A beautifully told story with colorful characters out of epic tradition, a tight and complex plot, and solid pacing. -- Booklist, starred review of On the Razor's Edge

Great writing, vivid scenarios, and thoughtful commentary ... the stories will linger after the last page is turned. -- Publisher's Weekly, on Captive Dreams

Wednesday, September 20, 2017

The Tyranny of Genes

PHYSICS, wrote Nigel Calder in Timescape, is the intrusive science, always butting in. Radiocarbon dating overturned a century or more of carefully built-up chronology in Egyptian king lists and other such things, as estimated by comparing strata in excavations. Consequently, it was resisted and denounced as heretical until it became orthodoxy.

It is possible these days to discover to whom you are genetically related and how closely by comparing the genomes found in various peoples. A consequence of this is that longstanding beliefs can be overturned. Two have recently come to TOF's attention, courtesy of Nassim Nicholas Taleb and Pierre Zalloua.

Now, TOF has always considered it unlikely that low density barbarian groups could replace high density civilizations. Barbarian invaders might impose their languages and customs, but not their genes. Hence, the Palestinian Arabs of today would be mainly descended from the same people who lived there before the arabs came huffing and puffing out of the desert: Basically Canaanites (with an admixture of Jews, Greeks, and Romans) who have perforce learned Arabic and pray facing Mecca. Likewise, most Turks today are actually Greeks and Cappadocians, rather than genetically Turkish. The cattle herding nomads never outnumbered the settled and civilized farmers of Anatolia. However, it is quite natural for a defeated people to imitate the ways of their conquerors. So even though France is genetically Celtic, it is linguistically Roman, despite invading Franks and other Germans. (e.g., Sowell, Conquests and Culture)

So genetic maps do not always match closely with linguistic maps. Southern China is an example. When we trace the progress of Indo-Europeans up the Danube Valley into the heart of Basque country, we may be seeing not the progress of a people but the successive adoption of iron implements and farming by aboriginal peoples who remained stubbornly in place.
Hence, the core of Europe was Celtic from Germany to Iberia and the Isles, but with a sizable admixture of Italics in the south and Germanics in the east. Some migrations are sizable and do leave genetic footprints. See Cavalli-Sforza, The Great Human Diasporas, pp. 149 and 153-157.
Taleb and Zalloua take note of two things in particular:

1. Ancient Greeks

Ancient Greeks were genetically as close to Ancient Levantines as they were to the so called “Aryans”. It just happened that Ancient Greeks spoke an Indo-European language while Ancient Levantines spoke a Semitic one. Ancient Greeks, it turns out, have an Anatolian origin (and Levantines are mostly Anatolian and Iranian). This was shown in recent paper, Lazaridis et al. (2017) who looked at Bronze Age and modern Greeks. And modern Greeks descend mostly from Bronze Age Greeks. In other words, if you want to see how [sic] ancient Greeks looked like, go no further than your local diner in Astoria, Queens.
IOW, you cannot claim a connection to Greece and the heritage of Western Civilization and preach hatred against Jews. Not if you want to be consistent. The dark blue in the above map indicates this Anatolian component, here labelled "Greco-Anatolian."

One is reminded of Colin Renfrew's "Anatolian Hypothesis" that the Indo-European migrations originated in Anatolia and not in the Ukraine. From Sumer-Elam he saw the Dravidian languages spreading east into Indus Valley and south into India; and Semitic languages spreading south into the Levant and then to Egypt and Arabia. Later IE migrations spread into Armenia, Iran, and northern India.

 Note: BTW, "aryan" was a social class, not a people. They were the descendants of the chariot lords who rolled out from the steppes into India and Europe; hence, Indo-European. (In Europe, the conquered indigenes were white ancestors to Basques; in India, they were dark ancestors of Dravidians.) In Iran, the word was spelled "aryan." In India, "arya," In Irish, it is "aire" (the y-glide is still there in the pronunciation). Irish recognized several grades of aire. The lowest was bo-aire, or "cow-lord," to qualify for which required that one own one cumal (= 3 cows). So if you don't own at least three cows, you cannot be a noble Aryan. In Greek, άρχων (archon) means a lord. In German, the word is "Ehre," which has shifted to mean "honor," which belonged only to nobles. Two countries still have names that mean "land of the aryans", viz., Erin and Iran.
The paper is more complicated than that. The focus was on comparing Minoans and Mycenaeans against folks from southwest Anatolia, a Neolithic sample, 332 other ancient individuals, and 2,614 present-day humans, including modern Greeks, Cretans, et al. In broad strokes, the study shows that there was genetic continuity in the Aegean from the time of the first farmers to present-day Greece, but not in isolation.

