Threading the Needle

The problem with the idea of an early, pre-Amerindian settlement of the Americas is that ( by hypothesis, and some evidence ) it succeeded, but ( from known evidence) it just barely succeeded, at best. Think like an epidemiologist ( they’re not all stupid ) – once humans managed to get past the ice, they must have had a growth factor greater than 1.0 per generation – but it seems that it can’t have been a lot larger than that, because if they had averaged, say, 3 surviving kids per generation ( r = 1.5) , their population would have exploded, filling up all the habitable territories south of the glaciers in less than 2000 years.

(1.5)^40 multiplies the original population by a factor greater than ten million !

A saturated hunter-gatherer population inhabiting millions of square miles leaves a fair number of artifacts and skeletons per millennium – but we haven’t found much. We have, so far, found no skeletons that old. I don’t think we have a lot of totally convincing artifacts, although I’m no expert at distinguishing artifacts from geofacts. ( But these were modern humans – how crude do we expect their artifacts to be?)

For-sure footprints we’ve got, and intriguing genetic data.

A priori, I would expect hunter-gatherers entering uninhabited America to have done pretty well, and have high population growth rates, especially after they become more familiar with the local ecology. There is good reason to think that early Amerindians did: Bayesian skyline analysis of their mtDNA indicates fast population growth. They were expert hunters before they ever arrived, and once they got rolling, they seem to have wiped out the megafauna quite rapidly.

But the Precursors do not seem to have become numerous, and did not cause a wave of extinctions ( as far as I know. check giant turtles.). What might have limited their biological success?

Maybe they didn’t have atlatls. The Amerindians certainly did.

Maybe they arrived as fishermen and didn’t have many hunting skills. Those could have been developed, but not instantaneously. An analogy: early Amerindians visited some West Coast islands and must have had boats. But after they crossed the continent and reached the Gulf of Mexico, they had lost that technology and took several thousand years to re-develop it and settle the Caribbean. Along this line, coastal fishing settlements back near the Glacial Maximum would all be under water today.

Maybe they fought among themselves to an unusual degree. I don’t really believe in this, am just throwing out notions.

Maybe their technology and skills set only worked in a limited set of situations, so that they could only successfully colonize certain niches. Neanderthals, for example, don’t seem to have flourished in plains, but instead in hilly country. On the other hand, we don’t tend to think of modern human having such limitations.

One can imagine some kind of infectious disease that made large areas uninhabitable. With the low human population density, most likely a zoonosis, perhaps carried by some component of the megafauna – which would also explain why it disappeared.

This entry was posted in Uncategorized. Bookmark the permalink.

84 Responses to Threading the Needle

  1. Random events occur in clusters. Every once in while you roll 20 sixes in a row. Sometimes a person or two makes it into a weird spot.

  2. To throw out another notion which may account for their downfall, perhaps they developed very silly ideas.

  3. Smithie says:

    It will all begin when they dig up a woolly in the permafrost which seems to have myxomatosis.

  4. TWS says:

    Small, not overly bright, no stone tools, not even shoes. The Senetilese seem like nice guys. If the precursors were like the Senetilese they would remain in tiny isolationist bands with tech that doesn’t rise to neanderthal levels of effectiveness for dealing with megafauna. Coastal gleaning is considered starvation strategy by most tribes I know. They have the same view of it as Joe Average does eating out of a dumpster.

    BTW I’m still pulling for my friend’s tribal legends of ‘stick indians’ small, unable to speak properly, can only whistle, have no shoes, nothing but wood tools (thus stick Indians) cannot hunt so they are relegated to coastal gleaning and vigorously defend that niche, (between the salt water and the outside edge of the driftwood). They also defend certain berry patches. They throw tiny rocks and try to lure women and children to abduct them.

    • Philip says:

      I’m not familiar with those legends but they sound interesting. Is there something good available to read on them?

      • TWS says:

        I don’t know about books. I heard those stories growing up as a kid. I spent most of my life there but I don’t think I read any books about them. My grandfather worked there, I did then my son. Heard a lot of stories from friends. There’s a white guy who has done most of the language preservation. I can’t think of his real name just his nickname. If my son can think of it I’ll post it.

    • Smithie says:

      I wonder whether there is any archaeological evidence that Amerinds wore shoes, in North America (I mean, beyond historical accounts.)

      I know that at least one of the Inca mummies has shoes. Maybe, the Atacama Desert would be an interesting place to look for signs of earlier inhabitants.

      • TWS says:

        Well I used to get moccasins for Christmas. That might be a clue.

