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

The twilight of American behavior genetics

Posted on April 6, 2022April 6, 2022 by Razib Khan

Many people, including some prominent scientists, have emailed me about the review of K. Paige Harden’s book The Genetic Lottery in The New York Review of Books: Why Biology Is Not Destiny – In The Genetic Lottery, Kathryn Harden disguises her radically subjective view of biological essentialism as an objective fact. It’s a pretty intense review. I thought it was mostly unfair, but even I winced at the punches that it got in. Give the devil his due?

There are two authors, M.W. Feldman and Jessica Riskin. I don’t know Riskin, but Feldman is an extremely eminent population geneticist whose influence is felt in others fields. He was an early founder of the field of cultural evolution, writing Cultural Transmission and Evolution with L. L. Cavalli-Sforza. From what I know, Peter Richerson attended classes on the topic taught by Feldman in the 1980s, and Richerson was Robert Boyd’s advisor, who was Joe Henrich’s advisor.

Indirectly Feldman is arguably the father of much of contemporary cultural evolution and cliodynamics.

As for what’s unfair in the piece, I think the below passage illustrates the method well:

“This polygenic index will be normally distributed,” Harden continues, now disguising an assumption—that there are intrinsic cognitive and personality traits whose distribution in a population follows a bell-shaped curve, a founding axiom of eugenics—as an objective fact.

Eugenics aimed to be an applied branch of hereditary science, and before the emergence of Mendelism, it was driven by ‘biometric’ thinking about continuous quantitative traits (though Francis Galton himself believed in “sports” and other noncontinuous changes). So the connection between eugenics and the normal distribution does exist…but the fact is that the normal (Gaussian) distribution is ubiquitous in science, and emerges out of the central limit theorem. Anyone with a cursory background in quantitative sciences won’t really associate the normal distribution with eugenics. How exactly was Harden going to write a book on behavior genetics without mentioning the normal distribution?

There is a lot of that sort of guilt-by-association and verbal sleight of hand. I think most fair-minded scientists will see what they did, but that’s not how the typical reader of The New York Review of Books will read the piece. Rather, they’ll see one hard-hitting salvo after another from two Stanford academics, one of whom is one of the most accomplished population geneticists of our day. How are they going to know about the ubiquity of the normal distribution and its centrality to much of modern statistics? Instead, most readers will experience an incredibly erudite and magnificent demolition of The Genetic Lottery and its presuppositions and implications.

Just to show you what I mean, let’s look at this passage. I’m going to add numbers that will help in the exegesis:

Harden condemns Jensen’s racism and rejects his assertion that social interventions are futile, but she doesn’t question his basic claim that genetic differences produce an [1] innate hierarchy of scholastic achievement. She also doesn’t acknowledge his dependence on fraudulent data from a 1966 paper by the English psychologist and geneticist Cyril Burt purporting to compare identical twins raised together and apart [2]. And nowhere does she cite the Princeton psychologist Leon Kamin’s 1974 devastating debunking of Jensen and Burt or engage with the critical problems Kamin raised there regarding twin studies in general, because of the impossibility of isolating genetic factors from environmental ones [3].

First, [1], there’s nowhere in the book that I recall Harden talking about “innate hierarchy.” One can see why Feldman and Riskin use this term, but the concept is repeated so often in the review that I believe most readers will believe this is exactly what Harden talked about in the book. No, it’s their interpretation and imputation. But repeated enough the allusive imputation becomes the literal fact.

