On the origin of whiteness

https://www.reddit.com/r/unitedkingdom/comments/7zi904/ancient_darkskinned_briton_cheddar_man_find_may/duohm8i/

The link above is the retraction and correction to the Cheddar Man hoax, copied and pasted on a reddit forum since the original is behind a paywall.  Given what it suggests, the question posed in the post title may well be impossible to determine.  First, let me back up... what do I mean by "whiteness?"

I'm mostly referring to the suite of characters in the "average" phenotype that is most common in Northern Europe.  Pale skin, light-colored eyes, and blond hair (by blond I mean "not black.")  Now, of course, plenty of white people don't have all of those characters.  My wife, for example, is about as white as you can get.  Her ancestry is English, Welsh and Danish.  She has very pale, freckly skin.  When younger, her hair was light, honey-colored blonde, but it is now a bit darker—although not black like an Indian or Hispanic or Asian or African descended person.  And she has brown eyes.

But, she comes from a population where those traits are not unusual—her father has pale blue eyes, and of her four siblings, three do as well, and two have retained blonde hair (naturally so, even) into adulthood.  Our family is the darkest in terms of hair color, eyes color and even to some degree skin color, even though we're absolutely white people, of the extended family.  Almost all of my nephews and nieces are blond(ish) with blue eyes and very pale skin.  So, I'm not suggesting that you need to look like a Nordic Aryan icon to be white, but I am suggesting that whiteness as a very broad phenotype includes those features in the population.  People from Pakistan and northern India may well have an Indo-European language, and R1a-Z93 Y-DNA haplogroup, but the Andronovo impact in their genetics is generally 25% or less; they are not white people.

Southern Europeans and MENAs are a bit trickier.  For purposes of this discussion, however, I'm excluding them, because they tend to be Mediterranean in physical features; darker skin, brown eyes and black hair being most common among them, and where it doesn't appear, it's usually rather easily traced to more recent admixture from farther north in Europe.

Now, I'm not making a value judgment about being white vs being Mediterranean or anything else other than the obvious that it's OK to be white.  Whiteness isn't the source of evil in the world; quite the opposite, in fact.  Whiteness is the foundation of the modern world, because everything "modern" is built on technological and social innovations that come from the white world.  Not only that, the Gospel was restored to a white nation which had been raised up and prepared to receive it and bring it to the rest of the world.  But that doesn't really mean a whole lot; we can hardly rest on the laurels of what someone else did, even if that someone else was part of the same genetic population that we're from, because everyone's responsible for their own salvation, both temporarily and spiritually.  It simply doesn't matter.  Except, of course, that it's interesting, and has been to people for a long time.

All that said, the link above should be a cautionary tale.  We simply don't know, even if we sometimes think we do, where these features came from, or when, or how.  On top of that, even if we could completely trust the results that we have, which we can't, we also have to be mindful of the fact that our samples are pretty scanty, and subject to a great deal of potential sampling bias.  It would be reckless in the extreme to suggest that all Western Hunter Gatherers (WHGs) were dark skinned and blue eyed because our untrustworthy data from Loschbour man suggests that he was, or that the Yamnaya were "somewhat" dark of hair, eye and skin, because the sampling of a handful of kurgan burials suggest as much for those individuals.  And yet, exactly those kinds of definitive declarations are common in the discussions for this particular question.

With those major caveats out of the way, what does the data tell us?  Unfortunately, it is often contradictory.  Here's some of the common declarations I've seen frequently:

• WHGs were dark-skinned with dark (although not necessarily black) hair and blue eyes, based on La Braña Man, Cheddar Man and Loschbour Man.
• That said, the population today with the highest amount of WHG DNA is on the Baltic Coast, which are among the whitest people on earth—its where the highest proportion of pale skin, blond hair and blue eyes occur.
• Early European Farmers (EEF) comes from a population where blue eyes may well have been found.  In Anatolia before they even spread into Europe, four out of 37 sampled individuals may have had blue eyes.
• That said, every population today with elevated EEF ancestry is Mediterranean in phenotype, and most likely they inherited that phenotype from their EEF ancestors.
• (That said, almost every European population has at least some level of EEF ancestry, and the EEF ancestry is probably itself "adulterated" with varying degrees of HG ancestry, varying quite a bit by locale.)
• The Globular Amphora Culture was once seen as the "whitest" prehistoric population and a likely source of "whiteness" in subsequent populations due to a few samples (5 of 6 were supposedly blond and blue-eyed.)
• However, the recently uncovered GAC mass grave in Zlota, Poland, has them as darker in all three features.  This has thrown a lot of the binary thinkers into a tizzy, because they aren't able to think with any nuance, and they weren't able to see the weakness of the data used to reach the conclusions that they had reached.  Given that GAC was about 60% EEF and 40% WHG (or maybe 70/30, depending on the report), that's not really surprising, though.
• The Eastern Hunter Gatherers (EHG) were supposed to have pale skin, but darker hair and brown eyes. They represent a hunter gatherer cline with the WHG population, though, although isolation by distance applies to them.  Much of their ancestry was Ancient North Eurasian (ANE) and the ANE's were supposed to carry the alleles for all three aspects of "whiteness" without being white themselves (in fact, their closest phenotypical descendants would be American Indians and some of the Siberian peoples.)
• Caucasian Hunter Gatherers (CHG) were supposedly relatively dark, which is probably an OK conclusion given that the Caucasus region today is peopled by "darker white" people.
• A handful of Yamnaya studies suggest "darker white" for the Yamnaya people too—brown eyes, dark hair and darker skin than the pale-skinned northern Europeans.  That said, their immediate successors, the Corded Ware Culture, is supposed to be "quite pale" in spite of the fact that they were "overwhelmingly Yamnaya-like" at a genome level.
• The prevailing opinion seems to be that whiteness actually came from the WHG/EEF mixture, and the Indo-Europeans only picked it up by taking their women.  However, not only do the most recent genetic studies suggest that at least in some areas, the locals didn't contribute much DNA to the subsequent population (this is especially true in the actual Corded Ware territory) but as the most recent paper I mentioned above about the GAC shows, the assumption that they were the source of all that "whiteness" is probably flawed to begin with.

