The Unz Review: An Alternative Media Selection
A Collection of Interesting, Important, and Controversial Perspectives Largely Excluded from the American Mainstream Media
 TeasersiSteve Blog
Big, if True: Are Humans Even More Genetically Diverse Racially Than Long Assumed?
🔊 Listen RSS
Email This Page to Someone

 Remember My Information



=>

Bookmark Toggle AllToCAdd to LibraryRemove from Library • BShow CommentNext New CommentNext New ReplyRead More
ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
AgreeDisagreeLOLTroll
These buttons register your public Agreement, Disagreement, Troll, or LOL with the selected comment. They are ONLY available to recent, frequent commenters who have saved their Name+Email using the 'Remember My Information' checkbox, and may also ONLY be used once per hour.
Ignore Commenter Follow Commenter
Search Text Case Sensitive  Exact Words  Include Comments
List of Bookmarks

Here’s an un-peer reviewed new preprint from researchers at Duke and Princeton:

New kinship and FST estimates reveal higher levels of differentiation in the global human population

by Alejandro Ochoa, John D. Storey

doi: https://doi.org/10.1101/653279
This article is a preprint and has not been peer-reviewed

Abstract
Kinship coefficients and FST, which measure genetic relatedness and the overall population structure, respectively, have important biomedical applications. However, existing estimators are only accurate under restrictive conditions that most natural population structures do not satisfy. We recently derived new kinship and FST estimators for arbitrary population structures. Our estimates on human datasets reveal a complex population structure driven by founder effects due to dispersal from Africa and admixture. Notably, our new approach estimates larger FST values of 26% for native worldwide human populations and 23% for admixed Hispanic individuals, whereas the existing approach estimates 9.8% and 2.6%, respectively. While previous work correctly measured FST between subpopulation pairs, our generalized FST measures genetic distances among all individuals and their most recent common ancestor (MRCA) population, revealing that genetic differentiation is greater than previously appreciated. This analysis demonstrates that estimating kinship and FST under more realistic assumptions is important for modern population genetic analysis.

It’s been quite a few years since I thought about the math of the Richard Lewontin controversy over human genetic diversity, so I’d appreciate it if commenters better informed than myself took a crack at interpreting the numbers.

Here’s my understanding of the background. Please let me know if I’m getting this wrong.

Many decades ago, Richard Lewontin pointed out that the majority of the genetic diversity within the human species is not allotted along the lines of race and sub-race.

You can get a hint of this just by comparing fraternal twins to identical twins. Fraternal twins are usually moderately different in looks, personality, athletic abilities, IQ, and so forth, while identical twins are clearly much less genetically diverse.

Even within the most racially homogeneous community, people differ a lot in personality, intelligence, size, etc etc.

Similarly, you can notice two people of different races who are fairly similar in important ways. For instance, Baseball Reference calculates similarities between the career statistics of baseball players and other players. Here are the 10 most similar batters to the Angels’ Mike Trout, a white superstar:

Similar Batters through age 26
  1. Frank Robinson (955.4) *
  2. Ken Griffey Jr. (940.0) *
  3. Mickey Mantle (935.5) *
  4. Hank Aaron (909.7) *
  5. Miguel Cabrera (896.6)
  6. Orlando Cepeda (877.8) *
  7. Mel Ott (877.3) *
  8. Eddie Mathews (866.2) *
  9. Andruw Jones (863.5)
  10. Albert Pujols (858.8)

Trout is most often compared to Mickey Mantle, another white superstar, but this calculation sees him as even more similar in performance to two black greats, the late Frank Robinson and Ken Griffey Jr.

Obviously, performance is not exactly equal to genotype, but it probably has some relationship.

So, a lot of the genetic diversity among humans is fairly random, rather than reflecting racial differences.

If I recall correctly, Lewontin estimated based on the early technology of his day that only 15% of genetic diversity in the human species was allocated according to racial and subracial ancestry.

This estimate is often used by the more sophisticated proponents of the Race Does Not Genetically Exist dogma. What they are really saying is, well, okay, race does exist genetically, but only to minor extent: the racial/subracial genetic diversity glass is 85% empty and only 15% full.