        • Smithie says:

          I don’t doubt that Amerinds in NA had shoes in pre-Colombian times. What I’m saying is, I think it is rather specious to argue, as some do, that these Pseudo-Andamese couldn’t have had shoes, since all we have is barefoot tracks. Untanned leather would not survive a few ten-thousand years.

          And the reality is that most people did not wear shoes regularly until relatively recent times. I have heard tell that there were a few families in Northern England who still didn’t wear them into around 1940. In any event, people walking in mud, may have taken them off.

  5. dave chamberlin says:

    If a very small number of people made an impossibility long trip across the Pacific Ocean then it is quite possible that the following generations were very inbred with a variety of physical problems and mental deficiencies. It bothers me that they were competent enough to make the nearly impossible journey across the Pacific but were thereafter incapable of making stone tools or exploiting the resources of two unpopulated continents full of delicious beasts that would not consider humans as dangerous.

    • Karl Zimmerman says:

      This is my guess as well. I mean, excessive inbreeding would not only result in a large number of genetic diseases and lower intelligence, but also reduced fertility. Not to mention if it was a very small number of individuals, it’s possible that (if for example, no adult men made it across) that a lot of cultural/technological knowhow would also be lost.

      I mean, we know the Neandertals themselves were very inbred by modern standards – likely because low population densities and repeated glacial bottlenecks reduced their effective population size again and again. I suspect this is one of the main reasons that they were replaced – and that African hominid populations repeatedly replaced non-African hominids. A stable/large population base in Africa meant the stay-behinds didn’t end up inbred, and would eventually lap those who expanded outward in terms of fitness. Until the last migration out at least.

  6. A more thorough take on what I have been thinking when listening to the people Razib is interviewing. “For-sure footprints,” and some weird Amazonian DNA that is related to the Andamanese. And now, some Mexican caves with pre-Clovis stuff, and over all, the nagging thought of “the coastline was in a different place.”

    Yet it didn’t take. The hung on enough to hang on, in a highly exploitable environment (as evidenced by the later Beringian migrations). Bruce Bradley points out that a lot of archaeologists write authoritatively about hunting who have never caught a rabbit themselves, but I think even academic anthropologists could have caught on if they could only make it through the first couple of generations. Yet these pre-Clovis folks didn’t. I suppose mere survival for a few thousand years is something, even if they did not thrive or leave more than a whisper of descendants.

    If I can make it there
    I’m gonna make it – anywhere,
    It’s up to you! New World,
    New Woooorld!

  7. Smithie says:

    Maybe, they just thought that giant sloths, glyptodonts, etc. were really cool and wanted to preserve them for future generations.

    • engleberg says:

      Maybe they were pets for giant sloths. Killed when they got too numerous, but otherwise cute and useful for massaging a tired sloth’s delicates.

      • Smithie says:

        Wouldn’t that be something if the bigger sloths were much more quick-witted than the smaller ones that survive today? At least some of them were omnivores, so it makes sense, in a way.

        Maybe, when the land bridge opened up, the giant sloths thought to take over the Old World, and turn us all into their slaves, but were stopped by the atlatl-wielders coming in the other direction.

    • jb says:

      Exactly! Unlike the irresponsible Amerinds who came after them, the Precursors were true stewards of the environment who deliberately kept their footprint small to avoid putting pressure on irreplaceable charismatic megafauna. I’ll bet they were vegans too!

      • j says:

        They grasped that they were a ticking “population bomb” and voluntarily extinguished themselves. They deserve statues on the Harvard campus.

  8. “The house the Blakeney’s built” + 10,000+ yrs

  9. Tooldandtired says:

    “But the Precursors do not seem to have become numerous, and did not cause a wave of extinctions ( as far as I know. check giant turtles.). What might have limited their biological success?

    Maybe they didn’t have atlatls.”

    Or maybe they did? And preferred to use them on their competitors instead of their game?

  10. syonredux says:

    Maybe they were just stupider than the Amerindians….

  11. syonredux says:

    Maybe they were just stupider than the Amerindians…

  12. capra internetensis says:

    IIRC it was Turner who suggested domestic dogs, to stop the hyaenas from eating you.

    • Jason says:

      I have a hard time believing that the hyenas were really much of a problem for the mammoth hunting ANE or their robust predecessors.

      • Anonymous says:

        I don’t know the answer to that, but I have had hyenas messing around outside my tent in the middle of nowhere and it was disconcerting.

  13. ḱwṓ says:

    I think you’ve conjectured before that perhaps the First* First Peoples had a mostly coastal technological toolkit, appropriate for living along shores and foraging shellfish/seafood, but not hunting megafauna.

