Quotations from China

“The genetic basis of intelligence has been ignored for a very long time,” says Mr. Zhao. “Our data will be ready in three months’ time.”

There is more here.  Here is some further explanation:

At the Hong Kong facility, more than 100 powerful gene-sequencing machines are deciphering about 2,200 DNA samples, reading off their 3.2 billion chemical base pairs one letter at a time. These are no ordinary DNA samples. Most come from some of America’s brightest people—extreme outliers in the intelligence sweepstakes.


Uh, information about outliers would not be very helpful. They're OUTLIERS.

Ah yes, all those brilliant people just wasting that time and money - when they could have just listened to ... Vernunft.

Oh dear, you don't know what an outlier is.

Here is a ball. Perhaps you would like to bounce it.

They are comparing highly intelligent individuals to those with average intelligence. While that might tell us only what makes individuals highly intelligent and not what causes variation in the normal range (the former would of course be very interesting in itself), in all likelihood high intelligence is just the end of a continuum in terms of genetic architecture. Steve Hsu, who is involved with the BGI project, has suggested that intelligence differences are mostly about mutation load, i.e., the fewer deleterious mutations you have, the higher your intelligence.

Has he? I thought that was Cochran's idea - though I wouldn't doubt that they commiserate from time-to-time. At any rate, it all appears to be rather academic at this point, as Vernunft has found a fatal flaw in their study (I'm being sarcastic, should that not be blazingly obvious to anyone).

See here for Hsu's views, based on preliminary results from the BGI study. The idea is not original to Hsu or Cochran, see this paper, for example.

Looking more closely at this, Hsu appears to argue that the deleterious variants are relatively common so it's not the same model exactly.

"People have chosen to ignore the genetics of intelligence for a long time," said Mr. Zhao, who hopes to publish his team's initial findings this summer. "People believe it's a controversial topic, especially in the West. That's not the case in China," where IQ studies are regarded more as a scientific challenge and therefore are easier to fund.

But critics worry that genetic data related to IQ could easily be misconstrued—or misused. Research into the science of intelligence has been used in the past "to target particular racial groups or individuals and delegitimize them," said Jeremy Gruber, president of the Council for Responsible Genetics, a watchdog group based in Cambridge, Mass. "I'd be very concerned that the reductionist and deterministic trends that still are very much present in the world of genetics would come to the fore in a project like this."

I guess the scientific study of intelligence is another one of those jobs that Americans just won't do.

Americans and Europeans.

Here's the irony. The formerly and formally communist regime has the people to undertake a study of innate intelligence that societies not committed even ritually to such an egalitarian philosophy are unable to conduct. Good. A non-white people from a land groaning under dictatorship can go ahead and uncover the genetics behind cognitive ability. Hopefully it can be put to better use than reaffirming the inheritance of natural selection like various race nationalists have the limited imagination to aspire to.

I doubt very much that, if anything definitive comes out of this, the PRC's policy response will conform to SPLC guidelines.

Here's the conclusion of my review of James D. Watson's 2007 book "Avoid Boring People:"

In his book's penultimate paragraph, Watson raises The Forbidden Subject:

"A priori, there is no firm reason to anticipate that the intellectual capacities of peoples geographically separated in their evolution should prove to have evolved identically. Our wanting to reserve equal powers of reason as some universal heritage of humanity will not be enough to make it so."

Watson then foresees the gibberish and lies that he has had to endure over the last week:

"Rather than face up to facts that will likely change the way we look at ourselves, many persons of goodwill may see only harm in our looking too closely at individual genetic essences."

(Here, Watson, a famously ornery son-of-a-gun, shows some unexpected—and, sadly, undeserved—graciousness in referring to his future demonizers as "persons of goodwill".)