The late genetic anthropologist Henry Harpending told me about 20 years ago that he thought Lewontin was being overly-generous and the real number should be even lower: 12.5%

Frank Salter and Harpending then pointed out in the early 2000s, however, that that 12.5% diversity figure is actually quite important: 12.5%, using Sewall Wright’s way of calculating relatedness, is the average figure for, say, an uncle and his nephew. So, say, Scotsmen, relative to Vietnamese, are about as similar to other Scotsmen as a Scot uncle is to his Scot nephew relative to other Scots.

That’s not huge but it’s not insignificant either.

But now these new researchers, with access to giant amounts of data, are suggesting that genetic differences among racial populations have been underestimated. I haven’t read the paper yet, and I’m out of practice at interpreting these numbers, so let me leave it at that for now.

 
Hide 43 CommentsLeave a Comment
Commenters to Ignore...to FollowEndorsed Only
Trim Comments?
  1. You’re not going to like it when you figure it out. It proves “race realists” wrong.

    • Replies: @SFG
    Not really. I don't know about some of the more mystical claims made by some people on here, but if you can show that a characteristic correlates with a behavior of interest, than it makes sense to act on that characteristic. It doesn't even have to be genetic!
    , @MarkU

    You’re not going to like it when you figure it out. It proves “race realists” wrong.
     
    I am becoming reluctant to take part in this or similar arguments because so often I get flamed by both sides. However I am curious about the way you worded your comment, as if you were a person who had already gotten there because you have processed the data. Perhaps you would care to enlighten me?
    , @bomag
    My tendency is to dismiss you, but like MarkU, I'd be interested in further explanation from you.

    I don't think you understand the issues in question. Any difference between races is going to be important; possibly a game changer if the difference is in a critical area.
  2. At least when it comes to black/white differences, one way in which we chronically underestimate the degree of difference is that we tend to compare whites not to blacks, but to mulattos and quadroons — that is to say to individuals who are partly or even mostly white. Our ‘black’ President was only half-black, and the numerous ‘blacks’ that he brought into his cabinet tended to be visibly even more white than he. In fact, I can only recall one individual in that cabinet who could reasonably be called ‘black’ rather than mulatto.

    It’s the same picture in the larger world. Outside of athletes, a disproportionate share of the ‘blacks’ whites actually have contact with — and therefore naturally draw on to form their notion of who blacks are — are mulattos. This leads to our underestimating the difference between blacks and whites. We think, for example, that it would be reasonably likely that a black could be intelligent enough to be, say, a junior college mathematics instructor. Actually, while it’s not impossible, it’s extremely improbable; most ‘blacks’ capable of filling such positions will be at least half-white. Actual, 100% black Blacks are a lot more different from us than we tend to assume.

    • Replies: @Steve Sailer
    My guess is that an increasing percentage of black junior college math instructors and CPAs and the like are 100% black because they are legal immigrants from Africa who got here through the postgrad system.
    , @Anonymous
    A disproportionate number of African-American athletes that people have heard of may be mulatto as well--there are many in the NBA who are the result of black NBA father/mulatto or white mother pairings. I imagine the same goes for the handful of blacks famous for playing white sports or positions, like Russell Wilson or, of course, Tiger Woods. Even in lower-class sports such as boxing and MMA, you will find plenty of mulattos like Muhammad Ali.
  3. @obwandiyag
    You're not going to like it when you figure it out. It proves "race realists" wrong.

    Not really. I don’t know about some of the more mystical claims made by some people on here, but if you can show that a characteristic correlates with a behavior of interest, than it makes sense to act on that characteristic. It doesn’t even have to be genetic!

  4. @Colin Wright
    At least when it comes to black/white differences, one way in which we chronically underestimate the degree of difference is that we tend to compare whites not to blacks, but to mulattos and quadroons -- that is to say to individuals who are partly or even mostly white. Our 'black' President was only half-black, and the numerous 'blacks' that he brought into his cabinet tended to be visibly even more white than he. In fact, I can only recall one individual in that cabinet who could reasonably be called 'black' rather than mulatto.

