This article is conflating language and ancestry. The seed of the confusion is in Reich’s research but the WSJ journalist blows it up to preposterous levels. Take India as an example. Most of the population is speaking some variant of an Indo-European (Indo-Iranian to be more precise) language but only a minority is genetically traced to Indo-European steppe people [1]
>> This article is conflating language and ancestry
From the article:
"DNA detectives, including at Reich’s lab, analyzed DNA samples from the remains of around 450 prehistoric individuals taken from 100 sites in Europe, as well as data from 1,000 previously known ancient samples"
Ancestry, not language.
"Reich’s award-winning lab at Harvard has one of the largest ancient DNA databases in the world and uses proprietary gene-analysis software co-developed by Nicholas Patterson, a British mathematician who once worked as a codebreaker for U.K. intelligence services."
Ancestry, not language.
"DNA evidence shows that the proto-Yamnaya population migrated from the Volga region to Anatolia"
Ancestry, not language.
"In many places, indigenous male DNA disappears upon the arrival of the Yamnaya, while indigenous female DNA is traceable in the following generations"
Ancestry, not language.
"Within years of their arrival, some 99% of the indigenous people disappeared, according to Reich’s analysis of DNA samples from the time"
Ancestry, not language.
I rate your claim that "This article is conflating language and ancestry" as false, and I award you no points.
This article's confusion is where it states "half the human beings alive today are descended from the Yamnaya." He thinks because half of the world population speaks an Indo-European language, and because the original speakers of the Proto-Indo-European languages were the Yamnaya culture (as Reich's research suggests), then half of the world population are descendants of the Yamnaya culture.
Archive version of the above science.org article "Where did India’s people come from? Massive genetic study reveals surprises
Analysis confirms Iranian influx, but also finds genes from Neanderthals and a mysterious human ancestor": https://archive.is/Wd4tP
That article says nothing about the percentage genetic component of the Indo European step people in the Indian population. It does mention a high genetic similarity to Iranians.
And interestingly Iranians are mostly not the descendants of the so-called Indo-Iranian steppe nomads (genetically). But they speak various Iranian languages.
Does a claim like this even have any meaning at all?
If you assume each generation is 25 years, then everybody alive today has 2^200 ancestors from 5,000 years ago. Which is obviously way more people than even existed in the world because your ancestors start overlapping, but the point is that you could probably make the claim that "half the human beings alive today" are descended from tons of groups of humans that existed 5,000 years ago. People travel and migrate and marry and genes get passed on at an exponential rate.
I'm glad someone gets the intuition of the Charlemagne paradox. In fact ALL people today are related if you go back 3000 years, but the aboriginees might have very little DNA from the pharaohs. The point is that ONE individual 1500 years ago traveling to Australia (and bringing the dingo) is enough to connect these graphs. The only question is: how related. 0.0000001%? ;)
It means that if you trace Y-chromosomes back 5000 years, you'll find that grand-grand-grand-...-fathers of 40% of people are concentrated in the area of the Yamnaya culture. Grand-grand-grand-...-mothers would be from multiple groups, yes.
> you could probably make the claim that "half the human beings alive today" are descended from tons of groups of humans that existed 5,000 years ago
Surely it's still notable to identify specific groups for which this is true, particularly when the group itself is primarily identified by an unrelated archaeological characteristic.
hard not to interpret these steppe horse cultures as being the centaurs of mythology.
lots of pet theories but this idea that much human language originated with them implies further to me that horsemanship was the origin of western ethics of stewardship and morality. riders require a unique ontology that includes sophisticated communion with other beings, and it's literally the approbation of nature. mythic surely, but it may also have some predictive power. fun stuff
I haven't read him. but the competence in it is so rare these days that the understanding has been reduced to being on the spectrum of yoga and pilates vs. its history as a much deeper classical art
Readit News
[1] https://www.science.org/content/article/where-did-india-s-pe...
The most common ancestry in the US is German, not English, but English is still the dominant language. Language isn't DNA.
From the article:
"DNA detectives, including at Reich’s lab, analyzed DNA samples from the remains of around 450 prehistoric individuals taken from 100 sites in Europe, as well as data from 1,000 previously known ancient samples"
Ancestry, not language.
"Reich’s award-winning lab at Harvard has one of the largest ancient DNA databases in the world and uses proprietary gene-analysis software co-developed by Nicholas Patterson, a British mathematician who once worked as a codebreaker for U.K. intelligence services."
Ancestry, not language.
"DNA evidence shows that the proto-Yamnaya population migrated from the Volga region to Anatolia"
Ancestry, not language.
"In many places, indigenous male DNA disappears upon the arrival of the Yamnaya, while indigenous female DNA is traceable in the following generations"
Ancestry, not language.
"Within years of their arrival, some 99% of the indigenous people disappeared, according to Reich’s analysis of DNA samples from the time"
Ancestry, not language.
I rate your claim that "This article is conflating language and ancestry" as false, and I award you no points.
Is the logical error clear now?
Deleted Comment
https://www.nature.com/articles/s41586-024-08531-5
Lazaridis, I., Patterson, N., Anthony, D. et al. The genetic origin of the Indo-Europeans. Nature 639, 132–142 (2025)
"supported in part by the National Institutes of Health and National Science Foundation" [1]
Also possibly this paper: https://www.nature.com/articles/s41586-024-08372-2
[1] https://hms.harvard.edu/news/ancient-dna-study-identifies-or...
If you assume each generation is 25 years, then everybody alive today has 2^200 ancestors from 5,000 years ago. Which is obviously way more people than even existed in the world because your ancestors start overlapping, but the point is that you could probably make the claim that "half the human beings alive today" are descended from tons of groups of humans that existed 5,000 years ago. People travel and migrate and marry and genes get passed on at an exponential rate.
r1b's population is only 190 million [1]
[1]https://www.razibkhan.com/p/the-haplogroup-is-dead-long-live...
Surely it's still notable to identify specific groups for which this is true, particularly when the group itself is primarily identified by an unrelated archaeological characteristic.
Deleted Comment
lots of pet theories but this idea that much human language originated with them implies further to me that horsemanship was the origin of western ethics of stewardship and morality. riders require a unique ontology that includes sophisticated communion with other beings, and it's literally the approbation of nature. mythic surely, but it may also have some predictive power. fun stuff
https://www.science.org/doi/10.1126/sciadv.ade2451