Watson ends his autobiography with a brief but telling exchange with then-acting Harvard president Bok:

"So I was not surprised when Derek asked apprehensively how many years would pass before the key genes affecting differences in human intelligence would be found. My back-of-the-envelope answer of 'fifteen years' meant Summers's then-undetermined successor would not necessarily need to handle this very hot potato.

"Upon returning to the Yard, however, I was not sure that even ten years would pass."

So your review is a series of quotations and terse commentaries? You would have made a superb philosopher in what historians were once honest enough to call the dark ages.

It takes a superb philosopher now to write the conclusion to a book review?

"So your review is a series of quotations and terse commentaries?"

I am of the opinion that James D. Watson's opinion on the future of genetics is more interesting than my own opinion.


A bit of humility that probably would have never occurred to Therapsid.

Because it takes humility to defer to a Nobel laureate, an actual scientist to boot, albeit one who has done no serious research for two generations, who just happens to agree with your racial theories.

No surprise that a racist misses a critical distinction - 'in our looking too closely at individual genetic essences'

'Individual' - we are only one race, composed of individuals. This statement has the advantage of being factually accurate, supported by empirical data, In contrast to the neverending evolution of how racists define race, to fit whatever notion they are peddling this generation.

Very sad statement from you, given that it is the opposite of reality. No serious evolutionary biologist denies the existence of race, which is literally as plain as the nose on your face. Did you miss the part about "peoples geographically separated in their evolution" (= races)?

"Our wanting to reserve equal powers of reason as some universal heritage of humanity will not be enough to make it so." - James Watson

Is Watson a racist? How about Steven Pinker?

The left fancies itself Defender of Science and takes great delight in mocking the cherished beliefs of evangelical Christians, for example, who they don't generally feel the need to consider 'people of good will.' Shoe's on the other foot here. Enjoy.

Personally, I wouldn't be surprised if it takes longer than the 10 to 15 years James D. Watson predicted. This stuff is complicated. The important question, though, is how likely is it that this day won't arrive someday: not very, right?

Just like the probability that NOBODY would use a nuclear weapon during the Cold War. Add up all those years, and it's not very likely, right?

In terms of probability, the likelihood that no one will ever find the genes for intelligence is more akin to the likelihood that no other country besides the U.S. would ever DEVELOP nuclear weapons during the Cold War, rather than use them.

Two comments:

1. This is why I'm waiting to have kids...so I can have superbabies!

2. If the world turns into a war between genetically enhanced superhumans and AI-augmented cyborgs, I am going to raid Bruce Sterling's home and find his time machine, dammit.

I think Larry Summers would disagree with the statement that the genetic basis for intelligence has been ignored.

Correct. It hasn't been ignored, its been actively suppressed.

"Most of the samples so far have come from outside of China." I heard in several interviews with Westerners who visited China how Chinese people would come up to them and ask them how to be creative or how they became smart. You know this study is just so the Chinese could crack the reason why European people are not only intelligent but creative. They want it for themselves. I suggest people avoid the study.

This sounds like a project driven by one ambitious guy, not China policy. Of course they crave creativity, but I don't think it is any surprise that it is lacking in China. Until very recently creativity hasn't really been embraced, or even tolerated. Rote learning is still the norm.

A sample of high IQ (mainly Westerners?), not particularly selected for creativity, compared against several thousand random people from the Chinese population, would not particularly seem that useful in identifying a "Western creativity" factor, even if it exists.

So that's why the guy in the Hong Kong Chinese restaurant wanted to have me spit into a cup.

I'm confused. The leftists and Democrats have sworn to me that there is no such thing as IQ.

I have it on the highest authority that this is impossible. Everyone is the same on the inside. We're a soft, gooey blank slate.

These people who tell me IQ doesn't exist also tell me race doesn't exist. They can't be wrong -- they have PhDs! In super-tough fields like Anthropology -- which is 4x tougher than that icky, western heteronormative Physics.

No one says IQ doesn't exist ; The controversy is as to how much of it is genetically determined.

If you run out, I have some straw over here.