    I guess my question is, how long should we expect it to take humans to re-develop that sort of toolkit ab initio? These were still behaviourally modern humans, and Australian indigenous peoples certainly seem quite able to hunt native megafauna (and did a number on the giant wallabies, giant venomous monitor lizards, marsupial lions, etc. etc..) pretty well- which is reasonable evidence that other Australasian-descendant populations should also be able do it pretty well. Of course that cultural toolkit would probably would have been lost as they moved along the coasts, but the length of time spent not regaining it is somewhat surprising.

    Feels like a missing puzzle piece here. Maybe some inbreeding from a very small settler population? But that should sort itself out over pretty short timescales, relative to “thousands of years” at least.

    It is certainly possible for populations or cultures to sit in a rut for a really long time and not move out of it, even when more productive niches are nearby- but I find it kind of surprising across an entire continent and many millennia, when you look at what BMH did in almost any other context, including groups who were cousins of the Precursors. At a minimum here, we’re looking at five thousand years, and probably much longer (just on balance of probabilities it seems kind of unlikely those footprints were the very first arrivals).

    Long time spent not re-inventing a woomera.

    • jb says:

      The Tasmanians supposedly went backwards after they got to Tasmania. They didn’t have a whole hemisphere to play with though…

      • dave chamberlin says:

        The Tasmanians looked a lot more archaic than the mainland aborigines in their facial features and I mean a lot more. They did not allow genetic testing of the last purebred Tasmanians, so we don’t know if they were a higher percentage Denisovan, but I would be surprised if they weren’t. They were isolated on Tasmania for 15,000 years so it is possible that while the mainland aborigines became a lower percentage Denisovan through interbreeding they remained the same.

        • I believe is it established that most PN-speaking Aboriginal groups are largely descended from the original settlers 40kyo+, and while it is hypothesised today that PN languages spread following the opening of Bass Strait, genetically I would expect Tasmanians are a subset of the PN-speaking mainland population (which seems to have had a reasonable amount of turnover, including east-west until relatively recently, for a long time). %denisovan more likely stabilised a long time ago.

          Many gaps in Australian genetic analyses however.

    • dave chamberlin says:

      You are right inbreeding usually does sort itself out via survival of the fittest. But key genes critical to high intelligence could be lost early on with very small population size and when they are gone the surviving population is a lot less sharp than those first arrivers who presumably fashioned themselves a pretty good boat.

      • dave chamberlin says:

        To state that more clearly a lot of genetic variation would be lost in a severe population bottle neck. It is reasonable to assume that resulting population could take a big hit on their intelligence. That would account for why this modern population of humans did not exploit the new world of South and North America and why their population stayed very small. Just trying to come up with a plausible scenario.

        • j says:

          Intelligence is not so important. Some African populations seem to be in the 50-60 range and survival was never a problem for them. Before the Bantu arrived, they owned
          the whole of Africa.

      • tickertape mummy says:

        Is it not entirely impossible that in a moment of sheer luck, the variance they lost snipped out all genes associated with lower IQ? Very unlikely, but perhaps they became genius anti-natalist nihilists and transcended this plane through meditation. We should respect their last wishes and vacate the continent.

  14. dearieme says:

    Have we got the wrong end of the stick here? Maybe lots of early modern human groups dispersed and died out, or damned nearly, and we just haven’t yet appreciated the frequency with which this has happened over the millennia. Maybe it’s nothing out of the ordinary except that we happen to have discovered two little lines of evidence that these poor souls once existed.

  15. Glengarry says:

    Unrelated, a question about “regression to the mean”. This seems sometimes to be interpreted as intelligence (say) always regressing to the population mean in successive generations of offspring.

    But is this true when we know more? Let us say two persons of known intelligence mate. Then it seems the offspring intelligence can be predicted by a random variable centered at about the average of the two, with some standard deviation (and assuming there is no ceiling). If this is the case, there is also a possible upside for the offspring and no systematic regression.

    Only if the high scorer’s bloodline mates with randoms drawn from a large population should intelligence also regress over time, because the randoms have average intelligence, which drags down the center point of the random variable.

    Wise readers, is this reasoning correct?

    • saintonge235 says:

      Not quite. Yes, an intelligent couple may have more intelligent children, but the odds are against it. Imagine two people playing poker, both with good hands, and you select five cards randomly from the two hands. It’s not likely the resultant hand is higher than both of the two source hands.

      If you visualize the Gaussian distribution, the two people in your example are already over on the right side. The offspring genomes that will result in higher intelligence are even further to the right on the curve, while the lower intelligence combos are to the left. There’s more of the curve to the left than to the right. Given time, we’d get a new bell-curve, possibly centered higher than the average of the population the couple was drawn from, but most likely lower than that of the two founding members.