Reference. Lazaridis, Iosif, Alissa Mittnik, Nick Patterson, Swapan Mallick, Nadin Rohland, Saskia Pfrengle, Anja Furtwängler et al. “Genetic origins of the Minoans and Mycenaeans.” Nature (2017).

2. Lebanese are Not Arabs

A recent paper showed that modern Lebanese are genetically very close to Ancient Canaanites. Further, in spite of claims in history books (e.g. Salibi), Lebanon experienced hardly any gene flow from the Arabian Peninsula. The presence of J1e haplogroup is less than 3%, a tenth of what is found in Syrians and Jordanians, and a fifth of what is found in Palestinians. This applies to all Lebanese ethnoreligious groups.
Further, the [genetic] distance between today’s Lebanese and modern Greeks is minimal, much closer than that between them and nonLevantine members of the Arab League.
Lebanese nationalist have always claimed that they are Phoenicians, not Arabs, and this would seem to be some evidence for that.

The writers of the paper tell us that they sequenced five sets of remains from the Canaanite city of Sidon, ca. 3700 years old and compared them to modern inhabitants of modern Lebanon. They found that a Bronze Age Canaanite-related ancestry was widespread in the region, shared among urban populations inhabiting the coast (Sidon) and inland populations (Jordan) who likely lived in farming societies or were pastoral nomads. This Canaanite-related ancestry derived from mixture between local Neolithic populations and eastern migrants genetically related to Chalcolithic Iranians. They estimated, using linkage-disequilibrium decay patterns, that this admixture occurred 6,600–3,550 years ago, coinciding with recorded massive population movements in Mesopotamia during the mid-Holocene. They concluded that present-day Lebanese derive most of their ancestry from a Canaanite-related population, which therefore implies substantial genetic continuity in the Levant since at least the Bronze Age. In addition, they found Eurasian ancestry in the modern Lebanese not present in Bronze Age or earlier Levantines, and concluded that this Eurasian ancestry arrived in the Levant around 3,750–2,170 years ago during a period of successive conquests by distant populations.

That would be around 1730 BC to 150 BC, from when the Hyksos chariot lords took Egypt and the Indo-Iranian chariot lords were pouring into Persia and the Indus Valley to when Alexander and his Seleucid successors conquered the area.

The upshot is that Greeks are much closer genetically to Levantines than they are to Nordics, even though the Greeks speak an Indo-European language and not a Semitic one. 

Marc Haber, etc, 2017. "Continuity and admixture in the last five millennia of Levantine history from ancient Canaanite and present-day Lebanese genome sequences." The American Journal of Human Genetics.


  1. Oooh, that's cool-- did they get a decent number of genetic samples to compare to? Since you're sharing it, I'd assume so, but there might not be enough to be a really good number so worth asking-- I know it says over three hundred ancient, but my grasp of what's a good sample over a period of time like that is...hazy.

    The Lebanon thing is also interesting.

    On an aside: Oh, thank goodness, you went a totally different direction than I feared-- you looked both at *populations*, and at *historic populations*, in contrast with the modern individual tests where people declare "I thought I was X, but my test says I'm Y, based on what area has a modern population most closely matching my genes!" *tiny type, much bigger on the website, explains that it's basically a "who is most like you right now based on our samples", but that's not what the ads imply*

    (I was really not looking forward to trying to defend that based largely on the example of "who are you more closely related to, your mother or your sibling?" Answer, for those who didn't have the same text books I got-- it depends, because you have half your mother's DNA, but your sibling can be anything from identical to no shared markers. So unless both parents are picture perfect examples of the locations they've got a test for, and first generation... you have a mess.)

