The identity of the earliest inhabitants of Xinjiang, in the heart of Inner Asia, and the languages that they spoke have long been debated and remain contentious 1. Here we present genomic data from 5 individuals dating to around 3000–2800 bc from the Dzungarian Basin and 13 individuals dating to around 2100–1700 bc from the Tarim Basin, representing the earliest yet discovered human remains from North and South Xinjiang, respectively. We find that the Early Bronze Age Dzungarian individuals exhibit a predominantly Afanasievo ancestry with an additional local contribution, and the Early–Middle Bronze Age Tarim individuals contain only a local ancestry. The Tarim individuals from the site of Xiaohe further exhibit strong evidence of milk proteins in their dental calculus, indicating a reliance on dairy pastoralism at the site since its founding. Our results do not support previous hypotheses for the origin of the Tarim mummies, who were argued to be Proto-Tocharian-speaking pastoralists descended from the Afanasievo 1,2 or to have originated among the Bactria–Margiana Archaeological Complex 3 or Inner Asian Mountain Corridor cultures 4. Instead, although Tocharian may have been plausibly introduced to the Dzungarian Basin by Afanasievo migrants during the Early Bronze Age, we find that the earliest Tarim Basin cultures appear to have arisen from a genetically isolated local population that adopted neighbouring pastoralist and agriculturalist practices, which allowed them to settle and thrive along the shifting riverine oases of the Taklamakan Desert.Zhang, F., Ning, C., Scott, A. et al. The genomic origins of the Bronze Age Tarim Basin mummies. Nature (2021). https://doi.org/10.1038/s41586-021-04052-7 See also... How the Shirenzigou nomads became Proto-Tocharians
Wednesday, October 27, 2021
Local origins of the earliest Tarim Basin mummies (Zhang et al. 2021)
Over at Nature at this LINK. It's nice to see yet another huge surprise courtesy of ancient DNA. Please note that most of the ancients from this paper are already in the Global25 datasheets. Here's the abstract:
Subscribe to:
Post Comments (Atom)
282 comments:
«Oldest ‹Older 201 – 282 of 282"The Steppe Hypothesis is in great shape."
Not wrt tocharian.
1-2 grains found in afanasievo doesn't cut it. Sorry. Afanasievo had no agriculture.
@Vasishta
"1-2 grains found in afanasievo doesn't cut it. Sorry. Afanasievo had no agriculture."
Yes it does make all the difference. Whether Afanasievo had agriculture or not isn't even that important. They were somehow in touch with it. That's enough to explain IE roots.
I wanted to say "And You know who has direct ancestry of them."
As for Urartu we will see ancient Urartians DNA next year. I hope both from Armenia and Turkey. Before getting them let's not forget that from Persian perspective Urartu and Armenia were synonyms in Darius Behistun inscription. So things are little bit more complicated than some people want to present.
And here are some qpAdm from Molgen made by a Georgian user btw. Tail prob is low, and proportions are not realistic. But over time things will become better.
----
Armenian.DG
Kura_Araxes_ARM_Kaps. 14.8%
ARM_LBA 63.8%
TUR_Arslantepe_EBA. 21.4%
tail: 0.112209
chisq: 16.856
And another with lower tail prob
Armenian.DG
Kura_Araxes_ARM_Kaps. 24.3%
ARM_LBA 59.0%
TUR_Alalakh_MLBA. 16.7%
tail: 0.0843576
chisq: 17.882
----
left pops:
Armenian.DG
Kura_Araxes_ARM_Kaps
ARM_LBA
TUR_Alalakh_MLBA
0.243 0.590 0.167
tail: 0.0843576
chisq: 17.882
right pops:
Mbuti.DG
Morocco_Iberomaurusian
Israel_Natufian_published
Israel_PPNB
Iran_GanjDareh_N
Turkey_N
CHG
Russia_MA1_HG.SG
Russia_Ust_Ishim.DG
Russia_Kostenki14.SG
Russia_HG_Karelia.SG
Russia_Siberia_Lena_EN
Brazil_LapaDoSanto_9600BP
Italy_North_Villabruna_HG
left pops:
Armenian.DG
Kura_Araxes_ARM_Kaps
ARM_LBA
TUR_Arslantepe_EBA
0.148 0.638 0.214
tail: 0.112209
chisq: 16.856
right pops:
Mbuti.DG
Morocco_Iberomaurusian
Israel_Natufian_published
Israel_PPNB
Iran_GanjDareh_N
Turkey_N
CHG
Russia_MA1_HG.SG
Russia_Ust_Ishim.DG
Russia_Kostenki14.SG
Russia_HG_Karelia.SG
Russia_Siberia_Lena_EN
Brazil_LapaDoSanto_9600BP
Italy_North_Villabruna_HG
“Who were the Picts, then?”
Who were the Britons giving Q-Celtic names to places in the midlands and north of England.? England still has loads of placenames that would be called Gaelic if they were in either Scotland or ireland. Why is no-one asking who they were or what their language was?
Picts are name Victorians gave to one or maybe two early mediaeval polities of eastern & northern Scotland, based on some mentions of the Cruithne in Irish annals. They were presumed by the Victorians to be the same polity beyond the Antonine wall mentioned by the Romans as “Picts and other Caledonian”, and commemorated as a defeated enemy by one or two Roman army trinkets. Whether they were meaningfully the same polities, or even in the same geographical location is only presumption. There’s three or four centuries of constant tumult in between.
As far as ogham inscriptions are understood, one of the most common words is maqqi (son of) as a name element. This is explained by some as somehow not being the language of the Picts but of Irish monks. Yet clearly the orthography in general does not follow that the Irish being used in Ireland. To decide they are P-Celtic speakers means discounting the most obvious and best understood bits of their epigraphs. I suspect that’s because the P/Q distinction is not a useful one. Let alone a defining one.
No-one as yet has given a satisfactory account of their language, beyond the likelihood that it was in the celtic group and divergent from the ones we know. Or rather, the Roman influence made the language in Wales and England diverge more. No-one has a clue what it was like before, during ir after the Roman period other than that to Romans, whatbthey heard in south-east England seemed to them like Gaulish.
The entire basis of supposing that the people of north-east Scotland were P-Celtic speakers (if it were a meaningful category) is a placename element “pitt/pett”. That element is mirrored further west in “both” and “bath; this is ignored. Some placenames that are cited as examples of P-celtic have mutated from “bath” to “pit” during the modern era. Like Pitfoddels in Aberdeenshire, still written Bathfoddels is older documents. Anyone familiar with Doric accents would not be surprised. This too has to be ignored.
There is no adna distinction established between adna got from burials of the Pictish area and period, and that of contiguous populations. Not much has been collected; Scotland’s soils tend to be pretty acidic.
Naive geneticists have made big claims based on modern dna without taking into account what large modern movements there have been in Scotland since the 9th century, such as the new Royal boroughs like Inverness, Aberdeen, Perth etc being planted with the Borders & Lothians descendants of minor-gentry refugees who had fled Viking and Norman invasions in England.
In short, as far as dna goes, from the little we have, the Picts that the Romans encountered were the same population as those around them. The early mediaeval people who left the carved stones were Christian at a time when much of England reverted to or adopted paganism; they were crazy about that whole macho/hierarchical cultural package of horses, hounds, & hawks; hunting deer and boar (and the nobs making a parade/procession out of it) they had collegiate clergy who travelled widely, they had promontory forts, probably still occupied some brochs, they had scriptorums, and their churches were amongst the first in Britain to fall prey to viking destruction.