    It's the same picture in the larger world. Outside of athletes, a disproportionate share of the 'blacks' whites actually have contact with -- and therefore naturally draw on to form their notion of who blacks are -- are mulattos. This leads to our underestimating the difference between blacks and whites. We think, for example, that it would be reasonably likely that a black could be intelligent enough to be, say, a junior college mathematics instructor. Actually, while it's not impossible, it's extremely improbable; most 'blacks' capable of filling such positions will be at least half-white. Actual, 100% black Blacks are a lot more different from us than we tend to assume.

    My guess is that an increasing percentage of black junior college math instructors and CPAs and the like are 100% black because they are legal immigrants from Africa who got here through the postgrad system.

    • Replies: @Chrisnonymous
    Yes, exactly. In my workplace, the mulatto Americans are quite obviously less intelligent than the (very) black African, who came as a foreign student.

    However, in general, my experience is that Colin is correct. In my personal life, the ratio of mullatos to actually black Africans among the blacks I've met is maybe 5:1 and, in media, public figures are largely mullato, while the mug shots of arrested suspects are at least 50% black, black, blackety-black.
  5. @obwandiyag
    You're not going to like it when you figure it out. It proves "race realists" wrong.

    You’re not going to like it when you figure it out. It proves “race realists” wrong.

    I am becoming reluctant to take part in this or similar arguments because so often I get flamed by both sides. However I am curious about the way you worded your comment, as if you were a person who had already gotten there because you have processed the data. Perhaps you would care to enlighten me?

  6. Similarly, you can notice two people of different races who are fairly similar in important ways.

    So, do we have enough data to tell us that the murder rate for white baseball players is the same as black baseball players?

    Or, do we have enough data to tell us that the bankruptcy rate for white football players is the same as that for black football players?

    As regards to the paper, I think it should be suppressed because some people will use it to justify claims that the ‘other’ is just not human.

  7. My understanding was that Lewontin’s 15% applied to most (but not all!) genes when considered individually, but that the whole argument flew out the window when you looked at many genes at once.

  8. Anonymous[921] • Disclaimer says:

    I’m confused about the 12.5% relatedness figure for uncle/nephew. The uncle would be on average 50% related to his sibling, the nephew’s parent, who would in turn be 50% related to the nephew. So it seems the uncle/nephew relatedness would be (50%)(50%) = 25%, no?

    • Agree: Prodigal son
    • Replies: @notanon
    mother
    , @Anonymous
    You have to go back to the common ancestor of the nephew and uncle, you can't look across siblings.

    A has two children, B1 and B2, each of whom receive half of her genes. B1 has a child, C, who gets half of B1's genes, i.e., 25% of A's genes. So C (nephew) has 25% of A's genes and B2 (uncle) has 50% of her genes. On average, what percent of her genes will they both have? 25% * 50% = 12.5%.
  9. Whatever the numbers are, we already know how they’re going to be interpreted. Whites are evil and privileged. Worst of all, we’re smirkers.

  10. @Steve Sailer
    My guess is that an increasing percentage of black junior college math instructors and CPAs and the like are 100% black because they are legal immigrants from Africa who got here through the postgrad system.

    Yes, exactly. In my workplace, the mulatto Americans are quite obviously less intelligent than the (very) black African, who came as a foreign student.

    However, in general, my experience is that Colin is correct. In my personal life, the ratio of mullatos to actually black Africans among the blacks I’ve met is maybe 5:1 and, in media, public figures are largely mullato, while the mug shots of arrested suspects are at least 50% black, black, blackety-black.

  11. Anonymous[921] • Disclaimer says:
    @Colin Wright
    At least when it comes to black/white differences, one way in which we chronically underestimate the degree of difference is that we tend to compare whites not to blacks, but to mulattos and quadroons -- that is to say to individuals who are partly or even mostly white. Our 'black' President was only half-black, and the numerous 'blacks' that he brought into his cabinet tended to be visibly even more white than he. In fact, I can only recall one individual in that cabinet who could reasonably be called 'black' rather than mulatto.