    • Coagulopath says:

      Another issue (aside from anon’s comment below) is that humans are diploidal: each of our cells has two copies of each chromosome, one from each parent. When meiosis occurs (the production of a sperm/egg), random chunks from each copy “cross over” to form a new chromosome.

      This means a child isn’t getting a perfect 25-25-25-25 blend of their grandparents’ genes, but an uneven mixture. It’s theoretically possible (although extremely unlikely) to receive half your autosomal DNA from one grandparent and zero from another!

      You can imagine the problems this causes when, say, one grandparent had an IQ of 140 and the other 80.

    • idk just some guy says:

      No, you’ve misunderstood how regression to the mean works. There is no regression to the mean in genotypic IQ, only in phenotypic IQ.

    • Jacob says:

      It’s more accurate to say that we regress to the mean of our grandparents and great-grandparents than to that of the general population. This takes place specifically because not all inherited predisposition to intelligence is attributable to additive heredity — some is non-additive, being attributable to gene x gene interaction effects.

      Additionally, the correlation between two spouses’ IQs is 0.4. A smarter spouse will be more likely to find someone less intelligent than themselves because most people are less intelligent than them, but they’ll probably find someone smarter than average because they’re still selecting on that trait. This probably takes place with some personality traits as well. The socioeconomic status of bloodlines thereby regresses to the mean at a slow but steady rate, retaining 0.7 – 0.8 of the status they had in the previous generation, and this is shockingly consistent across cultures. Greg Clark’s The Son Also Rises describes this in pretty great detail, and I always recommend the book to anyone interested in the fate of bloodlines.

      Three other implications of this:

      1) low-status bloodlines regress upwards at the same rate that high-status bloodlines regress downwards.

      2) most elite families probably rose as slowly as they will fall.

      3) regression is the most likely result, and therefore occurs to most families each generation, but it is not the only result. A minority of families in every generation will drift even further from the mean. This is driven by the same force that causes regression in most families: probability.

  16. anon says:

    No, the entire concept is that two parents averaging 130 result in offspring who average 115 or 118 or so, and NOT 130.

    The way to think about it is that we have an immeasurable, pure genetic IQ (GIQ). On top of that you have a modifier, an independent random variable with zero heritability (LUCK), which gets added/subtracted to your pure IQ. The sum of these two variables become your phenotype.

    So when you take two high scoring people, they (on average) will have had a good roll of the dice in the LUCK department. Thus their measurable IQ is larger than their GIQ.

    The kids will, on average, have the midpoint of the parents GIQ, but will average zero LUCK— hence the reversion toward the mean for measurable IQ.

    • Glengarry says:

      Do you by this LUCK factor mean ‘environment’? That is, the non-heritable 20% of IQ?

      (I’ve seen various values for this, basically 20-40% but 20% was the most recent one I can recall.)

      • anon says:

        I’m kinda outta my league here… but I’ll try.

        By luck, I mean environment + fortuitous genetic arrangement, i.e. being homozygous in the right spots so recessive genes get expressed.

        Say expression of dominant allele has effect of zero, while expression of recessive gives +1. Thus, somebody who is heterozygous gets zero boost to their phenotype. But they also have a genetic potential they can offer their offspring: through passing on the recessive, which has some chance of getting expressed should the 2nd parent also pass on the recessive allele. That would be reverting upward toward the mean.

        By the same token, you can get lucky across your genome and get a bunch of positive-value recessives that get expressed, Upon breeding, you will send a copy of that recessive to your kids, but, probabilistically, they’re not going to get the second copy as often from the other parent. They’ll be hetero and get no boost to their phenotype– hence reverting down toward the population mean.

        It’s these situations where phenotype =/= expected value of the genotype that create the math for a mean reversion.

  17. ohwilleke says:

    How about?

    (1) no dogs (a major survival edge for the main wave);

    (2) a small founding population that suffered technological degradation similar to what Tasmania experienced when it separated from Australia including the loss of maritime travel technology needed to reunite with kin in Asia or Beringia;

    (3) inbreeding depression greater than the main founding population of the Americas, and

    (4) challenges to thriving at hunting and gathering due to the ice age that caused the Last Glacial Maximum.

    Also, probably, in almost all of their range, they did go extinct after not too many generations. May have not reached South America until after the LGM if at all.

  18. Jacob says:

    Kind of a longshot but I wonder about inbreeding. Specifically, I think that technological innovation (maybe) could have allowed them to overcome their environment, and inbred people are worse at inventing things. All it would’ve taken was:

    1) a severe bottleneck at the time of arrival. Likely.

