Heteropaternal Superfecundation

Sperm competition is a factor in some cases of paternal uncertainty, but there are also many cases in which it is not. For example, Daniel Boone’s wife had a daughter, Jemima, when he had been gone for more than a year.  There is reason to believe that Jemima was fathered by Daniel’s brother Edward. Non-paternity, but no sperm competition.  Adoption also creates non-paternity without any sperm competition. Generally speaking, sperm competition ought to be more likely in a species that has a seasonal estrus cycle – all effective mating attempts have to be close together in time.

In a case where there is sperm competition, we can assume that the non-official parent fertilizes half the time. This means that 2 times the frequency of non-paternity is an upper limit on the fraction of conceptions in which sperm competition occurs.

Since every recent, high-quality set of genetic measurements shows that non-paternity is currently low and has been low for hundreds of years, usually under 2%, for those populations that have been studied,  the fraction of conceptions in which sperm competition occurs is less than 4% (in those places and times). We have an upper limit – but is it a good upper limit?  Is it close to the actual rate?

Once in a while, women give birth to fraternal twins that have different fathers. The frequency of this event can give us a direct measure of the fraction of conceptions in which sperm competition actually occurs.  In the presence of sperm competition, we only notice the cases in which the twins are discrepant.  The incidence of sperm competition will be twice that fraction.

So what is the rate?  This is the sort of question that could be answered definitively nowadays with a SNP chip, but here’s what we have: In an Italian study from 1992,  the frequency was  2.4% among fraternal twins whose parents were involved in paternity suits – people who presumably had reason to be suspicious. Generally speaking, nonpaternity is much more common in high-suspicion cases of this sort, usually at least 10 times more common than in the general population.  Another 1992 estimate was that 1 in 400 pairs of twins born to married white women had different fathers, which give a sperm competition rate of about half a percent – reasonably compatible with the Italian numbers.

Maybe we need to redo this with modern genetic techniques. Maybe Italian girls are special.  Maybe everything was different back before recorded history.

But if these numbers are correct and at all representative, sperm competition in humans is insignificant, and all the people talking about our specialized somatic and behavioral adaptations for sperm competition are wrong.

Someone mentioned that I am calling people loons who’ve authored of dozens of peer-reviewed papers and are at serious (sic) universities.

I can only say – somebody’s gotta do it!


This entry was posted in Genetics. Bookmark the permalink.

126 Responses to Heteropaternal Superfecundation

  1. Nick says:

    Are there any measures of the variability of this over time?

    I think some people believe that as the family unit in America is breaking down, non-paternity is becoming much more common, so that data from the early 90s may not be representative of the 2010’s.

  2. j3morecharacters says:

    Nice work and good that somebody (you) did it. If 0.5% of all births are from sperm competition where the “non-father” succeeded in its fertilizing the female, then about 1% of all fertilization takes place after sperm-competition. 1% is significant and explains the pump shape of human penis. It also stands to reason that the pump shape and action evolved as a response to sperm competition. Which may have been more frequent in the less settled and the more violent societies of the past.

    • gcochran9 says:

      read again: 1 in 400. .

      • j3morecharacters says:

        I did and it appears that in 0.5% of all fecundation events took place within a “sperm-war” scenario and that 99.5% of the fecundations “non-paternal” sperms were absent. If these numbers are correct and at all representative, sperm competition in humans is significant or insignificant (depending on the writer’s mood). What is your hieroglyph for “More research is needed”?

      • j3morecharacters says:

        On second thought, the frequency of contemporary human fecundation events within a “sperm war” scenario is probably even lower than 0.5%. The 0.5% figure presumes that the long-time partner aka husband and the non-paternal inseminator produce sperm of similar quality and are equally succesful. However, rapes produce higher percentage of pregnancies than regular intercourse (maybe excitation increases the chances of pregnancy – in cats it does). If non-paternals are in average more successful, then there is even less sperm war than estimated by Greg, and behavioural adaptations are even less plausible.

      • Anonymous says:

        j3morecharacters, I just clicked the “comments” intent to make such a point: the non-father was accepted, risk of scandal and all, precisely for his higher fitness (including better sperm), and at a more convenient time for pregnancy than the “father” would have, since ovulating women are more likely both to cheat AND to plea headache (and there is no reason to assume that these psychological adaptations are not further enhanced by physiological ones). But these increased chances of the non-father imply that there is even less sperm competition than what you’d have if the null 50/50 were true. However, this whole “no great amount of non-paternity you idiots” idea would indeed look even better when presented along with data from a greater variety of populations than the relatively egalitarian Bushmen circa 1970 or the emerging and contemporary first world; what about amongst the really rugged peasantry in traditional despotisms?

      • little spoon says:

        ” However, rapes produce higher percentage of pregnancies than regular intercourse (maybe excitation increases the chances of pregnancy – in cats it does). ”

        If that is the case, I think it is entirely because rapists choose more attractive/fertile looking women and not at all because excitement increases chance of pregnancy.

  3. JayMan says:

    Someone does indeed have to do it. We need more than a few someones, I think.

    And so many people still buy Why Is The Penis Shaped Like That? And I imagine they still will…

    • little spoon says:

      stupid penis head scoop thing. I’d like to see if there is a different rate of non paternity among Jews and ethnically European Muslims (Albanians etc). They didn’t even have penises shaped as nature intended, so if the shape is so important and all…

      In fact, why stop at the shape. I think desire for cunnilingus evolved as an added guard against sperm competition. See, that makes sense. It just does.

  4. 3LFreeL says:

    Even if the nonpaternity rate is below 1%, that doesn’t mean sperm wars are insignificant. That logic doesn’t make sense at all considering the enormous cost nonpaternity puts on a cuckolded male. Raising a child not related to you is a huge cost in terms of time, effort, and opportunity.I understand the apprehension that you have when people cite obviously false and astronomically high nonpaternity rates of ~10%, but even rates hovering around 1 in 400 are important. I also don’t know how reasonable it is to extrapolate prehistorical rates of nonpaternity from contemporary nonpaternity estimates. Certainly a non-trivial amount of nonpaternity is significant in explaining penis shape and male jealousy.

    • gcochran9 says:

      Non-paternity is not the same thing as sperm competition. Imagine that I just said that enough times to get the point across. Maybe Skewe’s number would do the trick.

      Imagine that you have a species with zero paternal investment. On average, a male in that species is going to have two surviving offspring. Imagine that one of those matings that otherwise would have resulted in his child is pre-empted by someone else’s super-sperm. His fitness drops by half.

      It is no different if our hero is a member of a species with high paternal investment. His fitness drops by half.

      The cost is the same.

      But there is a difference: paternal investment has a much lower payoff in that situation.

      Without sperm competition, did you expect the penis would be rectangular or something?

    • muhry says:

      Male gorillas exhibit sexual jealousy, so jealousy is not necessarily evidence of sperm competition.

  5. Stagg says:

    I always thought that the high frequency that men want to copulate with their mates was evidence that infidelity was a problem in prehistoric societies. By regularly copulating with his mate a man would ensure that her reproductive tract was constantly filled with his sperm and minimise the chances of her being fertilised by other men she had sex with when he wasn’t around.

    Doesn’t that count as a kind of sperm competition?

    • Anonymous says:

      It does whenever there is a competitor; however, there is more than impregnation concerns in human copulation frequencies. Just try to think of what reactions are stimulating, and what they may signify.

  6. Pingback: linkfest – 03/31/14 | hbd* chick

  7. szopeno says:

    My comment here has nothing to do with the blog’s entry, simply I had an idea and this is the place when a lot of knowledgeable people comments. Sorry for the author, please don’t kill me.

    Currently rates of incarceration for blacks are very high. People in jail should have lower number of children surviving; even if you have “wet visits” in USA, fertility should be affected; as only one parent is present, this should lower survival of the children (this is true even in Sweden); all in all, current black population seems to be under very strong selective pressure, removing the least intelligent and the most violent members of the population. If I am right, how quickly black population mean IQ could increase and how much of supposedly diminishing gap between mean IQ in white and black population can it explain?

