Why Genghis Khan won’t have had 16 million descendants.

Last Thursday, I posed the mystery of whether there are in fact 16 million direct male descendants of Genghis Khan. This factoid came from a 2003 study of some 2000 Central Asian men, of which 8% were found to share a common male ancestor around the year 1000 AD, give or take a century.

The adherents of this factoid basically point out what a stud Genghis Khan and his sons were. He had many wives himself and some of his sons and grandsons shagged everything they could lay their hands on in the territories conquered by the Golden Horde  led by his grandson Batu Khan. Some of them, like Kublai Khan, would easily have had hundreds of sons and Genghis himself probably had dozens.

The first hint of doubt that you should have is this whole business of the direct male line and what that implies: if you believe that 8% currently is directly from Genghis then you by implication believe they were not by his 3 brothers  and 2 half-brothers, who also were high-up in the Mongol hierarchy. You by implication believe they were not by the many uncles on his father side, nor indeed from any of the Mongol men who will have had the same male ancestor as Genghis Khan earlier in the line. You see, the Y-Chromosome attributable to Genghis is close to the one of his own father, his grandfather, his great grandfather, etc. You basically have to believe that there are relatively large and unusual genetic differences between his Y-chromosome and that of all the other male offspring of his own male ancestors to pick him as the sole source. Only then could you say the 8% were his and not the result of himself and a whole group of male  Mongols who shared a very similar Y-chromosome of an earlier male ancestor. Just on the genetic side, the odds of such a thing are pretty slim already.

Consider how many male Mongols you would thus have to ‘wish away’: Genghis came from an important family. Up his family tree were prolific male shaggers. What are the odds then that the contribution all the other men coming from the same male line would be outside the 8%? Miniscule, absolutely miniscule.

Consider further how little Genghis could actually himself have done during the conquest of the Khwarezmian Empire around 1219. Yes, he apparently reserved the most beautiful women for himself. But he led an army of up to 250,000 men, all raping and siring as well. Now, I don’t care what kind of super-stud Genghis was and how often he got first dibs on the prettiest women: he will not have been able to out-shag his fellow conquerors by much more than a factor of 10 (at best). Yes, he will have had many more women, but he will have had them once, whereas all his commanders and soldiers would have taken wives and concubines and looked after them for years, protecting the sons that came from those unions.

Consider moreover how the male descendants of Genghis would have been treated. Certainly his official sons had great status and protection. But any son he bore as a result of rape or one-night arrangements will have subsequently stood out like a sore thumb: the women would go back to their families and get other men who would not have looked kindly upon raising the son of someone else. You can be sure that many an illegitimate son of Genghis would have met an unfortunate and untimely death. It thus seems very likely that all the ‘illegitimate’ sons of Genghis would have had much less further genetic success than the average male in that time. The same goes for all the other illegitimate children of the recognised direct off-spring of Genghis.

It is this aspect of no special treatment for all but the recognised off-spring of Genghis that is crucial: the average male around at that time has 10 to 20 living male ancestors now. In order to get a higher success rate one needs some real advantage. For Genghis to have had 16 million if all the illegitimate ones had no special procreation advantage (and quite likely a big disadvantage) truly means that he and his recognised male offspring must have personally sired close to a million sons in that era. That is just beyond all reason. Even if 5 generations of his official descendants shagged themselves silly day and night, they couldn’t get close to that kind of number.

The only remaining possibility would then be that there was something special about Genghis’ chromosomes giving him and all his male descendants a true advantage in terms of procreation, even if they were illegitimate and had no help in terms of wealth or position at the start of life. That would not only imply that the current human race has supermen amongst it with some advantage that one should be able to measure, but in order to attribute them to just Genghis you then still need to believe this distinct true advantage was due to a mutation when the Y-chromosome was copied from Genghis’ father to Genghis himself. Very, very far fetched.

Put all this together and think of what happened in that time: up to 250,000 men raiding through Central Asia, making examples out of whole cities by slaughtering everyone and occupying huge territories with a population in the tens of millions. To the victor the spoils, with all the desirable young women married to or raped by the Mongols and their allies, of which Genghis Khan was but a single man. And the other men in the region of course were not all dead either, so it is not just the other 250,000 that vie for the 8% supposedly starting with Genghis. Various other waves of conquerors have swept through the region since.

What are the odds that Genghis Khan and his direct male descendants then managed a whole 8% for themselves? Or even Genghis Khan and his 10 closets male relatives? I would say pretty slim. What, on the other hand, are the odds of a fairly large slice of the Mongol horde managing 8% for themselves as a block of men? Pretty good, you would have to say.