    2) never developing the requisite effective population size to successfully select against the deleterious variants that drifted to higher frequencies during said bottleneck. Well, we already know that they never had fantastic numbers. Exactly how uncommon these people were is uncertain. The main argument against my idea is that this population threshold would be so low that, if my idea were true, we never would’ve found out about the precursors in the first place.

    3) to a lesser degree, recovery from this effect would depend on selection for general intelligence, of which we cannot be sure.

    How many times have similar scenarios played out — namely, population history having lasting consequences on phenotype, which become really apparent once another group full of fitter individuals shows up to dominate the same niche? You’ve pointed out that the 2nd effect probably gave us the Florians who, in turn, were liable enough to get wiped out by modern humans. We’ve talked about Neanderthal group sizes possibly yielding a similar effect, too.

    • dave chamberlin says:

      Inbreeding as we all know is the result of closely related couples mating and more recessive traits manifesting themselves. Another mechanism for inbreeding depression is the the removal of the fitness advantage of heterozygosity, which is known as over dominance. This can lead to reduced fitness in a population with many homozygous genotypes, even if they are not deleterious or recessive.

      As stated earlier inbreeding tends to sort itself out through time, people with decreased fitness are less likely to reproduce and can be sterile. But even when the inbreeding has sorted itself out the results of this very severe population bottleneck will be felt by the removal of so many alleles from the gene pool that heterozygosity is greatly increased. .

      Probably not scientific to say this but maybe it will make it clearer. Heterozygosity is the opposite of hybrid vigor. I am guessing that these people that made an almost impossible journey across the Pacific Ocean weren’t dumb at all. After all they made a craft that got there and provisioned it. But it was a small group that landed on a beach Their new home was a food foragers paradise. Giant turtles and the like just waiting for a hominid to come by and crack them open.They survived easily however they had lost so much variance in their gene pool that the following generations had severely decreased cognitive functioning and thereafter behaved entirely differently from the first arrivers.

      • dave chamberlin says:

        If Cochran wrote a book about this he could title it the 10,000 Year Fizzle

      • Jacob says:

        That’s basically what I’m thinking here. Although I will say that homozygous recessive genotypes are especially awful in the context of families because of ultra-rare loss of function variants. Most of us are protected by the fact that our mom’s ultra-rare LoF variants are different from our dad’s ultra-rare LoF variants. Having one such variant only costs us the opportunity for overdominance, but in inbreeding, you can get the same LoF variant on both sides. A bottleneck will grab some of these variants and make them much more common than they should be thanks to drift, a problem that can only be solved with selection.

      • Karl Zimmerman says:

        I think it’s worth taking into account a different species here…Cheetahs.

        Cheetahs seem to have suffered from two reproductive bottlenecks (100,000 and 12,000 years ago) which eliminated 90% or more of their genetic diversity. Modern cheetahs are so genetically similar you can graft skin from one to another without any rejection. Their sperm count is also awful, with 70% to 80% of sperm deformed. Now, cheetahs seem adapted well enough to their own niche, but they do horribly when taken out of it. It’s almost impossible to breed them in captivity, for example. They’re also more prone to disease/infection, and have high infant mortality for a cat.

        This just shows that if there’s a semi-stable niche, a genetically effed-up population can survive without purging their inbred genome of all deleterious effects…until someone else muscles in.

  19. dave chamberlin says:

    I guess I’m talking to crickets but I find the entire subject area fascinating. Scientists don’t have a clue about how genetics determines IQ, all they really know is that it does to the tune of approximately 75 percent. It was first pretty logically thought optimum brain function, say an IQ over 145, or .3% of the population was determined by the sum of all the genes that have a positive effect. But then the effect of individual genes was isolated and none had the effect of even one tenth of a point on IQ, so that is not the source of very much IQ variation. Cochran and others have theorized IQ variation is the result of mutational load. I don’t think it’s that either. Now mutational load is definitely the cause of very low IQ but I think absence of mutational load is not the source of very high IQ. I won’t go into the details as to why in a blog post but just leave it that this is an unsatisfactory answer. So some unknown factor or factors is causing optimum brain function to only express itself in a tiny percentage of the population via genetics. Another factor besides mutational load and genes of small positive effect is the largest factor in IQ variance and we don’t know what it is. The above mentioned Heterozygosity could be that unknown factor. I have no idea but it’s worth looking at. If we can look at a person’s DNA and pretty accurately predict their IQ like we now can predict height that would be an immense scientific achievement with future implications of designer babies that are born with a platinum spoon in their mouth.