    • j3morecharacters says:

      People in jail should have lower number of children surviving… but they dont.

    • Anonymous says:

      Despite this, blacks are constantly getting (genetically) dumber, just like everybody else is; the better ones tend to marry out, and the prison material didn’t go through all the trouble just for nothing.

  8. Patrick Boyle says:

    I think we have a lot more social evolution to undergo before we have worked out appropriate laws and rules for the use of SNP chips. The public simply doesn’t appreciate the technology.

    For example I ran into a series of YouTube videos yesterday that promoted the idea that Michelle Obama is actually a transvestite. She does appear to be a tall and muscular woman but most of the ‘evidence’ is based on finger length – not real persuasive for me. This question could be resolved – as well as the paternity of the daughters – with a few safe and simple tests.

    But we don’t know yet how to handle genetic information. We haven’t thought it through. Some day soon all Senators and Congressmen may have their genes in the public record. Will that be a better world?

    • Anonymous says:

      Extraordinary claims require extraordinary evidence. Michelle Obama is not a significant enough outlier to seriously question her sex.

  9. “In a case where there is sperm competition, we can assume that the non-official parent fertilizes half the time.”

    What allows us to assume that?

    • gcochran9 says:

      If you assume that the sperm competitiveness of two randomly drawn men are, on average equal, and that the circumstances are equal, you’d expect equal chance of reproductive success for each guy. Now circumstances might be different: if the non-official potential dad has to worry about a spear in his back, he will probably have sex with the lady less often, and his sperm will on average be older and less vigorous at ovulation. Or you could posit that females have various ways of favoring sperm from one of the two candidates, which might change things. Although people have talked about such mechanisms, I don’t think that there is any evidence that they actually exist in humans.

      • Sure, but circumstances aren’t always equal. If a woman has sex with her in-pair partner 3 times a week and with her extra-pair partner once a week consistently, we would expect to see a 25% non-paternity rate. This would be the case despite there being sperm competition for 100% of conceptions, so the non-paternity rate would underestimate the degree of sperm competition substantially. This, of course, works in the opposite direction as well, where you could potentially see 100% non-paternity in the absence of any sperm competition.

        In order for the 50/50 ratio to work, then, a woman needs to be fucking her EPM about as often as her IPM. I don’t think that assumption would hold in the world.

        • gcochran9 says:

          You’re misunderstanding your own scenario: the guy having sex once a week would probably completely miss two-thirds of the ovulations and have a 50% shot at the remaining one-third.

          The estimate from discordant twins tells you the actual rate that would be relevant to evolution. Half a percent – which is the most that you would gain if your sperm were more powerful than a locomotive and could jump tall buildings in a single bound.

          My new theory is that the people writing all this dreck about sperm competition are actually chimps.

  10. I should get a little more specific, I feel. So let’s say that there is a 2% non-paternity rate, on average, in human populations. Does this capture the sperm competition risk well? I don’t think so.

    First, let’s consider what percent of women admit to infidelity. That number, from a quick Googling (not great data, admittedly, but let’s just use it) says around half. So that 2% of non-paternity is clustered in only half of the population, meaning those men actually face a 4% non-paternity rate, while the rest face 0%. If we assume, as you did, that the extra-pair male has a 50% chance of winning (which might or might not be true, but let’s use it), then we might end up by saying that 8% is the upper-limit that men might face with respect to sperm competition; 0% is obviously the lower-limit.

    Now those numbers are by no means definitive, but they at least help us see some different points of view on this. One is that men need not have adaptations for persistently-high levels of sperm competition: half the men face no competition (assuming, that is, that the men not facing sperm competition within their mateships don’t face any when having affairs, which is dubious). But what are you to do if you happen to be in the other half of the population? Perhaps producing a little more sperm or having a little more sex would work better in those cases. Context-specific adaptations for sperm competition are still on the table in that case, I would think.

    I’m not suggesting that such adaptations necessarily exist; they might or might not. We’re certainly no chimps that respect, that’s for certain. I am curious, though: how much sperm competition you feel would be “enough” to drive the development of adaptations for it.

    • gcochran9 says:

      Only a fraction of the non-paternity cases involve sperm competition. What you want to look at is the fraction of heter0paternal fraternal twins. Double that. That’s the best estimate you’re ever going to get, something like half a percent.

      I could come some up with something else that happens half a percent of the time and write books about how important it is, how it is an important driver of human behavior, but I would be joking, unlike Baker and Bellis.

      The problem with evolutionary psychology is the people doing it.

      • “Only a fraction of the non-paternity cases involve sperm competition.”

        Agreed. That’s why I said 8% might be the upper-limit, given a few assumptions (like that non-paternity doesn’t exist absent cheating and in-pair men stand the same chance of winning as the out-pair ones); not the actual percentage of sperm competition. That is, of course, the upper-limit *after* adaptations for jealousy and mate-guarding have been at work. Effective as such mechanisms might be for avoiding sperm competition and non-paternity, they don’t stop it entirely. This hypothetical upper-limit is also not the rate the entire population faced, but the rate that the population of men being cheated on might face, in general.

        I’m not sure twins provide our best estimates of non-paternity, given that there might well be unknown variables floating around there. Twins are pretty rare, though useful cases to think about. From the comments on the last post, estimates of non-paternity hung around 2% in cases where paternity suits and twins weren’t involved, it seems, so perhaps this sample might not be the most informative for the general population. It might, but it might not. I wouldn’t hang all my hopes on it.

        The question I posed before still stands: how much sperm competition would you estimate needs to exist for it to matter, in some sense?

        • gcochran9 says:

          For evolutionary purposes, what matters is the average frequency, not the average frequency in some subcategory. So your 8% number is wrong. Go with half a percent: that’s the info we have.

          Then try to imagine a mutation that makes an unusually big difference in affecting sperm competitiveness: it boosts a sperm’s chance from 50% in a sperm duel to 55%. How long would it take for this to become common?

          It would boost carrier fitness by about 5 x 10-4. Let’s say that copy came into existence 100,000 years ago, 4000 generations. The probability that it would be lost by chance is 99.9%. If it did survive, how much would one expect its frequency to have increased in one hundred thousand years? Approximately (1.0005)**4000 , 7.38

          Wait! No! it’s only advantageous in men. Assume that it’s neutral in women. Make that 2000 generations: it increases by almost a factor of 3 in 100,000 years. If the population had remained constant, there would now be three copies, instead of one.

      • Sandgroper says:

        The annual probability of death or serious injury in a vehicle accident, if you include vehicle/pedestrian and vehicle/cyclist conflicts – that’s roughly the same frequency.

        What traffic engineers will tell you is that, no matter what they do to reduce risk (straighten out hair-pin bends in roads or whatever), drivers change behaviour to elevate the risk back to about the same level again, like taking the gentler curve faster, i.e. the risk level is not driving human behaviour. And it has had no significant effect, not so far – we haven’t evolved to be more risk-averse. I guess we could, in time, but it’s probably going to be faster to develop robot-driven vehicles or something like that.

        I guess you could publish a book in which you theorise about how we have made big changes in behaviour since the advent of the internal combustion engine, but we haven’t.

        This is almost slapstick comedy, but I can’t help but relate it. A government minister (now resigned due to a ‘nervous breakdown’) in Western Australia has just been charged by the cops because he drove his government owned ministerial car to a wedding reception, got absolutely rat-faced drunk, and then drove a distance of a couple of miles from the wedding reception to where he was living. During this journey, he managed to impact 4 other parked vehicles, a telegraph pole, and the front gate of his house.

        And he was the Minister for Transport.

