search this blog

Thursday, May 25, 2017

Anywhere but the steppe


Last week Scientific Reports put out a paper by Sarno et al. on the population history of Sicily and South Italy. I didn't blog about it at the time because I felt that it was generally a weak effort and not worth advertising. But people keep bringing it up in the comments section, so here goes.

If you download the PDF and do a search for "Africa", you'll see that the only time it comes up is in the bibliography. "Maghreb" doesn't come up at all.

Can anyone explain this? I can't. If you're doing a paper on the population history of Sicily and South Italy and you don't take a close look at the fairly recent North African admixture there, then at best you're naive and confused.

Also, the authors try to enter the Proto-Indo-European (PIE) homeland debate. They basically argue that Indo-European (IE) languages could not have arrived in Southeastern Europe from the Pontic-Caspian steppe because modern-day Southeastern Europeans overall don't pack much Bronze Age steppe admixture. They also claim that based on their admixture dating efforts (which may or may not be accurate) the steppe ancestry by and large arrived in the east Mediterranean during the early Middle Ages with Slavic migrations. Thus, they suggest that a better PIE homeland alternative to the Pontic-Caspian steppe might be West Asia.

These are very weak arguments for a number of important reasons. For instance, language change can happen without massive migrations from afar. Case in point: the Etruscans were a sizable non-IE speaking population in Southeastern Europe until historic times, and discarded their Etruscan language in favor of the IE Latin by being subsumed into the Roman Empire. Indeed, Southeastern Europe has been a bit of a hotspot for this type of thing; Razib has a little more on that and the admixture dating here.

Also worth positing is the likely scenario in which much of the Bronze Age steppe ancestry in Southeastern Europe has been diluted by more recent admixture from the Near East and North Africa. It's hard to say for sure to what extent without direct evidence from ancient DNA, but this is something that should have been considered in the paper.

I won't be blogging much from now on about population history papers based on modern-day samples, because such papers aren't usually worth blogging about.

Reference...

Sarno et al., Ancient and recent admixture layers in Sicily and Southern Italy trace multiple migration routes along the Mediterranean, Scientific Reports 7, Article number: 1984, (2017), doi:10.1038/s41598-017-01802-4

See also...

Late PIE ground zero now obvious; location of PIE homeland still uncertain, but...

102 comments:

Vincent said...

As I posted in another comments section, North African admixture is basically only in Sicily (from Arab occupation) and less than 5%, so why should it come up in a paper about significant CHG admixture all over SE Europe that potentially relates to the PIE homeland?

There's been no recent MENA admixture anywhere in Italy that would be enough to dilute any earlier ancestry from the Bronze Age. Your friend Razib said so himself. Same for South Balkan people - who cluster with Italians.

Davidski said...

@Vincent

As I posted in another comments section, North African admixture is basically only in Sicily (from Arab occupation) and less than 5%, so why should it come up in a paper about significant CHG admixture all over SE Europe that potentially relates to the PIE homeland?

North African admixture is the most significant exotic (i.e. non-East Med) admixture in Sicily.

Why shouldn't it come up in a paper titled Ancient and recent admixture layers in Sicily and Southern Italy trace multiple migration routes along the Mediterranean?

Seems like an obvious thing to look at to me.

There's been no recent MENA admixture anywhere in Italy that would be enough to dilute any earlier ancestry from the Bronze Age.

There's quite a bit. Where do you think all the non-steppe derived CHG admixture came from in Italy? It wasn't there during the Bronze Age.

It just depends how you define MENA. A lot of the CHG came from the Roman Empire and parts of Greece that are now Turkey, and therefore in modern parlance MENA.

Anthro Survey said...

@Vincent

"And if you look at how they relate to the Neolithic samples, they exhibit a lot of shift on the plane toward the steppe populations, parallel to the Levantines. In other words, a lot of the change since the Neolithic in Southern Italy is attributable to the influence of the steppe migration, not Roman era gene flow from Syrians."

Razib is incorrect here. Southern Italians are shifted quite PERPENDICULARLY to the Levant, if anything. He's prolly looking at a wrong set of dots(perhaps tuscans or north italians).

That South Italy has significant post-Neolithic MENA admixture is by now well established in the pop gen community----time for people in italanthro/italicroots to grow up and get over it and to stop bashing Iberia at every opportunity.

Their statistically sig shift in the PCA alone cannot be explained without some extra population that was more BE than anatolian farmers, not to mention all of the calculator and formal stats runs over the years.

Anthro Survey said...

@Davidski:

I think the aggrrgate population bringing all the extra CHG there is better modeled as a Syro-Levantine one, not as Caucasus-Anatolian. I.E.a ton of Natufian-like ancestry was brought along in the "package".

Davidski said...

@Anthro Survey

I think the aggrrgate population bringing all the extra CHG there is better modeled as a Syro-Levantine one, not as Caucasus-Anatolian. I.E.a ton of Natufian-like ancestry was brought along in the "package".

There wasn't just one big wave of this; it came in dribs and drabs over a long period of time with different people, and that's why linking it to PIE is total unadulterated bullshit.

Rob said...

@ Anthro

I would agree with Vince

Italian_Tuscan
"Hungary_N:I1496" 31.05
"Baalberge_MN:I0560" 26.9
"Armenia_EBA:I1635" 23.15
"Yamnaya_Samara:I0231" 19.2

Italian_WestSicilian
"Remedello_BA:RISE489" 32.3
"Armenia_EBA:I1635" 30.2
"Jordan_EBA:I1706" 12.05
"Barcin_Neolithic:I1099" 11.1

Italians are mostly MNE.
"MENA" is only 5 - 10% in south Italy.
The rest is Caucasus related stuff, which also arrived in the Balkans, with PIEs

Anthro Survey said...

@Davidski

I agree. But if we could find a center of mass of all these dibs/dabs, it'd somewhere around Hatay, Cyprus, Aleppo, etc. populations.

