[Epistemic status: I am not a geneticist, and even the geneticists I know aren’t sure about a lot of this. Take as speculation only.]
I.
PZ Myers argues against Stephen Hsu’s genetic engineering proposal here – a disappointing attitude toward mad science for a guy whose blog header is a crocodile-duck hybrid. The piece has a lot of errors, the worst of which other people have already discussed – but I want to talk about what I think is its strongest point. Myers writes:
Note [Hsu’s] estimate of the number of genes that contribute to IQ: 10,000. That’s half the human genome! Hmmm. I wonder if any of those genes play a role in other processes in human physiology that might be affected by his plan?
Here’s an analogy for you: let’s say a novice car designer has decided that the one quality of an automobile that is most important is speed, raw speed. He doesn’t know much about cars, so he asks more qualified engineers about what elements of the car contribute to acceleration and velocity, and they start off with the obvious…details of the engine, fuel mixes, etc. Then they’re talking tires. Aerodynamics. Weight. Pretty soon they have to admit that just about everything in the car is going to affect the speed at which it travels.
So our blithe designer decides that making a fast car is simple: we just look at each component of the car one by one, and we pick an available option for it entirely on the basis of which option makes the car go faster. We’ll easily be able to make a car that can rocket along at a thousand miles an hour, he thinks.
But we have to ask whether we would want a car where the seats and steering were optimized for speed, where safety options were discarded, where something like visibility or reliability were jettisoned for the sole virtue of going really fast.
This makes a lot of sense. A car in which every component was optimized for speed would probably be uncomfortable, unsafe, ugly, difficult to maintain, and otherwise not the sort of car you want to drive. A human in which every component was optimized for intelligence might well be unhealthy, ugly, physically weak, antisocial, et cetera.
And we can do more than just hand-wave at an analogy to cars. Natural selection constantly weeds out worse alleles and replaces them with better ones. If an allele increases intelligence enough to improve reproductive fitness, but has no negative effects, it should sweep across all human populations in an amount of time proportional to its fitness benefit1. We see genetic evidence of various alleles sweeping various human populations in both the distant and recent past. These do not include the intelligence-boosting alleles Hsu is talking about.
For example, Hsu cites Rietvald et al‘s finding that rs1487441 is linked to cognitive ability (though it only gives you 0.3 extra IQ points, typical of the generally unimpressive contribution of single genes). About 20% of both Europeans and Japanese have the (A:A) variant, which suggests that however many thousands of years it’s been since Europeans and Japanese diverged from each other isn’t long enough for the gene to undergo much selection. That means neither allele can have any overwhelming advantage, which means there must be some reason to have the opposite allele worth as much as 0.3 IQ points. I think this is the rigorous version of what Myers is saying.
Despite the fact that the race car argument makes perfect sense both analogically and rigorously, it seems to be wrong.
II.
What are the sorts of things we might trade off against intelligence? Perhaps fitness, height, attractiveness, health, longevity, social well-adjustedness?
But in fact none of these trade off against intelligence, many are strongly positively associated with it, and in some the link has been proven genetic!
People with high IQ tend to live longer. For example, a person with IQ 115 (85th percentile) is 20% more likely to survive to age 76 than an average person with IQ 100. One can of course posit many possible connections. Maybe high-IQ people are smart enough to eat healthy and exercise. Maybe rich people can afford both good schools and good doctors. Maybe good health behaviors protect the brain as well as the body and increase both IQ and longevity. But further investigation has cast doubt on all of these theories and strongly supports the hypothesis that no, the same genes that give you high intelligence also make you live longer. See for example the International Journal of Epidemiology: The Link Between Intelligence And Lifespan Is Mostly Genetic, which find genetics explain 95% of the correlation. A few of the genes linking intelligence and longevity may be already known; SSADH seems to be a contributing factor. My favorite study in this area, though, is one that is not yet complete: since all mammals are basically the same [citation needed] some London School of Economics researchers have developed an IQ test for dogs in the hope of checking whether the same correlation applies to them. Since canine intelligence doesn’t affect things like diet, exercise, or tobacco status, a positive correlation in them too would help solidify the finding. We’re still waiting on those results, but even without them the genetic hypothesis is looking pretty strong.
People with high IQ tend to be taller. This is interesting since height is often used as a measure of health and fitness during childhood, and since taller people get a bunch of advantages including being rated as more attractive and earning higher income. Once again we can imagine all sorts of possible confounders; once again studies find that the link is genetic. See for example Common Genetic Variants Explain The Majority Of The Correlation Between Height And Intelligence, The Genetic Correlation Between Height And IQ: Shared Genes Or Assortative Mating, Resolving The Genetic And Environmental Sources Of The Correlation Between Height And Intelligence, On The Height-Intelligence Correlation.
People with high IQ may be more attractive. This is the conclusion of a meta-analysis that finds a positive correlation between intelligence and body symmetry, usually used as a proxy for attractiveness unaffected by things like hairstyle and cosmetics; a second study failed to find this relationship. The jury is out on the positive link, but there certainly isn’t the negative link that Myers’ race car would predict.
People with high IQ commit much less crime – which is going to be our measure for social well-adjustedness here since it’s well studied. Once again it’s easy to think of possible confounders (I’ll add lead levels to the usual lot). Once again the studies show that at least some of the effect is genetic – here’s one on low-IQ/antisocial-behavior correlation in children, and here’s one cleverly linking fathers’ criminal history to sons’ vs. nephews’ IQ and then throwing enough statistics at it to find that the relationship is genetic. Likewise, the relationship between high IQ and low drug abuse seems to be genetic as well.
People with high IQ tend to be more physically fit. This is the conclusion of a study of 1.2 million Swedes. I don’t have any strong evidence that this relationship is genetically mediated (although Gottfredson on the fitness factor may be relevant here), I just want to note that, once again, there is less than zero evidence for Myers’ race car hypothesis.
People with high IQ have lower rates of heart disease, stroke, circulatory diseases, and diabetes. Intelligence may or may not decrease cancer risk, but again contra the race car hypothesis, it certainly does not increase it. Sibling designs suggest that shared family environment during youth is not responsible for the benefits; differential socioeconomic status as an adult may be, but this status is itself likely caused be the intelligence differences.
So despite its apparent plausibility the race car hypothesis crashes and burns.
III.
In one sense, this is bizarre. It’s as if somebody optimized every part of a race car for speed, and found that by coincidence this also made it the safest, most comfortable, and cheapest car on the road.
Yet if we think about it some more maybe it’s not too surprising. Consider Niels Bohr. He was a Nobel Prize winning physicist, professional football player, activist who helped Jews flee the Nazis, loving father of six children, and so healthy he kept doing science well into his seventies. And his talents show every sign of being at least partly genetic – his brother Harald was a leading mathematician, anti-Nazi activist, and Olympic silver medalist; one of Bohr’s children was also a Nobel Prize winning physicist and another was also an Olympic athlete. So it’s obviously possible to design a human with all-around great genes. Why does evolution restrict such designs to the Bohr family?
I can think of a few possibilities, all of which people who know more than I do are welcome to shoot down.
First, some of these all-around-beneficial genes could be good in heterozygosity but bad in homozygosity. We know something similar is true in the case of sickle-cell anaemia, which is mostly good in heterozygosity (protects vs. malaria) and very bad in homozygosity (causes sickle cell). This is exactly the sort of gene that should exist at a constant low frequency in the population, never getting more or less common. If the frequency got too low, then there’s no risk of two carriers mating, so evolution would encourage it as a free disease cure. If it got too high, evolution would discourage it – any carrier would probably marry another carrier and give their kids sickle cell. Suppose there are ten such genes, each of which grants higher intelligence on heterozygosity and has a frequency of 10% in the population. The average person will on average carry one such gene and have a 10% chance of a horrible genetic disease. Maybe Niels Bohr lucked out and carried all ten such genes without going homozygous on any. Maybe some other poor guy who is lost to history got all ten genes homozygous and died at birth of ten horrible genetic diseases at once.
This would make Hsu’s gene-editing project very promising; all he would need to do is give everybody one copy of the relevant genes (and then never let them mate). But the hypothesis can’t be quite right: I think it would predict that Niels Bohr’s children would have unusually high rates of genetic diseases. In fact, the children of great men regress to the mean a little bit but show no signs at all of being unusually cursed.
Second, we could be talking not about polymorphisms but about mutational load. That means that there’s some genome that works for humans (plus or minus a few hundred thousand polymorphisms that aren’t too important at this level of analysis) and genetic health is determined by how many detrimental mutations you and your parents randomly accreted. If your mother spent too much time near the local nuclear reactor when she was pregnant, maybe you get a few hundred extra mutations and end up with lower IQ, a worse heart, less attractive features, et cetera. And this is obviously true in the case of a literal nuclear reactor, but I’m having trouble figuring out what plays the reactor role in real life. I know Greg Cochran and others have talked about things like paternal age at conception, climate, et cetera, but he applies these only to differences between populations. I’m not sure whether it would work out to expect a big difference in mutational load between Niels Bohr and his underachieving next-door neighbor. Maybe Bohr came from a long line of people who lucked out and got hit by unusually few cosmic rays? I don’t know if this makes sense or not. Part of my problem might be that I still don’t really understand how mutational load ever decreases – I’ve heard “the most heavily-loaded people are weeded out by natural selection”, but it seems like that should only be able to slow the gradual universal dysgenesis.
This would also bode well for Hsu’s project. In fact, it would make it even easier; it would reduce to the modal genome idea (make a baby whose genome, at each location, has the nucleotide which is most common at that location among all humans worldwide) which could be done without even performing the groundwork to see which genes do or don’t influence intelligence.
Third, maybe all of these other good things are trading off against things that were important in the environment of evolutionary adaptedness but not today. Greg Cochran brings up infectious disease resistance; some commenters bring up calorie requirements. This latter seems especially plausible; the brain uses a lot of energy and energy was a scarce resource through much of evolutionary history. Either of these would explain why evolution kept the seemingly detrimental version around for so long, and why right now in our low-infection high-calorie modern civilization one allele or another seems to be an unalloyed good.
Again, this would bode well for Hsu’s project, although the supergeniuses so produced would probably want to stay well away from any malarial swamps.
Or maybe it is some mixture of all four possibilities – Myers’ trade-offs, heterozygosity, mutational load, and disease burden. The latter three could provide a sufficiently positive effect for intelligence to hide the negative effect of the first. I would be really surprised if something like this wasn’t true – the theoretical argument for the first seems compelling, and it ought to happen at least a little even if it isn’t the main source driving intelligence differences.
I was going to write that in this case we’d have to sort through every intelligence gene one-by-one to make sure we weren’t getting one of the trade-off ones, but maybe this isn’t true – a genius designed by Hsu’s method should have on average the same number of trade-offs versus unalloyed-goods as a genius born normally (right? or am I missing something?). Since most of us would prefer a natural-born baby with IQ 150 to a natural-born baby with IQ 100, it seems whatever trade-offs are necessary to reach that point are widely considered worth it. So unless there’s a difference I’m missing between normal recombination and Hsu’s method, we should be okay with designing an IQ 150 baby as well (from a purely health-related perspective, at least).
Whether this generalizes to creating an IQ IQ 200 or 300 baby depends not just on ethics, but on whether for some reason the costs and trade-offs of intelligence compound more than linearly. It’s possible, for example, that there are ten different genes coding for something that protects heart health, all of which can be traded off against intelligence. If you switch one gene from heart health to intelligence, whatever, you still have nine genes protecting your heart. But switching all of them at once would be a bad idea. I don’t know any reason to think this is true, but it’s a possibility that might give us pause between the IQ 150 and the IQ 300 level.
Overall though, I think the race car idea, despite its plausibility, is likely to be less of an impediment to genetic engineering than it might seem.
Only One Footnote, But It’s Really Long
1. This is an assumption I’m granting for the sake of argument, but possibly not true at the margin.
Consider that evolution doesn’t care about intelligence nearly as much as we do. The most recent common ancestor of Europeans and Japanese wasn’t going to use her intelligence to design a mammoth-seeking rocket. In fact, it’s not totally clear why humans did evolve intelligence before the modern age; sure, tools are nice, but early hominids stuck to the same tools for a million years at a stretch; that doesn’t exactly give a tight feedback loop to work with. The most convincing argument I’ve heard is the Machiavellian intelligence hypothesis which says that our ancestors used intelligence to navigate tribal politics and gain status within a social group.
But this theory would naively predict that the smartest person in high school would be the most popular. If intelligence is for gaining status, it seems to have diminishing returns beyond a certain point, which would explain why evolution didn’t generally make us more intelligent even though greater-than-average intelligence is clearly possible (eg geniuses).
In the rare cases where evolution did have an incentive to evolve higher intelligence, it did so quickly and effectively. Several highly mercantile societies independently evolved the same set of genes producing higher IQ. The most notable were the Ashkenazi Jews, who have an average IQ 12-15 points higher than their European neighbors and whose genes show strong signatures of recent selection for intelligence; this most likely occurred during the Middle Ages when they were the mercantile class of Europe, since non-Ashkenazi Jews show no such effect. The genes involved tend to produce sphingolipidoses when homozygous, which shows a pretty good reason why evolution didn’t do this kind of thing more often. Myers has previously dismissed this research, but I think wrongly – the paper itself considers and rejects the all the criticisms he raises (see pages 15 – 31). The very short summary is that Myers dismisses the genetic pattern as “variations amplified by chance,” but the expected level of chance variations can be calculated and this isn’t it. Ashkenazim have similar heterozygosity to other Europeans in neutral markers – ruling out a simple bottleneck – and the mutations involved are too potentially deleterious in homozygosity to persist for many generations by chance alone. Further, the mutations are all clustered in a few key pathways, many of which are clearly linked to intelligence. For example, Ashkenazim are at high risk for torsion dystonia, which is associated with higher IQ in sufferers.
So in response to the argument that evolution must trade off against something else, I would argue that evolution doesn’t share our exchange rate. Suppose that we could gain 20 IQ points at the cost of having larger heads that are harder to fit through a birth canal (remember, some of the known genes for intelligence are associated with head size, and the obstetrical dilemma used to be a big deal!). For hunter-gatherers, who had little use for IQ but lots of use for getting through birth canals, this was a bad deal and evolution didn’t take it. For moderns, who can use IQ points to cure cancer and explore space, and who have modern obstetric techniques, it’s a lot more attractive. So yes, let’s be cautious, but I think we’d all feel pretty stupid if we avoided bootstrapping our way to superintelligence out of fears of “things man was not meant to meddle with”, only to learn later that the whole problem could have been solved with c-sections.
The Mating Mind says it’s just runaway sexual selection; we developed intelligence (and music and religion and art) to show off what an abundance of fitness we have to potential mates.
However intelligence would be a very inefficient way to display genetic fitness. I mean it is hard to tell intelligence just from seeing someone in modern society, where we have lots of status markers and group association markers related to dress. Imagine how much harder it would have been to identify that in pre-modern times. Wouldn’t it make a lot more sense that attractiveness , height, and such were better sexually selected for. I think its much more likely that intelligence helps in the “not dying” part of evolutionary selections. Especially due to its correlation with reaction times. A high intelligence + good reaction times, would be hugely useful in evading predators , hunting, and warfare.
If you ignore technology (which only became a factor very recently, after humans had already evolved most of our intelligence), humans don’t seem to be that much better at evading predators, hunting, or warfare than animals with much smaller brains. On the other hand, we are amazingly good at language, suggesting that we’ve been evolving for that for a long time, and, surely not coincidentally, we seem to talk a lot, which almost certainly isn’t a new thing, as it seems our exceptional language skills must have taken some time to evolve. Just talking to someone can give you some clue as to how intelligent they are. Is it really that implausible to you that intelligence was selected for on the basis that it helped people talk mates into bed?
Bullshit. The oldest known stone tools are 3.3 million years old, roughly an order of magnitude older than the divergence of anatomically modern humans. If you don’t think stone axes and spears helped with hunting, you’re nuts. For that matter, if you don’t think language helped with hunting, you’re nuts.
The technology of fire was significant for a long period.
Humans are very good hunters compared with every other primate out there. And before that, humans were expert scavengers, with tools that could quickly separate meat from bone to maximize what was scavenged.
Not to pile on, but I don’t know why you think language is unimportant. It allows us to send out scouts and sentries, receive detailed reports, make and refine complex plans to coordinate our actions. One on one man versus tiger we’re going to lose, sure, but we’re not one on one and the ability to make long-range plans, learn from failures, keep a history, all of that makes us tremendously good hunters and predator evaders. Look at the holocene extinction, which is basically just animals that no longer exist because humans were so good at hunting them.
Granted, this is well after adoption of tools, but as caethan says, the earliest technology was adopted very, very early, before there were even modern humans.
It allows specialization. Among chimps, some chimps can’t go hunting, knowing that if they fail, they will still be able to eat the veggies that someone else gathered, just as the gatherers know that if the hunters get something, they will all get meat.
Language lets you write that spot down.
It also lets you talk reproductively fit girls into bed, yes. But no matter how good you are at that, if you don’t follow up with food and protection for mother and babies, you fail reproduction forever. Good luck sorting out reproductive fitness due to charm and reproductive fitness due to coordinated hunting tactics.
And from the other side of the equation, if you’re stuck trying to persuade the skilled hunter to support the child you just conceived with the starving artist in the next cave, you’ll want every trick in the book including language. If you’re trying to coordinate with your fellow new mothers on which roots and berries are worth looking for and which are not to be eaten under any circumstances, those also are worth writing down.
Depends on how easy it is for the mother (and her family) to get food. In a sparsely populated area with no winters to speak of, i.e. Africa, it is easier than in Northern Europe.
Furthermore, the mother’s family could provide some of that protection.
It also lets you strike deals with the others after your leg was injured and you can’t hunt for a month: if they will feed you and your kids, you will make sure no animals get at the meat that’s drying for the winter, etc.
Cuckolding works the opposite way, in (mostly) monogamous birds the female finds a very “fit” male, and then tricks a low fitness male (often one in his first mating season) into raising the chicks. Interestingly this seems to boost the males fitness the next mating season.
I note that as always in these discussions, not only in Protagoras’ comment but also in some of the replies, “hunting” is considered the activity most important to the success of ancient humans, as if humanity were carnivorous.
This is a pet peeve of mine. When people think of primitive times, they always imagine hunting, hunting, hunting. That’s misleading.
Repliers correctly noted that humans always used technology to hunt, but technology was also used to convert plants into calories because cooking makes tough plant matter such as tubers easily digestible. That’s an important part of human success.
Another point: I think that the reason animals may superficially seem similar to humans in their sheer survival skills is that they have physical attributes that compensate for their stupidity. They have fur (even in Africa) so they don’t need to make clothes, they often have physical strenght and speed greater than ours plus claws and fangs, they have digestive systems which allow some of them to subsist on leaves, and others to digest even bones. The various animals are supposed to fill different niches, and we fill ours – we use our brains to compete with them, brains that allow us to make weapons, tools, clothes, dwellings, hearths and so on, all stuff animals need less.
>very inefficient way to display genetic fitness
That’s the weird point about sexual selection that makes it different from other forms; it can be and is extremely inefficient because that’s pretty much the point. Peacocks can barely fly and lack the camouflage of peahens, but being able to survive *despite* that and still be able to put on a colorful show makes them very attractive to peahens.
Sure but I meant inefficient in the fact that it is hard to judge and estimate. In a prehistoric tribal society, it would be very hard to look at someone and determine their intelligence in that way ( unless it comes with another status marker). On the other hand physical good-looks and prowess are easier to be noticed for.
No, it’s quite easy to notice that someone is smart in a prehistoric tribal society. You simply notice that they tell good jokes, have the best stories, and remember all of the edible species of mushroom better than you. Human intelligence is primarily verbal intelligence, as mentioned above, and humans love to talk.
In prehistoric tribes, everyone knew everyone else very well, and they knew each other over the whole duration of their lives, so each person had a very good idea of everyone else’s traits.
I usually have a pretty good idea of how smart someone is after talking with them for ten minutes. Granted, “talking” is not something that would have been an option in early human evolution, but at that point we’d all have been living in small bands and had much longer than ten minutes to display our fitness to each other.
Thanks for mentioning The Mating Mind! Just finished reading it, and my impression is that everyone who tries to be serious about human rationality should read it, because knowing about natural selection and not knowing about sexual selection is one of those things that predictably turn people into “Vulcan rationalists”.
My impression after reading the book is that natural evolution alone usually produces something similar to paperclip maximizers, and it is mostly sexual selection responsible for all deviations from this pattern. Even things that clearly contribute to survival are likely to have developed by sexual selection first. Pretty much everything that we value about humans — intelligence, creativity, altruism, humor — is a result of sexual selection. Without acknowledging this, people are forced into inventing silly stories about how the true purpose of art is painting mammoth pictures that can serve as training targets for hunters.
Understanding sexual selection is the key to understanding status, attractiveness; what makes some people popular and other people creepy; etc. It’s pretty much inevitable for creating successful rationalist communities, and for making the concept of rationality attractive beyond the existing communities. This is a huge blind spot for our kind; something that doesn’t have it’s own LessWrong sequence; something that is usually not discussed explicitly; and when it is, the discussions are usually horrible precisely because people try to shoehorn the artifacts of sexual selection into natural selection framework which seems kinda convincing for a “Vulcan rationalist” but makes everyone else shudder.
This is a part of knowledge that definitely should not be ignored by a wannabe rationalist.
I don’t really understand the distinction you’re making here. Natural selection favors genes with higher fitness, where fitness is given by the expected number of offspring per individual. As long as a species can only reproduce sexually, it’s basically the same thing as sexual selection.
Are you trying to say that many people misunderstand evolution as selecting for something other than what maximizes offspring, e.g. not grokking that intraspecies social pressure can drive selection as strongly as external “natural” pressure?
People habitually misunderstand evolution exactly like that. You can produce vapors in people by pointing that the future’s going to look like the people who have babies.
Yes. At best they understand it on the “guessing the teacher’s password” level. They will admit that the intraspecies pressure is possible, in theory, but then they will still try to connect any specific trait to an external pressure first. And when that fails, the only “intraspecies pressure” considered will be some kind of hostility against the members of your species, like killing them, threatening them, or lying to them.
The typical mistake that people make in debates about evolution is presenting a fake greedy reduction, because according to their current understanding, that’s the only kind of reduction there is. Anyone suggesting otherwise will be suspect of not being a good reductionist.
Yes, we all know that in the end this is all about how many copies of the gene make it to the next generation. The thing is that there are many different factors influencing that. And the nature is not really a maximizer of anything other than immediate gains, which is how species sometimes evolve to extinction, or other unexpected things happen.
To say it simply, when your understanding of evolution can “scientifically prove” that peacocks cannot evolve extravagant eye-spotted tails, because that would obviously reduce their abilities to hide or run away from predators (and the tails aren’t even useful for killing other males of the same species), that’s a good moment to realize that your map does not match the territory. I have seen many highly intelligent people make this kind of mistake. Of course, not about the peacocks; usually it’s about humans.
I know, right? I never see any pop evo psyche or just so stories about mating the rationalist or rationalist adjacent community. And nowhere near sufficient attention to status and signaling. Definitely a blind spot!
Sigh. That’s exactly what I was talking about. People do discuss these things regularly, and they are making the same cringeworthy mistakes over and over again. Then the wiser ones reflect upon the creepiness and tell them to shut up. (Then the former complain about censorship.)
How about reading a popular science book instead? (I heard that Library Genesis is now offering digital books at a really low price.)
How about explaining what are you talking about in detail, maybe naming some of the cringeworthy mistakes, instead of being vaguely condescending towards a large group of people?
I mean, you seem to be saying that when someone calls you creepy because you were analytically discussing human social interaction and mating, then you just need to learn more about evolution and it will fix the problem. This sounds like a very silly idea. If you want to stop being “creepy” to “normal people”, you need to stop applying analytical thinking to those matters. (Not that I think this is completely bad advice. Most online discussions of this sort are full of tedious bullshit.)
Here is what I don’t get about the “mutational load” hypothesis: Wouldn’t the mutational load of every generation increase, because they inherit mutations from their parents, and then have mutations of their own? So if there was one genetic layout that worked right, and differences in ability are all about how far your mutations take you from the Proper Genes, wouldn’t genetic quality degrade with every successive generation? Yet, this is the opposite of what we see.
Mutational load would only increase each generation if people with high levels of load weren’t being selected against. If they’re dying off before they can reproduce themselves, their load doesn’t get passed on, and the average amount of load doesn’t get any larger each generation.
Purifying selection.
Except that less intelligent people reproduce more.
That’s a very recent phenomenon.
You have to take into account natural abortions which might disproportionately happen to embryos with high genetic loads.
They do now. They didn’t in the past. Wealth (which is highly correlated with IQ) used to be a great predictor of who would leave behind the most surviving children. The invention of welfare changed everything.
Also, load isn’t just about IQ. It affects everything in a person’s body. People with more load are more likely to be mentally ill, to be physically unhealthy, to be deformed, to be short, to have asymmetrical faces, etc.
Most of those traits are still being selected against, even with welfare. High-load individuals are definitely surviving more often than they used to (in wealthy countries), but I’m not sure they’re actually reproducing more than low-load people. People still don’t want to mate with ugly, short, mentally ill people.
Low IQ is correlated with load, but it’s far from a perfect predictor of it, so even if low IQ people are breeding more than high IQ ones, it doesn’t necessarily mean high load people are breeding more than low load ones.
If high load people are breeding more than low load people, that’s a big problem. I’m just not sure if it’s the case, though.
As far as I can tell, height only started making women more attractive a few decades ago.
At one point hormone treatments were given out to overly tall girls to save them from a lifetime of spinsterdom.
@Nancy Lebovitz
That may be true. I have no idea how long height has been an attractive trait in women for. But it’s been attractive in men for a very long time, which is enough to make shortness an overall negative trait for the species (because even if shortness is neutral for women, short women tend to give birth to short sons, who are worse at attracting their own mates when they are adults than they would have been if they were taller).
Also, tall men tend to father tall daughters, which spreads tallness genes into the sex that doesn’t actually need them.
Nancy, I just asked my wife who is a classics professor about this and she said in ancient (Greek and Roman) times for women being tall often was associated with being considered attractive.
Anon:
Shortness is not neutral for women; it is (today) a very large selective advantage.
At least, short women have many more children than average-height women do.
Also the invention of birth control, and the development a society in which (for most families) children are an economic burden, not an asset.
(A biography of Benjamin Franklin notes that Ben’s father became a chandler, making candles and soap, and “being labor-intensive, it furnished early employment for as many children as came along.” Josiah Franklin fathered 17(!) children.)
Nancy-
I think a Jane Austen character is praised as being “tall, but not too tall.”
Is height really considered attractive in women? By men? I’m a man, and I’ve never for a moment thought that. Nor can I recall having heard a man comment positively on a woman’s height, nor negatively on a woman’s lack of height. At least not in a dating/mating context. The adjective “statuesque” comes to mind, which might indicate beauty in a sense, but not in the sense that attracts male romantic interest.
I do have a female friend who is 4’10”, and that may be a bit of a handicap in the dating world, but I think she’s probably better positioned than the 6’1″ girl I know, at least in a one-on-one setting. At a party, the 6’1″ girl is of course noticed immediately, for better or worse. In any case, they’d probably both have more success if they were 5’4″ – 5’6″.
Models tend to be tall. I’ve been told this is because clothes fit better on tall women. Also, I imagine models are disproportionately selected by women and gay men, which may be tilting the scales relative to models that straight men would select. And in any case, the way that someone appears alone in a photo spread is only loosely indicative of how she would be judged in person.
Women’s sense of beauty seems to be more influenced by fashion/culture/social proof, hence anorexia and “thigh gap” and such. I think men’s tastes are more static, though there is certainly still some change and cultural variation.
@Michael Watts
“At least, short women have many more children than average-height women do.”
How much of that is because of an individual woman’s height and how much because groups where most people are short (e.g. Latin Americans of many sorts) have more children than groups where most people are tall (e.g. Danes)?
Take two women of identical height and similar in all other physical characteristics. One in high heels, one in flats. Have them walk into a club together. Follow the men’s eyes.
Heels affect more than height, though. Stance and gait, most importantly.
Is height really considered attractive in women? By men? I’m a man, and I’ve never for a moment thought that.
Yes, it is. If you’ve never thought that, it just means that different people have different criteria for attractiveness.
Nor can I recall having heard a man comment positively on a woman’s height, nor negatively on a woman’s lack of height. At least not in a dating/mating context.
Again, yes, I have. For women, above-average height is generally considered a plus in a dating context.
Is height really considered attractive in women? By men?
I can’t speak for anyone else but as an individual data-point I’ve always been fond of tall, leggy brunettes.
I think height is considered attractive in women, but that the point at which it shifts to a negative is a lot lower. I’d say most men start becoming less attracted to a woman as she passes 5’9″ or so (especially for the sexual attraction part, as opposed to the admiring as a statue part), whereas men seem to continue getting a bonus until 6’6″ or even taller (interestingly, I think the ideally aesthetic man, as opposed to the ideally sexy man is probably shorter than 6’6″, as the ideally aesthetic woman may be taller than 5’9″).
This dating site study seems to find that met prefer women who are about as tall as they are, are much less interested in women that are taller than they are, but are only slightly less interested in women that are shorter than they are. Make of that what you will.
@RKN
Heels are also signalling on some level.
I think most men are decent at stripping away the clothes, makeup, and so on to arrive at a decent neutral approximation of how attractive a woman is. What the clothing and heels and so on are doing is more signalling of interest/status/availability and intra-female status competition, rather than actually changing the true attractiveness of the underlying body.
Yeah, part of the reason I picked 5’9″ was that that is the average height for men in the USA. I think when a woman starts being taller than the average man it starts to be more of a negative than a positive on the dating market.
The important part about high heels is not so much that they make you taller, but that they make your butt stick out. A big butt is a good counterbalance to a pregnant belly.
http://www.dailymail.co.uk/sciencetech/article-2979043/Why-Kim-Kardashian-s-booty-attractive-Men-prefer-women-curved-spines-suggests-better-FORAGING-food.html
I agree with you regarding fashion in general being more of an intra-female status competition, but I think it’s clear that heels are genuinely worn to be more attractive to men (like onyomi says, for reasons of posture more than height). Strippers do generally wear heels, after all. Where intra-female status competition comes in is with the particular brands/designers/styles of the shoes.
Fun fact: high heels were originally introduced in Europe as a way for fashion conscious upper-class men to appear taller (check out Louis XIV’s pumps in this portrait), but they were quickly adopted by Italian courtesans, who found that wearing them greatly improved their earnings for some reason.
“Statuesque” is a backhanded compliment . . . until men are actually looking to marry.
Its known that men prefer certain secondary sex characteristics when looking for a short term encounter and others when looking to marry. I suspect that height and whatever else goes into being stately–good posture?–is more important for those looking to marry, although I do not believe it has actually been tested.
“less intelligent people reproduce more”
Maybe at some levels, but I bet that 85 IQ people have significantly greater reproductive success than 60 IQ people.
Among European gypsies this is not true. It is almost certainly true for most populations, but in most populations a 60 IQ is more likely to be due to some organic defect, specific disorder, or environmental insult, than due to mere variation resulting from genetic assortment.
Sure, if you don’t have recombination or selection, mutational load will increase.
But we do have recombination and selection, so this doesn’t happen.
