search this blog

Monday, September 2, 2019

Commoner or elite?


I recently started looking at the correlations between Y-chromosome haplogroups and social standing in ancient Europe, and was surprised by what I learned about the five currently sampled prehistoric Scandinavians belonging to Y-haplogroup R1b. I certainly wasn't expecting to uncover these stories about a mass human sacrifice, a bog body, and an Arctic circle warrior:

- The earliest Scandinavian in the ancient DNA record belonging to R1b comes from a grave site in what is now northern Norway (VK531, Margaryan et al. 2019). This individual has a genome-wide profile similar to that of local Mesolithic hunter-gatherers, but is dated to just ~2,400 BCE. During this time, Scandinavia was dominated by a "new" population associated with the Battle-Axe culture (BAC), with high levels of ancestry from the steppes of Eastern Europe. Since VK531 wasn't buried with any BAC grave goods, and indeed with no grave goods at all, it's possible that he may have been from a remnant forager population that was displaced and ultimately forced into extinction.

- R1b-U106 is today by far the most common R1b subclade in Scandinavia, but it's not yet clear how it managed to attain this status. Was it perhaps through elite dominance? The earliest ancient individual belonging to R1b-U106 is dated to 2275-2032 calBCE and comes from a Late Neolithic, likely post-BAC burial ground in what is now Sweden (RISE98, Lilla Beddinge, grave 49, southern skeleton, Allentoft et al. 2015). However, RISE98 wasn't buried in any way that would suggest he was an individual of high social standing. In fact, he was found in a mass grave, along with two other adults and two infants, possibly representing a human sacrifice. The only artefact in the grave was a bone needle. More details are available here.

- During the Nordic Bronze Age it became customary for Scandinavian elites to be laid to rest in richly furnished barrows, while commoners were buried in flat graves with few or no offerings. Human remains recovered from a "commoner" flat grave cemetery dated to the Early Bronze Age near the present-day city of Aalborg, northern Denmark, included the skeleton of a male belonging to Y-haplogroup R1b-M269 (RISE47, grave 3, skeleton 8, Allentoft et al. 2015). Keep in mind, however, that this might have been another case of an ancient Scandinavian R1b-U106 if not for missing data. A flint dagger was found alongside one of the skeletons in this cemetery, but RISE47 wasn't accompanied by any grave goods (see here).

- One of the most amazing archeological discoveries made in Scandinavia is the Trundholm Sun Chariot. Found in a peat bog on the island of Zealand, Denmark, in 1902, it's thought to be an Indo-European religious artefact dating back to the Nordic Bronze Age; a representation of a horse pulling the sun and perhaps also the moon in a spoked wheel chariot. Another important discovery in a peat bog near Trundholm dating to the Nordic Bronze Age was the body of a man belonging to R1b-M269 (RISE276, Trundholm mose II, bog find 1940, Allentoft et al. 2015). However, chances are slim that RISE276 was a charioteer or, say, a spiritual guru who accidentally drowned in the bog. Most Danish bog bodies are thought to have belonged to sacrificial victims or executed criminals.

- Interestingly, the earliest likely Scandinavian warrior belonging to R1b, and also R1b-U106, is from an early Iron Age burial in present-day northwestern Norway (VK418, Margaryan et al. 2019). This site isn't quite as far north as the grave of the above mentioned VK531, but it's still well within the Arctic circle. Apparently, VK418 was buried with some impressive weapons, potentially of "eastern origin", including a shield, spearheads and a sword. Who knows, he may even have been an elite warrior for his time and place?

The other two main Scandinavian Y-haplogroups, I1a and R1a, haven't yet been found in prehistoric Nordic remains from such, shall we say, depressing burials. That's not to say, of course, that they won't be sooner or later. RISE175, from Allentoft et al. 2015, is currently the only individual who fits the bill as a representative of the Nordic Bronze Age elite. He was buried in a barrow grave in what is now southwest Sweden and probably belongs to Y-haplogroup I1a. That's not much to go on, but perhaps it's a sign of things to come?


See also...

Isotopes vs ancient DNA in prehistoric Scandinavia

Who were the people of the Nordic Bronze Age?

They came, they saw, and they mixed

165 comments:

Matt said...

Regarding these last two posts, I'm not following the detail at the moment but is all this related to Kristiansen's 2017 - https://www.researchgate.net/publication/317335381_Identifying_commoners_in_the_Bronze_Age_burials_outside_barrows - "Identifying commoners in the Bronze Age: burials outside barrows"?

Davidski said...

Nope, it's more related to this...

https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0219850

zardos said...

I think its good to compare social stratification from later periods, but of a comparable organisational level with this finds.
You don’t just have an elite and commoners, but also a higher middle class, "specialists", like specific craftsmen and traders, but also, and thats very important, slaves.
In Northern Europe slaves were not that bad off in their daily life, but they had little rights, property and of course weaponry which belonged to them.
I doubt slavery was that common in the mobile, pastoralist warrior clans. But it became common most likely with extended crop farming, mining, big transports and more stratified, complex and more stable societies, which means the Bronze Age, since Unetice in the North.

I would say many so called commoners might actually be slaves, or had slave like status. Considering how free tribal members of f.e. GermanicS and Celts, were equipped, how the "wealthy burials" were equipped, it would have been a shame to bury the own relative so shabby
Even more so if some of those items were not that expensive. So rather there might have been rules to how certain people were buried. And obviously to deny a free family such things...I doubt it.

So most likely a lot of "commoners" without grave goods were in some way marginalised. The majority might have been slave-like and no proper members of the tribe.
This means they could be prisoners of War from elsewhere and are surely not more representative for the regional ethnicity in most instances, with the exception of a recent conquest by a foreign elite.

Davidski said...

@zardos

RISE98, RISE47 and RISE276 were probably locals based on their DNA and isotope ratios.

zardos said...

Executions could happen to anyone though which defiled gods, was a traitor or coward, or otherwise a criminal. Though such executions too would have been more likely to happen unfree and lowest level people in the social organisation obviously.
For some times and regions its no wonder more better equipped burials were found and just a few poor ones. Thats not always bias, but the way free tribal members were buried by their relatives. Only unfree and marginalised men were buried without that care.

Davidski said...

Yep, and it seems unlikely that the potential sacrificial victims from Lilla Beddinge, grave 49, were free.

zardos said...

@David: I don’t say its all the same, because scenarios can differ. But usually, the social status is heritable, which means you will see the same pattern for generations, but the foreign provenience of captured slaves or new elites only in the first generation in isotopes and just a little shift genetically between neighbouring groups.

Gaska said...

It seems good to me that you discuss the issue of VK531, because it also demonstrates the connection of the R1b lineage with the WHGs (such as Villabrunna, Iboussieres and the Latvians of the Kunda and Narva cultures). It also shows that the WHG reached the Arctic circle (66°33′47.8″ north of the Equator) which is an incredible feat considering the scarce material means of the Mesolithic-Neolithic in Europe.

Indeed, there is nothing in these sites in northern Norway that can relate this case of R1b with the CWC, BAC or BBC, because they never reached so far north, and also show another case without steppe ancestry in the haplogroup R1b, which is strange because everyone's commitment to link the origin of this lineage with the steppes. Extinct lineage? Why you always have to use this argument when R1b appears in Europe outside the steppes. What makes you think that?.

The appearance of this case of R1b in Norway greatly hinders the migration of this lineage from the Yamnaya culture (3,300-2,400 BC), I don't know if anyone will have any theory about it.

Davidski said...

@Gaska

As far as we know for now, VK531 belongs to R1b1a1 and doesn't have any steppe ancestry.

But then a bit later RISE98 belongs to R1b1a1a2a1a1 and has a lot of steppe ancestry.

See that's one of the many reasons why your arguments aren't convincing.

Gaska said...

RISE 98 is U106 (2.152BC), and I guess it belong to the BAC because it is too late to belong to the SGC. It is also another lineage with "tons" of steppe ancestry that has not appeared in the steppes (like R1b-L51-P312)-

On the other hand my arguments are not intended to convince anyone, they are categorical statements

1-Can you deny that R1b is a lineage related to WHG?
2- Can you deny that it is a feat to reach the Arctic Circle in the Norwegian Neolithic
3- Do you think that Vk531 belonged to the BBc or ​​the CWC?
4- Can you prove that it is an extinct lineage?
5- Can you deny that this case forces you to seek other explanations for steppe migrations?

Everyone knows that Kurganist explanations when they encounter cases like this are fundamentally that it is about extinct lineages, but I suppose you have sought other solutions. I guess you will think that U106 is also in the SGc with his father L51 and his relative P312, but again, where did they come? from the Yamnaya culture? really? We have to keep in mind that Vk531 is related to the Latvian hunter gatherers who controlled the Baltic Countries until they were displaced by the CWC. And where did those men go? They probably moved to Scandinavia and German Neolithic cultures. We have no reason to think they went extinct.

Davidski said...

@Gaska

Do you think that Vk531 belonged to the BBC or ​​the CWC?

It'd be strange if he did, since he doesn't show any steppe ancestry and doesn't belong to R1b-M269.

Angantyr said...

The R1b Kunda/Narva individuals were not pure WHGs, they were heavily mixed with EHGs. (It seems like the I2a Kunda/Narva individuals - unsurprisingly - generally had more WHG heritage.)

And we know that EHGs migrated west- and southwards along the Norwegian coast as long as 10000 years ago, to meet with WHGs and form the SHG population of the Scandinavian peninsula, so it's no surprise that an EHG paternal linage is found in northernmost Norway ~2,400 BCE. I don't think even the Nordic Bronze Age reached that far north.

Angantyr said...

@Gaska

RISE 98 is U106 (2.152BC), and I guess it belong to the BAC because it is too late to belong to the SGC.

And it is also too late for BAC, which ended ~2350 BCE in southernmost Scandinavia. (Further north it lasted until ~2300 BCE, but that doesn't help.)

Måns Sjöberg said...

Northern Norway is far away from the BAC area. Rather, this is traditional saami land.

epoch said...

A HG with a R1b clade that is ancestral to M269 is pretty interesting.

Gaska said...

@Argantyr.

As far as I know, our Latvian friends R1b-P297 were in the Baltic Countries for thousands of years and were 70% WHG and 30% EHG, we also have other WHGs R1b-L754 in Itaky and France, then any attempt to link R1b with the EHG seems unbelievable because we also know (as you said) that WHGs migrated north to Scandinavia to give rise to SHG. Then if we have WHgs R1b because we have to assume that it is a lineage belonging to the EHG

@Davidski-

Nice try but you know what the genetic composition of the BBC and the CWC was. In the first one we have a certain variety of male lineages-H2, I2a, G2, R1b-P312 and in the second R1a and even I2a. Then these cultures were not genetically homogeneous and Vk531 is also the proof that R1b is neither exclusive to the steppes nor is it exclusive to the BBC-I don't know if the Vk531 genome has enough quality to look more deeply at what subclades it could belong to, but surely more cases will appear in Scandinavia and then we'll see if they are M269 / L51 etc. or they aren't. Discarding it seems reckless

Gaska said...

Obviously-

Then if we have R1b-WHgs why we have to assume that it is a lineage belonging to the EHG?

Andrzejewski said...