As for [2], the Cyril Burt controversy continues down to the last I checked. He may have committed fraud, or he may not have. Feldman and Riskin take it as a given that the initial accusations of fraud were correct and not disputed. The reader won’t know the controversy about the controversy. As for [3], the reader will be unaware that in the 1970’s Kamin actually floated the position that the heritability of intelligence was ~0, severely undermining his credibility as a sagacious researcher. Second, the implication that there is an impossibility of isolating genetic factors from environmental ones is coherent in light of all the complexities of…complex traits. But this applies to many behavioral and nonbehavioral traits that are polygenic. Should we entertain the possibility that we can’t adduce the genetic aspect of schizophrenia due to philosophical quibbles about causality? (it’s very heritable)

If you want to know more about behavior genetics and genomics, I recommend this interview I did with James Lee, or this one from the fall with Alexander Young. As for Harden and The Genetic Lottery, I’m glum about the prospects for any project like this in this country. Though Harden is a tenured academic and comes out of a pretty good lineage (her advisor is Eric Turkheimer), ever since the bizarre piece in The New Yorker her reputation in many biological academic circles has taken a hit. From everything I know, she is sincere, earnest and a legitimate political progressive/liberal. But you wouldn’t know that from what some people say, and the lie repeated often enough becomes the truth. In 2006 I interviewed the famed population geneticist James F. Crow, and he stated that he felt “strongly that we should not discourage a line of research because someone might not like a possible outcome.” Crow died in 2012 at the age of 96, and from what I can see this sort of stance is mostly held by older academics.

A different wind blows in the future. Harden has tenure, and a band of her fellow travelers will continue to exist in American academia, but more and more they will be pariahs. Why? I’m not progressive so that’s beyond my pay grade. But there are other countries out there, so there will be places this sort of research continues.

The truth is what it is. No matter what the flock says.

Addendum: Bert Hölldobler, a long-time collaborator of E. O. Wilson has written a defense of him. Richard Lewontin was a great population geneticist, but I think his biggest impact will turn out to be the style of intellectual pugilism he promoted. It’s normative now among many younger academics. Here’s a section that jumped out at me:

It was a point that Dick Lewontin himself acknowledged when he showed up at my office the next day, apparently eager to soften what he had said. Although I respected Lewontin as a scientist and colleague at Harvard, I did not appreciate his ideologically driven “sand box Marxism.” When I asked why he so blithely distorted some of Ed’s writings he responded: “Bert, you do not understand, it is a political battle in the United States. All means are justified to win this battle.”…

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

Avars were Rourans

Posted on April 2, 2022April 2, 2022 by Razib Khan

genomes reveal origin and rapid trans-Eurasian migration of 7th century Avar elites:

The Avars settled the Carpathian Basin in 567/68 CE, establishing an empire lasting over 200 years. Who they were and where they came from is highly debated. Contemporaries have disagreed about whether they were, as they claimed, the direct successors of the Mongolian Steppe Rouran empire that was destroyed by the Turks in ∼550 CE. Here, we analyze new genome-wide data from 66 pre-Avar and Avar-period Carpathian Basin individuals, including the 8 richest Avar-period burials and further elite sites from Avar’s empire core region. Our results provide support for a rapid long-distance trans-Eurasian migration of Avar-period elites. These individuals carried Northeast Asian ancestry matching the profile of preceding Mongolian Steppe populations, particularly a genome available from the Rouran period. Some of the later elite individuals carried an additional non-local ancestry component broadly matching the steppe, which could point to a later migration or reflect greater genetic diversity within the initial migrant population.

No big surprises, but I think it is important to note that it looks like the East Eurasian Avar elites brought a lot of Iranian-steppe people as cadet elites. So a lot of the elite non-East Eurasian ancestry turns out to be non-European, and more Central Eurasian (probably Alanic and the like).

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

The Afanesievo live!

Posted on April 1, 2022April 1, 2022 by Razib Khan

Bronze and Iron Age population movements underlie Xinjiang population history:

The Xinjiang region in northwest China is a historically important geographical passage between East and West Eurasia. By sequencing 201 ancient genomes from 39 archaeological sites, we clarify the complex demographic history of this region. Bronze Age Xinjiang populations are characterized by four major ancestries related to Early Bronze Age cultures from the central and eastern Steppe, Central Asian, and Tarim Basin regions. Admixtures between Middle and Late Bronze Age Steppe cultures continued during the Late Bronze and Iron Ages, along with an inflow of East and Central Asian ancestry. Historical era populations show similar admixed and diverse ancestries as those of present-day Xinjiang populations. These results document the influence that East and West Eurasian populations have had over time in the different regions of Xinjiang.