    It's the same picture in the larger world. Outside of athletes, a disproportionate share of the 'blacks' whites actually have contact with -- and therefore naturally draw on to form their notion of who blacks are -- are mulattos. This leads to our underestimating the difference between blacks and whites. We think, for example, that it would be reasonably likely that a black could be intelligent enough to be, say, a junior college mathematics instructor. Actually, while it's not impossible, it's extremely improbable; most 'blacks' capable of filling such positions will be at least half-white. Actual, 100% black Blacks are a lot more different from us than we tend to assume.

    A disproportionate number of African-American athletes that people have heard of may be mulatto as well–there are many in the NBA who are the result of black NBA father/mulatto or white mother pairings. I imagine the same goes for the handful of blacks famous for playing white sports or positions, like Russell Wilson or, of course, Tiger Woods. Even in lower-class sports such as boxing and MMA, you will find plenty of mulattos like Muhammad Ali.

    • Replies: @Steve Sailer
    The 3 point line selects for blacks who grew up with their own driveway, who tend to be fairer.
  12. @obwandiyag
    You're not going to like it when you figure it out. It proves "race realists" wrong.

    My tendency is to dismiss you, but like MarkU, I’d be interested in further explanation from you.

    I don’t think you understand the issues in question. Any difference between races is going to be important; possibly a game changer if the difference is in a critical area.

  13. @Anonymous
    I'm confused about the 12.5% relatedness figure for uncle/nephew. The uncle would be on average 50% related to his sibling, the nephew's parent, who would in turn be 50% related to the nephew. So it seems the uncle/nephew relatedness would be (50%)(50%) = 25%, no?

    mother

    • Replies: @Anonymous
    Pardon?
  14. Many decades ago, Richard Lewontin pointed out that the majority of the genetic diversity within the human species is not allotted along the lines of race and sub-race.

    no expert but my understanding is this was a political con-trick because it only applied to single genes whereas race is a *pattern* of gene frequencies of many genes and when you compare many genes at once race jumps out very clearly.

    big jump in non-expertness here leading to big guess…

    i think what this paper is saying is that clustering effect (of many genes) is larger than expected and my guess is it will be because there is greater linkage equilibrium than expected and my further guess is that is because previous models looked at ancestry using a standard family tree diagram

    without taking into account your thing about double-duty i.e. “partially inbred extended family.”

    so more double-duty -> more linkage equilibrium -> more relatedness.

    90% guess though.

  15. Andruw Jones

    Why were three unimpressive Arubans, Sidney Ponson, Eugene Kinsdale, and Calvin Maduro knighted, but not Jones? Did Her Majesty Beatrix of the Netherlands have something against Curaçao?

    • Replies: @Steve Sailer
    Andruw Jones was a really good centerfielder for a decade or so, but then he got real swole, hit 51 homers, and two years later had just about the worst season ever for the Dodgers. In 75 games on a $16 million/yr contract, he hit .158 with 3 homers and 14 RBIs:

    https://www.baseball-reference.com/players/j/jonesan01.shtml
  16. Anonymous[921] • Disclaimer says:
    @notanon
    mother

    Pardon?

    • Replies: @bomag

    The uncle would be on average 50% related to his sibling
     
    Yes.

    [the uncle] would in turn be 50% related to the nephew.
     
    No. The presumption is that the uncles marry unrelated women (mother factor), so that cuts things in half.

    "I'd lay down my life for two brothers or eight cousins" (JRS Haldane)
  17. @Reg Cæsar

    Andruw Jones
     
    Why were three unimpressive Arubans, Sidney Ponson, Eugene Kinsdale, and Calvin Maduro knighted, but not Jones? Did Her Majesty Beatrix of the Netherlands have something against Curaçao?