Plus, from a purely linguistic point of view, to attribute a West Asian migration to the emergence of Italic is tantamount to going against the laws of thermodynamics for a chemist. lol I don't know what the consensus is on Greek, but every noob and their dog knows Italic shares an immediate node with Celtic and Germanic. I literally facepalmed so hard at that sentence of theirs.

Davidski said...

@Rob

The rest is Caucasus related stuff, which also arrived in the Balkans, with PIEs.

No, it's very late admixture mostly from the eastern Roman Empire.

Martin Clifford Styan said...

It has been clear for some time that in the Neolithic, Anatolia, the Balkans and Southern Italy were inhabited by people with Early Anatolian Farmer autosomal DNA and mainly Hg G Y-chromosomes. The modern populations have more CHG and haplogroups E and J2. This shows that there was a lot of migration into these regions from the east after the Neolithic. Some people are inclined to associate this with the spread of Indo-European, while others think this view is "total unadulterated bullshit". Whichever group is right, the migration or migrations from the east undoubtedly happened and presumably played a very important part in the development of these regions that are so important to the history of ancient civilizations.
I hope ancient DNA will soon be published and will tell us the chronology of the migration from the east, and whether there was a big wave of migration at any time, or only "dribs and drabs over a long period of time".

Rob said...

@ Dave
Possibly. I myself doubt that the odd merchant or gladiator from Roman Armenia would have such a profound impact in an Italy with million square of people.
Do you want to rethink your "theory"?

Anthro Survey said...

@Rob

Italy is far from a monolith: It's not really meaningful to consider it as a collective unit. Yes, the primary ancestral layer everywhere is MNE, but places in the south can and do easily exceed the 10% mark methinks. Particularly so in EAST Sicily and Calabria.

If that is qpAdm, how do the tail probs compare to the one Dave ran some time back?(where sicilians got ~30% Jordan EBA).

André de Vasconcelos said...

It's a matter of labelling, whether you call it 'MENA', 'West Asian' or 'Giacomo' is irrelevant. The CHG and relatively high levels of yDNA J and E are real in Southern Italy and Sicily, and they came somewhen since the EBA.

What I feel is that there is a racist undertone in some people who want to stay away from anything that isn't Europe when it comes to their ancestry, be it thre Middle East, North Africa or whatnot, and see that as a sort of an attack/insult. I was hoping the advances in population genetics would help eliminate this, as we're all so very much mixed, but stupidity will always be stupidity

Vincent said...

Non-steppe CHG came from Caucasus with some IE languages as Sarno says. Magna Graecia admix in South Italy is a fact but doesn't count as "MENA" since Greece is in Europe and it brought the same ancestry (EEF, CHG) and language family (IE) that were already in Italy. No evidence of any actual MENA admixture post-BA, outside of a tiny bit in Sicily.

Razib is correct. South Italians are shifted to the steppe (north from Anatolia and Levant), but more toward the CHG component than the EHG - for the reason Sarno says.

André de Vasconcelos said...

Thanks for emphasising my point

Arza said...

@ Alberto (cont. from last thread)
Though seeing comments about the Mycenaean civilization being an offshoot of Sintashta/Srubnaya or even of CWC, it seems that some people have definitely lost their grip on reality.

Can you elaborate why Mycenaeans could not have come from the north (CWC) or north-east (Sintashta/Srubnaya)?

Rob said...

@ Arza

"an you elaborate why Mycenaeans could not have come from the north (CWC) ?"
Do you suppose that the Myceneans spoke proto-Baltic ? LOL.

Rob said...

@ Anthro

"Italy is far from a monolith: It's not really meaningful to consider it as a collective unit."

That's rather a redundant observation, as no one who has a clue treats Italy as a monolith.
Which is why I presented at least two different regions of Italy, not to mention other testable populations.
Whether the proportions are correct will be seen with aDNA.It doesn't claim to be a definitive figure, but an overall picture.

Arza said...

@ Rob
Do you suppose

I'm asking for arguments and not for what I suppose.

Samuel Andrews said...

@Rob,
"Italians are mostly MNE.
"MENA" is only 5 - 10% in south Italy.
The rest is Caucasus related stuff, which also arrived in the Balkans, with PIEs"

MENA isn't just BA_Jordan. MENA includes everything from Iran to Turkey and Georgia to Saudi Arabia. Lots of people actually include North Africa in the MENA category.

Modern people in SouthWest Asia have a lot more Caucasus-type stuff than Jordan_EBA did. So using Jordan_EBA as the SW Asian or East Med reference doesn't work.

Italian's MENA ancestry(which includes the Caucasus stuff) can be modelled in a lot of different ways because Middle Easterners have varying degrees of CHG/IranNeo and EF/Natufian ancestry. It's just like how the European ancestry of Jews can be modeled in many different ways. It can be modeled as Polish or Italian or French or really anything.

In my experience modelling Italian's MENA ancestry Cyrpiot works best. Real ass Caucasin populations like Georgians or Adygei don't work. There certainly wasn't a direct migration from the Caucasus to Italy. Italians have a lot of Natufian-type stuff which could have only come from SouthWest Asia.

Olympus Mons said...

In this study, as it happened with the one ARAM linked in the previous post on Mtdna... everybody, e v e r y b o d y, runs way from South Ossetians.... It just blows my mind.

Olympus Mons said...

... And how lovely everybody is now interested in CHG without EHG admix in Europe. Well... lets see where it leads us to.

Olympus Mons said...

and @ Davidski!
Lets clear some shit up!

STOP!!
Alberto says "And CHG, we have Kumtepe6, 4600 BC, with CHG admixture. And Kumtepe is quite close to Varna, believe it or not."

Now when I said KUM6 has my shulaveri because it had CHG admix, you flat out called me an Idiot because Kum6 had no CHG. And I accepted. And said, ok, if KUM6 had no CHG admix than could not be my Shulaveri, and so I was being an Idiot.

So, is Alberto an Idiot as well?

Ebizur said...