This is a hard paper to read. The terminology could be clearer, and the narrative crisper. Perhaps this is just due to the need to compress into the length science wanted, I don’t know. But I find it hard to read, I doubt others will find it easier.

Here’s the main figure:

Here are my general summaries

1 – ~5,000 years ago the indigenous people of “Xinjiang” (northern half = Dzungaria, southern half = Uighurstan) were mostly distantly related to “Ancient North Eurasians” (the Tarim Basin mummies)

2 – Around this time early Indo-Europeans, the Afanasievo culture, begin to migrate south of the Altai pasturelands to the north. The Afanasievo are basically genetically 100% Yamnaya initially; their culture was really just an eastward migration of Yamnaya. They immediately start mixing with the indigenous ANE-related population, and also to some extent with small numbers of people from Northeast Asia.

3 – After 2000 BC two new entrants arrive on the scene. First, Indo-Iranians, second-wave Indo-Europeans who are different from the Afanesievo in having “Anatolian farmer” ancestry due to their provenance in the borderlands of Poland, Belarus and Ukraine. The Indo-Iranians, part of the Andronovo horizon, assimilate with the local populations, and almost certainly are responsible for some of the Iranian languages prominent in early historical records in the southwestern part of the Tarim Basin (e.g., Khotanese). Ad this time there is also a clear influx of Central Asian/Turanian ancestry related to the BMAC culture. These are basically pre-Indo-European Iranians, and you can think of them as strongly skewed to Zagrosian ancestry, along with some ANE/WSHG admixture, and minimal but nontrivial Anatolian farmer (the ancestors of the Indus Periphery population are zeroed out for Anatolian farmer).

If you look at admixture analyses of modern Uyghurs you see some Persian-related ancestry. This may in fact be Iranians from the medieval period, but perhaps more likely they’re due to the mixture between Andronovo and BMAC ancestry, which would perfectly mimic Persian ancestry.

At least period you also start seeing more East Asian ancestry with a southern affinity.

4 – By the historical period, you have the migration of populations from Central Asia, Siberia, and East Asia. Some of these you recognize. The Sakas and the Old Uyghurs for example. Others, you may not.

Looking at the admixture, one of the historical era samples is clearly Northwest Indian in ancestry. There is some evidence of Indian ancestry as early as the Bronze Age as well. This is a small component, but it is clearly not trivial. This particular result puts Kumarajiva in perspective.

The major finding of this paper seems to be that Xinjiang has not seen population replacement, but assimilation. The indigenous Tarim Basin ANE-derived population still make a substantial contribution to modern Uyghurs, as do the Afanasievo early Indo-Europeans, later Indo-Iranians, Xiongnu, Han Chinese and Central Asians.

Finally, it does look like the admixture between Afanasievo and ANE-derived indigenous populations dates to between 2500 and 3000 BC. Judging from the distinctiveness of Tocharian languages the authors claim that this means that it seems likely Tocharian is an Afanesievo language (and if that is true, then the original Yamnaya spoke something similar to Tocharian).

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

Ancient Africa may not have had as much deep structure as we think

Posted on March 29, 2022March 29, 2022 by Razib Khan

A weakly structured stem for human origins in Africa:

While it is now broadly accepted that Homo sapiens originated within Africa, considerable uncertainty surrounds specific models of divergence and migration across the continent. Progress is hampered by a paucity of fossil and genomic data, as well as variability in prior divergence time estimates. Here we use linkage disequilibrium and diversity-based statistics, optimized for rapid, complex demographic inference to discriminate among such models. We infer detailed demographic models for populations across Africa, including representatives from eastern and western groups, as well as 44 newly whole-genome sequenced individuals from the Nama (Khoe-San). Despite the complexity of African population history, contemporary population structure dates back to Marine Isotope Stage (MIS) 5. The earliest population divergence among contemporary populations occurs 120-135ka, between the Khoe-San and other groups. Prior to the divergence of contemporary African groups, we infer long-lasting structure between two or more weakly differentiated ancestral Homo populations connected by gene flow over hundreds of thousands of years (i.e. a weakly structured stem). We find that weakly structured stem models provide more likely explanations of polymorphism that had previously been attributed to contributions from archaic hominins in Africa. In contrast to models with archaic introgression, we predict that fossil remains from coexisting ancestral populations should be morphologically similar. Despite genetic similarity between these populations, an inferred 1–4% of genetic differentiation among contemporary human populations can be attributed to genetic drift between stem populations. We show that model misspecification explains variation in previous divergence time estimates and argue that studying a suite of models is key to robust inferences about deep history.

