search this blog

Thursday, May 16, 2019

Fresh off the sledge


As things stand, the closest individual to a Proto-Uralic speaker in the ancient DNA record is arguably OLS10 from an Iron Age tarand grave in what is now Estonia. I say that because:

- isotopic data suggest that OLS10 wasn't born where he died, and considering his elevated Siberian ancestry relative to earlier and most contemporaneous Baltic ancients, he was very likely a migrant to the Baltic region from the east

- the tarand grave tradition appears to be specifically a Finnic (west Uralic) phenomenon that probably spread from the Volga-Oka region, which is just west of where most people place the Proto-Uralic homeland

- OLS10 belongs to Y-chromosome haplogroup N-L1026, a paternal marker that is especially closely associated with Uralic-speaking populations and probably only appeared in the East Baltic region during the transition from the Bronze Age to the Iron Age

You can find more background info about OLS10 and other relevant samples in Saag et al. 2019 (see here). This is where he sits in my Principal Component Analyses (PCA) focusing on fine scale Northern European genetic diversity. The relevant datasheets are available here and here, respectively.



Note that OLS10 doesn't cluster strongly with any ancient or modern populations. To investigate this in more detail I ran a series of two-way qpAdm analyses, testing tens of ancient individuals and populations as potential admixture sources. These two models stood out above the rest in terms of their statistical fits, chronology and overall plausibility.

Baltic_EST_IA_0LS10
Baltic_EST_BA 0.826±0.045
RUS_Sintashta_MLBA_o1 0.174±0.045

chisq 12.527
tail prob 0.564048
Full output

Baltic_EST_IA_0LS10
Baltic_EST_BA 0.683±0.102
RUS_Mezhovskaya 0.317±0.102

chisq 13.811
tail prob 0.463864
Full output

Please note that RUS_Sintashta_MLBA_o1 isn't representative of the Sintashta culture population as a whole. It's a group of the most extreme genetic outliers among the Sintashta samples, and they may or may not have been Uralic speakers (see here). Interestingly, the Mezhovskaya culture population is generally associated with the Ugric branch of the Uralic language family.

I was also able to closely replicate these results with the Global25/nMonte method; down to almost one per cent. However, the statistical fits (distances) are poor, probably because the reference populations aren't the real mixture sources. This is in line with the fact that their Y-haplogroups are Q1a, R1a and R1b, rather than any type of N.

Baltic_EST_IA:0LS10
Baltic_EST_BA,83.8
RUS_Sintashta_MLBA_o1,16.2

distance%=4.7955

Baltic_EST_IA:0LS10
Baltic_EST_BA,69.8
RUS_Mezhovskaya,30.2

distance%=3.5783

I do realize that two Bronze Age samples from Bolshoy Oleni Ostrov, Kola Peninsula, belong to N-L1026, but adding them to my mixture models doesn't help. Little wonder, because the Kola Peninsula lies within the Arctic Circle, and I'm pretty sure that OLS10 and his N-L1026 came from somewhere just north of the mixture cline marked on the map below. Unfortunately, I can't test this directly yet due to the scarcity of ancient samples from this region.


See also...

It was always going to be this way

On the association between Uralic expansions and Y-haplogroup N

Uralic-specific genome-wide ancestry did make a signifcant impact in the East Baltic

142 comments:

Slumbery said...

@Davidski

Thank you, this a well written, measured post.

I would have never guessed Sintashta_O1, because that is a sample similar to Steppe Maykop. A mixture of WSHG + Yamnaya + some southern ancestry. And it is indeed a big outlier in Sintashta, in fact I do not think it has any Sintashta related ancestry or mixture at all.

Mezhovskaya is a big contrast, because it had Sintashta related / Sintashta-like ancestry and also had BHG admixture + like a tenth of the WSHG-like ancestry compared to S_O1, so it is stark different from Sintashta_O1.

The fact that the two best fitting populations in a two-way model are this different suggests that at least one of them is a complete miss.

Shaikorth said...

Srubnaya_outlier is similar to Sintashta_MLBA_o1 on average but there is considerable variation between the three Sintashta_o1 samples. One of the three has clear Sintashta ancestry.
For o1 and Srubnaya_o comparisons see figures S3.22 (both panels) and table S3.64 of Narasimhan et al. Srubnaya_o might actually represent the source as well as o1 average.

Davidski said...

Srubnaya_o produces by far the best qpAdm statistical fit for OLS10, with a tail prob of something like 0.88, but the analysis is based on much less than 100K SNPs, so it's difficult to take it seriously.

Huck Finn said...

Many thanks, very interesting reading. I wonder what makes the models prefer Mezhovskaya, in terms of quantitity, compared to Sintashta_o1?

Slumbery said...

It is probably a combination of:
-Mezhovskaya has much more last-period European farmer ancestry than S_O1. Since the test pop also have it, it can be only sourced from the local BA refetence. This is probably the main reason.
- Mezhovskaya had some EHG and BHG in it, the test population also did, but S_O1 did not. I am unsure of this effect on the quantity however. If anything this should worsen the fit instead. So mainly the first reason is behind the quatity difference.

Huck Finn said...

@ Slumbery: good points, thank you.

Kristiina said...

How about V10, where would he be on PC map? If I remember correctly, he is also a foreigner and carries more Siberian admixture than OLS10.

Huck Finn said...

Good question, Kristiina. If I recall it right these two gents, the other being N1c and the other R1a, are from the same Kunda tarand complex.

Samuel Andrews said...

Estonia_IA Finns are mostly of local Baltic origin. Uralic experts here seem to say Finnic language arrived in Finland via Baltic sea. If so, safe to assume early Finns had lots of Baltic ancestry. Which means they have recent common ancestors with Balts & Slavs.

"Balto-Slavic" drift is picked up in Finns in G25 PCA. I've noticed they get much better fits when Balts are used.

Samuel Andrews said...

3.4726"

Baltic_EST_IA:0LS10_1

Baltic_EST_BA,64.7
CWC_Czech,14.2
Bolshoy_Oleni_Ostrov,10.2
Unetice,7.9
EHG,3
Srubnaya,0
West_Siberia_N,0
Lokomotiv,0
Nganassan,0
Sintashta_MLBA_o1,0

Arza said...

Baltic_EST_IA:0LS10_1
Baltic_EST_BA 73%
RUS_Karasuk:RISE496 27%
Distance 3.1662%

Baltic_EST_IA:0LS10_1
Baltic_EST_BA 67.4%
RUS_Karasuk:RISE496 27.8%
POL_BKG_N_o1:N22 4.8%
Distance 2.9453%

Baltic_EST_IA:0LS10_1
Baltic_EST_BA 60.2%
RUS_Karasuk:RISE496 21.2%
CZE_Early_Slav 7.8%
FIN_Levanluhta_IA 5.4%
POL_BKG_N_o1:N22 5.4%
FIN_Levanluhta_IA_o 0%
Distance 2.8322%

Davidski said...

@Kristiina & Huck Finn

OLS10 comes out clearly more eastern than V10 in all of my PCA. For instance...

G25 East Baltic BA-IA transition2

But this doesn't contradict the result in the paper, where OLS10 is reported as having 6% Nganasan-related ancestry and V10 8%.

The 2% difference may or may not be real depending on the standard errors, and, in any case, the PCA are picking up all types of eastern ancestry, including I'd say Siberian ancestry not accounted for in the Nganasan-related ancestry proportion.

Both OLS10 and V10 might have still been migrants from the Volga-Oka region, but based on what I've seen, OLS10 shows more Uralic-specific ancestry.

Kristiina said...

Karasuk is probably just a proxy for a true Uralic N rich population because the available Karasuk yDNA is Q-M25, R1a1-PF6206 and R1a1-YP349. I do not think that anybody argues that Proto-Uralic was spoken in the Karasuk culture.

EastPole said...

@Arza
“Baltic_EST_IA:0LS10_1
Baltic_EST_BA 60.2%
RUS_Karasuk:RISE496 21.2%
CZE_Early_Slav 7.8%
FIN_Levanluhta_IA 5.4%
POL_BKG_N_o1:N22 5.4%”

PC1-PC3 plot is also very interesting:

https://i.postimg.cc/G349StVc/screenshot-498.png

PC3 explains 7.2691% of the variance, so it is significant. Poles are gold on the plot.