So the now-ubiquitous and most common R1b-U106 were the commoners during the BA and the I1a (foragers? EEG?) plus Corded Ware (R1a) were the elite?

Anonymous said...

As Narva culture (R1b1a1a) was autosomal close to the SHG, in the case of migration of its representatives into Scandinavia they very quickly become indistinguishable from the Mesolithic SHG.


Could the BBC (R1b1a1a) to be from the Narva culture (R1b1a1a) migrating through Scandinavia to the Netherlands where they mixed up with women of the SGC?

epoch said...

@Archi

No, because we have an Ertebolla genome - one extracted form the birch resin used as chewing gum - which was explicitly checked for that, with negative result.

epoch said...

https://www.biorxiv.org/content/biorxiv/early/2018/12/13/493882.full.pdf

"Interestingly, the Syltholm genome is entirely composed of WHG ancestry, suggesting that EHG ancestry did not reach southern Denmark in prehistory. Furthermore, it is striking that the Syltholm genome also lacks Neolithic farmer ancestry, despite its relatively late date."

Anonymous said...

epoch said...
" No, because we have an Ertebolla genome - one extracted form the birch resin used as chewing gum - which was explicitly checked for that, with negative result."

It is only one single genome of one time and only from Ertebølle, small size culture.

Gaska said...

@Archi

Why not?, according to my point of view is an option that must be taken into account

I4630- Latvia, Zvejnieki (7.271 BC)- HapY- R1b1a/1a-P297
I4432- Zvejnieki (5.997 BC)- HapY-R1b1a/1a-P297
I4626- Zvejnieki (5.738 BC)- HapY- R1b1a/1a-P297-
I4439- Zvejnieki (5.698 BC)- HapY-R1b1a/1a-P297
I4434- Zvejnieki (5.495 BC)- HapY-R1b1a/1a-P297
I4628- Zvejnieki (5.270 BC)- HapY- R1b1a/1a-P297
I4436- Zvejnieki (4.155 BC)- HapY-R1b1a/1a-P297
I4627- Zvejnieki (4.113 BC)- HapY-R1b1a/1a-P297

Here you have 3,000 years of genetic continuity of R1b-P297 in Latvia, local foraging societies were, however, not completely replaced and contributed a substantial proportion to the ancestry of Eastern Baltic individuals of the latest LN and Bronze Age, because modern Eastern Baltic populations carry the highest proportion of WHG ancestry of all Europeans, supporting the theory that the hunter-gatherer population of this region left a lasting genetic impact on subsequent populations-

The oldest farming culture in Scandinavia probably became established around 3,900 BCE. Norway was rather late in the game, but researchers believe they can also see traces of this 'wave' reaching Norway at the beginning of the Late Neolithic. But while people eslewhere in Europe had gotten the hang of cultivating crops and keeping livestock, the first farmers in Norway appear to have given up relatively early. They stopped growing crops after a relatively short period of time and returned to hunter-gatherer-fisher lifestyle-Towards the end of the Late Neolithic period, researchers again saw changes in the population."The settlement pattern changes very clearly when we approach 5,300 years ago. Coastal sites fall out of use, and long houses and grains begin to show up. Agriculture seems to have been introduced in Norway twice, and the second time it came to stay"

Vk531 could be a WHg descendant of the Latvians, but it could also be an European Central farmer because we have archaeological evidence of agriculture and livestock very close to the Arctic Circle-From the Stiurhelleren rock-shelter just below the Arctic Circle, six cereal kernels (five barleylHordeuin, one not identified to species), four sheep/goat bones and a single cattle bone were found in layers reportedly dated by four radiocarbon determinations of shells to between 4380f40 and 4170+100 b.p.(Hultgren et al. 1985). That equates to a date around the Middle Neolithic A/B phase. Stiurhelleren complements the disputed 10 cattle and sheep/goat bones from the far north settlements at Nyelv (Renouf 1989; also Engelstad 1985: 90). From Slettabn seven cattle bones were recovered from the Neolithic layer I1 (Skjdsvold 1977: 69). From the Skipshelleren rock-shelter, excavated 1930-31 (Bme 1934), a collection of 48 g of sheep/goat and cattle bones and teeth were radiocarbon-dated in the early 1970s. The direct date was reported as 4020f120
b.p./2830-2350 BC (Olsen 1976: 23f.)

Since we have R1b-M343 in Kromsdorf (2,602 BC), U152 in Osterhofen (2,542 BC), P312 in Spain (2,434 BC), the Netherlands and Alsace (aprox 2,400 BC) all of them older than Vk531, we can also think of a German origin of the lineage, some individuals migrated to Scandinavia (Farmers, no steppe ancestry) and others remained acquiring the steppe ancestry of the CWC

Anonymous said...

Narva was located between SHG and WHG. So who mixed and became like this.
"The earliest Scandinavian in the ancient DNA record belonging to R1b comes from a grave site in what is now northern Norway (VK531, Margaryan et al. 2019). This individual has a genome-wide profile similar to that of local Mesolithic hunter-gatherers"


This migration could be related to the Pit Ware culture.


zardos said...

@Gaska: But which group of farmers in or close to Germany have you in mind?
If BB R1b came not directly from the steppe, wouldnt French related farmers be more likely? They would have been closer to the Iberian centers of early, non-steppe BBC too and might have gained the advantage of earlier metallurgy before fusing with Western Corded groups.
So many scenarios are still possible for the steppe-BB emergence...

Gaska said...


Earlier I asked if anyone believed that VK531 could belong to the BBC or the CWC to see what everyone thought, but in reality it was a kind of trap question because there are certainly many possibilities that this individual was a Bell Beaker. Why ? you can see what Christopher Prescott says "No longer north of the Beakers" The present study extends the Northern European BBC geography into the Scandinavian Peninsula, potentially as far north as the ARCTIC CIRCLE.

The explicit definition of the BBC-region in Jutland provides a historical factor that helps to explain the transition to the Nordic Late Neolithic and the transformation of Scandinavia. It also challenges the predominant “local evolution” perspective inherent
to the last 50 years of Scandinavian archaeology. This is particularly true for the Norwegian case, where it has become increasingly apparent that around 2400–2350 cal BC, the Middle Neolithic B to Late Neolithic 1 transition, there was a dramatic reorientation of
material expressions, productive modes, economic organisation and ideology (Prescott 1996; Prescott and Walderhaug 1995).

Around the transition from MNB3 to the LN1 (2400/2350 cal BC) the above “Neolithic diversity” is transformed into an unprecedentedly unified cultural expression (the Nordic Late Neolithic/Dagger Period). On an even larger scale, the continuous patchwork
of small-scale groups extending from the Atlantic and into the taiga of Northeastern Europe was conceivably replaced by a cultural border between easterly groups and northwestern groups (Amundsen 2011; also Odner 2000).Based on radiocarbon dates and other contextual data this happened abruptly around 2400 BC. It took place from the southeastern part of the coast, in the interior, and along the coast, potentially as far North as the Arctic Circle.

The four key data are- BB culture/2,400 BC/R1b1a/Arctic Circle. Wonderful truth?, this means that it is perfectly possible to have a R1b BB (2.400 BC) in the Arctic Circle and most importantly without steppe ancestry.This would show that not all R1b Bbs would have steppe ancestry- That is why I also asked about the difficulty of explaining the R1bs steppe migration. It was important for everyone to remember that Vk531 does not have a drop of blood from the steppes.


Gaska said...

In Norway, the MNB3-LN1 agro-pastoral breakthrough, and the rapid establishment of farms
on a broad front, the transformation of material culture, technology, settlement, and architecture are most reasonably linked to influences from the BBC. The BBC argument is not only supported by chronological correspondence and dramatic events, but also material elements (Holberg 2000) like a bell beaker (Skjølsvold 1977), tanged-and-barbed bifacial points (Myhre 1979; Prescott and Walderhaug 1995; Østmo 2008; 2011), wrist guards and stylistic elements (Holberg 2000; Melheim 2011). It is reasonable to see the Bell Beaker regions in Jutland as the source of this transformation, though potentially one could argue that southerly western Norway and Jutland were subjected to the same processes. The distribution of type 1 flint daggers (Apel 2001; Prescott 2009; Sarauw 2008) indicate the routes of western Scandinavian interaction (Fig.2).

What drove BBC-people to move north along the coast of Norway and into the interior?
In terms of economy, it has previously been argued that the drive for pastures in societies with strong pastoral components could have been a motivation (Prescott 1995, 134). In terms of technology and production, the strong emphasis on metal re-renders the old idea of travelling prospectors as relevant (Childe 1957; Johansen 1983; Melheim 2011). On a more general level, the BBC-world’s drive for exotic objects (like pelts and antler, e.g. Prescott 1995, 137f.) could have fuelled a prestige exchange, creating a dynamic political
economy conducive to drawing ever new groups into alliances and networks. From the more political side of this political economy,competition between newly unleashed entrepreneurs could have generated a vying for new regional alliance partners

FrankN said...

Drago in the previous thread asked about my interpretation of the Tollense battlefield. I must say - the finds have left me as uncertain as anybody else, including well-known researchers such as T. Terberger, see
https://www.researchgate.net/publication/250308041_Gewaltsamer_Tod_im_Tollensetal_vor_3200_Jahren

The Tollense valley forms part of a major North-South axis, already used during the TRB period, that connects S. Sweden via Rügen, Potsdam, Riesa and Most to Bohemia. Therefore, an initial hypothesis was that the battlefield represents an attempt by locals to rob a trade caravan with armed escort that carried "southern" commodities, e.g. Slovakian copper, northwards. This hypothesis was a/o supported by frequent finds of "Silesian" bronze pins.
However, neither isotopic nor initial aDNA analyses supported such a scenario. Moreover, archeometalurgical analyses of contemporary Scanian bronzes speak against copper supply from Slovakia. Instead, Scanian swords were found to have either been made from Sardinian, and/or S. Iberian copper.
https://www.waughfamily.ca/Ancient/Moving%20metals%20II%20provenancing%20Scandinavian%20Bronze%20Age%20artefacts.pdf

Recent excavations of the battlefield area, but also along the A20 motorway that just runs 1 km east of the site, showed the existence of another major trade route, marked by a bridge over the Tollense and a wooden trackway leading to it. This route ran from the Lower Oder near Szczecin to NW Mecklenburg and ultimately Holstein, Jutland and Lower Saxony. It now looks that attackers coming from the west, equipped with bronze swords and partly on horseback, tried to conquer the Bridge, which was defended by bowmen and locals using wooden clubs.

