Guess which surnames died out in pre-industrial England?

Share on FacebookShare on Google+Email this to someoneTweet about this on Twitter

The surnames of the criminal and the poor, of course. Greg Clark provides new evidence for the “survival of the richest” here (and he thanks Nick Wade for the idea). From the abstract:

[E]vidence from…surnames…again shows the takeover of English society by the economically successful between 1600 and 1851, and the disappearance of the criminal and the poor. A man’s economic success in pre-industrial England predicted a permanent increase of his surname frequency, and hence his gene frequency, by 1851.

Confession: I, for one, had no idea that Elvis was a surname.

Clark’s papers have familiarized economists with the basics of genetics. It seems to be paying off: At the American Economic Association meetings this year, there was a session on brain evolution in the very long run, another on genetics and microeconomic behavior, and a third GNXP-friendly session where Clark presented the above-quoted paper.


  1. My first reaction before reading the paper is…. 
    Does the disappearance of the surnames of the poor and the criminal really mean the disappearance of their blood lines? 
    Specifically, were these people known for getting married, having children, and having those children bear their last names? 
    Could there not have been any bastards?

  2. If that work stands up to scrutiny, it’s a crackerjack.

  3. While the thesis is reasonable, that succesful working people leaves more living children than the scum of the society, the methodology seems to me weak. My opinion is that of a man of the street, it may have no academic value.  
    1. The selective value of social position seems to last for one, or maximum two generations. In our days no fortune survives more than two generations, except the highly endogamic Rothschilds. If so, one two generations of ¨bad luck genes¨ should not be perceivable after 7 generations.  
    2. The lumpen proletariat is not known to have less children than rich people, on the contrary. Malthus and Galton all talked of a dysgenic effect, which worries people today more than ever. People who spends most of their life in prison seem to leave the same number of children, or more, than rich people. Could it be that everybody has the wrong impression? 
    3. English names are very stable and sticking, but not immutable. People is known to have changed names voluntarily to avoid identification with known criminals, and/or to fake identification with famous people somehow related. When wills are in play, people takes the names of the rich relative. Sometimes the childless rich uncle demands change of name. I cannot quantify the effect of this thing but we are talking of small differences where everything may matter.

  4. Recent evidence from Spain suggests that social status is related to rareness of surname. I wonder if this is indicative of a dysgenic trend.

  5. The one thing that bothered me is that the study follows only the male line. Perhaps poor people have more daughters? Perhaps having more daughters leads to poverty? At any rate, a high female birth rate in a family would seem to pare the number of offspring carrying the family name.

  6. in the medieval period poor people did have proportionality more daughters. trivers-willard effect.

  7. How many of these indicted men were hung? 
    How many served long jail terms? 
    That would put a dent in your reproductive fitness right there, wouldn’t it? 
    I didn’t finish the paper, but assume Clark factored all that in.

  8. Sir Allen Stanford, recently charged in an $8 billion dollar Ponzi scheme, was born with the last name of “Tuck,” which sounds rather low-rent (Robin Hood’s Friar Tuck and anti-Nixon dirty trickster Dick Tuck are the only Tucks I can recall). Somewhere along the way he stopped using “Tuck” acquired the surname of a 19th Century railroad robber baron and university-founder. 
    So, the Tuck-Stanford Effect may influence the results.

  9. Checking in the GSS to see if a similar thing might be happening today, those who have been arrested or charged with crime have on average slightly fewer children (by 0.28) than those never arrested or charged. I guess crime doesn’t pay, at least not reproductively speaking.

  10. Going back quite a ways there was the banking tycoon Edward Tuck. Dartmouth University’s business school is named after him.

  11. @ Steve Sailer: 
    Clark checked for name change effects in two ways: aliases listed in wills and indictments (2% each), and minor spelling changes in name spelling over time.  
    No evidence that this was a driving force. Neither are perfect tests, but we don’t expect perfection out of science: We just expect future researchers to run more tests.  
    Especially when you look at the magnitudes involved (big–see Table 1 of the text), and when it already matches up with Clark’s direct data on the fertility of the rich (the rich had twice as many surviving children as the poor), this looks like big, big selection coefficients at work.  
    s=1.0 would be a literal reading, though of course Clark only has data on phenotypes not genotypes. Here’s hoping some enterprising grad student starts doing archaeogenetics with samples from Renaissance graves.

  12. Could this be the reason that China has so few family names? Could the ancestors of the Chans, Changs, and Chongs have been wealthier than other fogotten families?

  13. Is this so stastically insignificant?