@epoch
sure. in the same way NW Indians knew of horses and chariots. no need for steppe to explain any IE culture in India. See the shallowness of your point?
anyway, lets say i go 1 step further and agree that afanasievo had full scale agriculture(which it didnt). what does that have to do with tocharian and whats the link?
@genos historia
Your comment about the UK leaders, bell beakers and what “the leaders” want people in the UK to know — what a laugh.
Almost the whole of the UK media is in the hands of three billionaire families. Rothermere, Barclay and Murdoch. Murdoch empire media write whatever they think a majority of the population will want to buy, unless it’s overtly left wing, unless and until a solid majority left wing readership develops, when his papers become a bit more centrist to appeal to them. So for most of this century, that has been from solidly-right to reactionary.
The other two are unapologetically and consistently right wing.
By the way, our PM and his second in command Gove both worked for most of their lives for these same families; as does Gove’s wife, and the wife of another prominent Downing St mover and shaker, now shaken out, Cummjngs.
Even the one kinda-leftish tabloid tpin the UK, the Mail, is socially very conservative, reactionary, even.(Supported the Nazis, never retracted that support.)
For the last decade, “conservative” (actually radical) right-wing governments in the UK have had the appointment of the state broadcaster: to which they have consistently appointed their own politicians and lately the Conservative Party’s financial backers. These are all the top positions making policy in the BBC.
British governments have kept TV news in hand in commercial channels by means of the long-standing small contribution to news costs in return for the same control over their news values. Insanely, they are going to give that up, so they will lose that editorial influence; the only reason they can have for this is that they think their friends will buy it up.
Anyway, these are the “leaders” you imagine nefariously keeping the UK blind to our bell beaker heritage with some clever programme of lefty obscuration.
As if the internet didn’t exist, too.
So the conclusion you leaped to is just absolutely absurd. Utterly bonkers.
I need to reply here because I just noticed Targamos reply to my comment. Excuse me for offtop.
Targamos
You didn't explain the high fst btw Meskhetians and Armenians. Anyway.
Using Paleolithic/ Neolithic sources to model Armenians and Georgians just show that You ignore how ancient Near East functioned.
Armenians do not have real ancestry from neither Anatolia Barcin neither even CHG.
Armenians do have plenty of ancestry from unsampled regions of fertile crescent ( Nevali Chori for example ) that were lacking any WHG or EHG/ANE that is present both in Barcin and CHG respectively. WHG and EHG that easily mask any Yamna. Armenians offcourse have some KA ancestry also.
Those Fertile Crescent people legacy is well visible in Arslantepe, Alalakh, Hajji Firuz Chl. All this places to which Armenians had good affinity.
And when they are used in conjunction with Yamna then Armenians score decent amount of Yamna. Look at the excellent fit.
Target: Armenian
Distance: 0.9251% / 0.00925141
50.2 TUR_Arslantepe_EBA
36.0 IRN_Hajji_Firuz_C
10.0 RUS_Catacomb
3.8 Kura-Araxes_ARM_Talin
0.0 RUS_Darkveti-Meshoko_En
0.0 TUR_Kaman-Kalehoyuk_MLBA
But even this is not important. Because the proto Armenian is from Arm_LBA population not directly from Yamna. So obviously Armenians do have much more Proto Armenian ancestry than some people try to insinuate.
The fertile crescent ancestry is one of the main reasons that Armenians are the closest population to Neolithic South Caucasus.
The fertile crescent ancestry is very well visible in Y DNA. Neolithic lines are very diverse in Armenians which many Caucasian people lack because most of their Y dna are founder effects. And most of time in/after Bronze Age.
The irony of the situation is that Caucasian languages (including Hurrian who later expanded with Kur-Araxes) in most likelihood descend from those farmers. Which is visible in their Y dna but their autosomes had changed considerably over time. They were Caucasified while moving to North and later they got a lot off Steppe and even Turkic ancestry.
So please stop that pointless talks about rapes and brutalities. Otherwise we will move in a nasty direction.
As for Diau, Tao, Mushki etc. Anyone can claim whatever he want, especially in the absence of texts. But it's not what will matter at the end of the day.
With some effort I can find excellent etymology for Diau/Tao. Daos means wolf in Phrygian and in some Paleobalkanic languages. And there were plenty IE tribes with the root Dao. Like Dauni people. Dao/Dacians.
So what?
It's a DNA discussion. And based on DNA it's very very unlikely that genetics will give a support to most of this claims.
@Vasishta
"anyway, lets say i go 1 step further and agree that afanasievo had full scale agriculture(which it didnt). what does that have to do with tocharian and whats the link?"
Read that Mallory article you quoted from. One of the proposed steps has already been confirmed by aDNA. Also, do take not that no aDNA is available from areas where Tocharian A and B was spoken and that Tocharian C are just a number of loan words.
"One of the proposed steps has already been confirmed by aDNA. "
Ok, I will partially concede that 1 of those steps is done.
However, we have tarim_emba as well as afanasievo. They do not Harbor afanasievo ancestry. I do not know why you are hopeful that you will find it in between those time periods.
Refer
https://eurogenes.blogspot.com/2021/03/how-shirenzigou-nomads-became-proto.html?m=1
@ Aram
First of all using copper age and EBA samples to model a new population/group is quite outdated and anachronistic if you ask me.You def need more recent refrences to use.Your model also looks overfitting to me.Second,the steppe admix among Armenians is overrated just duel with it.It is barely 8-10% and some individuals have it even lower.We still have not seen those supposedly high-steppe admixed Armenians,wherever they hiding.And despite the fact that Armenians have tested a lot.'Steppe' admix among Armenians is ofc a secondary issue to worry about.What it matters is how they received so much Iranic and Levantine admix.Its obvious that their 'Iranic' and even 'Levant' admix are not just coming from the Fertile Crescent or northern mesopotamia/eastern anatolia area/regions.There are lineages among modern Armenians suggesting a Semitic and an Iranic-Persian contribution in their ethnogenesis.Ofc there is a possibility some of these lineages to have local origins/roots.In general 'Armenian highlands' and 'Armenians' as an ethnic group require deeper testing to found out about their ethnogenesis.So,i would suggest you to stop pretending the expert when it comes to Armenian genetics.
How can you even know if the steppe ancestry seen in Armenians primarily or entirely came by way of "Proto-Armenian" peoples?
@Vasishta
Hope has nothing to do with it. Tocharian A + B weren't spoken in the area from which we have Tarim EBA. The only thing that we have for that is Tocharian C. But its very existence is even debated,so a few loanwords can hardly indicate the place where this hypothetical language was spoken.
Literally the only thing that has changed by this aDNA is that the need for tying Xiaohe with Tocharian, a need that causes much problems, has disappeared. That the Tocharian speakers had Europoid appearances we know from the cave paintings.
Does the Black Sea Deluge have anything to do with the migration of CHG into Steppe? Progress into LDL? Sumerians from Anatolia into Iraq? @Aram @Togarma The Noah’s story is based on the Gilgamesh one. So it means that the progenitors of Sumerians might’ve lived on the Black Sea shores, right? Is there any significance that the Ark landed on mount Ararat in PD Armenia? (As in all surviving humanity allegedly derived from the Southern Caucasus). I mean, these legends must be some faint echos to obscure Prehistoric events, aren’t they?
