British class differences persisting down centuries?

By Razib Khan | July 22, 2012 11:01 am

People with Norman names wealthier than other Britons:

Research shows that the descendants of people who in 1858 had “rich” surnames such as Percy and Glanville, indicating they were descended from the French nobility, are still substantially wealthier in 2011 than those with traditionally “poor” or artisanal surnames. Artisans are defined as skilled manual workers.

As Steve Sailer observes strict adherence to surnames on a mass scale post-dates the Norman invasion by centuries. So the headline is pretty sensational. But I went and read the original working paper, and there is no mention of Norman or French names! The author of the piece in The Telegraph is probably right (i.e., a casual reading of history will show that Norman names are enriched in the English elite), but this is clearly another case of one having to be careful of the details when it comes to British media.

But the results in the paper are interesting enough. The biggest finding is that regression toward the mean is far less using this 200 year data set than might be extrapolated from modern 2 generation data sets. Another reason to be skeptical of economists waxing grandly on all they know from the little they know. Nevertheless, I do think that these sorts of studies are going to be interesting when they are synthesized with genetics. With very large data sets (i.e., thousands of whole genomes) and more powerful methods one might be able to see if the modern British elite (at least the portion with deep roots in the nation) does exhibit more similarity to the people across the English Channel than one might expect.

This might sound crazy, but the India caste system seems to b stratified in terms of ancestry. The Genographic Project even asserts that caste predates the Indo-Aryans, which I suspect is true. HAP has illustrated quite clearly that there are differences between South Indian non-Brahmin high/middle castes and low castes.

CATEGORIZED UNDER: Anthroplogy, History
MORE ABOUT: Genetics, History

Comments (12)

  1. Markk

    This is a real question by an ignorant American. Isn’t the Indian Caste System -defined- by ancestry? Wouldn’t it be rather surprising if there weren’t genetic stratifications? Maybe I just don’t understand but the last paragraph seems dog bite man, or some kind of irony I don’t get.

  2. Sandgroper

    I’m mildly amused – my father’s Norman-surnamed paternal grandfather was a brick maker and his Norman-surnamed maternal grandfather was a plumber. My grandfather became more upwardly mobile again. Yes, I realise the thrust is the other way around, and that Greg Clark showed that you get this downward mobility.

    The conventional wisdom I had gleaned from reading is that the Normans interbred so readily with the populations they invaded that ‘within 300 years they had disappeared as a distinguishable ethnic group.’ It would be kind of interesting to find out if that is actually true genetically. From historical sources, they do seem to have been enthusiastically exogamous, though – as I have been myself.

    I did one of those world name concentration searches and discovered a high population of people in India with my surname – a relic of British colonial occupation and mingling with the natives, I presume.

  3. wijjy

    Reminds me of the heiress problem in The genetical theory of natural selection by RA Fisher.

    His point is that inherited wealth is inversely correlated with fertility (the fewer childern that you have, the more each can inherit), so that the very richest would already be pre-selected for low fertility. I cannot find anything in the preprint about the relative numbers of descendents for the rich vs poor surnames, only for numbers of deaths.

    Trivia much beloved by UK satirists. The Bazalgette surname in the paper is the family of Joseph Bazalgette who created the London sewerage system, and, in the 1990s his great-great-grandson popularised the Big Brother TV programme worldwide.

    And some observations:

    1) The preprint bears no resemblance to the Telegraph article. The only surname mentioned is a Hugenot one – so an immigrant, and certainly not a Norman.
    2) Wealth can be directly inherited (so you can be counting the same money), so is less good a measure of class than the education statistics.

  4. Isn’t the Indian Caste System -defined- by ancestry? Wouldn’t it be rather surprising if there weren’t genetic stratifications?

    re: caste system. the fully elaborated caste system as we have it is the function of the last few hundred years, and was systematized and formalized under the british. additionally, the racial-ethnic origins have to some extent been grafted on recently. but, elements of this did exist in large measure before. the primary issue is that the for most indians jati, which is lower down as organizational structure, matters more than caste, and always has. only a few castes, brahmins, and some mercantile castes, had india-wide systems of self-conscious affiliation before the past few centuries. a brahmin was a brahmin everywhere. the other castes had some inchoate existence, but really crystallized within the last few centuries as the four-varna system became popularized. a ‘sudra’ in tamil nadu really has nothing to do with a ‘sudra’ in uttar pradesh. in contrast, south indian tamil brahmins do have some relationship, if distant, to north indian UP brahmins. so, the fact that brahmins are different in tamil nadu from non-brahmins is not surprising. but, the fact that nadars, a mercantile sudra caste, are different from dalits, who are lower on the caste totem pole, is somewhat surprising, because nadars were a localized community whose self-consciousness really cohered recently.

    second, basic genetics implies that over time gene flow will equilibrate over thousands of years. even if you had 1% intermarriage, differences will disappear. the classic case are the hui in china, who are now genetically 90% han, despite low endogamy. but it makes totally sense if you assume something like 5% per generation for 50 generations.

