Jews aren't really from there. Ashkenazi Jews are European converts, descended from the Khazars, and the DNA proves it.

The fuller answer

Population genetics answered this one in public, in journals with no stake in the fight. Every Jewish community anybody has tested carries a Middle Eastern component: Ashkenazi, Sephardi, Mizrahi, all of them. Ashkenazi Jews come out closer to other Jewish communities than to the Poles and Germans they lived beside for centuries. You can hold this ground with an evening and a search engine. That is unusual on this subject, and it is why the argument is worth having rather than swerving.

Two entries here do different halves of the job, and reaching for the wrong one wastes the exchange. This page is about descent, and it argues from genomes. The ancient connection is about the land, and it argues from prayer books, Ottoman tax registers and archaeology. If the sentence coming at you is that the attachment to the place was manufactured in the nineteenth century, go there. If it is that the people came from somewhere else entirely, you are in the right place.

The serious version, and the thing wearing its clothes

Concede the pedigree first. It costs nothing and it changes how the next ten minutes go.

This was not cooked up as propaganda. A historian at the Hebrew University proposed it in 1944. It reached millions through Arthur Koestler, whose The Thirteenth Tribe appeared in 1976, and Koestler’s own motive is the part nobody mentions. His biographers record that he believed proving eastern European Jews were Khazars would remove the racial basis of antisemitism, and so remove antisemitism.

It did the opposite. Academic reviewers took the book apart within months, one of Koestler’s biographers later calling it risible as scholarship. Neo-Nazi magazines greeted it as the political bombshell of the century. The Christian Identity movement adopted it as their primary source and sold it by mail order.

That is the fork. One branch is a genuine question about the ethnogenesis of a population, still argued in journals. The other is the Khazarian mafia meme, which says a hidden people is impersonating another, and which is the replacement conspiracy with different costumes. Work out which one is in front of you before you answer, because the same facts do different jobs against each.

Grant the Khazars, all of them

The khaganate was real. It sat north of the Caucasus from roughly 650 to 969. Its ruling class is widely held to have adopted rabbinic Judaism in the eighth century. Two medieval texts describe that, the Khazar Correspondence and the Schechter Letter.

A Hebrew University historian went back through those sources in 2013 and concluded that the conversion never happened at all. That is a real paper in a real journal and it is a minority position. Do not build on it.

Grant the conversion instead. Grant that Judaism has taken in converts throughout its history, which is a fact about the religion and not a charge against it.

None of it touches the question. Whether a khaganate existed is not in dispute. What is in dispute is whether the Jews of Warsaw and Vilna descend from it, and that is a question about genomes.

What the genomes say

Four studies, and the fourth is the one to lead with in a hostile room.

Y-chromosome work published in 2000 tested 1,371 men from 29 populations, including 336 Jewish men from seven communities. The Jewish communities clustered tightly together, pointing to a common Middle Eastern source population. A 2010 genome-wide study of fourteen Jewish diaspora communities against 69 non-Jewish populations traced the origins of most of them to the Levant. Another that year found each of seven Jewish populations forming its own cluster while sharing Middle Eastern ancestry. Palestinians, Bedouin and Druze came out among the closest non-Jewish relatives.

Then the 2013 study built to answer the Khazar question directly. It assembled 1,774 individuals from 106 populations. Crucially it went and got the samples that had been missing: 261 of them, from fifteen Caucasus and northern populations, none of which had been used in this argument before. The finding was flat. No particular similarity of Ashkenazi Jews to populations from the Caucasus, and their greatest shared ancestry is with other Jews.

Thirty geneticists signed that. It is the answer to the paper, not an opinion about it.

The concession, which is the sharpest thing on this page

Ashkenazi genomes are not overwhelmingly Levantine. Say otherwise and the first person in the room who has read the papers takes you apart.

One careful 2017 estimate puts Ashkenazi ancestry at roughly 53 per cent European against 47 per cent Middle Eastern. Southern Europe is the dominant European source. The maternal side goes further. A 2013 study argued that around 81 per cent of Ashkenazi maternal lineages have a European source, the four major founder lineages included. That is disputed, and it is not resolved.

Say it first. Say it before they do.