Privately some people have been grumbling about models of deep structure between very differentiated populations for a while. They claim this is just a bias in the model specifications because it’s so easy to think of gene flow happening in periodic pulse admixtures. But the reality is that Africa doesn’t seem to have had the same barriers as across Eurasia or between Eurasia and Africa, so how are these deep lineages persisting?

The preprint here shows that the data can fit a different model, one that they find more biologically and paleoanthropologically more reasonable. The discussion has an “out of Africa with total replacement” flavor, but here it is within Africa:

Multiple studies have shown a correspondence between phenotypic differentiation, usually assessed with measurements of the cranium, and genetic differentiation among human populations and between humans and Neanderthals 36,37,38 (see also Section 5.3). This correspondence allows predictions of our model to be related to the fossil record. The fossil record of Africa is sparse during the time period of the stems, but of the available fossils, some are very similar in morphology to contemporary humans (e.g., from Omo Kibish, Ethiopia 39,40), others are similar in some morphological features but not others (e.g., from Jebel Irhoud, Morocco 1,41), and others are very different in morphology (e.g., from Dinaledi, South Africa 42,43). If, as our model predicts, the genetic differences between the stems were comparable to those among contemporary human populations, the most morphologically divergent fossils are unlikely to represent branches that contributed appreciably to contemporary human ancestries.

This result would recenter Omo Kibish from what I can tell.

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

Depigmentation in Northern Europe

Posted on March 17, 2022March 17, 2022 by Razib Khan

Direct detection of natural selection in Bronze Age Britain:

We developed a novel method for efficiently estimating time-varying selection coefficients from genome-wide ancient DNA data. In simulations, our method accurately recovers selective trajectories, and is robust to mis-specification of population size. We applied it to a large dataset of ancient and present-day human genomes from Britain, and identified seven loci with genome-wide significant evidence of selection in the past 4500 years. Almost all of them are related to increased vitamin D or calcium levels, and we conclude that lack of vitamin D and consequent low calcium was consistently the most important selective pressure in Britain since the Bronze Age. However, the strength of selection on individual loci varied substantially over time, suggesting that cultural or environmental factors moderated the genetic response to this pressure. Of 28 complex anthropometric and metabolic traits, skin pigmentation was the only one with significant evidence of polygenic selection, further underscoring the importance of phenotypes related to vitamin D. Our approach illustrates the power of ancient DNA to characterize selection in human populations and illuminates the recent evolutionary history of Britain.

If you read this blog closely over the years you will know that I’ve noticed that the ancient DNA has yielded the general result that Bronze Age Europeans were somewhat darker in complexion than their modern descendants. This is without mass overall population genetic change in many areas. This is not always statistically significant, but you can tell a likelihood before something hits p = 0.05.

The ancient DNA temporal transects are pretty good for Britain. The best in Europe. And the above preprint seems to present a strong result of depigmentation over the last 4,500 years in Britain using powerful new methods drawing on ancient DNA sources. I strongly suspect this generalizes to much of Northern Europe.

