The words of the father

By Razib Khan | September 18, 2011 10:01 pm

Over at A Replicated Typo they are talking about a short paper in Science, Mother Tongue and Y Chromosomes. In it Peter Forster and Colin Renfrew observe that “A correlation is emerging that suggests language change in an already-populated region may require a minimum proportion of immigrant males, as reflected in Y-chromosome DNA types.” But there’s a catch: they don’t calculate a correlation in the paper. Rather, they’re making a descriptive verbal observation. This observation seems plausible on the face of it. In addition to the examples offered, one can add the Latin American case, where mestizo populations tend to have European Y chromosomal profiles and indigenous mtDNA.

In one of the more nuanced instances cited by Renfrew and Forster they note that though there is a fair penetration of both Austronesian mtDNA and Y chromosomes amongst Austronesian speaking inhabitants of New Guinea, Austronesian Y chromosomes are nearly absent from those populations which speak Papuan. In other words, the inference is that the native indigenous male elite perpetuated the Papuan language! I think the key issue here is social dominance, and the role of male cultural units. This is clear when you have a group like African Americans. Though mostly West African in ancestry this ethnic group clearly has ~20% European ancestry, mostly mediated through European males. This is simply a function of the social dominance of European males in relation to African males. And, it may be telling that African Americans are an English speaking Christian population, albeit with their own distinctive accent.

In the paper the authors note:

…It may be that during colonization episodes by emigrating agriculturalists, men generally outnumbered women in the pioneer colonizing groups and took wives from the local community. When the parents have different linguistic backgrounds, it may often be the language of the father that is dominant within the family group….

There is certainly some truth to this. But what I think that this narrative misses are the coarser and larger scale social units which expansive male cultural communities operate through. The spread of European males in Latin America was not uncoordinated. Often small groups of males decapitated indigenous male power elites, sometimes literally! Long distance travel, such as what the Austronesians engaged in, almost certainly must have entailed a minimal level of political and ideological commitment. They cite the example of the Genghis Khan haplotype, but that wasn’t perpetuated through a process of gradual demic expansion.

CATEGORIZED UNDER: Anthroplogy, Genetics, Human Genetics

Comments (13)

  1. Clark

    When the parents have different linguistic backgrounds, it may often be the language of the father that is dominant within the family group….

    Is there a formal study on this? I’d actually expect the opposite since the kids are around the mother far more than the father. I think it’d be more related to the local population the children are raised within though. Maybe the father largely controls that (they bring wives to their own community – which is what that quote might be hinting at).

    I’d actually expect the major issue is trade. As invasive languages become trade languages they would have more of an effect. That might be what underlies your point in your final paragraph. If a language is necessary for trade then there is an incentive for at least some individuals of a local community to learn it.

  2. There are some notable examples of conquering elites (probably male dominated) losing their language; the classical Romans in the Greek area and, as suggested recently at this blog, the Tutsis. The failure of Islamic invaders to produce language shift in South Asia despite producing religion shift, is also notable.

    Another interesting motiff along these lines is the Mongol imposition of a female dominate elite of Mongol princesses, with extraordinary power due to their connections in the larger empire, in Korea. This too did not produce languge shift and ultimate produced a successful counterrevolution.

  3. Subhan


    “The failure of Islamic invaders to produce language shift in South Asia despite producing religion shift, is also notable”

    The language shift did happen in south asia. Moguls used persian at official level. Then Arabic-Persian-Sanskirit mix Urdu/Hindi language spread.

    In 19th century, Urdu was chosen by British for schools in Panjab, even though it was not their native language

  4. Grey

    I think the dominant language outside the home would be the critical factor.

    If the invasive males are only in enough numbers to replace the native elite then the children would be bilingual but most of their conversations would be in the mother tongue and the invader’s language might peter out over a few generations. If the invaders replaced say 60%+ of the native males then again the children would be bilingual but the bulk of the conversations would be in the invader’s tongue.

    Put another way, if you divide the population into
    Sergeants, Merchants and Stewards
    then if the invaders just replace the aristocracy they have to learn the native language. If they replace both the aristocracy and the sergeants and stewards then they don’t.

    I think the Norman conquest of England is an example of the first and the Anglo-Saxon invasion an example of the second.

    Religon might follow the lower threshold because it’s more a matter of elite emulation whereas language is a mixture of elite emulation and practicality.

  5. Grey

    Also, it makes me think the varying proportions of maternal and paternal DNA across different sites may be more relevant than the whole in discovering prehistory through genetics.

    Also that the oldest still-existing DNA in a geographic region will be paternal DNA in the least desirable real estate *if* it matches the female DNA or the female DNA in the same real estate otherwise.

