For some time, we have known that many Jewish populations had mostly-Near Eastern paternal ancestry (looking at y-chromosomes) and mostly-local maternal ancestry looking at mtDNA). Autosomal admixture studies generally agree. This is easiest to see when the host population is fairly distant from Europe or the Near East, and thus has significantly different mtDNA types: it’s obvious in the case of Indian Jews. Roughly speaking, Jewish men settled distant lands, as traders or sometimes refugees and POWs. They married local girls, and later, mostly with the advent of Rabbinical Judaism, rules emerged that forbade further intermarriage – and presto, Roberta’s your aunt.
It’s a bit more difficult when comparing Europe and the Near East, since there has been a lot of population movement between those regions, most of it from the Near East into Europe in the form of the first farmers. So even though the mixed origin of Jewish populations (Near Eastern men and local women) was clear in a number of cases, it wasn’t so clear in the most important case, the Ashkenazi Jews, who make up most of the world’s Jews and and account for almost all Jewish intellectual accomplishment.
But even when the same mtDNA haplotypes are found in both Europe and the Near East, the sub-haplotypes are different – the fine details clarify the story.
Back in 2006, Doron Behar and company looked at Ashkenazi mtDNA. Four mtDNA lineages accounted for almost half: K1a1b1a, K1a9, K2a2a, and N1b. About 20% of Ashkenazi Jews have K1a1b1a mtDNA. Behar concluded that all of these lineages originated in the Near East. This was plausible for N1b (about 9% of Ashkenazi mtDNA), which is common in the Near East and rare in Europe (although it was common back in the LBK culture). He couldn’t find any closely related versions of the K1a9 and K2a2a lineages outside of the Ashkenazim – and went on to say that they probably originated in the Near East, based on nothing. He also concluded that K1a1b1a was probably Near Eastern, since the only close non-Jewish versions were found in Portugal, Italy, France, Morocco, and Tunisia: a conclusion which flew in the face of what evidence he had. It is if one knew that all the languages closely related to Russian (Polish, Ukrainian, Serbian, etc) were found in Eastern Europe, and then concluded that the Russian language must therefore have originated in South Africa.
In other words, Doron Behar is a liar. I was going to include something about the probable origins of Ashkenazi mtDNA (mostly Italian) and Behar’s follies in the book. I wrote it up (in a little essay titled “Special K”), but space prohibited, and anyhow liars are boring.
A new paper by Maria Costa et al (with Martin Richards as senior author) settles the issue. We have a lot more data now – more people, and more detail. Turns out that all of those major Ashkenazi mtDNA lineages originated in Western Europe – even N1b, fairly rare in Europe. The majority of the less common Ashkenazi mtDNA lineages also originated in Europe – probably mostly in Italy. Altogether, > 80% of Ashkenazi maternal ancestry is European – mostly Italian, but a bit from France and Germany as well.
You may have heard of Arthur Koestler’s Khazar hypothesis – the notion that Europeans Jews are largely descended from Khazar converts. It’s not true – these results show that it is impossible. Charles Murray suggested that selection leading to higher intelligence in Jews occurred a long time ago, as far back as the Babylonian Captivity. That’s not true either. It never made any sense, because there’s not a scintilla of evidence that Jews in Classical times were smarter than the average bear – but the Ashkenazim being half Italian crushed it yet again. If ancestral Jews had the genetic IQ magic, the Ashkenazim should be watered-down, closer to the European norm: but they’re not.
Lots of European admixture does not contradict our model of the evolution of superior Ashkenazi intelligence, because we think that the relevant selection occurred well after that admixture, during a period in which inward gene flow among the Ashkenazim was very low – as evidence by the fact that this study found plenty of Italian mtDNA, but little from Eastern Europe.
As Michael Balter cheerfully points out in Science, this result may be a bit troublesome to those that believe that Jewish identity descends through the female line. In that case, most Ashkenazi Jews – aren’t.
I haven’t heard anyone else mention this, but logically, someone who is Ashkenazi could now decide that he and his cousins are really the true heirs of the Roman Empire, rather than a member of the Chosen People. I’m sure that wouldn’t cause any trouble.
Doe this mean that the Palestinians have a better genealogical claim to the land of Israel than the Ashkenazi Jews? Maybe – but over the years, they’ve mixed too. They have a lot of South Arabian and African ancestry that wasn’t there 2000 years ago. That’s true of much of the Middle East – but that’s another post… I’m sure that modern DNA technology will answer this question anytime anyone cares to look, and obviously everyone will accept the verdict of Science, whatever it may be.
Anyhow, if Italy really is the Ashkenazi urheimat, that’s not so bad. I’d trade the Judean Hills for Tuscany in a New York minute. And even if trading homelands would require some toe-to-toe combat with the Italians – how hard would that be, really?