      • Anonymous says:

        Big deal; Mihallis Liappis (phonetically; Michales Liapes if you prefer traditional Greek to Latin transliteration), our former minister of transports, was caught a couple of months ago with fake plates on his SUV (to skip taxes); yesterday the secretary of the government was recorded in a conversation with a Golden Dawn MP (the one who physically attacked two women on tv) claiming that their prosecution is not his own fault, and calling his own prime minister “a faggot”. Today his son, a special forces sub-officer (sorry for my bad English military terminology; what do you call a sergeant as opposed to a captain?) stormed into parliament, and apparently kicked Golden Dawn MP Matthaiopoulos, fiancee of Golden Dawn’s leader’s daughter, who became MP as an engagement present–the party had another MP resign for him to take his place. And everyday I cannot read a newspaper without spelling errors.

    • Greying Wanderer says:

      “First, let’s consider what percent of women admit to infidelity. That number, from a quick Googling (not great data, admittedly, but let’s just use it) says around half.”

      Half of women are unfaithful every month or half have been unfaithful at least once over multi-decades of marriage?

  11. I can’t seem to reply to the responses for some reason, so I’ll just reply here to both threads instead.

    “You’re misunderstanding your own scenario: the guy having sex once a week would probably completely miss two-thirds of the ovulations and have a 50% shot at the remaining one-third.”

    I don’t think I’m misunderstanding my own scenario. If the EPM is having sex with the woman once per week (i.e. four times a month), I don’t know how you figure that he is missing 2/3s of the ovulations. Perhaps you could provide some more detail in that respect?

    As for the second thread:

    I couldn’t help but notice you avoided answering my question (“how much sperm competition would you estimate needs to exist for it to matter, in some sense?”), opting instead to provide information about other matters using other assumptions.

    Allow me to rephrase the question:

    Let’s say you start with a species that evidences 0% sperm competition: they’re a bunch of monogamous pair-bonders. Further, let’s imagine that a future ancestor of this species resembles chimps with respect to sperm competition (i.e. there’s a lot of it). How do you figure this evolutionary path would go?

    We could even use the numbers you did: a mutation that boosts the competitiveness of male sperm by 5% for winning the race. How much sperm competition would suddenly need to exist for that adaptation to get off the ground, so to speak?

    • Sandgroper says:

      He did answer your question, you just didn’t understand the answer, which is that the adaptation never gets off the ground.

      I’d like to know where you got that half of women admit to infidelity. I Googled infidelity rate for women in a committed relationship, and found numbers nowhere close to 50% ‘ever’.

      My own lack of meekness on this is attributable to me seeing that a lot of people seem to be inventing the ‘issue’ of sperm competition in modern humans and then examining how much of an issue it is, when the actual evidence I see from numbers is that the occurrence of it is so low that it is not of any significance at all. For the large majority of modern humans, this just doesn’t come up, certainly not approaching anything like the kind of frequency that would favour adaptations for it. The ‘penis pump’ is an invention of the imagination.

      Of course, inventing ‘issues’ so that they can be ‘debated again and again’ is not that unusual among the chatterati.

      • This was the first thing I came across when I Googled “what percent of women admit to cheating”: http://www.statisticbrain.com/infidelity-statistics/

        I’m not saying those numbers are necessarily accurate (I didn’t even dig up the source); I just wanted to use some numbers for the sake of the argument.

        That argument – again – being that adaptations for sperm competition could, potentially, be facultative, rather than obligate, in nature. Not that they definitely exist and take a certain form; just the possibility of them still being on the table.

      • Sandgroper says:

        1. I suspect you’re smart enough to know not to do that. 2. You are talking about adaptations in humans. Those are American data. If pressed, I am willing to admit that Americans are human, but not all humans are American (hard to believe, I know, but true).

    • misdreavus says:

      If I understand correctly:

      1) This paper* says the rate of heteropaternal superfecundation among white American women is roughly a quarter of a percent.

      2) Assume that any two men, taken at random, have roughly equal odds of fertilizing an ovum, given that both have intercourse with the same female within close temporal proximity to one another. (This would have to be the case, if there has been ongoing selection on human males through sperm competition. Relevant mutations are hard to come by, and selection would have driven any alleles of large effect into fixation long ago. )

      (And it’s also the case when there hasn’t been much selection at all. So the assumption holds true in both outcomes.)

      3) It doesn’t matter if the female only cheats on her partner 1/10, 1/100, 1/1000, or one millionth of the time. And it matters not if she has an exclusive tendency to be unfaithful outside (or within) her window of fertility. All those other scenarios are irrelevant to sperm competition. (To say otherwise is ludicrous from a biochemical perspective — just how is a sperm supposed to know whether or not it has been implanted in the right fallopian tube? Epigenetic imprinting? No, we only care about sperm from two different fathers that end up in the same womb.)

      4) Given assumption #2, the frequency of sexual events in which sperm competition becomes plausible is roughly double the figure in #1, or half a percent. This is basic probability theory. If the cuckold loses out half the time whenever sperm competition occurs (again, you don’t care about every other scenario!), twice the frequency of heteropaternity gives you the relevant frequency. 1/2 * 2 = 1.

      5) Knowing that opportunities for sperm competition are pretty uncommon, it is possible to calculate the chance that any positive allele that originates in a population will survive and rise in frequency over time. (See Cochran’s comment above.) Which is to say, pretty damn low. The odds don’t get much better for an allele that boosts a sperm’s chance to 60% or higher, and you very rarely see any new mutations that boost fitness in that way.

      *James, WH. “The incidence of superfecundation and of double paternity in the general population.” Acta Genet Med Gemellol 1993;42:257-62.

      • Paul says:

        1/2 * 2 = 1


        My God!

      • “This paper* says the rate of heteropaternal superfecundation among white American women is roughly a quarter of a percent.”

        OK. Twinning is also quite rate. Twinning with two different fathers rarer still. I’m not saying those numbers are irrelevant, but there are a lot of variables floating around there that we don’t know about, so I wouldn’t put all my eggs in that one basket. We might want to start with non-paternity in normal births.

        “It doesn’t matter if the female only cheats on her partner 1/10, 1/100, 1/1000, or one millionth of the time.”

        I think you’re making a whole lot of assumptions to get to that point. For instance, here’s some evidence that horses selectively abort pregnancies in particular circumstances involving strange males: http://www.epjournal.net/blog/2011/04/who-has-been-horsing-around/

        This isn’t just about sperm.

        “Given assumption #2, the frequency of sexual events in which sperm competition becomes plausible is roughly double the figure in #1, or half a percent…The odds don’t get much better for an allele that boosts a sperm’s chance to 60% or higher, and you very rarely see any new mutations that boost fitness in that way. ”

        OK; let’s talk about the assumption then. If a woman fucks her IPM once and her EPM once during her fertile window, we might assume that each has a 50/50 shot of fertilizing the egg. Let’s now saw that a mutation arises that has the following effect: it causes the IPM to be motivated enough to secure a second copulation within that fertile window. All else being equal, then, the IPM has now raised his odds of fathering the offspring from 50% to 66%. That’s not too bad. In fact, a 16% increase in competitiveness is more than triple the 5% mutation that Greg deemed to be “unusually big”, and the in question adaptation doesn’t appear to be terribly complicated at face value.

        So here’s a question to mull over: let’s take the hypothesis that humans have faced zero sperm competition over our evolutionary history; at the very least, not enough to forge any adaptations for sperm competition.

        What predictions does this hypothesis generate? Here’s one I could think of: when a man discovers his partner has been unfaithful to him, we should not expect (a) any change in the man’s interesting in having sex with his partner or (b) any change in the qualities of that man’s ejaculate over the next few copulations with his partner. Now sure, predicting the null is a bit of a bother, but those predictions do seem to flow naturally from the hypothesis.

        Do you think they’d be supported?

        • gcochran9 says:

          Jesse, if it happens so rarely, it’s not at all important, in an evolutionary sense. Let me tell you my best guess on the questions you ask: the sort of behavioral adaptations you’re talking about are unlikely to exist, but they are fairly likely to be reported by certain ev-psych types even though they don’t exist. In much the same way, Baker and Bellis talked about a study in northern England (in a block of flats in Liverpool) that had a nonpaternity rate of 20-30% – but that study has never been been published, and I think that it never took place. I think Baker and Bellis made it up.