The attackers are either identified with the Lüneburg Group, a culture with affinities to both the Tumulus Culture and the Nordic BA, or with early Urnfield (see Terberger above - this would a/o require a chronological revision, something that archeologists are hesitant to propose).
The defenders are currently seen as representing a western outskirt of the Lusatian culture (LC). The standard assumption is that the LC developed from Tumulus overforming Trzciniec. Early and sizeable LC presence in Vorpommern, a region quite afar from the Trzciniec area in E. Poland and W. Ukraine, plus apparent hostility between late Tumulus and LC, puts this assumption in question. The same holds btw. true for the early Lusatian presence in Brandenburg (e.g. Römerschanze Potsdam).
Unfortunately, so far there is very little known yet about Brandenburg during the MBA, i.e. after the Unetice collapse ca. 1600 BC, and the appearance of the LC around 1300 BC. Scarce finds there are commonly assigned to the Vorlausitzer Kultur (pre-Lusatian Culture, also known a/o as Silesian Tumulus Culture). For lack of settlement continuity to Unetice, pre-Lusatians are assumed to represent pastoralist immigrants. A typical pre-Lusatian artefact were bronze razors with horse-head shaped handles as e.g. also found with RISE61 (Kyndeloese/Zealand, R1a, Sr isotopic ratio consistent with an origin from around Leipzig, see previous blogpost).
http://archaeologie-welt.blogspot.com/2008/01/die-vorlausitzer-kultur-etwa-1500-1200.html

In summary – the Tollense finds call in question much of what was believed to be known about the MBA, and researchers apparently still struggle to place them in the wider context a battle of this size requires.

Gaska said...

VK531-The grave was situated above the natural harbour Skarsvågen at the NW end of the island Senja, overlooking the inner sailing route along the Norwegian coast. It was archaeologically excavated by P. Simonsen in 1964, who found no objects in the grave and no indications of a mound above-

Sailing Route-BB culture

Two wristguards are found in Norway (Rugland and Hitra) in association with a reworked type I flint dagger. This distribution is mirrored by some of the most extensive Scandinavian occurrences of the relevantly infrequent bifacial flint arrowheads with barbs and tang. These points are concentrated to South-west Norway, BUT ARE SPREAD NORTH ALONG THE COAST TO THE 62ND PARALLEL

The entire Norwegian west coast from north to south has sites belonging to the Bb culture, and the type of arrowheads is typical of the Iberian Peninsula

Therefore, we do have a Bb in the Arctic Circle, with a strong WHG component in its autosome and no evidence of steppe ancestry.


Anonymous said...

@FrankN

"Therefore, an initial hypothesis was that the battlefield represents an attempt by locals to rob a trade caravan with armed escort that carried "southern" commodities, e.g. Slovakian copper, northwards. This hypothesis was a/o supported by frequent finds of "Silesian" bronze pins.


Recent excavations of the battlefield area, but also along the A20 motorway that just runs 1 km east of the site, showed the existence of another major trade route, marked by a bridge over the Tollense and a wooden trackway leading to it."

This is not some ordinary battle to rob someone. No. This battle cannot be considered outside the context of the Zero World War - campaigns of the Sea Peoples, the destruction of Mycenae, trips to Egypt, destruction of the Hittite Empire and Troy, Philistinians/Muski, spread Cordoned Ware cultures, creating the Lusatian and Urnfields cultures, the Catastrophe of the Bronze Age, a radical change of cultures from elitist to egalitarian.

Andrzejewski said...

@All what happened that WHG related ancestry went almost extinct across the board during the BA where it was so dominant following the collapse of the farming communities across the continent b/c of plague? GAC was at least 20% forager. I also find it strange and odd that farmers had lots of mtDNA haps: H, V, HV, N1a, T, X, K etc albeit the dominant one among men was G2a. (J, E, T were also there but were very scarce before BA).

Could it be that Anatolia_N/Barcin populations were created by G2a men (Dzudzuana like or else) marrying women of diverse backgrounds back in Anatolia and therefore lots of farmers in Europe also had diverse phenotypes?

Davidski said...

@All

Here's a challenge to anyone who might know what they're doing...

What are the proportions of I1, R1a and R1b in the Viking Age and Iron Age samples from Scandinavia, and the proportions of commoners and elites within each haplogroup?

Simon_W said...

@ FrankN

"A typical pre-Lusatian artefact were bronze razors with horse-head shaped handles as e.g. also found with RISE61 (Kyndeloese/Zealand, R1a, Sr isotopic ratio consistent with an origin from around Leipzig, see previous blogpost). "

Are you suggesting RISE61 could be a migrant from the pre-Lusatian culture? What about his C14 date according to which he's dated to the much older Nordic MN B?

Would be interesting if he really was a migrant from MBA Brandenburg... AFAIK he's positive for the typical Scandinavian R1a marker Z287 and even for the downstream marker CTS8401, dated by yfull to 4200 ybp.

Drago said...

Frank
Thanks.
The concern is that if these guys are mercenaries / soldiers; they might not be representative of the local population. Hence; it would also be required to sample local “civilian” cemeteries associated with settlements

FrankN said...

Simon_W: My mistake - after all that reading about Danish and Scandinavian samples, I in the end mixed them up!

The guy buried with the horse-head razor was not RISE61, but instead RISE175, 1326-1192 BC, Abekas/SE, yDNA I1, 87Sr/86Sr 0.71277 (non-local).
http://www.archaeopress.com/Public/download.asp?id=%7B2862B30C-AB19-41C1-983E-D748A128A197%7D (pp. 51, 58)

Davidski said...

The paper doesn't claim that RISE175 was non-local, it just says that he may have been. An isotope ratio of 0.71277 certainly doesn't mean he was from outside of southern Scandinavia.

FrankN said...

Drago: "The concern is that if these guys are mercenaries / soldiers; they might not be representative of the local population."

Seems like bow & arrow were unusual for the Lüneburg Group, and in fact the whole Tumulus sphere, while occurring quite frequently east of the Tollense. As such, all those who died from arrows were most likely non-local.
Otherwise, however, I agree. In fact, we could do well with much more MLBA aDNA from Central Europe. Unfortunately, cremation was pretty popular during that time, and predominated LC "civilian" cemeteries (albeit there seem to have been some pre-Lusatian body graves).

[Someone has before asked about Schönfelder Culture aDNA. Schönfeld was also cremating. The closest in time and space we have to the Schönfelder Culture are the Quedlinburg BBs, but the extent to which they may incorporate Schönfeld admixture remains a matter of speculation].

Carlos Aramayo said...

@Davidski, Archi, and all,

The paper on Rakhigarhi will be released this September 5th, and will be somekind of coffin to "Anatolian hypothesis" and all variants of theories regarding Farming movements from Western to South Asia in Neolithic times, it´s all I can tell you by now.

Davidski said...

@Carlos

Thanks. Finally.

I've seen the older versions of the preprint a while back, so hopefully there's something new and awesome in this final version.

Carlos Aramayo said...

Yes David, but let's wait this three days.

Samuel Andrews said...

They better not do the 'parallel history of South Asia & Europe' crap. They do their best to make Andronovo & Yamnaya seem non-European.

I've seen Andronovo called 'Central Asian' because they lived in Asia for a few hundred years before coming to India.

The whole Indo European phenomenon is much more relevant to Europe than South Asia. Yamnaya lived in Eastern Europe. Corded Ware in Central Europe and served as the secondary homeland for Indo European languages. It is not a parallel history. It is better described as some Europeans went to Asia including South Asia.

I know I'm exaggerating but I feel like they don't want Europe to be the main player in any event in prehistory. So, they do their best to make Indo Europeans to be equally foreign to Europe & South Asia.

Samuel Andrews said...

Anyways, it will be interesting to see new ancient DNA from South Asia. The Iron age Swat Valley samples are a dead end. They're simply the direct ancestors of modern Sindhi (who are outliers in SOuth Asia) not a population heavily affected by the initial expansion of Andronovo into South Asia like people hoped.

I'm pretty sure people of mostly Andronovo origin migrated into South Asia. Because, as far as I can tell Indians don't have a noticeable BMAC signal. Brahmin are marked as different by their extra Andronovo ancestry not BMAC ancestry.

Btw, most Indians only have 13-16% Andornovo. That's not a lot at all. Expansion of Indo Europeans is much more relavent to Europeans' ancestry. Hopefully, researchers will acknowledge this important detail and do the parallel history stuff.

Samuel Andrews said...

If Anatolian & Iranian farmers are Middle Eastern. Then Yamnaya and certainly Andronovo are European.

Carlos Aramayo said...

@Samuel,

But we should remember that alternative hypothesis to steppe one were coined and promoted not by South Asians but by Europeans like Colin Renfrew, and recently by linguist Paul Heggarty.

I hope this debate can be followed by strict academic canons and not distrust to the other, as both sides have a common root in early proto Indo-European speaking people millennia ago. This common origin should join them in a scientific effort to know the evidence.

The forthcoming paper can be a proof of the union of specialists from India and abroad, toguether making science.

Carlos Aramayo said...


@Samuel,

I agree with you Swat samples are "not a population heavily affected by the initial expansion of Andronovo into South Asia like people hoped". Taking more samples from Mature Harappan period is needed, specially at 3C Harappa period (2200 to 1900 BC).

Drago said...

Carlos is confusing Renfrews hypothesis with Heggartys . They’re not actually the same

Davidski said...

Yeah, Renfrew's hypothesis was actually more coherent than Heggarty's, believe it or not.

Carlos Aramayo said...

@Drago,

My comment was not intended to match what Renfrew proposed to what Heggarty done with Renfrew´s hipothesis. I wanted to highlight both are "Europeans".

Carlos Aramayo said...

Anyway, Renfrew and Haggerty´s hypothesis, will be seriously challenged in the forthcoming paper.

Katharós said...

It’s actually difficult to tell who killed (sacrificed) who and why in retrospect. Foremost commoners have the tendency to sacrifice their own kind, as the Elite have the tendency to sacrifice people of higher status as an ultimate offering. “Everyone according to their possibilities” And to be quite frank, Christ is viewed as the ultimate Elite sacrifice and for the same reason Islam rejects the idea that a Messiah can be scourged "sacrificed" like a common criminal.

Davidski said...

A couple of these Scandinavian individuals may have been sacrificial victims. But it's impossible to be sure.

Perhaps the remains of RISE98 and RISE276 show trauma associated with specific methods of sacrificial execution, but I haven't been able to find any information about that.

The main point is that they weren't buried in any way that would indicate high, or even higher than average, social status.

RISE98 was buried with a bone needle, and may have been a commoner who died a natural death, while RISE276 was a bog body so he may have been executed for a crime.

epoch said...

@Samuel

There is some linguistic evidence that there was more than one movement of IE's into South Asia.

https://www.hf.uio.no/ikos/forskning/publikasjoner/tidsskrifter/acta/volum_77/ao_2016_cpz.pdf

Provided by FrankN, coming from this thread:

https://eurogenes.blogspot.com/2019/08/the-pie-homeland-controversy-august.html?commentPage=2

zardos said...

"[Someone has before asked about Schönfelder Culture aDNA. Schönfeld was also cremating. The closest in time and space we have to the Schönfelder Culture are the Quedlinburg BBs, but the extent to which they may incorporate Schönfeld admixture remains a matter of speculation]."

That's a pity. Because their role in the formation of later cultures, if they had any, especially Unetice could have been interesting, as would have been their yDNA.

Matt said...

@Sam, I still believe it seems difficult to tell if there is any further BMAC-like ancestry it South Asia pre and post-Harappa, without Harappa samples. The BMAC outliers are a long way from home if they are migrants from IVC, and may be locally admixed with BMAC or of Eastern Iran and of Afghanistan and not representative of who was in South Asia at the time. If they work, but have further BMAC ancestry which true Harappans lack, they will be "eating" that ancestry in models.