Could it be that to solve the PIE origin mystery we would need to look into now-submerged, under water Black Sea dwellings?
@ Genos
''The study claims to be about Britain. Yet they have a bunch of DNA from other parts of Europe.
Ok. That is bit a miscommunication. It makes those Y DNA results meaningless.
I'll wait till people at anthrogencia sort things out before saying anything.''
yeah I'm also waiting for the clarity from those yokels. What's their current theory ? Celtic emerged in Pennsylvania, with a second coming of ANE, written in the book of Mormon
epoch wrote,
"Hope has nothing to do with it. Tocharian A + B weren't spoken in the area from which we have Tarim EBA. The only thing that we have for that is Tocharian C. But its very existence is even debated,so a few loanwords can hardly indicate the place where this hypothetical language was spoken."
Looking at the latest version of the YFull mtree, I have noticed that the positions of the two specimens from the Qigeman site in Shanshan County (cf. Wang et al. 2021, "Ancient Xinjiang mitogenomes reveal intense
admixture with high genetic diversity") have been changed and are now more consistent with the haplogroup assignments provided in Table S1 of the original study.
QGM_4 (age 360-176 cal BCE, 2126 - 2310 BP) is now tabulated under Z4* on the YFull mtree; this specimen's mtDNA belongs to haplogroup Z4a1a according to Table S1 of Wang et al. 2021. QGM_M2 is now tabulated under U5a1g1a on the YFull mtree; this specimen's mtDNA belongs to haplogroup U5a1g1 according to Table S1 of Wang et al. 2021. (It appears that the QGM_M2 specimen has not been directly radiocarbon dated, but it seems to have been assigned a date of 2310-2126 BP on the basis of the dating for the QGM_4 specimen.)
The Qigeman site is located in the northeastern part of the Turpan Depression, near the Bezeklik Thousand Buddha Caves and within the area from which documents composed in a so-called Tocharian language have been recovered.
The Heyd interview was quite interesting... It seems like he's going into a dichotomy of the peaceful Yamnaya and the violent Corded Ware, from skeletal trauma. I wonder how much this is based on skeletons the early Corded Ware with more diverse (if still not very diverse) y-dna, against the later Corded Ware who have more of a narrowing / change of their y diversity. Maybe there is some situation where the groups seed into Central Europe, and then one group takes control of the landscape more violently.
I still have to say that when I look at the Bohemia Corded Ware transect, and then rescale time into generations (assuming 25 years per generation), then it looks like this - https://imgur.com/a/dJP7j0J
It looks like it takes about 20 generations to go from approximately 80% Progress_En ancestry (PNL1, VLI076 and OB003) to 60-55% Progress_En ancestry (in Bell_Beaker and CWC_late), which does seem to me about 1.5% admixture per generation, and it does look kind of a thing that increases over time rather than all being done in the first few 8 generations or something like (200 years).
(One thing of note was when I looked around for papers around signs of skeletal violence in late Neolithic cultures, Battle Axe culture in Gotland was mentioned... But these also seem to be the Pitted Ware Battle Axe culture, and not actually the genetic Corded Ware steppe migrants.)
There's obviously something to all of this of course.
....
Some articles in Spanish media today around the opening of a new exhibition / museum yesterday in Murcia around the El Argar early / proto civilization:
https://elnoroestedigital.com/la-almoloya-el-primer-parlamento-de-europa-abre-sus-puertas-en-el-convento-san-francisco-de-mula/
No information on the ancient dna work, but it seems quite likely to me that this is going to be timed close to the release of the new paper on El Argar.
Hardworking user on Anthrogenica, Altvred, has run some qpAdm models and y-dna calls for the samples from the upcoming British/LBA-Iron Age paper, so I've tabulated the results for the purpses of this comment thread.
First, here's how his model works on the "known" samples, the ones that Reich lab have previously published: https://imgur.com/a/tvEz8Q6
Basically it looks like his autosomal proportions are all correct, and his y-dna calls are correct *for males*, but with a caveat that two females in the set were called with y-dna groups that they obviously don't have.
So unless Reich lab got it wrong then and now an unpublished version have confirmed that they're male, I think that would be a problem for drawing too many conclusions from his y-dna calls, as it's possible that around 50% might be wrong or inapplicable.
And here's for the unknown set, which I've divided into EEF (no steppe) and then three tiers of level of steppe ancestry: https://imgur.com/a/cfTxVzt
(High steppe is basically above present day European range and getting towards Sintashta level, medium is more than Southern Europeans and getting up to the max in present day Northern European, low is at the level of present day Southern Europeans or below).
It looks to me like he's got many of the y-calls and proportions right, since e.g. most of the samples with EEF proportions have I2a1, for example, and there's relatively high representation of G2a in low steppe category, etc. Likewise J2b samples with steppe have a relatively low WHG proportion, consistent with SE European links.
But of course some things could be very wrong, so I wouldn't draw and conclusions from, e.g. the presence of good sized number of samples with I2a who have high steppe proportion. Too much unknown about these samples in terms of sex and location and time to draw anything out.
@Matt
Thanks for that , that is great. I notice a lot of U152 in the low steppe group. Conversely most of the med-high steppe is L21.
What are the odds that the Anglo Saxons came during the Iron Age, instead of the Middle Ages?
Linguistically it makes no sense for there to be 3 major IE waves (discounting Vikings/Normans).
Maybe the original IE wave was the Celts, and the Iron Age is the Anglo-Saxons that slowly expanded their influence. Plenty of historians go on about there being no archeological evidence of an Anglo Saxon invasion.
@DragonHermit,
Historical records are enough to confirm when the Angli, Saxon tribes came to England. It happened when written records were "abundant" in Europe.
People in formally Roman Europe talked about the Anglo Saxon invasion as it was happening.
British Romans, wrote a book a couple centuries later about how Anglo Saxons conquered their island.
Anglo Saxon monk Bede, said the English (name for all Germanic peoples in Britain) came from Germany a couple centuries earlier. He went into detail.
The Anglo Saxon chronicle in 900 AD, said they invaded Britain from Germany in 400 AD.
English priest in 900 AD, warned English that Danish would conquer Britain just as their ancestors from Germany conquered Britain 500 years earlier.
Everyone in Europe knew who the Saxon tribe was. They are documented by Romans going back to circa 200 AD.
Saxon tribes lived independently in Germany contemporary to the Saxon kingdoms in Britain for 400 years. In 700 AD, everyone in Europe knew the Saxons in Germany & in Britain were the same ethnicity.
The connection between Saxons in Britain & Germany was too recent for the invasion to be Iron age.
Then there's lingustics. Which is enough to show English language separated from other Germanic languages no more than 2,000 years ago. The connection with Frisian languages is more recent.
@Mat don't you think it is better when comparing different models; to parse the Sintashta burials of R1b-Z2103- Sintashta-_Corded Ware-_Eastern Bell Beaker component. Since R1a-branch of Sintashta differs from R1a found in Corded Ware and differs from R1b-Z2103 Sintashta in Potopovka I0246. They differ both in Kurgan burials and age of 4 and 2 wheeled transportation and DOM2 lineage.
For example even though Potapovka and Sintashta horses are basically contemporary there is a difference in the Tyumen HG like component elevated( 4x) in Sintashta MLBA 02 outlier I0246(Utyevka VI, kurgan 6, grave2)- compared to R1a I0419 sample.