M. Myllylä said...

@Samuel Andrews,

In Europe the Finns are genetically between Balts and Scandinavians, Estonians are almost fully Baltic. It is only common sense to think that two relatives separated by a sea have somewhat different ancestry and history. But there is a minor Asisn admixture too. We need Eurasian PCA to reveal it, But then, who were original Finnic speakers, did they live in Estonia or in Finland? Did they carry N1c1 haplogroup? We can say that N1c1 in Estonia and Baltic countries is not older than in Finland, and not even older than in Sweden. But even this proves nothing, because we don't know all routes of westward N1 expansions. Connecting the whole N1(c1) to the Baltic Finnic language branch is too ambiguous. Connecting L550 leads to another problem; why it is not unambiguously ancestral in Estonia? Archeology doesn't give unambiguous answers either. Tarand graves are only one local grave type among others in Estonia and Finland. Even if we think them, we have meet some problems. All oldest tarand graves are on the seaside and in islands nearest Sweden, Gotland and Finland. The later history proves about Scandinavian migration to east through the Volga water routes, so if we find something common with (humorously speaking) between Estonian Vikings and people of Volga Oka or Kama, does it prove the origin?

Andrzejewski said...

The Rurikid dynasty was N1C1.

M. Myllylä said...

@Andrzejewski,

Indeed, Rurikids were cosmopolitans, their roots are common in Sweden and Finland.

Shaikorth said...

Turns out that Baltic_EST_IA:0LS10_1 can also be modeled well as roughly 3/4 Baltic_BA 1/4 Maitan_MLBA_Alakul_o in qpAdm.
The latter is of course from Kazakhstan, far to the east of Sintashta outliers, so why does this fit work so well?

The answer may be in Narasimhan et al. preprint, the populations that have provided the best values (Srubnaya and Sintashta outliers as well as the BA Alakul outliers) are all part of the same cline of deep ancestry visible in figure S3.22:
https://i.imgur.com/2NlfnRM.jpg

Anthony Hanken said...

@M.Myllylä

Early Estonian Tarands are not really a local grave type. More of a western variant of older Volga grave types. They arrive with other eastern material culture around the same time.

This is why Tarands, N1c and Finnic is being connected to the Volga.
Linguists and archeologists like Parpola and Lang seem to agree that the language and culture came from the Volga during the Iron Age. Right now we lack aDNA from BA-IA Russian froest-steppe but N1c being found in Tarands supports this model.

N-L550 Tarand people seem to have been seafarers. Even though this branch is not the most common amoung modern Finns or Estonians they may have been able to exert their language due to Scandinavian and Baltic trade connections inland groups lacked.

Anthony Hanken said...

Another thing of note,

These Tarand samples are all coastal and propably spread to Estonia,Finland and Sweden rapidly. Not necessarily developing in Estonia and then spreading north and west.

Later burials to the south in Estonia may reveal more diverse N-VL29+ lineages.

M. Myllylä said...

@Anthony Hanken'

Parpola is not a geneticist. Has he really written that N1c1 can be connected to the root of Finnic languages? I wonder...

Although L550 is not the most common N1c1 line in Finland, it is very old in Sweden and Finland compared to the Baltic area. It is question about populational dynamics; the most common line can be the youngest one due to founder effects, wars, local famines, plagues etc. Today the most common North American HG is probably R1b, over 500 years ago it was not. And the American English then. But we know the answer because we have written history. These questions are really difficult in case we try to solve PREHISTORIC questions.

Andrzejewski said...

I'm fascinated by who the Saami are, and I'm not talking about the Finnic N1C1 (2/3 of them have that Y-DNA); I'm referring to where did the substantial non-Uralic substrate in Saami come from.

M. Myllylä said...

@Anthony Hanken

"These Tarand samples are all coastal and propably spread to Estonia,Finland and Sweden rapidly. Not necessarily developing in Estonia and then spreading north and west."

Another possibility is that Tarand grave culture came from Sweden. It would explain why many of the oldest ones are found from Saaremaa (western island). Soil in Saaremaa was quite poor and unsuitable for agricultural use. Best soil was in central and eastern Estonia. Estonians colonized Saaremaa very late, as a last corner of the country after they learn to be seafarers from "Vikings".

Huck Finn said...

It seems to me the only real point M. Myllylä has in this discussion is the lack of N1c in the ancient genetic material of NW Russia, that's available. Otherwise, fex both linguistic and archeological material points to a clear Kama-Volgaic -> Baltic contact network. Besides, it is rather frustrating to start discuss Viking Age connections based on late Bronze Age samples such as those of Kunda ie. OLS_10 and such. That being said, we have ancient N1c now not just from Kola Peninsula, but also from Estonia, Sargat kurgans of the Ural area and from the Hungarian elite burials. I look forward to hear about the model which explains the route from Kola to Hungary, for instance.

Andrzejewski said...

I got this one but I can "fish" for a full Hakkinen thesis on the subject:

https://en.wikipedia.org/wiki/Pre-Finno-Ugric_substrate

M. Myllylä said...

@Huck Finn,

I am grateful that I have that one point.

My mentioning about Vikings was in all contexts in quotes, answers to comments regarding 200-1000 years old haplotype frequencies in Finland.

Huck Finn said...

@ M.M.: so you're not seriously suggesting that Sintashta_o1 or Mezhovskaya type of features of OLS10 are based on Viking Age migrations? And yes, besides the absence of N1c in the BA/IA samples of NW Russia ex Kola you're not making a lot of sense.

Andrzejewski said...

From the Wikipedia entry: "In population genetics, Pre-Finno-Ugric substrate in Sami languages correlates with the spread of Y-DNA haplogroup I1 and mtDNA haplogroups H1 and U5b1b.[3]

Some examples of Kildin Sami words without convincing Uralic/Finno-Ugric (or any other) etymologies:[9]

Kildin Sami English
kut’t’k heart
nirr cheek
čad’z’ water
vuntas sand
poav’n hummock
k’ed’d’k stone
abbr’ rain
piŋŋk wind
ket’t’k’ wolverine
nigkeš pike (fish)
murr tree
cigk mist
mun frost
pin’ne to herd, to look after
čujke to ski
puaz reindeer
koan’n’t wild reindeer
luhpel’ 1 y.o. reindeer
sejjd deity
kipp’tε to cook
kuras empty
modžes beautiful
n’učke to jump
čacke to throw
tuллtε to boil
kuarktε to boast
лujx’ke to cry
nissε to kiss
madt trouble
tommtε to recognize
aps smell"

________________________________


A remnant of the WHG language with Y-Hap I1 in Northern Europe. Wow! (Although the entry says that it has NOTHING to do in comparison to the substrate in Germanic languages).

M. Myllylä said...

@Andrzejewski,

I read the Wiki article:

"In population genetics, Pre-Finno-Ugric substrate in Sami languages correlates with the spread of Y-DNA haplogroup I1".

I am not sure what this means in practice, though. Was the Pre-Finno-Ugric home land in Sweden, because the frequency of I1 is highest in Sweden, especially in Gotland?

M. Myllylä said...

A German philosopher said once that it is sometimes better to say nothing. But folks, please read my blog, I have always made it seriously.

Anthony Hanken said...

@M. Myllylä

I never called Parpola a geneticist, aDNA from Tarands will however support his model once we get samples from D'yakovo or Akozino-Akhmylovo.

I'm not sure why you would think Tarands come from Sweden. Seems to be going against what actual achealogists like Valter Lang believe. Houses of the Dead in the Volga are older than those found in Estonia or Sweden.

Go back and look at the discussion of BOO on this blog. It looks like a recently mixed group with one population coming from the south and the other from the east.

Just wait for samples from the massive blackhole we have in BA-IA west Russia before you assume all European N is decended from BOO.

Andrzejewski said...

@M Myllylä I don't know but it was ruled out that the Germanic substrate would be similar to the Pre-Uralic one...so I don't know WHAT it was...maybe the Germanic one came from Farmers instead (*IF* there was ANY substrate in Germanic at all).

Tesmos said...

