I sort of "kept my powder dry" on this thread because I didn't get a chance to read the whole paper and supplement until this morning. For personal reasons my time for this hobby is much more limited than it used to be. I must say that while there are things in the Limes paper which I think are problematical, especially as regards Albanians, given the disjunct with the yDna and their presence in the CNE cluster rather than the Kuline cluster, that paper looks like a masterpiece compared to this one, I'm sorry to say. The actual authors are second string, but I'm disappointed in Johannes Krause's supervision.
I think Leopoldo has made some very pertinent points, among which are the two immediately upthread.
"
1) the authors didn't communicate with the team that worked on the Danubian limes paper, and that's why they've written the already "debunked" line that "it was thanks to the Roman empire that southeastern Europeans came to occupy the place in the PCA between the rest of Europe and the near east" since the balkan cline already existed.
2)they didn't use the already published samples from Antonio et al 2019 (especially to integrate their very small dataset of imperial samples, also especially after verifying that Etruria and Latium were very similar both pre and post the imperial period)."
The first point, in particular, calls their whole analysis into question.
Jovialis' point about how it seems odd that the authors just make a blanket statement that there was no Iran Neo in mainland Southern Italy until the Imperial Age when they don't use any ancient samples from that area, ignoring samples which do, in fact, show such admixture is very well taken.
I also find it odd that the paper says the following: "As a result, the models that were found to fit the data best are those with a 38 to 59% contribution from Levantine
or Anatolian populations into the local/preexisting C.Italy_Etruscan gene pool (
Fig. 4B and table S4D).", yet then proceeds to show a graphic with only ancestry from the southern Levant. Did they, like the authors of the Limes paper, attempt to model the admixture with ancient Anatolian samples? If they did, where is it in the paper or Supplement?
There also has to be a degree of common sense when one analyzes data like this. Is it at all conceivable, to reword Leopoldo's other point about the conclusions here, that "slaves, soldiers, and mobile citizens" ONLY from the Levant went to the former Etruscan lands and all the Anatolian ones went to the Balkans? It doesn't make any sense. It also flies in the face of contemporary Imperial documents pointing to the large number of Greeks in Etruria. What, did all the Germanic and Gallic slaves get sent to latifundia and galleys and mines and died off, and what we are left with are the more educated people from the east? O.K. I'll buy that, as it fits with what the Limes paper shows, but again, ONLY from the Levant?
That's not to mention that they're talking about 6 samples for a 500 year sample. Why didn't they at least use the Antonio et al samples if they were going to draw conclusions about all of Italy? They do point out that burial practices might figure into the representativeness of those 6 samples, but then drop it there.
Speaking of this, where is the archaeological context for these samples, like what was provided in the Limes paper?
I'm intrigued by Archetypes' point that perhaps the Anatolian samples of the last millennium B.C. leaned more toward the Levant than prior Anatolian samples. Perhaps we'll discover that's true, but enough to make the admixture show almost 50% Southern Levant admixture?
I am really disappointed that the use of yDna seems to have gone out of fashion with the major genetics labs, to one degree or another. The weakest part of the Limes paper, to me, was the disjunct with the Albanian analysis and the yDna of Albania. How does their yDna support a 48% admixture rate with Slavs? If some E-V13 came with Germanics, ok, but Slavs?
Likewise here the yDna doesn't fit with the conclusions. To Ailchu's point, where are the Celts in all of this? There's only 5-10% of Germanic yDna in Toscana based on Maciamo's charts and graphic. That seems to fit the historical evidence very well; I mean, we're talking about a total of 60,000 Langobards, people, and a Gothic elite presence. The majority of the Langobard castles were in the northeast, and the strategically placed castles further into the peninsula could not and the records show did not have large numbers of actual Langobards manning them. Then, take into account the paucity of I1 and U-106.
In the Limes paper they explained this "problem" away by saying that the admixture was mostly female mediated, as also shown by their analysis of the X chromosome. That one leaves me scratching my head. Did the male lines die out because they were doing the fighting? Did the female Slavs, and Langobards in the case of this paper, find the locals irresistible? If the latter it would be a departure from the scenario in all cases of invasion and subjugation in which the arriving males mate with the local females, but their own females are sacrosanct. Typical male behavior, in other words.
All in all, I'm quite confused by this paper. It makes no difference to me whatsoever if Tuscans, who constitute at least 1/4 of my ancestry, or maybe if you take into account that eastern Liguria may be very like Toscana around Florence, 1/2 of my ancestry, are Anatolian or Levantine or Balkan or whatever. However, it has to make sense.