"earliest apparent chariot rider we have - "The Utyevka VI cemetery was located 0.8 km north-northeast of the village of Utyevka, Samara oblast, south of the Samara River. It contained some of the richest and most unusual graves of the Potapovka culture. Kurgan 6, grave 2 contained six humans: a male-female couple buried facing each other aged 15-17; and four children and infants too old and numerous to be the offspring of the couple—perhaps siblings, Horse sacrifices, shield-shaped studded bone cheekpieces interpreted as chariot-driving gear, weapons (three copper daggers, a flat copper axe, 16 flint projectile points), copper beads and rings, and other objects were found in the grave. Sample I0246 is from the 15-17-year-old male (confirmed genetically).")."
“ Plenty of historians go on about there being no archeological evidence of an Anglo Saxon invasion.“
The entire lowlands have Germanic cremation cemeteries
Even the Immobilists acknowledge that; they just pretend that this was all ‘cultural borrowing’
AS slaughter the Welsh. There’s no coincidence that there are nary zero Brythonic words in English, although I wonder about words like dog or boy w/o any clear Germanic cognates. Germanic aDNA is at least 50% if not 60-70 in East England and Lowland Scotland. Only after WWII the liberal revisionists started distancing British from German by claiming a phony “Celtic” lineages that didn’t really exist.
The Anglo-Saxon identity was formed in situ, not in the continent. Otherwise, Angles, Saxons and Jutes would have all have different dialects. They all spoke one unified language.
We still have no major archeological records of an Anglo-Saxon invasion in the Middle Ages. Not sure we have any genetic evidence either of that timeline. We know there's a "Danish-like" component to English genetics, but we don't know when it came to the isles. The genetic history tells another story entirely.
We're sweeping under the rug a major Indo-European invasion, if there are indeed 2 major IE invasions before 0 AD.
There's a very clear genetic difference between samples from Celtic and Anglo-Saxon graves in England.
https://eurogenes.blogspot.com/2019/12/avalon-vs-valhalla-revisited.html
The Anglo-Saxons are Germanic-like.
@All
https://drive.google.com/file/d/1WbT7fVlNB61h_9bZmK3TvVCxShVwB3jW/view?usp=sharing
Please keep in mind that these are preliminary coords.
So I won't be putting these samples into the G25 datasheets until I get the official genotypes.
the proto-Welsh mostly fled into Wales and consolidated around old Iron Age forts. This facilitated a physical and metaphorical revival of Brythonic identity, de-Romanization and polarizarion cf new Anglo-Saxon arrivals. In my mind; the archaeology has always been clear, although a handful of British archaeologists like some odd claims. But I’d look to the works of Katherine Hill & Christopher Scull. They’ve always been realistic & sober
@Davidski what do you make of this?
To me it seems strange that there could an IE people predating Celto-Germanic invasions in Britain and we have no clue they existed.
There are two main options:
1) British Beakers spoke a Celtic-like language, so when Celtic languages arrived in Britain and Ireland there weren't any non-Indo-European languages left to make an impact on them
2) British Beakers spoke a non-Indo-European language that their Corded Ware ancestors picked up from the western farmers that they mixed with.
Currently, the expert and general opinion out there is that the former is much more likely.
@DragonHermit
"Plenty of historians go on about there being no archeological evidence of an Anglo Saxon invasion. "
O, there is. The area of the Angles in Germany, aptly named "Angeln", shows an occupation hiatus just when the Angles migrated to Great Britain.
Thanks Davidski.
OK, first to validate, checking the samples that have previously been published and are being republished at higher coverage against their previous G25 hits: https://imgur.com/a/X5vjNC2
Although some population labels are different (which reflects differences between G25, how I have relabelled some G25 populations, and how Reich lab anno labels things and have in some cases relabelled samples over time), you can see there is a high degree of validation between these, and they're projecting pretty similarly. They're nigh on indistinguishable. So adds some confidence these samples are projecting well.
Second, here are the known sample IDs that I couldn't find in previous G25 data (presumably cos the old versions were too low coverage): https://imgur.com/a/DMcQqxL
They project pretty much like you'd expect, again adds some validation.
The position of Beaker I2565 (previously low coverage) looks potentially interesting, since this R-L21 guy looks relatively displaced on North Europe PCA towards the "Celtic" pole. (R-L21 call by teepean on Anthrogenica). but I haven't looked at these samples in any detail.
It's also interesting that he's one the Amesbury group ("companion"), though would have to read Olalde's supplement to get more information. Compared to Dutch Beaker I4067, who overlaps I2565 on the West Eurasia PCA, he's much more displaced towards the Irish/Celtic end of things.
(Here's a quick .csv file with the G25 data for all the above: https://pastebin.com/pYdSJuz3).
I'm going to have a look now at finding lowest distances for the other new samples - although we don't know anything about them, comparing their closest known G25 sample to the y-dna calls people have put out there might show something interesting.
Some time ago I proposed following model for Slavic expansion and diversity:
https://postimg.cc/64y9tHdK
Arza fund following Slavic Y-DNA in Celtic papier data:
I13780 - https://www.yfull.com/tree/R-YP5267/
is mixed Slavic - Germanic/Celtic
I13795 - https://www.yfull.com/tree/R-Z280/ (xZ92,CTS1211)
is French-like
I14193 - https://www.yfull.com/tree/R-Z280/ (xCTS1211)
is Sintashta-like
I18719 - https://www.yfull.com/tree/I-CTS10228/
is Bulgarian-like
I25524 - https://www.yfull.com/tree/R-M458/
is Lithuanian-like
I have added new G25 data with Slavic Y-DNA:
https://postimg.cc/tn10zh5C
It seems to confirm my model. The diversity of early Slavs was probably even greater as they mixed with adjacent populations:
Linguistically and culturally Sintashta was very close to early Slavs as links with Mierzanowice, Unetice and Trzciniec show.
Alright guys, here's a list of lowest distance samples for the unknowns: https://pastebin.com/U1ncsFsv
It looks like most are British or Irish like in their closest sample, but not all.
An obvious one here is I25524, who had in Altvred's data R1a1a1b1a1a~ and a Baltic_BA like mix of WHG:Barcin:Steppe, and so is unsurprisingly plotting closest to the Lithuanian like person from the Golden Horde, and sits on top of Lithuanians on PCA (https://imgur.com/a/nOR6SLW).
Another obvious one is I18241 who is closest to the Encrusted Pottery Culture and quite outlying on PCA, though has no y-dna call I can compare to (if they're even male).
I'd guess they might be a resequenced version of the Vatya outlier RISE479, although there are differences in how these samples project onto the North Europe PCA.
(Visually: https://imgur.com/a/GDil7Pg).
Samples I18183, I20515, I25509 will probably be of interest to Ambron as their best overall match is modern day or Viking Age Poles. If they are all securely Iron Age samples from somewhere then that's interesting in terms of what it shows. Overall samples closest to and close to present day East-Central European people are limited. (And there are also some Ukrainian/Lithuanian and Russian/Polish like samples in the form of I18226 and I16089, and a Hungarian like I5288. The former I18226 is pretty Baltic BA like. I18226 was called as R1b by Altvred but as y haplogroup also called as O as teepean, it's think it's likely that this person is female and the y call isn't right).
Also, some quick validation that Altvred's fits are basically correct: https://imgur.com/a/FpYBfb6
(Some overlap between different steppe thresholds on West_Eurasia_PCA; these are probably samples where they're on the margin between the groups, or where other factors like the HG:Barcin level and West:East European drift affect placement. There were also a couple with marginal steppe at around 6/7 level that I put into the "LOW STEPPE" group in my previous post but are really closer to EEF).