Father has a deleterious mutation in one of his copies of gene A, Mother has a deleterious mutation in one of her copies of gene B. Child 1 has both deleterious mutations, child 2 has neither; all else equal expected reproductive success of child 2 is higher. Scale up over many genes, variation in number of deleterious mutations between children of the same parents should scale as the square root of number of genes. A small fraction of deleterious genes get eliminated per generation, but the absolute number can be large; this elimination only has to balance against the mutation rate to keep mutational load constant.
Edited to add: selection to remove deleterious mutations might well be lower in modern times, but the resulting increase in genetic load would per generation would only equal the change in the removal rate, which is small relative to the total number of genetic mutations, so unless the total effect of genetic load is really huge you probably wouldn’t expect a very noticeable effect in a few genarations.
Good explanation.
Mutation count sort of does increase indefinitely. This is not a surprise because as time passes, species continue to speciate and diverge (and running it backwards, you get less genetic diversity, hence the use of ‘mutational clocks’). One man’s mutation is another mandrill’s fixed gene. But mutation load refers to harmful mutations. Whole genome-sequencing of parent-child trios indicates that a healthy live-birth child gets ~70 new mutations, but (almost by definition) none of them were fatal, and most are fairly neutral; which is how you get estimates like 70k very rare variants in each person. It’s the occasional harmful mutations which needs to be purged by purifying selection to keep a species from mutational meltdown; 1 mutation, 1 reproductive-death. Since many humans are sterile, or have no surviving descendants, so far it’s worked out.
Which is not to say that mutation load might not be a problem. But so far the GWAS and GCTA studies are not supporting it playing much of a role in individual differences. When a measurement-error-corrected estimate of common SNPs’ influence on intelligence is up to ~50% of variance and all genetics only accounts for 80% of variance, and studies hunting for rare variants affecting high intelligence keep coming up dry, it’s getting hard to see mutation load being important.
But a mutation load/purifying selection balance was one of the more elegant theories for why intelligence variants haven’t reached fixation given their apparent obvious utility, so this raises that question even more. Why are there genetic intelligence differences? Especially when the phenome studies (only some of which Yvain cited) show pervasive genetic overlap between genes which increase intelligence and genes which increase all sorts of other desirable traits like less schizophrenia risk? Right now I think probably the best theory is a resources or developmental one: a high quality brain, and body, are extremely metabolically demanding (did you know that in childhood, your brain and body have to take turns growing, because it’s metabolically impossible to do both?), and so any pro-intelligence variants runs into the risk of increasing vulnerability to famine and infection and injury; so you get selection for intelligence variants only up to the point where diminishing returns kick in hard, and then it’s better to have a more robust immune system or to stop growing early on and adopt more of a r-selected strategy, and then in the modern context where calories are abundant, education & intellectual pursuits apparently have been consistently dysgenic, so that eliminates any recent chance for driving pro-intelligence variants to fixation or even increasing their frequencies noticeably.
How can this be settled? Probably with ancient DNA. As more ancient DNA becomes sequenced, it becomes possible to look for selection signals. The current state of soft polygenic sweeps is pretty vague with only a few concrete examples of natural selection in European populations over the past few thousand years, but there’s clear evidence of a lot of polygenic sweeps, so like finding intelligence variants, it’s only a matter of getting more datapoints. If we see intelligence variants overall holding stable frequencies over thousands of years rather than overall increasing or generally drifting, then it’s probably the metabolic costs.
With the number of intelligence variants increasing to the 80s soon and ancient DNA regularly being sequenced, we may know quite soon! Exciting times. This is a golden age of behavioral genetics.
Can you link any phenome studies that I missed?
I still don’t (as I admitted in the post) really get the mutational load thing. Say that somebody starts with a mutational load of 0, and each child can have either 1, 2, or 3 mutations.
In the first generation he has three kids with mutational loads 1, 2, and 3.
Suppose evolution kills off the 2 and the 3 person for being unfit. The person with 1 has three kids with mutational loads 2, 3, and 4.
Suppose evolution kills off the 3 and the 4 person for being unfit. Now the 2 person has three kids with loads 3, 4, and 5…
So even if evolution kills off the most loaded people in every generation, it still creeps up with time, right?
The genetic load of a kid can be smaller than that of either parent since he only has a 50% chance of getting any one of his parents “bad” genes.
I’m still not quite there. Let’s say you get 30 de novo recessive deleterious alleles from each parent, with no overlap between them, so no loss in fitness as a result. Fine. Your kids only pass 15 of these on to their children but then add their own 30, and now you might be seeing some overlap. The only way it seems recombination works to get rid of them is if
1.) the mutations are very, very bad
2.) the genome is already at a really high entropy wig mutations from over the years, so that half of what you are passing down to your kids is suboptimal random junk.
Help me understand!
When the mutational load is small multiple of the mutations gained each generation, then just like you describe each generation ends up with more mutations. However, once the mutation load is sufficiently high it reaches an equilibrium: Say two parents each have 1000 deleterious alleles each. If each child has a 50% chance of acquiring each deleterious allele from each parent, then the number of deleterious alleles their children inherit should vary around 968-1032 (the standard deviation is around sqrt(1000)~32; actually I think it’s half that but this doesn’t affect the qualitative form of my argument). Accounting for new mutations the childrens’ mutational load should vary 998-1062. If only the most fit children survived to the next generation, then it would be reasonable to suppose that the next generation would also average at 1000 deleterious mutations. This is a far cry from “half of what you are passing down to your kids is suboptimal random junk”, as you suggested.
It’s interesting to look at this argument from the perspective of sexual vs. asexual reproduction. Imagine one of the parent was able to reproduce asexually instead. In that case, the child would be guaranteed to have 1030 deleterious alleles. This is the same in expectation as the 998-1062 alleles from sexual reproduction, but the variance is lower. Since the fitter children have more grandchildren, this means that a parent that reproduces sexually will have fitter grandchildren than a parent that reproduces asexually. Going forward one more generation, the result is that a parent which reproduces sexually passes more of their genes to great-grandchildren than a parent that reproduces asexually.
But when it comes to Y chromosomes and mtDNA’s, the load is just increasing indefinitely? Is there a reason why this isn’t as bad as it sounds?
But you’re still conflating new mutations with genetic load. You can have 3 children who each get 70 new mutations; for #2 and #3, they’re all harmless and so their mutation count has gone up, but their mutation load hasn’t; child #3, however, has a fairly harmful mutation, and partly because of that, only has 1 sickly offspring who dies while young, thereby extinguishing that harmful mutation from spreading. Children #2 and #3 go on to have grandkids without an issue. The number of mutations has gone up, but mutation load has remained constant.
Humans’ relatively low fertility can be thought of as a form of truncation selection: if you “roll the dice” and get all 1s, that embryo never starts to develop, or spontaneously aborts sometime during pregnancy.
Very nice to see two of our papers in SSC. Wish I felt half as confident about how to cast my vote for the Mayor of London today as I did about writing on Intelligence lifespan people and dogs!
I don’t know how well readers know Mark Ridley’s book (not Matt Ridley), Mendel’s Demon. Worth reading.
It’s a lovely exegesis of the relations between selection and the rain of mutation.
Mark frames this as a battle against mutational meltdown in a ‘Red Queen’ scenario (confusingly outlined in Matt Ridley’s book!).
I described the mutational load idea in our Author’s response to the critical letter by Kaufman & Muntaner to our IJE paper (http://bit.ly/1Ya2FDC). It falls out of the combination of pleiotropy, genes of small effect, and mutation rates. Whether or not variation in mutation load is a useful construct (a fitness factor – as the brilliant David Houle called it (‘Is there a g factor for fitness?’ 2000) is ultimately an empirical question.
To answer it we need better indices of overall mutational load. Such an index will come, but we don’t have a really excellent one yet. Studies with dogs and other non human animals have an essential role in this because they are free of lots of the confounders that beleaguer studies with people.
I’ve got a question for Rosalind about dog breeding. My vague impression is that the golden age of dog breeding was more or less the 18th and 19th Centuries, while my lifetime hasn’t seen much functional progress, at least not at the mass market level.
First, is this roughly correct? If so, then why?
Did humans pick most of the low-hanging dog breed fruit already? Is there just not much economic need for new working breeds, so most breeding energy goes instead into aesthetics? Has animal breeding become de classe? (It seems like in Darwin’s day it was a favorite hobby of educated gentlemen, but now perhaps breeding requires too much ruthlessness for contemporary educated tastes?)
Steve
@ Steve Sailer-
A dog breed today is not the same as a dog breed from 200 years ago. There has been tremendous change within the breeds over that span.
It’s possible that there are enough dog breeds to satisfy people, and it’s harder than it used to be to come up with a new breed. Also, a lot of breeds were developed from working dogs, and since there are a lot less people with working dogs [citation needed], that source of information is gone.
Actually, there are a lot of service dogs, but that doesn’t seem to involve highly specific body types. I haven’t heard of anyone working on developing new service dog breeds.
http://www.dog-breeds-list.com/akc-not-akc-recognized-dog-breeds
I’ve heard that border collie breeders don’t want to there to be a border collie breed because the AKC rewards appearance, not working ability or health.
I hear more about new cat breeds, like the munchkin. And I might as well mention our new feline overlords.
What about the explosion of deliberate crosses in the last 10, 15 years? I assume they don’t breed true?
@ Steve Sailer –
Not Rosalind but…
“Progress in dog breeding” can be difficult to define. The 18-19th century was the breed differentiation apex for domestic breeds for most of the West, but it can be difficult to separate out the development of new breeds vs the identification and clarification of regional landrace breeds. A lot of national/regional pride was expressed by the keeping and exhibition of “national native” breeds.
Animals in a single are differentiated into breed-specific population pools by time, distance, and culling out unwanted examples. Improved travel technology decreased the distance between different populations – which increased the possibility for cross-breeding.
Many people are resistant to culling dogs in the traditional sense of killing undesirable individuals (professional breeders less so than amatuers, and pet owners more than either, and American breeders less open to culling than Europeans) and not all defects are apparent until the animal reaches sexual maturity, so early pre-breeding neutering is not always possible.
With purpose-bred working breeds, there are definite standards & behaviors that the animals are chosen for. As noted, there are fewer and fewer opportunties for animals to “work”, and so lower-standard pet quality animals are more common. A new breed would most likely come from a new use for the dog, and there don’t seem to be many of those.
There are on-going shifts in existing breeds – English Bulldog’s increasingly grotesque body shape, the introduction of dappled/merle colors in a variety of short haired dogs (like Great Danes and Dachshunds), the American ‘show dog slouch’ in German Shepherd Dogs, the decreased prevalence of ‘collie eye’ among collie herding breeds, increased attention to preventing hip displasia by selecting against pedigrees that have poor OFA scores, and improvements in Dalmatians on two fronts – reducing congenital deafness and almost entirely eliminating the tendency to develop urate stones.
@ Nancy
I’ve heard that border collie breeders don’t want to there to be a border collie breed because the AKC rewards appearance, not working ability or health.
The AKC is an over-arching coordinating organization, and it’s hard to blame it for *everything* wrong with breeding dogs. But they do accept animals for registration based on reported pedigree, not on inspected appearance, certification of being free of health defects, and certainly not on any sort of performance.
Cats have been far less differentiated than dogs, so there is more to be done there.
What about the explosion of deliberate crosses in the last 10, 15 years? I assume they don’t breed true?
Correct. F1 crosses between two distinct breeds will often have a standard pheonotype, which can be more desirable than the original breeds, but the offspring from F1 crosses (and further) are a crapshoot.
Thanks for the informative replies.
Steve
If evolution kills off people with mutational loads >= 2 in one generation, it’ll probably keep killing people with that mutational load in subsequent generations. If fitness is related to mutational load, that just means that populations with more deleterious mutations die more.
You still need a mechanism to reduce it in some of your descendants, but other people have discussed that.
(Sexual selection is relative to populations, but that could go either way.)
Scott – the situation you just described, with mutational load creeping up, happens in systems without recombination and is called Muller’s Ratchet. With recombination, selection can act on segments of DNA instead of whole genomes, and this prevents the inevitable decay of the gene pool which would otherwise occur.
Gwern how does anime factor into all of this?
This does happen in species that don’t engage in any form of sex. It is called Muller’s ratchet. For example, it has been observed that all known species that do not reproduce sexually at all have very recent (on an evolutionary timescale) ancestors that did reproduce sexually. (This is including things like bacterial conjugation, which occur in species that also reproduce asexually.)
Except for bdelloid rotifers.
I interviewed geneticist Razib Khan who thinks we could get embryo selection for intelligence by 2020. https://soundcloud.com/user-519115521/interview-of-razib-khan
Your podcast is great by the way. You should make a post about it on /r/slatestarcodex.
PZMeyers probably thought Hsu’s program (which should be tried out for sure! so interesting!) would remind people of Hitler, so he’d better be against it.
Request fewer comments like this in the future; I think it’s destructive of good discourse to speculate too much on how people we disagree with have motives other than truth-seeking.
I’m not going to ban that kind of thing entirely, but the usual true, kind, necessary trio applies.
I read some comments on Myers’ post and was saddened that a lot of them were just insulting Hsu. Then I read the comments on Hsu’s reply and was saddened that a bunch of them were just insulting Myers.
Seems it’s very easy for comments to degenerate to just insulting the outgroup. Even though I would guess that in most ways Hsu’s and Myers’ commenters are quite similar!
I think the fact that one group regularly reads a blog devoted to insulting people and the other doesn’t is a pretty big difference between them.
I’m impressed at how little other narging there’s been on this post.
But also, I think this is one of the biggest impediments to human genetic modification, even just by selection for breeding — avoiding being hijacked by people who focussed on some sort of racial purity, and even if not, reassuring people that it won’t become that.
If so, it speaks well of Myers that he passed up the chance to call his hypothetical car the Master Racer.
Isn’t torsion dystonia dominant?
On investigation, yes. I could swear I read a study somewhere that was as I described it, but maybe it was another disease, or I misunderstood. Thanks for catching that.
Dominant with variable penetrance.
I know this is a technical term, but it sounds dirty as hell.
It means that you only need one allele to get the disease (dominant) but that not all people with the allele get the disease (variably penetrant).
Still sounds dirty.
Here is a fifth possibility — the nearly neutral model. Slightly deleterious mutations can remain fixed in the population, and slightly beneficial mutations can fail to fix in the population if the selective pressure for them is sufficiently small. The exact level of selective pressure is determined by the “effective population size”, which is the size of the population which would be expected to have the same level of genetic diversity as the actual population given ideal conditions. Due to genetic bottlenecks in the human past, and rapid population growth in the recent past the effective pop size for humans is quite low (in the tens of thousands). So, for humans the nearly neutral effect is strong.
See,
http://www.els.net/WileyCDA/ElsArticle/refId-a0001801.html
https://en.wikipedia.org/wiki/Effective_population_size
And I also have a sixth possibility. The effect on intelligence enhancing traits on chance of reproduction might be small or even nonexistent. Then you would expect no selection on these traits.
This might explain why high-IQ genes haven’t reached fixation, but I don’t think it would explain why they’re associated with so many other good things.
I think it does to a degree. If intelligence is only weakly selected for, but other attributes that could be also affected by intelligence boosting traits are very important, then you would expect that when an intelligence boosting trait occurs its effect on other attributes should be either positive or neutral. There are about three interactions you could expect here 1)”race car” traits — ie. intelligence boosted at the expense of other attributes, 2) intelligence only genes, 3) positive traits. If pushing intelligence correlated non-intelligence traits out of their normal range is strongly negative, you would expect there to be no race-car traits in the population. And thus when an intelligence correlated trait affects a non-intelligence trait, the effect will be positive.
@ Scott,
Imagine if intelligence was neutral on its own. It produced benefits, but also came at some costs that basically balanced them out as the race car hypothesis suggests. There are two basic ways to get intelligence into the population, drift (just the randomness of which neutral mutations make it) and co selection. Genes are not randomly passed down individually, but tend to be passed down in bunches (you don’t flip a coin for each individual gene, you flip a coin for a piece of the genome which has many genes). If you then have a gene that balances out the cost of intelligence that will let the intelligence confer advantages, and it becomes the combination of these genes that get passed down.
Now imagine that intelligence is actually a net negative on its own, but when matched up with several other traits it becomes positive. Then intelligence only gets passed down when it is in concert with these other genes (and so probably only on parts of the genome that get passed down together), so when you look many years later you only see intelligence correlated with these other positive traits because they were necessary just to get it off the ground.
Think of a Cheetah’s speed. Running really fast seems like a good advantage for catching prey- right until that prey changes direction! Cheetah evolution followed a trend of high speed + high agility because running really fast makes changing direction really hard, so the top speed only helps when selected with another attribute.
I think this is an important point. As I said above, I think we have intelligence because we live in packs and have dexterous hands, not the other way around. That is, there are probably many not obviously closely related traits and environmental factors that can change how useful more intelligence is–in some cases maybe even turn it into a negative.
Though unrelated to the racecar hypothesis (possibly related to the general fitness factor), I’m skeptical of Hsu’s program for different reasons.
The correlation with simple reaction time suggests that intelligence is likely to either be caused by or closely related to a generally physical property of the brain — “processing speed”, or something like that. This implies that it’s probably subject to basically physical limitations that would cause diminishing returns — the assumption of purely additive semantics of the different SNPs wouldn’t hold up, particularly toward the extremes.
To make an analogy, Hsu’s original article suggests IQs on the order of 1000 (meaningless number though this is). This implies around +60SD from modern averages. Modern averages in height, another very polygenic trait, are around 170cm, sigma ~7cm (for men). Would you expect a gene-editing program to be able to create a man six meters tall?
I agree 1000 is very unlikely, but we have firm proof that there can be people much smarter than average, and even having an average national IQ of 150 would cause a technological singularity really quickly.
It’s true that the ability to reliably engineer children with the high end of natural human intelligence, even only at the IQ-150 level, would itself be a huge step and is much more likely than ridiculous blow-the-top-off-the-meter levels. (I tend to anticipate that trying to do this will be harder than most people suggest; the assumption that any technology will immediately blow the top off its progression curve is rooted in a small number of examples which are very available, and it’s very plausible to me that we won’t manage this at all, or not for another hundred years. But prognostication is hard.)
That said, while the social effects of many babies suddenly being born with 150 IQ are likely to be very interesting, its effect on the technological singularity depends on a bunch of facts unknown. You could even formulate something like Lord Voldemort’s anti-singularity hypothesis, and suggest that the social problems arising from the sudden availability of such a technology would be likely to damage local conditions enough to keep much advancement from happening.
Can you name a single example of a quantitative trait that people tried to breed for and failed to move it 10 standard deviations?
I’m referring to technology, not breeding programs. The assumption is that any technology which we can do rudimentarily right now, we will inevitably be able to do much better very soon, and that the advances will continue to absurd degrees. This assumption is basically rooted in the great recent significance of electronics technology, and there isn’t a strong reason to believe it’s a universal law. Thus, we can’t necessarily assume that in twenty years, we will have gene-editing technology enormously better than we have today (in the same way that our computers are now enormously better than those of 1996, or perhaps more appropriately, computers of 2006 were enormously better than those of 1986).
If you tried an actual human breeding program for IQ, maybe you could move it 10 SDs that way. But that would take a long time, and probably have issues with ethics committees.
You are equivocating. Compare this comment.
I can imagine a relatively simple solution that could nudge the population towards higher IQs. Some dictatorial powers would be necessary, of course, but nothing too dramatic.
Make a law that only men with IQ 150 are allowed to donate sperm to sperm banks. (Frame it as a health issue, or even better as a women’s right.) Now you have created a mechanism that will gradually increase the population’s IQ, and doesn’t sound overly discriminatorily, so there is a chance the law would stay even after the dictator is gone.
How do you expect that to work, Viliam? Some people who want smart children will avail of the sperm bank (and associated fertility treatment, which is not cheap). This will select out the ambitious and better-off.
Meanwhile, those who can’t afford IVF (since if you’re going to go the route of having a genius baby, you want to make sure you conceive, and leaving it up to natural conception is too chancy) will be having babies the old-fashioned way, as will those who don’t want a genius baby, they want a child by their darling dear one which will have his cute nose.
So unless you make it enforceable and enforced law that all pregnancies have to be the result of donor sperm from your IQ 150 sperm banks, this will not gradually increase the general population’s IQ, it will create a sub-population of very smart children and a larger, inferior(?) population of IQ over a natural range.
And then conflict!
Yeah, the government would also have to subsidize IVF that uses the sperm of high-IQ donors.
Nice setting for a sci-fi story, but the IQ of the child would still be somewhere around the average of father and mother, so there would be a variance among the children created this way. Let’s suppose father’s IQ is 150, average mother’s IQ is 100, there is also some regression to mean, so the child’s IQ could be about 120. And there is nothing preventing those children to interbreed with the rest of the population (well, ignoring the possibility that some health problems that made IVF necessary could be heritable).
You’re comment has a lot of merit except for the conclusion which seems a little motivated. There’s already a “sub-population” of genius level individuals, they live among us. Are you saying all people with high IQ necessarily create conflict?
Oops *your
O Famous Hero of Note and Renown, what I mean is not that the population as currently constituted which mainly reproduces by natural means, and genius sperm banks have failed in the main, are riven by conflict due to scattered pockets of smarter people amongst the lumpen mass of the rest of us.
But if we set up a Guaranteed Genius Breeding Programme, and it’s confined to the better-off who can afford the costs of making sure they conceive a baby genius and raise it with the educational and environmental stimuli apparently considered most beneficial to making sure that the baby genius brain develops to its highest peak, and you get enough of these baby geniuses growing up to have baby genius offspring in their turn, then they not unlikely will form a clique, whether it be in the arts, sciences, politics, whatever.
Natural random shake of the genetic dice person going up for a job versus genius sperm bank person is going to lose out. Multiply that many times, when all the interesting and highest status employment goes to the genetically-gifted (because apparently we are all going to need to be computer programmers, inventors of better technology, entrepreneurs and founders of start-ups, or a combination of all to have any kind of employment in our Brave New Future), and you will get resentment.
Past history shows what happens when the masses decide they don’t like how elites are getting the cream. Look at Donald Trump – from joke to “Oh my God, he could very possibly be the Republican presidential candidate”. Despite all the name-calling about his support being racist homophobic bigots clinging to God and guns, he has undoubtedly tapped into a sense of dispossession and resentment among those considered, by the bien-pensant, as too stupid to know what is to their own advantage.
If people feel that they, or their children, are being disadvantaged unfairly, there need not be blood in the streets for there to be conflict: protests, demonstrations, political upheaval, protest votes for candidates who promise to stand up for the little guy. Where Viliam’s suggested “improve the population IQ gradually” scheme will run into the greatest risk is when it produces a sufficiently large sub-population of smarter people to be a perceptible threat to the ordinary natural range population, but not in enough numbers to be the majority or enough of a majority to control the society. I don’t know how many that would be; but if it’s not evenly spread out throughout all levels of society (“you too can have a genius baby even if you’re working three jobs, one of them as a burger-flipper!”), then it will tend to concentrate at a certain economic and class level and that will create resentment, perception of being set up to fail, fears that you and yours are being seen as fodder for the smart people to use as menial labour until they can build robots to replace you, etc.
Let’s say we want to bump the general level of IQ up to 130. At the moment, that’s around 2-3% of the population. What happens when Viliam’s or another programme gets that up to 5%? 10%? Not enough to be the majority, not enough to replace the lower IQ population as yet, but enough to start making their presence felt. Such individuals will still be rare but not as rare as today; indeed, they’ll be in the position of LGBT people and as claimed most people know someone who’s LGBT in their range of acquaintance.
It’s going to be a different attitude when “really smart Joe” is not some physics professor at an Ivy League college miles away from you and your concerns but is the guy interviewing for that job you – or your child – are also going for, and beating you out. And this happens in a range of situations where the 130 IQ people are now the ones making the decisions, running the businesses, appearing on talk shows with their vision of the future, employing you (or not employing you, as the case may be) and you find yourself developing an attitude about “those kind of people”.
“Meanwhile, those who can’t afford IVF (since if you’re going to go the route of having a genius baby, you want to make sure you conceive, and leaving it up to natural conception is too chancy) will be having babies the old-fashioned way”
I don’t follow that. Sperm is available in massively excess supply, so the market price for even high quality sperm should be low. Artificial insemination during a woman’s fertile period should give you about one chance in four of pregnancy. That’s surely a less expensive approach than IVF.
I don’t think there are enough babies born from sperm donors as such (i.e. unknown third parties) to move the needle. Also this group is already disproportionately intelligent (also tall and athletic). Maybe not 150 IQ intelligent, but probably around 120-130 average or higher.
Make a law that only men with IQ 150 are allowed to donate sperm to sperm banks. (Frame it as a health issue, or even better as a women’s right.)
Please don’t casually straw-man your opponents like this.
If I remember correctly, about ten percent of couples are unable to produce their own children. If we assume that many parents want smart kids, which seems plausible, the government doesn’t have to force the process, just not block it.
Sperm donation if the husband is sterile, egg donation if the wife is infertile, adoption—with a legal adoption market—or a host mother if the problem applies to both of them or if the wife can’t carry to term.
Using off the shelf technology and assuming half of the infertile couples give a high priority to intelligence, you could have five percent of the population come from the top tenth of a percent of the gene pool. That’s a pretty fast change.
Dosen’t that lend another portion to the Anthropic bias?
Namely, that there is a set intelligence level that civilizations can be at before they singularity themselves into spacedust?
See this picture on Hsu’s blog of chickens in 1957, 1978, and 2005. And this increase was done with technologies much inferior to what we currently have.
http://infoproc.blogspot.com/2014/10/big-chickens.html
I have trouble believing the “they were all raised in the same manner” caption, unless these are lab chickens versus farm/industrial chickens.
Regardless of your disbelief, that is what the researchers did: http://ps.oxfordjournals.org/content/93/12/2970.full
Was commenting less on the implausibility of artificial selection for chickens and more on the ubiquity of artificial selection on chicken feed.
With chickens we can reasonably hope that we can breed them to weigh twenty tons without hitting any unsurmountable obstacles inherent to their basic body construction, because we know that majestic ancestors used to roam the Earth with about that gravity. Though many necessary genes could’ve been lost of course, but at least in principle it’s doable because it has been done.
We have nothing like that for intelligence, and in fact its evolutionary history (with brain sizes stagnating and even decreasing in Homo Sapiens) is kinda concerning.
The current big problem with massive chicken breeding appears to be making sure they can still walk – at least well enough to reach the feeder. Pretty sure you’re right about upper bounds – I mean, we have ostriches and emus even today, which are damn huge flightless birds.
They’d need to get the tail back to scale to T-Rex size. Funny thing is people have put artificial tails on them to learn about dinosaur movement.
That may be the weirdest experiment I’ve heard of.
You mean the funnest. I knew I should have been a scientist.
I’m with Adam on this one.
“The correlation with simple reaction time suggests that intelligence is likely to either be caused by or closely related to a generally physical property of the brain — “processing speed”, or something like that.”
Chimps blow us away at reaction time.
Homonid brains (and presumably IQ) increased massively over the course of recent evolution. This was probably mostly not about just increasing general speed/efficiency of existing brains. I don’t know why you would expect that we’re close to some sort of natural upper bound right now given this history.
Whatever the ratio between chimp reaction time and human reaction time, among humans, IQ correlates with reaction time quite closely.
This is not about whether any entity can exist that is +60SD smarter than current humans (whatever that would actually mean). This is about whether you can create such an entity by stacking all the IQ+ human SNPs on top of each other. Intelligence when done human-style seems to involve literally speeding up the brain via reaction time; I haven’t bothered to look up the correlation coefficients, but presumably our hypothetical 1000 IQ human would thereby have some utterly implausible reaction time. The process of speeding up reaction time is certain to have some hard upper bound in the physics and chemistry of neurons, and when intelligence is done human-style, this should therefore correspond to an upper bound in intelligence.
This, of course, doesn’t touch on the possibility of wet-brain animals in whom the brain is wired differently so as to have higher correlation coefficients. It may well be that the process of evolving humans from proto-hominids involved increasing the correlation coefficient this way. But there’s no reason to assume you could get such an animal by stacking human SNPs on each other.
I’m pretty sure there are ways IQ could correlate with reaction time without IQ all boiling down to neuron-by-neuron signal processing time. Particularly since five minutes of Google puts the correlations in the .3 to .5 range.
According to Jensen’s “The g Factor”, if you combine reaction time measurements from several different tasks, you can get the correlations up to .7 or so.
There are plenty of examples of “super smart” people that don’t have exceptionally high reaction times–or even exceptionally high IQs. (I don’t have a citation, but a visit to a top university will prove this point quickly.) So I wouldn’t use reaction times as a gauge for what’s possible with general intelligence. Certainly as you go across species, reaction times aren’t a good measure for overall intelligence, which proves there must be other extremely important factors in intelligence.
It’s entirely possible that neuron signalling efficiency is the dominate source of variance in today’s human population. But as the population distribution changes (whether from natural selection, or a genetic engineering program), the dominate source of variance may change.
What is the average/peak chimp reaction time?
Chimps do not blow us away on reaction time, their reaction time is inferior to human reaction time IQ tests.
You may be confusing reaction time with a famous Japanese video of chimps doing well on quickly solving a memory test. These chimps are also outperformed by humans with a similar training regimen.
I intuitively expect wild animals to have faster reaction times than humans, and that fast reactions would be nullified by more thinking capacity; the same stimuli makes a cat strike out, whereas a big brained human stops to think it over a split second longer. We have more options for response to threats, so we have more to choose between.
Apparently, that’s wrong. Or at least between chimps and humans, or possibly mammals. (Insects, of course beat everything with flies having reaction times +10 times quicker than humans from the estimates I remember.)
Is reaction time just one thing, or might the different people and different species have different reaction times to different stimuli?
Maybe height is just plain different and has a lower ceiling? Really tall people are renowned for the stress their height puts on their body, but really smart people are not renowned for the non-social hindrances of being smart. Smart people don’t seem to be more susceptible to headaches in the same way that tall people are more susceptible to backache.
If we look at processing speed, computers have increased by billions of times since the 50s, but cars didn’t increase 0-60 times by the same magnitude in that time.
“Would you expect a gene-editing program to be able to create a man six meters tall?”
That might be a bit much. A 3 meter man is probably not impossible. All the tallest men seem to have problems with their spines and blood circulation, so you’d need to be able to supply them with genes for stronger spinal discs, and perhaps give them bigger hearts. Maybe humans don’t have any genes in any combination that can do that, but it doesn’t seem to me like the tallest humans are well optimized for being the size that they are and usually get there through overproduction of growth hormone.
Of course height specifically has different kinds of ceiling factors than intelligence specifically. But I don’t see any reason why we should assume intelligence (considered as processing speed) has no ceiling factors.
For another example, would you expect a gene-editing program to make a person who can bench-press a semi truck (on natural muscles and bones alone)? The ceiling factors in “muscle strength” are qualitatively different from those in either “height” or “neural processing speed”, but they’re still there.
I would expect anything with an IQ of 1000 (though again, it’s not defined what that would actually mean) to be doing something fundamentally different from a human brain, because it seems very likely that the processes a human brain uses just don’t scale up that far.