    Andruw Jones was a really good centerfielder for a decade or so, but then he got real swole, hit 51 homers, and two years later had just about the worst season ever for the Dodgers. In 75 games on a $16 million/yr contract, he hit .158 with 3 homers and 14 RBIs:

    https://www.baseball-reference.com/players/j/jonesan01.shtml

  18. Lewontin originally worked with blood types. Was there a alter re-do when genetics tech advanced a little more?

  19. Top 5 Spelling Bee contestants 2019 San Francisco Bay Area (~7M pop)

    • Replies: @AnotherDad
    Brahmins have had a literate tradition--to recite the Vedas and such--for a few thousand years. Memorizing stuff is right up their alley.

    But it's also sort of become their thing.

    Indians:Spelling Bees :: Blacks:Basketball
    , @Wilkey
    I could easily believe that certain Indian castes have been strongly selected for better memory.

    I could just as easily believe that spelling bees have become a thing. Even as a kid, in my very non-diverse school - maybe two Asians in my entire class in elementary school - it was the lone Indian dude who won every single spelling bee.

    My mom was into making me practice for the spelling bee. She was *really* into making me practice for the spelling bee, and I was peeved by the fact (even though I kinda wanted to win, too). But my mom could've drilled me on the words five times more than she did. So I can easily imagine a hardcore Indian tiger mom making little Sanjiv practice even more, which was probably a lot more than any other mom. And unlike most of the rest of our mom's, Sanjiv's mom didn't have a day job. So she got bored during the day and took it out on Sanjiv and his sister.

    Spelling bees just aren't a big thing for most white people. They're a hobby, not a career.
    , @FPD72
    They no longer keep going in spelling bees until the last kid standing. They now go twenty rounds, so if a kid spells twenty words correctly, he is a co-champion. The just concluded contest this year had eight winners, seven of whom were Indian. The lone exception was a white girl from Alabama.
    , @Expletive Deleted
    An odd pastime. Being muffin-tier English, I never spell out loud except when wrangling with overseas call-centre staff, who are absolutely certain that I am incompetent in my native language and in dire need of correction.

    https://www.youtube.com/watch?v=H-oH-TELcLE

    I either write the word absolutely accurately without even thinking about it (in pencil or fountain-pen; keyboards are another matter entirely, and primarily the domain of single ladies when I was young), or say it without having to visualise it.
    It's a frightful effort when I have to spell the things out. There's no connection between the two processes. Rather like the Chinese, I suppose. Hence the bizarre divergence of English (i.e. "British English") spelling and speech over the centuries.

    I note with interest the strange similarity of those unfortunate children in the picture with the incomprehensible call-centre demons I encounter. Perhaps they have to spell the words out in their heads before they can say them?
    , @Oleaginous Outrager
    Spelling bees? Tlk abt a dieing spclty in a txt age.
  20. @danand
    Top 5 Spelling Bee contestants 2019 San Francisco Bay Area (~7M pop)

    https://flic.kr/p/2g5UDin

    Brahmins have had a literate tradition–to recite the Vedas and such–for a few thousand years. Memorizing stuff is right up their alley.

    But it’s also sort of become their thing.

    Indians:Spelling Bees :: Blacks:Basketball

  21. Hypothesis:

    There are genetic differences that are selected for more or less evenly across every environment.

    Then there are genetic differences that exhibit strong selection in particular environments.

    Race and ethnicity are defined by the latter.

  22. @danand
    Top 5 Spelling Bee contestants 2019 San Francisco Bay Area (~7M pop)

    https://flic.kr/p/2g5UDin

    I could easily believe that certain Indian castes have been strongly selected for better memory.

    I could just as easily believe that spelling bees have become a thing. Even as a kid, in my very non-diverse school – maybe two Asians in my entire class in elementary school – it was the lone Indian dude who won every single spelling bee.

    My mom was into making me practice for the spelling bee. She was *really* into making me practice for the spelling bee, and I was peeved by the fact (even though I kinda wanted to win, too). But my mom could’ve drilled me on the words five times more than she did. So I can easily imagine a hardcore Indian tiger mom making little Sanjiv practice even more, which was probably a lot more than any other mom. And unlike most of the rest of our mom’s, Sanjiv’s mom didn’t have a day job. So she got bored during the day and took it out on Sanjiv and his sister.