Samuel Andrews wrote,

"MENA isn't just BA_Jordan. MENA includes everything from Iran to Turkey and Georgia to Saudi Arabia. Lots of people actually include North Africa in the MENA category."

I assume that the part of your comment I have put in bold is intended to be a joke.

I usually make efforts to refrain from trivial fault-finding, but I honestly laughed out loud.

Rob said...

@ Sam

"MENA isn't just BA_Jordan. MENA includes everything from Iran to Turkey and Georgia to Saudi Arabia. Lots of people actually include North Africa in the MENA category."

I was referring to MENA as more recent affinities with modern 'Middle Eastern' populations, encapsulated (not unreasonably, IMO) in my calculation as 'Jordan EBA".
It is important to stratify chronologically & specify geographically the contexts; as they mean very different movements.

Salden said...

https://c1.staticflickr.com/4/3803/33016272762_2f3e139a15_b.jpg

http://biorxiv.org/content/early/2016/06/16/059311

As this graph from Laziridis' work shows, Ashkenazi and Sephardi Jews (who outside of a fringe are accepted as having significant Near Eastern admixture) have an affinity towards South Italay.

Salden said...

For the record, Sephardi aren't included in the chart but they are shown from accepted studies to be close to each-other genomically if not same.

Samuel Andrews said...

@Ebizur,
"I assume that the part of your comment I have put in bold is intended to be a joke."

The average person views Morocco and Algeria and Libya as Middle Eastern. And I'm pretty sure many academics do. "Middle Eastern" has become synonymous with "Muslim world". Afghanistan is viewed as Middle Eastern, even Pakistan is. Hence Afghanistan and Iraq were the theaters of American wars in the Middle East.

André de Vasconcelos said...

Sammy, not wanting to sound like Cpt. Obvious here, but MENA means 'Middle East and North Africa' so the Maghreb is already included in the label as it's in NW Africa.
Personally I don't like the label as it includes peoples with very different histories and backgrounds, but that's what an Eurocentrist view does - getting all the muslims viewed as being roughly the same

http://istizada.com/wp-content/uploads/2016/09/mena-region-countries-list.jpg

http://italicroots.lefora.com said...

Only Eastern European Ashkenazi are close to South Italians. Sephardis and Western European AJs are much more MENA shifted.

http://italicroots.lefora.com said...

People talking about recent MENA DNA in Italy forget that both CHG and Anatolian Neolitich had more Natufian and Basal Eurasian admixtures than any modern Italian population. Moreover we know that CHG farmers invaded the Pontic steppe during the late Neolitich/early Calcholithic, but for some "experts" here all thé non steppe CHG admixture in South East Europe must have arrived during the Roman and Byzantine periods. Don't make me laugh.

Salden said...

I'll wait for you to source that.

Samuel Andrews said...

@Andre,
"Sammy, not wanting to sound like Cpt. Obvious here, but MENA means 'Middle East and North Africa' so the Maghreb is already included in the label as it's in NW Africa"

That's exactly what I said idiot. There's no reason for you to repeat it.

Folker said...

Just a reminder:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4062890/

"With less accuracy, information from HVI sequences also provides a phylogeographic perspective of U6 in Europe (Table 1). The largest U6 Maghreb component in Europe is found in Portugal (69.9%), then in Spain (50.0%) and Italy (53.0%), and decreases sharply in the Eastern Mediterranean (25.0%). No U6b representatives have been detected in Italy, although it is present in Iberia to the west and in the Near East to the east. Regarding the Canarian motif, 33% and 50% of the U6b haplotypes found respectively in mainland Portugal and Spain belong to the Canary Islands autochthonous U6b1a subgroup. Curiously, it has not been detected in the Portuguese island of Azores and Madeira or in Cape Verde either [58]. U6c is confirmed as a low-frequency Mediterranean haplogroup. All four identified U6 HVI components have representatives in Atlantic Europe. This Maghreb component could have arrived through Atlantic Copper or Bronze age networks, leaving the presence of U6c to Punic or more probably, Roman colonization."

So MENA in Italy = various migrations through time

André de Vasconcelos said...

First off, Samuel, everyone knows you're usually unpolite, or downright rude, to people, but I never gave you any reason to behave in such matter with me, so quit with the name calling. Ironically it makes you look like the idiot

Secondly not once did you mention 'North Africa', whereas 'Middle East' showed up four times in your reply, so I just wanted to clarify that MENA already includes that IN THE VERY NAME, not just on cultural affinity or popular perception.

Anyway, stupid way to waste my time

Nirjhar007 said...

Coldmountains and others,

The Mitanni origins explained (Part 1 ) : Were the Mitanni Aryans really Indo-Aryans?

http://new-indology.blogspot.in/2017/05/were-mitanni-really-indo-aryans.html

Note : I will not discuss on the post here , I am only linking it . If you want to discuss , please do it so in the blog . But please do read the post carefully first, before throwing comments and suggestions .

Olympus Mons said...

@Andre,
Be gentle... their world is crumbling down. Be gentle.

So much effort done by the Labs to drill bones in search of dna in the Balkans to find the "steppe" in it... and at the end it really did a trick on them.

So, they need time to adjust.

Lets just enjoy the fables, shall we?

Salden said...

Hello there Afrocentrist.

Alogo said...

It was an interesting paper that included some barely-studied populations. What they found seems to be that the South Balkans (Albanians and Greeks) have been recently overall impacted more by Northeast populations and South Italy and the Aegean more by Southeast ones. Their dating aside, I'm sure all sorts of populations movements might be included in that including ones that brought some 'direct' steppe ancestry. The Arbereshe are shown as relatively recent Balkan immigrants and the Greek speakers as older more 'native' populations that very likely spanned a greater area in the past, agreeing with the historical record.