          Look at this way: the best evidence we have, some very good indeed, is the actual rates of nonpaternity and sperm competition are drastically lower than anyone publishing in this field has ever admitted. The reasonable response is not to jump through hoops coming up with a scenario in which sperm competition somehow still matters, but instead to just admit that it was all a mistake.

          I could write a fascinating book about cainism in humans, chock-full of the ways in which it underlies all our social patterns, how it shapes history, all the special adaptations it has generated. I could put on my humanist hat and mourn the sadness of obligatory fratricide and the pain of natural selection. I could even imagine a world where cainism doesn’t exist, one where people say paradoxical things like “bare is brotherless back.”

          Of course, cainism does NOT exist in humans, but I’m thinking that I should maybe call my agent.

      • I should add on a quick thought:

        “All else being equal, then, the IPM has now raised his odds of fathering the offspring from 50% to 66%. That’s not too bad. In fact, a 16% increase in competitiveness is more than triple the 5% mutation that Greg deemed to be “unusually big”, and the in question adaptation doesn’t appear to be terribly complicated at face value.”

        Since the EPM’s chances drop to 33%, it swings the difference substantially. It goes from a 1-to-1 ratio favoring both males equally to a 2-to-1 ratio favoring the IPM

      • Greg,

        I have no idea why the “reply” option stops appearing after a certain point in the thread. Oh well.

        “The reasonable response is not to jump through hoops coming up with a scenario in which sperm competition somehow still matters, but instead to just admit that it was all a mistake.”

        We can disagree about that, then: I would say the reasonable response is to generate hypotheses and go test them; I suggested two possible tests already. They may well not pan out and that’s fine. They results might pan out in a way consistent with sperm competition, but be explained by some other factor. That’s fine too.

        “I could put on my humanist hat and mourn the sadness of obligatory fratricide and the pain of natural selection.”

        Just so we’re clear – again – I have been arguing that *facultative*, not obligatory, adaptations for some mild degree of sperm competition *might* exist in humans, so I’m not sure why you’re drawing that particular counterexample. Sperm competition is not an “either/or” type of thing. There are degrees of it, and these degrees are not always spread evenly throughout the population (that is, it’s more likely that certain women are more promiscuous than others – rather than all women are approximately equally as promiscuous – and cues to that promiscuity might result in particular sperm-related behaviors).

        Similarly, just because humans don’t commit obligate siblicide, it doesn’t mean we should stop research on the subject.

        I get it; I really do. You think there are some bad estimates out there for non-paternity. I’m not arguing that point. You think there is bad research or conclusions out there concerning sperm competition. Fine; I agree. You think Baker and Bellis might be liars. OK. But you seem to be making a number of assumptions in your arguments which, I think, warrant deeper consideration before granting them. I’m not even saying you’re wrong; just that there are some extra assumptions floating around here which might or might not be true.

      • Sandgroper says:

        In % terms, suicide doesn’t matter much.

        If I was researching suicide, I’d be looking less at environmental factors and more at multi-generational paternal age, and rates of mental illness and ‘academic morbidity’. The suicide rates among schizos and dipolar people are astronomical.

      • j3morecharacters says:

        CAINISM LIVES! Many statistical studies have shown that BIRTH ORDER has a strong impact on survival. Laterborns are heavier at birth, on average, and should thus be less prone to disease; however, despite all the possible advantages of laterborns, a large number of studies have indicated that children of high birth order tend to end up in a disadvantaged position during upbringing with regard to both health (Sears, Maccoby, & Levin, 1957; Nixon & Pearn, 1978; Kaplan, Mascie-Taylor, & Boldsen, 1992;
        Elliot, 1992) and educational achievement (Walld!en, 1990, 1992; Belmont & Marolla, 1973; Belmont, Stein, & Wittes, 1976; Breland, 1973, 1974). Taken together,
        these latter studies suggest that children born late in the sibling order may be more vulnerable to disease and death than their earlier-born brothers and sisters. This is from a Swedish study. If this is so among well-fed equalitarian Swedes, imagine what it must be among genocydical Mongols. The death of a Turkish Sultan unavoidable was followed by a fraternal massacre – his most surviving Cainistic son inherited the divan. I have more details for Greg’s book on the subject.

        • gcochran9 says:

          This is why even opening my mouth is risky. Make a joke with a new idea in it, no matter how bullshit, and you always risk it growing until it overshadows the Earth, like the Garuda bird.

      • j3morecharacters says:

        For the Book on Cainism: Michael Corleone orders the death of his brother Fredo in The Godfather Part II. BTW: Only sons are safe.

    • misdreavus says:

      Damn opinion polls and whatever people claim they do with their private parts. People are liars. You’re going to have to find the truth some other way, which requires not being a typical evolutionary psychologist. And since it’s hard to do any experimentation to test this hypothesis (what, is some NYU grad student going to pay a bunch of hobos in Manhattan to screw each other?), this is as good as we’ve got.

      • j3morecharacters says:

        CAINISMO in the Spanish Royal House! Now that Greg opened my eyes, I’m finding evergrowing evidence of the prevalence of human cainism. All Spain is excited as new evidence seems to show that His Highness the King Juan Carlos de Borbon y las Dos Sicilias is a Cainist, having shot his younger brother Alfonso in the forehead. It was not a “gun accident” as described in the official communiqué at the time but a textbook example of cainism. The scientific enigma of why the laterborn die younger is close to resolution.

      • j3morecharacters says:

        Sandy, I know that for you all Africans look the same, but Bekele is Oromo, Falashas are Agaw. Oromos run, Falashas walk. End.

      • j3morecharacters says:

        Oromo = Oranges; Agaw = Apples.

      • j3morecharacters says:

        CAINISM not only exists in humans but it has been extensively studied under the confusing title of “siblicide”. http://en.wikipedia.org/wiki/Obligate_siblicide In our days of subfertility, when each baby is a precious gift and people has to travel to the end of the world to find a crippled baby to adopt, siblicide seems unconceivable, yet I am old enough to remember when the birth of a new “mouth to feed” was seen with dismay. There was such an overabundance of babies that they were abandoned at churches and left on the doorsteps of wealthy families, and foundlings hospitals couldnt cope. Now that Greg has mentioned it, Cainism seems to me perfectly plausible in humans, only that in the rich and subfertile Western societies we live in, it has been utterly deleted from collective memory.

      • j3morecharacters says:

        Siblicide is the monstruosity, but it would be hubristic to accuse you of its creation. I’m stopping here.

      • Sandgroper says:

        I’m just addressing your original point. I think you are missing some important things here, but never mind. You wouldn’t be the first person to miss something.

  12. RS says:

    > Non-paternity is not the same thing as sperm competition. Imagine that I just said that enough times to get the point across. Maybe Skewe’s number would do the trick.

    What might help is naming some of these pops that have been around or below 2% for at least a few hundred years. Because a lot of peoples, if you look at how they lived in 1800, they probably were about the same 2,000 years ago. And in essential lifestyle, probably many millennia before that, though this is a bit more tenuous. Other peoples have probably changed a lot recently. Do you have 2%-ish data for diverse hunter pops?

    Otherwise one could say, well, the ridge of the penis clears semen from other males, this evolved under 12% nonpaternity, which hasn’t been around for about 2,000 years, but the ridge was neutral over that time so we still have it.

    Contrariwise, if we know a number of different hunter pops worldwide had 2% nonpaternity in 1800, it’s at least likely they all or almost all had 2% nonpaternity in 10,000 BC and 15,000 BC. In 15,000 BC they were probably a little less impressive between the ears, but lived the same broad pattern of life. If the penis ridge functioned in sperm comp, and has been very-near neutral for 17,000 years, I would think it would be somewhat deteriorated for lack of purifying selection. I’m not literate in developmental bio, but for petes sake we are only ~300,000 generations removed from chimps. Penis ridge might not vaporize in 800-1,000 generations under near-zero selection, but I think you would see variation in it across individuals and groups ; instead every man looks pretty identical. Therefore it must be under selection re sexual pleasure/bonding or something — and if it ever cleared other males’ sperm, that was probably sometime way before out-of-Africa.