    Spelling bees just aren’t a big thing for most white people. They’re a hobby, not a career.

  23. In 75 games on a $16 million/yr contract, he hit .158 with 3 homers and 14 RBIs:

    That alone would make Andruw the Ruth, Cobb, DiMaggio, Williams, and Ott of the Netherlands Antilles. I thought the Dutch went easy on drugs, but that apparently doesn’t apply to steroids.

    Some honor well below knighthood got Ferguson Jenkins out of doing time for smuggling dope home to Canada.

    Bert Blyleven appears to be the top Dutch-born baseball player, but I may be missing somebody. Kenley Jansen and Andrelton Simmons, both from Curaçao, play for the Dodgers and Angels, respectively.

    They don’t seem to be knights, either. Anticuraçaoseisme!

  24. @danand
    Top 5 Spelling Bee contestants 2019 San Francisco Bay Area (~7M pop)

    https://flic.kr/p/2g5UDin

    They no longer keep going in spelling bees until the last kid standing. They now go twenty rounds, so if a kid spells twenty words correctly, he is a co-champion. The just concluded contest this year had eight winners, seven of whom were Indian. The lone exception was a white girl from Alabama.

  25. anon[247] • Disclaimer says:

    One thing to remember about Lewontin’s estimate is that it’s essentially a measure of neutral genetic variation, i.e. differences between populations that have arisen due to chance over many generations, rather than differences due to natural selection. The vast majority of genetic differences between populations are neutral rather than due to selection. Some are, of course, due to selection but unless they’re really obvious like skin color, it’s not something that’s easy to establish. So, because neutral differences accumulate over generations, FST is essentially a measure of how long populations have been separated from each other, rather than how dissimilar they are in terms of genetic differences that we would consider consequential.

    What this new paper argues is that traditional FST estimates are based on assumptions that are in practice violated in human populations. Once you correct for this, a much larger share of (neutral) genetic variation is between-populations than has been appreciated previously.

    Another problem with Lewontin’s approach is that it is based on averaged locus-by-locus analysis of variation that ignores the correlation structure of genetic differences. Once you realize that genes are not isolated entities but that the presence of a given allele predicts the presence of other alleles in other loci, you will understand that to assess the overall similarity of individuals, you have to simultaneously consider many genetic variants. If you use a multivariate method like PCA, you will discover the common-sensical fact that populations are not differentiated by the presence or absence of this or that single allele, but rather the differentiation arises from the fact that populations are characterized by genome-wide patterns of genetic differentiation resulting from founder effects, inbreeding, drift and selection.

    • Replies: @Steve Sailer
    Thanks.
  26. @anon
    One thing to remember about Lewontin's estimate is that it's essentially a measure of neutral genetic variation, i.e. differences between populations that have arisen due to chance over many generations, rather than differences due to natural selection. The vast majority of genetic differences between populations are neutral rather than due to selection. Some are, of course, due to selection but unless they're really obvious like skin color, it's not something that's easy to establish. So, because neutral differences accumulate over generations, FST is essentially a measure of how long populations have been separated from each other, rather than how dissimilar they are in terms of genetic differences that we would consider consequential.

    What this new paper argues is that traditional FST estimates are based on assumptions that are in practice violated in human populations. Once you correct for this, a much larger share of (neutral) genetic variation is between-populations than has been appreciated previously.

    Another problem with Lewontin's approach is that it is based on averaged locus-by-locus analysis of variation that ignores the correlation structure of genetic differences. Once you realize that genes are not isolated entities but that the presence of a given allele predicts the presence of other alleles in other loci, you will understand that to assess the overall similarity of individuals, you have to simultaneously consider many genetic variants. If you use a multivariate method like PCA, you will discover the common-sensical fact that populations are not differentiated by the presence or absence of this or that single allele, but rather the differentiation arises from the fact that populations are characterized by genome-wide patterns of genetic differentiation resulting from founder effects, inbreeding, drift and selection.

    Thanks.

  27. @Anonymous
    Pardon?

    The uncle would be on average 50% related to his sibling

    Yes.

    [the uncle] would in turn be 50% related to the nephew.