@Carlos Aramayo: You mean theories specifically tieing IE languages to migration to South Asia by Western Asian farmers in Neolithic times, and not migration of Western Asian farmers to South Asia in a general sense? Because the latter, though not necessarily at a neolithic time (could even be relatively recent), seems necessary to explain the genetics. (Of course if this did not happen by Harappa that's a big shock for the already shocked Renfrew idea, but a big shock for models as presented by Narasimhan too).

@Katharos, that is an interesting point. To try and solve it, are there any means we have to learn about status of these individuals, removed from consideration of their grave goods? E.g. biological status indicators like height, diet, skeletal indicators of malnutrition. Even if it's not specific to the individuals and just a wide survey, that could tell us whether sacrifices ('bog bodies') in the periods in question tend to be disproportionately 'unfree' or 'common' or 'elite'.

Davidski said...

@Matt

The BMAC outliers are a long way from home if they are migrants from IVC, and may be locally admixed with BMAC or of Eastern Iran and of Afghanistan and not representative of who was in South Asia at the time.

Not really.

From what I've seen, only some of them can have recent ancestry from Turan. The others should form their own cline to ASI that will look different from the Turan > ASI cline.

But obviously we'll have to revisit this issue after we've both had a look at the relevant new data.

Matt said...

@Davidski: Sure. We should discuss again after Thurs, although like with Shirenzigou, it may take until we are able to look at the samples in a more G25 PCA and rich qpAdm way.

The others should form their own cline to ASI that will look different from the Turan > ASI cline.

It's possible, only that SIS2 and Gonur2 don't look much like that to me thus far. In reprocessed G25, the three samples so far look like a straight cline from SIS3 to BMAC, albeit the less ANE+Anatolia shifted samples in BMAC, but not a cline to IranN: https://imgur.com/a/tDaygU7

There's some higher dimensionality that can't seemingly explain, but it's less clear to me what those drift paths are about (may have something to do with CHG and East Asian outgroups?), and BMAC overall seems a better fit than IranN overall...

Anonymous said...


Aryans invaded Northern India (Pakistan) at the final stage of the Andronovo culture, in fact, post-Andronovo, in time of the Cordoned Ware culture. Cordoned potteries are present North of the Indus, it is the same as in the Russian steppes/Central Asia. In fact, this invasion simultaneously the invasion of the Sea Peoples in the Mediterranean. South of the Indus, the Arias penetrated much later, therefore, no longer have any connection with Andronovo.

Matt said...

@epoch, may or may not relate to that article (as I haven't had time to read it) re:Indo-Aryan structured dispersals, but was having a look at one of latest Jena/Planck papers on lexical phylogeny (https://www.jbe-platform.com/content/journals/10.1075/jhl.17009.ver - "Detecting non-tree-like signal using multiple tree topologies"* (since we're discussing their models again).

Paper is not sci-hubbable yet (as recent and small journal I guess), but trees are here (https://zenodo.org/record/2653209#.XW4bIYGYXIU), though not based on new data.

These do not use the Chang paper constraint (attested ancient varieties must be ancestral to extent ones) and in light of your comment (to get to the point), it was interesting to me that the degree to which Vedic Sanskrit was a non-ancestral side-branch to Indo-Aryan languages attested today is quite strong, while the comparative effects for Ancient Greek, Classical Armenian, Latin, Old Irish, Old Church Slavonic, Old English (in descending orer) are all quite weak by comparison.

Although when Garrett has presented the issue in conference he's mainly talked about Latin as the example, the effects of compressing dates in Chang from making ancient attested varieties ancestral to modern (to avoid the issues they believe arise from homoplasy) probably arise mainly from the effect of their change within the Indo-Iranian branch.

That all seems potentially reasonably interesting in light of talking about substructured Indo-Aryan and Indo-Iranian dispersals?

*which seems what it sounds like; generally it seems like the lower P multiple topologies for IE (which give rise to the non-treelike signal) do involve allying Indo-Iranian to Balto-Slavic and Greco-Armenian in some respects.

Davidski said...

@Matt

There's some higher dimensionality that can't seemingly explain, but it's less clear to me what those drift paths are about (may have something to do with CHG and East Asian outgroups?), and BMAC overall seems a better fit than IranN overall...

Indians more or less form a cline between SIS3 and Sintashta.

PCA of ancient West Eurasia

Hodo Scariti said...

Also the Narasimhan's paper revised is waited... wonder when it will be published...

Matt said...

Hmmm, that combination seems fairly off cline to me: https://imgur.com/a/UNCWB3l

Taken broadly enough, even Yamnaya+SIS3 would get somewhat close for many South Asians though, just would not work as a cline for all.

Matt said...

Mentioning using the West Eurasia 9 PCA on South and Central Asians, I was curious to run it on the ancient South and Central Asian and Steppe and PCA that I did on G25, and which broadly replicated the PCA from Narasimhan. Since you can get more samples on there which fell off the G25.

Doing so, it looks to me like more samples are outliers that are labelled as such, and the dataset labelling is fairly conservative.

See: https://imgur.com/a/NdwreV5

The following look like outliers to me:

I6893 - PAK_Swat_Saidu_Sharif_IA (My label: PAK_Swat_Saidu_Sharif_IA_o2)
I1992 - PAK_Swat_Udegram_IA (My label: PAK_Swat_Udegram_IA_o)
S3261.E1.L1 - PAK_Swat_Udegram_IA (My label: PAK_Swat_Udegram_IA_o2)
S8527.E1.L1 - TKM_Geoksiur_En (My label: TKM_Geoksiur_En_o)
I1787 - TKM_Gonur1_BA (My label: TKM_Gonur1_BA_o4)
I6217 - TKM_Gonur1_BA (TKM_Gonur1_BA_o3)

(Also relabeled I1792 as TKM_Gonur1_BA_o2, since the differences between them and I1783 in West Siberian ancestry is large and it makes the plot cleaner).

Particularly of interest re South Asian is probably I1992 and S3261.E1.L1. Udegram represents the earliest Swat sequence, and I1992 respectively looks strongly steppe_MLBA shifted while S3261.E1.L1 looks similar to Indus Periphery. Although I'm not sold on the remaining samples as admix between these two, combined with the early date, that's interesting and gives some more credibility to modelling the samples using Indus Periphery (though in truth I'm not sold and it may be explained with extra Iran N ancestry instead).

I6893 from Saidu Sharif, much later and from the Buddhist Period site, looks like an admix between a more strongly Steppe_MLBA population and more ASI than typical, as previously discussed.

The remaining "extra" outliers look Anatolian or West Siberia admixed, except I1787 from Gonur, who looks almost like an Iran_N survivor or something harder to place.

Datasheet if anyone wants: https://pastebin.com/T4ictuTP

But yeah, probably best to discuss this again after Thursday!

Anonymous said...

Due to the Vedic Indo-Aryans were synchronously the Urnfields and Lusatian cultures, they also only cremated.
This changed only with the advent of Buddhism which denied cremation, so the genetics of the Indo-Aryans will have to be told by Buddhist burials.

Samuel Andrews said...

90% of what the harvard ancient DNA people do is awesome. This includes what they do for Indo Euroepan-related stuff. They're great in almost every way.

But, I hope they end the misconceptions about early Indo Europeans. Which are....

-They lived in Central Asia.
-They were master horse drawn warriors similar to Scythians, Huns, Mongols.
-They are as relevant to Asia as to Europe.

The Nova documentary on 'Horse Warrior' Indo Europeans is the biggest public coverage this topic has gotten. It was full of misconceptions on Indo Europeans and left out so much important information.

I say researchers, should intentional explain how the Indo European phenomenon happened before the Scythians/Turks/Mongols and was very different. That they weren't horse warriors. That they lived in eastern Europe not Central Asia.

Andronovo & its decendants did become Central Asian horse warriors. But, the spread of Indo European languages in Europe had nothing to do with Central Asia or the horse or the wheel.

Gaska said...

@Sam said "The whole Indo European phenomenon is much more relevant to Europe than South Asia"

That will be for the Anglo-Saxons who are culturally close to the Fiji Islands, not for the Basques who are not only Western Europeans but also speak a non-Indo-European language. Kurganists only have "steppes" in their brains, and for us Basques your "steppe ideology" is funny, absurd and pathetic. The British Anglo-Saxons and their American relatives are going to make a universal ridicule but it doesn't surprise me considering that you are already practicing with Brexit etc etc.

@Sam Andrews said-"90% of what the harvard ancient DNA people do is awesome"
.
90% of what Harvard says is awsome because it matches what you think, the remaining 10% is wrong because you don't like their conclusions. For me and for many Europeans Harvard is not only an overwhelming disaster but it is becoming a lab with which it is better not to collaborate. Last year more than 100 French,Portuguese and Spanish archaelogists (including the president of European archaeologists) signed a manifesto against the provisional conclusions of Olalde's last paper for considering it biased and little rigorous. Olalde said (2.018) that there was no genetic exchange between Iberia and the other BB regions because in Iberia the mit-lineage H3 has not been found in this period when in Spain there are more than 10 samples of this haplogroup in the Neolithic and Chalcolithic including the BB period. Is it that they have not studied enough?, or do they not know enough about genetic databases in Spain despite being Spaniards? This is a disaster. Obsession with the steppes has turned European genetics into hell.








Davidski said...

@Matt

Hmmm, that combination seems fairly off cline to me: https://imgur.com/a/UNCWB3l

You're trying to connect the wrong dots there. As I've been saying, it's likely that some of those Indus_Periphery samples have minor Turan ancestry.

But SIS3 most certainly doesn't have any such ancestry, and the SIS3 > Sintashta cline is where most Indians are sitting. That's because they don't have any perceptible Turan ancestry.

Of course, South Asia is more diverse than just the Indian cline. In Pakistan and surrounds there are all sorts of very recent influences, from Central Asia, Arabia, Iran and even Africa, but these are irrelevant to the genetic structure of the IVC people.

Anyway, I think you're in for a major facepalm moment later this week. :)

Matt said...

@Sam, I would agree that it's a distortion that tries to oversimplify the situation to present Europe and South Asia as particularly parallel, in genetic terms (leaving aside language). What happened in SA and Europe is not particularly more parallel than what happened in China and SE Asia, or within the Middle East, etc. I don't think it convinces South Asian people either, or placates the people who viscerally would object to any steppe/European related ancestry in South Asia.

They should present the data directly, and not try to make explicit a parallel that is not particularly the case.

@Davidski, would you mind posting up a version of your plot that does connect SIS3 and Sintashta then (The right dots?), to your mind, and which cuts through the Indian cline? Your claim seemed to be "Indians more or less form a cline between SIS3 and Sintashta", hence I drew a cline cutting through SIS3 and the South Asian cline (which did of course not point to Sintashta, because it wouldn't!). I don't remember you saying that you thought the Gonur2 or SIS2 samples have any Turan ancestry though rather than being direct migrants from the Indus.

My claim is that the Indus_Diaspora 3 may be mixed, particularly other samples than SIS3 may be quite heavily admixed with local people from BMAC and SIS, and that if the Harappan samples are actually more like SIS3 (or even slightly more AASI!) then models like Chad Rohlfson lays out here may be more relevant for South Asia, involving significant extra BMAC, rather than the model presented by Narasimhan - https://populationgenomics.blog/2018/08/08/another-look-at-south-asian/. (That in turn offers some different potential routes for expansion of IE languages than the scenarios suggested by Narasimahan.)