Davidski, are you still going to post about the Nordic Bronze Age soon?

M. Myllylä said...

Everyone, read the text behind this link and put dates into perspective of supposed N1c1 frequencies, Finnic home lands and different burial types in Sweden, West Estonia and Volga Oka. The time gap between burials seems to be smaller enough to be explained by randomness of archaeological finds. Nevertheless, after searching more info, it convinced me that there is no archaeological consensus about mortuary houses and Tarand graves.

https://books.google.fi/books?id=r-HuwSrQjvQC&pg=PA120&lpg=PA120&dq=volga-oka+mortuary+houses+age&source=bl&ots=YAATeAPjKx&sig=ACfU3U2y-UavnGA0fzf2b1MopDFTUxjAJA&hl=fi&sa=X&ved=2ahUKEwj3v_fWnKPiAhVSAxAIHX0GAzUQ6AEwAHoECAMQAQ#v=onepage&q=volga-oka%20mortuary%20houses%20age&f=false

capra internetensis said...

Might be worth trying Krasnoyarsk MLBA outlier I6717 for Estonia IA, though he may have too much Baikal HG.

In G25 I6717 is fairly Karasuk-like, but seems particularly close to Bolshoy Oleni Ostrov. IIUC where it comes from on the upper Yenisei is a mixed Fedorovo-Karasuk area, and the main Krasnoyarsk MLBA are Sintashta-like.

Actually in G25 you can model the shit out of all kinds of northern Eurasians with BOO or Krasnoyarsk O. Might be something close to original Arctic EBA and/or N-L1026 people. I6717 is Q1a2-M346(xQ1a2a-L53), though, not N.

Parpola suggests that ancestral Samoyeds are associated with the Cherkaskul culture, which seems to be mostly near the Urals, but according to some Russian archaeologists extended all the way to the upper Yenisei. He also connects it to Mezhovskaya.

Davidski said...

@Tesmos

Davidski, are you still going to post about the Nordic Bronze Age soon?

Yes but not imminently.

Davidski said...

@M. Myllylä

The theory that tarand graves spread into the Baltic region from the east along with people speaking Proto-Finnic has now been very strongly corroborated by ancient DNA data, so if there wasn't a consensus about their origin then there is now.

I'm sure that once you get your head around all of the relevant genetic, archeological and linguistic data from the Mesolithic to the Middle Ages, you'll realize that your opposition to the mainstream theories about the origins and expansions of Uralic and Finnic languages is a lost cause.

Davidski said...

@All

By the way, one point that I should have also made in the blog entry is that OLS10 is unlikely to have come from just east of Estonia, because the Saag paper had Iron Age samples from Ingria, and they were surprisingly western in comparison to the Estonian Iron Age samples, and especially to OLS10.

Gabriel said...

And what is the ultimate source of European ancestry in Finns? And do Uralics in general share similar European ancestry?

Davidski said...

@Gabriel

And what is the ultimate source of European ancestry in Finns? And do Uralics in general share similar European ancestry?

At the most basic level, hunter-gatherers from across Northern and Eastern Europe, Late Neolithic farmers from across Northern and Eastern Europe, and Bronze Age pastoralists from the Pontic-Caspian steppe.

Most other Uralic speakers share these ancestries, but from somewhat different proximate sources.

For instance, Finns have a lot of Baltic and Scandinavian influence, while Uralic speakers from deep in Siberia have practically none.

FrankN said...

Andrzej: Thx for pointing to that Wikipedia article - I had been looking for a list of pre-Saammi substrate for some time.

The posted wordlist has very little in common with the pre-Germanic substrate, except for:
k’ed’d’k "stone" to tentatively be linked to German Kies "gravel, flintstone, in mining: non-ore bearing rock" [d->t->z is a well known (high) German sound shift, c.f. PIE *dn̥ǵʰwéh₂s ->PGerm *tungǭ ->German Zunge "tongue". Kluge connects Kies to Lith. ziedra "gravel, grain" and proposes an otherwise unattested PIE *gei- as shared root, even though the endings obviously don't match. This etymology fulfils my personal criteria (inconsistent structure, lack of widespread IE parallels) to show that we are actually dealing with a substrate term - in this case possibly shared by Saami, German(ic) and Baltic.

I am certain that Germanic contains several kinds of non-IE sub- and adstrates: A good third of proto-Germanic entries in the Swadesh 100 list lack convincing (P)IE parallels/ etymologies, including terms like “hand”, “sing”, ”blood”, “breast”, “drink”. As concerns "hand" (*kant prior to the proto-Germanic sound shifts), P. Shrijver connects it to Proto-Finno-Ugric *kati "hand", and proposes that both families ultimately borrowed the term from a lost Fennoscandian language he dubs as "Language of Geminates". Another shared root cited by him is Saami *kuti "fish roe", which he connects to Germanic terms like cunt or MLG kut "entrails, roe" [German Kutteln "tripes (cooked testicles)" should also belong here.]
https://www.academia.edu/38396354/Lost_languages_in_northern_Europe.pdf

However, that language, which has according to P. Shryver also left phonetic trails in Saami as in N. Germanic and even Scotch Gaelic (<-Pictish?) cannot have been the only source of Finno-Ugric/ Saami substrate - otherwise we should find much more parallels between Saami and Germanic substrate.

The source language of “sing”, ”blood”, “breast”, “drink” apparently didn't have an issue with intra-syllabic consonant clusters. To the opposite: What a/o sets apart Old Norse from Proto-Germanic is heavy syncopation, i.e. loss of unstressed vowels, producing consonant clusters(for an example, compare original English "do not" to syncopated "don't"). The most parsimonous explanation of this phenomenon is Old Norse being strongly influenced by a heavily syncopating substrate, which in all likelyhood also supplied terms like “sing”, ”blood”, “breast”, “drink” to Germanic. I believe we are dealing here with post-Pitted Ware (which displays many similarities to Scottish Grooved Ware), and ultimately post-Ertebölle horizons.

The opposite of syncopation is polnoglasie or full vocalization, i.e. entering extra vowels to break up consonant clusters. The phenomenon is best documented in, and characteristic of East Slavic, c.f. OCS mlěko->Russ. molokó "milk", OCS gradъ->Russ. górod "city". Its most parsimonious explanation is language switching from a source language that prohibits intra-syllabic consonant clusters. In the case of East Slavic, that source should most likely have been Uralic [Altaic also prohibits intra-syllabic consonant clusters, but tolerates inter-syllabic ones, so Altaic speakers would have been satisfied with turning OCS mlěko into something like molkó.]


Against the background of Uralic "resistance" against consonant clusters, I find it intriguing how much consonant clustering is present in the pre-Uralic Saami substrate. Apparently, while Saami were able to implant their language across N. Fennoscandia, pre-Saami phonology and phonotactics survived and ultimately won over Uralic. I haven't yet assessed the process from a DNA perspective, but it surely could provide for a nice case study.

Gabriel said...

What I meant is, do Uralics share common Bronze Age European ancestry from a specific source? If they don’t, does this mean proto-Uralics were lacking in European ancestry, and Finns acquired it as they moved west? Do any have possible Sintashta-related influence?

Davidski said...

@Gabriel

Populations associated with the Volosovo, Garino-Bor and other southern Urals cultures that probably spoke Proto-Uralic and early Uralic languages had to have been largely of European forager ancestry, because they were located in the forest zone west of the Urals. So this type of European ancestry associated with the earliest Uralic speakers should be found in practically all present-day Uralic speakers.

Also, as the early Uralic populations expanded into areas occupied by people derived from Yamnaya, Poltavka, Corded Ware, Sintashta, Srubnaya and Andronovo groups, they would have acquired and subsequently spread steppe and farmer ancestry by mixing with these peoples, just like they mixed with Balts and Scandinavians when they reached the Baltic region.

There's also bound to be minor BMAC ancestry in some Uralic groups, mediated via late Andronovo and Srubnaya populations moving from Central Asia back into the forest steppes near the Urals.

FrankN said...