What does this mean? Well, it means that the common male ancestor of whom 8% of all East Asian men derive is most likely not Genghis Khan but rather a male ancestor further up in his male line and the male line of a whole slice of Mongol men. Some guy living around the year 1000 or even earlier who will, by the time of the Khwarezmian Empire and the Golden Horde that swept through East Asia, had hundreds or thousands of male descendants in prominent places in the Mongol genetic pool, poised to take advantage of the raping and other female-acquiring possibilities that came with that massive conquest.

You should hence not be surprised to know that after writing this I was alerted by my research assistant Miriam Carter to two genetic studies doubting the 8% number, though purely on the basis of genetics. A 2012 study by Russian geneticists suggests the direct male descendants of Genghis Khan have all but died out (which I don’t believe either). Perhaps more importantly, another study (Abilev et al. in Human Biology 2012) attributes the 16 million to a whole tribe currently living in Kazachstan, the Kerey, who functioned as elite units in the Mongol armies and were indeed exceedingly well poised to spread their genes around during the Mongol conquests. That tribe and the Khan family were close so it is possible that the earlier male ancestor spawned both the males from the Khan family and much of the Kerey.

One has to wonder whether the authors of the original study were fully aware of how unlikely it was that Genghis Khan was really the direct male ancestor of the 8%. After all, they labelled their study ‘The Genetic Legacy of the Mongols’ and not ‘The genetic legacy of Genghis Khan’. In their study they are furthermore quite careful to acknowledge that their best-guess for the period in which the common male ancestor lived was the year 1000, not the year 1200 which was when Genghis Khan was still around. He is on the outer fringes of their confidence bands. Yet, the authors do say explicitly that Genghis Khan fits the bill for the kind of person they are looking for and it is that what was taken up by the world’s media. The factoid that went into the world media certainly helped to get more attention for their study. It made their careers.

So, once again champagne for the many who had similar doubts. Diet water for the rest.

This entry was posted in Uncategorized. Bookmark the permalink.

24 Responses to Why Genghis Khan won’t have had 16 million descendants.

  1. Alan says:

    If you’re excluding illegitimate sons, you need to bear in mind that by the fifteenth century the legitimate line had almost completely died out. Dayan Khan 1464–1517 was universally recognised as the last direct descendant of Chinghiz Khan through Ögedei and Mongke, although that would not have included the branch lines descended through Jochi and Chaghatai.

    • Paul Frijters says:

      we are talking the genetic male line, so the illegitimate ones would surely be the vast majority. My argument quite strongly relies on saying the illegitimate ones will have had no special advantages relative to other men born in that time, but I would not go so far as to suggest they will all have died out. After all, Genghis alone reportedly had up to 2000 females. If you as a rule of thumb say that 10% of his one-off encounters resulted in pregnancy, you get 100 illegitimate sons. Nowhere close to a million but hard to believe that kind of number would not secure him some genetic male representation down the ages (which is why I am skeptical of the first Russian study. I suspect the family line they use to identify the Genghis Y-chromosome will not actually go back to Genghis at all). And some of his recognized progeny were even busier…

      The naming conventions are interesting, as your use of the name Chinghiz makes clear. They guy went by many names but the ones used mostly are Timujin and Genghis Khan. I use the latter name so Western people living today know who I am talking about.

  2. Alan says:

    Temujin was a personal name. Chingiz Khan was a title assumed after the khuriltai in 1206. Chingiz is thought to be a better approximation of the sound. The meaning of Genghis/Chingiz is unclear.

    Previous aspirants to the Mongol leadership, including Jamuqa, Temujin’s great rival, had used ‘Gur Khan’, ‘supreme khan’. Temujin’s father-in-law was granted the title Ong Khan, ‘royal khan’, by the Jin nomad dynasty that controlled North China.

  3. JM says:

    Show me the math. Let’s assume that Ghenghis has 32 children (leaving aside the social construct of “legitimate hier”), that each of those through social privilege has 16, then 8 in the next generation, then 4, then all asymptote to 2 per generation thereafter (quite modern reproduction levels, those Mongols, hey?)

    It only takes 14 generations, or about 280 years to get to 16 million offspring. Now this leaves out a lot of detail – disease, disability, death in war, etc – but it does show that 16 million is not an outlandish number, particularly because we’ve got more than 280 years available.