On that specific issue, every place in Europe currently has enough steppe for a steppe ancestry for core PIE, at the very least. After all the most 'extreme' scenarios like that of Robert Drews, who however preferred an entrance from the East for Greek and not a steppe homeland for PIE, assumed there would have been something like an 1:9 ratio with the original IE speakers diluted into the general population not too long after. So there's some wiggle room there. Areas like the Aegean and South Italy/Sicily seem like areas of secondary Indo-Europeanization so to speak, anyway. Either way, I thihk that's focusing on the least interesting part as far as this paper is concerned imo.

Olympus Mons said...

Hummm? Actually a south caucasus centrist you mean?

Or a balkan...ish...west south black sea centrist??

Having said that. Guys, north africa wont bite! For fuck sake.

epoch2013 said...

Do we know if neolithic European famer married patrilocal or matrilocal? In other words were farms inherited by sons in law or sons? Do we have anything on that? It may be of importance to a number of things I noticed.

Gioiello said...

Il suo test, anche se fatto su 3268 SNPs su 16569 (19,7%), è ottimo, anche perché l'aplogruppo K è testato particolarmente bene, io ho sempre pensato perché è l'aplogruppo di una buona parte di ebrei ashkenaziti, che sono K1a1b1a e altri gruppi K minori. Praticamente il K1b1c è testato integralmente con solo un no call e uno SNP non testato:
A73G G94A A263G A750G T1189C A1438G A1811G A2706G G3337A A3480G A4769G G5913A C7028T A8860G G9055A T9698C G9962A A10289G A10398G A10550G T11299C A11467G G11719A G11914A G12372A C14167T C14766T T14798C C15946T T16224C T16311C T16519C
A12308G no call
A15326G untested
Anche se non possiamo sapere senza un FMS se lei ha altre mutazioni, posso dire che in gran parte non ha le mutazioni private degli esemplari conosciuti:
i3001869 MT 146 T
i3001922 MT 152 T
i3002111 MT 199 T
rs3901846 MT 499 G
i6000025 MT 650 T
i3001940 MT 709 --
2971 not tested
i4000933 MT 4802 T
i6000266 MT 4970 A
i3002271 MT 5811 A
5834 not tested
6040 not tested
6617 not tested
6935 not tested
i4001048 MT 8545 G
9489 not tested
i4000775 MT 12397 A
12609 not tested
13026 not tested
i3001172 MT 13145 G
13794 not tested
i3001250 MT 14053 A
14078 not tested
15529 not tested
i3001552 MT 16051 A
i3001562 MT 16070 A
i3001776 MT 16266 C
i3001799 MT 16289 A
i3001839 MT 16318 A
i3001847 MT 16325 T
i3001891 MT 16365 --
per cui potrebbe appartenere all'aplotipo di base finora presente in solo due serbi:
15. KT697998(Serbia) Davidovic2016 K1b1c 02-SEP-2016
A73G G94A A263G 315.1C A750G T1189C A1438G A1811G A2706G G3337A A3480G A4769G G5913A C7028T A8860G G9055A T9698C G9962A A10289G A10398G A10550G T11299C A11467G G11719A G11914A A12308G G12372A C14167T C14766T T14798C A15326G C15946T T16224C T16311C T16519C
16. KT698029(Serbia) Davidovic2016 K1b1c 02-SEP-2016
A73G G94A A263G 315.1C A750G T1189C A1438G A1811G A2706G G3337A A3480G A4769G G5913A C7028T A8860G G9055A T9698C G9962A A10289G A10398G A10550G T11299C A11467G G11719A G11914A A12308G G12372A C14167T C14766T T14798C A15326G C15946T T16224C T16311C T16519C
Riguardo all'origine lei mi chiede un possibile collegamento con i Sarmati che avrebbero abitato il suo paese natale, ma, se loro erano prevalentemente soldati, certamente non le hanno trasmesso il mitocondriale.
Si è detto spesso che l'aplogruppo K sarebbe di origine mediorientale, ma è ormai chiaro, dagli ultimi esemplari di DNA antico pubblicati, che K1 (con gli Y I2, C1 ed R1) era presente nei cacciatori raccoglitori dell'Europa, per cui considererei senz'altro europeo il suo aplogruppo, e la presenza di K1b1b1 in Anatolia 6500-6200 aC non abolisce questo, essendo gli anatolici del nord più collegati con gli europei che con i mediorientali e gli iraniani antichi.
Riguardo al possibile collegamento del suo cognome Carosso con l'ebreo Carasso non darei molto peso, essendo Carasso un Y R-M269*, che, sebbene un outlier nel Jewish cluster, ha solo meno di 1000 anni ed è molto probabilmente una introgression da R-M269 iberici.
Naturalmente un FMS sarebbe utile anche per capire quanto il suo K1b1c possa essere antico in Italia, essendo anche il Piemonte una delle zone chiave della diffusione di Y R-L51 e anche molti mitocondriali nel popolamento dell'Europa moderna (questa è almeno la mia teoria del "Rifugio italiano").

Gioiello said...

From my friend Tino Mattoli, an R-L277 of noble family from Umbria who is waiting to know if his origins were from Alans or not. So far in his subclade only Italians. He lives in Chile, has 5 sons and thirteen grandsons. One of the Italians who are not becoming extinct. Unfortunately I cannot post the maid's pic.

The Uruguayan Maid
The Uruguayan maid asked for a pay increase.
The wife was very upset about this, and decided to talk to her about the raise. She asked: "Now Maria, why do you want a pay increase?"
Maria: "Well, Señora, there are tree reasons why I wanna increaze."
"The first is that I iron better than you."
Wife: "Who said you iron better than me?"Maria: "Jor huzban he say so."
Wife: "Oh yeah?"
Maria: "The second reason eez that I am a better cook than you."
Wife: "Nonsense, who said you were a better cook than me?"
Maria: "Jor hozban did."
Wife, increasingly agitated: "Oh he did, did he?"
Maria: "The third reason is that I am better at sex than you in the bed."
Wife, really boiling now and through gritted teeth. Wife: "And did my husband say that as well?"
Maria: "No Señora...."The gardener did."
Wife: "So, how much do you want?"