Good on the Chinese. I've always felt that there has been a conspiracy of silence in the west regarding iq research. Too many people take the view that it's best not to delve too deeply in case we don't like what we find.

But science is about truth, and it is the truth which leads to advances. It's deception and lies which hold us back.

Everybody accepts that certain physical characteristics are divided unequally among different ancestral groups, but apparently intelligence is a no no.

It is more than a conspiracy of silence, there is active suppression, both legal and moral, particularly in the US. This kind or research is being actively discouraged by Institutional Review Boards in the US (known as Ethics Committees in Europe). The weasel word to look for is "group harm".

Most researchers in the US have to undergo obligatory annual ethics courses by the CITI. (https://www.citiprogram.org/Default.asp) The CITI makes sure that researchers considering this type of research think twice: "New insights into the concept of group harms in vulnerable populations such as minorities and workers in a workplace setting and the use of Community Consultation to prevent injury to special social structures."

Here is a paper studying the concern of potential research subjects. These authors want to make sure such research is nipped in the bud:
"Differences were observed between white and black respondents, with 64% of black and 34% of white respondents reporting that their willingness to donate a blood sample for genetic research would be reduced substantially due to concerns about genetic research findings being used to discriminate against people by race or ethnicity."
'Patient Perspectives on Group Benefits and Harms in Genetic Research': http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3104869/

“The genetic basis of intelligence has been ignored for a very long time,” says Mr. Zhao.

Apparently Chinese researchers don't believe in literature reviews at the start of their research. Quite frankly, if he gets this wrong, how much trust can you put in any of his work?

The word "ignored" is his way of being polite about the dominant mode of thought in the West in recent decades.

Yeah! We need to go back to those better decades. Like the 1850s I guess, when Steve could have bought and sold flesh to show those uppity ignoramuses who had the whip hand.

when all you have is snark, you'd better be good at it. and you are not.

Not true. He also has hatred. But the force is not strong with this one.

If the only thing preventing the return of black slavery in the West is a lack of evidence that there are measureable differences between races ("population groups", if you prefer) as a result of genetic variation, I don't think you're going to like the future.

BS. In the west now, if we find weakness in anything, first instinct is to subsidize it.

Whether you believe America is in decline absolutely or relatively, either way American thinking is "on the way out" to some degree. It might still be unfashionable to hold "racist" ideas in 20 or 30 years, but the "racist" will be well funded by Chinese or Russian backers.

diversity is holding back america's intellectual advance. russia and china don't have the albatross of needing to pacify large, poorer-performing minorities or their self-aggrandizing sponsors.

True enough. They have the albatross of pacifying large, poorly performing majorities.

I suspect they won't find just one gene or gene complex for high IQ. If there were such a thing, I would expect there to be a distinct peak on the IQ spectrum from those that have it. The fact that IQ is well-approximated by a Gaussian is a strong hint that lots of genes (and probably other factors) each contribute a little.


Brains are really, really complicated, and thus likely require a lot of genes each doing a bit.

If Nobel prize winners average only 145 whereas some are up to 160 and higher ,once we discover the magic pill , do we only have prople gene-splice enough of it to go up to the Nobel level ? So that we can have more Shockleys?
And if we discover that IQ-wise genetically perhaps Germany was superior , we should regret having won the War?

I think it's a lot less likely that Nobel IQ is only 145, and a lot more likely that standard IQ tests become "noisier" at the upper end. I very much doubt that there's a strong statistical difference between someone who gets 199 vs. 198 out of 200 questions correct, or someone who scores perfectly in 60 minutes instead of 65 minutes.

Since the variance of a Bernouli estimator is sqrt(n * p * (1-p)), the further p is from 0.5 (i.e. the closer the IQ sub-sample is to getting almost all questions right or wrong), the higher the estimator variance.

Haven't had my second cup yet, but I don't think that's right. As p grows, sqrt(npq) shrinks.