You wouldn’t expect a human to pump a semi-trailer because we have well-established physical scaling laws for muscle vs volume and living examples of creatures like elephants. On the other hand, the existing material I’ve read about suspected neural scaling for computing power like Hofman 2015’s http://www.gwern.net/docs/iq/2015-hofman.pdf “Evolution of the Human Brain: From Matter to Mind” or Herculano-Houzel’s stuff gives no indication that we are exactly at any absolute limit of primate brain architectures and there may yet be as much room above us as there is above chimps.
“Smart people don’t seem to be more susceptible to headaches in the same way that tall people are more susceptible to backache.”
Of course there are lots of mentally healthy smart people, as there are lots of healthy tall people, but I have the subjective impression that smart people are more likely to suffer various mental disorders like anxiety, depression, bipolar, OCD… don’t know if there’s a real statistical correlation, and it could be confounded by smart people seeking mental health diagnosis at a higher rate, but still that’s my subjective impression.
I think smart people are likely to talk and write more vividly about their mental problems, so there might be an availability bias.
Probably true, but I think there’s also a potential causative factor. I’m pretty sure part of what makes me smarter than average is precisely the same thing that makes me a little bipolar and OCD. I probably couldn’t get even more intense creativity bursts and obsessive attention to detail without a tradeoff in worsened bipolar and OCD; or, at least, it might be harder to achieve that without a tradeoff.
170? I understand this is a global and gender average but it seems a little short. There are plenty of countries with average heights for women well above this, which would imply we should work from around there in engineering taller people.
For all we know, there IS a way to make people 6’6 without any drawbacks. It just hasn’t been reached by evolution because all the pathways it could follow lead to crippling inbetween states. This is more in-line with the goal of 150 IQ babies.
170 is the American average, much higher than the global average. Young Dutch women average 170. I doubt that there is any country where the average woman is 170.
Where are these amazing lands of the Amazons where the average height of women is >170 cm?
Double reply to Doug and JK: Turns out the plenty part was an overestimate, but nonetheless there are several countries with female heights at or above 170cms. Per the following studies: http://www.mjssm.me/clanci/MJSSM_March_2015_Popovic_29-36.pdf (2014, measured), http://www.sciencedirect.com/science/article/pii/S1631069105001319 (2005, measured) and http://www.nature.com/pr/journal/v73/n3/full/pr2012189a.html (2009, self report); Bosnia-Herzegovina, the Dinaric Alps and Netherlands are all at or above that line. (measured studies give Dutch women at slightly beneath 170, sometimes by just 1mm) There are also plenty of near-170 countries – Norway, Germany, Denmark, Montenegro and Belgium would all round up to 170 as the nearest 5, and Italy, Austria, Lithuania, Slovenia, Finland, the Czech Republic, Israel, Greece, Spain, Sweden, Samoa, Serbia, Croatia, Slovakia, Azerbaijan, Tonga and Poland round up to 170 as the nearest 10 by one or more studies. This is all just from the wikipedia article on human height.
This isn’t even what I wanted to talk about – I made the classic mistake of putting something more interesting in an offhand opening than the body of my post.
Nope. I can read wikipedia, too. But can you read my comment?
If you could read mine without being a snot first, I’d be happy to help you.
Douglas’s point, made in his original comment, is that the average female height is that much in some countries only in the younger (and numerically smaller) cohorts. Older women are shorter. The average height may exceed 170 cm in those countries some day, but this is not that day.
… but this is not that day
You have to have known that someone was going to go look up the heights of Miranda Otto, Liv Tyler, and Cate Blanchett, right?
They average 172 cm. Now you know where to find the land of statuesque Amazonian goddesses.
That point has been addressed JK. There are more important things to talk about.
I got a flashback to RotK when I finished reading your post, so you’re good in my books.
Is Ashkenazim IQ really 12-15 points higher? This seems like an extreme estimate. Anyone have sources?
Their IQ is pretty high. I found this paper by Gregory Cochran that has some interesting information on Ashkenazi IQ:
It seems to me there is a large difference between 12-15 and “half a std” (7.5 right? I have seen 16 used as SD some places)? 15 is twice the effect size of 7.5 and even 12 is 160% of the effect size of 7.5. A dispute of this size seems problematic to me.
Anyway thank you for the source!
It’s a pretty big difference, yeah. I don’t know which estimate is better. It’d be nice to get some newer studies with large Ns, but I don’t know of any.
You’re welcome!
Ashkenazim perform differently on different subtests, so the size of the gap would depend on what subtests are aggregated and with what weights.
Typo:
I assume that first “causes sickle cell” should be “provides disease resistance” or similar.
You beat me by a minute!
The sickle cell gene protects against malaria.
Thanks, corrected.
You go on to quote Cochrane’s argument later about Jews but don’t apply it here.
In related groups, IQ correlates with less mutational load as well as better genes for intelligence but when a specific breeding population has been selected under conditions where intelligence has a much higher payoff and so can incur more costs, you get decreases in fitness, height, attractiveness, etc – in addition to blatantly obvious genetic disorders that are deleterious when homozygotic but less harmful when heterozygotic.
Interesting! So your theory is that intelligence can still be pushing up against an upperGroup boundary even if the individuals in that group displaying the highest intelligence aren’t affected? That the smartest Ashkenazim might be on average more attractive and taller and “more desirable” in most aspects but he or his family must be paying for it in other ways? If this is true, couldn’t we examine the roots of the higher intelligence and cut the chafe? An in exact analogy would be finding another effective mutation besides sickle heterozygosity that stops malaria and just using that instead.
Sort of – the problem is that intelligence in Ashkenazim is associated with deleterious genes. The example Cochran used that I remember was Tay Sachs (which you’ll note is a neurological disorder); heterozygous, you’re fine and likely smarter; homozygous, you usually die by age 4.
If you could control for those and make groups that have the same genes that are trading off [attractiveness, height, neuroticism, etc] for intelligence then I’d guess that the higher IQ group would overall be taller, more attractive, etc. Within the group, the mutational load hypothesis should hold. Going back to the race car analogy – make groups of each of major modifications (air bag removal, no power windows, etc.). Within those groups, the faster car is the one built closer to spec.
“But all of these things are strongly positively associated with intelligence, and in many the link has been proven genetic!”
I’m only part way through the comments, so someone may have already made this point, but if genetic IQ correlates positively with genetic traits that give increased reproductive success, the obvious interpretation is that IQ, at least in the environment we evolved in, gave decreased reproductive success. The logic is the same as Scott already sketched.
I don’t know if it is true, but it isn’t absurd. There is an obvious conflict of interest between me and my genes, and higher IQ may make me better at promoting my interest at the expense of theirs. That’s particularly plausible for females, since they bear the biological cost of producing offspring.
With any luck, the trade-off is that we need to consume large amount of calories and are only fertile at a very specific time of the year.
This is already true.
I think that the perceived negative correlation between intelligence and popularity is mostly a quirk of late 20th-century pop culture, and is greatly exaggerated. First, popular kids aren’t dumb, no matter how much we nerds like to comfort ourselves with that fact. My memory of high school suggests that the most popular kids in school were also high achievers with good grades, probably with IQ of 115+. The people with low IQ were not exceptionally popular. And the unpopular, nerdy kids… well, I remember plenty of unpopular nerds who were, in fact, of very average intelligence. “Nerd” attributes (poor dress, social awkwardness, bad grooming, focus on fringe topics) caused unpopularity, but they didn’t cause intelligence, and intelligence doesn’t cause nerdiness. It’s merely the prominence of this stereotype that makes us think of the socially awkward nerd as an exemplar of high intelligence and assume that intelligence should anti-correlate with social success.
It does take a certain amount of verbal wit to keep the esteem of the masses. Or a really good physique. Maybe intelligence evolved as a secondary sexual selection strategy?
Yeah, I was struck by this idea throughout. I’m not as up on the literature as I could be, but it’s not a stretch to imagine that what we call “intelligence” (poorly-defined at the best of times, really) is really social adaptability, albeit a particular form thereof. If a person is skilled at learning the appropriate dress, posture, speaking patterns and other factors that influence popularity – which are skills to be learned like any other – then it’s not a stretch to consider that they’re more intelligent in at least some way. And more popular people may be afforded more advantages in some way – teachers might like them more by the same process that makes them more likable – leading to better educational outcomes, giving them an advantage in other areas like getting into good universities or whatever.
So, yeah. I’d posit that popular people are, in some important way, actually very intelligent, in the same way that successful business people are, in some important way, very intelligent. Maybe this is g at play, maybe it’s a flaw in how we perceive/measure intelligence, I’m not sure (although I’m leaning towards the latter). Hell, cultural factors on how intelligence is perceived are huge, so it seems silly to disregard them to enforce some split.
“popular kids aren’t dumb, no matter how much we nerds like to comfort ourselves with that fact.”
I think it’s partly driven by nerds wanting to compensate for their unpopularity so as to shield their ego, but the idea of the “dumb bully” also gets driven by children’s media, as a sort of low rent version of equality where yes, we’re all different, but each of us has traits the other doesn’t, so it all evens out.
Of course, trade-offs only have to occur at limits, so there’s no intrinsic physical reason that someone can’t be better than you at everything, it’s just that it hurts a lot to know that Chad is not only stronger than you, better looking than you, and more popular, but you can’t even outwit him either.
There’s also the idea that when it comes to bullies (not that the popular kids are always bullies like its a movie), they are compensating for a lack of self-confidence in themselves. Teachers love to push this idea because it fits the same compensatory trade off concept, where yes, he’s bullying you, but that’s just because he’s a loser.
Last year, a study came out showing that bullies had the highest self-esteem, social status, and lowest rates of depression.
“Bullies” isn’t synonymous with “popular people”, so you can have mostly smart popular people and mostly dumb bullies at the same time.
When I was in school, the bullies were mostly lower-class kids of middling popularity and modest academic achievement. The popular kids tended to be academically successful as well, although they weren’t the most academically successful.
I went to a large high school and didn’t know enough people to have a good idea of the intelligence of various cliques; but there were four National Merit finalists in my year, and of those, one was a nerd, one was a goth, and two were artsy kids.
I also suspect (but don’t know) that the idea “nerds are unpopular” is tied to our specific historical moment. Does anyone know if this perception existed in other times?
I’m not sure, but a good place to check would be the beginning of the Industrial Revolution. Were the luddites anti-nerd bullies? There was probably a lot of talk about machine men with machine minds around that time.
Recently nerds have gained some semi-popularity through the celebration (and bastardization) of nerd culture, with people self-identifying purely for liking things associated with geeks in the past. This may have something to do with all the comic book movies that have finally made something obscure accessible to the masses, as well as the rise and rise of gaming culture as it becomes more mainstream.
Of course, I think it’s going to sour again and is in the process of souring. The idea that everyone will be replaced by machines in the workplace gives programmers and software engineers tremendous power, and since they are tied to the nerd identity in the public’s mind, nerds will receive a backlash in the coming decades.
Were comic book readers ever smarter than average?
More specifically to question of human intelligence, what about the hunter gatherers?
I do not know of any literature on the topic, but I have worked with two African hunter-gatherer tribes (including administering some cognitive tasks). In both cases, my sense was that intelligence is an excellent positive predictor of social status.
Does Odysseus count?
I can’t say about other times, but will report that Chinese international students I’ve happened to mention it to (n=8 or so) are reliably baffled by the idea that smart/studious kids are unpopular here. They usually have a strong assumption in the other direction.
I believe there is a positive correlation between high intelligence and nerdiness. (Adding the obvious that “positive correlation” does not mean “it’s the same thing”.) It’s just a bit more complicated, and yes, there are people of average intelligence with nerdy hobbies, and people with high intelligence without nerdy hobbies.
I suppose the people with average intelligence having nerdy hobbies is what happened after the nerdy hobbies became publicly associated with higher intelligence. It shouldn’t be used to reason retroactively that the nerdy hobbies were actually never associated with it.
As an analogy, imagine that all the cool kids at your school would start wearing red shirts on the same day… because for some reason they decided that red shirts are cool, and because all the local cool kids know each other, so it was easy to synchronize their decision. So on Monday, the red shirts are strongly correlated with being cool. The next say, many not-so-cool kids notice what is happening, and some of them start wearing red shirts too, just to appear cool. On Wednesday, both the cool kids, and the kids who want to be perceived as cool and have good observing skills wear red shirts. Meanwhile, someone writes a blog with the observation “wearing a red shirt will make you belong in the cool crowd”, and everyone reads it. On Friday, even the greatest losers with zero observation powers wear red shirts, because they have read on the blog that this is the cool thing to do.
In this model, wearing a red shirt on Friday doesn’t prove much… but we shouldn’t use that as a proof that even the kids who wore red shirts on Monday were actually losers.
Somehow similarly, I believe there is something that draws many intelligent people towards nerdy hobbies. And when it becomes known, there is also the obvious reason why some less intelligent people mimic them in order to be also perceived as intelligent. (This is similar to the known “fake geek” phenomenon.) But this doesn’t mean that everyone with nerdy hobbies is only doing it to appear highly intelligent, because it all had to start somewhere. (Where would the idea that “highly intelligent people have nerdy hobbies” come from, if all nerds would be merely stupid posers?)
Higher IQ adds skills, but reduces the compatibility with the average population. A person with IQ seems impressive, but a person with IQ seems weird. (From your point of view, both of them have “magical” skills, but only the former uses them for meaningful purposes; that is, meaningful to you.) If the average IQ of population is 100, people with IQ 120 become leaders, and people with IQ 150 become weirdoes. Some people with very high IQ may actually have phenotypically lower social skills simply because they didn’t get enough experience interacting with their peers during their formative years. — That does not mean that everyone with lower social skills has IQ 150! It simply means that there is more than one causal path to lower social skills: having low IQ is one, being isolated from one’s peers is another, having a specific disorder such as autism is yet another.
I don’t think less intelligent people with nerdy hobbies are only doing it because of status.
Many nerdy hobbies– especially the fiction– include fun of the ordinary sort.
Also, a lot of people find television and movies more accessible than books, and these days, a lot of nerdy books books are available as television and movies.
We’ve seen lots of originally-nerdy interests go mainstream in just the last few decades. I mean, video games made billions from mainstream audiences with fantasy and sci-fi concepts that would have previously been thought of as niche interests of weirdo nerds. Comic books basically conquered the globe, with Hollywood as an intermediary.
I think “nerdy hobbies” are mostly things with a core of genuine mass interest, but pursued in an unusually motivated or effortful fashion. Typically someone finds a way to keep a fair chunk of the core enjoyment while dispatching with most of the required effort (especially mental effort), and it crosses over.
I remember the times when even thinking about internet made one a weird nerd.
(“So, you can use a computer… to connect to another computer… in a different country… so you can type messages to people you don’t know and will never meet. Huh? Why would anyone want to do that, when you can simply go out and talk with your friends? Oh, I guess it’s for the ones who don’t have real friends. Well, I think you should just work on your social skills instead of trying to solve the problem using technology.”)
The concept of “pen pals” has been around for a while. Also, all activities cannot be neatly divided into the two categories “what all normal folks do all the time” and “what only the despised low-status outcasts do”.
I think it depends where you are. I moved all over growing up; in the fancy private school I ended up at, the very popular did well (but still copied from “nerd” homework occasionally); however, in the urban public schools, popularity was strongly +correlated with athleticism and conspicuous failure in or disregard for grades.
There’s also that very long-term study on high-IQ children — can’t remember the name but it’s pretty well-known — where the conclusion was that very high IQ is positively correlated with social difficulties and medium-high IQ (~120 IIRC) with interpersonal success.
They also gave examples from history to support this, in the essay wherein I encountered it, but that’s kind of a toss-up IMO — for every Sidis, there’s probably a Bohr.
Is there any research about what children get bullied for? And not just for contemporary sub-cultures, but also a historical overview.
I sometimes felt that one of the reasons I was bullied was because I was smart. I mean, the main things were that I was poor and had been homeschooled, but there was also this feeling that knowing the answers in class made me a “show-off,” reading big books made me “weird,” using big words made other people feel uncomfortable and dumb. So the bullying was about keeping me in my place when I was deviating too far from the average. People don’t really like that much, for whatever reason.
Jane Goodall observed of chimpanzees that rank in their status hierarchy appeared to correlate very little with size, strength, intelligence, age, or any other property apart from “drive to rise in the status hierarchy”. The top chimpanzees were distinguished primarily by being willing to work harder, take more and bigger risks, and generally persist in their pursuit of the goal of rising in the hierarchy, until they reached top status. (She also noticed that the material benefits of high status were actually pretty meager–occasionally a bit more food or sex, but not much more.) I think the analogy to human popularity carries over quite well…
The study is behind a paywall, so I can’t check myself, but did they do anything to test the difference between “Smart people commit less crimes” and “smart people get caught committing less crimes”?
(I’d suspect that both are true to varying degrees, just curious.)
I was able to access the paper through my university login, and it doesn’t look like they addressed that issue.
Here’s how they described their methods:
Another thing in that quote that jumped out at me: in some social environments certain criminal activities are part of being well adjusted to that environment.
As they say… the smart criminals don’t break laws; they make them.
Fiction really muddles people’s idea of crime, by portraying it as something like conscious cost/benefit decisions to commit crimes. Characters commit crimes because Reasons, which reflects the audience’s expectation more than it reflects reality.
In reality, people commit most types of crime because of factors like:
– Provocative or difficult life situation
– Bad emotional control
– Hyperbolic discounting
– Poor understanding of how to accomplish even basic immediate goals
– Failure to comprehend or consider consequences
All of those should negatively correlate with intelligence.
Source: worked briefly in public defender’s office, talk to people who still do.
Useful data, but I’d also like to see data from the subset of criminals that don’t use the public defender’s office. It wouldn’t surprise me to see a correlation with intelligence or rational planning there as well.
I was going to say what John said, that your sample seems likely to be skewed toward the left end of the criminal bell curve rather than randomly drawn, but aside from that, you’re also only drawing from the pool of criminals that got caught.
Though I still suspect you’re largely correct. But that’s just a matter of intuition, which is even less reliable. It just seems obvious to me in my own calculus regarding what to do with my life that there are so many ways out there for an intelligent person to attain a good life without having to bear the risks of criminal behavior at all, that crime doesn’t seem worth it. There’s a bit of ‘privilege’ in that line of thinking, though. If I’d grown up in a different environment, say late 90s DRC or something, the calculus could easily change and whatever intellect I have may have been more fruitfully applied to militia-building.
Yeah. The US studies indicate that over the past 50 years the biggest reason people commit or don’t commit crime is lead. And lead => low intelligence
You’re probably right. As I said before, I expect both to be true, but something can be true and also not be demonstrated by some particular study. (And I would absolutely expect that even when high IQ people commit crimes for bad reasons they’ll be more likely to be able to deal with the resultant fallout in a way that keeps them out of trouble.)
This is always the first thing that comes to mind whenever anyone brings up the apparent negative correlation between criminality and IQ. It seems that the data could be equally well explained by either “smart people are just plain less likely to become criminals” or “dumb criminals hold up liquor stores; smart criminals become embezzlers, black hat computer hackers, and email scammers; really smart criminals become Wall Street executives and steal money in entirely legal ways.”
Private military contractors, shady lawyers, international smugglers, etc…
This is a wonderful plan with surely no negative long-term repercussions. =P
Wouldn’t the reduction ad absurdium of the mutational load theory be that paramecium (or whatever very far back early ancestor) were hyper-intelligent? Is it all explained by the somewhat recent lack of selection pressure, so now we at the far side of an equilibrium balanced between intelligence that keeps us alive and breeding despite the mutational load slowly eroding the intelligence genes?
(edit: addressed in above comments, of course)
Nitpick: I expect that a car optimized for speed would be very pretty– we tend to like the looks of things that can move fast. I agree it would be deficient for human use.
Notion from Michael Vassar, which I hope I’m getting right. There are two routes to intelligence. One is excellent overall health, and the other is single exaggerated characteristic.
Torsion dystonia isn’t very common, even among Jews, or at least I don’t know anyone who has it.
My best guess is that your extended footnote is correct — evolution doesn’t value things the way we do.
To take just one example, why aren’t all men hunks with high muscle mass and low body fat? That’s well within the power of natural selection. In fact, we’ve lost muscularity recently (since archaic homo sapiens and Neanderthals), even more relative to more primitive homonids, and chimpanzees are much stronger than we are.
But muscle is expensive, and fat is useful for storing energy. If you could reduce muscle mass and get away with it, that would be the thing to do in the ancestral environment. Now, in our food-rich modern environment, the calculus has changed.
I imagine that the most likely explanation for why all the +IQ alleles haven’t become fixed across all human populations is just that they all have (relatively small) costs in terms of energy usage or whatever, and their benefit in the ancestral environment was relatively small, so they were pushed up to the point where the average person had enough of these alleles to reach the point where marginal benefit equaled marginal cost, and stopped there.
But now we live in a society where the benefits (at least as we value things — maybe not in reproductive fitness!) of IQ are much greater than in the ancestral environment, and plausible costs much lower.
(As an aside, if each allele has a very small effect, it could take a *very* long time to achieve fixation. I mean, what’s the fitness advantage of a zero-drawback +0.3 IQ point allele in the ancestral environment? 1% more kids per generation? Probably more like 0.1% or 0.01%. But then we’re talking on the order of 100s of thousands of years to achieve fixation in the population. But given a polymorphic trait like this, a better question is do we see steady evidence of selection on the set of +IQ alleles *as a whole*.)
That’s a good point. As is my habit, let’s put a number on it. I think the equation for number of generation to fixation is T=2ln(N)/s where N is the species population and s is the genetic advantage. Homo sapiens has existed for around 150,000 years which makes 10,000 generations. Human population has usually been a few million since the agricultural revolution, let’s say 1 million before that. That means that for a mutation to reach fixation it must have had:
s > 2ln(1,000,000)/10,000 = 0.3%. That’s 3 extra surviving kids out of 1,000. It seems pretty high for a ~1 IQ point mutation, so it makes sense that most of these are very far from fixation.
Almost certainly the two major selective pressures against high intelligence are or have been:
1) Maternal death in childbirth. Intelligence is strongly correlated with head size well down into infancy. Anyone who’s seen both a farm animal and a woman give birth can appreciate that bit in Genesis about “in sorrow shalt thou bring forth children”. Big human heads make childbirth hard.
2) Increased calorie requirements, particularly in infancy. Brains are expensive. The brain burns about 20% of an adult’s calories, and as high as 50% of an infant’s calories. Brain size and caloric requirements correlate with intelligence, and high intelligence would be selected against in environments prone to sporadic famine. Smart kids would be more likely to die if their mothers’ milk production goes down.
You’ll also note that neither of these pressures are really a problem in modern societies: we know how to do C-sections and we’ve got more calories around than we know what to do with.
One notes that fire used as a technology was probably a major factor in allowing intelligence. Basically, cooking food predigests it. This enables people to get more calories from it, and get them while expending fewer calories in the digestive process.
Once we hit the threshold of using fire to cook, we had a world of opportunities.
I have a vague hunch that head size is related to maternal hip width, and that wider pelvises might contribute to slower running speed and lower running efficiency.
How important was running speed/efficiency? Probably pretty important in hunter-gatherer environments, and moderately important in violent settled environments (e.g., when running away from a losing battle, it pays to run fast).
On the other hand, I don’t have much evidence for this.
Olympic running champions, for example, tend to have narrow hips.
Humans are awfully slow, and I imagine running speed is primarily useful in agricultural battles, and not much help if a lion starts chasing you. On the other hand, humans were endurance hunters for a long time and we have some of the best endurance running amongst mammals.
On the other hand, I’m not so sure that brain size *within* the human population is a particularly *strong* correlation (and the evidence is a bit shakey?), once you control for actual disorders that physically inhibit brain growth. But I haven’t read a lot on that and stand to be corrected.
Regarding the calorie consumption of brains, I don’t know much about it but I suspect that it’s a bigger problem than it would seem from the proportion of calories it normally uses, because a starving body can’t get rid of its calorie-burning brain in the same way that it gets rid of its calorie-burning muscles.
I suspect that the brain is a huge factor in the minimal short term survival calorie requirement. So small changes in its caloric requirements are probably a bigger deal than it seems.
Also, the brain can’t easily run on fat; it requires glucose which the body can make out of protein if necessary, but this is a problem since most of our emergency energy storage is in the form of fat. This makes the problem of caloric requirements of the brain during famines even greater.
The brain can also run on ketone bodies made from fat, which is what starts happening if you eat no carbs (or no anything) for three days. For this reason, humans can actually fast much longer than most animals.
“the most heavily-loaded people are weeded out by natural selection”,
That means 50% child mortality that was common through the ages even among wealthy classes. About 150 years ago this mechanism came to a halt among western populations, so our genetic load must be increasing and we should see more health and mental problems.
We still have high rates of natural abortion.
Weeding out tends to get me heavily loaded.
I think this doesn’t prove anything. If you ask someone “would you rather have an IQ 150 baby or an IQ 100 baby” they are considering the hypothetical holding everything other than IQ fixed. People aren’t considering the trade-offs worth it, they’re not considering the tradeoffs at all, because they don’t know what they are! The right way to consider this question is to come up with some candidate horrible tradeoffs and play “would you rather” with prospective parents. I think plenty of parents would choose an IQ 100 baby over, say, an IQ 150 baby that grows up to be a miserable depressive and offs itself by age 30.
The right way is to consider the the tradeoff as it is in reality, i.e., look at the lives (or probability distributions over lives) of actual people with 150 IQ vs 100 IQ and ask which one you want for your child. (Which people may already be doing if you ask them that question, to the extent that they have accurate ideas of the lives of 150 IQ and 100 IQ people.)
Scott Aaronson probably has an iq around 150. I’m not sure I would want to be him.
Tenured at UT Austin working in a field he loves with a wife and two children? From the outside that seems fairly good. Of course, there are other things you could be with that degree of intelligence: a lawyer or a doctor are obvious choices, but certainly there are many other routes in life, and it’s hard to think of one where intelligence isn’t helpful.
In the long run, life worked out well for him but look how miserable he was before he got there.
I would guess that most of those factors that led to misery were environmental.
Hi there! I unfortunately can’t answer the question of whether someone else would want to be me. But I can tell you that, from my perspective, life was indeed miserable for a little over a decade, but is fairly pleasant now, all things considered. (And perhaps knowing firsthand how miserable life can be made me appreciate non-misery all the more, while also giving me more empathy, if still not as much empathy as a “normal” person has.) But Dana and I only have one kid so far, not two.
If there were a lot more IQ130+ people, people like Aaronson would be a lot happier. It would also be much more feasible to have gifted & talented programs and schools and early college admissions and other accommodations that seem to help out prodigies.
Some general observations I’ve had are that (a) people who are not permanently satisfied tend to be more creative (for obvious reasons: if everything is already perfect, you don’t need to change anything and can get an office job or work in banking). People who are more creative genuinely produce more noticeable things, and we tend to associate them with being more intelligent, whereas socially adjusted high-iq patent clerks (or unsocialized weirdos, https://www.youtube.com/watch?v=ggur-Ca2nzs) will not tend to be identified as intelligent. (b) There are trade-offs between social fitness and other abilities. For example the PI of this study http://www.pnas.org/content/97/8/4398.full.pdf was quoted in the Scientific American:
“Maguire thinks that The Knowledge may enlarge the hippocampus’s posterior (rear) at the expense of its anterior (front), creating a trade-off of cognitive talents—that is, taxi drivers master some forms of memory but become worse at others. ”
This gives a partial explanation to both the “nerd effect” and the “tortured genius” perception that’s separate from mutational load.
Yeah, that’s a sentiment I’d echo if I was given the choice, speaking as a major depressive from the age of about 14, now in my mid 40s, high intelligence does not always equate to success in life. I used to have suicidal ideation multiple times a day back in my teens and twenties. One thing that helped was talking down a friend from their own suicidal thoughts. I won’t have children because of the mutational genetic load my dear father left me. I couldn’t kill myself while my partner is alive, but I am fairly certain I’ll be dead in the next 10 years anyway (males on my father’s side generally die in their late forties to early fifties from complications related to aforementioned genetic mutations).
A kid who’s emaciated is going to be — not only dumb — but also weak, ugly, unhealthy, and maybe antisocial. In that vein, a $1,000,000 Ferrari (despite its speed) is probably going to have a sturdier chassis than a $100 pre-owned Jeep. The inherent engineering trade-offs don’t matter as much given the disparity between the resources invested into the Ferrari vs the Jeep.
Hypothesis: Is it possible that Mother Nature indeed imposes a trade-off between IQ and {everything else}, but scientists don’t notice this because the effects of genes are outshone by things like the effects of nutrition? Or has this been aleady accounted for by Hsu and friends.
I kinda agree but I think that it becomes clearer if we say that isn’t how much we feed a person, it’s how much food a person is “designed” to require.
Assuming CRISPR gets better (a safe bet IMO), the most important short-term use of this may be in cellular therapies. E.g.,
If you’re already injecting stem cells into someone, why not inject “genetically spellchecked” stem cells?
If you’re already culturing someone a new liver for implantation, why not make it a “spellchecked” liver?
Obviously this wouldn’t produce a result as profound as growing a “spellchecked” person from scratch, but the messy CRISPR stuff could all be done in vitro, which would make it about a million times easier and safer (and you’d be much less of a target for the FDA). It could also help people who are already alive- like us.
I figure in 10 years it’ll be a no-brainer to add a ‘spellcheck’ step to any cellular therapy. The problem will be creating a framework that can deal with both low-impact common mutations (like what Hsu talks about) and high-impact super rare / de novo mutations (like what the MacArthur Lab is working on) and suggest how to best spend whatever “edit budget” we have in the future. More thoughts here: https://goo.gl/8ulE1K
In the car example. We know that what directly improves speed is the engine. Other factors reduce the speed originally generated by the engine (e.g. weight, drag, tire slip, fuel impurity, etc). Every car component has weight. So in a sense, every car component affects speed. But it would be misleading to claim that, say, the radiator generates speed since its weight varies depending on the model.
Though 50% of the human genome is related to IQ, I doubt that more than 1% are directly responsible for IQ. I find it more likely that the other 49% is responsible for things like metabolism. E.g. if I have malformed blood cells, the lack of oxygenation will surely affect my cognitive ability. But if Hsu’s genetic-engineering project consists of improving IQ, would it not be more effective for him to focus his research on the 1% of genes that are directly responsible for IQ, rather than the trillion little knobs & levers that affect metabolism?
If he wanted to understand the brain and improve it, yes, it would probably be better to focus, but if he wants to obtain a superhuman brain as fast and easily as possible, no.
I don’t see how that makes a difference. :/
We can’t turn a Jeep into a Ferrari by adding better tires. So why do we expect to obtain a superhuman brain by increasing metabolism? Are you saying that Hsu is willing to take a Faustian Bargain to obtain a superhuman brain? (e.g. crank metabolism to such a ridiculous level that the rest of the body is nothing more than an enormous small intestine)?
“Kaneda!”
“Tetsuo!”
“Janet!”
“Dr Scott?”
“Rocky!”
I think the race car factor only kicks in above a certain “IQ” level. Sure intelligence is helpful, but there’s a peak in the curve beyond which the benefits of being more intelligent and adaptable are relatively insignificant and outweighed by the drawbacks of, say, being an asocial neurotic wreck…or is that just me?
(“IQ” in scare quotes because I have doubts that intelligence can be measured or reduced to a scalar, and I think IQ tests mostly measure how good you are at taking IQ tests.)