I don't have any idea what the Harappan samples will actually be like of course, it just seems rather a big assumption that the average of the Indus_Periphery (or Reich's fantasy points) are representative of them.

Davidski said...

@Matt

Would you mind posting up a version of your plot that does connect SIS3 and Sintashta then (The right dots?), to your mind, and which cuts through the Indian cline?

Here you go...

Sintashta > Indus_Periphery cline PCA

Keep in mind that this doesn't have to be too precise for now, because there are other data points that make this solution more likely than others, such as the high frequencies of R1a-Z93(L657+) in upper caste Indo-European-speaking Indians.

Where does L657 come from in your mind and how did it reach such high frequencies in Indo-Aryan speakers?

Also, I did actually comment in a couple of blog posts that Indus_Periphery Gonur2_BA and the Swat_IA samples might have some recent ancestry from Turan...

The South Asian cline that no longer exists

The protohistoric Swat Valley "Indo-Aryans" might not be exactly what we think they are

But even if this is true, it doesn't have any linguistic implications, because most Indo-European-speaking Indians don't have this type of ancestry. Unless they're from some relative outlier groups in the north, they're basically a mixture of SIS3-like and Sintashta-related populations. Some have higher or lower ratios of local hunter-gatherer ancestry, but that's to be expected, since SIS3 is just one individual and can't be representative of the entire IVC population.

You're surprisingly stubborn about this, I must say.

Samuel Andrews said...

My history class will do its first day on ancient India on September 5th the same day that ancient DNA papers is supposed to be published.

Andrzejewski said...

Some very interesting read re: Scotland’s genetic profile:

https://www.thesun.co.uk/tech/9847652/scotland-genetic-map-revealed-scientists/amp?utm_medium=Social&utm_campaign=sunmaintwitter&utm_source=Twitter&__twitter_impression=true

Chad said...

I went back and did SIS3, BMAC and Andronovo in the pleft for all Swat samples. They all still want at least as much BMAC as Andronovo, to 3x as much. I'll post outputs tomorrow.

Davidski said...

@Chad

You really should wait for the Rakhigarhi paper.

Drago said...

Matt/ Davidski
In general; in playing with PcA based projections; what samples are included can affect the clines generated. Eg sometimes you want to exclude samples to have less clutter & focus on one region; but that plot might sit slightly differently to a world PCA

Davidski said...

It's just one of the lines of evidence, along with other types of plots, formal stat-based graphs, uniparental markers, etc.

Anonymous said...


The Indo-Aryans can only be judged by Buddhist burials.
Burials in earlier cultures refer to Dasa, Dasyu, Shudras.

natsunoame said...

…to make general statements only on the basis of DNA analysis is not simply elemental but idiotic. There are, however, many contemporary examples of such a language of a similar origin groups being different, and conversely different DNA speaks a similar language.
Human flows were intense and diverse.
To try to explain these processes without taking into account history, archeology and cultural parallels, old toponymy (giving indications about people inhabiting this area during this particular period), language parallels, religious beliefs and deities / which are of great importance to people in the past /, is just a jumble of scientific work and of course the conclusions we can expect accordingly.
If you consider continue this way go to a fortune teller better …she can guess at a higher rate with her methods.
Your later posts sounds closer to that anyway.
I am sure DNA samples are pretty easy to manipulate/replaced, more easier than history. There are too many interests here, political too, to leave that results without control.

zardos said...

The results are not that easy to manipulate and since new, more samples will come in, which might be processed by different people rather sooner than later, any forge not involving the majority of scientific workers is at constant threat of being exposed.
So any kind of deliberate manipulation would have to be a huge conspiracy involving so many different people from different backgrounds that it is not feasible, especially not for long.

The results can be distorted and misinterpretated though, but even that was much easier in the other disciplines you mentioned. Or do I have to remind you on all the crap produced by the "New Archaeology" since the 1960s with 90 percent attributable to Cultural Marxism and "political correctness" which grew out of it.

The great outlines being now known for more than 120 years, at least for the metal Ages. Some truths were just questioned for the sake of it, because they were "unpleasant tales" for the new grand narrative of a "naturally peaceful and tolerant" humanity which was just "poisened" by material competition and "bad" ideologies/ religions.

The good thing about aDNA studies is that they deliver hard data, which manipulations can be grand too, but these are self-limited and oftentimes have a date of expiration.
So all sides, regardless where you are coming from ideologically, have to adapt their views and position to sometimes unexpected results.
At some point to argue against some of the more obvious truth from genetics will become like someone arguing for a flat Earth or against evolution in general hopfully: Only morons, insane and delusional subjects.

From were do you expect more reliable data? Seriously.

Samuel Andrews said...

About Anatolian languages. No way had they been in ANatolia since 3000 or 4000 BC.

What it probably is, is ANatolian languages originated OUTSIDE Anatolia. The language family is that old but it hadn't been in Anatolia for that long.

Samuel Andrews said...

Also it is pretty obvious the whole Late Proto Indo European thing that linguistics talk about is simply Corded Ware.

Almost all Indo European languages in Europe derive from Corded Ware. Indo Iranian derives from Corded Ware. Hence, most Indo European languages share common origins in 2800 BC.

The expansion of Indo Europeans can be summarized as.....
Russia in 4000 BC.
Took over most of Europe in circa 2800-2000 BC.
Then from Europe, they went into Central/South Asia in 1800-1500 BC.

The IE expansion in Europe between 2800 and 2000 BC had the biggest demographic/genetic impact. But, it had NOTHING to do with chariots or horses.

Chariots & Horses were important in the Indo European expansion in Asia between 1800 and 1500 BC.

But, arguable the expansion in Europe was more impressive and important and widespread. So, the idea ALL Indo Europeans were charioteers and master horse warriors needs to end. It was only the Indo Iranians in Asia who did that.

I guess Myceneans & Hittites might have gotten chariots from late Indo Europeans in eastern Europe related to Andronovo (Indo Iranians).

Anyways, only the second late half of the story of Indo European expansion involved horses.

zardos said...

While I agree on the rest, the presence and importance of horses in the CWC can be debated.

zardos said...

Also, while I agree on Anatolian languages to be from an early branch in Europe in general, the exact archaeological culture which might be the vector for the early spread of IE in Anatolia is what matters. Your position?

Andrzejewski said...

@Samuel Anatolian languages date in Anatolia circa 1,800 BCE, after defeating non-IE (CHG rich ANF?) Hattian and Hurrians. They intermingled with them so extensively that the R1b signal got swamped

Andrzejewski said...

@Samuel “Also it is pretty obvious the whole Late Proto Indo European thing that linguistics talk about is simply Corded Ware.

Almost all Indo European languages in Europe derive from Corded Ware. Indo Iranian derives from Corded Ware. Hence, most Indo European languages share common origins in 2800 BC.”

If you read my posting history, you’ll see that I was saying it repeatedly all along: Late PIE is not anything to do with Yamnaya but CWC one’s.

But apparently some posters disagree with me, thinking that Corded Ware spoke some archaic form of Balt-Slavic...

Besides, CWC being an offspring of Sredny Stog may readily explain the non-IE agricultural substrate in European IE languages: its from GAC and Cucuteni influences within Sredny Stog (which themselves were ~18% EEF), not from any substrate from later expansions.

Leron said...

The great Danube appears to be the missing link between CWC LPIE and proto-Anatolian.

Anonymous said...

Troy after 2000 BC is wagon's culture with horse breeding, so Luwians already know how. They did not know chariots, but they knew wheel and wagon.

PIE obviously didn't know chariots and domesticated horses, but know wheels and wagons.

Mycenaeans knew and used the chariots and the domesticated horses with the Babyno culture.

Aryans migrated from Europe to Asia circa 2200-2100 BC.

natsunoame said...

zardos
...as I wrote earlier, information from DNA analysis alone means nothing in itself, only part of the puzzle.
You cannot interpret the results only with knowledge in this field, because they are misleading.

Anonymous said...

@natsunoame

In the case of genetics you have sources (BAM files, etc) and you can check on them all that is written, but in the case of archaeology and anthropology sources no one has, they are usually seen one or two people and in most cases they are no longer available and no one can not check.

Carlos Aramayo said...

@David, and all

Here´s the official communication in press, tomorrow is the day for the paper's publication on Rakhigarhi, and the team will have a public presentation in India next September 6th.

Niraj Rai commented on Twitter:

"...Who were the custodians of Indus Valley Civilization? How farming in South Asia arose? Our first ever ancient DNA results on IVC will be published online on 5th September, 2019. We will share more information on our findings on 6th at National Museum, New Delhi..."

You can find more of it in:

https://tinyurl.com/y6job46v

Regards...

Leron said...

Not to mention that archaeology is not immune from fraudsters and forgeries. One famous case being Luwian expert Mellaart, who used his expertise to create hoaxes that pushed his personal fiction about the Luwians. Actual remains may have actually been covered up or destroyed because it went against his narrative.

Samuel Andrews said...

@Andre,
"If you read my posting history, you’ll see that I was saying it repeatedly all along: Late PIE is not anything to do with Yamnaya but CWC one’s."

You got that one right. Congrats. It didn't click for me that LPIE is is just Corded Ware till last week. Yamnaya probably spoke somekind of basal IE language. My guess is Anatolian languages are Yamnaya who fled to the Middle East after LPIE Corded Ware invaded Russia.

The Yamnaya thing which Harverd geneticists keep saying is wrong. Yamnaya is not a big player in the expansion of IE languages. Corded Ware (Late Proto Indo Euroepan) is the big player.

Samuel Andrews said...

Thanks for the info Carlos. It's kind of disappointing though. No Bronze age DNA from India yet.

Anonymous said...

Yamnaya did not speak on the PIE. The realities of the Yamnaya culture and PIE vocabulary do not correspond to each other. The working version by using the genetic is only that Yamnaya spoke para-Tocharian proto-language.

Matt said...

@Davidski, that.... seems visually pretty loose to me (and not Sintashta+SIS3).

I think if you're proposing that Sintashta+SIS3, then that's basically pretty close to proposing that ASI+Sintashta can synthesis the modern South Asian cline, because SIS3 is pretty close to the base of the South Asian cline (although other Indus Diaspora sampled by the Reich Lab may have been more so): https://imgur.com/a/lQVWXLx

That really seems to go against all the academic work (and blogging on fits) on the profound and strong influence from IranN/BMAC related populations have in the South Asian cline beyond a Sintashta+ASI synthesis, and which is folded into the average of the "Indus_Diaspora", and I have trouble accepting that.

Re; R1a-Z93, I'm not proposing no influence from Steppe_MLBA in what changed in South Asia, so that's not inexplicable. There certainly has to be some Steppe_MLBA ancestry to make sense. Only that Steppe ancestry may not be the only thing that changed, and instead there may be influence from BMAC, as covered by Chad and others consistently with the archaeology, more so if it turns out the Harappans were quite SIS3/ASI like. This could allow for other models for dispersal of Indo-Iranian, ones that might explain deeper time depths and conflicting features than proposed in the models of repeated dispersals from Steppe MLBA.