I agree though that IQ doesn't sort well at the upper reaches, as that's not really its purpose. But I think that's more about the design of the test than about the mathematics of statistics.

Depends on which test you're considering. Most tests are normed around the average IQ, for obvious reasons, but some are designed to be sensitive at the high end. If you only take a single test, it's almost certainly going to be a general purpose one, but if you score high enough you might end up taking a second test to get a more accurate result.

I'm aware that some IQ tests are designed to be sensitive at the high end, but it's not possible for, say, people with an IQ of 150 to design a test that will distinguish correctly between people with IQ's of 190 and 200. People at the very top see things other people don't see. The difference is not quantitative but qualitative.

And I don't think psychological testing attracts too many people with IQ's above 150. It's a thoughtful profession but not in the same category as advanced mathematics, physics, or biology.

Measured IQ consists of the average score for short-term memory, reasoning and verbal agility which are handled by different parts of the brain. If were looking for genes for intelligence I would look at people who were outliers for each of these abilities not people who have a high average score since they are more likely to be determined by a single gene.

I suspect that people who excel in the real world would score higher on one of the subtest than most people with IQ of 160

There are various subsystems in the brain which can be identified.

They seem to have a positive correlate in quality though, appear to have a shared genetic basis, and they're all useful for thought time. Finding variants which boost all factors would be more desirable than any specific one.

I read this, and then an interview of a statistician from Google. A tale of two statisticians!


Good find! My first thought, on reading the part about the Google guy, was "Great - another genius being employed in using advanced techniques in service of the ancient art of separating dumb suckers from their money." But the techniques he's using to figure out how to get people to click on ads will probably lead to advances in things like driverless cars or other aspects of what will hopefully be the next wave in automation.

this was the part of the article that caught my eye:

"The scientific challenge is significant. Consider the genetics of height, which, like intelligence, is a complex trait governed by many different genes, each one with a tiny influence.

Attempts to find height-related genes didn't yield any reliable hits until the number of DNA samples exceeded 10,000. By studying more and more samples, scientists have now identified about 1,000 genetic variations that partly explain why some people are taller than others. Those results are replicable—and they hold true whether a person is from Iceland or Japan."

The sample requirement is telling us something about how complex genetic decompositions are. While I work with very different data, if I could only find a precise relationship with 10k observations then I'd be cautious in my conclusions. I suspect the take away from the IQ study, like the height one, will be, yes there are some genetic combos associated with high IQ, but those explain only a small fraction of the high IQ observed. And do we have good metrics of how much difference a few IQ points really makes in the real world and do those differences rise or fall when the high IQ person is place in an environment with more (less) high IQ persons?

This all strikes me as a reasonable scientific endeavor, it's the inferences that will be tricky.

"And do we have good metrics of how much difference a few IQ points really makes in the real world..."

Why do you say "a few IQ points"? We do have some pretty good idea about say, a full standard deviation's worth of IQ points. See Linda Gottfredson.

And about the "inferences" - I wouldn't worry too much about that - as far as what may be discussed in public, any findings from these studies will be surely ignored in the west.

Right now, ignoring this BGI endeavor, the importance of intelligence in success and failure in a modern economy, its heritability, and its unequal distribution among different groups of people are obvious to anyone observing the real world around us and confirmed by statistical measures available to anyone who cares to look. If these "inferences" are completely ignored when the evidence is as plain as the nose on your face, I seriously doubt that the appearance of some impenetrably complex genomic study is going to suddenly make the scales fall fall from our eyes.

Why do you say “a few IQ points”? ...well if you need a large sample, I generally assume you are trying to parse some small effects. if standard deviation differences are dropping out of the data you don't need a big sample...unless the drivers are super rare.

"...these studies will be surely ignored in the west." uh, did I miss something or was an article *in anticipation* of the study results in the WSJ? Maybe we disagree on the definition of "ignore?"