IQ is correlated with a ridiculous number of things that an intelligence test should predict. It may not be perfect, but it’s measuring something important that reflects the real world. With the knowledge we have now, denying the importance of an IQ test is pretty irrational.
Oh I’m sure there’s correlation, but intelligence is multidimensional and if rationalism means treating a deeply flawed measurement of a single projection of a multidimensional vector into an arbitrary scalar space as the sole arbiter of one’s value as a human being, then I prefer to be irrational.
No one’s saying your value as a human being is determined by your score on an IQ test. Wrong Species is pointing out that it’s empirically false that “IQ tests mostly measure how good you are at taking IQ tests.” I think it’s surprising that the single scalar produced by an IQ test is a strong predictor of outcomes, but it seems to be true.
IQ is correlated with a lot of things, some of which are negative.
I wonder how are intelligence and personality related, if at all. Openness to experience aside…
Would IQ have any effect in how someone evaluates, for example, the possibility of making all predators vegetarians via genetic engineering, at the “gut level”?
I’m not opposed to people changing themselves in whatever ways they see fit, I hope the ones that want to be superintelligent grey goo find success, but I would tread carefully. Even if you manage to establish some sort of global “basic income” all you get is a happy zoo in place of what used to be the real world.
What does prediction involves, ethically, in a deterministic world? How free are you if some superintelligence can predict your entire life and even thoughts? Even if it does nothing, how free are you exactly? Is it all right if it wants to have fun watching the simulated death of your future grand-grand-grandson but pretends not to be able to predict that well, so you might keep some sanity?
So we get to the point where every intelligent being has to become more intelligent or be just another specimen in the zoo. But then some people believe animals obviously have a much better time in nice zoos than they do in the wild (And you are a deluded, romanticizing dirty hippy for suggesting otherwise!) meanwhile others feel this is terrible ignorance and hubris; Those animals would be better euthanized in the name of dignity (And you are a scary, inhuman p-zombie if you fail to see this!)
I hope intelligence has no significant effects on personality, aesthetics and all that. Would not be comfortable optimizing the world for intelligence otherwise, even if it somehow made people more likely to feel and judge like I do.
If we were talking about AI I could understand your concerns but it seems pretty ridiculous to believe that smarter people won’t care about aesthetics.
I don’t mean they won’t care about aesthetics, they most certainly will. I wonder if some aesthetic persuasions might be overrepresented among them on account of related genes and how IQ might or might not play a part itself when it comes to all that. Both sides of the example I gave are “aesthetic appreciation”.
It is worrying because those 300 IQ people are probably going to rule the world one way or another, even if they do so in conflict with each other or by indirect means. An overrepresented personality trait or whatever among them would have global effects.
I admit my original comment is pretty exaggerated, I’m not suggesting smart people are going to be able to predict the death of your unborn grandson… I think the principles still stand though. And for all we know, it could be that a 300 IQ quickly makes a 500 IQ person and so on, and now the difference between a superintelligent AI and very smart people is not so clear anymore.
IIRC, in Lewis Terman’s classic study where high IQ individuals were identified in childhood and followed up through life one of the findings was that personality-wise the study participants were almost like a random sample from the population. That is, strong selection on IQ did not select for personality at all. This is consistent with meta-analytic results showing that the IQ-Big Five correlations are very low, except for Openness (which is the same as self-reported intelligence in many ways). This would suggest that eugenic selection for IQ would have hardly any effect on personality variation.
As far as personality goes, there’s also the issue that Big Five is proving a very tough nut to crack. Much harder than height or intelligence. We can right now do meaningful selection of inches or IQ points, but next to nothing for Big Five (except maybe Neuroticism, which seems to be overlapping with mental diseases).
I wonder how useful the Big Five model really is. Its not hard to imagine two people with very similar Big Five scores having different and maybe even opposing ethical values and aesthetic sensibilities.
I don’t buy those being 100% the result of our conscious Free Will or our cultures, genes related to traits people exclude from “personality” could play a part, maybe intelligence itself too.
Would the proportion of hippie luddites to grey goo transhumanists remain constant if you raised everyone’s IQ by 50 points? If it remains constant, would they be more radical in their views, or more compromising?
The worst possible situation would be one where IQ does make us more likely to pick any of those sides and also very optimally makes us immune to understanding the other side’s points, at the gut level at least.
I bet that intelligence also correlates with depression, high functioning autism and other mental problems, so I don’t think it’s all positive. It’s interesting that intelligence is correlated with being physically healthy but also (possibly)mentally unhealthy.
Correlations between IQ and mental disorders are negative or zero. There is no evidence that higher IQ is associated with any negative health outcomes.
Autism might be associated with a lower mean IQ but a higher variance of IQ so autistics might be over-represented among those with extremely high IQs.
https://pumpkinperson.com/2014/10/05/iq-and-autism/
Why do you put the Bohr brothers’ political views and activity in with their list of obviously genetically-influenced positive traits?
probably to show that they’ve not been left psychopaths by whatever things affected their IQ and health.
Why would it? Why would genetic traits that led to a certain outcome in the past necessarily lead to the same outcome now?
For example, modern culture affords ways to use one’s intelligence towards other ends than increased social status. Maybe nerds are folks who aren’t in the habit of using their intelligence this way. Paul Graham has an essay exploring this point. (I don’t necessarily agree with it). Maybe the modern cultural idea of the antisocial nerd is a self-perpetuating meme that encourages people who identify as “smart” and “different” to segregate themselves socially.
Suppose the sense of taste evolved to make sure we ate foods that gave us a variety of nutrients. Then should we conclude that people with a great sensitivity to taste must be the strongest and healthiest among us? How is that different than the argument above? (I do actually feel like I’m missing something, and would like to know).
On a different tack, I wonder why the stereotype of “smart –> low social status” is usually applied to kids. We don’t usually say this about smart adults, right?
Kids are forced to be around other kids who bully them. Adults are not.
Also, if you harass an adult in a way comparable to a kid bullying a kid, you’ll probably end up with a restraining order; when an adult appeals to the authorities, it has a good chance of actually working.
“Also, if you harass an adult in a way comparable to a kid bullying a kid, you’ll probably end up with a restraining order; when an adult appeals to the authorities, it has a good chance of actually working.”
From what I’ve read about domestic violence, getting help from the authorities isn’t guaranteed, though I’ll grant that the odds are better than for bullying between children.
Also, workplace bullying is fairly common, very debilitating to the victims, and since it involves the victim’s income and generally isn’t violent, generally isn’t taken to the authorities.
“Why would it? Why would genetic traits that led to a certain outcome in the past necessarily lead to the same outcome now?”
Yes, that’s the part that makes it “naïve”
I understand Dilbert as hinting towards that.
Dilbert started in 1989, long before the first dot com boom, let alone the rise of the current tech giants. Being in software was MUCH lower status then. I’m not sure of its status now, having watched its rise from the inside I’m too close to it.
It seems a bit dangerous to start depending on c-sections as a matter of course. Wouldn’t that risk turning survivable disasters (a nuclear war, for example, since hospitals tend to be located in cities) into extinction-level events?
While you’re developing larger skulls, why not develop larger pelvises to go with them? We don’t need to run as fast as the early women did.
Large pelvises won’t have that magical 0.7 hip to waist ratio men are attracted to. So we’ll have to tweak that too.
They will if you also increase the waist size. Which automatically happens if women just become taller and proportionally larger, which is pretty much what is already happening, and may be well one of the correlation between height and IQ.
I don’t think 0.7 was ever shown to be more attractive than even lower WHRs; it’s more like a reasonable benchmark for ‘has a great figure’ that doesn’t dip in to numbers rarely seen in humans. A few of the early Playboy centerfolds supposedly had WHRs as low as 0.55 or so.
Eliezer Yudkowsky used to talk about “Algernons” – people with high intelligence but low average fitness.
You’re right that people a little bit more intelligent than average don’t seem to exhibit this loss of fitness. But Hsu’s idea takes us to the far end of the bell curve.
My intuition: massive and aggressive tinkering would at best produce someone like Kim Peek – a barely functional human being with a photographic memory of phone books.
edit: will elaborate later
Memorizing phone books does not require a lot of intelligence; it’s inherently pretty easy. My intuition is that people who are really good at hard-for-people-easy-for-computers tasks are not the end process of cranking intelligence to the limit, but rather have an altered mental structure that makes them architecturally more computer-like. My intuition is that this is a totally separate phenomenon.
Yeah, my apologies, looking over this again I misunderstood what Hsu is proposing.
The born children. You don’t know how many were unviable and miscarried early, perhaps without Bohr’s wife realising she was pregnant.
I’m not convinced that the race car model “crashes and burns”. Suppose that, analogous to humans, we were unable to do direct experiments on cars- all we can do is look at the data that is naturally generated. You’d still have a whole mess of data spanning over 100 years. And you’d see a strong correlation between speed and lots of other nice variables like safety, fuel efficiency, and comfort, because cars have generally gotten better over time. There’s a more general lurking factor of “technology” there- the more technologically advanced cars just do better at everything. (of course there are many counterexamples, like early specialized racing cars, I’m just talking about the general trend, which is all you can measure from the rough correlation in the general population.)
Likewise, human IQ is strongly correlated with other nice outcomes because of “health”. But so what? We all want to improve health, but that’s not an easy thing to improve- mostly you just have have to avoid obvious sources of damage. For a person who is already as healthy as we know how to make them, it seems unlikely that we can improve IQ without sacrificing something else.
I like this. It’s basically a continuous version of Simpson’s paradox rather than a binary version.
And even for cars of the same era, you’d see a positive correlation: a high-end BMW is faster, nicer, safer, more comfortable, etc. than the typical Toyota Corolla. In order to see the tradeoffs, you’d need to look at very specialized vehicles, or you need to look at the price tags.
In the case of humans, the “price tag” may be something which is not easy to observe, e.g. because survivor bias. Maybe embryos with Bohr-like genes are more likely to die in early pregnancy, in a way that causes a small but significant fitness loss which offsets the fitness gains of Bohr-like intelligence and physical prowess.
Another way that the car model is misleading that I haven’t seen anyone point out so far is that genes are selected for being compatible with other genes in the population, but car parts aren’t. If you pick all the “best” car parts on the basis of speed, they could end up just being incompatible, but if genes worked like that, then sexual reproduction often wouldn’t work. Thus, genes are sort of forced to operate in a mostly independent way, which makes it more plausible that different IQ-related genes could have separate, additive effects.
It’s not misleading at all. We are talking about additive variants here. Additive variants are found and selected for their average effect across the different genetic backgrounds of literally hundreds of thousands of people. If there were a variant which had some sort of interaction or dominance or epistasis such that it worked only in conjunction with another specific variant and never otherwise, it would basically not appear in the GWAS or GCTA results. The evolutionary argument is helpful for establishing that we shouldn’t be surprised to find tons of additive variants, but it is not necessary since the results are safe by construction.
“If there were a variant which had some sort of interaction or dominance or epistasis such that it worked only in conjunction with another specific variant and never otherwise, it would basically not appear in the GWAS or GCTA results.”
This is wrong.
“Additive variants are found and selected for their average effect across the different genetic backgrounds of literally hundreds of thousands of people.”
Those hundred thousand people don’t have independently varying genetic backgrounds. To the extent that they are genetically similar, they represent the same genetic background.
The variants effects are well captured by additive models, and additive models accurately predict a lot of things, such as short term response to selection. This does not imply that the variants themselves work by inherently additive mechanisms, nor that everything will be accurately predicted by additive effects.
Quibble:
Radation around nuclear power plants is very low. Quoth US Nuclear Regulatory Commission:
It doesn’t change the conclusion, but I’d rather not reinforce the belief that nuclear power plants are dangerous for the citizens living close to them.
Yup. I’d go as far as saying that people who work in nuclear plants probably receive less radiation than people who work outdoors.
One of my favorite notes from XKCD what if was that
“You may actually receive a lower dose of radiation treading water in a spent fuel pool than walking around on the street.”
Assuming you stay near the surface and don’t swim down the the fuel rods.
Can confirm. I’m a Rad Worker living on the east coast, occupational exposure < 100 mrem/year. I get *less* exposure in my current position than I did living in Denver (high altitude + background rock = 120 mrem/year).
I took that part to mean that she was standing next to the actual reactor, not just the power plant.
Here’s a silly hypothesis: what if, like the harmful homozygote examples, having too many genetically intelligent people is detrimental to reproduction in a society as whole? Maybe it leads to a lower birth rate, or a higher suicide rate, or (silly) susceptibility to invasion.
Maybe it leads to a lower birth rate
Given all the talk (both approving and not) about “Smarter people have fewer kids, it’s the idiots who are breeding”, this is a real risk.
Evolution is pretty bad at getting rid of genes that help the bearer but hurt society. At least, we’d need an explanation for how it accomplishes this beyond just “group selection”
This made me think about homosexuality, which obviously hurts the individuals’ reproductive fitness, but may help his/her relatives or tribe somehow. Related, I heard recently that being gay correlates to high IQ as well. If true, it would seem to support the planners=/=breeders theory within a tribe, or something. Maybe smart people are more useful to your tribe as a whole if they aren’t dealing with a bunch of kids.
I remember reading somewhere that some of the most capable birds in some species assume the most dangerous positions when flocking, like the vanguard or the spots at the ends, many of those birds are homosexual and don’t get kids to worry about…
I can’t remember where I read it.
Supposedly the female relatives of gay men have more children, which has led some to the conclusion that there’s a gene that does both. I’ve seen – can’t remember where – the proposal that the link serves a purpose: tending to have fewer children of their own* they are more able to devote resources to helping care for the children of female relatives, especially if the fathers get eaten by bears or whatever.
*even in societies where it is hidden, gay men would probably have fewer children, due to less interest in sex with women, leading to fewer accidental pregnancies, leading to fewer total pregnancies, I would think.
In general I agree, and think myers is both wrong, and a total asshole (WHAT IF WE MAKE BETTER PEOPLE AND THEY TURN OUT TO BE NERDY LOSERS LIKE YOU HUH STEVEN TSU??), but something you should be aware of is epistasis. Epistasis is interaction between allele effects. Human genetics people are fond of saying that it’s negligible because additive models work so well in practice. However studies in animals, where we can manipulate the genetic background, tell us that epistasis very common – it’s just that additive effects work well to capture itbecause allele frequencies are low . This means that the effects that genes have in one genetic background can be different from ones they have in another (hence, partially, terrible reproducibility in GWAS).
So the upshot of all this is that we can probably make ourselves a Nils Bohr, but once we get up past that to very high IQs, we will have altered the background enough that we might well start seeing the kind of weird genetic covariances that Myers was afraid of – this happens in cows, a lot of their traits like milk production have no reached the point where pushing the trait further inevitably pushes other things we don’t want further as well. This would be less of a problem to the extent that the genetic load hypothesis is true, but I don’t know of strong evidence telling us how important genetic load really is.
I always thought that “ancient, dying out race” space opera trope is rather implausible. Now there is argument for a policy that will likely lead to exactly this outcome. C-sections for everyone, really?
Ya, I feel vaguely uncomfortable with measures which, while fine in our current society, would mean large portions of the population being unable to breed without the help of modern technology.
But the population of earth is high enough that without modern agriculture almost all of us would die. Having just, say, 1% of the population being able to reproduce post-apocalypse might be a good thing if the apocalypse reduced the carrying capacity of the earth by 99%.
Problem is if the short term fitness of the 99% is higher.
Say there’s a disaster, short term, who’s most likely to survive the first year of cannibalism and blood rain: the 1% of natural humans or the extra-intelligent extra-strong Übermensch?
Even if the 1% have a long term advantage they have a short term disadvantage to the other 99%, less likely to win in conflicts for food and resources etc. Remember that the 99% aren’t going to lay down and die of their own free will even if they know they can’t reproduce themselves.
I expect that some of the 1% will be able to recruit and arm members of the 99%,
At that point, the only remaining question is, will that work better as a series of novels, a movie, or a game?
I expect that some of the 1% will be able to recruit and arm members of the 99%
Recruit them how, and arm them with what?
Notwithstanding the common usage established by OWS et al, in this case the 1% are likely at the bottom of the pre-apocalypse socioeconomic ladder, a Gattica-esque genetic underclass.
Excellent point I hadn’t considered.
Some of the 1% presumably have stashes of food and weapons.
Some of the 1% have probably been in situations where it was useful to be able to work with/lead the 99%. It doesn’t seem likely to me that the 1% only work for each other.
Also, there could a multi-level hierarchy. I bet there’d be plenty of people in the 130+ range who are working for the 1% (the work is more interesting and higher status) and wrangling the 115s (if the business includes that many levels) who are wrangling the 95s.
The wisdom of fictional evidence tells us that however small minority there is, some members will always survive. If anyone tells you otherwise, regardless of what evidence they used, they will be proven wrong at the beginning of the sequel.
Agreed. That’s why I think humanity should be working on some kind of contingency plan to keep civilization going in the event of some near-cataclysmic fate. I had in mind some kind of training program for many people scattered across the globe(in order to maximize the number of locations able to kick-start an industrial revolution) and global stockpiles of the most important technology to get everything started.
In 1800, before fertilizers, pesticides and so on, world population was about one billion. 99% is way too high.
There was a lot of technology in 1800, most of it knowledge in peoples brains. We no longer have these pre-industrial skills and without modern technology we would likely go back to the stone age.
There are enough weirdos around that I doubt Precursor technology is as lost as all that.
” Having just, say, 1% of the population being able to reproduce post-apocalypse might be a good thing if the apocalypse reduced the carrying capacity of the earth by 99%.”
If the 1% produce children who are all able to reproduce that can work. But if the 1% are mostly people with the genes for producing big headed babies but who happen to have unusually wide hips for reasons only in part heritable, or people who are at the low tail of produced baby head size but whose children will revert towards the norm, you have a problem. In the first generation the population falls to one percent of its original value–but it keeps falling thereafter, because only a fraction of that one percent can themselves breed.
C-sections for everyone, really?
The “planet of large-brained people bred for superior mental powers but now a dying race” was a Star Trek episode 🙂
Don’t forget whatever meddling would be required to make people not find our big heads grotesque and actually mate with each other, unless people are fine sacrificing that too…
People are remarkably good at accepting willing mates given some time. Most people like tall partners yet dwarf couples are pretty common. Ditto most other physical traits.
Well, if we’re going to mess with genetics anyway, we could make additional changes so that twins and triplets are a lot more common and the gender ratio is skewed in favor of females. C-sections have been done for a long time; they were just almost always fatal before modern medicine. But even if the apocalypse occurs and each woman can only breed once, it won’t be so bad if she always has twins.
Because there are no known complications associated with twin pregnancies at all?
After all, science has proven parents are irrelevant after birth, so no harm done there.
I’m not sure that hunter-gatherers had little use for intelligence. Niven said, “How much intelligence does it take to sneak up on a leaf?” — I’d say, “Which mushroom?”
Before agriculture made the world safe for stupidity, you had to know a lot of fine distinctions about which plants to gather and how to process them — and teach all this to your children. As leader of a small group, you had to know how to look at clouds on the mountain and calculate whether it will freeze up there, and, remembering the summer’s weather up there, what plants are now likely to be tasty foraging up there. And what animals also find them tasty and are likely to forage upon you if you go up there.
Just surviving day by day took a lot of intelligence, applied on a lot of subjects, under a lot of pressures.
Also, intelligence varies a lot among herbivores– elephants are very smart.
Elephants are also highly social, so the question is “which comes first” is socialization/cooperation the trait that makes intelligence worth selecting for or is intelligence the trait that promotes socialization.
Niven said, “How much intelligence does it take to sneak up on a leaf?”
Birds can eat yew berries without harm, humans can’t (if they swallow the seeds). The foliage is toxic to cattle and horses, and the pollen is very bad for humans. If you go by the rule of thumb “see if an animal eats it and lives, then it’s safe for you to eat” but aren’t smart enough to see which animals live or die, you’re not going to live very long.
Niven said, “How much intelligence does it take to sneak up on a leaf?”
No he didn’t – one of his characters said, “How much intelligence does it take to sneak up on a leaf?”
Another character reflected that the answer depends on the plant the leaf is attached to.
Yeah, the Kzinti were not exactly that universe’s exemplars of applied intelligence.
The race that managed to completely negate the militaristic Kzinti threat without ever having to expose their own cowardly bodies to danger, oh, wait, those were the dumb leaf-eaters.
Yeah, this brought to mind a passage from Pinker’s “How the Mind Works”, which might be a bit dated now, but is still awesome:
Intelligence is also very helpful in dealing with human’s number one threat: other humans.
I sort of feel both arguments. It would be an amazing coincidence if the evolutionary selection pressure to find a local optimum in breeding likelihood exactly matched what was best for human happiness or for the survival of a society as a whole. So there’s almost certainly untapped potential in making smarter/healthier humans.
But OTOH, it’s the sort of thing that’s not completely understood and is very complicated, I’d be amazed if there’s not SOME unfortunate downsides hiding there.
I don’t understand mutational load as such, but I assume biologists understand it.
Delish
Repeated Caesarean sections are not problem-free:
I suppose if you’re only going to have one or at most two children, this is a small risk (remember, this is major abdominal surgery, not a quick-fix alternative to the traditional way). But I imagine if we get to the point of engineering genius babies (with optional enlarged heads), we’ll be culturing them in artificial wombs rather than exposing them to the risks of the uncontrolled environment of the natural womb.
I remain dubious about the whole notion because I don’t think we have a useful working definition of what intelligence is; why is it when we talk about “high IQ”, the usual example given is that of a physicist? What effect would it have on the world if it were composed entirely of physicists, even genius ones? Every single person optimised to be a physicist – yes, probably some of them would turn to being inventors of robots to do the menial and vital tasks necessary such as growing food, building houses, and running the economy, but the fact remains: if we measure IQ on “can solve maths problems”, does this mean that a population of mathematicians is necessarily going to be the most capable at making robots to grow and cook the food to keep them alive or do the other tasks? Or sitting down to design an economy? Or treating psychiatric illnesses?
We laugh at Plato’s notion of government by philosophers, but is government by physicists likely to be more successful?
As far as I know, many c-sections are caused by doctors insisting that women lay on their backs during childbirth — because that’s more convenient for the doctor, but less convenient for the newborn who must now move uphill over the tailbone, so the intervention is required more often.
I suppose that allowing women to give childbirth in different positions would be a perfectly acceptable cost for increasing the average IQ of a population to 150.
That’s fine when we’re talking about 150 IQ. My siblings are all pretty smart and also were in the 90th percentile for head size as babies (seriously, we all looked like little aliens) and my mom gave birth to us all naturally — no drugs, no tearing. But what about when you go even bigger than that? What about when the baby’s head size is half again as large? Squatting isn’t going to help you with that, and even an unusually large pelvis is probably not going to be big enough.
Yeah. Well, at this point my whole plan is “when the population will have 150 IQ on average, someone is going to find a solution to this problem”. 😉
One possibility is babies being born at an earlier developmental stage.
It’s a possibility, though human babies are already born very “premature” compared to most other animals for precisely this reason. This is why the human baby is much more helpless than, for example, a baby giraffe. All we need now is a pouch of some kind like a kangaroo… no more need for baby slings and the like.
Not to disagree with your point, Onyomi, but it would be better to compare to a kitten than to a giraffe. Predators are generally born more helpless (eg, blind and unable to walk) than prey.
Regarding development at birth for various species – be careful to consider litter size, and don’t assume too much based on generalities.
Litter bearers tend to have smaller young because they are carrying many more at a time, and the more young at one time, the less developed they are at birth. (With cod being the ur-example of quantity over quality.)
Also – european rabbits are born more blind and helpless than puppies. However, jackrabbits are a precocious species, as are guinea pigs.
It could be the other way around and a failure of representing smart people. Lower IQ people may find it easier to point at people capable of doing high level physics and maths as representative of having IQ 150, but miss where other IQ 150 people work. Could also be that physics and maths are the only domains challenging enough to draw in IQ 150 people. Are you making the point that people < IQ 150 can tolerate the more menial tasks of looking after the mundane things of average living?
That’s a bit of a strawman. You’re jumping from “Physicists are an example of someone who’s very intelligent” to “Optimizing for intelligence means optimizing for being good at physics and nothing else.” There’s no reason to believe the “and nothing else” bit.
IQ is correlated with a lot of things besides “good at physics.” Income, crime rate, etc. Unless physics professors are so well-paid that they skew the average for everyone else, I would suspect that high IQ has other applications than physics.
Also, AFAIK, a modern IQ test has a lot more than math problems on it. Have a gander at the list of stuff on the WAIS test. Vocabulary, logic, visual problems, spatial reasoning, memory, processing speed…
Heck, even the SAT tests more than “can solve math problems.”
Although vocabulary is clearly a major component of intelligence, I don’t see how it’s possible to test in a culture-neutral manner. The SAT used to be biased towards the kind of people who’ve heard of regattas, etc. And how do you even begin to compare these things across language barriers?
Daily reminder that “the black-white difference is generally wider on items that appear to be culturally neutral than on items that appear to be culturally loaded.”
(I recall seeing that the black/white gap on the regatta question was in fact smaller than the average gap on SAT questions in the year that the regatta question was used, but I can’t find it online.).
Most white people haven’t heard of regattas, either. I see that question as mainly about filtering people who went to Andover or Dalton from the hoi polloi. Race is part of it, but only part.
The regatta question also finds the compulsion readers. I’m not upper class WASP, but I probably knew what a regatta was by the time I took the SAT.
Yeah, what Nancy said. I went to a public school and never knew anybody that sailed. Nobody in my family older than me even went to college. We weren’t “cultured.” I still knew what a regatta was and I got a perfect SAT score because I spent much of my childhood reading like a maniac.
The highest I ever placed in a spelling bee was 2nd for LA region, but look at the kids that actually win. The winning words last year were ‘scherenschnitte’ and ‘nunatak.’ You really think anybody knows those words because they’re white or they’re rich? Well, first, obviously not, because the winners were both Indian, not white. Second, nobody knows those words unless they’re the kind of person who obsessively studies words, which is exactly the kind of person colleges are looking for.
It also filters the Police fans.
I’ve heard of regattas, and if “the SAT used to be biased towards the kind of people who’ve heard of regattas”, then that means it was biased towards a lot of short, dumpy, bad-tempered bog-Irish who read a lot 🙂
I imagine the “regatta” question was likely strongly affected by location at least as much as class. I hadn’t heard of a “regatta” until I moved to the Philadelphia area, where regattas on the Schuylkill River are big events. You don’t have to be of the prep school class to have heard of them.
I never said it was a perfect filter. And just because a bias can be overcome it doesn’t mean the bias was never there to begin with.
(And yes, I also read voraciously and knew what regatta meant, but that’s not the point!)
Is there any evidence that regatta questions or the equivalent–culturally loaded questions–ever made up a significant fraction of the SAT? My suspicion is not, that this is a case where a guess got converted into a fact because a lot of people wanted to believe it.
Aegeus, I was saying that every time someone uses an example of “high IQ individual”, they go to a physicist or a mathematician.
I’ve never yet seen anyone use “most successful basketball player” or the like. Is the assumption that athletes are stupid? Some of them apparently go to college and even if they’re there on scholarships for sporting prowess so the college team can win, they must manage some kind of degree at the end.
Intelligence does indeed link up with a lot of other attributes, but as I said, every example I’ve seen quoted is “hard science, especially mathematically-linked” and not something else. So if you’re talking about “let’s breed IQ 200 babies”, I don’t imagine most people expect the genius baby to grow up and be the world’s greatest composer.
If memory serves, the average IQ of NFL players is somewhere in the decent-but-not-exceptional range, and I expect other sports are similar (adjusted for the needs of sport and position). So they’re not dumb, and there are plenty of examples of very smart people who were very athletically successful, but success in sports isn’t coupled tightly enough to IQ to be a good proxy.
(The notion of “kinesthetic intelligence” that was popular in the ’90s, and which some hippies still like, seems to be false. Or, if not totally false, then it doesn’t buy players more than a good coach would.)
I’ve never yet seen anyone use “most successful basketball player” or the like. Is the assumption that athletes are stupid?
The most successful athlete I know is a former pro football player who finished his career as a full professor of engineering and radiology, so, no.
But success in STEM fields can be achieved on the basis of pure intelligence with a bit of willpower. Success in professional athletics requires intelligence, the right physique, more willpower, and teamwork. So, while the best athletes are likely to be fairly smart, there won’t be as much of a concentration of smartness as there is in fields where you just have to sit at a desk and exercise smartness.
I doubt there is any field in which the best performers don’t have substantially above-average intelligence. But the more requirements you put on the job beyond “be smart”, the weaker this effect will be.
I imagine the adjustment by sport and position is quite significant. I’d put money on the smartest athletes being those in individual sports where there’s direct competition against another athlete, and thus a major need for short-term tactical adjustment: tennis, combat sports, etc.
Completely anecdotal, but the people I know from BJJ who are most successful in competition are all very sharp.
At a slight tangent, my old explanation for the Flynn effect was modern obstetrics. Someone who knew more about the question than I did assured me that while my explanation gave the right sign it wasn’t adequate to explain the size of the effect.
I cannot see how attractiveness itself can affect average IQ in a society with equal gender roles. For example, I may be attracted to someone with high IQ and repulsed by someone with low IQ, but the same reasoning applies to the persons I’m attracted to – I may be rather disgusting option for someone with much higher IQ than me. So most probably I’ll end up with someone of the same IQ, and overall effect of IQ vs attractiveness will be zero.
These arguments may fail when attractiveness depends on IQ in some nonlinear way, or if there is some gender inequality. Anyway, the most straightforward way to evaluate effect of IQ on attractiveness is to discard proxies such as “masculinity”, “beauty” etc., and create instead a cross-population 2D distribution of IQs of parents, which would say what are the chances a male with IQm and female with IQf would mate to give a viable offspring in a given society.
The next obvious thing to do is to trash “height”, “fitness”, “health”, “longevity” and “social well-adjustedness” and go directly to a number of children and their IQs (of course all IQs should be age-normalized). To summarize, the basic model should incorporate distribution of IQ vs attractiveness for both genders, distribution of IQ vs number of kids, IQ of parents vs IQ of kids and age-normalization function. Barring inflow/outflow of the population, I think the model should describe IQ dynamics rather well. Maybe it would even be possible to evaluate the effect of artificially created geniuses on the general IQ of the population.
Does anyone have references to any of the data above? The model looks rather easy to implement and fun to play with.
I was thinking about this: not only in recent times, but even throughout most of history, there doesn’t seem to be a trend of more attractive and/or more intelligent people having more children. In fact, now it seems like the opposite. Of course, there are cases where the emperor gets a bunch of wives or Genghis Khan fathers a million children, but I’m not sure it’s IQ which selects for being an emperor or khan–probably more like ruthlessness or something.
Of course, children of the wealthy and beautiful tend to have more resources, but for most of history, it wasn’t like the king’s children got vaccines while poor children did not. In many cases it was probably just king’s children dress in silk and take baths, while poor children run around in burlap sacks, getting beneficial exposure to microbes that reduced their chance of dying of smallpox.
In a way, it may not be surprising that this happens: the threshold for surviving and reproducing in human society is lower than the threshold for being really successful, and probably has been for a long time. Ugly people marry other ugly people and stupid people marry other stupid people. And then they have roughly the same number of children as smart people. Their children’s lives aren’t as nice, but they also probably survive and reproduce at a rate not much lower than that of the rich, beautiful people (in fact, there is a general stereotype that the poor are very “hardy,” while the interbred aristocracy are sickly). There is a certain level of ugliness or stupidity where you are probably unlikely to find a mate, but it’s probably quite low.