Gaspar said...

History states the arabs arrived in North Africa AFTER the fall of the Roman Empire. So any arab mix in Sicily is not ancient.

Gioiello said...

Gaspar, I don't like to speak about autosome, because it is, as Hegel said, like the night where all the cows are black. History is made of singular stories. People don't speak of Arab intake in Sicily, but of Berber one anyway, and Berbers aren't Arabs (see the struggle for speaking Amarigh again etc.) Anyway "Arabs" of Sicily were deported from Federico II to Apulia, and exterminated after a revolt, even though some "Arab" Y may be found there to-day. But also about Berbers we are speaking, with Ya Kem and other Berbers, if E-M81 is really from Northern Africa or not, being very recent his MRCA, and above all of E-L19, very old also in Europe, so I said that I am waiting for the aDNA for saying where it came from....

Matt said...

Another new Southeast Europe adna paper y'all - https://phys.org/news/2017-05-ancient-dna-evidence-hunter-gatherers-farmers.html

(Wonder if the Mathieson preprint got rushed a little cos of this...)

Salden said...

No one here said they had Arabian admixture. What was brought up was Maghrebi and Levantine mixture. Really.

http://italicroots.lefora.com said...

There is no Levantine or Natufian admixture, only a lot of Neolitich/Chalcolitich admixture from Anatolia and some Bronze Age/Iron Age admixture from the Steppe. The Steppes of Russia and Ukraine were overrun by Caucasus farmers who mixed with local HGs during the Neolitich, but for someone this admixture arrived in South Eastern Europe only from the Roman Empire.

Arza said...

Thanks Matt!

Chan_meso is intriguing.
Page 3, footnotes
(C) ADMIXTURE analysis. ADMIXTURE results are shown at K = 17. The Spanish and Romanian hunter-gatherer samples are composed entirely of the "blue"
component, which is also found in other European hunter-gatherer samples, with the exception of the oldest Spanish Mesolithic sample, Chan_Meso, which also
has a "lilac" component found in South Indians.

Matt said...

@ Arza, f3 from supplement, Chan vs Canes:

http://i.imgur.com/XUqVofr.png

Chan more related to Upper Paleolithic European samples (Goyet, El Miron, Vestonice). Canes shares more drift with other WHG, Natufians, etc.

Apóstolos Papaðimitríu said...

No Etruscans in SE Europe.

Davidski said...

No Etruscans in SE Europe.

Italy is in SE Europe, therefore Tuscany is in SE Europe, hence Etruscans lived in SE Europe.

Gioiello said...

@ Apóstolos Papaðimitríu

Timeo Danaos et dona ferentes, for that I was banned from Anthrogenica, thanks to your compatriot Costa. I don't say more about old history, otherwise Davidski deletes my posts. Greek is spoken from 10 million persons, Latin from more that 700 million. You very likely don't know history and all the rest. I wrote to the great linguist Schrjiver, who thought that the first migrants from Anatolia spoke Hatti. I said that very likely they spoke Etruscan, seen that Hatti isn't spoken anywhere in Europe, whereas Etruscan were spoken in Lemnos, and, after the migration to central Europe, was spoken from Rhaetians, Camuns and Etruscans who migrated to central Italy. You are the demonstration as to how Greeks are so little since 2300 years. I make you note that every contry of Europe has been tested, deeply, except Italy. You'll get many surprises when it will be done beyond Villabruna.

Rob said...

Err Dave how is Italy in southeast Europe ?

Rob said...

https://goo.gl/images/D65m9C

Davidski said...

Don't have time to argue bullshit semantics.

Italy is often described as a Southeastern European country, especially in terms of genetic clustering.

Alberto said...

@Arza

Sorry, I don't think I can even start to elaborate on that. If you do think that Mycenaean Greece and Srubnaya (which are contemporary) are sister cultures coming from the same people, ok. Things are bad enough here as they are (see Sam's reaction to his own idiocy) to start this other debate, so I think I'll have to drop it for the time being. Maybe next time.

Arza said...

@ Alberto

OK, fine. Maybe next time.

But just for a clarification - it was you who brought Srubnaya to the table. I wanted to talk rather about Mycenaean-like fortress from 1750 BCE with Mycenaean-like figurines inside.

Anthro Survey said...

@Rob

Southern Italy is a southeast Euro area genetically and would have been culturally, too, if Norman rule didn't latinize it by bringing settler colonies from France and North Italy there.

Rob said...

@ Anthro

Not that I have an issue with it, but I've not come across anyone with a right mind who would refer to Sicily as part of SEE culturally or genetically.
. SEE implies the Balkan Slav countries, romania etc
Not only modern differences, south Italy would lack many of the Neolithic copper age links between SEE and Eastern Europe
It's definitely a "Mediterranean European" country.
The clustering which is confusing some people here relates to their affinities to certain Greeks or what have you .

Davidski said...

Almost all of Italy is classic Southeastern Europe genetically, and also historically and broadly speaking geographically.

Tuscans overlap with Greeks on many PCA for a reason.

Anthro Survey said...

@Rob

Southern Italy should not be grouped together with SW Euro cluster proper, tho: North Italy, Iberia, Southern France.

We have yet to obtain more ancient DNA to vindicate it as part of SE Europe----If it turns out that by the Etruscan period S.Italy was displaying strong signals of admix from East Med area. (The Etruscan samples to the north so far look rather Iberian.)

Anthro Survey said...

I somehow think the admixture causing Tuscans to cluster where they cluster is more recent than in southern Italy. We'll see. It's also noteworthy how strong their particular r1b lineage there is. Could of course be a founder effect but could also indicate strong influence of BB, Urnfield, Villanova and/or other steppe-rich groups.

Culturally, Tuscany is and has always been interchangeable with places like Provence unlike S. italy. In fact, Tuscany has been the western european cultural archetype.

Tuscany never had the kinds of close connections to the east med over the ages that south italy did.

Davidski said...