Btw, the reason I had been looking for a list of Pre-Saami substrate was entertaining the idea of a maritime movement along the Arctic Sea to explain documented (e.g. Jaeger 2015), but so far unexplained parallels between IE, especially Germanic and (Insular) Celtic, and Chukotko-Kamchatkan. In fact, there are quite a number of shared maritime HG terms, e.g. lax (salmon) that can be traced from Germanic via Balto-Slavic, Finnic, a couple of Chukotko-Kamchatkan and NW American languages to ultimately Sundanese lauk "sweetwater fish". Similar chains can be constructed for "seal", "whale", or "cock" (c.f. Greenberg's proposed Amerindian *cac "large bird").

Then, however, I came across S. Nikolaev's recent reconstructions of Proto-Wakashan-Nivkh-Algic (see https://en.wikipedia.org/wiki/Algic_languages for respective links). If correct, these reconstructions would not only - importantly enough - present a second case, after Yenissean-Na Dene, of a trans-Pacific language family, they could also provoke the need to fundamentally review everything we have so far assumed about the origin of PIE.

Essentially, S. Nikolaev's reconstructions point towards Nivkh-Algic not only sharing isoglosses with Germanic and Insular Celtic, as possibly explainable by maritime contacts along the Arctic Sea, but with PIE as a whole. Not just a few isoglosses, but in fact more Swadesh 100 (and beyond) isoglosses than PIE has in common with Proto-Uralic. To highlight just one example: "Old Europe" mentionned in a comment to the last post PIE *weik "settlement", attested a/o in Indic, Latin and Baltic. S. Nikolaev reconstructs a more-less identical Nivkh-Algic root meaning "dwelling" a/o attested in Algonquin wig-wam "tent".

Ebizur said...

That sort of long-range comparison of short, phonologically simple morphemes cherry-picked from among a great number of languages is not statistically robust.

One might just as well compare those words for "dwelling," "settlement," etc. (one must be wary of granting too much semantic leeway, too, as is often done in such long-range comparisons) to Chinese 屋 "house; room; building, shelter" (Mandarin , Cantonese uk1, Hakka vuk ~ vug5, Minnan ok ~ og4, Sino-Vietnamese ốc, Sino-Korean ok, Sino-Japanese woku > oku).

M. Myllylä said...

@Davidski

"The theory that tarand graves spread into the Baltic region from the east along with people speaking Proto-Finnic has now been very strongly corroborated by ancient DNA data, so if there wasn't a consensus about their origin then there is now."

I am very disappointed that I can't see the aDna connection, despite of feeling my head decent. The problem in many tests using modern and ancient dna is in understanding of the difference between Asian/Siberian gene pools. Tests fail because so many Siberian groups went through founder effects and became distorted in test results. Another problem is that we, and even practiced researchers do not fully understand how algorithms work. This has been true from the first studies, for example Nelis et al. 2009. I am erring too, but still I have permission to observe things being odd and illogical. Sure there are things I don't know, but there are also things which can be seen peculiar just on the base of basic Finnish history. But most of us are not aware of this side.

Dragos said...

Looks like it’s game over for the Steppe fan boys

“”Leaked News:
Romans had predominantly Hellenistic Greek origin and Etruscans were admixture of Italian Bell Beakers + Italian Mid Neolithic.
Unfortunately I can't reveal my source because they are probably reading this forum, but I can assure you it's reliable.”

Of course; Romans predate “Hellenistic” ; so what they mean is Balkan type ancestry

Arza said...

@ Dragos

Heh, nope. If they are like Greeks they will show the same post-CWC Central European ancestry as Mycenaeans in Lazaridis et al.

We are heading towards more realistic scenario of the spread of IE languages. Of course it won't be liked by many, because most of IE branches will come out as a result of secondary expansions from Central Europe.

Dragos said...

Arza
Sorry buddy
Too little too late
There’ll be no R1a in Mycenaeans
There was a secondary expansion from Carpathian - Balkan area to north

Davidski said...

The Romans in that PCA obviously have steppe ancestry. Some have quite a bit more than the Mycenaeans.

But many also look like they have recent Middle Eastern and perhaps North African ancestry, which is what is pulling them so far south.

Etruscans just look like a two-way mix of early farmers and Bell Beakers, plus maybe some Roman input.

Nothing surprising.

Davidski said...

And there will be many more samples in the paper. Hehe.

Dragos said...

Ah yes, the old Iron Age African migration to the Appenines
Hannibal’s grandfather ?

Gaska said...

This is every day more fun-

But what about Latin?, because it is more related to Celtic than Greek

Let's see what sites have analyzed, if they are truly Etruscan-r1b and descended from the Italian Bbs, then finally the BB culture did not speak IE. And the consequences can be devastating for the origin of that languagein the steppes.

It is only possible to understand a much later expansion of IE languages ​​in Western Europe

Davidski said...

@Dragos

Remember that Roman outlier from Britain? Well, expect more of where that came from.

Leron said...

Gaska: Many believe there’s an IE substrate lying underneath the Etruscan language, or that it originated as a split before proto-IE.

Andrzejewski said...

So Etruscan might be another Non-IE B.B. language, after Basque

Matt said...

The same source also stated:

"Picentes, Samnites and Umbri were very close to Etruscans. This makes things even more complicated. but I think best explanation is that proto-Italic was probably language of Italian bell-beakers." (e.g. predominant Iron Age groups of Italic speakers in Central Italy, east of the range of early Latin speakers; no / little additional CHG).

Seems that some Italic speaking populations had admixture from Anatolia MLBA / Minoan like populations, others did not.

Greek speakers per se, (presuming the aforementioned were not) may not be relevant at all? Doesn't seem likely to me that the Romans as a whole were Greeks who forgot that they were ("predominantly of Greek origin"), rather more likely that they had a lesser degree of ancestry from a more concentrated CHG source than Mycenaeans...

Though this all depends on when and where the "Romans" were from however. The city may have drawn in some fairly different populations, even early on.

Davidski said...

Yeah, as you'll see, Picentes, Samnites, Umbri and Etruscans will together sit along an earlier cline of EBA Italian samples.

I don't know how to interpret that.

Leron said...

Matt: What you mentioned parallels what the ancients said about “Pelasgian cities” in Crete, Greece, Italy and Spain. In preparation of the paper’s release I’m going to read up on the Italian Pelasgian cities and later see if there’s a higher proportion of LMBA Anatolian and CHG in those areas.

Samuel Andrews said...

Sounds like Etruscans cluster with modern northern Italy (Beaker+Italy farmer). If, so Etruscans will be just one more example that Indo Europeans sometimes adopted language of locals. Even if you think Beaker R1b U152 didn't speak IE but spoke Basque-Iberian, Etruscan is a different language than that too.

IMO, the Romans cluster near modern central/south Italians who technically cluster near Myceneans (but it is a pseudo-closeness) But hopefully they get early Latin samples not samples from the city of Rome (which could be of mixed origin).

Samuel Andrews said...

Looking at Italians' position in G25 PCA, and at Y DNA (E-V13, J2b), I said they have direct Balkan ancestry not just Anatolian/NEar Eastern.

Matt said...

@Sam, was going to say something similar re; "Even if you think Beaker R1b U152 didn't speak IE but spoke Basque-Iberian, Etruscan is a different language than that too." (though at more than twice the length and not saying much that anyone already familiar with that discussion didn't know ;) ).

(Of course, if we ended up admitting that there was any switching at all, the basis for thinking "R1b U152" (in Central Europe) "didn't speak IE but spoke Basque-Iberian" seems a little sketchy).

Well, either that or.... "There was no switching! Etrusco-Vasco-Iberian was all originally one Bell Beaker language!" (it's not like it's hugely more or less established than Vasco-Iberian as far as I can tell) or should there be any non-R1b-M269 found at all (even 10-20%) - "Yes, there was some switching so there are vastly different dynamics in Northern Italy, see the clear archaeological evidence of XYZ! So there definitely wasn't switching in Iberia/SW France". Cynically I expect to see both...

Samuel Andrews said...

Tuscans have 25% Yamnaya ancestry. Southern Italians have 20%. Sicilians have 15%.

There's a very low chance any historical incursions from northern Europe made an impact on central-southern Italy.

Therefore, all of this Yamnaya ancestry was in place in Italy in the Iron age. I also, guarantee not all of it is from Bell Beaker R1b U152+ descendants but that some of it is from the Balkans.