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

Why my Substack posts are better and worse than ancestry calculators

Posted on March 15, 2022March 15, 2022 by Razib Khan

Of all my Substack posts, Ashkenazi Jewish genetics: a match made in the Mediterranean has been the most popular of the paid posts. It prompted this response from a reader:

The issue here is that my Substack is doing something different than what personal genomics companies are trying to do. My Substsack post is giving a survey of a whole population and its history, a personal genomics test is trying to give an individual estimate that is intelligible. When 23andMe or the other companies tell you are are 99% “Ashkenazi Jewish” it is simply giving you confirmation that you’re within the range of variation typical for Ashkenazi Jews (there is some suspicions from genealogy enthusiasts that 23andMe smooths out differences between Galicianers and Litvaks, for example).

Imagine that 23andMe told its Jewish customers that they were 45.3% Northern Levantine, 40% Southwest European, and 9.7% Northern European. How would they interpret it? Sophisticated users would understand this points to a deep history of admixture, but most users are not sophisticated. They want to know that they’re Ashkenazi Jewish, and how Jewish they are (most will be nearly 100%, but some people may have non-Ashkenazi cryptic ancestry).

When I worked for Embark Vet. one of the issues that the canine DNA test was having is that we were looking for wolf ancestry in dogs, as some customers with F1’s or backcrosses wanted to test their pooch. But, it turned out that you had to be careful because some northern Arctic dog breeds kept coming back with “wolf” ancestry at low fractions because they did have ancient wolf ancestry. But that’s not what the wolf test was designed to pick up.

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

Tibetans as the compound of two populations

Posted on March 9, 2022March 9, 2022 by Razib Khan

A new paper looks at some ancient Tibetan genomes:

Present-day Tibetans have adapted both genetically and culturally to the high altitude environment of the Tibetan Plateau, but fundamental questions about their origins remain unanswered. Recent archaeological and genetic research suggests the presence of an early population on the Plateau within the past 40 thousand years, followed by the arrival of subsequent groups within the past 10 thousand years. Here, we obtain new genome-wide data for 33 ancient individuals from high elevation sites on the southern fringe of the Tibetan Plateau in Nepal, who we show are most closely related to present-day Tibetans. They derive most of their ancestry from groups related to Late Neolithic populations at the northeastern edge of the Tibetan Plateau but also harbor a minor genetic component from a distinct and deep Paleolithic Eurasian ancestry. In contrast to their Tibetan neighbors, present-day non-Tibetan Tibeto-Burman speakers living at mid-elevations along the southern and eastern margins of the Plateau form a genetic cline that reflects a distinct genetic history. Finally, a comparison between ancient and present-day highlanders confirms ongoing positive selection of high altitude adaptive alleles.

Y haplogroup D is found at high frequencies in Japan, Tibet, and the Andaman Islands. It strikes me this is evidence of a Paleolithic substrate, though the graph above shows that it diverged really deep in Eurasia, and in the text they say it’s only in Tibet.

The Tibetan ancestry seems to have been found in the Himalayan zone by 1450 BC, so rather early.

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

Seeing the world through other eyes

Posted on March 8, 2022March 8, 2022 by Razib Khan

As most of you know I am the child of Bangladeshi immigrants to the US. I don’t make much of my “identity” because it rests lightly on me, and is not a major concern. I’ve been to Bangladesh twice in the last 40 years. My views on ascriptive identity are old-fashioned, you should listen to me because I am a human, not because of my sex, gender, class, race or religion. My experience and background are not trivial, but neither are they the most important thing.

But sometimes they do matter. Recently I saw this Tweet:

This person lives in Washington D.C. and refers to herself as “Tree-hugging, granola-crunching, whale-saving, ACLU card-carrying, liberal Democrat; world traveler; tennis fanatic; animal lover; political junkie and activist.”

I think it is understandable that Lithuania is angry considering its geopolitical circumstances. The cancelation of the shipment seems petty, but it’s obviously within their rights, and for historical reasons, Lithuanians are extremely passionate about the current conflict in Ukraine and look very negatively upon Russia.