(Altvred fit around 355/836 samples, so these plots don't represent the full breadth of the sampling necessarily, and I've excluded republished samples here).
Foxvillage
The rationale of using this specific references is very simple. They are the closest to historic Armenia. And second reason is that they do NOT have steppe ancestry at all (except the KA). Younger samples already have steppe so it's meaningless to use them. I think it's quite obvious that if You want to measure total steppe ancestry You need to choose such references that completely lack steppe. Then see how much steppe is added.
As for overfiting. Many people confuse low distance and overfit. Overfit is not the low distance per se. Overfit is when to much similar pops are used to get low distance. In that model the number of source pops is minimalistic. So there is no reason to believe that there is an overfit. But if You want that I reduce the number of source pops. Here just with three pops. Notice still an excellent fit.
Target: Armenian
Distance: 0.9357% / 0.00935746
50.6 TUR_Arslantepe_EBA
38.8 IRN_Hajji_Firuz_C
10.6 RUS_Catacomb
Target: Armenian_Erzurum
Distance: 1.3163% / 0.01316339
55.8 TUR_Arslantepe_EBA
33.2 IRN_Hajji_Firuz_C
11.0 RUS_Catacomb
So it's still 10 percent. My point was very simple. The supposedly anomalous low level of Steppe (lower than 5 percent ) in Armenia is a cliche based on poor sampling and poor understanding. And yes I agree that average will be something like 8-10 percent. Eastern Armenians will have higher than average , while the western one lower than the average.
As for Semitic and Iranic influences. Do You think Semites are from Moon and they have some unusual autosomes? Look at that Alalakh_MLBA which was known to have mix Semitic , Hurrian and maybe Luwian population. Do You see there anything remarkable except the typical Semitic J1-Z1828? So it's not clear what You mean. Nobody denies that Armenians have some Semitic and Iranic but so what? Can You be more specific how much it is and prove that they exceed 10-15 percent in sum.
Do this change the fact that Armenians have close affinity to Fertile Crescent affiliated pops like Hajji Firuz Chl and Arslantepe. And have numerous Y DNA on most basal positions of farming lineages. And why those haplotypes couldn't have live natively in southern parts of historic Armenia which btw in some places overlap with Fertile Crescent.
Do I say something that is new to anyone here?
Btw all Iranians have BMAC ancestry. While Armenians have very low affinity to BMAC. And how much R1a Armenians have You must know better than me because You claim to know Armenian DNA better than me.
Copper Axe.
I was not saying that all Steppe in Armenia is from Proto-Armenians. Steppe ancestry was present in Kura-Araxes from the preceding LC influx. Steppe ancestry came with Iranians. So 10 percent is total of all events.
But to be honest. The Iranian impact in Armenia is lower than I and many others imagined.
Dutch IA are really on an "oscillating" position! Between Isles/ 'Celtic' and Nordic/ 'Germanic '
As such there is some cultural difference in a coastal zone (Noord-Holland/NH, Zeeland) and inland Nijmegen. At Nijmegen there were some important La Tene foundings for example.
Who has a hunch????
https://i.postimg.cc/QxvtCkty/Unknown-46.png
https://i.postimg.cc/RVtmsrSg/Unknown-47.png
Matt, thank you! I also follow these genomes on AG and on Arza's blog.
My error
I was wanting to say J1-Z1853 in Alalakh. I just checked and it reaches 4-4.5 percent of Armenian Y dna.
Andre
Without ancient DNA we can only speculate about Sumerians. But based on the presence of cognates in basic vocabulary btw Hurrian and Sumerian Kassian proposed a northern origin of Sumerians.
I guess this is quite probable. But we need aDNA.
My pet theory is that J1-P58 (prominent among Semites) is an Ubaid marker.
So... I'm really curious to find out which one of these samples is the famour "Amesbury Archer" (the "King of Stonehenge").
We know from Reich's presentation that he's in the set: https://imgur.com/a/Ki7B21d
And we can guess the following fact about him: He's likely to have a similar proportion, but very slightly lower, proportion of EEF ancestry to I2416 from Amesbury Down at about the same time, and to I2462 woman from Thanet a little later. We can know this from a G25 comparison of the published samples to Reich's presentation (which I've also included in the above). He also would seem likely to be close to both above and beyond this due to his location (although we can't prove this from Reich's plot).
So, one way to do a sift for this is to check to see which samples are closest to I2416 and I2462 out of the unknown samples: https://imgur.com/a/o9hOYkc
The results show that, of the samples in the Celtic paper, the only ones who are in the top ten closest matches for both I2416 and I2462 are:
I16099: lowest distance to I2416, 7th closest to I2462, called as I2a1ba1b by Altvred and I-L1229 by teepean. (G25 Proportions: 46:31:23: Barcin:Yamnaya:WHG)
I17017: 8th closest to I2416, 4th closest to I2462, appears more likely to be female to me as teepean called them as Q-Y16849, which seems unlikely. (G25 Proportions: 45:36:19: Barcin:Yamnaya:WHG)
The proportions of I16099 would also fit OK with being very slightly lower in EEF than I2416 (49:25:26) and I2462 (47:33:20).
So on balance I16099 seems most likely *BUT* please note this is just a guess and not certain at all, and just based on trying to find the best overall match in the set for two high EEF Beaker-CA samples we have.
There are many other new samples in the set from Reich lab who would seem to be outliers with I2416 and I2462 like ancestry proportions, so it's very possible I16099 could be any of those. (Also I16099 does seem to be a little off in WHG:Barcin ratio; though that may be consistent with an origin from Switzerland as is sometimes suggested by isotopes?).
In any event though, although the Archer is thought to be related to I2565 ('The Companion'), it's interesting that this sample must have a very different ratio of EEF, if the Archer does have a similar ratio of EEF to I2416 and I2462. 'The Companion' definitely has a very average level of steppe ancestry (came out at 58% Yamnaya vs 55% for all GBR_Beaker, including high EEF samples). If I16099 was found to be a relative of I2565, it would probably be a dead cert he is the Archer though.
@Wee
It seems more and more probable that two languages were spoken in Pictland or had been spoken at least, and that the one found on Oghams was not brittonic;
that said, NO, the theory of a Britttonic language in Pictland is not based ONLY on the PIt- placenames. Other typically P- Celtic placenames existed there, containing 'carden', 'pert', 'lanerc' (cf Welsh 'Llanerch'), 'pevr', 'aber', 'pren', 'tref-' or '-tref'; a small handufl of 'bath-' placenames adapted (analogy?)later in 'pit-' (loaned itself later by Gaelic speakers) doesn't reject the big niver of typical Pit- placenames of Pictland. A Brittonic etymology seems very reliable for 'pit'.
We know little yet, it's ready, but the known language of Picts has always been named 'Brittonic' by foreigners of the times (high Middle Ages), they never looked for another name marking a difference from other tribes of Britain.
Even the other mysterious language, almost inattested in Toponymy, could have been Celtic, or Proto-Celtic or at least I-Ean (BB's ?)
THat said, that Q- Celtic languages could have been spoken BeFORE in the Isles is almost certain, Pictland or other places.
Q6 and P- division is not only based on this lone stop: Q- and P- Celtic tongues show other regular phonetic changes for the most in consonnants evolution, spite both sorts as Celtic tend towards strong consonnants lenition as a whole; differences maybe born by differences in the place of stress in words.