And this may also explain the existence of caste and class systems throughout world civilizations. It’s not that the untouchables don’t, sometimes, have really smart, children, but they may be descended from a particular stratum which was at one point, whether fairly or unfairly (perhaps the children of a conquered people, say), were considered less fit; but being an untouchable doesn’t prevent you having lots of healthy children with another untouchable. Their lives are just less likely to be pleasant.
This also seems to point to how evolution seems to function less strongly, or, at least, differently, once you reach a high level of abundance. The kings of medieval Europe, to say nothing of most of us living in the developed world today, had way more than they strictly needed to survive and reproduce. The fact that genes of non-kings survived too, then, should not be surprising.
Of course, there are cases where the emperor gets a bunch of wives or Genghis Khan fathers a million children, but I’m not sure it’s IQ which selects for being an emperor or khan–probably more like ruthlessness or something.
I’d bet good money that Genghis Khan was a very smart, charismatic, physicaly and mentaly healthy guy. Likewise, I doubt many fools survived Byzantine politics.
Also, Genghis Khan was able to create an empire that survived his death. Which helped many of his children survive, and some of them even live in prosperity.
Having many children is less useful when the next king kills them all.
Yeah, I’m sure he was at least reasonably smart and obviously must have been really charismatic. It’s commonly said of unusually successful men like TR that they are super high energy, sleep little, etc. as well (though part of that maybe after-the-fact myth-making), and it seems like that sort of trait is probably also important.
Related is the issue of disturbingly hot young Stalin. The takeaway is that all the people who gain huge political power despite seeming, in retrospect, to be unconscionably bad or stupid were, at some point, at least, really, really charming and calculating.
But I do think the nerds often comfort themselves by saying that the jocks must be all brawn and no brain. Pretty much any president of the US is smarter than average, if not way smarter than average. Most of the people who criticized Bush Jr as dumb and who are now criticizing Trump as dumb are, of course, not as smart as them. What they are really saying is “they don’t subscribe to my tribe’s aesthetics.”
“Pretty much any president of the US is smarter than average, if not way smarter than average.”
My father’s opinion of Nixon, who he had interacted with, was that he had a very high IQ.
Wealthier people had more children surviving in to adulthood at least as recently as 19th century Germany (and it wasn’t a small difference, if you looked at landowners versus day laborers for example). It was not until about a hundred years ago, even in the most developed countries, that food was no longer a serious limiting factor in terms of how many children you could raise. Even in societies where people wouldn’t starve outright, because of the Church or other social institutions, the deprivation and malnutrition of the poor left them a lot more vulnerable to various diseases. Indeed you can also show that, counter to your idea of the poor children gaining some kind of beneficial exposure to microbes, wealthy people were somewhat less likely to die of the Black Plague back when that was a thing.
Measuring IQ vs number of children and then calibrating IQ(time) function to predict average IQ in 5-10 years looks to me like a rather simple and straightforward experiment. I would be suprised if no one have done anything like it.
They may have to marry people with serious hereditary diseases (because the healthier ones were already taken by the competition), which would reduce the number of offsprings that reach adulthood and reproduce.
As a very simplified model, imagine that there are three components of attractivity — pretty face, smart brain, healthy body — each one of them on scale from 0 to 10, and the total attractivity is the sum of these three numbers. Free choice for both genders would mean that having a pretty face or a smart brain increases the expected health of your children (because you mated with someone who achieved the same total attractivity by compensating less pretty-face points or less smart-brain points by more healthy-body points). So even if everyone has the same number of children, people with prettier faces or smarter brains will have more grandchildren, because their children will be more likely to reach the reproductive age.
(Note: This argument is based on an argument from The Mating Mind, which is one of the reasons I recommend this book to everyone who wants to discuss human sexual behavior and human evolution rationally.)
Jack Cohen, a reproductive biologist and author has argued that part of what could have driven up human IQ is puberty rituals. Standard format is that you have a tribe, before children (sometimes just the male children) can officially become adults and do all the adult things like get married within the tribe etc they have to go through some rite or ritual. Often there is significant pain involved.
Examples include:
Wearing a glove of ants for several minutes.
https://en.wikipedia.org/wiki/Maw%C3%A9_people
Scarification:
http://www.news.com.au/travel/world-travel/crocodile-scarification-is-an-ancient-initiation-practised-by-the-chambri-tribe-of-papua-new-guinea/news-story/79973a3cfa6733ad225d94dc01db124a
Whip initiation:
https://prezi.com/rf0h8hvaqf9d/fulani-whip-initiation/
You’d assume that between 2 groups the one which didn’t risk it’s members health through ceremonies that involve significant risk of infections would win yet those groups still exist.
It may be that such ceremonies acted as a crude form of eugenics, selecting for tribe members best able to make choices based on deferred rewards(terrifying pain now in exchange for the perks of adulthood later) and choosing deferred rewards is fairly strongly correlated with things like SAT scores and educational attainment which are pretty strongly linked to IQ.
Although evolution pushes the sex ratio to 1:1, the optimal sex ratio for a tribe is to have more women than men.
Why?
Optimal in what respect?
There’s a lot of work being done there by “optimal” and some by “tribe”. But the basic idea is that one man can impregnate many women — wombs are a scarce resource. For a given population size the higher proportion of fertile woman to men (assuming at least one) the more babies you can have going at a time.
Of course that ignores the whole surviving to reproductive age thing and also at least flirts with the is/out fallacy.
It’s not just so much the “surviving to reproductive age” thing as the assumption that the number of wombs is a bottleneck.
Which tribe has greater reproductive fitness, tribe A, with 50 men, 150 women, or tribe B, with 150 men, 50 women? Tribe A, because they have more wombs available? Or tribe B, because they can despoil tribe A?
I mean, you have to be a cool f—ing customer to claim that evolution takes us away from reproductive fitness.
A short-term advantage can turn into a long-term disadvantage; an advantage for an individual can be a disadvantage for a group or even for the whole species.
Therefore this argument is not valid in general.
“I mean, you have to be a cool f—ing customer to claim that evolution takes us away from reproductive fitness.”
Sex ratio is an odd case. It isn’t that someone who produces equal numbers of male and female offspring will have more reproductive success than someone who doesn’t, or that a group that does will.
It’s that in a society which produces more female offspring, producing more male offspring leads to more reproductive success, in a society that produces more male offspring, producing more female leads to more reproductive success, giving an equilibrium of roughly equal numbers.
An explanation which I believe was first offered by R.A. Fisher.
In domestic and wild ruminants, the data indicates that more dominant better fed females tend towards male offspring, while lower ranking females in poorer condition produce more females. This is theorized to happen because a fast-growing male may grow up to be the sire of many offspring, but even a smaller/weaker than average female will produce *some* offspring.
Sometimes I wonder if our modern Western medical problems are due to the fact that our bodies expect to be put through traumatic trials such as these, and to be infested with parasites, and when the maximum stress we experience turns out to be having to ask a girl to prom, the body/brain doesn’t know how to react/calibrate.
To be fair: even in brutal societies people can get deeply upset about social situations. Winning a mate is about the most important thing possible as far as your genes are concerned. Far more important than floods, fires and famines.
There was an infamous news story of a prisoner dubbed the “Booty Warrior” who described that in prison having some booty was more important than food or drinking water.
Good examples.
But I’m pretty sure that much of the benefit from painful initiation rituals is psychological. They’re not evolutionarily better off by suffering pain & risking infection & maybe killing a few men. They are because they did so as an initiation ritual to prove their loyalty to a group.
selecting for tribe members best able to make choices based on deferred rewards(terrifying pain now in exchange for the perks of adulthood later)
This assumes that members of the tribe actually had the choice. I’m inclined to guess they didn’t. Does anyone know different?
I tend to regard these kinds of things as attempts by the older men to restrict competition from the younger ones. Over time some of them seem to have gotten out of hand.
Actually, attractiveness is uncorrelated with IQ. Most studies that find links have attractiveness judged by raters who are aware of other characteristics of the subjects. When you have raters who rate only attractiveness and are blind to everything else, the link disappears.
No relationship between intelligence and facial attractiveness in a large, genetically informative sample
Luck. Genetic load is the accumulated burden of deleterious mutations one carries. These mutations are only slowly selected out, and the ones we possess often go back many generations. The distribution of load is going to be roughly normal in the population. Some individuals – and some families – carry more than others.
The overall load of the population remains mostly constant, because new mutations appear at the rate old ones are selected out.
Actually, the intelligence of all peoples from historically cold-weather farming societies (Northern Europeans and Northeast Asians) evolved to their modern levels during this time. This process was detailed by Gregory Clark in A Farewell to Alms.
This is a pretty good post.
>Actually, the intelligence of all peoples from historically cold-weather farming societies (Northern Europeans and Northeast Asians) evolved to their modern levels during this time. This process was detailed by Gregory Clark in A Farewell to Alms.
If this is true then why do people from Spain have roughly the same IQ as people from Denmark? Now obviously a few outliers don’t disprove a trend but looking at a map of European IQ, it doesn’t seem like there is that much correlation between the two variable until you get to the balkans. What evidence do you have that it was specifically cold weather which caused the increase in IQ?
It would, yes, but a small amount of sophistication is all that’s needed to rescue it. If human intelligence is explained by the fact that allows people to navigate tribal politics and gain status within a social group, under such circumstances it would presumably become highly optimized for that task, including extensive evolved defenses, like boredom, against the intelligence being diverted to other tasks that would not improve reproductive success.
Someone of high raw intelligence with a defect in those defenses against diversion would demonstrate unusually high achievement in other fields not because they are natively “more intelligent”, but because they would be willing to spend the time and effort to perform abstract reasoning rather than scheme to get laid. The persons of high intelligence with the best defenses against diversion would still be bright overall, but they would, thanks to lack of interest, never become serious masters of fields of study other than navigating human social relationships.
This goes back to the fact that there are lots of things that are easy for a four-year-old to do are very difficult for sophisticated computers, and vice-versa. That it takes a very bright person with full-strength defenses to do X does not mean that X is necessarily abstractly a difficult computational task. Tests that specifically measure the ability to do such tasks would correctly sort the raw intellectual power of what we might call “neurotypical” people, but overcredit those with weakened defenses against diversion. In which case, things like high Ashkenazi Jewish intelligence might not indicate a high potential for increasing underlying human intelligence, but rather representing what benefits that can be reaped when the typical defenses against diversion are selected against, allowing more effort to be put into general computational tasks.
One could conceive of it as human brains already being “Myer’s race car”, with brains over-optimized for zero-sum sexual competition, and that we can make human being a lot more “effectively” smarter by reducing that over-optimization in favor of something that allows brainpower to be used for something more useful — general intelligence.
This is extremely weak. Most actual cars that are faster are also better in other ways. That in no way proves that if you make a car as fast as possible, it will improve in all other ways as well. The same thing applies to intelligence.
But it disproves that you can’t make a car much faster than average without making it useless.
See the Mazda3.
“early hominids stuck to the same tools for a million years at a stretch”
yes, but
While the nature of the tools didn’t change much, tool-making techniques in fact evolved a lot over prehistory and tools constantly got refined, so that while early paleolithic and late neolithic hominids technically use the same tools, in practice there is almost as much difference in design and making between their respective sets of tools than there is between, say, this: https://en.wikipedia.org/wiki/File:Stephenson's_Rocket_drawing.jpg
And this: https://en.wikipedia.org/wiki/File:Number_4468_Mallard_in_York.jpg
(Though admitedly it took a lot more time to cross that distance for stone tools).
Though admitedly it took a lot more time to cross that distance for stone tools
Enough time that reproductive fitness through superior tool-creating ability was almost certainly irrelevant. In modern Silicon Valley, being a better coder maybe means your startup succeeds where the other guy’s does not, and his girlfriend ditches that loser to become your trophy wife. More generally, in an era where significant technological progress occurs every generation, there is a steady job market for people smart enough to design better tools, and the pay for such jobs can be translated into reproductive success.
Ca. 500,000 BCE, if you’re the guy who invents a new type of hand axe, great, maybe you become the most popular caveman in the tribe and have twenty genius cavebabies as a result. The next technological advance worth noticing is scheduled for 490,000 BCE. Whatever keeps your clever offspring alive and reproductively fit between now and then, it isn’t the clever ability to create new and better tools.
I think that ancient people kept losing and reinventing technologies them over and over.
That’s an interesting alternative hypothesis, and plausible in a preliterate society, but is there any evidence for it?
It’s hard to say, since not a lot of that stuff survives well, but lithic tool styles often persisted for many thousands of years, which points away from reinvention.
Except that nowadays fertility rate is negatively correlated with income in the developed world. So getting a higher paying job actually translates into less reproductive success, on average.
Though it is a bit silly to talk about “successful reproductive strategies” in a world where simply donating to a sperm bank allows a man to become one of the most prolific fathers in human history.
Women don’t pick sperm at random, though. They want a photo and a resume of your achievements. If you’ve got a degree from Harvard and have composed symphonies and look like Jim Caviezel, sure, your sample will be used as often as the clinic will allow (probably 10 or so times, because they want to reduce the risk of accidental inbreeding). But if you’re an average Joe, homely-looking and not a college grad, sorry, nobody’s going to want your sperm.
Someone linked a story on here a few weeks back of an unemployed, mentally ill man with a criminal record that managed to trick a sperm bank into telling women he was a neuroscience Phd. Apparently had three dozen kids.
I imagine those on here that have reified the selfish gene metaphor into a religion think he is some sort of saint or at least hero.
Anyone know whether there’s a genetic correlation with being a trickster?
All I’m sure of is that it’s a very strong drive for some people.
This seems to have correlation vs. causation issues. Does the effect persist when you control for parental income?
Does the effect exist before controlling for anything? No, not if you look at male income, as in John’s comment that started this.
Suppose what intelligence gives you isn’t the ability to make a better design of hand axe but to be better at the difficult job of making a good hand axe of the current design?
I agree.
When designing race cars, or people, it is best to take the holistic approach. Never mind that a car with a giant engine might not be very comfortable, unless each part has been designed in relation to the whole , it would likely not even be particularly fast (or move at all).
Regarding intelligence and popularity: it’s my sense that popularity is a recent phenomena which isn’t really related to survival or breeding opportunities. I mean, old King Henry wasn’t ‘popular’ was he? Genghis Khan? How many children did Shakespeare have? Was David Bowie ‘popular’ in the sense of having attractive personal qualities, or was he just elected by the zeitgeist? I mean, modern, urban, society is quite unusual – right? Where we actually get to choose our allies…
For what it’s worth, in my experience, popularity seems to be most strongly related to how often you smile, how friendly you are (how social you are). That is, the most popular people are the ones who like socialising the most. If you had two people equally interested in socialising, with equally nice smiles, I would guess that the more intelligent person would win.
Also, if intelligence is related to the environment, why wouldn’t people who live in the desert be more intelligent that people who live in the South of England? Surely it is easier to live in Surrey than the Sahara?
(I suspect that intelligence is most useful, not for social jiggery-pokery, but for bloody inter-tribal warfare. That’s where we need our intelligence – not to make you *like* me. Maybe the people in the desert aren’t as clever because the clever people didn’t want to go there?)
We definitely can’t extrapolate from contemporary experiences of youth social dynamics backward. I think you’re very right about that. We throw together up to thousands of kids who are complete strangers, force them to sit still for eight hours a day doing things that most kids find boring and don’t want to do, and this is not at all what youth was like for humans living in communities of kin well below the Dunbar number learning the traditions and technologies of their specific community.
“That is, the most popular people are the ones who like socialising the most.”
At a slight tangent, I long ago observed that I was a better teacher when I liked the students.
Does anybody have a good source for IQs of national leaders, high-ranking politicians, etc? The last time something like this came up, I tried to see if I could find things like IQ results for, say, US presidents. I had a hard time doing so. One academic has done estimates, but his methods seem awfully subjective.
The only hard, tested IQ numbers I’ve been able to find for the leading elites of any nation at any time are, the IQ tests given to the Nuremberg defendants. The majority of the defendants (who made up a big chunk of the highest leadership) had IQs in the mid-120s to mid-130s. That’s ~top 2% of the population. Only 2 were above 140. Lacking any good reason to believe these numbers are atypically high or low for national leadership groups, I would feel safe betting that most high-ranking bureaucrats and so forth hover around that point. Possibly relevant: by many accounts, due to the dysfunctional nature of Hitler’s leadership style, this was a group of men with an ability to participate in a cutthroat social environment.
It’s generally accepted that it’s hard to relate to people who are significantly smarter or dumber than you, right? If someone with an IQ of 100 can comfortably interact with someone 30 points in either direction (to make up a number), that’s almost everybody they will interact with, especially in a smaller society (like a hunter-gatherer group). Someone with a 115 is still going to be able to deal with most people. 130 and it starts getting harder, but they can still get along with half the group, and the problem-solving abilities are worth it. For someone with a 145 IQ, in a group of dozens to maybe a hundred or a hundred and 50 IQ, there would be a couple dozen people they could interact with comfortably, and that’s not enough to keep everyone else from deciding you’re a snooty nuisance and letting a bear eat you. And as stated in the main post, what exactly is an IQ that high needed for in the stone age, if it’s compromising the ability to interact socially with people?
Even in a much larger, modern society, where there’s much more ability for people to sort, it’s way more likely that someone will have to deal with a person of average intelligence than extremely above average intelligence, and IQs above a certain point are so rare that an IQ above 130 or so – which, if there’s a 30 point bridge, equips them to deal with up to IQ 160 – isn’t worth it for, say, a politician: a politician with an IQ of 130 can comfortably deal with the occasional rocket scientist, and still handle dealing with half the population, meaning they won’t come off as too distant and incomprehensible when dealing with the people they have to convince to vote for them.
Jonathan Wai has published studies where the IQs of business and political leaders are estimated based on the selectivity of the colleges they attended. See here, for example.
Going to a selective college is associated with being rich and having political connections, which is also associated with being business and political leaders. You could argue that some of that is because of IQ anyway, but not all of it is, making for a huge confounding factor.
Some commenters and I noticed this a while back: there are two factors, throughout the animal kingdom, which seem to strongly predispose to intelligence (i. e. which seem to make having intelligence worthwhile). One is being a social animal. Intelligence is more helpful for social animals than non-social animals because of the advantage of adroitly navigating group dynamics.
The other is having dexterous hands. The octopus, for example, is the smartest non-mammal sea creature, though it isn’t social. Not only is a larger neuronal network likely necessary just to manipulate all those arms, the having of all those arms makes cleverness more useful: you realize you can use a rock to smash open a snail, for example.
Being a sort of dog-octopus, humans are the smartest animals. I would say this also sort of makes sense of the “why aren’t the smartest people in high school the most popular” question. They may have more octopus intelligence (which I’d guess is more what we’d call “autism-spectrum,”-type intelligence) than dog intelligence (which might be more like savvy, charisma, etc.).
And yeah, I think the brain being an incredibly energy-hungry organ is, along with the birth canal, probably one of the biggest limiting factors. It burns something like 20% of the calories despite being 2% of the body weight. Like being big, tall, having a fast metabolism, and not being prone to depression, having a big brain is a big tradeoff in an energy-scarce environment in a way which makes it seem like an almost unalloyed good today.
So, yes, let’s all engineer ourselves to have super high-energy brains so we can eat pizza without getting fat and be super smart. Sure, we won’t survive as long on a desert island, but I’ll take that chance.
“Being a sort of dog-octopus, humans…”
I don’t care how the rest of this paragraph goes, this is the best way to begin explaining anything.
+1. Or maybe +6.
Basically, I just came here to reveal & revel in my dog-octopus nature.
Do you prefer the term Octopug, or Dogtopus?
I somehow find the idea of a furry, sociable octopus less disturbing than that of an eight-legged, long-fingered dog. Tangentially, I only recently realized the similarity between bats and long-fingered vampires.
A furry, sociable octopus sounds like a great pet. Eight arms means four hugs.
@ onyomi
I somehow find the idea of a furry, sociable octopus less disturbing than that of an eight-legged, long-fingered dog.
Yes. A dog brain would not be happy in an octopus body. Dogs like bounding along in a well-supported, four-legged body. Octopi can do their slither/pour movement just as well with or without fur and water as not, sfaik
Edit: Even if a dog’s brain were big enough to operate twice as many, and much more wiggly, appendages, I don’t think zie would _like_ it. Dogs want to get on with simple, direct action, soonest.
“Dogs want to get on with simple, direct action, soonest.”
That would be jet propulsion.
Admittedly, you’d have an underwater dogtopus, but seals have fur, so I don’t see any insurmountable problems.
Sea lions are definitely water dogs.
Onyomi:
I don’t speak more that a smattering of Dutch, but I know that their word for seal translates literally as ‘sea dog’
https://nl.wikipedia.org/wiki/Zeehonden
Dog-octopus! Nice.
I like this so much I’m going to self-identify as it.
Preferred pronoun is the octothorpe.
Flying birds have remarkably compact brains.
Engineering based on crow brains might solve the energy cost and birth canal problems of increasing human intelligence, if we’re smart enough to figure it out.
I recall reading a post by Eliezer a while back about someone whose skull cavity was like, 90% fluid due to some problem, but who was still, to all intents and purposes, functional, because the rest of the brain had basically just squished into a smaller space. While I find it difficult to believe that that extreme level of compression had no longterm deleterious effects, it does seem to point to the idea that our current skull size may not be absolutely necessary for our current level of brain power, which also implies that we could probably have a lot more brain power, potentially, without bigger skulls.
I also seem to recall reading somewhere that a big skull size was correlated with a certain kind of intelligence much more strongly than others; if I’m recalling it correctly it had to do with spatial conceptualization: like, how good are you at navigating a large wilderness, or, in the case of the blue whale, which, after all, has a much bigger brain than us in absolute terms, a vast ocean.
But since modern society doesn’t usually demand that you track a caribou across hundreds of miles of indistinct tundra, it seems also conceivable that the types of intelligence we want more of are not the kind that literally require more head space.
Peter Watts discussion of this topic.
Not just “functional”, but IQ 126.
he was able to live with that much brain matter at the point when he died.
it was discovered when he came in to hospital after a motorbike accident.
With conditions similar to his like brain tumors it’s not uncommon for it to only be discovered after a head injury from a traffic accident because the person has a fit/seizure while driving when the condition finally affects too much of the brain for it to handle it.
So he probably wasn’t born with that little brain matter and it was likely discovered just at the breaking point.
In normal brains we know a lot of our brain matter dies as we age and the die-off is indeed part of the process that allows pathways to be optimized.
So you may only need some of your brain matter to do fine at age 25 but if you had that much at age 1 you might be a vegetable barely able to learn without the raw grey matter to pattern match enough data to lay down those pathways you need.
That being said, children tend to recover better from Hemispherectomies than adults.
That occurred to me, too. Not the shrinking brain part, but how much of the brain is devoted to computationally expensive but not necessarily intelligent functions like basic signal processing, motion tracking, handling kinesthesia and proprioception, all the background tasks we accomplish effortlessly that we’ve had so much trouble translating into robotics. If we’re willing to sacrifice that but boost planning and problem-solving, we could squeeze much more intelligence out of the same volume of brain mass.
Though finding and selecting for genes positively correlated with IQ seems like a comparatively easy goal. Selecting for planning but against motion tracking might not even be possible.
I remember some study found the brain areas usually tasked with face recognition (It said this is probably the one thing brains do the best, orders of magnitude better and faster than any AI) were being partially used during chess games, but only by the really good grandmasters and so on…
Peter Watts I think had this story (Or maybe in one of the novels) where they make a “language of faces” in order to harness that cognitive power and use it for something else. Trippy stuff.
Haldeman did this too, in “None So Blind”. Not his best writing, but a cute idea.
Isn’t spatial visualization ability the thing that lets you drive cars, read maps, solve geometry problems, and mentally rotate objects? It’s not just for hunting caribou.
I’m not saying we should get rid of it or not increase it if we reasonably can. Only that, if it turns out that geometry ability is a particularly costly ability in terms of cranial real estate, there might be ways to increase other types of intelligence without having to deal with a freakishly large alien head.
“Dog-octopus” sounds like the output of one of those neural network image processing toys.
Great comment. The human hand is a stunningly versatile tool. The human brain devotes a lot of attention to its working. Link.
I prefer to think of myself as a hyper-evolved, futuristic, thunder-monkey.
You’re assuming that relationship between intelligence and physical health, attractiveness, fitness, etc. is a monotonic relationship. I don’t think that’s a given at all.
Going back to the car design analogy: a lot of generally good car design features will make a car go faster, an efficient engine, good aerodynamics, solid frame, etc. But when you start optimizing for speed, like a race car, you start losing reliability and fuel efficiency.
On why humans evolved intelligence, my impression is that rather than social skills, it was adaptability (which our sociality would also have been a part of). The human ecologinal niche was “hyper-versatile omnivore.” Today, the most versatile animals are also the smartest. Evolution initially gave us intelligence to make us more adaptable, but then accidentally gave us technology and civilization as a byproduct, which led to a runaway increase in the value of intelligence.
The reason evolution is so stingy in giving out intelligence in the first place is that before that point, its advantages are little more than any other attribute, only really useful if you’re a social animal, or in an ecological niche demanding versatility.
I suspect that engineering smarter humans is a more promising strategy to avert global catastrophic risk than anything MIRI might do with type theory. Not that type theory isn’t cool. Type theory is cool.
I wouldn’t be so sure about this. If everyone is brought to high but not infallible IQ, then the number of people with the wherewithal to screw something up horribly (at least in certain fields like AI, which your MIRI comment alludes to) also dramatically increases. On the other hand, if a select few are engineered to come out way ahead of the rest, they might BE the global catastrophic risk. A super-genius madman or group of such is pretty scary if they’re among the only super geniuses around.
Which isn’t to say that engineering can’t possibly help, but the process introduces its own risks and it’s not clear to me that it would make things overall less risky.
This suggests a question relevant to both MIRI and the effects of genetic engineering for high IQ–are smarter people, on average, nicer or less nice?
My casual impression, from people I’ve known, is nicer. For a couple of extreme examples, Von Neumann seems to have been a nice person, Feynman perhaps not.
Somewhere I came across a reference to someone who had known Smith, Ricardo, and Malthus, and commented that it was to the credit of a field that its founders were such nice people. Ricardo clearly had a very high IQ, the other two possibly as well.
@ David Friedman
Results of a common cause? Children raised by nice people learn nice behavior, and are more free to develop their intelligence.
Am I the only one that finds the prospect of creating ridiculously intelligent humans less interesting than the prospect of bringing the stupid up to our level?
There may be unsustainable genetic tradeoffs at very high IQ levels, but there certainly aren’t at the ~120 level. Suppose cheap genetic editing allows us to do something like, say, alter an embryo after conception at a low enough price that it can be provided for free, either by government or by charity. The deal might be something like “we’ll modify X genes in your child to raise their IQ by 15 points.” Even if this doesn’t work for people at IQ 140, it should work for people at IQ 80.
I live in a high-IQ bubble and don’t know how people outside of it think — but as far as I know, wanting better opportunities for one’s kids is pretty universal. Such a policy, if accepted, might go a long way to counteract the dysgenic effects of dumb people having more kids. Anyone concerned about technological unemployment of people with <X IQ should find this interesting, too.
A couple concerns about focusing on bringing everybody up to 120 level rather than on making a few 200 or 300-level supergeniuses: a very tiny minority of very smart people are responsible for most of the big advances in technology. Therefore, having just a handful of IQ 300 people might advance technology more than having millions more IQ 120 people.
Moreover, it is known* that smart people in boring jobs are unhappier than stupid people. So long as we have crap jobs that need doing, making all the people doing them have IQ 120 instead of 100 might not increase happiness.
The area where I do find this most intriguing is politics. Right now politics feels intolerably dumb because it must, necessarily, gravitate toward an IQ 100ish level of discussion. Whether raising the average to equivalent of today’s 120 would make everyone see the wisdom of smart policies and informed voting or just make them more apt to go in for disastrous utopian schemes, however, I’m not sure.
*Dothraki citation practice
Technological advances and smarter politics aren’t the only advantages, though.
If higher intelligence causes people to commit fewer crimes, live healthier lives, behave more financially responsibly, and so on, raising intelligence for everyone average or lower by one standard deviation could have huge benefits for society as a whole.
And Error’s example was a standard deviation boost for someone at the example IQ of 80, which is more than a standard deviation below the norm. That takes someone from the bottom 15% of the population, to just below the median.
On the other hand, keeping the masses stupid may be necessary for governments and other social authorities to retain control over the people. A more intelligent general populace may well lead to greater disorder as a growing number of people take firm ideological positions and war (literally or figuratively) with each other or with the system.
The masses aren’t stupid, though. The masses are average. Someone of IQ 80 is considerably below average – about 10th percentile. Bringing them up to 95 would put them at around the 35th percentile.
Being able to lift the average to what is today considered smart, or very smart, or genius – a society where the curve centres on 145, let’s say – would have completely unpredictable results. That’s less than 1% of the population today. Even among highly selected elite groups IQs that high are rare.
In contrast, a society where everyone below average was raised to what is now the average, but nothing changed for those above the average, would be weird in that there would no longer be a bell curve distribution, but it would probably be more predictable.
I agree that 80 -> 95 is much less concerning in this regard than 100 -> 145. Still, I would be curious about social effects. My (unsupported) guess is that an IQ 95 individual is more likely to join Twitter culture wars than an IQ 80 individual, or failing that more likely to do so in a way that influences others. I suspect there is a correlation between intelligence and caring deeply about abstract ideas.
And there’s a whole lot of hostility on Twitter over such ideas, which sometimes has ramifications off the internet as well. A significant increase in the number of people getting involved could be harmful to society.
Lacking any data about the intelligence of people who get involved in Twitter culture wars and such, I get the feeling that it’s still something for the higher end of average, at least (which raises the distressing thought that the people on comment sections might actually be on the higher end of average).
What % of the population now is involved in the Twitter culture wars, vs what % attends undergraduate university? Perhaps there would be more people doing it in a strictly follow-the-leader fashion.
You are right that it might have the effect of making a lot more people care about culture war stuff, though. I hadn’t thought of that.
Still, which is worse: the culture war, or the sort of social problems associated with below average intelligence?
Right now, you’re correct that the culture squabbling is meager compared to actual murders and the like. No argument there, and I see why it makes some IQ engineering an attractive prospect.
I fear that the culture war is still in a stage of germination and that we haven’t seen it truly explode yet. When I look at some of that stuff, the hate between groups that I see is almost staggering…if things accelerate, I honestly do worry about violence on a medium or large scale.
The prospect of such violence, with belief and purpose behind it, concerns me more than low IQ crime which, while more frequent than desirable, is scattershot enough that it seems unlikely to do much damage to the social order. Ideological war or revolts seem much more threatening in this sense, and I think the low IQ not caring about the relevant ideas as much has been a significant buffer against the situation getting too out of hand yet.
Of course, this is all largely speculative, but I hope I have explained myself better. Significantly increasing the number of people who care to associate with a “cause” may be dangerous. (Initially I was thinking of 100->120 due to onyomi’s comment, which would be a much more numerous group, but 80->95 to a lesser extent)
Perhaps included with the free intelligence gene fixing would be some aggression gene fixing too? (Honestly, I think if we could find the genes that make people more or less moral, that should be a higher priority than fixing intelligence anyway.)