    Now at this point this simple model is starting to skate pretty close to some uncomfortable truths. In evolution, the advantages of the parent – whether genetic or social – are visited upon the children. There is a reversion to the mean but that can be countered by social advantage. It all depends on how stable the social advantage is across the generations.

    • Paul frijters says:

      JM,
      Take your own maths. 2 kids is one boy, meaning the Y-chromosome no longer begets more descendants during the 2-kid period. Within your maths, the number of male descendants is then 16 times 8 times 4 times 2, which is 1024.

      Yes, survival of social advantage is key.

      • JM says:

        Paul, on the face of it, yes you are right. But not quite.

        The question is “descendants”, not “direct male lineage” which is what happens if you only follow the Y-chromosome. The girls get a look in too.

        So if you count the way I was, and even allow for pretty high infant mortality rates, 16 million is not outlandish if you allow for a few more generations.

        You actually concede this point yourself when you say “the commo male ancestor of whom 8% of all East Asian men derive is most likely not Genghis Khan but rather a male ancestor further up in his male line

        I would only disagree with the “most likely” bit. Who is more likely to be the common male ancestor? An extremely successful and wealthy “shagger” like ol’ GK? Or his relatively less successful and obscure great-grandfather?

        I’d put my money on Ghengis.

        Now the other thing is that ascribing responsibility to someone “further up the line” is in itself a bit of a cop out. Go back enough generations and we’re all related. Homo Sapiens is reckoned at one time to have been only a small group of around a dozen individuals in a valley in Africa. (ie. the “Eve” hypothesis)

        Go back further, and you and I are both related to the gorillas in the Melbourne Zoo. (I actually had the pleasure of explaining this point to my daughter when she was very young and we were watching a gorilla in a zoo.)

        There is no reason to doubt IMHO that a famously successful shagger like Ghengkis wouldn’t have 16 million descendants after a few hundred years. (His idiot brother who never got on with girls and died in battle… well that’s a different story)

        • Paul Frijters says:

          Genghis Khan might have had way more than 16 million descendants carrying some of his genes (indeed, very likely), but not 16 million direct male descendants who will have the same bit of his Y-chromosome (which was the question in the original post and the question in the first sentence of this one).
          The original post already covers all the stuff about common ancestry.

        • Michael Price says:

          “I would only disagree with the “most likely” bit. Who is more likely to be the common male ancestor? An extremely successful and wealthy “shagger” like ol’ GK? Or his relatively less successful and obscure great-grandfather?”
          His great-grandfather. Firstly the amount of mutation supports an origin about 10th century, not 12th. Secondly any descendents of GK are also descendants of his great-grandfather. Thirdly when GK won, his relatives, including his fellow carriers of a particular Y chromosome did too. Well except for the ones he killed. But the rest did very well in the mating opportunities stakes.

  4. Alan says:

    Of course, legitimate heir is a social construct. It is nevertheless a non-trivial fact that the acknowledged male descendants of Chingiz Khan though 2 of the 4 acknowledged sons were reduced to a single individual by 1487. It’s possibly also non-trivial to this inquiry that 3 of the 4 sons appear to have died of alcohol poisoning.

  5. Alan says:

    What is the historical evidence for Chingiz Khan as a famous shagger? The Mongols did not have a harem tradition and did not adopt it until Ögedei became great khan. Jochi, the eldest son was born after his mother had been kidnapped and spent the entire pregnancy in the hands of another tribe.

    • Paul Frijters says:

      his biography, his multiple official wives, the details of the conquests. Why? Have you got a reason to suspect some grand historical falsification of the records one can find about him online?

      • Alan says:

        I should have answered this some time ago. Chingiz Khan’s reputation is massively overstated as famous shagger, as destroyer of cities and the rest.

        Partly it is confusion with Timur (aka Tamerlane) partly it is the result of wildly inflated stories by Persian historians in the pay of the Mongol Ilkhans. For whatever reason, the Mongol government in Iran actively encouraged belief in their propensity for genocide. Later Muslim sources report figures that are simply impossible like the Mongols killing 97 million (97 million is not a typo) people in Baghdad when that city fell to Hülagu.