It's likely that the Etruscans were originally from the Aegean, and if so, Tuscany was an East Med colony for a long time. I guess we'll see if that's true when Etruscan genomes are published.

Samuel Andrews said...

@Andre,
"First off, Samuel, everyone knows you're usually unpolite, or downright rude, to people,"

I reacted like that because a lot of people who post here are douche bags and your post looked rude to me.

Rob said...

There's no evidence of colonisation from the near east or Aegean in Etruscan territory
It is only Central European urnfield influences which occur just prior the Etruscan state

Anthro Survey said...

@Rob

I was also thinking along these lines for Tuscany.

Although, there could have been a very minor influence there at the time and mainly manifested itsef in changes to local material culture---kind of like Phoenician influence in ancient Catalonia.

Davidski said...

@Rob

There's no evidence of colonisation from the near east or Aegean in Etruscan territory
It is only Central European urnfield influences which occur just prior the Etruscan state.


We'll see about that when the Etruscan genomes come out.

Rob said...

@ Anthro

It's been a while since I looked at it, but it does seem to be a 'package' of cremation, iron working, etc arriving from central Europe, thus more than some cursory influences.
The curious aspect is this C.E. impact coincides broadly but not quite exactly where Etruscans emerge, whilst the 'periphery' along the eastern littoral continued to be more dispersed groups with Tumulus burials.
Of course, "Orientalizing' influences arrived, but this was with Greek colonists.

@ Dave

I know you're a raving Kurgan devotee, but don't be so defensive, we're just discussing. I know you're very well read, so When do you propose the migration occurred ?

Gioiello said...

@ Davidski

"We'll see about that when the Etruscan genomes come out".

History is complex. If were confirmed my theory that Etruscans were the descendants, at least in part from a genetical point of view, of the agriculturalists that from Northern Anatolia colonized Europe till Central Europe and after from there migrated southward leaving in their pathway the Rhaetians and Camuns till Etruscans in Central Italy, how may you demonstrate that from an autosomal point of view? It isn't that more demonstrable from single haplotypes and from the language and culture?


Davidski said...

I'm not being defensive, but I can see parallels between the Corded Ware story and the Etruscan story, with a lot of opposition to a migration origin for both.

So let's wait for the Etruscan genomes, which are probably coming very soon. They may well be very similar the the Early Bronze Age Anatolians, because there's some merit to the claim that Etruscan was a Tyrsenian language from the Aegean coast.

Nirjhar007 said...

Just heard that the big Greek one will be out very soon , probably within a few days . Going to be awesome .

Davidski said...

The big fat Greek ancient paper: soon at Nature.

Rob said...

RE: Greek genomes. Awesome !

RE: Etruscnas. The story of CWC & Etruscans is different.
The idea that CWC are from somewhere around the steppe was always popular and had strong intimations. Just read Coon's work. It's just that archaeologists got a but too cute with immobilism.

Etruscans are different. There is no clear path & a number of possibilities exist, such as local Neolithic / Copper Age descendants. I'll go over the materials. .

Anthro Survey said...

@Rob

I meant cursory influences from the Near East. Central Euro influences were substantial both in terms of DNA and culturally speaking. No question about this!

epoch2013 said...

That is interesting. Apparently a derived SLC24A5 in early mesolithic Spain. And quite some diversity as well. I wonder how both these samples would score on the Asian, Middle-Eastern and ANE admixture found in Fu et al.

Furthermore, we know that Magdalenian can be modeled as a mixture of Goyet116 and proto-Villabruna. Lately La Brana has been modeled as WHG + 35% extra something, related to K14. Why doesn't this paper do D-stats like this?

Mbuti Goyet116-1; Chan-Meso, Canes1-Meso

Are these samples out yet? Will they get out?

epoch2013 said...

O, sorry, wrong thread.

Apóstolos Papaðimitríu said...

Davidski said:
| Italy is in SE Europe, therefore Tuscany is in SE Europe, hence
| Etruscans lived in SE Europe.

If Germany is Central Europe, then Italy is South Central Europe or something.
SE Europe is Albania, Bosnia, Bulgaria, Croatia, Greece, Kosovo, FYR Macedonia, Moldova, Montenegro, Romania, Serbia, Slovenia and East Thrace (Turkey), some of them partially.

@Gioello
Whatever. I don't consider Greek someone who is moderator in Anthrogenica. I know much more about the history of the region than what you think though and try to use arguments.

Davidski said...

If Germany is Central Europe, then Italy is South Central Europe or something.
SE Europe is Albania, Bosnia, Bulgaria, Croatia, Greece, Kosovo, FYR Macedonia, Moldova, Montenegro, Romania, Serbia, Slovenia and East Thrace (Turkey), some of them partially.


You keep yapping on about modern geography, and missing the point in my post.

What the hell does the Slavic Balkans got to do with the language shifts from non-IE to IE in the Roman and Greek world in Etruria, Sicily, Crete and Cyprus?

Gioiello said...

@ Apóstolos Papaðimitríu

"@Gioello
Whatever. I don't consider Greek someone who is moderator in Anthrogenica. I know much more about the history of the region than what you think though and try to use arguments".

That person usurps a glorious Greek name, but he belongs to the owners of those blogs, he too a lackey of the criminal firm. Nothing to do with me and you, except he will discover that his ancestors came from the Sea Peoples, but he won't ever be like you and me. The Costa I spoke about is a true Greek, a J2 of Venetian origin. The other is only a J1...

Samuel Andrews said...

Italy can only considered SE Europe genetically.

Geographically you can call it anything from SouthWest Europe to Central Europe to SouthEast Europe. No matter how you put it Italy looks pretty isolated on a map because it's a peninsula.

But of course that might have meant nothing to ancient gene flow, as Italy probably received as much or more genetic shifts since the Mesolithic as any other region of Europe.

Apóstolos Papaðimitríu said...

@Davidski
Why the language shift in Cyprus has to be different than the language shift in Poland? In Cyprus Greek texts are dated to the 11th century BC. The first Polish text is written c. 1370 AD.