    No. The presumption is that the uncles marry unrelated women (mother factor), so that cuts things in half.

    “I’d lay down my life for two brothers or eight cousins” (JRS Haldane)

    • Replies: @TBA
    Are you sure? I get these paths between uncle and nephew.

    Via grandfather: (1/2) * (1/2) * (1/2) = 1/8
    Via grandmother: (1/2) * (1/2) * (1/2) = 1/8

    In total 2/8 = 1/4.
    , @Anonymous
    You are saying that cousins are 12.5% related which I do not debate! But then shouldn't the relatedness of uncle to nephew be twice as high, or 25%? This halving mother factor only comes into play when considering the uncle's children (first cousins), not the uncle himself. Whoever I marry and have kids with doesn't affect how related I am to my nephews.
    , @ben tillman
    Or four uncles.
  28. @danand
    Top 5 Spelling Bee contestants 2019 San Francisco Bay Area (~7M pop)

    https://flic.kr/p/2g5UDin

    An odd pastime. Being muffin-tier English, I never spell out loud except when wrangling with overseas call-centre staff, who are absolutely certain that I am incompetent in my native language and in dire need of correction.

    I either write the word absolutely accurately without even thinking about it (in pencil or fountain-pen; keyboards are another matter entirely, and primarily the domain of single ladies when I was young), or say it without having to visualise it.
    It’s a frightful effort when I have to spell the things out. There’s no connection between the two processes. Rather like the Chinese, I suppose. Hence the bizarre divergence of English (i.e. “British English”) spelling and speech over the centuries.

    I note with interest the strange similarity of those unfortunate children in the picture with the incomprehensible call-centre demons I encounter. Perhaps they have to spell the words out in their heads before they can say them?

  29. @bomag

    The uncle would be on average 50% related to his sibling
     
    Yes.

    [the uncle] would in turn be 50% related to the nephew.
     
    No. The presumption is that the uncles marry unrelated women (mother factor), so that cuts things in half.

    "I'd lay down my life for two brothers or eight cousins" (JRS Haldane)

    Are you sure? I get these paths between uncle and nephew.

    Via grandfather: (1/2) * (1/2) * (1/2) = 1/8
    Via grandmother: (1/2) * (1/2) * (1/2) = 1/8

    In total 2/8 = 1/4.

  30. Anonymous[921] • Disclaimer says:
    @bomag

    The uncle would be on average 50% related to his sibling
     
    Yes.

    [the uncle] would in turn be 50% related to the nephew.
     
    No. The presumption is that the uncles marry unrelated women (mother factor), so that cuts things in half.

    "I'd lay down my life for two brothers or eight cousins" (JRS Haldane)

    You are saying that cousins are 12.5% related which I do not debate! But then shouldn’t the relatedness of uncle to nephew be twice as high, or 25%? This halving mother factor only comes into play when considering the uncle’s children (first cousins), not the uncle himself. Whoever I marry and have kids with doesn’t affect how related I am to my nephews.

  31. @bomag

    The uncle would be on average 50% related to his sibling
     
    Yes.

    [the uncle] would in turn be 50% related to the nephew.
     
    No. The presumption is that the uncles marry unrelated women (mother factor), so that cuts things in half.

    "I'd lay down my life for two brothers or eight cousins" (JRS Haldane)

    Or four uncles.

  32. Ultimate phenotypic heterogeneity may be greater than estimates of genetic heterogeneity due to the fact that so-called “loss of (protein) function” in an allele (recessiveness) can ultimately phenotypically “function” in the homozygous state as well as in the heterozygous state. With as little as one dominant allele in the same population, 2 taken 2 at a time with replacement and you get 3 phenotypes. Compare this to the case of a population with N “functional” (protein producing) alleles (dominant) providing the same ultimate phenotypic function — N taken 2 at a time, will produce no ultimate phenotypic variation.

    I don’t know what the quantitative reality is but it does seem to be something worth taking into account in phenotypic vs genotypic heterogeneity estimates.