    • RS says:


      If the penis ridge had once functioned in sperm comp, but had been very-near neutral for 17,000 years, I would think it would be somewhat deteriorated for lack of purifying selection.

      • Anonymous says:

        And of course you’d expect the chimp penis to be more, not less weird than the human one, and to exhibit the same kind of weirdness but to a larger degree.

  13. misdreavus says:

    Some possible compounding factors:

    1) This model assumes that the rate of dizygotic twinning is evenly distributed among females. It seems the frequency of DZ is too rare for anyone to calculate reliable estimates of heritability (has anyone seen otherwise? Jayman?), but we know that it does vary by ethnic group — from roughly 0.5% among East Asians to 2% and higher among west Africans. If somehow there is a correlation between cheating and the odds of dizygotic twinning (which might be the case), that should inflate the frequency of estimated events where SC (sperm competition) is possible. If otherwise, our upper bound is too high.

    2) Maybe women were a lot looser during Caveman Times, and have grown relatively chaste since the great Puritan revolution of the 1960s. Maybe the Aka Pygmies temporarily refrain from their swinging, wild orgies, and polyamorous behavior whenever all those meddling anthropologists are within earshot. Who knows, positive selection for sperm motility, etc. might have been a lot higher during the late Pleistocene, if not among hunter-gatherer peoples living today. (Pbbbbt)

    3) That paper by W.H. James is wrong, and guests on the Maury Show constitute a representative stratum of society. Sometimes, seventeen paternity tests on behalf of a single baby just aren’t good enough. (Maybe the manosphere is finally on to something, for once.)

    4) This tendency might vary strongly by ethnic group, so who knows? Do west Africans and Bantus have special adaptations that the rest of us haven’t heard about? Might be. (Once again, see: the Maury Show.)

  14. dave chamberlin says:

    Sperm competition in humans is insignificant, I got it. Lusty bastard makers and their female counterparts are far and few between and any scientist who thinks otherwise is watching too much Maury rather than looking at the actual data.

    Here’s an unrelated question that interests me. I have read that when men kill each other in a high enough percentage (directly after World war 1 this occurred in several European nations) the women in the following years have a much higher percentage of baby boys than the usual near 50%. Perfectly good evolutionary reason for this to happen, you need a higher percentage of men if you live in an incredibly violent society where the Malthusian limit is never approached because of a constant state of war with your neighbors. What I want to know is how in the hell this works, it’s bugged me for years and I thought maybe Greg knows why.

    • misdreavus says:

      More social scientists ought to watch the Maury show. They might learn a valuable lesson or two.

      • Sandgroper says:

        I didn’t know what it is, so I just caught a clip of it – there was a black girl crying and saying “You DID father my baby when I was 14!” and the audience were all sort of making that chimpanzee hooting noise that people make at football games.

        I think I’ve got the idea, approximately.

      • Sandgroper says:

        In case anyone takes offence, there were clips of white girls, but I just picked that one because she looked relatively bright and attractive. I didn’t notice any Asian-Americans, but I understand they have some issues.

    • gcochran9 says:

      It’s insignificant over the past few hundred years in the populations that have been looked at. It might be more important in some unexamined population, or at some other time. There was a blood-group survey from about 50 years ago that found realistic nonpaternity numbers for whites in rural Michigan, around 1%. But it found a 10% rate in blacks, and I don’t have recent evidence on rates in blacks, except for the Dogon.

      More research is needed!

    • anon says:

      This is Fischer’s principle.

      1. Fewer males means that a newborn male has better mating prospects than a newborn female, and therefore can expect to have more offspring.

      2. Therefore parents genetically disposed to produce males tend to have more than average numbers of grandchildren born to them.

      3.Therefore the genes for male-producing tendencies spread, and male births become more common.

    • gcochran9 says:

      The change is tiny, a quarter of a percent after WWI. And sometimes it goes the other way.

      • dave chamberlin says:

        Damn, I know better than to bother Cochran with a question that a little bit of research on wiki could have just as easily answered. Thanks for answering.

  15. j3morecharacters says:

    It may work in the same way as it always did in China and India.

    • Sandgroper says:

      Female infanticide happens even when the male/female ratio is already extremely high. It happens a lot less in Shanghai than in impoverished rural areas.

      The interesting thing to do is to plot the gender imbalance in India by state, and compare it to the level of literacy and economic well-being.

      • j3morecharacters says:

        That is a mystery. If the male/female ratio is extremely high, infanticiding the daughter and sparing the son must necessarily reduce the probable number of grandchildren. This idiocy should be bred out pretty fast, yet is widespread. Maybe having a larger number of fighters in the village is more important than any gender imbalance.

      • Toad says:

        Abortion, not infanticide. Before abortion was legal, the sex ratio was normal. Think of it as first generation “designer babies”.

      • Sandgroper says:

        Simply not true.

  16. Sandgroper says:

    “This idiocy should be bred out pretty fast” – You would expect so, wouldn’t you? No. It does flip over pretty fast with education/economic improvement/urbanisation, to the extent that people start preferring daughters, which suggests it has to do with labour for low-tech agriculture. Girls are a burden, just more mouths to feed, and when they grow up, they go away to a distant village to join their husband’s family. They don’t contribute to continuation of the male line.

    All of that changes with modernisation, education, urbanisation and ageing population. Girls stay closer to their parents, and look after them in old age. So far, this flipping has happened in Japan, Hong Kong and South Korea, and is starting to happen in Mainland China, first in the urban centres. Less female infanticide and less gender imbalance occurs in the (mostly southern) more prosperous Indian states with higher literacy rates, which also have lower population densities because educated girls have fewer children.

    • j3morecharacters says:

      Sand, Greg’s quantitative analysis of “sperm wars” has rather shaken my faith in sociological/behavioural explanations. May be on the Y chromosome sits an egoist, gender-Nazi gene that causes parents to love sons and genocyde daughters.

      • Sandgroper says:

        Agree. But there are real empirical observations – when people let girls go to school, the birth rate drops. With urbanisation, gender preference reverses.

        If there is such a gene, I evidently don’t have it. I told my wife if it was a boy I was going to flush it down the toilet. It was a girl.

    • Dahlia says:

      I’ve heard that it is exactly the opposite: wealthier, urbanite states have been more skewed. It has taken a change in morals to achieve a more natural gender ratio: South Korea has been out ahead due, I’ve heard, to growing Christianity or its influence.

      • muhry says:

        Here’s a passage from the article “The Global War Against Baby Girls:”

        “For the country as a whole, SRBs since 1982 have consistently been lowest for China’s cities, and highest for rural areas; in the 2005 mini-census, reported SRBs were roughly 123 for rural areas, 120 for towns, and 115 for cities.”

        However the author adds:

        “After all, fertility levels are decidedly lower in urban than in rural China, meaning that a smaller proportion of babies born in China’s cities are higher-parity births, which tend in China to be overwhelmingly male.”

        Fewer higher parity births in urban areas could be part of the reason for a more normal sex ratio in cities.

        The author also mentions that the sex ratio imbalances are largely a Han phenomenon whereas other ethnic groups have more normal ratios.

        Here’s the article if you want to read it.


      • Sandgroper says:

        And Japan and Hong Kong have been ahead of South Korea due to what, exactly?

      • Anonymous says:

        You’re correct about the ratios vis a via urban versus rural, but the article emphasizes it is not due to education.

        I read this when it came out (and others) and was trying to draw from memory in my initial comment. He always does great work. He may be the one who mentioned South Korea.

