Substack cometh, and lo it is good. (Pricing)

Five things paleogenetics tells us about the human past

Since I’m flogging Enlightenment Now, I thought perhaps I should remind readers that Who We Are and How We Got Here: Ancient DNA and the New Science of the Human Past by David Reich is out in 1.5 months. For years people have asked me about a book to read to understand what genetics has to say about human history. This is that book.

And yet before you get there, what do you need to know?

Here are five things you should know. Five things that we know with a very high degree of certitude.

  1. Many (most?) modern populations clusters we perceive as clear and distinct date to the last 5,000 years. To give a concrete example, the genetics that we find to be typical of Northern Europeans only comes into being ~5,000 years ago, with the Corded Ware populations. To my knowledge none of the prior populations along the North European plain exhibit the mix of characteristics and ancestries typical of modern Northern Europeans in any way, shape, or form.
  2. Concomitantly, many of the physical characteristics we find typical of modern populations are probably relatively recent configurations due to natural selection.
  3. Non-African populations, whether European, Middle Eastern, South Asian, (South)East Asian, Amerindian or Oceanian, derive from a population expansion that dates to ~50,000 years BP. These populations experienced a bottleneck on the order of 1,000 to 10,000 breeding individuals.
  4. Modern humans are old. Population structure within Africa of modern humans dates to at least 200,000 years before the present, and perhaps even earlier.
  5. Population turnover was ubiquitous. Change was the only constant.

6 thoughts on “Five things paleogenetics tells us about the human past

  1. By Ussher’s calculations, the Bible also had a year-zero at 4004 BCE. These “Priestly” genealogies are more Mesopotamian in mindset than anything else (presumably exilic).

    It’s as if Semitic civilisation retained a vague memory back to the first Akkado-Sumerian tablets, in the fourth millennium BCE, but no further.

  2. I’m REALLY looking forward to this book!

    “Many (most?) modern populations clusters we perceive as clear and distinct date to the last 5,000 years.”

    I would say this as “genetically speaking, humans are a product of the Metal Ages.” This makes the massive genetic changes of the last 500 years in the Americas and Australasia part of the picture. Archaeological trends had been against migration and replacement in the past 50,000 years, which had the effect of making the last 500 look sui generis. In fact, it seems to be in continuity with these past processes, which were just as brutal as the conquest of the Americas (and probably involved disease, as well).

    I wonder if the publication date of this book, late March, will coincide with the dilatory Indian ancient DNA paper. I understand some of the research was supported by Harvard, where Reich is based. I can’t imagine he will leave South Asia out of his book, and doesn’t want to hold himself back in his magnum opus in order to protect confidences. Do you think his lab will put pressure on the paper’s authors to publish before late March?

  3. So what we think of today as the three major races (outside the Americas and Australia) are in fact (relatively) new breeds? Or is it just mostly Caucasians?

  4. @Luke, I’d say that’s a question that’s pretty difficult to answer. A quick take below:

    On the one hand, talking about race = cluster, and the question is “Does 3 clusters to explain West Eurasia+East Eurasia+Africa even make sense in fairly ancient periods?”, then a 3 cluster solution to a set where we have the most distant ancient West Eurasians at about 10,000 BC (EHG, WHG, Natufians, Iran_Upper Paleolithic), ancient East Eurasians from the same time and Africans, would probably still find the most optimal solution would be to place all the West Eurasians in a cluster, then the East Eurasians and Africans in another.

    (For a basic version see here: https://imgur.com/ME3MqS9 based on Fst, though including AG3-MA1 instead of EHG, and EHG are basically in between AG3-MA1 and WHG. I’ve got a table with EHG, but it doesn’t look hugely different enough and I have not got time to dig out atm. Note the struture of the tree).

    But it has to be noted, the degree to which the 10,000 BC West Eurasian groups would be best fit as a cluster together would be much shallower relative to the outgroups than present day West Eurasians. So the idea of 3 clusters (races) tells us less about the overall structure of the populations.

    On the other hand, if we are talking about race = phenotype cluster, then that’s actually harder to answer. Pre-ancient dna we only had hard tissues, and while Dienekes Pontikos found some solutions which show that the presence of “Caucasoid”, “Mongoloid” and African skull morphologies, others show clustering of paleolithic humans without regard to geography, mostly due to greater robusticity that they share. When it comes to soft tissue features like skin colour, hair form, etc. the WHG genetics shows that either early West Eurasian groups had different phenotype or even if early West Eurasian groups had similar phenotype and our understanding of the genetics is incomplete, then the genetic basis was quite different between groups (and probably with at least some effect on their phenotype).

Comments are closed.