    In European terms that might be North Wales or the remoter parts of Ireland, Scotland and Brittany. North wales in particular because although the language is Celtic there is a distinctive physical type there which is shortish, darkish, curly brown hair, brown eyes – (i always assumed they were the basis for Tolkein’s hobbits) – which is quite distinctive.

  6. I think the dominant language outside the home would be the critical factor.

    yeah, language socialization in “peer groups” is probably driving this i suspect.

  7. Grey

    Too many comments but anyway…

    If one standard pattern where a population was seemingly completely replaced was:

    1) Whole-population colony settles at the mouth of a river. Mostly maternal incomer DNA with some native.

    2) Male-only or male-majority conquest groups expanding radially from original colony replacing native males on a village by village basis but *only* within a distance that can be militarily supported by the original colony. All or mostly all native maternal DNA. Incomer language remains dominant however because still linked to main site.

    3) Whole-population offshoot colony from the original (now with some maternal native DNA) settles upriver a few miles. A bit more native maternal admixture

    4) Repeat of the radial male-only conquest groups again within support distance of the new colony.

    Rinse and repeat.

    If so then a pre-historic expansion might still be visible in some places by a plot of the proportion of incomer and native maternal DNA i.e if native maternal DNA was dark blue and incomer light blue the initial colony would stand out and the various daughter colonies also as they gradually shaded from lighter blue to darker blue along the path(s) of expansion.

  8. Roger Bigod

    There may have been a male preponderance in the first settlement of the Saxon Shore, but the expansion over England took several centuries. The Y haplotype was R1a, and there’s only about 10% in the British Isles data base. I have trouble imagining the social processes and institutions that produced the complete language replacement. The Norman Conquest had little impact on the gene pool and didn’t replace the language. I’ve never seen it put this way, but English is a creole of Medieval Norman French.

    There’s probably more to be learned from the variety and sharp contrasts of the Asian and African examples of language replacement. This is in keeping with the WASP striving to be bland and boring, within limits.

  9. Grey

    “There’s probably more to be learned from the variety and sharp contrasts of the Asian and African examples of language replacement.”

    No doubt. My point was simply they’re two known examples of invasions of the same place where one led to language replacement and one didn’t. The one that didn’t is known to have been elite replacement only. The other is less clear cut but there must have been a difference of some kind because it did lead to language replacement.

  10. Jamie

    Fascinating article – the comments are even more interesting! First time I have ever heard that “English is a creole of Medieval Norman French!” If so, then I can relax instead of complaining to myself that I still haven’t learned French! There is still so much to be learned about genetics, cultural and linguistic development. This is a wonderful conversation.

  11. “I have trouble imagining the social processes and institutions that produced the complete language replacement.”

    One process that is probably pretty pertinent is that the conquerers probably had a single language that was highly homogenized following the practices of their own ruling court. In many cases, however, the conquered people would have had a hodge-podge of somewhat related dialects with the dialects of geographically distant communities being mutually unintelligible and each individual dialect being spoken in a consistent way by only a small community of illiterates.

    In a divide and conquerer kind of linguistic scenario it is somewhat less difficult to see how the language of the conquerers that carried both access to power in the community and horizons beyond one’s own local community, could be seductive.

  12. Roger Bigod

    I checked some papers on the Anglo-Saxon invasion, and one states that the replacement of Celt by Anglo-Saxon Y markers was 65-100% in the Midlands, a typical area.

    If this is correct, it wasn’t a matter of elite replacement. Any speakers of a Celtic dialect remaining would have been under considerable pressure to adopt Anglo-Saxon.

    The source I remembered stated that Anglo-Saxons replaced only 20% of the gene pool of the British Isles. But this ignores the enormous local variation.

    Also, R1a is a minor haplotype in the Anglo-Saxon gene pool and represents only about 5% of the British population. It may have arrived with Viking invaders, as well as Anglo-Saxons.

  13. Slyfox666

    The Norman Conquest of 1066 replaced virtually all of the Anglo-Saxon (and remaining Celtic) nobility with Norman masters (Robin Hood notwithstanding). French became the language of the nobility, government, justice, and the bourgeoise. It remained so for several hundred years. Even the word Parliament is a corruption of the French “Parle Mont” (speaking place). Parliament was ceded (grudgingly) legislative power by King John after the battle of Runnymeade in 1215. By the time of Shakespeare’s Henry V, English had become a meld of Anglo-Saxon and French words.

    Vestiges exist in the American judicial system. In many states, the baliff enters the court room shouting the words “Oyez, Oyez, Oyez” (French for “Hear Ye”), the ____ court is now in session, Judge _____ presiding.

    Another surviving genetic vestige was discovered during WWII. The Germans bombed many areas other than London. After one particularly devastating raid, the general populace was asked to donate clothing and shoes to those who had lost everything. The authorities discovered that the shoe sizes worn in at least two different areas were vastly different. The dividing line was approximately along the trace of Offa’s dike.


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