Dragos said...

Davidski
Yes I think during Roman Empire period there’s be exotic ancestry from all around the Mediterranean
So it depends when the samples date to

Gaska said...

sam what percentage of yamnaya ancestry we have the Basques and the rest of Spaniards?

If U152 did not speak Italo Celtic, where did Latin come from? It can not come from the Greeks, can it?

The theory of R1b hijacking Neolithic languages ​​throughout Europe does not seem serious to me

Do you know what I'm thinking knowing what happened in Spain and probably in Italy?

Gaska said...

Truly the Italians are a very interesting people, much more than the Basques

Simon_W said...

Very exciting news there, I didn't read the original posts on Anthrogenica, but I suppose it's from that upcoming paper. The info that the Umbri, Sabines and Picentes were - just like the Etruscans - like Italian Beaker + Italian Neolithic clearly contradicts the interpretation that the Italics were South Italian-like folks from the Balkans. The exact time and place of these "Hellenistic"-like Romans will be crucial… But at least it's an established fact that Latin and Faliscan were a different branch of Italic than Umbrian, Sabine and Picene, so there may well be room for genetic differences. And at least it seems more and more likely that the Etruscans didn't have an Aegean or Anatolian origin, which is great to know.

You see, there certainly is some Balkan influence in prehistoric (eastern-)central/southern Italy it's archaeologically attested. But as far as I see it, its influence wasn't very Anatolia_BA-rich. Because:


distance%=2.4111

HRV_MBA

Barcin_N, 60.6
Yamnaya_Samara, 33.2
WHG, 6.2
Anatolia_EBA, 0

distance%=3.981

HRV_Early_IA

Barcin_N, 1.5
Yamnaya_Samara, 34.8
WHG, 3.7
Anatolia_EBA, 0

But, distance%=5.2989

Beaker_Sicily_no_steppe

Barcin_N, 62.4
Anatolia_EBA_Ovaoren, 19.6
Anatolia_EBA_Isparta, 8.2
Yamnaya_Samara, 5.5
WHG, 3.8
Ganj_Dareh_N, 0.5

Anatolia_BA admixture seems to have affected Sicily, and by inference also Southern Italy, early on and in considerable measures. But all the Balkan influence from Dalmatia seems to have lacked it, at least in a significant degree.

Gaska said...

Saluti Amici

Samuel Andrews said...

@Davidski,
"Remember that Roman outlier from Britain? Well, expect more of where that came from."

I doubt that, considering southern Italians are so similar to Greek Islanders/ancient western Anatolia. Looks like Italians have ancestry from Anatolia.

Simon_W said...

Yeah I think Southern Italy and Sicily got it from Anatolia/Greece and maybe also from Bronze Age Cyprus, which is still terra incognita in ancient DNA terms. But probably Anatolia_BA-like with possibly some additional Levantine affinity. And it looks logical that central Italy, to some degree is an admixture between northern and southern Italy. It seems unlikely that they acquired their Anatolia_BA-related admixture in an entirely different manner than the South.

Bob Floy said...

@Sam

"Looks like Italians have ancestry from Anatolia."

What kind of ancestry do you have in mind? I assume you mean something other than the usual EEF type stuff.

Matt said...

Witty as ever Drobbo.

natsunoame said...

What exactly means Greek or Hellenic DNA !? In the light of official historical records including Herodotus the Greeks came from African continent. So, what is this DNA you are pointing so often like Greek?! May be some autochthonous one which obviously isn’t Greek at all. Someone to have any clue!? I like myths and legends too but this pretend to be scientific blog or no?

Simon_W said...

@Rob Floy

Lol, of course they mean something Anatolia_BA-like, something with significant additional CHG/Iran-related input, not the original Barcin-like ANF/EEF Farmers.

Bob Floy said...

@Simon

Yeah, I was hoping that someone would go out on a limb and try to(speculatively) narrow that down a bit.

Simon_W said...

@natsunoame

Lol, what a crazy stuff you're posting there. I have seen this sort of theory before though. But there's no evidence in the autosomal DNA of ancient Greeks (Mycenaean, Empuries2) for an African origin of any sorts. At least not any more recent than OOA.

Simon_W said...

BTW, something other off-topic that I noticed today: The position of Hallstatt_Bylany:DA111 in the North European PCA is really amazing. Between the Irish and the French and very far from the early Slavic Bohemian (and from the modern Czech). That's admittedly not what I had expected prior to the publication of the Hallstatt Bylany samples. I had predicted here that they were going to be more eastern, in spite of their Celtic cultural affinity. The other Hallstatt_Bylany sample is indeed more eastern, but he has Scythian-like steppe admixture. In any case this all confirms the impressive parallels between ethnic affinity and the position in that PCA. The ancient Celts were like the modern Irish in the North and like the French in the South, but not like modern Germans or Czechs!

natsunoame said...

I like fairy tales too but really prefer scientific literature...
Anyone familiar with the works of the ancient chroniclers knows very well that Greece has been inhabited by other peoples long before the Hellenes appeared in Europe. Herodotus [3] II-56, Strabo [4] V-2.4, VII. 7.1, Pavzini [5] VIII-1.4, Pliny [6] IV.III. 8-IV, IV. V. 19-vi, IV, VII. 27-viii-29, Efor and Hesiod (cited by Strabo) [4] V-2.4 are unanimous on this issue. The Thracians and Pelasgians are the most ancient settlers of Greece. For the Greeks, the old authors say they came from Africa. Herodotus [3] VI-53, Strabon [4] VII-7- 1 and Pliny [6] VII-LVI-195 explain this clearly.
So Mister Simon can you argue with FACTS and written history?

natsunoame said...

And I am still waiting for this specific Greek DNA pls? Simon ? Someone ?

Dita said...

@Samuel Andrews

"Looking at Italians' position in G25 PCA, and at Y DNA (E-V13, J2b), I said they have direct Balkan ancestry not just Anatolian/NEar Eastern."

If it is connected to EV13 and J2b, then its Albanian and nothing to do with Greeks. Greeks have no J2b (except arvanite settlement locations) and all the EV13 they have has Tmrcas dating to the late bronze age/dorian invasions. All their Ev13 has low diversity and bottlenecks which all point to a northern origin ~3000 years ago.

Salden said...

Ban the Afrocentric already. Africa beyond the northern regions (who are racially separate from the rest of the continent) was not relevant across written history until well after the fading of the Greeks and Romans.

Bob Floy said...

I'm half southern Italian, and recently found out that I belong to Y-haplogroup E-V22, which is not on the EV13 branch.

Samuel Andrews said...

First Horse Warriors. May 15 2019.
https://www.pbs.org/wgbh/nova/video/first-horse-warriors/

natsunoame said...

...the Albanians themselves were mentioned in the Balkans not earlier than the 12th century. What exactly can be their role in this particular time then? They are OUT of records obviously and don't match linguistically either.

Samuel Andrews said...

@Bob Floy,

E-M123 is more rare E1b1b1 clade in Italy & southeastern Europe. It is more common in the Near East.

Andrzejewski said...

@Samuel Andrews "Tuscans have 25% Yamnaya ancestry. Southern Italians have 20%. Sicilians have 15%.

There's a very low chance any historical incursions from northern Europe made an impact on central-southern Italy.

Therefore, all of this Yamnaya ancestry was in place in Italy in the Iron age. I also, guarantee not all of it is from Bell Beaker R1b U152+ descendants but that some of it is from the Balkans."

The Germanic invasions ie Ostrogoths and Lombards/Longobards impacted the DNA of Northern Italians but almost none of the Central or Southern Italians, let alone that of the Sicilians.

Bob Floy said...

@Sam

Yeah, I saw a good map a few days ago, it's practically non-existent in Europe outside of southern Italy and parts of Spain. You know a lot about Italy, what are your thoughts on that? Phoenicians? Or something more exotic? It's barely present in Greece, but there's a big hot spot around what's now Istanbul.

Andrzejewski said...