Anyway, the article Carlos links provides Rai saying "Comparing Rakhigarhi with data from modern Indian populations, we have concluded that they have more of an affinity with the ancestral South Indian" (e.g. ASI) "tribal population compared to the North Indian population".

Now lets just wait a day or two and see how ASI they were! (And whether Reich's claims of the post-Harappan formation of ASI through truly massive absorption Indian Hunter Gatherers were right at all, then massive dispersal through India somehow, or something close to ASI formed much earlier).

Samuel Andrews said...

I bet, Corded Ware were the ones who killed those 3 Globular Amphora families. The massacre happened 50-100 years before Corded Ware expanded across Central-Eastern Europe.

They were killed with axes which was the favorite weapon of Corded Ware. Globular Amphora used axes too but.....

Andrzejewski said...

@archi “Yamnaya did not speak on the PIE. The realities of the Yamnaya culture and PIE vocabulary do not correspond to each other. The working version by using the genetic is only that Yamnaya spoke para-Tocharian proto-language.”

Tocharian is most likely an Andronovo, not Afanasievo product, so it pulls the rug from under your feet. I suspect OTOH that Yamnaya has to do more with a CHG mother language

Leron said...

@Andrzejewski Wouldn't CHG have more to do with Maykop and the proto-North Caucasian languages?

Anonymous said...

@Andrzejewski
"Tocharian is most likely an Andronovo, not Afanasievo product, so it pulls the rug from under your feet."

We don't have a single genome that would be reliably owned by non-Aryan Tocharians, so the proto-Yamnaya (R1b-Z2103) -> Afanasievo (R1b-Z2103) -> Tocharians (R1b-Z2103?) hypothesis is still alive and not rejected. Archaeologically Andronovo is not associated with the non-Aryan Tocharians.

Drago said...

@ Sam

“About Anatolian languages. No way had they been in ANatolia since 3000 or 4000 BC.

What it probably is, is ANatolian languages originated OUTSIDE Anatolia. The language family is that old but it hadn't been in Anatolia for that long.”

We all enjoy Sams “hey everybody....” proclamations; however your need to qualify that with some sort of evidence & synthesis
In any case; the reality could be very much the exact opposite of what you suggest - probably movements into Anatolia (began as early as M5) with Anatolian languages forming within anatolia

Drago said...

@ Matt
I haven’t looked much into that region; but wouldn’t Turkmenistan IA <—> SAS -3 cline work well ? (Even if some would argue that’s too late for IA, etc)

Davidski said...

@Drago

Turkmenistan IA has a lot of BMAC-like ancestry, so it only really works in South Asia for eastern Iranians and Indo-Aryan groups living among eastern Iranian populations.

It doesn't work for Indians by and large, because Indo-Aryan Indians basically lack ancestry from BMAC.

They're essentially a two-way mix between Steppe_MLBA and Indus_Periphery.

Anonymous said...

The inhabiting of the Anatolians (Luwian) out of Anatolia before Mycenaeans time Herodotus described, and they had inhabit from Thrace to the Ionian sea. Teucrians as we know is the Trojans, and the Trojans spoke Luwian language, what remained of the inscription, and theirs Luwian names.

Hdt. 7.20.2 "neither is the one of the Mysians and Teucrians which before the Trojan war crossed the Bosporus into Europe, subdued all the Thracians, and came down to the Ionian sea, marching southward as far as the river Peneus."

Of course, no resettlement of Trojans in Europe was not, it is not confirmed by archaeological data. Just in the times of Herodotus and later by all and sundry, drew their genealogy from Troy, and accordingly, if Balkans were people related Luwians, they considered themselves natives of Troy, not by their ancestors.

Drago said...

Davidski
However the Tkm_IA sample dates to 1200 BC ?
That would bring it to the realm of possibility ; matches when the data in Narasimhan preprint shows MLBA ancestry appearing in swat
Anything earlier (2000-1500) wasn’t convincing

FrankN said...

On Anatolian: There are some IE borrowings already in Sumerian, but most of them relate to commodities (wool, copper, slaves, etc.) that may have been traded from quite afar.
We also have "horse & wheel" terminology in Sumerian, Akkadian and Hurrian that connects to IE terms. However, here, it is unclear whether e.g. Sum. sí-sí 'horse' ↔ Akk. sisium 'horse' ← Hurr. issi(a) 'horse' ~ PIE *h1ekwos 'horse' ultimately go back to (satemised) IE, or represent independent IE and Hurrian borrowings from a third, yet unknown source.

A term that acc. to most linguists was clearly borrowed from IE, most likely Anatolian, is Akkadian targumannu "interpreter, translator" (c.f. Engl. dragoman, Medieval Latin dragumannus, Middle Greek δραγομάνος, Spanish trujamán, Ge'ez (Classical Ethiopic) t-r-gw-m-, Turkish tercüman etc.). The term is attested from ca. 2000 BC onwards. The best accepted etymology is from PIE *trg "to turn" (c.f. Lat. torquere "to twist", German drechseln "to turn wood", see Engl. "to convert" for the semantic relation between "turning" and "translation"). -annu is a well known Hittite (Anatolian) suffix turning verbs into agents, roughly comparable to Engl. -er (work->work-er).

In this case, we are not dealing with some commodity, or a regional wanderwort, but a Superstrate term of high political and/or commercial relevance. The fact that Akkadians borrowed this superstrate term from IE (most likely Anatolian) signifies that around 2000 BC Akkadians were dealing with an IE-speaking unit of substantial political and/or commercial weight, a weight certainly superior to anything expectable from recent pastoralist entrants.
Morever, Akkadians should have been in direct diplomatic and commercial relations to this IE-dominated unit, so we are talking about an area closer to Akkad than, e.g., Troy. Kanes (Kültepe), which relates to the Hittites self-designation as nešili, might fit here, but other areas such as SE Anatolia (around Antakya), where a mixed Luwian-Hurrian population is attested from ca. 1600 BC onwards, look equally possible.

natsunoame said...

Archi
Well, well... they reached even more than that. And they left something more than culture and knowledge.

“Pelasgians, after wandering over most of the habitable earth and subduing most of mankind, settled down on that site, and that from their strength in war they called their city Rome.“ -Plut.Rom.I.1.

Pliny the Elder does not hesitate to admit that in Latium, the alphabet was brought by the Pelasgians: "... litterarum usus in Latium eas attulerunt Pelasgi." -Plin.VII.193. As the kings of the Sylvian dynasty and descendants of Aeneas and Askani, Titus Livy refers to the Thracian names Attis, Kapus and Capetus -Liv.I.3.

“The Thracian party become the ancestors of the second colonizing race, Firbolgs.” -Encyclopaedia Britannica, Dictionary of Arts, Sciences, and General Literature, 9th ed. Vol.V, Henry G. Allen and Co, 1833, p.299.

Prof. Karl Müller: "The Thracians who settled in Pieria, near the Mount Helikon, being the creators of the veneration of Dionysus and the Muses, as well as the creators of Greek poetic art, the people of greatest importance in history of civilization. ”- The History and The Antiquity of the Doric Race, translated by H. Tufnell, GCLewis, J. Murray Publ. London, MDCCCXXXIX, p.10

Davidski said...

@Drago

TKM_IA DA382 is dated to 911-799 calBCE, and wasn't the first person with steppe ancestry in the area. So yeah, people like that may have had some impact on the Iron Age populations in the Swat Valley.

Drago said...

Right; it’s Kashkarchi that’s 14-1200 BC; and that sits near Sintashta
If that’s the case; the implication is Swat valley samples lie somewhat off that cline

Chad said...

My output will have to wait a day due to my baby being sick, but it is pretty clear that BMAC input is there in the Swat Valley. The interaction is between Uzbekistan, Tajikistan, Andronovo, and IVC. Sapalli culture goods are very visible all over the Gandhara Grave Culture. I'm doing both runs with standard sets and including outliers. Not sure how this was missed.

Davidski said...

I agree that the Iron Age Swat samples have BMAC or BMAC-related ancestry. This is visible in their uniparental markers too.

But actual Indo-Aryan-speakers from India are a different matter. They're basically a two-way mix between Andronovo and Indus_Periphery, plus often some extra ASI.

So no matter what we say about the Swat samples, they're not the game changer when it comes to the Indo-Europeanization of India, because they had no role in this.

Chad said...

I have to disagree. Swat Valley, plus ASI and SE Asian works. Making inferences about India without any samples from there is pure speculation. Even my own, but it works. There were too many movements into and within India to make everyone that simple mix. Without data from India no one can say who came in or how IE reached there.

Davidski said...

Swat Valley, ASI (or whatever you're substiting it with, considering there's no ASI genome) and SE Asian might work, but only as an abstract solution.

You should be working with Indus_Periphery, because that will get you closer to the real answer.

Carlos Aramayo said...

@David,

I was made a mistake, I put Andronovo instead of Sintashta. Anyway, is the mixing between Sintashta (2100-1800 BCE) and IVC cline, or is it between Andronovo (2000-900 BCE) with ICV cline?

Davidski said...

@Carlos

Swat valley samples show no ancestry related to present day inhabitants over there.

Maybe they do to some groups, it's hard to say for sure, but anyway there does seem to be excess Anatolian ancestry in much of Pakistan and surrounds, which is probably related to the later movements of west and east Iranic speakers into the region.

But I would like to know if you consider Andronovo chronology from 2100 to 1800 BCE is accurate or not, in order to check the possible period of admixture between Andronovo and Indus Valley people.

I'm not up to date with the Andronovo chronology so I can't comment about that.

We really have to wait to see what the new samples in the final version of the Narasimhan et al. paper are like. I've heard that the Swat outliers look interesting, with a lot of steppe ancestry, and one apparently has an early date.

Davidski said...

And the individuals from the main Sintashta and Andronovo clusters are very similar.

This is the type of steppe ancestry that most South Asians have, so it's hard to say which Sintashta-like steppe group it came from exactly.

If and when R1a-Z93(L657+) shows up in these groups we might be able to get more specific than just Andronovo/Sintashta-related.

Carlos Aramayo said...

@David,

Yes, we should be more specific regarding Andronovo, which two widely spread and maybe its earliest "subcultures" are Fedorovo (1900-1400 BCE) and Alakul (1800-1500 BCE) as Parpola (2017: 248-249)in his paper called "Finnish vatsa - Sanskrit vatsá" suggests. Regarding Sintashta I think the most updated chronology can be found also in this paper as (2200-1800 BCE)based on Chernykh (2009: 128-133).

Gaska said...

Italian data begins to arrive

"The Bronze Age migration from the Pontic-Caspian steppe region has been linked to the arrival of the Indo-European languages in mainland Europe. Our identification of an additional substantial component in Italy possibly arriving in the Bronze Age raises the possibility of multiple Indo-European waves into the continent. Similarly, the persistence in Italy of non-Indo European languages into historical times (e.g., Etruscan) could be linked to reduced SBA penetration along the peninsula"

The game has just begun, we are going to have fun

Davidski said...

Sounds like nonsense, because early Italics are basically a mix of Europe_LNBA (with lots of steppe) and Europe_MN (with no steppe). They don't have any significant extra CHG.