        The Chinese, Indian and Central Asian sources describe Chingiz Khan as a barbarian conqueror but they do not report any special brutality or an overweening taste for shagging. The Secret History of the Mongols, a document intended only for immediate members of the Chingizkhanid dynasty, makes no mention of genocide or famous shagging. Chaucer devotes an entire chapter, The Squire’s Tale, to Chingiz Khan under the name Cambyuskan:

        This noble king was called Genghis Khan,
        Who in his time was of great renown
        That there was nowhere in no region
        So excellent a lord in all things.
        He lacked nothing that belonged to a king.
        As of the sect of which he was born
        He kept his law, to which that he was sworn.
        And thereto he was hardy, wise, and rich,
        And piteous and just, always liked;
        Soothe of his word, benign, and honorable,
        Of his courage as any center stable;
        Young, fresh, and strong, in arms desirous
        As any bachelor of all his house.
        A fair person he was and fortunate,
        And kept always so well royal estate
        That there was nowhere such another man.
        This noble king, the Tartar Genghis Khan.

        You may care to read Genghis Khan and the Making of the Modern World by Jack Weatherford. While not all his arguments are persuasive, overall he makes a reasonable case for a reconsideration of Chingiz Khan’s reputation.

        The Mongol empire was structured around chokepoints, trade routes, and highly mobile forces that could operate at intercontinental distances and attack from unexpected directions. Substitute ships for horses and you begin to get some idea of how the Mongol empire worked.

        It was said that a naked virgin could carry a bag of gold from Vienna to Seoul without being troubled. In many ways the Pax Mongolica looks a whole lot more like our world of globalised violence and globalised trade than we care to think.

  6. Alan says:

    What biography? What was the number of official wives? What details of the conquest?

    • Paul Frijters says:

      I am sure you can find them yourself. I find this a tad childish.
      If you are serious though, start from the wikipedia entry (http://en.wikipedia.org/wiki/Genghis_Khan), which is admirably full of other links. On the wives and kids front it says


      As previously arranged by his father, Temujin married Börte of the Onggirat tribe when he was around 16 in order to cement alliances between their respective tribes. Soon after Börte’s marriage to Temujin, she was kidnapped by the Merkits, and reportedly given away as a wife. Temüjin rescued her with the help of his friend and future rival, Jamukha, and his protector, Toghrul Khan of the Kerait tribe. She gave birth to a son, Jochi (1185–1226), nine months later, clouding the issue of his parentage. Despite speculation over Jochi, Börte would be his only empress, though Temujin did follow tradition by taking several morganatic wives.[21] Börte had three more sons, Chagatai (1187—1241), Ögedei (1189—1241), and Tolui (1190–1232). Genghis Khan also had many other children with his other wives, but they were excluded from the succession. While the names of sons were documented, daughters were not. The names of at least six daughters are known, and while they played significant roles behind the scenes during his lifetime, no documents have survived that definitively provide the number or names of daughters born to the consorts of Genghis Khan.[22]”

  7. Alan says:

    The problem is that biography does not read ‘famous shagger’. If anything that biography tells us about someone who was extraordinarily restrained in terms of the number of wives and children by comparison with his contemporaries. Where Chingiz Khan’s behaviour is unusual for his times, it actually points in the opposite direction. Most contemporary rulers would not have kept Bortë as empress or recognised Jochi as legitimate. If you were to research the matter a little more deeply, by for instance investigating the series of nuptial decrees issued at the marriage of each daughter, you would discover someone who appears to have held an attitude to women that was fairly exotic for its time. I am not minimising the the impact of the Mongol conquest, just questioning the famous shagger meme.

    A whole lot gets written about ‘barbarians’ in general, Mongols in particular, and Chingiz Khan most particularly of all, that just does not correspond a lot to any known facts. So much so that even questioning the horror show biography gets characterised as childish.

    The sole evidence advanced for the famous shagger thread in this thread is that Chingiz Khan was a famous shagger because Chingiz Khan was a famous shaggier.

  8. Alan says:

    If you were seriously looking for the Asiatic megashagger, have a look at Kublai Khan. Same genetic line. Massive harem. No sign of his grandfather’s attitude to women.

  9. JM says:

    I’m getting confused about what this argument is about. Is it about fact or interpretation?

    It seems people accept that 16 million males in central Asia currently living have a common male ancestor from about 1000AD. That seems to be the fact.

    An interpretation applied to this fact is that common male ancestor could well be Ghengis Khan, who sticks out in history as a pretty good candidate.

    Interpretations can never be proven with certainty, only likelihood. However, the original post seems to go through hoops to deny any likelihood whatsoever to the completely plausible hypothesis that this common male ancestor was a certain prominent historical figure. In fact, it argues that the common male ancestor is more likely to be a far less prominent historical figure who was nonetheless closely related.

    Personally I think that is an absurd position to adopt. Occams Razor. Go for the simplest and most likely answer, but recognize that is contingent upon better evidence that may arise.