@Dragos "@ Matt. / Sam
No no.
I think what happened is that PIE came from Siberia (a/p Andrzejewski's linguistic deductions)"

I have never said or stated that PIE came from Siberia. All I was hypothesizing was the possibility that it might've originated with the EHG groups roaming the Steppes and not with the CHG ones, but I'm increasingly becoming more certain that if it were the Piedmont groups speaking a putative PIE and later on spreading this speech to both Khvalynsk and Sredny Stog II as @Samuel said, then it must be the CHG-rich groups where this speech originated and not some R1a1 Ukraine HG (Bug Dniester/Dnieper Donets) nor with the "Samara Hunter Gatherers"...

Andrzejewski said...

@Gaska "sam what percentage of yamnaya ancestry we have the Basques and the rest of Spaniards?

If U152 did not speak Italo Celtic, where did Latin come from? It can not come from the Greeks, can it?

The theory of R1b hijacking Neolithic languages ​​throughout Europe does not seem serious to me

Do you know what I'm thinking knowing what happened in Spain and probably in Italy?"

Basques and Etruscans have 25% Steppe, which means their Kurgan origins may exceed many IE speaking nations. Basques and Etruscans can both come from some BB non-IE speaking group although they don't seem to be related according to latest linguistic work; therefore it may turn out that Bell Beaker was a complex link of diverse speakers of non related languages - Steppe, EEF, WHG - and that the linguistic landscape was an evolving and changing one. It was kind of proven that Basque and Sardinian Nuragic were related but some have attributed it to Basque migrants on the Sardinian shores rather than a closeness of some allegedly EEF language relic.

And going off on a tangent here re: Sardinians, I surmise that some of the R1b that the ancient Jews during King David time and onward possessed might've come from the Shardana migrants from Sardinia as part of the larger "Sea People" migration causing the "Bronze Age Collapse"; I further believe that the word "Sard" has something to do with the phrygian city Sardis on the Aegean coast.

Dragos said...

@ Andre
That’s because you assume that PIE must come from that steppe is a truism. Anyhow, even if pre-pre-pre-PIE is a north Eurasian paleo-language, with some southern influences later acting on it; why are you so sure it’s liked to specifically to Yamnaya, EHG or R1? These ANE impacts were already present for Millenia before that; percolating throug WHG & even Middle Neolithic farmers

Andrzejewski said...

@FrankN "Btw, the reason I had been looking for a list of Pre-Saami substrate was entertaining the idea of a maritime movement along the Arctic Sea to explain documented (e.g. Jaeger 2015), but so far unexplained parallels between IE, especially Germanic and (Insular) Celtic, and Chukotko-Kamchatkan. In fact, there are quite a number of shared maritime HG terms, e.g. lax (salmon) that can be traced from Germanic via Balto-Slavic, Finnic, a couple of Chukotko-Kamchatkan and NW American languages to ultimately Sundanese lauk "sweetwater fish". Similar chains can be constructed for "seal", "whale", or "cock" (c.f. Greenberg's proposed Amerindian *cac "large bird").

Then, however, I came across S. Nikolaev's recent reconstructions of Proto-Wakashan-Nivkh-Algic (see https://en.wikipedia.org/wiki/Algic_languages for respective links). If correct, these reconstructions would not only - importantly enough - present a second case, after Yeniseian-Na Dene, of a trans-Pacific language family, they could also provoke the need to fundamentally review everything we have so far assumed about the origin of PIE."

I hope @Davidski agrees with me, because here is what I think happened:

I. 24,000 years ago there was MA1. The origin point of "Ancient North Eurasians" was to the west of Lake Baikal. One group traveled up and along the Yenisei River in Siberia to become the Kett or whatever they were. Whether or not they had linguistic ties with the Na-Dene as proposed by Vaida remains to be seen (after 24,000 years!). Another group traversed eastbound TRANSBAIKAL all the way to the banks of the Amur River and became another ANE (or "Paleo-Siberian") group as the Nivkh. Some people claim that the Ainu, the Korean and some Tungus groups have lots of Nivkh substrate, both linguistically and genetically. (Intriguing to figure out who the AINU actually were and their plot of the PCA charts, BTW). It was asserted that many shamanistic beliefs of the NW Pacific Coast Native Americans such as the Gwinchin and the Haida are almost identical to those of the Nivkh.

II. On their long trek towards Beringia, these Nivkh-like bands came across ENA (East Asian) groups e.g. Ulchi. The latter are nowadays speakers of a Tungus language, and it turned out Native Americans share their East Asian aDNA with the Ulchi (which lack Ancient North Eurasian component at all).

Andrzejewski said...

@Dragos "@ Andre
That’s because you assume that PIE must come from that steppe is a truism. Anyhow, even if pre-pre-pre-PIE is a north Eurasian paleo-language, with some southern influences later acting on it; why are you so sure it’s liked to specifically to Yamnaya, EHG or R1? These ANE impacts were already present for Millenia before that; percolating throug WHG & even Middle Neolithic farmers"

98% of the posters here plus leading experts in the field including Dr. Anthony, Dr. Reich and Dr. Lazaridis posit that PIE was a Steppe language. We only differ on the DETAILS, i.e. was PIE an EHG language, a CHG one, an amalgam of both or a brand new creation following the advent of what's termed as "Steppe ancestry".

What is YOUR idea regarding PIE? That it was the lingua franca of the Cucuteni-Tripolye Culture?

Andrzejewski said...

III. Another group of ANE became the AG3, which happened to be among the first in the world to have a genetic mutation for blond hair circa 14,500 years-before-present. That was 10,000 years AFTER MA1. Somehow following the LGM these R1 rich groups, originally mammoth hunters moved into Eastern European Steppe. We don't even know how PIE was created, but if what @Samuel Andrews contended a few days ago, it may've been that the Piedmont (or maybe Progress?) groups are the first speakers of PIE or whatever shape it took 7,000 years ago, transmitting it in turn to the Samara HG and the Ukraine HG, thereby creating the renowned Steppe ancestry by transforming Samara into Khvalynsk, Sredny Stog I into Sredny Stog II and everything else we associate with PIE society: burial with the red ochre and supine position, taming and mastery of the horsemanship, kurgan tels and mounds, the PIE mythology and religion which gave rise to European folklore and Rig Vedic sourced Hinduism/Buddhism, hierarchical and individualistic society where the "heroism" takes center-stage, etc.

Andrzejewski said...

@Samuel Andrews "First Horse Warriors. May 15 2019.
https://www.pbs.org/wgbh/nova/video/first-horse-warriors/"

I watched it online a few days ago. Dr. Anthony was featured prominently in it but I largely discarded his theory since his book came out in 2005. One thing that is highlighted and is interesting is the possibility that the Steppe pastoralists were greatly outnumbered by their neighbors but the plague (=Yersinia Pestis) spread with them, weakening other societies which had no or lesser immunity to it (settled Cucuteni Tripolye comes to mind!). According to the NOVA broadcast, female were much more vulnerable to the disease, resulting in male yamnaya kin groups raiding their neighbors and stealing their females as wives.

The importance of this segment is that if it was indeed true and factual then it could potentially explain how come most of Europeans have either R1a or R1b (with minor farmer T, E1b1b or G1a or HG I2 or Ia) whereas most of the female European DNA comes from farmer societies to their west - H1, H3, T, N etc.

Dragos said...

Andre
98% of the posters here don't have a clue about what theyre talking about, especially you
David Anthony's model has already been disproven.
You seem fixated on dumb-down models revoling around EHG . ANE , Siberians.

Andrzejewski said...

About the term "weik": it's more likely that the PIE's referred to themselves as something to do with this term or a derived one, meaning "community" than anything like "Aryan", although I wouldn't rule out that their ruling elite might've referred to themselves with this now-controversial term.

Samuel Andrews said...

"First Horse Warriors. May 15 2019.
https://www.pbs.org/wgbh/nova/video/first-horse-warriors/"

99% accurate. Wrongly call all Steppe people Yamnaya. Really, it was mostly Corded Ware, bell beaker, Andronovo. But that is a small error. Not sure if their success was due to horses. I thought the idea of Eneolithic horse-riding R1b/R1a was dead.

They make Yamnaya clothing in documentary look like classical Scythians (who were mostly of Yamanya decent). Not sure if that is accurate. But, whatever that's not very important.

Andrzejewski said...