    • Jacob says:

      I’m open to hearing alternate ideas but, personally, I think mutational load does explain a significant chunk of the variance

      Also, if having greater than average load predicts low IQ, having lower than average load should predict higher IQ, even if not necessarily to the same degree.

    • Cat Rationalist says:

      Isn’t it that there are some problems with measuring mutational load. Genotyping chips usually considers tag SNPs which more frequent enough that certain design threshold.
      If low IQ is determined by rare alleles and everyone has some, but different ones — it’d be difficult to find them, even with huge sample sizes.

  20. jb says:

    How does the fact that individual genes have small effects refute the hypothesis that IQ is “determined by the sum of all the genes that have a positive effect”? Wouldn’t that simply mean that you need a greater number of such genes?

    I ask because I’m interested too, and for a long time I have had an uncomfortable intuitive sense that the larger the number of genes required, the less discrete inheritance becomes, and the more you run into the problem of simple dilution of inherited variation that always puzzled Darwin. If a very large number of genes is involved, and IQ is just the sum of those genes, then how is it possible that a child’s IQ could be anything other than the strict average of its parents? How could you expect any significant stochastic variation up or down? Wouldn’t the law of large numbers prohibit that? Is that what you’re saying?

    • dave chamberlin says:

      Yes I am. The variance in IQ is far too large for it to be determined by the sum of genes that barely have any effect at all on IQ. There is something else that causes that 15 point standard deviation and nobody knows what it is. Mutational load or lack there of has been suggested but I find that unlikely as well. Fact is we know very little about our most important feature, human intelligence and how inheritance determines 75 percent of it. I am not an expert in the field so I want to stop there.

      • Anonymous says:

        Huh… perhaps if you compare a test genome to the Cambridge Reference Sequence and score a malus for each difference you would get a usable model for intelligence. But then as you say, regression to mean would be invalid and children’s IQ would be an average of their parents. The alternative would be that IQ is determined by groups (complexes) of covarying genes (or gene controlled effects). One could build a model with varying number complexes and see when the transition from continuous to regression to mean occured.

      • Coagulopath says:

        But how can you know this? We still don’t have all the genes!

        Look at height. Twin studies predict that 80% of height variation is genetic, but for years geneticists could only explain 40-50%, even with massive GWA studies involving millions of people.

        Then, in 2019, Peter Visscher’s team did whole-genome sequencing on a smaller number of people, and was able to recover 79% heritability. The GWAS missed rare SNPs.

        I’m not saying that will necessarily be the case for complex traits like IQ, but the science is nowhere near finished.

      • JAMES SHEARER says:

        “… The variance in IQ is far too large for it to be determined by the sum of genes that barely have any effect at all on IQ. ..,”

        Why do you think this? If I have figured correctly n genes each having two equally likely forms with an effect of 30/sqrt(n) on IQ will have a standard deviation of 15. So for example 1 gene with effect 30 so half the people have IQ 85 and half 115. Or 4 genes with effect 15 so 1/16 with IQ 70, 4/16 with IQ 85, 6/16 with IQ 100, 4/16 with IQ 115 and 1/16 with IQ 130. So 900 genes with effect 1 would do it. Or 90000 genes with effect .1.

        • jb says:

          Humans don’t have 90,000 genes — it’s more on the order of 20,000.. And they won’t all be involved in intelligence. If 10,000 genes were involved, by your math they would all need to have an effect of .3, which is very much out of line from anything anyone has found.

        • dave chamberlin says:

          I am guessing that we have not yet found the primary genetic determinate for the large variation in IQ. Certainly the best guess we have so far is 1) mutational load or lack there of and 2) summation of genes which have a small positive effect. There is something missing, some perfect balance that makes optimum brain function or IQ above 145 so rare. I am not guessing what it is but simply saying we have a very poor understanding of how the brain works and an almost complete ignorance in how genetics determines IQ to the extent it does. Stay tuned to this subject area because breakthroughs are eventually coming in this area. I’m hopeful that artificial intelligence will crack this problem this century. People fear artificial intelligence, I fear actual stupidity much much more.

    • Esso says:

      Is it possible for single faulty variants to have a small additive effect in population level statistics, but a large effect for an individual, maybe because he is also faulty in some other function that would compensate? Maybe it’s like that game of broken phone, where small faults multiply to a large effect.

      It need not be a big flaw when it’s in the head, my father always says.

    • Jacob says:

      “If a very large number of genes is involved, and IQ is just the sum of those genes, then how is it possible that a child’s IQ could be anything other than the strict average of its parents?”

      This pattern actually does predict the bell curve seen in the data (within and between families), so the shape of the distribution makes sense. Since alleles are independently inherited, the odds of getting significantly higher-than-expected or lower-than-expected IQ are not high, leading to a big lump of people in the middle of the distribution. However, it remains possible to get lucky or unlucky, which explains the tails.