      • Sandgroper says:

        Muhry: A few points – The ratio was already high before the one child policy was introduced. Female infanticide, selling female children, etc were traditional cultural rural practices. Ethnic minorities get exclusions from the one child policy, believe it or not, as well as other affirmative action measures. The ratios in northern India are even higher than among Han, and female infanticide is common, particularly where sex-selective abortion is not an available option. Urbanisation coincides with both lower fertility and lower ratios – you can’t necessarily infer one from the other, except that both coincide with higher education and higher economic well-being.

        Reporting it as a global war is a bit alarmist, I think – he has not reported the countries that have more women than men. China now recognises it has a massive problem, and that the one child policy was a mistake (because it amplified existing traditional cultural practices) and probably unnecessary anyway – the big concern now is ageing population.

        The data for Hong Kong are skewed by the prevalence of Mainland women coming to Hong Kong to give birth, which has been a big problem. The local ratio has been dropping. Hong Kong has more women than men, even if you exclude female overseas contract workers.

      • Sandgroper says:

        He’s wrong – decrease in ration correlates with education. He seems keen to show this is all due to the availability of abortion. From your perspective that may be doing great work. If the thrust of the argument is that teaching Christian morality and banning abortion is the solution, good luck with that. What Christian missionaries in Africa are finding is that what works is actually letting little girls go to school.

      • Sandgroper says:

        I suggest the solution is to keep feeding vodka to Russian men, and export the surplus Chinese men to Siberia. That could solve all sorts of problems, apart from breeding a worrying new race of Central Asians.

        We could get Vlad the Very Short Impaler to teach them suitably manly Orthodox ethics and then cut them loose on Pussy Riot.

      • Dahlia says:

        I don’t have an opinion, just trusted the writers; Eberstadt has a good reputation. If other evidence came along that upended the consensus or changed Eberstadt’s mind, I’d go with it.

      • Dahlia says:

        I’ve done a little bit of reading on the one child policy over the years, but haven’t reread Eberstadt’s paper. These reports seem to come in batches and their authors run the gamut with respect to ideology.

        Some of the key things I remember (keyword, please anyone correct if mistaken) from the last go around were
        that some 40% of the world people act on an anti-daughter bias; moral outlook is the main driver of this; and South Korea has been out ahead in changing their anti-daughter bias.
        You mentioned Africa, but I can’t remember it being discussed as a place with an anti-daughter bias, the focus seemed mostly eastward, but my memory is bad.

      • Sandgroper says:

        Just look at the graph in the Eberstadt paper that muhry linked to, and tell me how South Korea has been out ahead of Japan on gender imbalance.

        While you are at it, overlay his map of India with the data on literacy and GDP by state. India has no one child policy and never did have.

        It depends what you mean by moral outlook. I don’t know what different people mean by that.

      • Sandgroper says:

        And don’t get me wrong, I’m not kicking at you – I have a powerful dislike of people with an anti-daughter bias.

      • Sandgroper says:

        The other thing you could try is overlaying this guy’s IQ map on Eberstadt’s map.

        http://theslittyeye.wordpress.com/2011/11/19/iq-geography-in-china/ It’s not that easy if you are not familiar with some of the detail, because Eberstadt’s map is by County, and Slitty Eye’s map is by Province – it helps to know, for example that people are not uniform in terms of fine grained culture/traditional practices across provinces, but you can see that in the provinces around Shanghai, for most of Shandong Province and a fair bit of Guangdong Province, the gender ratio is pretty close to normal, and that tends to correlate with IQ, which of course tends to correlate with educational level and level of income. The poorer hinterland to the west of the east coast higher developed/more prosperous belt is where you tend to see the really high areas of gender imbalance.

        What I see is a general trend by traditional culture, exacerbated by the one child policy, but strongly mediated by intelligence/educational level. Oddly, the one child policy has been unevenly applied but has tended to be less complied with in the areas where you see the greater gender imbalances. In China, the poor, more rural, less educated, less intelligent favour sons much more strongly. So, the less compliant have been having multiple sons, rather than mixed families. The more wealthy and educated are the ones who should have better access to ultrasound/pre-natal gender selection.

        That is pretty much the pattern I see in India too.

      • Sandgroper says:

        My reference to Africa was simply that what they have found there is that educating girls is about the only thing that strongly reduces overall birth rate.

      • Matt says:

        Of course, if Trivers-Willard is true, poorer folks should tend to have more daughters naturally, so I wonder if male offspring preference this isn’t to a certain degree just observation of a “low status” trait…? (and “elite emulation”, common to all humans and oh how those “bandwagoning” Chinese do love to emulate the Junzi, almost as much as Westerners like to make themselves better than him so they can punch him in the face).


        There’s a degree to which this makes sense – boys are the “riskier” strategy (even without polygamy, reproductive variance is higher), and risk is best pursued from a position of strength*, which is one of nutritional plenitude, which occurs more in high SES persons. Thus spontaneous miscarriage of female foetuses (of which sex selected abortion is an artificial extension).

        (Greg Cochran at this point is free to chime in with “This is insanity” if that’s the case).

        This would have a few odd implications historically –

        It would suggest that high SES men should have been more common historically than women, because high SES families tend to have boys and SES is heritable.

        And thus men of genius and social accomplishment really should have been historically somewhat more common than women (even assuming a man and woman from the same SES background are similar), and men would have been perceived as “more intelligent” (and actually been more intelligent).

        It would also place a higher brake on assortative mating by SES that may be less true today (if the milkmaids tend to have daughters and queens tend to have sons, then princes and milkmaids will be more common).

        This all may be less true in today’s world of society wide nutritional plenitude.

        *or alternatively perhaps a “nothing to lose” position of weakness, depending on how the curves work out – but extremes either way.

      • muhry says:


        I posted the article for the numbers – which I assumed to be correct – more so than for the author’s view point. I read the article when it came out and remembered the data, largely forgetting the rest.

        It sounds like we’re pretty much in agreement. I didn’t think male preference was new and I accept such things as schooling having contributed to helping normalize the ratios.

        If I read you correctly, you’re saying that the ratio difference between the Han and ethnic minorities is due to the one child policy. In which case the ethnic minorities probably don’t have a natural ratio, but rather a more normal one due to lesser rates of infanticide. After all, their rural living would probably be similar to Han, such as patrilocality, dowries and other reasons thought to contribute to male preference.

      • Sandgroper says:

        muhry: Nnnooo, I’m not really saying that, just that they don’t, so there is less likely to be some kind of distortion or magnification of traditional practices by a politically imposed rule – yeah, probably a more natural ratio. In some areas like Xinjiang and Tibet, the ratios are very obviously different, but a lot of the ethnic minority areas are not that easy to spot, and the Hui are basically closely integrated with the Han ethnic majority, to the extent of living in the same villages, and using the same shared communal fields.

        There are 55 ‘official’ ethnic minorities. If you look at how they plot genetically on a PCA plot, some plot precisely with the Han ethnic majority, so they look to be ‘ethnic’ in the sense that they are culturally different but genetically not really distinguishable (e.g. the Hui are Muslims, but they just look like Han Muslims, although you might find a small West Eurasian component if you looked at individual genomes), some look to be both genetically and largely culturally identical, they just wear big funny hats that look like birthday cakes or something, and some are quite different, and you get people like Uyghurs who are very obviously an admixed East-West Eurasian population.

        There are lots of Tibetans and Uyghurs who have migrated to cities in other provinces, so trying to plot the distribution of ethnic minorites now would be a nightmare.

        You get ‘odd’ groups, like matrilineals, where the women never marry or cohabit with adult males, they just have random visiting boyfriends when they feel like it, then they bring the kids up without any paternal intervention. They might make for an interesting ‘parenting’ study, I guess, although they are obviously pretty behaviourally different to begin with (and no, Jesse, there’s no sperm competition, because they space the visits far enough apart).