    • part of the point is that Ghengis Khan does not stick out as a pretty good candidate at all for the title of the sole originator of the 16 million. Too late in history and requiring the fantastical coincidence that he was a mutation with respect to his dad.

      So your Ocram’s razor just boils down to ‘Why not pick someone famous from 2 centuries later when you dont know anyone from 1000AD?’. Lazy, just like the journos that bought the original line..

  10. joe says:

    I like to comment on.. we seem to forget , war at that time…

    it’s Winner takes all…. not only the wealth, but the women were raped.

    granted , I don’t believed GK populated all 16millions by him self… but substantial of his men did carried out such atrocity across Asia..Eastern Europe..

  11. John adam says:

    If he and his sons just had 100 offspring between them, and if there are roughly 18 generations since then, if each offspring had only 2 children , math says its possible he could have had many many more than 16 million carrying his code. He Would have had about 26,000,000 if you follow this math, since people in asian countries typically had many More than 2 children(culturally if daughters were born the couple would keep trying until they had a son, the 1 child per family rule is only a few generations old, and a direct response to this custom). So yea, 16,000,000 sounds right. i’m going to go with the Phd’s that devoted years to this study, not some internet troll that says__Nu-Uh! i call BULLSHIT! … Lots Of Love… JA.

  12. Peter Booth says:

    Fascinating discussion. It’s easy to accept the main thesis that Ghenghis Khan can’t have had anywhere near 16 million descendants. It’s also easy to accept that the individuals from that time who had most descendants won’t necessarily be “celebrities” who are well known today.

    What about the details ? What sense can we make of “the average man alive at that time has 10 to 20 descendants alive today?” If we were to list every man alive during ghenghis khans lifetime, and somehow compute or estimate the number of present day living descendants what would be the characteristics of that variable? It’s minimum value would be 0. It’s a discrete function – no one has 1.5 descendants. Would it be multimodal? would it have mode of zero and monotonivalky decreasing or have a mode greater than zero? How long a tail would it have?

    I wonder whether a dataset like familysearch.org, with its well know issues with duplicate personas and realistic length of say 500 years have similar properties?

  13. peter says:

    The comments miss questioning what I think is the key point of the original argument: “any son he bore as a result of rape or one-night arrangements will have subsequently stood out like a sore thumb: the women would go back to their families and get other men who would not have looked kindly upon raising the son of someone else. You can be sure that many an illegitimate son of Genghis would have met an unfortunate and untimely death. ”

    What is your evidence for this? This strikes me as projecting back in time our current-day values and preferences. Why would being a bastard son of somebody so incredibly famous at the time not be an advantage? The notion that many of his bastard sons would meet an untimely death (presumably in contrast to whatever mortality rates affected males at the time in general) seems to me to be an absurd argument.

    In my own (personal) genealogy, I trace back to bastard children of old Scandinavian royalty. Looking at the names and titles of that progeny, and their continued success to marry “good families” down the generations, I would say the opposite would have been true.

    By the same reasoning, somebody like a great conqueror leaving bastard children from “rape” also strikes me as modern-day projecting. I doubt the Great Khan needed to “rape” to procreate, by the same above reasoning. In fact, I would doubt that the difference between a “legitimate” and “illegitimate” son back in those days, and in a Mongolian conquering culture no less, would be anything like our view today.

    If that is not the case, do you have any sources to support your contention? That the values / cultures in question would reasonably have responded as you describe?

    • Paul Frijters says:

      yes, that is the key bit in the argument. And yes, we’re arguing plausibility here since there are no records that allow us to trace families that long ago (nor should we believe any records going back that long, or even more recent one, since genetic testing is new and cuckolding is old).

      Why would I think bastard and illegitimate children would not have been looked upon favorably? Where did I get that outlandish notion, without even giving a reference? Well, for instance, in the UK it only became possible in 1926 for a bastard to become legitimate. In the centuries before then? Not so easy. In the UK bastards could not inherit. Several laws on inheritances have applied.
      The laws throughout Asia? Well, there have been many as we’re talking about a long period. But marriage was an institution and illegitimacy not looked upon well pretty much anywhere.

      Deeper reasons for the negative effects of illegitimacy? Simple evolution: looking after the children of other men is not a winning evolutionary strategy.

      And of course there are famous exceptions, but note for my argument that even a few generations of ‘equal footing’ with the legitimate children is not enough. to get to 8% of the population one needs many generations of illegitimates having special advantages.

Leave a Reply

Your email address will not be published. Required fields are marked *

Notify me of followup comments via e-mail. You can also subscribe without commenting.