@Dragos "You seem fixated on dumb-down models revoling around EHG . ANE , Siberians."

Or ones revolving around the mixture of CHG-WHG to produce PIE's. I do acknowledge Yamnaya having 18% EEF+WHG ratio stemming most likely with earlier interactions with Cucuteni-Tripolye.

What do YOU think happened to C-T? Did they go the way of the Botai into extinction, or did they contribute anything to any historic or modern Balkanic population of any sort?

"98% of the posters here don't have a clue about what theyre talking about"

I am sure @Davidski would not agree with you LOL

Andrzejewski said...

@Samuel Andrews "99% accurate. Wrongly call all Steppe people Yamnaya."

Because Yamnaya which likely grew from Repin came to encompass all Steppe area from Pontus (Black Sea) to the Caspian. @Davidski calls it "Pontic-Caspian Steppe". I thought that horseback riding didn't evolve until the onset of Sintashta but I may be wrong here. I also heard theories that horses used to be used as food before being tamed on the Steppe. If NOVA was correct about Pestis infecting and impacting females more than males it might answer why there was a much more male-replacement of the European population than female ones: first of all, there were much less women around, and secondly - Yamnaya needed females which they took from the conquered populations. Moreover, if Steppe Kurganists had more immunity to the plague (think Spaniards v. Aztecs, Inca) then it could explain the downward spiral encumbered by CT Culture and other Neolithic ones, enabling in the process for the initially outnumbered Steppe groups to take over and multiply.

Samuel Andrews said...

@Bob Floy,
"Yeah, I saw a good map a few days ago, it's practically non-existent in Europe outside of southern Italy and parts of Spain. You know a lot about Italy, what are your thoughts on that? Phoenicians? Or something more exotic? It's barely present in Greece, but there's a big hot spot around what's now Istanbul."

I don't know much about Y DNA. There's signs of Levantie ancestry only in southern Italy but not in central/northern. Greeks dominated southern Italy from 7th-3rd centuries bc. I don't think the Phoecians made any colonies there. Phoecians were in North Africa, Spain, Sardinia.

Davidski said...

@Samuel Andrews

First Horse Warriors. May 15 2019. https://www.pbs.org/wgbh/nova/video/first-horse-warriors

Can't watch this right now. Do they mention Sintashta at all?

Dragos said...


Nothing new . Evidence for horses at Botai
A lot of talk about Yamnaya; but no evidence.

Dragos said...

But they managed to get one thing right at least- they point out that the main Yamnaya expansion was to Northwest Europe; therefore not relevant for actual IEs; who are attested in Anatolia and SEE

Andrzejewski said...

@Dragos I'm curious: why are you so personally invested in the "PIE is NOT from Yamnaya/Kurgan/Pontic-Caspian Steppe" position?

Dragos said...

I see nothing wrong with a modified “kurgan” model of some sort ; but I would have thought that it’s clear by now that the model propounded by Anthony is erroneous

Samuel Andrews said...

@Davidski,
"Can't watch this right now. Do they mention Sintashta at all?"

No. They only mention Botai & Yamnaya. Are you asking because Yamnaya weren't the master horse riders & that it was really Sintashta.

Ric Hern said...

What is interesting for me is this that Shaikorth posted :

distance%=5.1415"

Bell_Beaker_FRA_C

Anatolia_Barcin_N,48.6
UKR_Mesolithic,36
RUS_Karelia_HG,8.6
GEO_CHG,6.8

"distance%=5.6525"

IRL_BA

Anatolia_Barcin_N,38.2
UKR_Mesolithic,37.2
RUS_Karelia_HG,13.2
GEO_CHG,11.4

Interesting that CHG went down in France...So some CHG that spread along a Southern Route from Anatolia did not reach France during the Chalcolithic/Early Bronze Age....

Davidski said...

@Samuel Andrews

Yeah, obviously, Sintashta is really the first documented equestrian culture.

@All

Early Italic speakers have bucket loads of steppe ancestry. You can put that in the bank.

Davidski said...

@Andrzejewski

Please make an effort to check your facts before posting here.

Haak and Lazaridis didn't write any thesis on horseback riding in the Bronze Age.

Ric Hern said...

For me personally the fact that Celtic, Germanic and Italic shared some Linguistic similarities on Proto Level points to more close Geographic contact between these groups before the Nordic Bronze Age....

Dragos said...

@ Andre
Given that you mention about CT disappearing; you might wish to know that Yamnaya also “disappeared “ from the Balkans hundreds of years before the earliest IE are attested. You might also wish to know that Yamnaya got stopped in their tracks in Pannonia by Kostolac groups. Curiously it is descendants of the latter which expand to Anatolia & Greece; but also north- central Italy and even back to Central Europe
Given that they were still interacting with steppe groups; they also had some of that magical steppe admixture that you froth about
But ; essentially; I see no evidence that Yamnaya was IE. indeed; the axioms of Yamnaya are extinct (Afanasievo); or non -attested / non-IE (BB).

Andrzejewski said...

@Ric Hern "For me personally the fact that Celtic, Germanic and Italic shared some Linguistic similarities on Proto Level points to more close Geographic contact between these groups before the Nordic Bronze Age...."

Linguists find a link between those two and Tocharian.

Andrzejewski said...

Not sure about writing about the horseback riding (I was wrong if I did) but Haak & Lazaridis sure did write an essay how a massive migration from the PC Steppe (they said “Russia”) is responsible for the spread of IE languages across the rest of Europe and also throughout parts of Asia:

https://www.nature.com/articles/nature14317

But @Dragos would call them “idiots” because they don’t share his anti-Steppe bias

Dragos said...

@ Andre
I guess you answered your own Q - even Davidski thinks youre a clown, and he is very sympathetic to the steppe hypothesis.
As I suspected, my attempts to outline the problem with the steppe hypothesis went over your head. You will continue to pontificate about matters you have no clue about
Tell me - have you even been to Europe ?

Andrzejewski said...

@Dragos https://www.nature.com/articles/nature14317

I have just posted AGAIN FOR THE SECOND TIME the link to Dr. Haak & Lazaridis' 2015 article. Have you read it? Do you think that these scientists are stupid?

Ric Hern said...

Yes, for me personally the similarities between some words in Tocharian to Germanic, Celtic and Baltic/Slavic words is absolutely striking....

Dragos said...

Yes I have but I don't think you have
Haak 2015 showed that CWC can be modelled as 70% Yamnaya + 30 % MNE, on average.
What's your point ?

Bob Floy said...

@Sam

Thanks.

"Greeks dominated southern Italy from 7th-3rd centuries bc. I don't think the Phoecians made any colonies there. Phoecians were in North Africa, Spain, Sardinia."

Phoenicians actually did have colonies in western Sicily, and for awhile(during the last Punic war, I think) a presence in Calabria, but of course it's true that the Greeks were the dominant party there. I am curious, though, where do you think Italy's big Anatolia_BA footprint came from? I mean, if you had to guess?

Dragos said...

@ Davidski- help the clown out please

Ric Hern said...

I will not be surprised if Proto-Indo-Europeans actually sounded more like Tocharian. Unlucky for most the Late Attestation is a thorn in the foot that Nutters will use in all kinds of ways..

Davidski said...

@Andrzejewski

You need to slow down a bit and try to do more careful background reading before getting into these debates, otherwise you open yourself up to easy attacks and create discussions that don't produce anything useful.

For example, the Haak et al. 2015 paper (not thesis) doesn't claim or suggest that Anthony was wrong. In fact, it obviously backs up Anthony in some very big ways.

But it also stops short of saying that the Proto-Indo-European homeland was on the steppe, and indeed the lead author seems to be very sympathetic to a PIE homeland south of the Caucasus, along with his colleagues at Max Planck. See here...

Dead cat bounce

Are you surprised? Well, you shouldn't be. You should have known all of this already after posting here for many months.

Ric Hern said...

@ Dragos

How precisely will you personally describe the Origins of Tocharian ? How, Where and When ? By ways of which Cultures ?

Samuel Andrews said...

@Bob Floy,
"I am curious, though, where do you think Italy's big Anatolia_BA footprint came from? I mean, if you had to guess?"