        The numbers feel pretty right, but just to note that recently the one child policy has been further relaxed, which people tend to read as an official admission that it was a bad idea, and that they are easing in gentle stages back towards a ‘no restriction’ policy, the total population is now on a downward trend, which can really accelerate once it gets going, and the Chinese government’s real concern now is ageing population, which I see Eberstadt has waxed lyrical on (and I personally think somewhat alarmingly or alarmistly) in another paper. Population projections have a way of *never* turning out to be correct. But then, Eberstadt is an economist, and naturally he would see any population decline as ‘bad news for growth in business’.

        The Japanese are supposed to become extinct around 2050, but I don’t believe that will actually happen either.

        Razib Khan’s take on this is that the one child policy probably made little difference in overall population terms – the trend is clearly for reduction in birth rate with urbanisation, and that with modernisation, economic development and massive rural migration to the cities, the birth rate in China would have dropped off naturally anyway.

        So, the one child policy was a knee-jerk panic reaction to the problem of population control, it was a mistake, they didn’t need to do it, and they probably just succeeded in somewhat magnifying an existing cultural problem in relation to male-preference, but which will diminish with modernisation/education/economic improvement (plus maybe the people who incline more to male preference have just ensured their genes will vanish from the gene pool by having sons who will never get to marry and have children because of the shortage of wives).

        Apparently they also admit that simplification of Chinese characters was also a mistake they didn’t need to make, but they’re stuck with it now. The folks in ‘Greater China’ are damned if they’re going to change from traditional characters, so the Mainlanders are just going to have to labour under that particular disadvantage.

        Eberstadt’s thing about the ‘dying bear’ is also somewhat out of date now – the Russian population has bottomed out and is starting to pick up again. I only glanced over that one, but he didn’t seem to take cognisance of the massive recent past migration out of Russia as an impact on the in-country population, which now seems to have stopped, pretty well.

        So – I’d say Eberstadt and Co. are fine as a viewpoint, but most people are pushing some ideology or other, and it never hurts to be skeptical and form your own view from how you read the data.

      • Sandgroper says:

        Spot the differences, apart from lots of funny hats. Some are obvious. The Kazakhs are lactose-tolerant. I don’t know about a lot of the others. Most won’t be, but it can be hard to tell without genotyping – some adults who are not can drink a glass of milk without any obvious digestive problems. Experts will tell you that is definitely not the case.

        The dowry thing works the other way with Han – the male has to pay the female’s parents. You would think this would make females more valuable, but evidently not. Apart from those odd matrilineal people, the patrilocality works the same way as elsewhere.

        Razib has posted quite a bit of stuff on genetic substructure in China over the years, e.g.: http://blogs.discovermagazine.com/gnxp/2009/11/population-substructure-within-china/#.Uz5sfa2Swvc


        Within Han, you also get traditional cultural substructure, as well as genetic substructure. An obvious example that a lot of people know about is that the Hakka never had female foot-binding, although they are genetically and linguistically obviously Han. But then foot-binding was confined to wealthy elite, and Hakka were traditionally confined to marginal agricultural land in a lot of areas, so maybe they just never got rich enough to afford the ‘luxury’ – it’s recounted as a source of cultural pride by modern Hakka, though.

        I had a PCA plot of ethnic minorities, but just can’t find it at the moment.

      • Sandgroper says:

        I think what I am trying to say is whether or not ethnic minorities exhibit more normal or natural ratios is not going to be a yes-no question – it depends on which ethnic minorities where, and there are a lot of them, many of whom are likely to be culturally indistinguishable from Han on this, but who may be a bit more ‘relaxed’ because of policy allowances made for ethnic minorities. Others, like Uyghurs, do look to be really culturally different, but that is not too surprising, because they obviously are genetically and culturally distant from Han.

        To actually resolve how the ratio varies with ethnic group would require huge amounts of very accurate data down to almost a personal level, and I question whether such data would be available or accessible, aside from some very broad inferences, like Uyghurs and Tibetans are obviously different from Han, particularly given that how closely the one child policy has been applied and complied with has varied a lot anyway.

      • Sandgroper says:

        This is quite informative. I wish Zack had not used yellow in the PCA plots. But then, I’m appreciative that he does this stuff, so I’ll shut up about it.


        It is evident that Tujia, She and Miao are ancestrally not too different from Southern Han. There is a lot of within-group difference in Miao (Hmong) from Christian to animist, so they clearly cannot be regarded as a culturally homogeneous group, even though they are grouped as one ethnic minority in China.

        Naxi and Yi are not too ancestrally different from Northern Han. There is a cline running from Northern Han to Yakut, which is not surprising, and a cline running from Northern Han to Southern Han and then Dai . This is also not surprising.

      • dave chamberlin says:

        I like reading your analysis on China Sandgroper, it is (not by my choice) a Euro-centric world I live in and I rarely read much about the 1.43 billion people who live in China. Hans Rosling’s excellent blog Gapminder gives me the following information on the birth rate in China.
        1992 2.2 children per women
        1994 1.9
        1998 through 2002 1.5
        2003 through 2010 1.6
        2011 and 2012 1.7

        It amazes me that there are 100 articles on the worsening environment and little to no reporting on plunging birth rates around the world, even though they are hopelessly intertwined issues. China is not exceptional in it’s birthrate, similar or lower rates are common throughout Europe. I wonder if this bias towards having sons when you expect to have only one child will change. In the west at least you are far more likely to be taken care of in your old age by a daughter than by a son. Although future projections of world population have been notoriously inaccurate in the past thanks to unpredictable changes in family sizes, there is a marked stabilization in birth rates in first and second world countries in the last few decades, as the above information on China indicates. We are now projected to reach a population maximum of 9 billion by 2070 and then start decreasing rapidly. I actually think this population maximum will happen earlier than this because the wild card in these projections (the shit is hitting the fan in Africa and the middle east) is not included. I regret to inform the gloom and doom fan club that the impending violence from those places will spread to the rest of the world is simply not supported by the facts. Powerful advanced countries have never been over run by weak backwards countries, it works the other way around if it works at all. A far more likely scenario is the nations that overpopulate to levels that trigger utter anarchy and violence will be sealed off. That may sound cold hearted but I’m not into wishful thinking or it’s opposite, dwelling on worst case scenarios. Extrapolate out a 1.7 children per family birth rate for just one century and you get a population reduction of roughly 50%. No sense in trying to refine the projections as no one knows what is going to happen regarding either global warming or chaos and population crash in Africa. What does look predictable is most of the world is finally getting it’s shit together regarding reducing population in a peaceful and voluntary method.

      • Sandgroper says:

        Dave: “In the west at least you are far more likely to be taken care of in your old age by a daughter than by a son.” Same. Girls are girls.

      • Sandgroper says:

        Dave: I keep waiting for the Beautiful Pink Army to invade, but they just won’t do it, damn it.

      • dave chamberlin says:

        You are better than I am Greg at thinking these things through. Everyone has been hopelessly wrong so far in projecting future population trends, why should I be any different. But still it is an incredibly important issue so I think it should be speculated on, with a very humble attitude as to how accurate such predictions can be constantly added.

  17. j3morecharacters says:

    We beautiful people make daughters.

  18. Paul Conroy says:

    Completely off-topic, but I just read this:
    “Sexuality and Gender Role in Autism Spectrum Disorder: A Case Control Study”

    It basically says that being on the autistic spectrum (ASD) is correlated with being gay.

    But IMO, ASD is highly genetically inherited, and is basically selection for higher IQ – albeit crude selection.

    Then it occurred to me that maybe being gay is also related to some sort of selection for higher IQ – via another crude selection.

    This should be easy to test:
    Q: Are high IQ groups more likely to have ASD members?
    Q: Are high IQ groups more likely to have gay members?

    • 420blazeitfgt says:

      iirc autistics don’t have higher IQs, but asperger’s sufferers do only because the definition of asperger’s says that your IQ has to be =>70

    • Matt says:

      ASD men are less masculine in the sense of having less social confidence for sure though. Male ability to “read” others maybe needs to reach a certain minimum level in order to have the social confidence to behave in a bombastic alpha male style (or even the less bombastic beta male style), even though “reading” others is a generally female trait.