  33. Reading the first couple pages and the conclusion, it looks like the whole paper is a new way to calculate Fst without the problems that previous Fst measures had.

    If you look at the middle of this page:
    https://en.wikipedia.org/wiki/Fixation_index

    It gives a pretty nice definition of Fst where you count the number of pairwise differences in two individuals in either two different populations or from within the same population. However, this has problems when you have small populations and populations of different size (when sampling from two pops). The paper looks like it has a fix for this issue because it looks like it does not need to assume any structure.

    As a computer programmer I absolutely despise the way math people write computer code using math symbols. They leave out explanations for 10000 different things that you must get correct for you to implement their “algorithm” the same way. For example, X transposed is meaningless unless you define the orientation of X to begin with, which they never do. There is just conventions that you just have to know.

  34. @Anonymous
    A disproportionate number of African-American athletes that people have heard of may be mulatto as well--there are many in the NBA who are the result of black NBA father/mulatto or white mother pairings. I imagine the same goes for the handful of blacks famous for playing white sports or positions, like Russell Wilson or, of course, Tiger Woods. Even in lower-class sports such as boxing and MMA, you will find plenty of mulattos like Muhammad Ali.

    The 3 point line selects for blacks who grew up with their own driveway, who tend to be fairer.

  35. Anonymous[241] • Disclaimer says:
    @Anonymous
    I'm confused about the 12.5% relatedness figure for uncle/nephew. The uncle would be on average 50% related to his sibling, the nephew's parent, who would in turn be 50% related to the nephew. So it seems the uncle/nephew relatedness would be (50%)(50%) = 25%, no?

    You have to go back to the common ancestor of the nephew and uncle, you can’t look across siblings.

    A has two children, B1 and B2, each of whom receive half of her genes. B1 has a child, C, who gets half of B1’s genes, i.e., 25% of A’s genes. So C (nephew) has 25% of A’s genes and B2 (uncle) has 50% of her genes. On average, what percent of her genes will they both have? 25% * 50% = 12.5%.

    • Replies: @TBA
    Yes, but uncle and nephew are related via two common ancestors. Shouldn't they have 12.5% in common from each?
  36. There is more genetic diversity within a given race than there is between races.

    Any fool knows this. Except your type of fool.

    • Replies: @anon
    At times I wonder if you're actually a LARPer.
  37. What makes blacks inferior is the extent of their inbreeding with the Irish. The more the stupider.

  38. @obwandiyag
    There is more genetic diversity within a given race than there is between races.

    Any fool knows this. Except your type of fool.

    At times I wonder if you’re actually a LARPer.

  39. So, a lot of the genetic diversity among humans is fairly random, rather than reflecting racial differences.

    Damn, Steve, ain’t you learned nothin’ yet? All Diversity Is Not Created Equal!

  40. @danand
    Top 5 Spelling Bee contestants 2019 San Francisco Bay Area (~7M pop)

    https://flic.kr/p/2g5UDin

    Spelling bees? Tlk abt a dieing spclty in a txt age.

  41. The paper shows there are at least 11 Races in all:
    >Euro
    >Caucasus
    >Amerindian
    >East Asian
    >Aussie Abo
    >Pajeet
    >West Africa
    >North Africa
    >East Africa
    >Islander
    >Islander

  42. @Anonymous
    You have to go back to the common ancestor of the nephew and uncle, you can't look across siblings.

    A has two children, B1 and B2, each of whom receive half of her genes. B1 has a child, C, who gets half of B1's genes, i.e., 25% of A's genes. So C (nephew) has 25% of A's genes and B2 (uncle) has 50% of her genes. On average, what percent of her genes will they both have? 25% * 50% = 12.5%.

    Yes, but uncle and nephew are related via two common ancestors. Shouldn’t they have 12.5% in common from each?

Comments are closed.

Subscribe to All Steve Sailer Comments via RSS
PastClassics
Which superpower is more threatened by its “extractive elites”?
The “war hero” candidate buried information about POWs left behind in Vietnam.
What Was John McCain's True Wartime Record in Vietnam?
Are elite university admissions based on meritocracy and diversity as claimed?