But what about Bangladesh? Here I can actually offer some personal perspective, because my parents grew up in Bangladesh (East Pakistan), and much of my family lives in Bangladesh. On the whole, feelings toward Russia are warm, if somewhat distant and abstract. On a geopolitical level, Russia has been a “friend” to both India and Bangladesh for decades. This is not just a theory at the scale of the nation-state, there were personal connections, as Indians and Bangladeshis traveled to the Soviet Union to study, and the USSR sent advisors to the subcontinent. On the merits Indians and Bangladeshis may not be comfortable with the Russian invasion, but should they turn their back so quickly on a relationship that goes back decades? Will Western countries embrace India and Bangladesh with open arms to reward them for their actions?

For Bangladesh, there is a more concrete historical reason for Russophilia: the Soviet Union was in the end on the side of India and the soon-to-be Bangladesh during the 1971 conflict with Pakistan. Because the US was a staunch ally of Pakistan, the official government’s position was to ignore evidence of massive human rights atrocities being reported by their own diplomats. The Bengali civilian death toll is usually given to be in the range of ~100,000 to 2 million. The latter figure actually comes from Pravda, and I think there is reason to be skeptical that 1 out of 33 Bengalis in East Pakistan were killed. But the ~100,000 figure is possibly too low. In any case, it wouldn’t be a trivial death toll even if it was around 100,000, and the need for widespread abortion clinics after the war attests to mass rapes (the rape had a eugenic intent, a Pakistani general asserted that they would “change the race of this bastard nation”).

The Nixon administration even took some threatening moves with naval power once India intervened and was clearly going to defeat Pakistan, aided by the Bengali nationalist left-wing militias. The Soviet Union mobilized its own naval power to check the US. People of my parent’s generation remember these events with some clarity (my mother was shot by Pakistani soldiers).

In 1972 Bangladesh was founded as the “People’s Republic of Bangladesh.” The name should make it clear that Bangladesh’s origin was as a secular socialist left-nationalist nation-state. Over the decades many things have changed, in particular, the rise of a more Islamic self-conception and the shift away from socialism to export-oriented capitalism. But the founding myth of a socialist nationalist struggle remains, and people of my parents’ generation remain strongly influenced by 1970’s Third World socialism.

The Russian invasion of Ukraine has been a disaster for geopolitical stability, and now the world economy. It’s been a disaster for Ukraine, and Russia is not really benefiting so much in material terms. I am personally terrified of the increased risks of nuclear war. All that being said, are the Russians intent on a war of total subjugation laced with genocide? My own understanding is that they thought Ukrainian nationalism was a paper tiger and that the corrupt government would fall and they would take over quickly. The Ukraine invasion is far more important than the genocide in Bangladesh in the early 1970’s (that targeted Hindus and intellectuals) because the fate of the world hangs in the balance, even if the probabilities are low. But to be candid on the grand scale of humanitarian disasters I doubt the civilian death count will reach anything like what happened in Bangladesh. Would Bangladeshis really want to sacrifice the old friendship for abstractions about the international order? Or a humanitarian crisis of far lesser magnitude than what they themselves went through two generations ago?

In the years after 9/11 the US went through foreign policy disasters because it refused the understand the world that it tried to change. There are other histories and other viewpoints out there. You may not agree with them, but they are there nevertheless.

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

Liberals are wordcels par excellence

Posted on March 8, 2022March 8, 2022 by Razib Khan

Matt Yglesias asks why more selective colleges are more left-wing. There are several reasons, but I think one of them is that liberals just tend to have somewhat higher verbal, and that really matters at the “tails” of the distribution.

Below is some data from the GSS. I limited the sample to non-Hispanic whites (which means it’s from the year 2000 and later), and stratified by low, average and high vocab score (0-5,6-8 and 9-10), and crossed that with political ideology. Then, I also broke it down by those with noncollege education, those with bachelor’s degrees and those with graduate degrees. There’s just a bigger pool of very verbally strong liberals (or, being verbally strong makes you liberal, I don’t know).

		no college
	liberal	moderate	conservative
low vocab	33	43	35
average vocab	57	51	58
high vocab	10	6	6
		bachelor’s
low vocab	10	19	18
average vocab	58	65	64
high vocab	32	17	18
		grad degree
low vocab	6	8	11
average vocab	46	57	62
high vocab	49	35	28