Simon_W said...

Well! I've got some interesting anecdotal evidence from my own Global10 results to share. As some of you might remember, 1/4 of my ancestry is from Cesena and surroundings, in the Emilia-Romagna (the rest from Germany and German speaking Switzerland).

Using nMonte I can model myself very well as:

French_East 34.2
Maros:RISE373 25.8
German 22
Polish 13.25
England_Roman_outlier:3DRIF-26 1.35
Jordan_EBA:average 1.3
Armenia_Chalcolithic:average 1.3
Mozabite 0.8

This would mean that my Italian grandfather would have been approximately:

86.66% Maros
10.2% Armenia_ChL + Jordan_EBA
3.14% North African

All in all a fairly modest MENA contribution.

However, if I delete Maros from the sheet of samples, nMonte takes the French_East as a substitute and gives me an impossibly high French_East score. So I have to delete those as well. In consequence I get modelled like this:

German 48.15
Hungary_BA:I1504 17.2
Polish 14.05
Iceman_MN:Iceman 9.9
Jordan_EBA:average 4.05
Armenia_Chalcolithic:average 4
Anatolia_Chalcolithic:I1584 1.6
Mozabite 1.05

So my Italian grandfather would have been approximately like this:

38.82% Ötzi
19.22% Hungary_BA:I1504 (aka BR2, the LBA sample)
37.84% Armenia_ChL + Jordan_EBA + Anatolia_ChL
4.12% Mozabite

Which would mean a very massive MENA influx.

Obviously the Maros culture from early Bronze Age southern Hungary packed a lot of MENA admixture. It must have been similar to a mix of Ötzi + BR2 + Armenia_ChL + Anatolia_ChL + Jordan_EBA. The question is: Did the largest part of the MENA admixture in the Romagna come from such a Balkan/Hungarian Bronze Age culture, or was the Hungarian/IE influence in the Romagna more BR2-like and the MENA influence from a later period like the Roman and early Byzantine Age?

Considering the currently available dating efforts with Alder and similar methods, which clearly suggest a Roman or even post-Roman date for admixture between North European-like and MENA-like pops, I tend to believe the second option to be true.

Simon_W said...

BTW the 37.84% MENA influx according to the second model are similar to the 33% Cypriot-like admixture Hellenthal et al. calculated for the Bergamasque sample.

Simon_W said...

But southern Italy is a different story, and at least a large part of the Cypriot-like / West Asian / CHG related ancestry may have arrived at a much earlier date than in the north. That's what the Alder results of Sarno et al. suggest. The admixture event for the South Italian and Sicilian clusters between Sardinian-like and West Asian pops dates to before 1000 BC, which is the upper limit for the method. And in contrast to what I said in a previous thread, this is not in conflict with Busby et al. 2015. The latter paper showed (in Figure S3) for the South Italian cluster itali8 an admixture event dated to the Roman age that only brought Levantine and North African admixture, while the Greek-like and the Caucasus- and Armenia-related admixture was already there.

Simon_W said...

As for the Etruscans, I agree with David about a probable Aegaean origin. The paper by Busby et al. 2015 showed in one sample of modern Tuscans, tsi70, small slices of Southeast European and East Caucasian ancestries on the local side of the Roman Era admixture event. While the large amounts of Armenian-, Cypriot-, Levantine- and North African-related admixtures were on the other, non-local side of the admixture event. Which means that at least a little bit of Greek-like and Caucasus-related admixture was already there prior to this. And I would ascribe this to the Etruscans, although it would mean that their share in the MENA admixture of modern Tuscans was small.

Simon_W said...

@ Rob

You're certainly right that a link between Etruria and the Aegaean is archaeologically hard to establish at the right time, in the Bronze Age, and that the Orientalizing phase of the Etruscan culture has nothing to do with their eastern origin, because it came too late and was mediated by Greeks.

But personally I don't believe that the Protovillanovans were those who brought the ancestor of the Etruscan language to Italy. I think the former were Indo-Europeans, and more exactly speaking I think they were ancestral to Latino-Faliscan, like I explained in a recent thread on Anthrogenica. This hypothesis is for sure much more likely than the interpretation of the Protovillanovans as Sabellics. Because the concentration of Protovillanovan sites in the area of the Latins and Faliscans is rather dense, much denser than in most of the Sabellic lands. And most of all, they stuck to the cremation rites introduced by the Protovillanovans clearly longer than most of the Sabellics.

Anthro Survey said...

@Simon

I always reference the Etruscan genomes to which we were introduced here by you: the ones you mentioned clustering together with Iberians.

Is there a chance you can project them onto a complete West Wurasian PCA? The one in the original article is far from satisfactory.

I don't discount the possibility of LIMITED genetic introgression from the Aegean, btw. It was just more of an elite dominance type of thing over a population resembling aforementioned Etruscan genomes and the BB gentleman from Parma.

Also, what predecessor culture would you attribute Umbro-Sabellic languages to if not proto-Villanovan?

Davidski said...

@Apóstolos Papaðimitríu

Yeah, obviously, because language change is always accompanied by texts. And language change in Late Neolithic Poland via a massive migration of steppe pastoralists was obviously very similar to the language changes in historical city states of the East Mediterranean.

Please note the sarcasm and stop wasting my time.

Simon_W said...

@ Anthro Survey

No, I can't project them; even if I had access to these genomes I wouldn't be able. The best I can do is to rotate the PCA and to put it near a similar European PCA, like one from Günther et. al 2015, for comparison. One of the Etruscans has a Bulgarian-like position:

https://jpst.it/_zRP

I wonder why they never published a paper about these...

As for Sabellic, I'm pretty sure it was the language of the Apennine culture, and I guess that it got to Italy early on with the Cetina culture that linked eastern Italy with the Balkan coast.

Simon_W said...