How is one to interpret that big mess of contradictory and confusing data?  My approach is more or less the following: parsimonious null hypothesis would suggest looking at modern populations, interpreting their "mix" of ancient populations, and working backwards from there.  This is especially important when the DNA for ancient populations' physical traits interpretation can't be trusted, as we've seen with the Cheddar Man situation and the probable sampling bias that we have with the data that we do have.  The best scenario is when ancient DNA markers and subsequent physical traits are aligned with each other.  When they are in conflict, the subsequent observed traits trump, I believe, the fundamentally untrustworthy DNA data.  I'm also not a fan of trying to invoke very rapid mutation or genetic drift as... well, not impossible, but hardly parsimonious.  I'm also not very fond of just-so stories about vitamin D and sunlight, for the simple reasons that 1) there are other populations that thrive just fine in polar regions without having become pale, and 2) it wouldn't have made any difference anyway unless they were running around naked so that their skin would be exposed to the sun, which doesn't make any sense in a polar or near-polar region.  Too much cold weather!

So, all that said, what do I think is the likely interpretation of various ancient populations?  What I think likely is that the WHG probably had pretty white skin and maybe a high preponderance of blue eyes.  The EHG probably had white skin and a preponderance of blond hair.  All three might have been traits common to both groups.  The EEF and CHG were probably somewhat Mediterranean in appearance.  Subsequent populations probably had similar traits, depending on their level of admixture.  So, I can accept that the Yamnaya probably weren't classically Nordic—they're classically a mix of EHG and CHG with some EEF (and slight WHG) admixture on the western fringe, although genes for light (especially reddish) hair and blue eyes and paler skin were probably not uncommon in the population.  The slight admixture of populations like Narva and GAC and TRB, which were mixed hunter gatherer with relatively slight EEF admixture (compared to cultures further south) probably boosted the signal so that the Corded Ware peoples came out Nordic.  Where later cultures derived from CWC imposed themselves on areas further south, like the BMAC territory and the Swat valley, and even southern Europe, the subsequent populations came out eventually considerably less Nordic.  But even where Yamnaya directly came into contact with EEF, as in the Balkans and later Greece, as well as Italy and Iberia, etc. the physical type was described (of at least the aristocratic upper class) as more closely related to Nordic phenotypes than Mediterranean phenotypes.  It seems, in fact, that red hair was surprisingly common, although it's been largely "bred out" except in places where it complemented the boosted signal from the natives (Scandinavia and the Baltic area) or the natives contributed relatively little to the genome (the British Isles.)  The Greeks and Romans have a shockingly high number of aristocratic personalities and families that are described as red, the Thracians were described as red-headed, the Chinese described many of their neighboring steppe barbarians like the Wusun as red-headed, etc.  And on and on.

(As an aside, I think it's unusual and maybe significant, that reddishness can be common among people that don't have red hair.  It's pretty common knowledge that beards are often much more red than hair.  Even my one blond and blue-eyed son has very red stubble when he hasn't shaved in a while.  I also think it's interesting that the three "white" traits do seem to bundle themselves together.  That same son of mine is the only one of my kids with blue eyes (out of four.) He's also the only one who retains blond hair into adolescence, and he also seems to have the palest skin of the family and has the hardest time tanning.  Now, granted, he's hardly a representative sample, but I think that that's not an unusual situation too that the three traits tend to boost each other's signal when combined, and that the alleles for each probably influence the alleles for the rest in some fashion.)

Now, all that said, I'm not dogmatic about that.  But I do think that it is a compelling null hypothesis, backed up by compelling (albeit often circumstantial) evidence.  To convince me that this picture is wrong?  I'd need to have better confidence in the DNA interpretation, which will probably come in time, and we'd need more samples.  Probably a lot more samples.  Our sampling is just a tiny sprinkling today.