      Variance is inevitable, therefore, and the real question is why variance exists to the degree that it does (rather than a much smaller one). What might stretch out the curve to the point where your place on it actually matters? I can think of two decent explanations:

      1) Assortative mating. Humans are known to assortatively mate by IQ, so I think that does maintain our current distribution of intelligence to a significant degree. In all likelihood, this has always happened, and always will happen.

      2) It requires very little genotypic variance, relative to the amount that could theoretically exist, to actually make someone smarter or dumber relative to other living humans.

      The wild implication of 2), if true, is that someone with purely human DNA could theoretically be far more intelligent than anyone who has ever lived. This could be accomplished by cherry-picking embryos or, thanks to the existence of polar bodies and single-cell DNA seq, ova, and doing this for multiple generations in a row.

      Note that this is already basically how selection works: some alleles are favored, some aren’t. In a massively quantitative trait, tons of potential for genotypic variance already exists, and selection can crank that shit like a dial.

      • dave chamberlin says:

        Think of what humanity will become if we can look at a person’s DNA and accurately predict their IQ if they develop normally.

        You think the industrial revolution was a transformative revolution of mankind? You ain’t seen nothing yet. For starters parts of the world will gladly improve the IQ of the next generation by in vitro fertilization of ten eggs and only implanting the smartest one in ten. Mutational load would be decreased dramatically right away and keep decreasing every generation. The following generations, thanks in part to their higher intelligence, won’t stop there. A deeper understanding of optimum brain function as controlled by DNA will surely follow and i would argue that it will become probable that genius will become average for parents who choose it in those parts of the world that allow it.

        The abortion is murder folks will of course become apoplectic in opposing any such option of throwing out nine of ten fertilized eggs and the stage will thereafter be set for at the very least a Cold War and possibly a hot one, if this first and easiest option is taken. But who knows what will happen. I need to stop there with my futurist babble because Futurists have a terrible track record of predicting the future and I will too. They always make the mistake of thinking multiple steps ahead and I caution against doing that. Each step in predicting the future is unlikely and instead of running off to a step five scenario that has a one in a million chance stick to step one and weigh the probability of various options of what humanity will do when it understands how DNA predicts IQ. We know it does because identical twins are very close in IQ.

        Fun stuff to think about

        • Jacob says:

          Your insightful comment deserves a more detailed response, but for right now I’ll comment that we can throw a bone to the religious right: egg selection.

          If you did single-cell DNA seq on the polar bodies, you could somewhat-accurately infer the genome of an unfertilized egg, and then only fertilize eggs you intend to implant. This should be roughly 70% as effective as embryo selection, not 50%, because a ‘smart’ egg is fertilized by an ‘average’ sperm more often than not — the variability in outcomes goes down as you add more info, so capturing the first half of info gives you more than half of the variance.

          This solves the abortion element of it, but not the fact that you’re making test tube babies. Will they accept this? Well, no, probably not. Still against the natural plan or whatever. Pearls before swine.

          The conservative rejects our help despite the fact that it’s objectively a good idea, but the leftist rejects our help specifically because it’s a good idea. We have to sidestep politics entirely, ergo we have to eschew any attempt at honesty about our intentions with respect to IQ in particular.

          The solution? Market embryo selection/egg selection as a way of reducing the odds of childhood cancers, type 1 diabetes, schizophrenia, Huntington’s, etc. Religious people would still hate us for it, but this time we can at least get the cult of Fauci on board. We never have to tell them that we’re trying to make people smarter.

          • dave chamberlin says:

            I like your thinking. Once we know how our DNA pretty accurately predicts IQ, I don’t see how the anti genetic engineering folks will stop a technology that makes the next generation considerably smarter. They may suppress it for a long time in parts of the world, but we already know laws are basically futile in suppressing black markets where there is high demand. I can see the opposite sides getting furious with each other along the lines you propose. One side accused of being pro genetic engineering of freaks, and other being called the new Amish. “All it is mind glasses, giving the next generation a crystal clear mind, how is that a bad thing!” “Abominations, they are abominations!” It will make today’s most emotional political arguments look tame.

            My opinion on the matter is simple. If it works (a big maybe right now) then there will be no stopping it. Whatever emotional opinion rules the day in various parts of the world will only delay the inevitable, a society that can multiply it’s percentage of it’s population of high IQ people wins.

    • Cat Rationalist says:

      You’re forgetting that genes that sit near on a chromosome usually inherited together. Here’s a quote I found somewhere

      There are typically between one and four recombination events per generation in a chromosome, depending on its length [21]. This corresponds to about one recombination event per every hundred million bases [22].