High IQ enclaves like LANL or Caltech don’t seem to turn into dystopian anarchies. A Brave New World is a good book, but I’m pretty sure that part is a misfire.
I actually think it’s more likely than not that a higher IQ populace would lead to better, not worse, politics. That said, Caltech is not a sovereign nation.
I think raising up IQ toward average would lead to better politics too. I mean, they’re going to vote! It would be nice if they could comprehend at least some of the issues they’re supposed to be deciding! (Not *all,* because who can? Not me.)
Average IQ makes you way more employable than 80 IQ, thus you won’t be “the impoverished masses” that it seems people are afraid of. It probably means you’ll get along better socially. You’ll get the joy of reading good literature. And, yeah, you might make an ass of yourself on Twitter, but you’d be in good company.
What I want to know is, is it ethical to deliberately keep some people dumb, if you could fix it, just because you are worried about the consequences for everybody else? Surely any slight problems that arise for everyone else would be minimal compared to the vastly increased quality of life in the subjects!
@Anonymous: But it goes past murder, or even crime. If higher IQ does in fact lead people to live healthier lives, or take stupid risks less, or manage their personal finances better – would those things be outweighed by a possible escalation in internet shitstorms?
You worry about an escalation into the real world – a lot of online drama is very inward-facing, hardly the sort of thing that’s going to lead to a civil war. It’s not as though left-wing campus politics from the early 90s onwards – sharing a lot in common with the left-wing bits of the internet culture war – has led to bloodshed.
The people involved in all this online hatred – which is alarming, I grant you – seem often to be the sort of person who might fantasize about their preferred flavour of covering themselves in the blood of their enemies. But probably no more than that. Every situation in history I can think of where intellectual types did get their hands bloody was only because of some other big disaster: if you’re worried about the new activist left, well, the Russian revolution only happened because of the first world war. If you’re worried about the new far right, well, ditto for the rise of the Nazis. Any situation that will lead to clashes of armies marching under banners featuring different Pepes will only come about in the aftermath of something else.
I hope you’re not right, basically. I have some reason to put stock in this hope: I think you’re overestimating the degree to which the average person – who, after all, is in the 50th percentile of IQ – cares about politics and so on. The average person doesn’t really follow the news, isn’t a member of a political party, is increasingly less likely to vote, etc.
I think a lot of us exist in a bubble where everyone cares about politics or culture war or whatever. But that’s not the norm. Additionally, if it does come down to it, is it harder to rile up a mob of people with IQ 100 than IQ 80?
> Every situation in history I can think of where intellectual types did get their hands bloody was only because of some other big disaster: if you’re worried about the new activist left, well, the Russian revolution only happened because of the first world war. If you’re worried about the new far right, well, ditto for the rise of the Nazis. Any situation that will lead to clashes of armies marching under banners featuring different Pepes will only come about in the aftermath of something else.
By golly, what a weird theory. Spain was neutral in World War I, the reasons for their bloody, genocidal civil war were purely political. Yugoslavia was non-aligned in the Cold War, the reason their genocidal civil war started was basically the perceived economic unproductiveness of the Serbs (but naturally, religious tensions surfaced during the war). The Syrian Civil War turned from mass protest to armed rebellion, and the role of “social media” (i.e. Twitter) in this process is well-documented.
@Publius Varinius: Yugoslavia’s collapse into civil war was to some extent the result of the death of Tito, who had been holding the show together, and it’s hard to imagine that Syria would have turned to chaos without the influence of events across the Middle East and North Africa.
I’m not saying every single time there was conflict, it was the result of a major power vacuum or something similar – but the thought of the various participants in the internet culture war getting into actual serious physical conflict seems unlikely without something larger going terribly wrong.
> Yugoslavia’s collapse into civil war was to some extent the result of the death of Tito […]
> I’m not saying every single time there was conflict, it was the result of a major power vacuum or something similar – but the thought of the various participants in the internet culture war getting into actual serious physical conflict seems unlikely without something larger going terribly wrong.
The Serbian “nationalist” movement got serious around 1986, rapidly spreading across Yugoslav academia. The narrative was that Serbians were victims of assimilation and “cultural genocide” at the hands of the Croatian elite. This was well after the Tito died.
Anyway, I think it’s dangerous to dismiss the probability of civil wars based on “bad things would have had happened in the past”. After all, 11 years is a long time, and one can always find minor events to blame:
– say, a policeman shooting an unarmed man, sparking existing tensions in a predominantly black city, leading to three waves vandalism, looting, rioting, arson
– or political division caused by loudmouthed, controversial businessman running for president
– two major attacks by religious terrorists killing 147 people in a capital city, and making “terrorist attack” a more likely cause of death than “traffic accident” in that city for that year
etc.
@Publius Varinius: Again, I hope I’m right and this particular Anonymous is wrong.
The US has seen worse upheaval than currently: for instance, the urban riots in the 1960s seem to have been much worse than anything recently – and have there been any riots recently? It seems to me like BLM means that the immediate response to police gunning down a black guy is an organized protest camp, rather than a riot.
Additionally, this didn’t begin with the prospect of upheaval – it began with the question of whether raising people with below-average intelligence to the average would escalate the internet
culture war to the point of real-world consequences beyond those that already exist, and whether that would be a good tradeoff.
Here’s another angle on the subject of intelligence and potential upheaval: would the world be a better place if every wannabe spree killer and/or terrorist was as smart as Anders Breivik and Timothy McVeigh? Looking at the accounts of many mass killings and attempted mass killings, there are many instances where only dumb mistakes on the part of the killer prevented far more people from dying. For example, Eric Harris and Dylan Klebold might have killed as many as 500 people if they hadn’t sucked at bomb-making.
And given how terrorists are disproportionately likely to be engineers, I’m not confident that making people smarter would also reduce the number of potential mass killers.
Only vaguely related, but is there any information about whether smart people are less likely to abuse their kids?
@NN: True. That might be an issue if there was an across-the-board IQ boost.
@Nancy Lebovitz: Abuse physically, or emotionally/psychologically? I would guess the first is correlated with lower IQs.
@dndnrsn:
The US has seen worse upheaval than currently: for instance, the urban riots in the 1960s seem to have been much worse than anything recently – and have there been any riots recently?
The fact that serious urban riots happen all the time – and almost never lead to escalated culture wars or civil wars – is evidence against your “May 6, 2016 at 10:21 pm” hypothesis. My point being: you see bad stuff before every escalated culture war only because bad stuff always happens in any 10 year time interval. For example, the description above refers to the 2014-2015 riots in Ferguson, MO.
However, culture wars escalate all the time for minor reasons and sometimes the accompanying ideology comes from academia (e.g. the “Serbians are victims” meme came from Yugoslav universities).
@Publius Varinius:
Do serious urban riots happen all the time?
If higher intelligence causes people to commit fewer crimes,
Mighty big “if” there. The available evidence is equally consistent with the theory that higher intelligence enables people to get away with more crimes.
Is that true of all crimes, though? There certainly seems to be a correlation between lower IQ and violent crime. It is easier to conceal (as opposed to get away with) some crimes – violent crimes tend to be harder to conceal.
I can believe there’s a ton of white collar crime committed by people with 130+ IQs we don’t know about, but it’s doubtful there’s a whole bunch of murders where nobody has noticed a body or noticed that someone’s missing.
higher intelligence enables people to get away with more crimes
Exactly. Why go the messy route of hitting people over the head and robbing their wallets for chickenfeed, when your smarts mean you can fine tooth comb the regulations to take advantage of loopholes meaning you can set up investment scams, take investors for all they’ve got, then put the assets in your wife’s name, declare bankruptcy, walk away and set up another company and repeat the whole cycle again?
The technological problem of preventing most sub 110 IQ (or whatever) births looks to me far easier than achieving a 300 IQ birth. In fact, given draconian enough policies you could more or less achieve the former with existing technology, whereas I’m fairly skeptical about Hsu’s 10,000 allele => 1000 IQ idea.
On the other hand the social problem of implementing the former looks a lot harder to me than the latter. Even if we can do better than forced sterilization and abortions, and go with modifying embryos, that still requires well nigh universal IVF and concomitant effective birth control for all sex.
In short, I don’t think either program is reasonably likely in the near to medium future though for different reasons. For more likely is that some relatively small group of people — mostly those with high IQs already — use technology to make sure their children are on the upper end of what’s we see today, with some edging up over decade long periods of time.
For technologies, broadly defined, that raise average IQ, consider fornication and adultery. If status correlates well with IQ and low status women often get pregnant by high status men they are not married to, IQ’s might rise pretty fast.
My impression is that in 18th and 19th century England, it was pretty common for female servants to get pregnant by their employer or his son. How many of the resulting children survived to adulthood I don’t know, but there have been societies where high status men made some provision for their illegitimate children.
I have no idea how common it was for married women to produce children fathered by men of higher status than their husbands, but I gather that the pattern is fairly common among “monogamy tempered by adultery” bird species.
On the other hand, the rate of false paternity in modern societies seems to be pretty low–a percent or two in several large scale studies.
Where “modern” means the past 500 years.
Do we have decent data that far back? I believe the studies I have seen were of people presently alive.
This must be a massively simplified model of values to optimize for, surely, not the way you think about this issue for practical purposes, right? Because this comment (and the whole discussion, starting with the OP, in truth) reads a bit like a not-very-savvy autocrat taking a look at society in search of adjustable parameters, and seeing (e.g. here) basically two knobs to turn: technological advancement, and workplace satisfaction. Which he’ll pursue to the extent of massive, dystopian interventions in people’s reproductive freedom, failure to use his powers to prevent large numbers of stupid people essentially for sweatshop purposes, and deliberately keeping the level of political discourse dumbed-down as a preferable alternative to the “reasonably smart” falling into the uncanny valley of bad rationality and starting to dream up utopian socialist communes, or something. (I think we can all agree that under such plans, SSC would be a likely hotbed of political weirdness marked as among the first to go.) I get a whiff of Brave New World from all of this.
That’s quite a lot of values to get thrown under the bus in favour of a high modernist dream of a high-output, well-oiled system. Please, people, have some sense of perspective. I get that it’s all abstract discussions here, but visions like this are why people looking in from outside call us beep bops. You’re all too inclined to prioritise resource spending on elites rather than on the majority, and I do not suspect entirely rational reasons for this. There is real value in focusing upon raising the overall intelligence waterline, if that comes to be in our power, in ways that we can barely grasp right now; it might well modify crowd dynamics, massively improve the efficiency of the education system, push some industries (think prolefeed) into obsolescence, and yes, contribute to your dear old technological advancement in fields that depend heavily upon grit — and we might still get our occasional once-in-a-generation genius who gets a few tens of new theorems discovered and named after him or her. But under some chosen set of societal values that’s not good enough, and we have to fast-track humanity into sci-fi levels of technical prowess in a few generations, whatever that would take.
It’s not a world I would like to live in. It would be inherently loathsome for the vast majority of people to be reduced to the meat puppets through which their elite rulers, be they republican or, as many here fancy, private, implement their visions for societies, whose breeding, birth, existence is carefully supervised for state purposes. But our hypothetical autocrat doesn’t take that into account. My point being, unless you (general “you”) can see beyond a few narrow parameters to be adjusted at any cost and into the full range of human and societal values, you are not fit to rule and your vision does not merit supremacy among mainstream theories of governance.
Conceivably, a higher general intelligence would probably automate some features of those boring jobs.
There’s also the benefit of outsourcing our work, too. There’s plenty of low-IQ poor people willing to do the crap stuff!
“Whether raising the average to equivalent of today’s 120 would make everyone see the wisdom of smart policies and informed voting or just make them more apt to go in for disastrous utopian schemes, however, I’m not sure.”
The degree to which elite liberal arts colleges are political monocultures, and not very bright monocultures at that, suggests a negative answer to the question.
There’s a lot more going on there than just everyone being above average intelligence.
I don’t think bringing everyone up to 120, or even 100, is strictly more interesting than going for a lot of people at 200 or 300, but I do think it’s very important and doesn’t get nearly the attention it deserves.
In the same spirit, there are people who live to be 90 or so in good health. It’s genetic. It’s being studied, but the good thing about it– compared to greater life extension– is that we know 90 in good health is possible.
Yes. It’s not that some people drop dead at 60 and some people live until they’re 120. It’s that some people are falling apart, physically and mentally, by their mid-60s, while others are still going strong. The second group is probably going to live longer – but their quality of life is going to be significantly better, and it’s easier on everyone else too.
In a non-genetic engineering context, no, you aren’t the only one, that is the main focus of educational theory and experimentation at the k-12 level.
Am I the only one that finds the prospect of creating ridiculously intelligent humans less interesting than the prospect of bringing the stupid up to our level?
Loss/accident analysis post WWII led the US military to conclude that having people of an IQ less than three standard deviations below average was overwhelmingly associated with fatal and catastrophic accidents. Congress was persuaded to pass a law forbidding the enlistment of anyone who scored that low (even in universal drafts) and as a practical matter, the military is extremely reluctant to take anyone who scores more than two standard devs below average. (the practical lower limit hovers somewhere around 1 -1.5 SD. Standards are higher when the military is shrinking and/or there are many volunteers, lower when the numbers need to go up or the volunteer pool drops(*).
So no, not just you.
(*) Size of the volunteer pool is most accurately tied to the overall economy – lousy economy = more volunteers, good economy = more people trying other things. International affairs are a much less predictable metric.
I heartily agree. Kind of how I feel like trying for immortality when so much of the world doesn’t even have clean water is a strange priority. When some people don’t have basic opportunity, that seems the thing we should work on before giving some people extra opportunities.
I imagine a world in which some supergeniuses use their intelligence to acquire most of the world’s wealth and resources, create robots to work for them so they can lay off all their employees, and all the stupid people are left out in the cold to squabble over marginal land and water. Is this really what we want?
And in answer to “well, what if they get on Twitter?” — well, it seems a reasonable price to pay for their ability to get a job that will support themselves as manufacturing and other menial jobs vanish. Their ability to enjoy literature and learn science. Their ability to participate in a positive way in society instead of always being unwanted and unliked.
Think of Brave New World, in which most people are deliberately damaged before birth to make them dumber, so that they will be satisfied with doing menial jobs. Would it be ethical to do this? Then why is it ethical to withhold a treatment that could help, for the same reason?
The trouble is, I don’t know how this is to be done, exactly. IVF is expensive and genetic engineering would be more so. And the numbers of people that could benefit are large. Then there’s the question of how you would convince 80-IQ mom and dad to participate. Seems they’d be a bit suspicious of your help — how do they know you’re not going to sterilize them, give them somebody else’s embryo*, or whatever?
[*this being the most obvious and cheapest way to give less-intelligent parents a genius child]
The tradeoff argument makes sense when considering attributes outside the observed viable human range. Even if desirable attributes are positively correlated in the normal range they may become negatively correlated at the tails of the distribution, and this is in fact expected.
Also, while the effects of individual alleles may be roughly linear in the normal range, saturating non-linearities will probably occur at the tails.
This is why I think that Hsu is overselling his proposal by promising IQ 1,000. This is like saying that we could make 16 meters tall people by a mere selection of the genes that control height, except that in reality very tall people have a number of health problems and Robert Pershing Wadlow, the tallest recorded man at 2.72 meters, needed a cane and special braces to walk, and died at age 22 because of an infected wound caused by such braces, therefore ultimately because of his height.
On the other hand, it seems that it should be possible to use allele selection to create healthy people with IQ at the top of the observed range, or even a bit above. There are healthy and generally successful people with IQs in the 160-170 range, possibly above (it’s difficult to say since standard IQ tests may introduce extra measurement error at the tails), so it should be possible to use genetic editing to achieve this.
What about reverse causation? If IQ is positively correlated with height, attractiveness, etc. could, say, being taller cause slightly higher IQ via, say, better experiences in schooling? (maybe that’s already been accounted for, or just dismissed as implausible)
Actually, I think you may be right — taller people have larger heads, larger heads means larger brains, and that’s the pathway. Or so I’ve read … someplace. (If only I were taller, maybe I’d remember the source! 😉 )
This is why I posted links to the studies finding that the correlations are genetic.
I don’t think Scott is optimizing hard enough here. Specifically, I think that there is a lot of fuzziness around what exactly Scott, Myers and Hsu really mean when they say ‘optimized for intelligence’. When saying that people with IQs of 115 are 20% more likely to make it to age 76, should we really take it as evidence that ‘optimizing’ for intelligence doesn’t affect other positive traits?
Imagine if Scott was making this argument 50 years ago, using the same criteria, IQ 115. The Flynn effect is about 3 points per decade (though I don’t know if that is true for the 1st std. dev as well), so all of those Optimized 115 IQ individuals would today be just about average, or to put it another way, the average person got about 15 points smarter.
This doesn’t sound like aggressive optimization to me, and I can’t imagine that this is what Myers was talking about. In fact Hsu is writing about IQs of 1000, which are a very different proposition. Can people with 115 IQ tell us that much about even 190, let alone 1000 IQ?
So the 1000 IQ that Hsu mentions sounds completely made up, but what if we imagine he was saying that we optimize for intelligence such that the average person would be as smart as the smartest people today, around IQ 160? Do they live longer? Are they more social? Are they taller and Fitter and more attractive? That is the kind of analysis I think I would need to be convinced here, at least until we start talking about IQs in the heavily theoretical range. Alternatively Scott might just have meant that we can optimize the mean up about 15 points without too much trouble, in which case his epistemic uncertainty seems a little unwarranted 🙂
If we go back to the car analogy it turns out that we can make the same arguments about optimization. The ‘Car factors’ that Scott picked out were speed, safety, comfort and cost. All of the are positively correlated in the kind of cars that we actually see on the road. Aesthetic appeal, accelleration, handling, these are also pretty well correlated. If you take the equivalent of your 115 IQ, perhaps a BMW M2, you could look at it and say that optimizing for speed also optimized those other factors, or you could reevaluate and say that you were no where *near* optimizing speed!
As it happens we have the advantage of knowing exactly what happens to all those other factors when you start to selectively optimize. If you optimize Speed, Accelleration and Handling while keeping a minimum of safety and comfort you get an F1 car. If you Optimize speed a little more you can forget about handling completely as you break the sound barrier in your ThrustSSC. And then if you forget about safety too you could always enjoy your last moments in one of these. So the idea that there will be no tradeoffs once you truly start to optimize only makes sense if the genes for IQ really are the exact same ones as the genes for our other desirable traits.
However! The proposed explanation for IQ correlation with other positive traits – that they rely on the same genes – also seems a bit gummy to me! If IQ is heritable and athletic ability is heritable, and we know that already, do we really need a new mechanism to explain why intelligent, athletic Niels Bohr had athletic, intelligent children? Perhaps I’m misunderstanding the article, but the evidence that, eg: the link between intelligence and lifespan being genetic, does not mean they are actually the same genes, only that someone with gene A for IQ is more likely to have gene B for height. Perhaps having gene B compensates for some retarding effect that A might otherwise have on height.
On the surface that doesn’t seem unlikely. The positive traits we’re talking about all seem like good proxies for sexual fitness. If smart people become wealthy, and wealthy people attract healthy, attractive mates will their children be more likely to be wealthy, attractive and smart? I don’t see why not. Likewise, in any given academic, what will tell you more about the height of their children: Their IQ, or their own height?
So, will making ourselves smarter require any tradeoffs? No, as long as smarter means “someone who would be considered average in 50 years regardless of what we do”. But Maybe otherwise. Are the traits for IQ and a host of other positive factors linked on some fundamental level? Who knows! But it looks to me like modelling them as independent genes that each contribute to sexual selection would work just as well.
Genes are randomized. That’s why you are entitled to infer causality. What process could ensure that over the entire population gene A is always accompanied by gene B rather than ~B, when B vs ~B is constantly being randomly inherited during conception? If A!=B, it should not be possible to take a polygenic score for intelligence and in an entirely separate cohort, use it to effectively predict grades, misbehavior, cardiovascular health, etc.
I agree with that reasoning but no evidence was presented along those lines. Height and Intelligence are both polygenic traits and they in fact are not always seen together. To me that is enough evidence to say that A!=B (for some or most A)
I’ll let someone else restate my theory
He goes on to say that:
that is, that even between siblings one can inherit the genes for intelligence without inheriting the genes for height, etc, which would be bizarre if A=B
Unfortunately for me/(us?) later on in that same post we get to a study that actually tries to unscramble this exact question:
So I guess that I would have to revise my thought here a little, but given that half the genetic correlation between height/IQ is not pleiotropic, it still seems a reasonable conclusion that optimizing for intelligence could have a negative affect on height.
Maybe the high IQ genes are, in fact, sweeping through humanity. There’s some evidence of that (Flynn Effect).
The Flynn effect is going too fast to be genetic.
I’ve seen this claim before, but what is the evidence for it?
The Flynn effect tracks modernization. What else correlates with modernization? Lots of things, for instance a strong reduction of infant and mother mortality during childbirth.
If the Obstetrical Dilemma exerted a strong selective pressure against greater intelligence until modern times, then lifting this pressure with modern obstetrical practices could have a huge impact on the average intelligence in the span of few generations.
Child malnutrition and child mortality also strongly negatively correlate with modernization. The brain of a small child may account for nearly half of their energy consumption. This causes an environmental effect: in malnourished children brain development is stunted, making them less intelligent, but it may also cause a significant genetic selection effect: children with brains that are just too large or metabolically active for genetic reasons used to starve to death. With modernization this does not happen.
“If the Obstetrical Dilemma exerted a strong selective pressure against greater intelligence until modern times, then lifting this pressure with modern obstetrical practices could have a huge impact on the average intelligence in the span of few generations.”
Not just because of evolution towards higher IQ. If deaths in childbirth were selectively weeding out large headed/high IQ children and their mothers, you would have an equilibrium where the genotype was smarter than the phenotype. Introduce modern obstetrics and phenotype rises to genotype. My old explanation of the Flynn effect.
And it’s happening in societies which as far as we can tell are dysgenic with respect to fertility.
I wanted to return, again, to Scott’s hypothesis that intelligence isn’t linked to sexual success/social status.
This may be true in high school (although I’m actually far from certain of that– what literature I’ve seen on the subject suggests that most popular students have higher-than-average IQs), but in most cases, it’s not true beyond that world. (Indeed, you could even argue that a lot of nerds are trading off present-day social status/sexual success against future status/success.)
Indeed, think about the cultural image of high-average-IQ professions. Doctors, lawyers, politicians, financiers, corporate executives, humanities and social science professors, journalists, artists, economists, naturalists, biologists, and clergymen all tend to be considered more attractive because of what they do. The only exceptions crop up in the non-life science STEM fields. Even in this case, though, the romantic problems faced by intelligent people seem to be grounded in our specific historical moment. Richard Feynman, Erwin Schroedinger, John von Neumann, Albert Einstein, and John Nash were all noteworthy womanizers, even though some of them were deeply strange people (Einstein is often retroactively diagnosed with some Autism spectrum disorder, Nash was schizophrenic, von Neumann was such an absentminded cloudcuckoolander that he would read while he was driving).
If I were to make a guess about why the STEM fields have trouble now, I would guess that it’s because a lot of them today have cultures which deliberately encourage a certain degree of slovenliness. (I have a BA in pure mathematics, so I have a lot of experience with this.) I would imagine that if the typical male computer science undergrad shaved off his ill-advised facial hair and traded in his ill-fitting xkcd shirt and cargo pants for a well-fitted graphic tee and jeans, he would have a lot more romantic success.
Von Braun was another well-known womanizer.
Yeah – I think that the emergence of the thoughtful, but sexually frustrated and self-loathing young man is a by-product of our experiments with mad liberalism.
An intelligent man who grew up in a basically christian society (even if he wasn’t a christian) knew his worth wasn’t tied to his bang number, and knew that there were social implications to sex (I would guess that even if Einstein was quite the stud by early twentieth century standards his level of promiscuity wouldn’t be at a level to be noteworthy these days – there have of course always been a hard-core of sexual libertines, but my impression is that they were an insignificant minority until recently).
A thoughtful young man growing up at any time since the 60s has been told that essentially, the meaning of life is to bang – uh – and if that comes into conflict with his gentler sensibilities – more fool him.
I mean… I’d opt out of sex too, if I believed half of the rubbish that was promoted by the liberal sex people.
I’m acquainted with a fair number of left-wingers, and I’m pretty sure none of them either believe or have promoted the idea that the meaning of life is to bang. Not even the one who actually is a sexual libertine. Most of them would probably attribute the notch-on-the-belt mentality to dumb-jock college culture — hardly a bastion of liberalism, though probably not conservative either.
I think we’re in danger of straying into the dark swamp of the antonymic homonyms here… but can we agree that attitudes towards sex became far more liberal (socially) – there was the idea that what you as an individual did did not affect society (which is obviously an idea in conflict with much left wing sentiment) – and also that intellectually there was an increased desire to see the internal life of humans as a product of impersonal forces rather than life as a product of the personal (god).
So, you’re right – very few people are going around saying “the meaning of life is to have sex” – but I think that is the implication that many people will draw from the basic ideas that have been heavily promoted – there are few negative implications to sex, we are animals that exist only because we have a desire and capacity to have sex.
Though, I think the worst of this has passed, and we seem to be heading into a new age of prudery (though a weirdly distorted feminist flavoured modern prudery, to be sure.)
Are we really heading into a new age of prudery? I mean, people have been predicting this for a while now – back in the late 80s/early 90s there were books and articles and such about how fear of HIV would lead to less promiscuity, and about how feminist activism would lead to less promiscuity … Katie Roiphe wrote a book on each topic, even.
And indeed we do see that millennnials have less sex, and start at older ages, than their parents did when they were young.
Though it is a bit of an open question how much of that is due to a new age of prudery, and how much of that is due to internet porn.
There is also this problem in Japan called “herbivorous men”–men who aren’t interested in dating, sex, marriage, or children. This is attributed to shut-in/agoraphobic tendencies (NEET/hikikomori) and possibly also internet porn.
Great, that’s what we all needed, more implicit associations between male sexuality and (literal, carnivorous) predation.
Why is it a problem?
Let’s put it this way, which side are you on that this is not a problem?
I’m under the impression that this “college culture” has been actively promoted by or even invented in pop-media targeted at teens and young adults.
I don’t want to claim it’s a conspiracy, maybe this stuff just sells better, but I seriously doubt that the typical college frat party looks like what you see in the movies. To the extent that it does, it’s probably college kids trying to emulate the movies than the other way round.
Agreed. I’m remembering (but couldn’t find on google) studies asking college kids how much they drank and had sex vs. how much they thought their classmates, on average, drank and had sex. The findings were consistent: self-reported behavior was more modest than peer expectations. Usually substantially more modest.
I don’t think it’s a conspiracy, just a combination of the salience heuristic and normal human preoccupation with other people’s sex lives. Which preoccupation also explains the media promotion: yes, that stuff sells better.
@LCL — I wonder how much of that is driven by the same network effects that make your friends likely to be more popular than you. If popularity correlates with drinking and promiscuity, then accurately observing your friends could mislead you about the population average.
Agreed about frats.
Anecdote: A few years ago, after our father died, my brother asked his old fraternity for help cleaning out our father’s condo. The job was more unpleasant than it sounds like; our father was a hoarder and the place was effectively a garbage dump.
He got the help he asked for. Some half a dozen guys he’d never met and never known volunteered and they got it sorted out over a couple days — for no other reason than that the request came from a frat brother.
I was floored. I lost a lot of my contempt for fraternity culture after that.
I’m pretty sure none of them either believe or have promoted the idea that the meaning of life is to bang
They’ve pretty consistently denounced or ridiculed anyone who suggests that maximal banging is in any way antithetical to the Good Life. That, plus some of the left’s chosen role models, may not have been intended to send such an enthusiastic pro-casual-sex message, but I think it does have that effect.
If you name some of your anti-rape events “Slutwalks”, yeah, I get the deeper meaning but you generally don’t go with that sort of thing if you aren’t comfortable with the obvious interpretation that most of the audience is going to take away.
It’s not really a complaint about the left as such. They are still mourning the death of God. On the one hand maybe it was a bad thing, but on the other hand he was dead in the West before any of them were born. Maybe it’s time to get over it.
Get over it and on with your plans, you mean.
Nah ain’t got no particular plans. Just sick of this silly psuedo-nostalgia for an era none of you actually lived through.
Speak for yourself
Yeah… I don’t think it’s nostalgia… more of an informed critique.
I was born a love child of 1970’s libertines, and experienced all of the usual shit associated with.. that. Not to complain – could have been worse – but, you know how all of those really tragic and stupid atheists were brought up in religious households? I’m like the opposite of that. I was brought up to believe that metaphysics equaled… I dunno like someone smearing themselves in dung or something. Like literally – the look on my father (a physicist)’s face when he said the word metaphysics… I thought that metaphysics was the same thing as astrology. I thought that believing in god meant you were stupid. And now… I understand everything. I know what morality is and why it is important. I understand why it isn’t stupid to believe in God. And I know that the smug, self-satisfied atheists are wrong, both in terms of the misery their ‘ideas’ will inevitably bring, and the certainty with which they think they are supported by logic.
So… on one level.. yeah… maybe I’m just like those moronic atheist fellows. But at the same time… uh… I am your children.
I sympathize with your object-level positions, but come on. Self-awareness isn’t degeneracy.
“I sympathize with your object-level positions, but come on. Self-awareness isn’t degeneracy.”
Hmmm…. I thought it was kind of cool, the way I suggested that while my views might be stupid, everyone who disagreed with me would have children who thought the same way.
How much more self-aware do you want me to be? (Without entirely conceding my point)
Anyway, I honestly feel like I’ve got a better understanding of morality than most people (in that I don’t feel anyone has been able to demonstrate where I’m mistaken).
So… this is morality – it’s basically egoism – the fucking moral innovation is to realize that it’s in your own self interest to imagine that other people (weird shapes and sound that give you food) are the same as you. Other people cannot exist, in any meaningful sense, except as an extension of our own emotional state – THEY DON’T EXIST – there is just your feelings projected onto some sights and sounds which are imposed on you. But if you don’t consider others to be the same as you, your life sucks. That is the moral message, that is how everything ties together… also puddleglum.
> So… on one level.. yeah… maybe I’m just like those moronic atheist fellows. But at the same time… uh… I am your children.
Children of extremely tall parents tend to be not as extremely tall. Children of extremely smart parents tend to be not as extremely smart. Parent-offspring regression is hardly surprising to the SSC audience.
So I am inclined to be receptive to your thinking (see User Name), but do we really know that the “Sexually Frustrated Young Man” is a modern development?
Do we really know that the Japanese Herbivores are a new development, either? I mean, what numbers do we have to go on? What did the Great Authors say?
I would strongly suspect that young men have always had the horn – but also that what they do with it, and what the horn means to them, depends on the cultural context. Herbivores clearly aren’t an entirely modern development -monks, right – but I think intellectually/culturally its quite new to view the herbivores/monks/celibates as the lowest of the low, as opposed to the elite.
[And even from the perspective of a horn-master it makes sense to ‘view’ them as an elite… right? Get those other chumps to forego sex and you win! Which is exactly the best way to undermine anything good. My personal view is that i don’t give a flying fuck about evolution. I’m more interested in beauty – evolutionary theories can only provide explanations for my motivations, it can’t determine what they are.]