I can only analyze Ust' Ishim as a European IUP representative since none of the IUP genomes geographically from Europe are in the Global25 dataset, but Ust' Ishim can represent them to a great extent:
https://drive.google.com/file/d/1jatGsd0Y5wOEcmJ6BjkObfeOT71jC4nj/view?usp=sharing
As you see, Ust' Ishim is somewhere in the midst of early differentiated East and West Crown Eurasians genetically, belongs to neither of them but is probably from the earlier undifferentiated Crown Eurasians.
@ Aram
From historical Armenian lands we got a few samples mostly from kura-araxes culture,the samples from eastern Armenian(Areni etc) and the 3 samples from Arslantepe witch are closer to western Armenia/eastern anatolia.We lack DNA from the LBA/IA period especially from the western historical Armenian area.We lack DNA from Urartu and we have zero idea how Armenians arrived in those lands and how much genetic impact these IE folks had to the natives.Without ancient DNA we just guessworking.I will insist that using a copper age sample to model a modern population is anachronistic for me.It is too old despite if you can get good fits.I can agree that a sample/refrence like Arslantepe/Alalalkh it is very likely to cover a huge bulk of their autosomal.But they need something extra 'Iranic'that it seems to me to be more recent.Armenians are genetically very close to Assyrians as well.I am not avoiding the fact intermarriages between these groups especially during Roman/Byzantine and Ottoman times.Anyway lets hope that we are going to see some Armenian DNA in the near future.I heard that neolithic samples will come out from Armenian highlands.Very interesting to see what component was native there.
@weure
Dutch have something northwest in their autosomal.For some reason they shifting towards to British/Irish.I think both the Germanic tribes and the Celtic people assilimated Beaker folks when they settled in the low countries.Dutch along with some Belgians and Flemish folks can have similar looks/phenotypes with British/Irish.'Keltic Nordic' is one of the examples.I associate this type of look with bronze age beakers(R1b).You barely can see this pheno/look outside from Britain/Ireland and Low countries.NorthwestBlock culture was probably a Beaker related culture.
@Aram “ Btw all Iranians have BMAC ancestry. While Armenians have very low affinity to BMAC. And how much R1a Armenians have You must know better than me because You claim to know Armenian DNA better than me.”
Were Iran_N, BMAC and Dravidian related?
@Aram “ Without ancient DNA we can only speculate about Sumerians. But based on the presence of cognates in basic vocabulary btw Hurrian and Sumerian Kassian proposed a northern origin of Sumerians.
I guess this is quite probable. But we need aDNA.”
Doesn’t seem that way. Sumerian is apparently a “language isolate”. They either came from Anatolian farmers or perhaps a Kelteminar/Botai like WSHG pop.
@weure
Finn, the samples are from around Nijmegen because that is where the only Dutch burials are found that are usable.
@weure
https://studenttheses.universiteitleiden.nl/access/item%3A2607845/view
Based on the new Dutch IA samples, the headline is that the old Frisians of IA and Roman times were clearly genetically clearly genetically affiliated with the Britons on the other side of the North Sea. We could label them as: North Sea Celts. The affinity in language on both sides of the North Sea is most lately stated by Schrijver (2017).No wonder in the case of the West Frisians regarding their poll position towards the Isles. Later on in migration times in the fifth and sixth century the "Anglo-Saxon" amalgam came in....And old English and Frisian (after migration time) was clearly Germanic with a Celtic substrate. I guess the picture fits.
The question is what caused this affiliation in the case of the West-Frisians I would go for an Tumulus/Urnfield connection because the Hoogkarspel culture (1800-400 BC), as part of the Elp culture, was Tumulus/ Urnfield derived. And I guess from there the net was also thrown across the Isles...But may be there a better explanations that caused the affiliate in genetic (and linguistic) sense!
As user Ebizur on AG noted there are mtDNA samples available from the Iron Age where Tocharian A was spoken, from Qigeman in Shanshan County. The paper that published them assigned one them (QGM_M2) to U5a1g1 and another (QGM_4) to Z4a1a.
https://www.science.org/doi/10.1126/sciadv.abd6690#F1
And that is interesting as the Afanasievo sample I5272 from Elo Bashi has mtDNA U5a1g1 (published in Narasimhan 2019)
Off topic: The data from the Transeurasian languages paper with a fair number of Bronze Age genomes from Japan and Korea (https://www.researchgate.net/publication/353278242_Triangulation_supports_agricultural_spread_of_the_Transeurasian_languages/fulltext/60f0ddec9541032c6d41dee5/Triangulation-supports-agricultural-spread-of-the-Transeurasian-languages.pdf) was uploaded to the ENA in August
https://www.ebi.ac.uk/ena/browser/view/PRJEB46162?show=reads
I guess it would be easier and better to get genotypes, but is any brave person with the requisite knowledge and an interest up for tackling these .bams?
The other adna paper by Lara Cassidy's group, from September, also has the data here:
https://www.ebi.ac.uk/ena/browser/view/PRJEB43762?show=reads
It would be cool to look at both to compare their conclusions.
@ Matt
Triangulation supports agricultural spread of the Transeurasian languages
by Ning
https://edmond.mpdl.mpg.de/imeji/collection/59JGAaOpSxRb96Vh?q=
@weure
The name of "De Vlie" undoubtedly comes from Flevo and has a clear Germanic origin. Frisians spoke a Germanic language by the time they encountered Romans.
@epoch, thanks for the publication, interesting. Yeah most of the old Frisians during IA were incremented, 'Aak' and the Hoogovenman are the exceptions to the rule.
@Foxvillagers 'Dutch have something northwest in their autosomal.For some reason they shifting towards to British/Irish.I think both the Germanic tribes and the Celtic people assilimated Beaker folks when they settled in the low countries.' Of course there is a BB component in it, but the Dutch IA samples are very clear muted compared to BB. I think the Tumulus/ Elp culture, a network of a warrior elite had a genetic impact. So we must not forget that this had changed the genetic profile (during BA). The R1b U016 branch of one of the IA West Frisians is Z306, https://www.yfull.com/tree/R-S497/, may be around DF96/98. The Jimonice/ Unetice sample from 2000 BC is DF98. So the Tumulus/ Urnfield wind blowed from Central Europe, Southern Germany to NW Europe. This modified the (post) Beaker population in the NW.
In case anyone was interested, I put the "Celtic Paper" G25 on a Neighbour Joining Tree with a selection of (fairly randomly) selected frame samples - https://pdfhost.io/v/n0Lnc.ZNj_tree_rotated
Searchable and may provide another way of looking at which samples are close to which other samples.
@epoch ....and the Rhine clearly Celtic and the Frisians chieftains Malorix and Verritus that went to Rome too, the names of Drenthe and Twenthe have a Celtic root, so we can start endless debates....
Peter Schrijver has shown that North Sea Germanic (after the migrations!!!) has a Celtic substrate, also based on the vowel system. Earlier on came Kuhn with the Nordwestblock (nor Germanic nor Celtic). Much reason to doubt if the Frisii spoke clearcut Germanic (in Jastorfian sense).
We simply don't know what they spoke, no sources, no evidence. All speculative bla bla bla (incl. myself).
Nevertheless the IA Frisians were connected with the neigboring Isles and the Rhineland/ German inland. That were also the most near and most developed area's (in IA!).