Then use it, because that paper is their best source and it does not say what they need. It found almost no maternal ancestry from the North Caucasus, which is precisely what the Khazar hypothesis predicts. And it puts the male line the other way, predominantly Near Eastern. The Y-chromosome work supports that, estimating European contribution to Ashkenazi men at 13 to 23 per cent.

Read the two sides together and a picture falls out that nobody designed as advocacy. Levantine men, European wives, over centuries, in Europe. That is what a diaspora looks like in the data. It is also why thirty-three Jews buried at Erfurt in the 1300s come out as a close match for Ashkenazi Jews today.

Where this backfires. Delivered as a list of percentages it sounds like somebody defending a pedigree, which is a losing posture whatever the numbers say. Two figures, then move on.

Why the Khazar paper does not carry it

The objection is technical and it is short, so you can carry it.

There was no Khazar DNA. No population is known to descend from the Khazars, and no ancient Khazar genome was available to compare against. So the 2013 paper used modern Armenians and Georgians as stand-ins for Khazars, and Palestinians as stand-ins for ancient Israelites. Everything it reports about Khazar ancestry rests on those two substitutions being right.

A later method by the same author drew a separate critique in 2016. It compresses about 100,000 genotypes down to nine components and works from 26 reference populations. On people of mixed ancestry it returns roughly the midpoint between the parental sources rather than an origin. Some of its outputs land in open sea.

Be fair about the man. He published in a peer-reviewed journal, and he has said his interest was medical, improving disease studies that assume a homogeneous population. He is entitled to be answered rather than accused. He was answered.

“You are not real Semites anyway”

Different move, different answer, and the word is ours before it is theirs.

Semitic is a label for a family of languages. It was coined by German scholars in 1781, from the name Shem, and it covers Arabic, Amharic, Tigrinya, Aramaic, Maltese and Hebrew. It has never named a race, a bloodline or a region.

Antisemitism was coined a century later, in Germany, in 1879, by a man who founded a League of Antisemites. Wilhelm Marr picked the new word over the ordinary German Judenhass precisely because it sounded racial and scientific rather than religious. It meant hatred of Jews on the day it was invented. It has never meant anything else.

Now name the move rather than arguing the etymology further. This is a claim that a compound word means the sum of its parts, and it gets applied to exactly one word.

Then step past it, because the word game has arrived in place of an argument. Even if every Ashkenazi Jew were Turkic to the last chromosome, it would say nothing whatever about the Jews of Baghdad, Sana’a, Casablanca and Gondar. The Khazar claim was never about them, and they are the largest bloc of Israeli Jews.

Then ask what the answer is for

You have now won a factual argument. Take one more step, because winning this one on its own terms costs something.

Nobody asks a Frenchman for a genome. This question is put to one people, and answering it politely, every time, concedes the premise underneath it: that ancestry decides who may live where.

So put the burden back. If the test came back the way they want, what happens next? Who leaves? Where do they go, and who moves them?

A position that would dissolve a state on genetic grounds owes a description of what follows, and it is almost never asked for one.

That question protects you from your own evidence as well. A world that hands out countries by genome is not one anybody is campaigning for. Saying so out loud is what stops this page being turned against the thing it defends. The legal and moral case for the state does not run through anybody’s DNA. It runs through why the state exists.

The lines

Take the shape, not the wording.

Somebody went and looked. 1,774 people, 261 of them from the Caucasus. Ashkenazi Jews came out closest to other Jews.

That paper used Armenians and Georgians as stand-ins for Khazars, because nobody knows who the Khazars’ descendants are.

Yes, Ashkenazi genomes are roughly half European. Levantine men, European wives, centuries of it. That’s a diaspora, not a conversion.

The maternal study they quote also found almost nothing from the North Caucasus, which is the one thing the Khazar theory needs.

Grant me the Khazars entirely. Now tell me how that touches the Jews of Baghdad, Sana’a and Casablanca.

If the DNA came back the way you want, what happens next? Who leaves?

Before somebody says this to you

Two sentences with a real day in them.

Action. “On Wednesday evening I will read the abstract of the 2013 Human Biology paper and the abstract of the 2017 PLOS Genetics one, and write three numbers on a card: 1,774 people, 261 Caucasus samples, 53 per cent European.”

Coping. “If somebody quotes the European admixture at me, I will agree with the figure before I answer it, say Levantine men and European wives, and then ask what happens next if they are right.”