  2. IIRC, most Europeans aren't actually related to Basques; the Basques and Sardinians have a number of weird genetic markers not shared with the rest of Europe. I think the people the Indo-Europeans conquered would be "Celts", with the understanding that that's not a linguistic classifier yet. Because Celts are as close to a "generic European" as you're going to get. (Except in northern Norther Europe, where you're probably dealing with the ancestors of people like the Finns.)

    The most interesting and/or amusing thing, to me, is that there is no such thing as "black"; if two African groups didn't share territory, they're likely no more related to each other than to Swedes. With the relatively recent overlay of the Niger-Congo peoples (especially the Bantus) in the last 3000 years, admittedly.

    If you want to piss off white supremacists and black nationalists simultaneously, point out that nobody in Africa—not cities like Great Zimbabwe, not the vicious militias of the Congo—has anything to do with black Americans, who are from the Senegambia and adjoining regions (okay Timbuktu might, since it's in Mali and some of the "black diaspora" came from there).

    Black Americans voting for Barack Obama—whose ancestors came from the opposite side of Africa from theirs—makes about as much sense as people from Hyderabad voting for an Irishman as "one of us".

    1. A good point, but it mostly indicates that "African American" isn't about slavery, it's about being visibly black in the US and identifying as such.*

      Sort of like how Pelosi is a "Catholic Politician," or (according to my Mormon friends) Reid is a "Mormon Politician," but they don't have much to do with the actual faith.

      It's sort of like how my grandmother had a picture of JFK, the Pope and Jesus on a wall-- and I couldn't say who was at Jesus' right hand.
      It doesn't quite fit, because all the information on JFK not being a very good Catholic seems to be fairly modern in folks knowing about it and the news back then was very scared because he was so Catholic, but it's a comparison.

      * thinking about it, this would mean that "racism" in the US isn't about slaver, it's about rather nasty Democrats forcing everyone to conform to their biases, IE Jim Crow. I don't think they checked to see where you were from if you looked Black back then?

    2. On the other hand, though, if Barack Obama was "representin'", he'd sound like Thurston Howell III. He was raised by old-money WASPs, and that's what he really is; if you've ever seen his mother's high-school yearbook picture, she looked like H. P. Lovecraft in Stepford Wife crossplay.

  3. WRT #2--So, does this mean the Phalange were right after all?

    (BTW--do they even still exist?...)

    1. They were right that the Lebanese are not Arabs; they weren't right that, for example, their language is not Arabic. Many Lebanese nationalists insist Lebanese Arabic is actually Phoenician, but that would look more like Neo-Aramaic, which, like Phoenician, is Northwest Semitic; what they speak in Lebanon is definitely Arabic, albeit with a strong Northwest Semitic substrate. (It would be like if a French nationalist took nos ancêtres les gaulois so literally as to claim French was a Celtic language. Celtic and Italic are roughly as close as Northwest Semitic and Arabic, but nevertheless French is demonstrably an Italic, specifically Romance, language, not a Celtic one, even though it has, IIRC, the strongest Celtic substrate of any major Romance language.)

      Of course, "Arabic" is mostly a legal fiction; calling what they speak in the Levant, Egypt, and Arabia the same language, is like calling the French, Spanish, and Italian of the Early Modern period "Latin". They generally have to switch to Modern Literary or Classical/Quranic Arabic to be understood by each other, much like Early Modern Europeans switching to Humanist or Church Latin.

  4. I can't decide which part of G. K. Chesterton's essay "On Celts and Celtophiles" to quote here, so I'll just recommend the whole thing. It's Chapter 13 of "Heretics", pretty short, and well worth the read.

  5. "has anything to do with black Americans, who are from the Senegambia and adjoining regions"

    This isn't entirely true. Some slaves came from Portuguese East Africa, especially after the British Navy shut down the slave trade from West Africa. I'm a black American, but my Y DNA (paternal line) originated in southern Mozambique.

    1. Good point, I hadn't considered that. Though the descendants of Portuguese slaves presumably make up a fairly small percentage of black Americans. (Then again, the Navajo word for "black person" literally means "dark Mexican"—that's where the first black people they met were from—so even with a population like black Americans, who are mostly descended from only a few groups, it's risky to over-generalize.)


Whoa, What's This?