"the importance of intelligence in success and failure in a modern economy, its heritability, and its unequal distribution among different groups of people are obvious to anyone observing the real world" I observe (and live in) the real economy and it's not clear to me that higher IQ (particularly high up in the range) is always better, particularly if it come at the expense of other attributes.

Claudia, I am afraid the cat is slowly being let out of the bag. But keep up the good fight, the Inquisition managed to uphold the Ptolemaic world view from 1543 until at least 1633.


It's amazing to me that you're trying to preemptively downplay the significance of this. Assume it is just a few points and assume it's only in very limited circumstances that it matters. So what? It might matter a lot in those very limited cases and it might lead to enormous advances on other margins. Besides, it's not funded by the USA. What probability would you assign that NO big findings will emerge that invalidate your prior views? 99%? That remaining one percent still seems like it could have pretty large payoffs relative to a great deal of well-funded research on more trivial issues.

The fact that so many are reacting to this study with snark or outrage suggests that the science of intelligence has been actively ignored -- at least relative to its potential importance.

if standard deviation differences are dropping out of the data you don’t need a big sample…unless the drivers are super rare.
As a quantitative hereditary trait, like height, intelligence is assumed to be caused by many genes of small effect. That's why they're so heavily sampling from the far-right end of the distribution - by going out to 4 sigma, they hope to uncover some of these - or many - of these genes. The social impacts of standard-deviation size differences, on the other hand, are not small.

Maybe we disagree on the definition of “ignore?”
I think so - by ignore I mean we won't take the results seriously and apply them to policy prescriptions. If - and that's a big if - BGI finds genetic evidence of intelligence, my prediction is that these will not be taken seriously in the west. For example, the concept of "disparate impact" will not be reconsidered no matter what BGI finds.

"it’s not clear to me that higher IQ (particularly high up in the range) is always better"

Of course it's not always better, but in the aggregate, the mean economic success of people with IQ's of 130 will be substantially higher than those with IQ's of 70.
Are smart people always happier, or is intelligence all that matters for a healthy life? Of course not, but no one seems to care about that. When Larry Summers mused that men might be disproportionately represented at the highest levels of IQ, the outrage was nuclear. Suppose he said that women have better social intelligence than men - would men have been outraged? Of course not. Women would have still been outraged, because he would have been suggesting that women are better talkers while minimizing harder intelligence. That's because no one cares about these other supposed modes of intelligence or social abilities - everyone wants to be though of as smart in exactly the way that IQ measures it. Any other suggestion is an insult.

I think a lot of the reason liberals get uncomfortable is that the people who put forward arguments favoring heriditary intelligence tend to be out-and-proud eugenists. Just take a look at this post from ziel's blog: http://lyingeyes.blogspot.com/2013/01/theyve-got-devil-in-them.html

"everyone wants to be though of as smart in exactly the way that IQ measures it. Any other suggestion is an insult." Disagree. I don't know my IQ but I am sure it's well below several of you here and several of the people I interact with in real life. I am still productive...actually some of my proudest professional achievements have been translating the really-high-IQ person insights into something that others could use. The work would not have happened without the genius on the team but it would not have happened without me either. I am interested in this work, but I see it as a baby step not some game changing work...it's far too uni-dimensional to be the latter. And yes, there's all kinds of outrage in the world, but frankly I hear more of it your responses than in my comments.

If it becomes possible to enhance IQ through genetic engineering everyone will find a way to do it. Even if its a black market. The incentive will be too great.

Worst case the Chinese end up making a lot of money off it, but people will find a way. It's easy to be a blank slatist when there is nothing at stake. When its your childrens future? People have already shown that they break their own rules when that is the case.

More likely, charlatans will incorporate the science of IQ in the spiel for quack medicines claimed to increase IQ. How would you tell that it is quackery, given that it might take 10 years to show any benefit in your child? China has a very vigorous quack medicine industry.