  5. Jaldhar


    I think even for Brahmanas, jati is more salient than varna. For instance the Pancha Dravida (the five major groups south of the Vindhyas) do not dine or intermarry with the Pancha Gauda (the five north of the Vindhyas.) South Indians also do not traditionally marry Brahmanas belonging to other Vedantic sects making these ideological units into endogamous jatis. Each considers the other as inferior and morally suspect. (Northerners eat meat. Southerners marry their nieces — both mentioned by Kumarila Bhatta c. 6th century AD.) These divisions have been there for a long time. I would go as far as to say that Varna is a totally theoretical concept to all but the revivalists and jati is the only socially valid definition of “caste.”

  6. #5, agree with a lot of what you say. but what i’ve read indicates that even before the british the constructs of brahmin and outcaste were found all across the subcontinent in a coherent sense. and surprising to a great extent this is validated genetically. in contrast, sudras in tamil nadu are very different from sudras in punjab. the distance between a tam brahm and a punjabi brahmin is not so great (though it exists).

  7. Karl Zimmerman

    3 –

    I remember reading a study in graduate school which attempted to determine the scope of inherited wealth in the U.S. IIRC, they looked at families in The Social Register (which has been in active publication since the 1880s), and found that on average, substantial inherited wealth in the U.S. lasts only five generations, mainly because inheritance subdivides fortunes more rapidly than the average scion can build them. I’ll try and see if I can find a citation somewhere.

  8. Jason Malloy

    But I went and read the original working paper, and there is no mention of Norman or French names!

    The data they are discussing is from Gregory Clark’s next book (PDF) which covers 1000 years of class mobility in England, and which has been online, in evolving form, for a few years.

  9. Leor

    Heisman’s Suicide Note contains a vision of a genetic history of England. According to him, the civil wars and the American civil war can be understood as conflict between the Normans and Anglo-Saxons.

  10. Miley Cyrax

    “This might sound crazy, but the India caste system seems to b stratified in terms of ancestry. ”

    So call me, maybe.

  11. AG

    “even if you had 1% intermarriage, differences will disappear. the classic case are the hui in china, who are now genetically 90% han, despite low endogamy”

    For hui people, many believe they are sinicized uyghur people since words `hui’ and `uy’ sound quite alike.

    China is a true melting pot in which many ancient ethnic groups melted away like xianbei, xiongnu, ect. Manchurian is on its way out as ethnic group. Ehnic young koreans are not interested speaking korean any more. Han culture seems never having issue with intermarriage. It does not matter which ethnic groups were ruling. The result seems always same.
    Ming dynasty (which kicked mongol out) actually made law to forbbiden marriage within minority ethnic groups. i,e. It was illegal to marry your own kind unless you were ethnic han. For racial minority, only intermarriage was legal. You can call this a gentle way of ethnic group eradication.

  12. Sandgroper

    ‘gentle way of ethnic group eradication’

    Yes, they tried that in Australia. It was called ‘breeding out the Abo.’ What we might think of as ‘genetic swamping’.

    I gave a hunch this helps to explain the relatively relaxed Han attitude to exogamy – confidence that they can absorb other ethnic groups and cultures. History pretty much shows the confidence is justified.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Gene Expression

This blog is about evolution, genetics, genomics and their interstices. Please beware that comments are aggressively moderated. Uncivil or churlish comments will likely get you banned immediately, so make any contribution count!

About Razib Khan

I have degrees in biology and biochemistry, a passion for genetics, history, and philosophy, and shrimp is my favorite food. In relation to nationality I'm a American Northwesterner, in politics I'm a reactionary, and as for religion I have none (I'm an atheist). If you want to know more, see the links at


See More


RSS Razib’s Pinboard

Edifying books

Collapse bottom bar