The extra CHG, or rather Aegean/Near Eastern ancestry, arrived in Italy during Roman times, because Romans are clearly shifted southeast compared to early Italics.

Drago said...

@ Leron
What did Mellart obfuscate ?

Gaska said...

I think they could have said multi Indo-European waves into mainland Italy,

To obtain temporal insights into the emergence of the differences between Northern and Southern Italy in relation to SBA and ABA ancestries, we performed the same qpAdm analysis on post-Neolithic/Bronze Age Italian individuals.

Iceman and Remedello, the oldest Italian samples included here (3.400-2.800 BC), were composed of high proportions of AN (74 and 85%, respectively).

The Bell Beaker samples of Northern Italy (2200 to 1930 BCE) were modeled as ABA and AN + SBA and WHG. Although ABA estimates in these samples were characterized by large standard
errors (SE), the detection of steppe ancestry, at approximately 14%, was more robust. In contrast, Bell Beaker samples from Sicily (2500 to 1900 BCE) were modeled almost exclusively as ABA, with less than 5% SBA (data file S4).

These results suggest a differentiation in ancient ancestry composition between different
areas of Italy, dating at least in part back to the Bronze Age.If as you say, the early Italics have "lots of steppe ancestry", it does not come from the Italian Bbs and much less from the Sicilian or Iberian BBs, it had to enter the Bronze age or even the Iron Age.

In any case, if the rumors about the Etruscan genomes are confirmed (R1b-M269 and a case of U152) are true, then the situation in Iberia would be repeated and it means that we were not wrong in denying the linkage of P312 with IE- The status of the Italic peoples will depend on which of them has been analyzed and the age of the samples.

No one doubts that Rome meant the significant increase in exotic ancestry in the Italian peninsula

Matt said...

@Drago, just by using the reprocessed West Eurasia 9 plots and without running models, looks TKM_IA roughly in the right neighbourhood for a two-way model with SIS3, though perhaps a tad steppe rich for the Swat_IA (may work better for Brahmin folks and Rors, who some would argue are preferential to use, rightly or wrongly).

See: https://imgur.com/a/J8DO5r0

Doesn't seem clearly preferential to the various Steppe_MLBA+BMAC outliers though or to Haji Firuz BA (though I'm sure it's temporarally much better than HFBA).

Other outliers might be more solid on simple two way models since slightly less steppe rich, but I'm not sure we should get too hung up small differences in sample position on two-way models with outliers. That said, nMonte / other algorithm modellers may find something more interesting.
Not gonna do anything further for now as all may be obsolete by this evening.

(Not that they'll necessarily clear it all up. Depending on what the sample sets from Rakhigarhi show and when sampled we may not be able to exclude for'ex a model with two layers, with Parpola's supposed migration from BMAC at 1900 BCE, and then slightly later transfers of Steppe MLBA ancestry at 1600 BCE - 1400 BCE, or something very much almost exactly at the same time. Who knows...)

Pastebin of datasheet for the above: https://pastebin.com/sTWfxitw

(Datasheet and plots includes two more Gonur outliers I missed including last time who have some steppe ancestry - I1789, I2122, who I've labelled as Gonur o5. There really are a lot more from Narasimhan and others who've got onto "West Eurasia 9" compared to G25, so in some ways more useful for a look at the ancient South-Central Asian scene, despite truncation of most ENA rich samples! And potentially for modelling.).

Anonymous said...

@FrankN
The Hurrians may have borrowed the name of the horse from the Mitanni, and the Akkadians from the Hurrians. But in the Luwian language the horse was aswa and this is the original name of the horse in the Luwian language.

Lat. torquere "to twist", German drechseln "to turn wood" is not from PIE **trg, but from PIE *trk. Middle Greek δραγομάνος -> Medieval Latin dragumannus etc. is borrowed from non-Indo-European language.

Davidski said...

@All

The following Italian populations have been added to the Global25 or updated. Let me know if there are any clear outliers or other issues.

Italian_Abruzzo
Italian_Aosta_Valley
Italian_Apulia
Italian_Basilica
Italian_Calabria
Italian_Campania
Italian_Lazio
Italian_Liguria
Italian_Lombardy
Italian_Marche
Italian_Molise
Italian_Northeast
Italian_Trentino-Alto-Adige
Italian_Tuscany
Italian_Umbria
Italian_Veneto

Same links as always...

Getting the most out of the Global25

Simon_W said...

Wow cool stuff! That's what I've been waiting for, thanks.

zardos said...

From the new Italian paper:
"Contrary to previous reports (4), the occurrence of CHG as detected by our CP/NNLS analysis did not mirror the presence of SBA, with several populations testing positive for the latter but not for the former (Fig. 2 and fig. S5, A and B). When we compared this analysis and the one using a different CHG sample (SATP) (5), the two were highly correlated (Spearman ρ = 0.972, P < 0.05; fig. S5F). We therefore speculate that our approach might, in general, underestimate the presence of CHG across the continent; however, we note that even considering this scenario, the excess of Caucasus-related ancestry detected in the south of the European continent, and in Southern Italy in particular, is notable and unexplained by currently proposed models for the peopling of the continent."

SATP is related to IE, while ABA was an independent influence which effect in Europe is low and mostly restricted to the SE.

Davidski said...

Yeah, it's just extra and more recent Near Eastern influence in Italy due to the Roman Empire and Muslim conquests.

The authors of that paper must be very confused if they don't know this.

zardos said...

Well, I guess that other prehistoric and earlier historical events like Phoenician and Greek colonisation played in too.
But whatever it was, its unrelated to the IE question.

zardos said...

Btw, didnt found a list for yDNA for ancients and moderns while reading on the phone, is there any?

Andrzejewski said...

@archi “The inhabiting of the Anatolians (Luwian) out of Anatolia before Mycenaeans time Herodotus described, and they had inhabit from Thrace to the Ionian sea. Teucrians as we know is the Trojans, and the Trojans spoke Luwian language, what remained of the inscription, and theirs Luwian names.”

It wouldn’t surprise me if the Philistines are Trojan refugees

Andrzejewski said...

@Leron “ Wouldn't CHG have more to do with Maykop and the proto-North Caucasian languages?”

I don’t think so. I used to think that early PIE had something to do with an EHG language, however I realized that it sounded or had nothing in common with EHG languages like Narva Culture’s, Volosovo or Pit Ware Culture. Furthermore, Yamnaya bore no resemblance to Yenisseyan or WSHG ones. Hence, it probably originated with a Piedmont or Progress group right north of the Caucasus.

The last thing that clinched the deal for me was the phonology of the (reconstructed) PIE language with lots of odd laryngeal consonants akin to the NW Caucasus languages like Adyghe or Ubykh

Anonymous said...

@andrzejewski
"had nothing in common with EHG languages like Narva Culture’s, Volosovo or Pit Ware Culture"

Narva Culture is not EHG! It is WHG. Narva Culture came from the west.
About Volosovo no data available.
Pit Ware Culture and Yamnaya is synonymous.

"Yamnaya bore no resemblance to Yenisseyan"

Yenisseyans is Q1! Yamnayans is R1b-Z2103! They are not related in any way.
The Yenissenyan language is the Sino-Caucasian. You have contradiction.

"Hence, it probably originated with a Piedmont or Progress group right north of the Caucasus."

Hardly.

"The last thing that clinched the deal for me was the phonology of the (reconstructed) PIE language with lots of odd laryngeal consonants akin to the NW Caucasus languages like Adyghe or Ubykh"

It's only version. Laryngeals are archaism in languages, they were everywhere.

The closest relatives of PIE are in Siberia (Uralian, Altayan), so it really is the language of EHG.

Andrzejewski said...

@archi “Pit Ware Culture and Yamnaya is synonymous.“

I meant “Combed Ware”. A non-IE non-Uralic substrate in Eastern Europe. Some researchers believe it has something to do with either SHG or Saami, or both.

https://en.m.wikipedia.org/wiki/Pit–Comb_Ware_culture

Now, this is what PIE allegedly sounded like:

https://m.youtube.com/watch?v=S7kxmcz971E

Sounds similar to Adyghe or even Kartvelian.

PS: what proof do you have that PIE has any resemblance to Uralic or Altaic, except for a few cognates, usually a Sintashta —> PU borrowings?

Simon_W said...

@Davidski

"Italics are basically a mix of Europe_LNBA (with lots of steppe) and Europe_MN (with no steppe). They don't have any significant extra CHG.

The extra CHG, or rather Aegean/Near Eastern ancestry, arrived in Italy during Roman times, because Romans are clearly shifted southeast compared to early Italics."

The fact that the Sicilian Beakers are modeled as almost exclusively ABA in that paper shows that it's not all from Roman times and that there was structure involving ABA admixture even prior to the Roman era. The only leaked info in conflict with this, that I read about, is that the Samnites were North Italian-like too. But the Samnites were a particularly warlike Italic confederation centered around the mountainous interior of Southern Italy where ABA ancestry is still lower even in the present era. The coastal areas of Apulia, Calabria and Campania were probably different and more ABA-rich even before the Romans; that's my prediction.

Simon_W said...

@Davidski

Was there an issue with the samples from the Piedmont and the Emilia-Romagna that you didn't include them?

Simon_W said...

And just a detail, but it's called Basilicata, not Basilica.;-)

Simon_W said...

It's really a shame Raveane et al. didn't consider my suggestion to include samples from the ancient Levant in their analyses. Well, probably they didn't read the comments here.

Anonymous said...

"I meant “Combed Ware”. A non-IE non-Uralic substrate in Eastern Europe. Some researchers believe it has something to do with either SHG or Saami, or both."

There were many migrations to North Eastern Europe for example from Siberia, the last such not directly related to the Finno-Ugric visible on the Bolshoy Oleniy Island in the Barents sea.


"Sounds similar to Adyghe or even Kartvelian"
This hypothesis does not have not a single proof, the absolute fantasy of the author. The only language with a "laryngeal" is Hittite, where it sounded like uvular.


"PS: what proof do you have that PIE has any resemblance to Uralic"
Linguistics says so, and practically all linguists.

zardos said...

The ABA BB is quite interesting and probably represents the often postulated link of, especially Southern, BB to Eastern Mediterranean cultures and traders. Would be nice to check how far the ABA influence reaches in BB and generally in early BA. Not just in Italy.

Andrzejewski said...

@archi “Yenisseyans is Q1! Yamnayans is R1b-Z2103! They are not related in any way.
The Yenissenyan language is the Sino-Caucasian. You have contradiction.”

Botai were R1b and Q1a (latter was also in Khvalynsk), both of EHG/WSHG derivation, ultimately of AG3/ANE source. But their languages couldn’t be more different. Yenisseyan is also Q and of ANE and it’s very different from PIE. So your theory falls flat here.

zardos said...

@Andre: You know what language Botai people spoke? I think your approach might be too simplified.
We all dont know for sure, but isnt a language spoken by EHG with other, significant and formative, influences the best bet?

Andrzejewski said...

@zardos @archi we know that Botai didn’t speak anything close to PIE...

Anonymous said...