I would actually guess Greece. They're a people from near Anatolia who had a big presence in Italy.

Bob Floy said...

@Sam

I would think so too, I only ask because several posters here, more knowledgeable than me, have suggested that the Greeks didn't make much of a genetic impact there.

Thanks.

Dragos said...

@ Ric
I’m actually starting to think that Tocharians came with the Andronovo horizon ; but then fell ‘out of the loop’

Bob Floy said...

@Dragos

"I’m actually starting to think that Tocharians came with the Andronovo horizon ; but then fell ‘out of the loop’"

How to explain the "centum" character of their language?

Dragos said...

Bob
Satem is merely one (potentially late) set of isoglosses shared amongst some IE lects; and imperfectly so.

Bob Floy said...

@Dragos

I was under the impression that "centum" characteristics were more conservative, that is, closer to the PIE stem, while "satem" characteristics were a later innovation. I also thought that indo-Iranian languages are far removed from Tocharian in a number of other ways. Are you suggesting that Andronovo wasn't really associated with IA languages? Or that they actually spoke multiple IE languages? How would your scenario work? I mean, hypothetically.

Dragos said...

Bob; that’s what I just suggested - satem is a later development; centum is the default
There is not a consistent typology for tocharian; some doubt that is so far removed as initially thought; similarly formmuch of the early hum-drum about its supposed Celtic affinities (not surprisingly; made by 2 Irish scholars:))
The corollary would be that CWC is an early Northern IE

Dragos said...

The Satem Languages of the Indo-European Northwest. First Contacts?
Henning Andersen

Bob Floy said...

@Dragos

"some doubt that is so far removed as initially thought"

That's interesting. I do remember reading years ago that they called themselves "Arya".

"the early hum-drum about its supposed Celtic affinities"

Yeah, and other researchers swore up and down that it was close to Germanic. Never seemed to realistic to me.

EastPole said...

@Dragos

“satem is a later development; centum is the default”

No, this is not true. PIE was neither centum nor satem. The centum–satem division forms an isogloss in synchronic (!!!!!) descriptions of Indo-European languages. It is no longer thought that the Proto-Indo-European language split first into centum and satem branches from which all the centum and all the satem languages, respectively, would have derived.
There are centumisatin and satemisation processes going on all the time in all languages.

Dragos said...

Whatever you prefer. But that’s my point- the C/S split is not so definitive or diagnostic

Dita said...

Albanian is the only surviving IE language that is neither Satem nor Centum. Its astounding how much it is not even taken into account when analysing the IE situation, especially since its in the middleof Greece, Rome, Danube, etc all critical points for IE phenomenon.

“Thus, the three-way IE reconstructed voiced ~ voiceless ~ voiced aspirated system of obstruents has been reduced, as in many IE dialects, to a double opposition: voiced ~ voiceless; and the outcomes of the three dorsal series suggest that Albanian, like Luwian, may have originally retained this three-way opposition intact and therefore is neither centum nor satem, despite the clear satem-like outcome of its palatal dorsals in most instances.

The evidence for this is the palatalization of original PIE labiovelars, but not plain velars, before front vowels.”

B 2018 De Gruyter
“Handbook of Comparative and Historical Indo-European Linguistics

Vinitharya said...

The French language is quasi-Satem, with ancestral 'c' centum forms having undergone a kind of satemization. Campus 'field' is champ, pronounced initially with a 'sh' sound; Caput 'head' is chef, again, pronounced with same sound as champ; and even the Latin centum, hundred, is cent, pronounced with a sibilant.

Simon_W said...

Well, I've just checked it and seen natsunoame isn't banned as Salden had suggested, so in case he still reads here let me reply to his question. These are models I made with samples of ancient Greek DNA applying the Global25/nMonte method, they show the autosomal ancestry of these people in percent:


distance%=1.8477

GRC_Mycenaean

Barcin_N, 51.2
Anatolia_EBA_Isparta, 21
Anatolia_EBA_Ovaoren, 14
Yamnaya_Samara, 11.5
Yoruba, 0.9
Ganj_Dareh_N, 0.8
WHG, 0.3
Han, 0.3


distance%=2.6186

Iberia_Northeast_Empuries2

Barcin_N, 50.6
Anatolia_EBA_Isparta, 22.5
Anatolia_EBA_Ovaoren, 14.3
Yamnaya_Samara, 12.3
Han, 0.3

Mycenaeans are Bronze Age Greeks from Greece, and Empuries was a Greek colony in northeastern Spain. As you can see, they were very similar. The slight Yoruba admixture showing up in the first model is noise, the same holds true of the Han admixture. You see, the results show that the ancient Greeks were a three-way mix of early European Farmers of the Barcin-type, Bronze Age Anatolians and Yamnaya from the Pontic-Caspian steppe.

Simon_W said...

@Bob Floy

"I would think so too, I only ask because several posters here, more knowledgeable than me, have suggested that the Greeks didn't make much of a genetic impact there."

A slow spread of Anatolia_BA ancestry across Greece and beyond, to Southern Italy and Sicily is indeed the most probable explanation, but this doesn't mean it was all brought by ethnic Greeks. Because, as my model of the Sicilian Beaker showed, it started long before there were Greeks in Greece.

Simon_W said...

Really impressive that even the Samnites were like the Umbrians and Etruscans northern-like. I misread them as Sabines last Saturday. For all who are not familiar with the Samnites: They were Oscan-speakers located in Southern (!) Italy.

I could imagine that the Latin/Faliscan branch of Italic was the result of an earlier Italic wave and because of this had more non-Italic ancestry. But given the leaked info that central Italy in the EBA was EEF-like this explanation isn't very satisfactory. Of course there's still the old story that the Latins mixed with Trojan refugees under the lead of Aeneas at the end of the Bronze Age. Would be a huge surprise if that turned out to be true, it always sounded like wishful thinking of some Roman wannabe Grecoids.

Davidski said...

@Simon

Central Italy during the EBA wasn't all EEF-like. The report you saw must have been based on a few samples from a particular location.

My information is that EBA Italy was very heterogeneous, with a cline that stretched from EEF to Northern Europe.

Simon_W said...

@Davidski

I was drawing upon the info provided months ago by Ryukendo at Anthrogenica who was at a presentation of the main author of that big upcoming paper. According to him, Lazio in the EBA, at least up to 1700 BC, was Sardinian-like. So yes, I wasn't correct saying "central Italy" as Lazio is only a part of it. But at least it means that there wasn't much Anatolia_BA ancestry up to 1700 BC in the place where the Latins later lived.

However, northern-like people in (parts of) peninsular Italy during the EBA that you say you have read about are certainly interesting. This predates even the fairly homogenous Apennine culture of the MBA. Could be in part Beaker-derived and in part from the Balkans.



Simon_W said...

Maybe, or even more so, it's also associated with the Polada influence attested in central Italian EBA pottery. The Polada culture was located in Northern Italy, but exerted some influence upon central Italy. And it had been in turn influenced by cultures from beyond the Alps like Singen and Straubing.

natsunoame said...

Simon
Your copy-paste text means nothing but just showing to the world you are naked in this theme...
First of all you have to understand which are the greeks in real, but for that you need not just some DNA results...you need to read some historical records and archeological datas, not to mention political problematic here. Just like Macedonia and Romania, Greece is a political creation from XIX century, means really new country on the map. And their DNA reveals that. About North Greece, the history from XX century we are talking about, you can check by yourself why they wear Bulgarian blood and some even speak in Bulgarian in the Greek Parliament in XXI century, they still know their ancestors language.
Second..just like in practical science you need experimental and control group, but if you don't have that your attitude means nothing like at the moment. Because I can assure you furthermore there are tones of literature speaking about the autohtone people in todays Greece, including their culture, old cities names/Corinth, Mycenae, Athens... which are not created by Hellenes,even J. Chadwick admitted the fact/, Gods and of course their language and they were not greeks. When you realise that we can speak again. In other case its just a lost of time, mine and yours too.
Anatolia old maps can reveal you which tribes have inhabited these lands and again this aren't greeks at all, exactly like we are 100% sure Ilion/Troy wasn't greek too, but was destroyed by them.
P.S. Check about Danaus tribe.