  21. teageegeepea says:

    This is off-topic, but the domain for your website “The 10000 Year Explosion” appears to have been taken by someone else.

  22. Soren Gard says:

    Off topic, but has anyone read the recently released “The Dawn of Everything: A New History of Humanity” by anthropologist David Graeber and archaeologist David Wengrow. It looks fascinating but I’, not equipped to evaluate it. This crowd however seems to be.

  23. Esso says:

    Mr. Cochran,

    please ignore the ethical tendence, but do you think there could be something to this:

    That is: selection of livestock and antibiotic/probiotic feeding also selecting for obesogenic microbiota, and those microbes spreading to humans and causing the post 80’s epidemic of overweight.

    • dave chamberlin says:

      I don’t know about that but a very clear reason for the increase in obesity has already been found. Americans have increased their caloric intake by 23% since 1961. It’s a wonder that we aren’t fatter than we are.

  24. Nomen Est Omen says:

    Robert E. Howard’s take on prehistory was closer to the truth than the one promulgated by archaeologists in the past few decades…

    So maybe H.P. Lovecraft was right and the pre-Indians were being predated by eldritch creatures:

    “The Call of Cthulhu”

    […] The region now entered by the police was one of traditionally evil repute, substantially unknown and untraversed by white men. There were legends of a hidden lake unglimpsed by mortal sight, in which dwelt a huge, formless white polypous thing with luminous eyes; and squatters whispered that bat-winged devils flew up out of caverns in inner earth to worship it at midnight. They said it had been there before D’Iberville, before La Salle, before the Indians, and before even the wholesome beasts and birds of the woods. It was nightmare itself, and to see it was to die.

  25. epoch says:

    There is a similar thing this side of the Pacific. There is a small genetic signal of a unknown pre-OoA wave of human settlement, only to be found in Aboriginals and Papua’s. Come to think of it, even substantial Denisovan can only be found in both Aboriginals and Papua, even though very tiny Denisovan admixture can be found in East-Asians.

    Both admixtures look to have been picked up before crossing the Wallace line.

    So a suggestion: A failed first attempt (trapped in worsening climatic conditions due to LGM, perhaps?) of which the remnants are picked up only by one group who became ancestor to Suruii et al.

    By the way, is the dating of the White Sands prints undisputed?

  26. James says:

    It seems to me they mostly died out, probably a combination of insufficient technology, low population size and perhaps climatic change. The first wave (if we are to take for granted the inferences about shared affinity to both Tianyuan and Onge) must have arrived there very early, perhaps earlier than 35kya, perhaps as early as 40kya though I don’t know if climatic data can sustain that. From what I know, East Eurasian lithic technologies at the time were quite simple, so perhaps that’s why there is a lack of artifacts.

    But speaking of early/failed attempts, perhaps you come across this by any chance?
    Seems like there was indeed an earlier OOA that mostly died out, but not before it left some traces in western Neanderthals. I wonder how accurate these inferences are of this type of ancient AMH-Neanderthal gene flow but seem legit thus far.

    • epoch says:

      That is not the only evidence we have for earlier AMH in Eurasia.

      • LOADED says:

        it is immensely difficult to understand that the severity of problems with these categorizations of people based on genetics has created more problems than solutions.

      • James says:

        There seems to have been a failed sort of OOA about 200kya. I think the Broad institute people had said something similar, and they likely did contribute to some Neanderthals.

        Though honestly speaking, a contribution to eastern Neanderthals but not Denisovans or European Neanderthals sounds dubious to me unless that wave was only around South Asia or maybe Southeast Asia.

  27. Gordon William Marsden says:

    the population may have been closer to the comet hit in 12900 years ago than anyone else , one and done

  28. Henry Scrope says:

    Looking fwd to your Christmas book recommendations Dr Cochran, some brilliant ones in the past.

  29. James Bier says:

    I don’t see how this is different from what happened in Eurasia. There is spotty evidence for modern humans in Europe and Asia from much earlier than 60,000 years ago, but the group that left Africa c. 60,000 years ago took over, whereas the earlier out-of-Africa groups either died out or their genetic signal got swamped by the later group.

  30. Mike1 says:

    This is all crazy talk. The Mormons arrived on ships from Israel and settled the US in antiquity.

  31. anonymous says:


    Or make your own:
    described in
    I made my first atlatl/dart set based on descriptions in this book. Be warned, if the dart shaft isn’t flexible enough, it will snap in half on launch.

    Be warned, you need a LOT of space, and they are a LOT of fun

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s