I suspect that our educational system might play a role, specifically the custom of splitting children into age groups. It emphasises the weirdness of high-IQ people among their age-peer group, while in a more natural setting they would simply hang out with slightly older people, which in turn perhaps could make them perceived higher-status within their own age group.
Also, in classroom everyone has to spend the same time, regardless of how well they progress. Imagine that children who learn their lessons faster could go sooner out to the playground… in other words, imagine that the intelligence would directly translate into something that everyone values. This would probably increase the status of intelligence itself.
I sometimes have a feeling that large parts of the school system are designed to remove the advantages of highly intelligent students (at least on elementary and high school levels, but it already comes to colleges, too).
It would most likely lead to resentment. The socially dominant don’t like it when the non-socially-dominant start getting the advantages over them.
You may as well say that if Jews became bankers in medieval times and got richer, and everyone values money, that would increase the status of Jews.
In some ways it did. Status isn’t uncomplicated.
I feel that “are positive qualities like attractiveness, health, longevity etc correlated with IQ across the entire population?” and “… with very high IQ?” are two very different questions with potentially very different answers. Most of your references seem to answer the former, while we are interested in the latter.
For example, I remember hearing that high IQ is correlated with all kinds of sensory processing disorders, from this blog even, though quick googling didn’t give me very trustworthy sources. Another obvious thing to check for would be epilepsy and migraines, if higher IQ is associated with higher reaction speed, that would be probable results of overclocking it.
I mean, the genetic load theory in particular could have an extremely nasty interpretation/consequence: what if the natural IQ of an “undamaged human” is as low as say 120? Then on one hand when you look at the entire population you’d see all sorts of correlations of IQ with other nice traits (both causative and not), because the 120+ part of your population is not statistically significant. On the other hand, removing the junk would get you only that far, so then you’d start flipping switches that move the brain outside the normal operating conditions (probably as a cumulative result of course) and hello.
“Optimization” implies “optimum.” From a bare-bones engineering perspective, there is no reason whatsoever to think that the optimum value of a process that we have never seen fluctuate above ~160 lies at 1000 or greater.
Like, you can indeed optimize a car for speed (known as drag racing), but you cannot in fact build a car that goes 2000 mph, especially by limiting yourself to the “genes” of car design present in consumer-grade cars.
If you conceptually broke the design of computer circuit boards into “genes” and then “bred” the optimal computer, you would run up against physical limits quickly, and your design might not be much better than the existing best design.
Brain-sized lumps of goo are probably notionally capable of something like IQ 1000 cognition, but this does not imply that we can construct such a brain from the palette of available human genes.
I think all this talk of “IQ 1000” is making communication unnecessarily difficult, though. It’s a meaningless statement. What do you expect a person with IQ 1000 to be able to do? What is their max score on Dual N-Back? How good is their episodic memory? Should they be able to instantly derive Calculus by watching a falling apple? Infer GR from a single frame of a bent blade of grass?
Yep! This is also a big part of why I’m skeptical of superintelligent AI.
The IQ 1000 is a measure of rarity, not of a cognitive ability. That is, it’s possible for a IQ infinity brain to exist, and it won’t even be a very smart one, it just should be smarter than the smartest possible human brain. The dumbest superhuman.
Theoretically, an IQ of 1000 is not impossible, but no existing IQ tests can give a corresponding score. So, as moridinamael just said, we don’t know what an “IQ of 1000” would actually look like, or how we would recognize it, in practice.
But an “IQ infinity brain” is logically impossible because infinity is not a number. Maybe “infinite intelligence” is possible, but you still have to define “intelligence” somehow.
“But an “IQ infinity brain” is logically impossible because infinity is not a number.”
I think the point of the IQ infinity claim was that we define IQ numbers by the distribution, with some fixed number of points per standard deviation. Suppose we knew the full distribution of scores on IQ tests for an infinitely large population–what the observed distribution would approach as population size increased without limit. Further suppose that there was a maximum to that distribution. It makes sense to say that anything above that maximum is an infinite number of standard deviations above the mean, hence an infinite IQ.
My impression is that a lot of the smartest people (or at least a lot of the people who because famous for intellectual accomplishment) a century or so ago were sickly as children. This meant they didn’t spend their time running around with other children and had to entertain themselves by learning and thinking.
Does this have any chance of fitting in with the idea that intelligence and health are correlated?
Or is some of this historical accident? At least some of the children who were treated as sickly had harmless heart murmurs.
Remember western Europe went through several purges in the last hundred years where several countries killed off nearly half of their breeding-age young men. If you were sickly and didn’t have to fight, that was a huge advantage.
Also, whatever the long-run relationship is between trait X and surviving long enough to reproduce gets thrown out the window a bit when it starts coming down to pure luck of the draw whether you happen to be in the proximity of an exploding mortar or not.
Hypothesis: superintelligence is Macbeth.
Also, bootstrapping our way to superintelligence does seem like a good way to feel less stupid.
Can someone please clear up how we’re defining “intelligent”, again?
“That which is measured by IQ tests, and which correlates pretty well with everything else people ever mean when they say ‘intelligent’ but if there’s any doubt we’ll fall back on the IQ score as a measurable variable”.
Maybe Bohr came from a long line of people who lucked out and got hit by unusually few cosmic rays?
Or got hit by them in just the right ways?
(Alternatively, this proves the Bohrs, somewhere back down the line, are obviously Millenarians from Anathem and can control things like that at the quantum level, re-running history at each interaction with a cosmic ray until it comes out how they like.
Well, or something like that. The book’s not exactly explicit on the details of how they do it.)
Trying to predict and design the phenotype you want using the results of gene association studies is probably the wrong approach. I think this may be part of what Meyer’s meant with the race car analogy. Achieving the complex phenotype you want (e.g., super intelligent) is going to be much more complicated than merely assembling the correct “parts list” in the genome. Because the most immediate cause of phenotype is a very complicated interaction of not only gene products, many of which are pleiotropic, but a whole bunch of other active bio-molecules interacting with each other inside the Proteome. Add to that the influences environment have in shaping phenotype, especially in humans, and the problem only becomes harder.
Also, raising a child who is much more intelligent than its parents is a hard problem in itself.
My assumption is that if 150 IQ parents raise a 300 IQ child, the result is a crippled person with 300 IQ, at least compared to someone with 300 IQ who grew up in a 300 IQ society. (Cite: various science fiction)
It might make sense to go for 300 IQ by gaining 20 IQ points every generation or two.
When I think of how hard it is to keep my normal children out of trouble … just imagine a kid twice as smart as me. How fast would they figure out the baby gate then?
Doctors are smarter than population average, and yet most doctors don’t invent new drugs or medical procedures. Are doctors smarter just because they are better at navigating the tribal politics of health care systems? Or maybe you are underestimating the amount of intelligence needed to be a successful hunter-gatherer?
Yes. This. Underestimating & wild conjectures about hunter-gatherer lifestyles are extremely common, and allow you to argue almost anything. They don’t help.
But the other thing that didn’t help was misreading “doctors” as “dolphins”. Although it is true that dolphins are quite smart and still fail to invent new drugs.
Are doctors smarter just because they are better at navigating the tribal politics of health care systems?
And now I want a comedy sketch with a faux-Attenborough voiceover as we observe white-coated bands of doctors roaming the savannah, encountering rival tribes (perhaps surgeons?) and otherwise coping with the day-to-day struggle of tribal survival 🙂
This should please you as much as it did me.
Er, no. There are lots of animals with “hunter-gatherer” approaches to survival. None of those lineages have had runaway intelligence explosions of the sort seen in the hominid lineage. Therefore, the intellectual demands to successfully survive pursuing hunter-gatherer living cannot plausibly explain the explosion in brain sizes of one specific lineage of hunter-gatherers in the last few million years. (That is, if brains were driven by survival pressure, we should have sapient raccoons.)
The same works on pretty much any other survival selection theory. If the explosion of brain size from the human-chimpanzee concestor to approximately 150,000 years ago was driven by survival selection, there should be lots and lots of other cases of similar explosions, as natural variability in brain size gets grabbed by natural selection.
But, if the driving pressure for brain size was an existing sexual selection pressure in a specific lineage, then we’ve got a plausible reason why it would only cause an explosion in such lineages. This is especially true if, all else being equal, increased brain size is generally anti-survival; then we have survival selection pressure generally preventing increased brain sizes in all lineages, regardless of sexual selection pressures for enlarged brains. All we need to find is a unique reason for survival selection to ease off in one specific lineage where there’s a sexual selection pressure for bigger brains, and we’ve explained the unique explosion in brain size.
As it happens, when proto-humans became bipedal, they seem to have lucked into the new, calorie-rich niche of persistence hunting, which none of their close relatives did. That would translate to substantially reduced survival pressure, which would provide the margin for a substantial increase in brain size . . . if, of course there was an existing sexual selection pressure for intelligence.
(Note here that if brains were driven by survival pressure, the reaction to reduced survival pressure would have, if anything, been a decline in average brain size, as smaller-brained individuals would have been more able to survive.)
So, the question is, what’s the sexual selection pressure that worked on the human ancestors with chimpanzee-sized brains to push them to grow bigger ones? That’s where the Machiavellian hypothesis comes in — social apes have politics that translate to sexual success, and more intelligence means more ability to manipulate those politics.
If you want to dismiss that answer, sure, you can. But you’re going to have to come up with something better than “big brains help survival” as an alternative hypothesis, because it’s ruled out before we even get to the question.
But they were generally not as successful as the early hominids, who managed to colonize a huge part of the planet even before becoming anatomically modern sapiens.
Moreover, other animal species with comparable body size tend to have fangs, claws, or just brute strength and speed. Humans are physically good at long-distance slow-speed running, and they are tall enough to see above the high grass of the African plains, but that’s it. And yet they managed to do well in quite different environments.
Sexual selection and group proto-politics also exist in other species, but they did not cause an intelligence explosion.
Sexual selection also exists in species other than peacocks, but it didn’t cause them all to have peacock tails.
That’s the whole point. Sexual selection is a force that creates traits unique for one species. Human-level intelligence is unique for humans. Thus the conclusion that it probably evolved through sexual selection.
However, the whole “Machiavellian hypothesis” seems to me like refusing to update. Like saying: ‘yeah, I admit that sexual selection was somehow involved, but I still insist that the true reason had to be something about killing and avoiding to be killed, because that’s the only selection mechanism I really accept.’
Just like the peacock’s tail is a random thing that pleases the peahen’s eye, it is possible that human intelligence was just a random thing that happened to please human senses… probably through speech.
It also pleases the human eye.
“but that’s it”
Humans are also very good at throwing things.
But the things that humans can easily throw aren’t very effective against the typical things that humans want to eat or that want to eat humans, unless they are specifically engineered to be so.
Trying to take down an African buffalo by throwing rocks and sticks at it seems like a perfect way to get gored or trampled. If you combine certain types of sticks with certain types of appropriately knapped stones to create a javelin then you have something that you could effectively throw at a large animal to kill it.
If you’re throwing rocks and sticks, you’re throwing them at things like rabbits, snakes, and roosting birds. (See Aboriginal use of Boomerangs for hunting) That or you’re using them to chase some other predator/scavenger off a carcass.
As for buffalo, that’s why we invented spears.
“Trying to take down an African buffalo by throwing rocks and sticks at it seems like a perfect way to get gored or trampled.”
Well, I wasn’t explicit about the point I was trying to make, so I can’t be too touchy about you refuting a point that I wasn’t trying to make, or even about you mocking me for the ridiculousness of the point that you concluded I was making. I’m not sure, though. why you would conclude I was trying to make a point about successfully throwing improvised projectiles to take down any big game at all. (Much less taking down today’s African big game that has coevolved with protohumans hunting them with specialized weapons for hundreds of thousands of years.)
The point I was trying to make was that the list of human specializations that I was responding to (that I quoted from) was incomplete. In particular, if long-distance running is on the list, I think throwing should be too. I don’t know how exactly how human throwing interacted with prehuman African buffalo predator-prey ecology. But I don’t see why my point depends on African buffaloes in the least. The general specialization for throwing *something* (not necessarily sticks and stones) at *something* (not necessarily buffalo) seems pretty unmistakable, without depending on any particular just-so story about how it came about.
@Hlynkacg
But then you’ll be eating rabbits, snakes and birds, assuming that the local lions and hyenas don’t take them away from you and maybe even eat you while they are at it.
Your smarter neighbors who have developed the spear, instead, eat buffalos and wear lion furs. Seems like intelligence gave them a fitness advantage.
Boomerangs aren’t random sticks, they require intelligence and culture. And Aboriginal live in environment where they are the apex predators and there are neither large game animals nor other large predators.
But then you’ll be eating rabbits, snakes and birds, assuming that the local lions and hyenas don’t take them away from you and maybe even eat you while they are at it.
Yes, absolutely.
Sport hunters targeting the biggest, most impressive-looking game they can find is a recent aberration. Actual hunter-gatherers mostly ate, and still mostly eat, rabbits, snakes, and birds. And chase off the hyenas by throwing things at them, which is usually enough to make a hyena or even lion decide to steal dinner from someone without a targeting computer behind their binocular vision.
If your tribe gets forced into an area where there are buffalo but no rabbits, sure, they’ll either develop buffalo-hunting techniques or die. Until fairly recently, evolutionarily-speaking, I believe they mostly died. Now, we fetishize those tribes as the hunter-gatherer ideal, and mostly ignore the vast majority of rabbit-eaters.
The entire reason that there aren’t any large game animals or large predators in Australia is that the Aborigines killed all of those large animals 40,000 years ago.
And such a broad colonization before achieving anatomically modern brain size proves that Scott and I are not “underestimating the amount of intelligence needed to be a successful hunter-gatherer”. H. erectus-sized brains will, as you point out, do just fine for survival; now you need to come up with a different mechanism for brain size increase past that point.
Which is why there’s a rest-of-the-paragraph after the bit you quoted.
I agree with Steven. Also, I am much less certain about “are doctors smarter just because they are better at navigating the tribal politics of health care systems?” than you might think.
I concede that intelligence may have a positional value in many cases. But it seems that intelligence also has a clear absolute value as evidenced by the correlation between average national IQ and national development.
Anyway, the fact that people who are smart enough to navigate the tribal politics of health care systems makes it possible for complex health care systems to exist. More generally, people smart enough to do complex politics make it possible for complex societies to exist. This seems to create a group fitness advantage even in addition of the individual fitness advantage that is selected for.
If Dunbar’s number hypothesis holds for humans, then the fact that human hunter-gatherer groups are typically larger than chimp troops is due to the higher human intelligence. Larger groups are generally more efficient, yielding a group fitness advantage.
Alternate Title: The Powergamer’s Guide to Genetic Engineering
It’s conceivable that some of the associations with intelligence you mentioned in Part II might help explain why these genes weren’t selected for in the ancestral environment. For instance if the personality traits that lead to criminality and short life expectancy in the modern age (Aggression, risk-taking behaviour, lack of empathy etc. etc.) were actually a fitness advantage in less organized societies.
You probably know this, but the crocodile-duck hybrid is mocking creationists.
My guess is that intelligence was more useful for a group than an individual in our evolutionary history. It’s good to have a smart guy in your tribe but more smart people lead to diminishing returns while energy requirements increased linearly. The optimal strategy would then be to have a few smart guys to do all the thinking stuff while energy is saved on average or below average individuals who could still contribute by doing the heavy lifting.
Btw, how many genetic studies are done on modern hunter-gatherer societies? Seems like the optimal point of reference for questions regarding early human history.
point
I recently read The Secret of Our Success, which I strongly recommend. At the risk of oversimplifying a very interesting book, and at the risk of having a hammer and seeing nails, I think part of what it has to say is that intelligence evolved not to gain status but to let us recognize and evaluate status. A hundred millennia ago we fortuitously slipped over a boundary where group knowhow exceeds what an individual can learn from first principles, after which there was a tremendous advantage to figuring out who to emulate.
Thence comes “prestige”, which is different from dominance, stemming from admiration of skill but feeding on itself as we learn to choose mentors in part by whom others seem to emulate. Thence come traditions that detoxify sketchy food, developed a little at a time as people evolved to emulate those who are healthy and long-lived, even though the culture that develops these traditions has no idea why they are following them. (We have frankly astonishing metabolisms compared to other animals — we do not naturally handle toxicity as well, and we have weak jaws, small teeth, shrunken stomachs, and short intestines, because we have offloaded the responsibility for food prep to our culture.)
This is all a bit of a digression from what you’re really talking about, except to suggest that what we have evolved is not necessarily what Hsu is trying to optimize — which might help explain why natural selection hasn’t already taken care of it.
http://biorxiv.org/content/biorxiv/early/2016/04/27/050534.full.pdf
I’ve wondered about this for a long time:
Brains take a lot of calories to keep running.
Does that mean I can lose weight by thinking hard?
Depends. There are two cases.
Case 1: Brains take an average of 20%, but can go as low as 19% if you relax or as high as 21% if you think really hard.
Case 2: Brains take an average of 20%, but can go as low as 10% if you relax or as high as 30% if you think really hard.
I have no idea which one is true
Probably the former, because thinking is only one of many functions brain does. (Maybe if you close your eyes while thinking hard you actually save energy.)
Probably Case 1. Look to “thinking hard sports” like chess or go. Anecdotally, I don’t believe top-level players have much increased calorie requirements during major tournaments. This thread references a paper that tried to estimate it in chess players, but I don’t currently have my Springer access to look at the paper, itself.
This article talks about these issues
http://www.scientificamerican.com/article/thinking-hard-calories/
energy consumption does not very much based on a cognitive task.
Thanks!
That answers my question once and for all.
A quick Google search once upon a time told me that no, increased brain energy consumption during intellectually taxing activities does not amount to anything significant in the grand scheme of weight loss.
Think in terms of the physical quantities involved — joules and respiratory rate (an important way through which metabolized body fat leaves the body) in motion vs. sedentary metabolic processes. An interesting, though obviously flawed analogy to get the point across is the energy consumption of a computer vs. a washing machine or ventilation-based heater. It still takes more energy to move mass around, and let’s also remember that a lot of challenging intellectual activities (like taking an exam) are sedentary.
Obligatory mention: my source is vague memory of some pop-sci articles helped by handy references from pidooma.com, but if you want a more technical explanation, this is an excellent opportunity for having some mental workout as you research brain metabolism while you’re testing your theory on whether mental workouts help with weight loss. Yo, dawg.
Scott, can you please anchor-link your footnotes? It’s pretty simple and it makes them much, much easier to read.
Are we really living in a “low-infection” time? I’d love to see a citation for this. On the one hand we’ve solved many serious illnesses with antibiotics and vaccines. On the other hand we live in densely populated cities in a globally connected world. Anyone with kids knows they get a new cold every month or two. Newly mutated flu strains originate every year and sweep around the world. I find it hard to believe hunter-gatherers faced the same constant exposure to new viruses.
I know it took you a month to write this post, but you could have taken the time to read Myer’s follow up or at least linked to it, which shows why Myers is so mean to Hsu.
Specifically this line:
Myers is completely dismissive to everything Hsu’s saying because he believes no one knows what intelligence is.
Further he rejects the idea that you can target complex behavioral traits that have many contributing factors using the same process you target single traits. Even when you have large amounts of genetic diversity to work with, you can’t target complex traits that way. Intelligence is more complex than say, Race Horse running speed, which, despite intensive breeding for, has stagnated for decades probably because additive genetic variance eventually exhausts itself in the face of strong selection.
Myers, elsewhere, takes the position that G is a Statistical Myth. He also notes the differences between animal intelligences is barely understood and that differences within species intelligence, is just as poorly understood.
These people cannot have a debate because Myers believes that Hsu isn’t asking the right questions. And he knows Hsu doesn’t have the right answers, because no one has the right answers to the right questions. Maybe P.Z. Myers is wrong, G exists, but that doesn’t undermine the fact that Hsu, as an outsider to the field, is asking the wrong questions. The fact that Hsu has detailed and intelligent answers to the wrong questions is just irrelevant. Myers, as an ‘expert’ in the field, at least knows what the questions are.
Further, Myer’s dismissal that he should spend his time with Social Work instead of pursuing this research is part of a much larger metalevel debate. Hsu’s ‘research’ is advocating for a direction in science that, those with any understanding of the difficulty of genetically modifying human beings to select for cognition will immediately see is deeply disturbing. Is Hsu aware of the ethical problems, or does he not care? Even if Hsu is correct on every point, Myers can still be right in trying to shut down the argument, because it’s a dangerous science.
There’s an episode of Sliders where Einstein wrote a letter saying nukes are impossible no one should pursue the research, and so no one did. I’m not saying Myers is Einstein, but a good case can be made that genetic modification of human intelligence is as problematic as nuclear weapons. It would not surprise me if this is the unstated consensus position within the field, in the same way Anthropologists sanction scientists that use their expertise to help militaries conquer countries, or Psychologists and Doctors sanction their members who help torture prisoners.
The thing is though, you don’t have to go that big to make a complaint against what Hsu’s doing.
If you go to the Chinese government and you tell them you can crank out 1000 IQ babies using techniques that the West considers “too unethical” to pursue, you can get a ton of money, while producing nothing of scientific import. Money that could go to say, genetically mapping endangered wildlife like the Panda, is instead going to swabbing ‘smart people’.
Whatever, look at these big chickens.
Myers specializes in the development of the zebrafish. He has no expertise in or experience with psychometrics, quantitative genetics, or genomics. The fact that he’s a geneticist doesn’t mean that he’s an expert on all the myriad topics that go under the rubric of genetics. From the mistakes he makes in his comments it’s easy to see that he does not have even amateur-level knowledge of many of the issues under discussion. In contrast, Hsu, while a physicist by training, has worked for years in this field. As to the ethics of genetic engineering, I think Hsu’s perspective is that the genie is or will be out of the bottle anyway, so it’s better that the technology is safe and legal and government-controlled rather than the opposite.
Intelligence is more complex than say, Race Horse running speed, which, despite intensive breeding for, has stagnated for decades probably because additive genetic variance eventually exhausts itself in the face of strong selection.
Not actually the case. It’s a social problem, not a tech one.
Animal breeding has resulted in extraordinarily leaps in desired outcomes – more milk, more meat in chicken and turkey breasts, specific colors, and so forth. However, horse racing has lagged considerably behind this for several reasons.
Firstly, racing horses are far less of a commodity than beef cattle – there is a great deal of aestetics and “branding” involved in selection. Secondly, they come into actual use (so their output can be measured) much slower than cattle. (We are ignoring the rapid gains that can be made in chickens because that’s just not going to be matched by larger, slower growing animals.) Thirdly – and more importantly – the measurements used to compare animals are not uniform.
And the measurements are not uniform because the racing PTB are not interested in creating a level playing field where a random horse with the right genes can be more valuable than one of Secretariat’s sons. And this is because the output of the horse racing industry is *not* “fast horses” but instead “big crowds at racing parks”. Focusing on the speed means losing the names that draw the crowds for at least a human generation – and horse racing as an entertainment industry is forever on the edge of failing.
What would a good measure of racing horse speed look like? And why don’t we have them now, when we measure all the races with clocks and such? It should be a simple matter to say that well, horse A’s offspring ran their races a little faster than horse B’s, and therefore someone who wanted a fast horse should choose horse A over horse B to be the mate of their broodmare.
But races are run on tracks that are highly individualized, under weather conditions that are even more so. In order to say that a horse who ran this race was faster than this other horse who ran a similar distance on a different day across the country, we would have to agree that these two tracks are the same, and give commonly agreed on modifiers for dry track vs wet, grass vs dirt, and other factors. That’s where the social bits come in. Racing standards are run by people who own race horses. No one is going to agree to standards that make their animals look less valuable. And without standardized measurements, we can’t compare two (or twenty, or two hundred) things and get repeatable valid results.
Cattle breeders (and chickens, and sheep, and all the others) have also faced these issues – and have overcome them. Cattle breeding metrics are constantly being modified as more information is learned (ie, another crop of calves sired by a particular bull goes to market, and their weight, gain, marbling, health and backfat are all measured and sent out to interested parties.) As the knowledge of heritability increases, more and more things are being measured and selected for – dairy cattle now have metrics for “longevity”, which is almost as hard to measure as intelligence, and is something dairymen have been trying to get a grip on for some time.
It all comes down to repeatable measurements – can you get them, do they give you information on what you are looking for, and will you act on the measurements you get?
If you’re not going to make a purchase decision, or a breeding decision, or a treatment decision, or a hiring decision based on the data you get, it’s of limited use.
Very interesting.
I’d add that horses seem to have been breeding themselves for speed long before scientific breeding got started a few hundred years ago; colts in a big field appear to spontaneously race each other to see who wins and presumably the fastest enjoyed reproductive advantages. So, horses a few hundred years ago were already close to their physical limits in speed, while chickens back then were optimized for something other than being hefty.
I mean, people enjoy watching and betting on horse races because, in part, the horses seem to be naturally into it. Running fast is part of a healthy horse’s mission statement.
On the other hand, being a giant chicken is kind of grotesque and unnatural and we’d prefer that it happen out of sight in chicken ranches.
Now this sounds kind of Aristotelian, but, then, Aristotle probably enjoyed a good horse race.
Eh. On the one hand, yes, absolutely, horses make a landscape more beautiful, but if you look at wild horses, they make growth/size compromises that are not reflected in domestic populations. ‘Wild type’ horses match the available feed levels, and tend to stagnate at about the size of Arab (or Welsh) horses. To get the elevated size and muscle mass of racing throughbreds – much less that of draft horses – the young horses (and their dams) have to be heavily supplemented with protein.
And yes young horses spontaneously race each other – but so do goat kids, wolf cubs, and a whole host of other animals. Many things feel the need for speed.
On the other hand – if one limits the amount of food available to a commercial broiler bird, it grows to about the same shape (but larger in size) as its undifferentiated forebearers. It is only with unlimited access to feed (and long periods of light, and properly balanced diets) that allow the maximum weight gain that gives the poor secondary effects (poor heart function, gait abnormalities). (And a modern layer won’t provide as many eggs without proper diet and light, either.)
(The modifications for modern broilers do include an increase in appetite, which motivates the bird to eat more than otherwise.)
In a way, this points back to the differentiation in horses for work/muscle mass – in the pre-combustion engine ages, horses were more differentiated from wild type in order to pull plows and wagons, not for speed.
And while chickens don’t naturally run in straight lines, the males do naturally fight like the dickens. However, it is seen as immorally cruel to select for this trait and then bet on the resulting conflict. *shrugs* People, whatchagonna do.
Go on then, make the case.
There’s an episode of Sliders where Einstein wrote a letter saying nukes are impossible no one should pursue the research, and so no one did. I’m not saying Myers is Einstein, but a good case can be made that genetic modification of human intelligence is as problematic as nuclear weapons.
And in “Last and First Men” (1930), Stapledon had a conspiracy of scientists shocked by the first use of nuclear weapons plant the “nuclear energy is impossible” meme so firmly that when the coal runs out in 6000 AD nobody can figure out how to build a reactor.
I didn’t buy it then, and I don’t buy it now. I am, for example, mildly curious as to how in the alternate Sliderverse, a letter by Albert Einstein persuaded the Nazis to abandon their nuclear weapons program.
If the story is that everyone but the Nazis abandoned nuclear weapons because Saint Einstein told them to, and the Nazis didn’t quite get a nuclear monopoly before they were conquered by conventional means, then yeah, everybody lucked out in that universe and that can happen. There’s still an obvious, ugly failure mode in that general strategy, though.
The posters on the AlternateHistory.com forum seem pretty confident that 1940s Germany didn’t have the resources to build a nuclear bomb under any circumstances, even if Hitler hadn’t sabotaged the nuclear weapons program through mismanagement like he did to so many other government programs.
Of course IOTL even the Manhattan Project wasn’t able to create a nuclear weapon until after the Nazis were defeated. So I’m not very confident in the Nazis being able to create one on their own even under the best imaginable circumstances.
The Nazis had by far the worst nuclear weapons program at the time, even behind both of the Japanese programs. (Yes, the Army and Navy both had their own.) They had made several mistakes in the basic physics, and weren’t getting the priorities or organizational support they needed. But the chances that the British, French, Soviets, and Japanese all decide to abandon their programs on Einstein’s say-so? Miniscule. His influence was important in starting the crash program we got, but even if he’d gone against it, it probably wouldn’t have delayed things more than 10 years. Maybe less.
My understanding is that a lot of top physicists argued that a hydrogen bomb was impossible, either because they wanted it to be or because there really were very hard problems to be solved. It got developed anyway.
We made our own missteps. Trinity used a backup design — until relatively late in the design process, work was proceeding under the assumption that gun-type weapons using plutonium would work with the technology of the time. They wouldn’t; with the isotope mix available for that element, spontaneous fission would make the device predetonate.
It’s impossible to say how much time this would have saved if it’d been realized earlier, but the Manhattan Project certainly didn’t take the most efficient path to a working weapon. On the other hand, the Nazis didn’t have anywhere close to the basic research capacity that the Allies did, even if their engineering was top-notch at the start of the war.
No time at all, actually. The Manhattan project was rather unique among technological trailblazers in that they basically threw everything at the problem, including at least three different methods of Uranium enrichment, and Plutonium at the same time. They used a massively redundant path to avoid missteps costing time. This is very expensive, but also effective.
Note that while Trinity was a plutonium implosion device, Little Boy was a uranium gun-type device, and was ready at essentially the same time. And that the arsenal was cleared out after Little Boy and Fat Man, and we would have had to wait for more fissile material to show up before we could have made more bombs. Even if they’d realized 6 months earlier about the spontaneous fission problem in plutonium (and I’m actually not sure your history is correct there, but I’d have to check references), it wouldn’t have made more than maybe a few weeks difference in terms of Trinity, and probably 0 difference at Nagasaki.
The posters on the AlternateHistory.com forum seem pretty confident that 1940s Germany didn’t have the resources to build a nuclear bomb under any circumstances
Is there some specific shortfall they have identified? Because the Nazis had plenty of uranium metal, and they had an industrialized economy roughly an order of magnitude larger than that of North Korea today, and, well, Boom. Boom Boom. Boom. No boom today. Boom tomorrow?
Nazi Germany historically had an underfunded, poorly managed nuclear arms program that made several critical technical errors early on and never recovered. North Korea was able to avoid most of the unforced errors by reading everyone else’s “don’t do that” reports and memoirs. But the idea that, if we were rerunning history, we could safely cede the Nazis a nuclear monopoly on the grounds that they would surely muck it up, seems to me unwise.
Likewise the idea that we should cede to any future CRISPR-equipped Nazis a monopoly on master-race-breeding programs. Either beat them at their own game, or make damn sure they never play that game. And I’m not comfortable with the level of internationally-enforced control over human reproduction that might be necessary for the latter plan.
I don’t think that nuclear weapons in the 1940’s were as much of an unstoppable war ender as we think they were.
Neither side in Europe exterminated enemy civilians as much as they could have done, fearing that the enemy would have done the same. Poison gas was never used.
Thus, if only one side of the war in Europe had obtained nukes and used them against civilians (as the US did to Japan), the other side would have pulled out all the stops and retaliated with poison gas against civilians in a way Japan could not have done. This means that nukes would have been less of an advantage.
If the Allies had obtained nukes earlier, the fear of poison gas retaliation, plus the great odds of achieving their goals regardless, may have been enough to deter them from using them on German cities.
If Germany had obtained nukes first, the fear of poison gas retaliation, plus the consideration that most of the Allied territory would have remained out of nuking reach so it wasn’t possible to win, may have been enough to deter them from using them on Allied cities.