Things began definitely to chance when the Anglo-Saxon amalgam, very clear Germanic speakers from Schleswig-Holstein and surrounding area's dropped in (fifth/ sixth century) and began to dominate the scene also in Friesland.
Nevertheless the old Frisian language (from after the migration time)- like old English- still contained much 'celtic' relicts....
@moesan
Far from being culture of the high middle ages, (ie starting at the end if the first millennium AD) there was no Pictish polity or language remaining by then.
You are the only person who I have ever seen claim such a thing. From most of the placenames you have cited, you also appear to think their territory was a couple of hundred miles south of where anyone else locates its southern edge.
@ Moesan
I do agree with you that the Ogham of Pictland is probably not Brittonic.
I also think that casting the question in terms of a Brittonic/Gaelic split is not even useful, since we have no clear idea what Brittonic was before it replaced half its vocabulary with Roman Latin versions. Or even what it was like for a few centuries after it did.
Almost all we know is what it and its offspring, Breton looked like quite a long time after the Romans left. Which is half latinised.
The Roman opinion of tribes on the south east of England adjacent to the French coast, soon after invasion, was that they spoke something that sounded to Romans to be close to Gaulish. We don’t even know whether that was true for, or tyoical of, the rest of what is now England, and nor do we have any explanation for the placenames in England that resemble Gaelic more than Welsh.
@ weure
i agree. It is beyond doubt that the 'Old (true) Frisii'' are different to those of Frankish times, who borrowed their name. The latter are Anglo-Saxons
I too have doubts about the Germanic character of old Frisii, but seems etymological evidence is meagre
@weure
"Nevertheless the old Frisian language (from after the migration time)- like old English- still contained much 'celtic' relicts...."
Which contradicts the theory that the old Frisii spoke Celtic *and* the area was completely depopulated because if the area really was devoid of people, how would the Anglo-Saxon newcomers pick up Celtic relics?
Foxvillager
New ancient DNA will offcourse help for better understanding. But some conclusions can be already made. One important thing is that the apparent similarities of two pops can be misleading. Here is a very simple model showing that Assyrians are in reality more related to Iranians. And indeed they have affinity to Jordan EBA as one would expect from a Semitic pop. While Armenians do not look particularly related to both of them. Kurds have strong affinity to BMAC as one would expect from an Iranian pop.
Target: Armenian
Distance: 0.8977% / 0.00897739
57.2 TUR_Arslantepe_EBA
26.6 IRN_Hajji_Firuz_C
9.4 RUS_Catacomb
3.6 RUS_Maykop_Novosvobodnaya
3.2 TKM_Gonur1_BA
Notice the presence of Jordan EBA in Kurds which is consistent with their geographic position relative to Armenians.
Target: Kurdish
Distance: 1.4331% / 0.01433144
33.8 TUR_Arslantepe_EBA
28.2 TKM_Gonur1_BA
13.4 IRN_Hajji_Firuz_C
12.8 RUS_Catacomb
11.8 Levant_JOR_EBA
Notice the presence of BMAC (Gonur) in Assyrians.
Target: Assyrian
Distance: 1.0742% / 0.01074218
49.0 TUR_Arslantepe_EBA
17.0 IRN_Hajji_Firuz_C
14.0 TKM_Gonur1_BA
12.8 Levant_JOR_EBA
7.2 RUS_Catacomb
And here is an individual apparently from East Armenia with higher steppe than average. Higher than the average Kurdish by the way. But still no extra affinity to BMAC. I hope next year new modern Armenian samples will be published with regional diversity.
Target: Armenian:GS000013745
Distance: 1.6416% / 0.01641606
54.8 TUR_Arslantepe_EBA
19.4 IRN_Hajji_Firuz_C
13.6 RUS_Catacomb
12.2 RUS_Maykop_Novosvobodnaya
And here finally another pop that on PCA and on Admixtures look close to Armenians but has quite different proportions. High Levantine and strong affinity to Iran as one would expect from a Semitic population who moved to Georgia via the Persian Empire.
Target: Georgian_Jew
Distance: 1.2962% / 0.01296205
56.4 TUR_Arslantepe_EBA
17.4 Levant_JOR_EBA
17.0 TKM_Gonur1_BA
9.2 RUS_Catacomb
PS Btw we don't have until now any Medes or ancient Persian sample. I guess they surprise a lot off people.
As for good and bad models. The age of samples is not a big issue. After all even with paleolithic era samples one can see the difference.
What is more important for good modeling is that all source samples have more or less the same age. Which will exclude shared ancestry issue.
Samples must be as unadmixed as possible. For example using Hajji Firuz BA is meaningless because it is a hybrid pop.
Samples should not be very close to each other. And source samples should not be very numerous.
That is fully sufficient to figure out who is who.
@Aram In your opinion, were Hurro-Urrartians related to Barcin farmers or more to an Iran_N/Chl or CHG population?
"Were Iran_N, BMAC and Dravidian related?"
Only as per crackpots.
David, is this data enough to form the coordinates of G25?
https://ngdc.cncb.ac.cn/gsa-human/browse/HRA000810
https://academic.oup.com/mbe/article/38/11/4908/6329832#supplementary-data
@Aram actually it is quite possible that F38 from Hasanlu and I2327 from Hajji Firuz, two EIA samples from around northwestern Zagros Mountains, were Mede-like ancient Western Iranics. Some experts (like Parpola, Kuzmina, Kurochkin, R. Dyson, C. Young etc.) have argued for an Iranic presence in Hasanlu Tepe IV or even V. It seems like these two samples both have BMAC ancestry in addition to Zagros_ChL and Steppe_MLBA, which probably means that they were either Mitannis or early Western Iranics (Dumezil, Francfort and some others had argued for a Mitanni presence in Hasanlu V). Some may argue that they can not be Aryans since they are not R1a but; 1. Haplos =/= languages and 2. We have some samples from Poltavka Culture (which is generally associated with Proto-Indo-Iranians) that have the same R1b subclade as their paternal haplogroup.
And I agree with you: Although Armenians have low WSH ancestry, it is still possible that they have a considerable percentage of Proto-Armenian (maybe something Lchashen-Metsamor-like?) ancestry.
@Andrze BMAC is mostly Iran_N (I prefer Zagros_N since the word "Iran" could be confusing for some people) derived. The origins of Dravidian is still a mystery. There is a strong tendency among Indologists to think that some of Avestan and Sanskrit vocabulary have BMAC-language as their origin; yet the etymology of these hypothetical BMAC words is non-Dravidian. So it may not be the case that Dravidian is related to either Zagros_N or BMAC.
@epoch it was not completely depopulated, Westergo most probably was, but other North Dutch parts definitely not, Westergo around Wijnaldum was also the most Anglo-Saxon/ Jutish like in the North Dutch area.
@Vladimir
I already ran those samples after someone from Anthrogenica emailed me the genotypes.
The G25 coords should be in the relevant thread at Anthrogenica.
Andre
My point in this thread was that beside Anatolia_N and Iran Neolithic related groups there was another group in the middle of them which we can label Central Farmers or Central Fertile Crescent farmers. They moved to North. Settled in valleys of Highlands and Caucasus. Probably mixed with some local foragers , which resulted in the increase of CHG and reexpanded as Kura-Araxes.
My feeling is that Hurro-Urartian was a language of the farmer side, rather than CHG side. Despite the increase of CHG. But I am fine with the CHG version also.
