid such’s diagnosis. I’m not bad at this type of thing any longer, having trained myself or more like grown to be able to patiently resolve such problems in a systematic, foolproof fashion.

Does that mean I enjoy this type of thing? No, not at all, though I find it tolerable, more or less. Too autistic for me. It does not have the depth that mathematics has. It has not the beauty of poetry or of music. It has not the wittiness of words or the expressiveness of (human) language. Nor does it have the significance on the world that politics has. There are more meaningful to be doing than programming, though needless to say there is much demand for it as the world now runs on computer programs, which are written mostly by politically incompetent and often socially awkward who answer to morons with MBAs.

I’ve come to notice that programmers tend to be very narrow. They only know programming. There are of course exceptions. Mathematicians and to a greater extent physicists are more broad, and more deep. It makes them very boring to talk with. The people who are more well rounded who are in programming are often, from my observation, in it for the easy money, which is of course paltry relative to what the parasites of our society suck in, but nonetheless a very good sum by the standards of ordinary folk.

There is of course another world of programming, that of the incompetents, who often know only Java and barely know any computer science even. They’re far from the functional programmers who I work with. This industry is so in need of grunt labor that those people manage to find their way into six figure salaries. Yes, this includes places like Google and Facebook. There are Google engineers who don’t know what the difference between stack memory and heap memory is and who think C++ pointers are scary, who make 200k a year or almost. I won’t talk more about them. Waste of breath.

Some speculation on the possible detailed linkage of brain structure to cognitive function

I recall seeing, probably first on Steve Hsu’s blog, that brain structure is highly heritable, as high as g is, surprise surprise. In fact, g, currently measured non-physically through cognitive tests, is, to my intuition, unguided by detailed relevant knowledge of the area, in its scientific essence a function of brain structure that could be computed through data obtained from a direct neurobiological measurement (through MRI and what not).

On this, I have hypothesized a future where we will be able to predict with reasonable accuracy, to a fine grain of detail, behavioral traits as complex as one’s personality or one’s cognitive strengths and weaknesses (like what subjects one will find more intuitive), which have a high heritable component, and not only that, the course of their development across age. A clone of me who lived in that world could be told at age n or so you will become like X, feel subjectively like Y, be at X \pm \sigma across age group at Z, etc, at age n+k, k > 0, k \in \mathbb{N}. Of course, if say n = 6, I would only respond with a sense of awe and incredulity, but across time, I would attain ever deeper and more realistic understanding of what was going on, as can be said for virtually every subject matter for virtually every human that has ever lived on this planet. Perhaps this is pushing it a bit too far. Learning to this degree of precision may be too much; there would not be enough data, even if we sequence every human on the planet and run on a computer cluster orders of magnitude larger and more powerful than, say, Google’s jobs which continuously improve the models. However, I do believe we will get somewhere. We will reach the point where we can deliver with reasonable level of confidence personalized medicine and education and such for every individual in a way that based on rigorous biostatistical data as opposed to on blind human intuition, which of course still works well, most of the time, as is done now. What other comes to mind is, as you can probably guess, the possibility of evolving a population vastly different from our own, perhaps one with ability far beyond that of homo sapiens (say, a population wherein everyone is smarter than John von Neumann (which I highly doubt, as that would be rather non conducive to the well functioning of the world)). As for that, as for what should be done, I have little to say other than that we’ll have to decide for ourselves. I do think it would be not a bad thing to use this newly found knowledge to eliminate severe dysfunction and suffering on this planet, which could be done through the likes of preventing the birth of, say, those prone towards severe depression or towards becoming serial killers.

We’ll have to see what happens. Maybe I will be able to see something exciting, possibly even to the level just described, in my lifetime. I am still young, after all.