A new documentary about that BGI project is now online; most of it is English, with some bits in Mandarin, and the subtitles in Dutch. An English version is also coming soon. And for anyone interested in signing up for the study, it seems that BGI is still looking for volunteers. That link and more at http://theboostbook.com/2012/12/documentary-about-bgi-and-the-search-for-iq-genes/

The effect of genes could be tricky. The allele for sickle cell anemia only causes the disease when you get it from both of your parents. When you only have it from one parent, you don't have the disease and you do have increased resistance to malaria. IQ could be like that -- the genes responsible might exert their positive effect when you only have one copy. Two copies might give you schizophrenia or autism or something like that.

Lefthandedness is something like that. IIRC, more high IQ people are left-handed than in the general population, but the average IQ of lefties is lower than the general population.

" 'If you can identify kids who are going to have trouble learning, you can intervene' early on in their lives, through special schooling or other programs, says Robert Plomin, a professor of behavioral genetics at King's College, London, who is involved in the BGI project."

I didn't realize that high-IQ people are the ones that have trouble learning and need early intervention through special schooling.

You also don't understand the study design. They compare high-IQ people to average-IQ people to enhance statistical power. The objective is to understand the genetic architecture of IQ at all levels of intelligence.

Here is what I think will happen:
They will publish a paper this year. It will list ~ 1/4-1/3 of the coding part of the genome as "explaining" 30-50% of variation in intelligence. There will be tons of hype coming from all directions. The critical validation study - predict IQ based on ~1000 new genomes - will not be done.

So, here's something we can bet on:

After Larry Summers got kicked out of Harvard's presidency for (among much else) having an informed opinion on gender differences in IQ at the far right edge of the bell curve, he was replaced by interim president Derek Bok, co-author of a book supporting affirmative action, The Shape of the River. Bok's interim term ran from July 1, 2006 to June 30, 2007, after which point he was replaced Dr. Faust, to whom Larry had given $50 million in reparations.

So, we can date the conversation between interim president Bok and James D. Watson, the co-discoverer of the structure of DNA, about when the genes responsible for some of the racial differences in IQ will be discovered to to 2006-2007. Call it January 1, 2007 for simplicity. So, Watson's initial prediction was 2022, then he thought 2017 possible.

So, here's a question: When will come The Date that Watson and Bok discussed?

A. Before 2017
B. 2017-2022
C. After 2022
D. Never

I'd probably put my money on C, but lots of people have more expertise than me.

The exact quote is "how many years would pass before the key genes affecting differences in human intelligence would be found" (emphasis mine)

If the definition of a "key gene" is "having much higher effect size than most" (say, > 3 sigmas of the overall distribution), then they will be the first identified reliably and it will be definitely before 2017. Problem is, together they will account for a miniscule proportion of intelligence.

If you are betting on something like "when will we be able to look at a genome and predict IQ within 16 points interval with 95% probability", then I am with you, it will be after 2022.


only stupid people want higher IQ, people with high IQ realize how miserable it makes them (often). I don't think it's all that obvious to assume that abnormally high IQ is somehow evolutionary better. It seems Average IQ is more of an evolutionary advantage than individual high IQ. Better to ride the wave than stick your neck out.

To be honest, reading that having a high school dropout running such an important lab which is intended to discover and shed light on our intelligence doesn't give me lots of confidence in the project. THIS IS NOT A START-UP ENVIRONMENT WHERE A DROPOUT CAN EXCEL. At the same time, I hope Mr. Zhao just provides some leadership and salesmanship, not the hardcore scientific knowledge nor the experience of interpreting the research results. With my own 140 IQ, i would speculate that this profoundly difficult task - explaining human IQ and the influencing genome - probably takes more than just a co-author of a paper on cucumber genome. :-)

He dropped high school because he already was successful as a researcher.

Patrik, you can choose to believe the BS. I don't.

Comments for this post are closed