For example hašhur means apple in sumero-akkadian and hinzur means the same in Hurrian. The extra -n- in the middle of word is a regular nasalisation for Hurrian. Like in Puratu>Purantu.
Btw we have South Caucasian Neolithic farmers sample. It's Y dna is a usual fertile crescent stuff the H2 popular in European Neo also. While autosomaly it was looking something like this.
Target: AZE_Caucasus_lowlands_LN
Distance: 1.7712% / 0.01771214
58.8 TUR_Arslantepe_EBA
20.8 IRN_Hajji_Firuz_C
20.2 RUS_Maykop_Novosvobodnaya
0.2 TKM_Gonur1_BA
The model is ahistoric. Because younger sources are used to model an older sample. But I give it just to have a possibility to compare it's proportions with modern pops. As You can see it is not like modern Caucasians , who are rich in Maykop, despite the fact that it can be a good candidate for being their linguistic ancestor.
Those farmers were neither Anatolia neither Iran but in the middle.
As for the language of BMAC. I don't think it was a Dravidian related language. Should it be the case there would be traces in Iranian languages. Which is not the case.
@vasistha & all
What is the likelihood that BMAC actually contributed genetically to South Asians (yes even the North-west South Asia so heavily fetishized on anthrogenica lol) in any capacity?
From the current UZB/TKM samples it doesn't seem like there has been any contribution at all, that is to say if most of them are even BMAC [Bustan?]. They aren't close to any modern population and sit way off axis on well-thought out PCAs.
Anthrogenica vehemently defend the idea of significant BMAC contribution but will concede it as "BMAC-like" when pushed. This is the problem with taking nMonte for something it is not and a misunderstanding of stat artefact(s).
Nezih Seven
The Celtic case, the Greeks in Empirus with zero steppe and Roman empire/languages expansion all shows that after initial dispersal of IE people there were secondary dispersals which spread daughter languages.
I analysed numerous Y dna clusters. Iranians have many y dna clusters outside R1a some of them from BMAC. The same about Greeks, the same about Armenians which is not exclusively R1b/I2c.
All this means that sort of Caucasian supremasistic style comments like "look how little Yamna You have" is a crap.
Yamna is important but is not anymore relevant to daughter languages expansions who could have an autosome shift.
In the same manner the true Iranian ancestry is much higher than the Andronovo.
And yes Armenians have a lot off Lchashen type ancestry.
As for NW Iran Iron Age. Iranian names are attested there. One of famous one is the Bagdatu the ruler of Uishdish which is close to Hajji Firuz. So everything is possible. Except I am acautious person so I would like to see more aDNA to make judgements
@ Aram
Which substrates do people propose to exist in proto-Iranian & proto-Armenian ?
@Wee
For High Middle Age, it’s a question of definition ; Bede was born in the 7th Cy end ; I consider we are in core Middle Age after the 1000’s only, and the 7th is no more Antiquity for me. But you’re right, I misinterpreted a sentance I read ; in fact, yes, Bede distinguished between Brittonic and Pictic ; what level of differences, it isn’t said, and Bede wasn’t a linguist, I suppose.
Concerning the placenames I evoked and the southern boundaries of Pictland, I took as reliable that the limits in East were the Firth of Forth between Fifeshire in North and Lothians in South, roughly said. In West, the pre-Gaelic frontiers are not very sure, but it seems Brittons (a tribe of northern Domnonians according to someones) had been settled around Dumbarton in what has become N-Stratclud, in a later Cumbric zone ; who north of them, and when ? Uneasy to answer precisely.
& : It’s true toponymy gives us a few clues, and the most of the placenames I took as references were denser in South Pictland than in North ; elements in placenames depend on geography profiles too, we cannot consider we ‘ll find the same density for every word or word element in the whole country.
This apparently more southern distributions of the elements I cited (except ‘aber’) could be explained by a late enough colonisation by Brittonic speakers of a previously « foreign » territory (under ancient « true » Picts ?) ; the seemingly Brittonic settlements in West could support this ???. But if we rely on more ancient localisations of Caledonian tribes, we see Cornovii in most than a place, as far as North Scotland (Caithness, the Cornauii in Ptolemy) ; but is was before the historical Pictland. What we have at hand is rather scarce to be sure. On another side, the most of tribes names of preceding periods among Caledonians were rather Celtic ones. Would have been « Oghams genuine Picts » new not-Celts intruders from elsewhere? Not completely resolved.
Concerning Welsh and cousin languages (rather of Cornovii-like origin: Cornish, Breton, spite Vannetais dialect of Breton shows at the same time a few continental Celtic and Welshlike inputs), I cannot say « latinised Brittonic », because their syntax and phonetics shows very poor Latin influence, and the High Middle Age Breton remnants qualified as Cumbric seemingly was very close to what could have given these modern languages. I don’t discard other dialects here and there which could have presented mixed traits, between typical Gaelic and Brittonic langages, but the traces which remains are rather in favour of already well formed P- dialects almost allover Britain, except maybe (?) some parts of far Northern Scotland, at least at the time of the so called Anglo-Saxons invasion. Some Armorican Breton terms show ties with Cumbric, plus the most of Armorican Bretons and some Welsh ancient high lineages seem sourced from ancient Stratclud, some of these last ones qualified of ‘Corneu’ ; all that seems to support as well small differences as a relatively great unity between P-Celtics of old Britain after Roman times. Maybe through the great influence of first Cornovii who gave a lot of auxilliaries to Roman troops here and there between North and South before Rome withdrawal and whose elite apparently integrated other high lineages of North before their retreat southwards in today small Cornwall and Brittany, if we rely on genealogies.
@Aram
The Empuries Greeks are very similar to Mycenaeans and they do have steppe ancestry.
MH_82
The usually accepted theory is that the main substrate in Armenian comes from Hurro-Urartian linguistic group. Greppin expressed doubts on this suggesting that Urartian substrate in Armenian is exaggerated but as far as I know his opinion is not widely accepted.
As for the substrate in proto-Iranian. They might have it. But the name of that language is unknown. For example the word for camel in Iranian derived from *uštra is usually proposed to be such example.
Similar examples are attested in Akkadian udru and Urartian ultu from which Armenian ult meaning camel.
But as usual there are other opinions also, like the *ustra is native IIr term.
https://iranicaonline.org/articles/camel-sotor
Davidski
If I recall correctly one of Empuries Greek sample didn't have Steppe. The I8208
The correct link
https://adnaera.com/wp-content/uploads/2019/03/empuries2.png
The beauty of Loulan looks ambiguous. Could be European but could just as well be Siberian/Amerindian, judging by her features.
Maybe R1b1c predates the formation of the Steppe genetic signature by several millenia and is better seen as an ANE associated marker.
EHG and ANF are connected through WHG.
@yamz
What is the likelihood that BMAC actually contributed genetically to South Asians (yes even the North-west South Asia so heavily fetishized on anthrogenica lol) in any capacity?"
Swat adna definitely has bmac like ancestry on top of swat local and steppe. Now this bmac like ancestry could be local (survived in pockets nearby) or even SiSBA1 like.
No such ancestry is detected yet in modern indians. Where we also have the added problem for qpAdm modeling that we lack any proximal source database.
Modeling modern indians with shahr outliers and onge is already taking a lot of liberties, and I do not rely on g25 for more than telling me about broad affinities of the target.
The Native Anericans broke off with the Eastern Eurasian population first at 40,000 ybp, and then with the ANE population last around 25,000 you and into the Bering Strait and into the Americas.
Post a Comment