I’m looking at abstracts on Ashkenazi genetics from ASHG 2013 and SMBE 2014 – by the same group, with Shai Carmi as the lead author. They did 128 whole genomes, 50x deep.
They concluded Ashkenazi Jews were about 50% Middle Eastern and 50% European. In the 2013 abstract, they were pretty specific: they estimated the European ancestry fraction at 55% , plus or minus 2%. ( In our book, we had a crude estimate of about 40% European ancestry.) They estimated the split between Europeans and Middle Easterners at about 9000 BC: which sounds about the right date for the entry of the Sardinian-like farmers. From other data (mtDNA) , and from the fact that you see almost zero WHG or ANE in Ashkenazi autosomal genes, one can conclude that the European admixture was mostly Italian, with some southern French. Very little German or Slavic – by that time serious endogamy had set in..
By looking at IBD segments, they conclude that there was indeed a bottleneck in Ashkenazi ancestry, ~350 individuals, followed by a rapid expansion. IBD analysis should pin this down quite accurately. They estimate that this was about 800 years ago, but I would bet money that it was a little earlier – more like 1100 years ago. In other words, the founding bottleneck, the time when the ancestors of the Ashkenazi Jews moved from Italy into the Rhineland, not a later persecution bottleneck. I don’t think the population ever dropped that low in the persecutions. I say this because of what the history looks like to me, and because of certain things that suggest that our models of recombination may be a bit off. 350 isn’t a terribly tight bottleneck, as long as it doesn’t last long – and it’s hard to see how any bottleneck could generate several unusually deleterious recessives that concentrate in a few metabolic paths. Or make you smart.
Many people looking at Jewish population history have boggled at the idea of a small group expanding to a few million in a thousand years or so, and have come up with various scenarios other than Italy -> Rhineland -> Polish-Lithuanian Commonwealth model, such as Koestler’s Khazar theory or Wexler’s idea that the Ashkenazi Jews came from the East. Koestler and Wexler are both wrong, by the way – the genetic evidence is quite clear.
There was never anything particularly particularly improbable about the Ashkenazi population expansion. Moderate prosperity, which the Ashkenazim had for most of their sojourn in Europe, easily allows a family to average 3 surviving kids. Given that rate of growth, a population increases by a factor of more than one million in 35 generations. Ask the French Canadians, or the Puritans, or the Boers.
But since anyone who can’t understand this by now never will, I look forward to the first revisionist history of the Amish. Do you really believe that ~200 Mennonites landing in Pennsylvania in the 1700s could number a quarter of a million by 2010? Doesn’t there have to be a deeper, more subtle explanation? Where do you think Martin Bormann ended up, huh?