@ André de Vasconcelos

"What I feel is that there is a racist undertone in some people who want to stay away from anything that isn't Europe when it comes to their ancestry, be it thre Middle East, North Africa or whatnot, and see that as a sort of an attack/insult. I was hoping the advances in population genetics would help eliminate this, as we're all so very much mixed, but stupidity will always be stupidity"

I couldn't have said it any better. The behaviour of these guys is obvious and ridiculous. They should be sent to London to visit the British Museum, so they could see the greatness and magic of ancient Near Eastern cultures.

Rob said...

@ Simon W

I don't find the idea of separate waves of Italic migrations into Italy very parasiminous, on the basis of linguistic data, which favours one "Proto-Italic" event

Arza said...

@ Simon W

This is as close as I can get with DoHA:
https://1.bp.blogspot.com/-9YvZzMHx5r4/WSmCu1OvhoI/AAAAAAAAAMM/IDMku2YtiwYAChXj4LOmCxDR52frkkE5wCLcB/s1600/Etruscans_PCA.png

It's a rotated and scaled 3D view of PC1/PC2/PC3.
Indeed they look Bulgarian or rather generally Balkan-like, but as you can see there is a second option. Due to differences in third dimension some Spanish and Portuguese samples create a "tail" that becomes visible when I rotate the plot. And this "tail" is placed exactly like Etruscan samples.

https://1.bp.blogspot.com/-V0EOgVWvfGk/WSmFt-LQ8PI/AAAAAAAAAMY/G66TIDn2pBo59cHUV_4Y8IjPujbHRUnPwCLcB/s1600/Etruscans_PCA_sup.png

http://italicroots.lefora.com said...

@Rob

"I don't find the idea of separate waves of Italic migrations into Italy very parasiminous, on the basis of linguistic data, which favours one "Proto-Italic" event"

Well that's like claiming that there were no separate waves of Indo-Iranian migrations because, on the basis of linguistic data, there was one "Proto-Indo-Iranian" event"

http://italicroots.lefora.com said...

Also look how the most isolated S.Italians from Bovesia and Pelopponnesian Greeks from Mani turned out almost Cypriot like in Stamatoyannopoulos et al. This new paper about Minoans and Maykop will be a real pain for some here who think that ancient Romans/Greeks were Germans/Swedes who then mixed with Lebanese slaves. LOL.

Gioiello said...

@ Rob
"This is certainly a minority view today, eg Silverstri and Garrett,"

Are you speaking of Domenico Silvestri? I knew his book about the "Indomediterraneo" of more than 40 years ago. After no news about him. I did other.

Rob said...

@ Italicroots

Yes, the idea of multiple waves of Italic people migrating into Italy - just as the idea of multiple Greek invasions (eg of Ionians, then Dorians, etc..) is an outdated minority one today (eg Silverstri, Chadwick, and Garrett). Rather, the formulation of Italic languages is seen to have occurred within Italy. Thus to speak of a distinctly "Sabellic migration' which can correlate with an archaeological horizon seems rather speculative.

Simon_W said...

@ Rob

Well, admittedly, it may seem odd to postulate a Proto-Italic homeland outside of Italy when all written evidence for Italic is only found in Italy. This would seem to imply that almost magically the various Italic branches at various times all migrated to Italy.

How much parsimonious hence the idea that there was one Proto-Italic migration followed by a local differentiation.

A problem with this reasoning is just that there is no written evidence of the languages of Bronze Age Hungary and nearby areas. What do we know about the language of the eastern sphere of the Urnfield culture? It would be pure speculation to call it Celtic.

Apart from that I don't think it could be proven with hard linguistic evidence that Italic differentiated in situ. I mean all the linguists can do is to show similarities and differences. They could argue that the Italic languages are quite close to each other, so they cannot have split too long ago. But then there is also the phenomenon of convergence, that they may have influenced each other, thereby reducing the differences again. It has always been a controversy how much of the commonalities were inherited and how much due to convergence. But sure, it's very easygoing to postulate one single migration followed by insitu differentiation. That's parsimony. Apart from this I don't see much of an advantage of this theory.

But sure I could imagine that both Sabellic and Latino-Faliscan were spoken in the Apennine culture, and that the Protovillanovans didn't have a lasting impact anywhere.

Simon_W said...

@ italicroots.lefora

Griko speakers from Calabria and southernmost Peloponnesian Greeks are one thing, the ancient Latins are a different thing. I suspect that the original Latino-Faliscans at the time of their invasion were Hungary_BA-like. So neither Swedish nor Cypriot-like. But then they mixed with the locals and acquired more southern ancestry, maybe Iberian-like. Greek- and Caucasus-related admixture in southern Italy may be quite old and some of this must have diffused northwards, probably more easily after the Roman unification of Italy than prior to that time. There's also that Aeneas myth... I suspect it's more myth than reality, but I hope we'll be able to test it eventually. I suspect there's no single explanation for Near Eastern and Caucasus related admixture in Italy that holds true for all regions. Apart from slavery there was also voluntary migration within the Roman empire, of traders, soldiers, magistrates etc. There was even an emperor with Syrian roots. And then the spread of the early Christians who had originated in the MENA region. Lastly, the Muslim expansion must have caused a certain exodus of MENA Christians to the lands that were still under Byzantine control at that time.

Simon_W said...

@ Arza

Thanks for the interesting effort. But the topography of PCA plots depends on the samples included, hence a West Eurasian PCA looks very different from a Europe-only PCA, and a worldwide PCA is again very different. And I see there's a Natufian sample in your PCA, so I'm not sure that they can be compared...

Simon_W said...

Yeah I really don't think slavery was the main reason for MENA admixture in Italy, because the slaves were really from everywhere, there were also Gaulish, Germanic, British slaves, also Iberian ones and many Greeks.

Gioiello said...

@ Simon_W
"There was even an emperor with Syrian roots"

But very likely he left a few descendants!

Poise n Pen said...

If language and genes don't go together though then you should stop talking about yamnaya then. If they do then it is highly odd that SE europe does not have much more input when they had to have had IE languages first.