      ASD men, if high functioning could also often tend to behave in a rationalistic (rather than socially informed fashion) and systematizing way that would impede masculine behavior. “Being a brute is so irrational, especially since I’m obsessed with maths and not obsessed with dominating others, so I won’t do it”.

      I could also see a certain kind of “autist” moving towards homosexuality – they may think if you can’t read women and successfully mate with them, why not take a “rational”, instrumental view of sexual pleasure and release? (shades of Waterhouse’s rationalistic attitude towards “self relief” in Cryptonomicon here, to a lesser extent that novel’s depiction of Turing). Although I will not press this speculation strongly if questioned by the agents of the Gay Germ Jihad!

    • J says:

      Prisoners and prep schoolers engage in gay sex because they don’t have access to females. Maybe isolation from women for prolonged periods of time triggers the same surrogate drive in ‘Spergs, and because they don’t know of an alternative, and because it’s trendy to identify as A Gay Man or, in the cuckoo land of the internet, a pansexual cisgendered freakshow, they decide they’re committed homosexuals.

      Just one hypothesis among many.

  19. Greying Wanderer says:

    “People in jail should have lower number of children surviving”

    I don’t know what the stats are but some thoughts based on anecdotal stuff…

    1) Jail may have been more successful at achieving this in the past and the success dipped as the criminal population outstripped the prison population from the 60s onwards. Now the prison population has caught up the eugenic effect of prison might start to kick in again however…

    2) One of the counter-factors is if you have mass unemployment among a segment of the population then *relatively speaking* that makes crime very much pay in terms of reproductive success among that segment and that effect might outweigh the effect of prison.

    3) The age of puberty going down means the juvenile limit.in the criminal justice system has a bigger effect i.e. they might already have had their kids before they reach the age of adult sentencing.

    4) Gang culture magnifies (2) and (3) because gang culture means girls don’t get to choose who the father is.

    5) Welfare system. Before welfare criminal’s kids would have a hard time thriving.

    So for prison to have the eugenic effect it had in the past I think it requires:
    1) The correct number of prison places. (This requires very low levels of young male immigration or constant prison building.)
    2) Increasing the age of puberty or lowering the criminal juvenile age.
    3) Full employment.
    4) Serious attempt to deal with gang culture.
    5) Welfare penalties on kids of criminals.

    All of which used to be the case one way or another.

    • Greying Wanderer says:

      “All of which used to be the case one way or another.”

      Rather, all of which used to be public policy.

  20. Greying Wanderer says:

    “In much the same way, Baker and Bellis talked about a study in northern England (in a block of flats in Liverpool) that had a nonpaternity rate of 20-30% – but that study has never been been published, and I think that it never took place. I think Baker and Bellis made it up.”

    It’s highly likely that there are tower blocks in Liverpool where 20-30% of the men are living with women whose kids aren’t theirs but the men know they’re not theirs because the women already had the kids when the men started shacking up with them. I wonder if it’s possible for researchers from the old “normal” culture where couples were couples to not understand the rate of relationship churn in the welfare sub-culture and so test non-paternity without actually asking if the current bf/husband is fully aware they’re not the actual father?


    Jesse Marczyk
    “Just so we’re clear – again – I have been arguing that *facultative*, not obligatory, adaptations for some mild degree of sperm competition *might* exist in humans”

    There are a lot of cultural adaptations to avoid non-paternity so even if sperm competition was important once it seems likely the pressure to select for it will have been going down for a long time.

    Also if the advantage of any trait requires both increased reproduction *and* the offspring going on to reproduce then (anecdotal again) but I’d say known or even suspected non-paternity often has a dramatic and damaging influence on male attitude to their purported offspring so I don’t think the benefit to the cuckolder is as clear-cut as it might first appear.


    @female infanticide

    If a village can support 200 adults then killing 20% of the infant females and forcing the surviving 80% to have 10/8 more children than they otherwise would have means you only have 180 adults which gives you a solid Malthusian buffer with the same number of male workers/defenders. It makes perfect, if cruel, sense. Increased prosperity would reduce the need to do this – with a time lag as culture adapted to the new conditions.

    What’s interesting is how/why Europe went down a different route (hajnal late marriage etc).


    “Male ability to “read” others maybe needs to reach a certain minimum level in order to have the social confidence to behave in a bombastic alpha male style (or even the less bombastic beta male style), even though “reading” others is a generally female trait.”

    I’d say this was very true. Being able to read others is very useful in individual conflict and even more so in leading small group conflict.

    • Sandgroper says:

      The more I look for the Hajnal line, the more I don’t find it.

      • Greying Wanderer says:

        I guess the first step would be looking for where female infanticide wasn’t the norm.

      • Sandgroper says:

        That’s difficult. Which European cultures killed female babies as the norm during which time period? I don’t actually know, but I’m guessing that Finns, Southern Italians and Irish farmers didn’t do it routinely. I’ve heard that Russians ate their babies during famines, but I have never seen the assertion backed by any data, and there didn’t seem to be any suggestion that they preferred to eat female babies. Even if they did, how would you know?

        It’s routinely theorised that all nomadic hunter-gatherers practiced infanticide, but I have never seen any suggestion that it was selective. And I never heard anyone suggest it for nomadic herders/Steppe people, even before they used horses for transport.

        Of course we could just assert stuff. We wouldn’t be lonely.

      • Sandgroper says:

        I’m inclined to like your idea, that it’s something that happens in labour-intensive farming communities operating at the Malthusian limit, but I don’t know how we demonstrate that, unless we could get some data from, say, pre- and post-Black Death populations in the same region.

  21. RS says:

    > And of course you’d expect the chimp penis to be more, not less weird than the human one, and to exhibit the same kind of weirdness but to a larger degree.

    You can’t count on that, evolution being highly contingent. Some adaptation can be quite fitness-enhancing in abstracto, and yet never actually appear.

  22. georgesdelatour says:

    “Powerful advanced countries have never been over run by weak backwards countries.”

    The adjectives “powerful” and “weak” make your claim a tautology rather than an empirical observation. If instead you’d simply said that “relatively advanced countries have never been overrun by relatively backwards countries” you’d realise that statement is wrong.

    Athens lost the Peloponnesian War; The western Roman Empire fell to the Barbarians; Muhammad’s Arab tribesmen destroyed the Sassanid Empire and most of the Byzantine Empire; Genghis Khan’s warriors mortally wounded destroyed the Abbasid Empire and his grandson took over China, etc.

  23. Gamera says:

    “Maybe everything was different back before recorded history.”
    Yes it was.
    There was no widely available birth control.
    So even in 1992 when birth control was available to everyone in Italy, you estimate that half a percent of fraternal twins born there had different fathers. Is there any reason to suspect that the greater portion of the potential sperm competition wasn’t tied up in latex bladders and tossed in the trash?

    • misdreavus says:

      Yet another commenter who failed to understand the premise of post, entirely.

      Doesn’t matter how many sperm are tied up in prophylactic tubes. Sperm competition can only occur when the sperm enter the birth canal. You’re confusing two very different things, behaviors that influence the odds misattributed paternity and sperm competition itself. Only what’s in the birth canal counts!

  24. j3morecharacters says:

    In military terms, Sparta was more advanced than Athens. If the competition would have been in stageing tragedies or in philosophical debate, Athens probably whould have bested the Laconic Spartans.

  25. davmlaw says:

    Nobody’s looked at whether vanishing twin syndrome is more common between half sibs than sibs. Maybe sperm competition is being underestimated by foetal competition!

  26. luxlucis7 says:

    I randomly ended up here looking for more information while reading the book “sperm wars” and loved your blog!
    Do get in touch if you’d like to do something together : thepowermoves dot com

Leave a Reply

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

WordPress.com Logo

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

Twitter picture

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

Facebook photo

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

Connecting to %s