In the first case – would the Germans really have had serious capability to use poison gas against Allied civilians? Their strategic bombing capability – as I understand it, that was the delivery method advocated by pre-war theorists who proposed use of poison gas against civilians – was limited in 1940, let alone by the mid to late war. While they had a more advanced poison gas research program, their ability to deliver it against civilian targets would have been the weak link.
In the second case, I find myself wondering what the tactical battlefield utility of the atomic weapons of the 1940s would have been.
Thus, if only one side of the war in Europe had obtained nukes and used them against civilians (as the US did to Japan)
What if one side had obtained nukes and used them against not-civilians (as the Luftwaffe did with conventional weapons in all of its victories)? Logistics targets, for example. The obvious targets for Germany’s first two A-bombs, if they had been available at the time, would have been Omaha and Gold Beach at Normandy, ideally about a week after the landings. Without the artificial harbors, and with ten Panzer divisions close at hand, “Normandy” rapidly becomes another name for “Dunkirk”.
Then the Liverpool dockyards, and if someone in London decides to interpret that as an attack on civilians, so what? Liverpool was the only place left in England capable of offloading large quantities of e.g. avgas from tankers, so there’s a bit of a kink in the plan where the Allies retaliate by sending B-17s full of mustard gas against German cities.
If Germany had obtained nukes first, the fear of poison gas retaliation, plus the consideration that most of the Allied territory would have remained out of nuking reach so it wasn’t possible to win, may have been enough to deter them from using them on Allied cities.
The most likely delivery systems for any Nazi nuclear weapon were the Vergeltungswaffen. Consider the implications of the name, and if your German is a bit rusty, I’ll give you a hint: “Vergeltung” does not mean anything remotely similar to “war-winning”.
Finally: While nuclear, chemical, biological, and radiological weapons are all linked, in law and public consciousness, as “Weapons of Mass Destruction”, one of these things is very much not like the others. Even the crudest nuclear weapons are vastly more lethal than any war gas, and more so if we are talking about old-fashioned phosgene and mustard gas against people with gas masks (and yes, in WWII, gas masks for civilians were a thing, just in case).
If nuclear weapons don’t guarantee immediate victory through shock and awe, mustard gas certainly won’t. And we know from experience, in WWII and elsewhere, what happens when you kill civilians through strategic bombardment in a less than immediately war-winning scale and fashion. It doesn’t encourage restraint. The prospect does not deter. It does result in a grim determination to kill and kill and kill every last one of those fucking bastards or until surrender unconditionally, because what they did is so far beyond the pale that nothing less is acceptable. The only way that ends with less than genocide or total defeat for one party, is after years of stalemate characterized by both sides using nuclear and chemical weapons against enemy civilians.
And if one side is smart enough to attack logistical targets first, the other side will rapidly lose the ability to attack enemy civilians with nukes and/or gas.
I would recommend a basic book on intelligence – I’ve heard Stuart Ritchie’s “Intelligence: All That Matters” is pretty good. Statements that g is a statistical myth or doesn’t exist haven’t been tenable for decades. This whole field is pretty well understood and if Myers as an outsider doesn’t understand it then he should learn.
By way of analogy, in software design, there are improvements in processing that involve tradeoffs, but now and then there are breakthroughs that are simply better all around. Once discovered, they are simply superior to older ways of doing things.
For example, I can recall about a quarter of a century ago reading in the newspaper about an Indian-American computer scientist who had come up with a better algorithm for routing long distance phone calls that was estimated to save the telecom industry something like a billion dollars per year.
My guess is that evolution blunders into some of these all-around better techniques now and then, although I would have no idea how to quantify this.
If you know beer, you know everything #2/OK.
If it hadn’t been for bee, we wouldn’t have a Copenhagen interpretation. Niels “superman” Bohr able ovattract top talenr to his Copenhagen institute thanks to funding from the Carlsberg brewery.
And William Gosset (a.k.a. Student) was employed by Guinness. But that doesn’t mean that you should drink several pints before posting on SSC.
Reminds me a bit of this kid in my AP classes who got a perfect SAT score, started self-studying multivariate calculus from a textbook after he took AP Calc BC, and was a member of the swim team. I just looked him up and apparently he’s had quite the academic career since then; Bio-X undergraduate fellow at Stanford, MD-PhD student at Harvard/MIT…
Some people just get all the good genes.
I know a similar girl. Popular young lady, nationally ranked Individual Events, and a competitive athlete as well.
More information would probably personally identify her, but she apparently has two doctorates and is working at Tesla.
And here I am, commenting on the Internet, pretending to be Lord Commander of the Night’s Watch….
The worst is that these people often seem to be very nice and well-adjusted.
Don’t worry, there’s intelligent and athletic maladjusted cynical assholes too. Probably more common because no one ever comments on them.
I suppose the consolation is that there are not a lot of them, hehehehe.
I did some Facebook-stalking this morning. Most of the “Smart” Kids from high school ended up in law firms or academia. Yuck.
The absence of spouses is particularly off-putting.
Of people I knew in university, I’d divide them into 5 kinds:
1. Across the board perfect. Smart, good looking, hard working, athletic, nice, etc.
2. Seems like 1, but speak with them for 5-10 minutes and they just seem off – where 1 seems naturally affable, 2 comes off like they read in a book somewhere that being affable is important, and they do it really aggressively.
3. Some combination of positive traits, but never all of them, and with major flaws of the kind that are unpleasant for other people. Where 2 is pretending really hard to be perfect, 3 is too busy getting coked up and mean/cheating/backstabbing.
4. Some combination of positive traits, but with major flaws of the kind that are unpleasant mostly for them. Most commonly some sort of depression/anxiety situation.
5. No dramatic positive or negative traits.
Among those who I can see becoming Important, 1-4 are represented. Among the future Unimportant (among whom I count myself) 3-5 are represented.
I am OK with the Chinese going first on this.
I agree that human experimentation makes me nervous on several dimensions.
In general, human medicine doesn’t advance all that fast, so I think there is time to think through the implications ahead of time if we get started now.
I’m not sure that the evidence from those correlational studies is as clear cut as you make it out to be. There is, after all, a difference between phenotype and genotype. For example, low-functioning autistics tend to have low IQs, so a study of this sort would probably find a negative correlation between IQ and autism. Yet we know from family studies that physicists, engineers, and mathematicians are more likely than the general population to have children with autism. That would seem to be strong evidence that at least some genes that increase intelligence and/or other qualities that make a person good at STEM fields also increase a person’s likelihood of being autistic.
The long list of legendary geniuses with serious mental problems, from John Nash to Nikola Tesla to Howard Hughes to Isaac Newton, just to name a few, is also enough to make me wary, regardless of what various correlational studies of the general population have found.
Finally, there is the general observation that humanity’s track record with genetic engineering is mixed at best, with multiple examples of animal breeders accidentally creating animals with traits that result in lifelong excruciating pain. The example of dairy cows that have been bred for maximum milk production suffering from mastitis seems especially relevant to this discussion, and like a pretty good match for Myers’s race car analogy. It also seems like the sort of thing that would have been missed by correlational studies like the ones referenced in this post, since I think that it is highly likely that if someone had done a similar study of cows in the wild, they would have found that milk production positively correlated with all sorts of measures of general health.
I would think that the genes that correlate with both STEM careers and autism are not intelligence genes but rather are related to particular personality types. In the study you linked to the only significant differences between STEM and literature students at Cambridge were autism (more common in relatives of STEM students) and manic depression (more common in relatives of literature students). As the literature students are smart, too, you could argue that manic depression genes increase intelligence, but nobody seems to be arguing that (and affective disorders are in fact negatively or not at all correlated with IQ in the general population). Similarly, without further evidence there’s no reason to believe that autism genes are IQ genes.
Hence why I said “intelligence and/or other qualities that make a person good at STEM fields.”
But for the record, college students that major in STEM fields have higher IQs on average than students of most other majors, including literature majors.
In answer to your question about how mutational load ever gets weeded out: Let’s adopt a model where every new individual gets a fixed number of new novel bad mutations each of which has a small negative effect. Let’s also say that the chances of any given individual reproducing are some fixed percentage if they’re above the fiftieth percentile for the population as a whole, and some lower fixed percentage below that threshold, netting out to the population being stable. The difference between these two percentages might be very small.
Now we get to discuss asymptotics. If the population starts out with no bad genes, they’ll get more each generation, with natural selection hardly making a dent in the amount of mutational load. Over time the effect of natural selection will get more pronounced though. Specifically, the amount of natural selection will be proportional to the standard deviation in the average mutational load. Since the standard deviation is based on the square root of the mutational load and the amount of mutation is fixed, there will always be some mutational load for which natural selection balances out mutation and the population stops getting sicker.
Note that this produces the interesting prediction that doubling the mutation rate or halving the advantage rate will quadruple the mutational load. If we make the assumption that each individual has 25 bad novel mutations and that an individual’s reproductive success is 25% genetic, that suggests that each individual has (25/.25)^2 bad mutations, or about 10,000. But if each individual has only 5 bad novel mutations and reproductive success is 50% genetic, it will be about (5/.5)^2, which is 100.
For the record, c-sections are pretty awful, and increase the risk of your uterus spontaneously bursting in subsequent labors. NOT FUN.
They’ve become super routine in China due to a combination of lazy doctors and the one-child policy: if you’re only planning on having one baby anyway the risk of c-section seems a bit lower, though still hardly ideal.
I think your argument fails in section II. You cite how high IQ is associated with late-life longevity, height, decreased crime, and lower rates of some diseases. (I think the attractiveness and physical fitness links are extremely weak or based on extremely limited studies, so I ignore those.)
The longevity gains and decreased diseases are likely two ways of looking at the same thing, and give little advantage unless the longer-lived high-IQ types continue to be productive or contribute to society. I’d like to see a study that shows otherwise. In the absence of that evidence, living a little longer (i.e. more octogenarians and up) may well be a societal disadvantage. Of course this longevity gives absolutely no reproductive advantage, and as a potential societal resource drain, it may be a net negative, despite the occasional long-lived sage.
Height advantages give taller men some advantages in many societies, but are a huge penalty for taller women. Do you know many really tall women? It’s a tough life. As a net for society, this is probably a wash. And I’m not sure there’s any evidence that the taller men, who are disproportionately chosen to be leaders, are really any better at the things they’re chosen for. Being taller than average is subtly intimidating on an unconscious level, but I’m not sure that matters in modern society. Moreover, in a society where everyone is getting taller, humans would quickly reach the point where increased height starts to have serious health consequences. (I don’t think anyone has called out that it appears that most of the extremely intelligent geniuses are male, so maybe in this genetic engineering we end up with a race of super-tall, super-smart men and average height, pretty smart women. Ugh.)
Decreased crime is where I think your argument really crashes. The genetic traits that correspond to increased crime (impulsivity and lack of long-term planning, sociopathy, resistance to authority, etc.), while mostly negative when it comes to criminal activity, are potentially very important to society and human development. People with these traits are often risk takers (in the positive sense), iconoclasts, and explorers. I’m not sure I want to live in a world were everyone always looks before they leap, and carefully considers the societal consequences of every action.
The counterargument to the list you made is to consider other traits that are associated with very high IQs. I’m just speculating here, but some of the traits I see in this group are perfectionism and excessive reliance on rationalism, decreased social skills, desire for isolation and solitary pursuits, and decreased interest in family (including having children). Really really smart people tend to be loners or social misfits, tend to be focused on esoterica (at least as far as the rest of society is concerned), and tend to be conservative when it comes to risk-taking. It also appears to me that the extremely high IQ folks tend to be humorless and judgmental towards those with less IQ, making consensus and cooperativity problematic. I realize these are cliches, not backed by any studies (I haven’t looked), but this is my experience. Gregarious, athletic Neils Bohr-types appear to be the exception that prove the rule.
And ultimately, the random matching of our existing genes has produced only a few Bohrs and Heisenbergs (that we know of) among the billions of humans born, and never (as far as we know) anyone with an IQ remotely near 300, whatever that means. This alone tells me that the negative loading of all these apparent IQ genes in combination is not viable for a functional ‘fit’ human.
Perhaps a better avenue is to really figure out what a 200-300 IQ means. Then look at the closest examples known, and consider their ‘fitness’ as models for what we want for ourselves, our children, our society. You write that we all prefer children with IQs of 150 to those of 100. I know this isn’t true of most people — you, like me, are very bright, and I can think of few worse tragedies than having a child who was much less intelligent then I am. People of average intelligence, however, don’t lust for super-smart children. But I’ll accept that society would be perhaps better off if we had more ‘somewhat smarter’ people in general. That said, I’m not at all sure we want a society where the average intellect is the current equivalent of a 150 IQ, much less a society where it is dramatically higher on the limited things measured by typical IQ tests.
I don’t know of any direct measurement of the competence of tall people, but more than half of the income benefit of height is due to tall people being smarter.
How strong is the correlation between height and head size?
“Gregarious, athletic Neils Bohr-types appear to be the exception that prove the rule.”
What do you think “exception that proves the rule” means?
I don’t know about athletic, but I have known a fair number of very smart people who got along well with other people.
“decreased interest in family (including having children). … It also appears to me that the extremely high IQ folks tend to be humorless” … .
When I was about ten, I met Sir R.A. Fisher and asked him what kind of a knight he was. His response was that he was a knight bachelor with five daughters and two sons. That appears to contradict both conjectures quoted above. As does my observation of other very high IQ people.
Isn’t IQ, and ancestral hominid intelligence in general, correlated with Neoteny?
So the tradeoff in paleolithic is this: genes to be extra smart means you’re going to lose a lot of fights to your peers, you won’t bring home as much meat, etc.
But in modern times, being able to fend for yourself, and reproduce by age 16 is no longer so important. Just as genes for good cardiovascular health after 65 were irrelevant back then.
So the “racecar” makes sense when we realize traits for a high quality of life in modern times – curiosity, longevity – are different than the traits which were advantageous for 100Kya to 10Kya.
Yeah it could be that one of the tradeoffs of many of the good genes is slower maturation. We humans already have ridiculously slow maturation, but maybe it would be better if were even slower.
Two thoughts, before I read any of the 511(!) comments.
Race car analogy – there are probably some genes which improve intelligence *and* other factors, and others which are tradeoffs. Some simple aerodynamics fixes can make cars both faster *and* more fuel-efficient, without cramping space or looks much. (And some changes can keep working on speed and efficiency, but start to cost interior space after a while.) Those genes which don’t have bad tradeoffs, or whose tradeoffs are limited, will spread, while those with bad tradeoffs will remain limited.
Intelligence for navigating social status hierarchy: Teenagers are generally as intelligent as adults but lack *knowledge* (which is mostly not genetic). Where your likely mate (for most of your children) was figured out before age 20, being too smart would be a handicap (because you don’t know the rules for exactly how much you can show off), but being a little smarter than the group average likely would be a benefit. As society gets more complex, there are sub-hierarchies for the more intelligent, rewarding higher intelligence, but also slowing reproduction, as such sub-hierarchies are tolerated because they’re more productive, and they’re more productive because they have more knowledge. Which means you get married later in those social groups.
1. Genetic engineer bigger birth canals
2. wait.
We don’t need bigger birth canals under current circumstances. We have modern obstetrics. Deaths in childbirth are very rare in developed societies.
Genetic modification absolutely will unleash higher intelligence eventually, just not yet.
“is intelligence, and more of it, really that good for human beings?”
Clearly large fractions of humanity and society want to increase intelligence with education, and that is enough to justify biological treatments. “Is it good”, is a less relevant question.
I have a few comments on the practical side of making smarter humans:
We are confident that genetics cause various forms of intelligence, but we can’t exactly see which genes and how they do it. I think it is wise to focus on other problems that are much more tractable and need to be solved and come back to this in another decade or so when our general understanding and toolsets are much more advanced.
Standard lab techniques for making GMO mice involve killing/discarding lots of individual mice and at least two full generations. It’s fine to raise a bunch of mice in captivity, force breeding, and kill/discard lots of them, but you can’t do that with humans. Even mad scientist types would get queasy about breeding and discarding lots of humans that didn’t go quite as planned. This will get better over time as tools advance, of course. This is an area worth investing into.
The more near term genetic modifications would be fixing well characterized diseases and less serious defects. Human genomes have tons of mistakes and undesirable transposons and remnants of ancient retro viruses called EVEs. Address the simpler, more tangible changes first, and then move on to the more complex ones, like general intelligence.
Also, exploratory changes focus on plants and animals first, not humans. Only after we are pretty sure we have a good treatment, will we apply to humans.
@ Massimo Heitor
Also, exploratory changes focus on plants and animals first, not humans. Only after we are pretty sure we have a good treatment, will we apply to humans.
A limiting factor there is, when the lab animals become smarter than the scientists.
My layperson’s understanding of “genes for intelligence” are from the GWAS for all the intelligent people who had the genetic testing done. Are there any studies the opposite way? That is how sure are we that these combination of genes blesses one with intelligence under normal circumstances?
I bring this up because all genes associated with diseases don’t always manifest as diseases. And many with such genetic profile may never end up being identified as the one with a particular disease. I am just wondering what would be the size of “control group” of this GWAS studies consider this possibility and other factors influencing the manifestation of intelligence. I guess my question is, did we find any normal people with this combination genes and saw why their IQ is 100 yet?
Also, what about intelligence and infectious disease resistance? At species level, intelligence seems to have exploded after antibiotics came in to scene. If we end up with antibiotic resistant common infectious diseases again, how good are these intelligence genes hold up to that?
Intelligence and height are always seem to be about men rather than women. Considering that more premature babies will result due to increasing averaging intelligence (what with big head, birth canal issues), more time will be required to be bringing them up to speed to normal communication levels. Are really intelligent women interested in this sort of thing (i.e., spending enormous amount of their life bringing up babies)?
Bringing up young children isn’t an overly intelligent job and gets more and more boring and frustrating as the more intelligent you are and used to communications at your intelligence level.
It seems to me, highly intelligent men could be good at providing resources but highly intelligent women would be bored out of their head in early childhood of their children and avoid having more of them. There must be some average intelligence women around whose delight in bringing up babies (particularly toddlers) exceeds their need for intelligent communications (wit, humor, maths, whatever…) for genetic success. This combination of intelligent fathers and average intelligence mothers seems to be most productive for genetic success rather than both intelligent parents either bored with semi-coherent toddlers or delegating their child rearing. This should damp the natural selection for high intelligence?
Not that delegating child rearing is bad, just that there is some need for not highly intelligent women for handling semi-coherent toddlers that are unrelated to them. Robotic advancement may help some, but human babies still seem to prefer human faces.
From historical examples, I haven’t known highly intelligent men handling the early childhood duties well.
May be I am missing something regarding studies on all these issues?
Edit: wow, rather a long comment,
TL:DR; main issues:
1. pathways for manifestation of intelligence
2. infectious disease resistance and intelligence
3. child rearing and IQ
It seems unlikely for most of human history that the joy from bringing up children, and women’s desire not to have more of them, played much of a role for most of the population. And… maths? Seriously?
I’ve been thinking a little more about this genetic super-intelligence project. I’m not sure what the goal is. When I was younger, I would have been instinctively all for it, but now I wonder. Why do really smart people think more intelligence, beyond the level of even ‘normal’ genius, is desirable, especially for one’s own children? And why are so many people so sure the potential consequences will be manageable? On that second point, a couple of natural experiments seem to have already occurred, one involving math and engineering prowess, and the other involving creativity.
There appears to be a strong connection between autism-spectrum disorders and math/engineering prowess. This is a cliche, but having been a psychiatrist in the south bay of Los Angeles years ago, and having treated scores of engineers, as well as working with programmers and computer scientists now, this seems real. What if the price of boosting your children’s math and problem-solving skills is to load them up with autism-spectrum genes?
The second natural experiment involves exceptional artistic creativity. A substantial portion of great artists and writers have suffered from bipolar disorder, and a fair number from debilitating depressive disorders. This linkage also appears to me to be pretty solid. Obviously, most people with mood disorders do not become important artists or writers, just as most people with autism don’t become successful engineers or programmers. But there does seem to be some substantial general (in terms of populations) trade off between certain kinds of raw intelligence (math and engineering) and social intelligence, as well as a trade off between extreme applied creativity, and mood stability.
This probably isn’t enough for a major program, but being intelligent is fun as a well as useful. If (probably as some sort of utilitarians) we wanted to increase the capacity for complex fun (why complex? because *I* like it), what would we do?
This is the point I was trying to make above; sure, intelligence is connected with a host of other things. But what I meant by “optimising for physicists” is that seems to be the kind of intelligence that is meant when people in general talk about “high IQ” – that kind of mathematical ability. If, say, China does institute a genetic engineering for genius babies programme, the kinds of tests they will use to determine if the baby genius is on track will be ones for mathematical/STEM abilities.
Nobody is going to give a rat’s ass about “This kid could be the next Mozart or Shakespeare”. They’ll be looking for the next Einstein, Bohr or von Neumann. If baby Einstein II is also tall, handsome, athletic, can play three musical instruments and speak six languages, that’s a bonus. If baby Einstein II is short, tonedeaf and monolingual, it won’t matter so long as they can be a genius physicist, engineer, software programmer, etc.
That’s what I meant by “a world of physicists”. And if we’re lucky and intelligence piggybacks on the traits for height, health, longevity, good hand-eye co-ordination, etc. then we’ll have that better world of improved humans.
But at the kind of levels (IQ 200-300) they’re dreaming about, we don’t know that. And we won’t know, until we’ve embarked on a programme of genesplicing embryos, implanting them in surrogates, and aborting the ones that have the identifiable bad traits that genetic counselling of today picks up. But it’s when the babies have been born and are developing to at least toddlerhood that we’ll know if there is a correlation between increased intelligence and likelihood of personality disorders – and if we’ve fucked them up, it’ll be too late then to go “Oops!”.
Unless we’re going to have a nice, neat programme of euthanasia running alongside to take care of the mistakes, and maybe we will. Why not? It’ll be merciful to put them out of their misery, right?
Yes, I agree that in this discussion most people immediately started talking about the theoretical possibilities, without agreeing on what kind of intelligence is sought, and what the goals would be (for individuals, and for society). The original article, as you say, is all about advancing physics theories, as if that is a worthy goal at the exclusion of any other considerations. That’s a very very narrow target, and not necessarily one that has definite benefits for either society or the individuals.
I’d like more discussion of what the real goal is. Would it be OK for parents to not opt for genetic physics mastery, but instead opt for genetic mods that lead to extreme height, because they want a basketball star, or extreme attractiveness, because they want the most beautiful child? I know such parents who would go that route in a heartbeat.
I’m just rambling out loud, but we also seemed to have completely skipped over the discussion of what this would mean for the human beings that we would be bringing into existence. If the project were successful, these people would be, by definition, freaks of nature. When I’m around people who are a few standard deviations below my IQ, I feel pretty lonely and disenfranchised. Imagine a world where virtually everyone, including your parents, were idiots compared to you.
“Would it be OK for parents to not opt for genetic physics mastery, but instead opt for genetic mods that lead to extreme height, because they want a basketball star, or extreme attractiveness, because they want the most beautiful child?”
No and yes. The first is a rat race (with negative side effects), the second isn’t.
“Imagine a world where virtually everyone, including your parents, were idiots compared to you.”
None of your peers are modified?
Not sure why one example of modifying for one type of physical extreme is a rat race, and the other is not. There are currently far more negative effects from people trying to become more attractive via conventional means than there are for people trying to enhance their athleticism.
I assume that this genetic modification thing will be expensive and unusual, so there would likely be fewer peers. But perhaps not. Then there’s the question of ‘social connectedness.’ From my experience, people at the highest end of the IQ spectrum are less socially connected. As I wrote above, they tend to be isolating loners. They live in the world of ideas, not in the world of friends and family. So even if there are many hyper-intelligent peers, I suspect for most it would still be a lonely, disconnected life.
Again, this is not referring to helping people with what is now fairly average intelligence bumping up a standard deviation or two in intellect. This project is about moving people into a range that we don’t even seem able to define, but would be standard deviations above the greatest known geniuses who have ever lived, people who per the original article would be far far beyond the intellect of Newton and Einstein.
“Not sure why one example of modifying for one type of physical extreme is a rat race, and the other is not.”
Because the advantage of height in basketball is entirely positional.
“I assume that this genetic modification thing will be expensive and unusual, so there would likely be fewer peers.”
Why? Rich people aren’t going to use a completely untested procedure and it isn’t clear why this would be substantially more expensive than IVF.
“Imagine a world where virtually everyone, including your parents, were idiots compared to you.”
I think that assumes a faster rate of change than we would expect. As more and more of the genetics of intelligence is discovered, the level of the modified offspring gradually rises. Since there is a distribution of intelligence already for other reasons, the improved person of generation X is the peer of (say) the top ten percent of generation X-1.
An increase of a standard deviation per generation would be pretty fast in terms of its effect on society. I expect my IQ is at least one standard deviation higher than that of most people I interact with, and they don’t come across to me as idiots, just as people who are somewhat less good at certain sorts of things than I am.
No, the identification of genes affecting IQ will be pretty much all at once, just like the identification of genes affecting height was.
What do people do with post-high school education? There are many forms of intelligence, of course. But ultimately, it’s clearly something people want and covet.
As I just wrote above, this proposed project doesn’t refer helping people become more capable of post HS education, or even post-graduate education. Yes, that’s what many people (especially the well-educated) want. This proposed project is completely different. Do people really wish they had intellect that was far far beyond what the greatest geniuses have every displayed?
If the project’s goal is to find a way to bump up the average intellect by a standard deviations from the current average, and perhaps come up with a higher percentage of people 3 or 4 standard deviations above the current average, then that’s a different subject.
I think its possible (and I think likely) to see an overall correlation between IQ and many other positive attributes, just as pointed out by Scott in this article, but then for things to go in the opposite direction at the upper end of the scale. Perhaps IQ related deficiencies and mutation loading are responsible for the variation between 80 and 120, but to reach higher levels you start having to make trade-offs with other abilities, such as social ability, disease resistance or whatever else.
I think the shape of the human population fits with this trade-off model. If a trait is super-effective, you’d imagine evolution would be pushing the entire population in that direction and culling harmful alleles. The population would display a very even distribution (think gazelles all being good at running). The kind of high variability that you get in the human population (at the upper ends of the scale) seem to me to be indicative of population making trade-offs with many other important traits. Trait groupings might even form coherent strategies that mean the effectiveness is very contextual too (moderately higher intelligence might not be that useful to a super-athlete, who should just focus on the physical realm, for example).
One thing in all this discussion that I find quite suprising is that in a scientific/rationalist (SSC/LW/etc.) subculture in which Chesterton fences are a popular idea, there is a strong willingness to commence the rewriting of our own fundamental fabric. A couple of loose correlations about intelligence and related traits doesn’t remotely approach understanding of how 1000s of alleles influence the most complex machine known to humanity (humans). There is plenty of reason to believe it our understanding is still quite primitive. And understanding an individual doesn’t equate to understanding the individuals interaction with society and culture. We’re totally lacking a discussion of the social aspects of this sort of work at the moment – for example a start would be considering game theory and what might happen when there is a differential of intelligence-seeking amongst power-seeking individuals compared with non-power-seeking individuals.
I fully support the eradication of unambiguously horrific genetic diseases using genetic technology, but altering the germline based on vague notions of the improvement of humanity is basically bulldozing one of the few Schelling points we have. Suppose the race to design the most intelligent child begins – what happens if we start to find alleles that assist intelligence, but result in negative externalities (like mild psychopathy). If intelligence really is super-effective, I wonder if virtually anybody would survive the endgame of a competitive escalation, at least if we don’t thoroughly think through the coordination problems and safety mechanisms first. I’m not saying disaster is certain, but I’m really surprised there isn’t more caution when the paths being discussed are so fundamental.
I also think its worth keeping in mind the fact that perfecting humanity as if they are instrumental to some other object is fallacious – humans are the ends, not the means, and intelligence is instrumental to us, not the other way around.
Propagation of DNA sequences is the end. Humans are the means.
I disagree, propagation is not the end, its the means of survival. And as human DNA is an inseparable, defining part of humans, it’s not accurate to characterize them as means in the context of the statement I was making.
Realistically:
1. Get gene editing techniques working, safely, and reliably. We aren’t there yet. If you saw how transgenic mice are made, you would understand this isn’t remotely ready for human use.
2. Offer treatments for obvious, well characterized horrific diseases like Tay Sachs.
3. Offer treatments to remove less serious diseases like Sickle Cell. Sure, there’s some heterozygous advantage if you live in the third world and are at risk of dying of malaria. And first-worlders can live reasonably normal lives with homozygous sickle cell. If I had the choice for myself and offspring, I would eliminate it.
4. Offer treatments for other minor well characterized genetic problems. Humans have lots of genetic defects that cause small problems. If I had the choice for myself and offspring, I would eliminate these.
5. Offer treatments for enhancements. For example, make this part of my brain larger. We are knowing some genes that can affect brain nucleus size.
I would say that complete understanding of the genetic basis for general intelligence is too far off and is not a tractable problem to be tackled in the near term.
ctrl + f “epigenetics”
Phrase not found
…
That’s a good test to use for checking if an article on genetics is worth reading. If there’s lots of babble about epigenetics, especially of the neo-Lamarckian variety, the article can be safely skipped as nonsense. I skimmed the study you linked to and it seems to be another demonstration of the multiple comparisons problem.
Pretty ironic that Hsu uses modern broiler chickens as an argument in favor of genetic engineering for intelligence, since they are a perfect example of the “race car” objection. Broilers bred purely for size are well known for having serious heart and skeleton problems and in many cases are not even able to walk properly.
Scott’s arguments about the correlation between IQ and other positive traits are not an answer to to this objection. Unselected broiler chickens (from 50-100 years ago say) probably also showed positive correlations between size and other desirable traits, but this predicts nothing about what can happen when a trait is pushed very far beyond its natural range of variation.
Anyone who has spent much time around purebred show dogs or thoroughbred racing horses can tell you plenty about how breeding programs that optimize for certain specific traits can lead to animals that have all sorts of wacky health problems. I don’t know if the same sort of thing would happen in in Steve Hsu’s hypothetical superhumans, but neither does anyone else.
I haven’t noticed any discussion here of Heinlein’s ingenious idea for what I think of as libertarian eugenics–let couples choose, from among the children they could have, which ones they do have. Do it by selecting separately on egg and sperm, combining the egg that has the mother’s musical ability and doesn’t have her poor circulation with the sperm that has the father’s good memory and doesn’t have his bad heart. That avoids a lot of the problems with other versions of eugenics, but could produce quite rapid improvement.
How do you tell what genes are in a sperm or egg without destroying it? You get the full genetics from any ordinary cell. Produce egg or sperm in vitro, destructively analyze the bodies thrown off in the process without touching egg or sperm, and subtract.
From Beyond This Horizon, one of Heinlein’s less successful novels but with a number of interesting ideas in it.
The correlations between IQ and various other “positive” indices raise a paradox: why does IQ exhibit additive heritability? As you notice, there do not seem to be tradeoffs between IQ and other traits; and IQ has probably undergone extensive positive selection.
One possibility is that IQ = mutation load, and variation is the result of mutation-selection balance. Another is that the positive manifold is really not so extensive as most psychologists think, so IQ-type cognitive abilities were not important to success until very recently.