@andrzejewski How do you know what language Botai spoke? You write some nonsense as if there were only PIE and Yeniseian, while there were hundreds of different paleo-Siberian languages unrelated to each other. Yeniseyan in Siberia has spread in the 1st Millennium CE. The modern state of languages is very late, it is associated with the expansion of several languages, and before that in Eurasia there was a wealth of languages more than in America. So all your statements are simply wrong.

Anonymous said...

^^-paleo-Siberian- read paleo-Asian (all extincted)

Bronze said...

@samuel andrews
I understand blood is still dripping from your pussy after getting fucked by a superior indian man.

But youre factually wrong. The history of europe and south asia is parallell. Indo-europeans are not closer to europeans than they are to south asians at all.

Yamnaya was not even remotely european, not genetically and not phenotypically. Andronovo are not ancestors to south asians. Andronovo is just a mixed offspring of central asians moving into eastern europe and mixing with the native european women there, while bringing R1a and ANE into europe.

So no, Europe is not relevant at all to the origin of indo-europeans. You pathetic dickriding whore.

Anonymous said...

@Bronze The Ravings Of A Madman. Full Nonsense.

Simon_W said...

Not even worth commenting Archie, Bronze's comment is bound to be deleted within hours.

zardos said...

Btw, related to Italians, whats currently the oldest aDNA sample for U152?

Simon_W said...

Ah crap, Davidski is busy with the Rakhigarhi thread.
You're just lucky Bronze.

@zardos

Good question! Jean Manco's site had a very useful regularly updated list of ancient yDNA, but sadly it's down since she's dead.

zardos said...

Oh, I visited the site and saw it was down. Didnt know it was because of the owners death. R.I.P.

Simon_W said...

@zardos

Indeed, RIP.
Well, I checked my old notes on Allentoft et al. 2015 and there's a South German Bell Beaker with R1b-U152, RISE563. There may be more Bell Beakers with that marker, but probably none from older cultures, as it was the Bell Beakers who spread R1b across central and western Europe.

Simon_W said...

As for the new Italian samples in the Global25, it really strikes me how much less ABA they get when using the Global25/nMonte method, compared to their ABA scores in Raveane et al. 2019!

For instance, Figure 2D in that paper shows the different proximate ancestries according to the CP/NNLS method, and the South Italian clusters 1 and 3 are almost entirely ABA. And most North Italian clusters also get some ABA, though on a much lower level. Furthermore Figure S5G, showing the ancestry proportions according to the qpAdm method, suggests that most Italian samples have around 50% or more ABA ancestry, even Sardinians. That's really weird if you compare it to the Global25/nMonte results that I got:

Eastern Sicily gets 35.3% ABA, Calabria 38% and Apulia 35.5%. Campania also scores comparatively high with 36.6% ABA, using this method. And the available North Italian samples all completely lack ABA. There is 0.5% Natufian in Lombardy and there is 1.1% Natufian, 1.1% Iberomaurusian and 0.4% Sub-Saharan in Liguria, but no ABA! Likewise Sardinia has no ABA, just some 1.1% Iberomaurusian.

This discrepancy is weird and hard to understand. But in my opinion the lower ABA makes more sense, because, what's the chance that there was a near complete replacement of the ANF by ABA in Southern Italy? How should that work out?

Another thing that struck me is the really low WHG (I mean other than what's in the Steppe ancestry) that most Italian samples get. Even Lombardy gets a mere 5.2% WHG. Marche has 2.6%, Umbria 2.9% and Southern Italian samples get less than 2%. On the other hand Sardinians get a comparatively whopping 12.7% WHG, hence they are the Italians with the strongest WHG. I think this explains why Sardinians often look rather European-like, whereas even North Italians may often have small faces and weak chins almost like Arabs.

zardos said...

I don't think that chins/faces being mainly the result of such relatively minor admixture differences, but rather bigger ones and selection, also from within the same ancestral people.
However, the question is from where ABA ancestry came to Southern Italy in the first place. If it came f.e. via Greek islands or SEE, it would have picked up most likely local ancestry and this would mean a considerable replacement for places like Sicily and Calabria by Eastern Mediterrenaean/SEE people.

Survive the Jive said...

@Samuel Andrews

"Andronovo & its decendants did become Central Asian horse warriors. But, the spread of Indo European languages in Europe had nothing to do with Central Asia or the horse or the wheel."

Why are you so sure IE people in Europe did not depend on horses? We know horses went extinct in Britain and were reintroduced by the Beaker folk (who wore garments made from horse hair). The presence of the wrist bracers in british beaker burials also suggests possible mounted archery

weure said...

Based on a Dutch study of the Y-DNA landscape of the Netherlands.

Tentative conclusion about R1b U106 and the influx of the Anglo-Saxons in migration time (early middle ages).

The Saxons/ Chauci (mainly L47 and U198) are mostly represented in the eastern parts of the North Dutch area (Groningen and Drenthe).

The Jutes (Z9 and Z18) are mostly represented in the Western parts, especially in Westergo/ Friesland, the Jutish and Frisian power center was Wijnaldum (see Finn in Beowulf).

https://www.nature.com/articles/s41431-019-0496-0

zardos said...

What I don’t understand is, why they used no subclades for I in the Dutch study. Might have been interesting too.

Simon_W said...

@zardos

But look, Iberians in general tend to have positive chins and not overly thick lips, even though their steppe ancestry isn't much stronger than in Lombardy for instance:

[1] "distance%=3.7411"

Spanish_Castilla_Y_Leon

Barcin_N,53.3
Yamnaya_Samara,29.9
WHG,11.7
Morocco_Iberomaurusian,5.1

I do have the impression that Italy has more of the araboid looking types, but hey it's terribly subjective and I may well be wrong.

Lombardy:
Barcin_N,60.7
Yamnaya_Samara,33.6
WHG,5.2
Natufian,0.5

More important is another observation: North Italian Beakers had 12.6% WHG and 15.8% Yamnaya. Remedello, even the late one after 2134 BC, RISE486, had even more WHG, 17.4% with 0% Yamnaya. So one thing seems clear: Modern Italians are not primarily descended from ancient North Italians. North Italians around 2000 BC had plenty of WHG, while modern Italians, even northern ones, have much less WHG. It means that there was an influx of a Barcin-rich, WHG-poor population, from southeastern Europe most likely. But at a time when ABA didn't permeate southeastern Europe yet.

Simon_W said...

@zardos

"However, the question is from where ABA ancestry came to Southern Italy in the first place. If it came f.e. via Greek islands or SEE, it would have picked up most likely local ancestry and this would mean a considerable replacement for places like Sicily and Calabria by Eastern Mediterrenaean/SEE people."

I can't quite follow that thought. If ABA from SEE picked up local ancestry in Southern Italy, it would be diluted. Pure ABA man mixing with local woman would mean 50% ABA offspring. But according to Raveane et al. South Italian cluster 3 is more than 75% ABA. Of course it's possible, but prima facie I find the 35.5% ABA I got with Global25/nMonte more realistic.

Samuel Andrews said...

@Simon_W,

The ancient DNA teasers say Near Eastern admixture became widespread in Italy in the Roman era. Etruscans, Samnites, Umbri cluster together relatively near modern Northern Italy. Romans have a lot of Near Eastern admixture and cluster with modern Southern Italians.

Looks like a big demographic change happened in central Italy when the Roman empire emerged.

Simon_W said...

@Sam

I'm eagerly waiting for those papers. Still, ABA admixture is much stronger in modern central Italy than the ultimately Natufian ancestry. Using my usual array of populations, I get 22.3% ABA in Umbria and 1.6% Natufian. In the Marche it's 20% ABA and 4.2% Natufian. In Lazio 25% ABA and 3.9% Natufian. In Tuscany 10.9% ABA and 2.2% Natufian. Of course neither Natufians nor ABA existed in the Roman era, so a few percent Natufian translates into a much higher figure of Roman age Levantine admixture.

zardos said...

I doubt the Phoenician-Greek South was as Northern-Central European before the Imperial times, but lets see with what they come up.

The more bird-like facial profile is indeed more common in the Eastern Mediterranean than the West. Its represented in Minoan and Mycenaean art and even more extreme in Hittites as you know.

75 would be huge indeed.

Samuel Andrews said...

Looks like there could be Roman Italian ancestry in East coast of Spain.

0.577"

Spanish_Peri-Barcelona

Basque_Spanish,42.8
Italian_Veneto,39.9
France-like_Italy Medieval,13.6
Moroccan,3.7

ǵenh said...

I doubt that all of Natufian in modern Italians is due to Roman age Levantine admixture. Traces of Natufian using nMonte can also be found in Mainland Greeks, Albanians, Macedonians, Montenegrins, Bulgarians, Romanians, Moldovans, Bosnians, Croatians, and further north. Among the ancient samples traces of Natufians are also found in some Mycenaean samples, in some LBK samples from Germany and Austria, in the Greek Neolithic samples, and so on.

zardos said...

I'm sure Iberians got their fair share of Roman ancestry, more than Gothic for sure.
There were a lot of Roman veterans and citizens in the provinces since the times of Pompeius.

Simon_W said...

@genh

Careful. Small traces of exotic admixture in some few individuals are sometimes just noise due to poor data quality. Also, with Global25/nMonte it always matters what other populations were included in your models.

My usual list of ancient samples to cover the basic variation in West Eurasia is:

Yamnaya_Samara
Barcin_N
WHG
Anatolia_EBA_Isparta
Anatolia_EBA_Ovaoren
Ganj_Dareh_N
Natufian
Morocco_Iberomaurusian
Han
Yoruba

Applying this, on population averages only, not on individuals, I don't find any Natufian admixture in Minoans, nor in Mycenaeans, nor in Greeks from Empuries. They all have Zero Natufian. Moreover, the Bell Beaker from Sicily also completely lacks it. I also checked Klei10 from late Neolithic Northern Greece. He's completely without Natufian.

Simon_W said...

@genh

This is what I get for German LBK:

[1] "distance%=1.9816"

DEU_LBK_N

Barcin_N,95.8
WHG,4.2
Yamnaya_Samara,0
Anatolia_EBA_Isparta,0
Anatolia_EBA_Ovaoren,0
Ganj_Dareh_N,0
Natufian,0
Morocco_Iberomaurusian,0
Han,0
Yoruba,0

And this for the Peloponnese Neolithic:

[1] "distance%=1.298"

GRC_Peloponnese_N

Barcin_N,82.8
Anatolia_EBA_Isparta,15.6
Ganj_Dareh_N,1.6
Yamnaya_Samara,0
WHG,0
Anatolia_EBA_Ovaoren,0
Natufian,0
Morocco_Iberomaurusian,0
Han,0
Yoruba,0

Such are the facts. BTW, Vinca, Lengyel, Baden, Vucedol, Maros, Hungary_BA, Croatia_MBA and Croatia_early_IA also lack Natufian admixture. So sooner or later you'll have to accept that the Natufian admixture we can see in modern Italians arrived there very late. Leaks from one of the upcoming papers suggest that it started to show up in Italy during the Iron Age.

Leron said...

I'm waiting for someone to drop ancient aliens theory from the way some of these comments are going...

Ellejay said...

Gedmatch thinks my grandmother is only 7.6 generations to MRCA for RISE98 Sweden from 3.7 thousand years ago...gotta love that random genetic recomb.