Thoughts from Klyosov R1b data and the IE problem

secherbernard

Elite member
Messages
57
Reaction score
5
Points
0
Location
Antony, France
Y-DNA haplogroup
R-L21
mtDNA haplogroup
U6a7a1
Anatole Klyosov has published a very interesting web page about R1b history:http://www.turkicworld.org/turkic/60_Genetics/Klyosov2010DNK-GenealogyEn.htm
This is a short summary of Anatole thesis:
1) R1a and R1b are both from Alta?
2) 12.000 years ago R1a migrated and arrived in Balkans
3) Between 6.000 and 8.000 years ago R1b migrated westward from Alta? and arrived in the north of Black Sea. This R1b migration produced the cultures of Samara, Khvalynsk and Yamnaya
4) From here, 2 R1b migrations: the first one to Balkans and Italy and brings R1b-ht35 in Europe, the second one to Caucasus, Near East, North Africa and Iberia. Then R1b-ht15 spreads in occidental Europe with Bell Beaker culture.
5) R1a migrated to North Europe and produced the Corded Ware culture, then he went eastward to the Steppe between 5.000 or 4.000 years ago and produced the culture of Andronovo. Then R1a spread to Iran and India.

Anatole Klyosov thinks that R1a spoke proto IE language and R1b spoke proto turkic language. He doesn't agree with Kurgan hypothesis. He sees that R1b migrated from east to west and that R1a migrated from west to east.

I think that Klyosov data and chronology can be understood under the Kurgan hypothesis.

If R1b arrived first in the Steppes during Samara, Khvalynsk and Yamnaya cultures, I think R1b spoke proto IE language. The first migration of R1b to Balkans and Italy can be related with Yamnaya migrations.
When R1b arrived in East Europe, he comes in contact with R1a people (remember that Klyosov thinks R1a are in Balkans before 3.000 BC). From this contact, R1a changed his language and became IE with satem language. R1b kept his centum language.
Then R1a spread to the north of Europe:it is the Corded Ware culture.
From the south-east of Corded Ware, we can see an eastward R1a migration from Corded Ware culture to Sintashta culture, via Middle Dnieper, Fatyanovo, Abashevo cultures.
Then from Sintashta culture R1a migrated to south-east in Andronovo culture, and then spread to Iran and India.

I think this R1a and R1b scenario could be in agreement with Klyosov data.
 
Anatole Klyosov thinks that R1a spoke proto IE language and R1b spoke proto turkic language. He doesn't agree with Kurgan hypothesis. He sees that R1b migrated from east to west and that R1a migrated from west to east.

But their is no evidence of Turkic substrate in Centum languages. How would R1b people lose so fast all their Turkic vocabulary ?
 
But their is no evidence of Turkic substrate in Centum languages. How would R1b people lose so fast all their Turkic vocabulary ?

It's utterly impossible from the linguistic side. In addition to what you said, Proto-Turkic is *much* younger than the Indo-European languages, and the Turkic languages probably did not start to split up until 1st century BC or so, and the earliest that Turkic languages arrived in Europe might be with the Hunnic invasions.

The oldest attested Centum language, in contrast, would be Mycenean Greek from the 15th century BC.

It should be added that there is also zero evidence in Basque for Turkic loans.
 
Last edited:
Anatole Klyosov has published a very interesting web page about R1b history:http://www.turkicworld.org/turkic/60_Genetics/Klyosov2010DNK-GenealogyEn.htm
This is a short summary of Anatole thesis:
1) R1a and R1b are both from Alta�
2) 12.000 years ago R1a migrated and arrived in Balkans
3) Between 6.000 and 8.000 years ago R1b migrated westward from Alta� and arrived in the north of Black Sea. This R1b migration produced the cultures of Samara, Khvalynsk and Yamnaya
4) From here, 2 R1b migrations: the first one to Balkans and Italy and brings R1b-ht35 in Europe, the second one to Caucasus, Near East, North Africa and Iberia. Then R1b-ht15 spreads in occidental Europe with Bell Beaker culture.
5) R1a migrated to North Europe and produced the Corded Ware culture, then he went eastward to the Steppe between 5.000 or 4.000 years ago and produced the culture of Andronovo. Then R1a spread to Iran and India.

Anatole Klyosov thinks that R1a spoke proto IE language and R1b spoke proto turkic language. He doesn't agree with Kurgan hypothesis. He sees that R1b migrated from east to west and that R1a migrated from west to east.

Dienekes used a more accurate dating formula to come to the following conclusion:

Southeastern Europe (the Balkans)
Thus, Balkan haplogroup I seems related to a Bronze Age origin, with R-M17 being substantially older, and deriving perhaps from northern Balkan Neolithic or alternatively intrusive Kurgan populations.
http://dienekes.blogspot.com/2008/07/how-y-str-variance-accumulates-comment.html
 
This theory doesn't make much sense to me. It completely disregards many essential points, such as :

1) the presence of the oldest clades of R1b in the Middle East and South Asia.

2) the early migration of R1b-V88 from the Middle East (where it is still found today) to Africa.

3) the distribution pattern of Indo-European languages, which obviously branched apart from a common source near the Caucasus/Black Sea.

4) archaeological evidence that domesticated animals like cows and sheep spread from the Caucasus region to the Pontic steppes, then eastward to Central Asia, and not the other way round (hence the migration of herders in the same direction).

5) archaeological evidence that the kurgan/tumulus culture and artefacts associated (including bronze weapons) spread from the North Caucasus/Pontic Steppe westward towards the Balkans and eastward towards the Urals and Central Asia, not the other way round.

Furthermore, the proposed migration scenario is very hard to conceive. How could R1b be the last to arrive from the Altai, settling in the Pontic steppe after R1a, but completely bridge the whole Balkans to move to Western Europe and almost completely disappear from Ukraine, Russia and the Altai region ? It makes much more sense that R1b came from the Middle East from Central-South Asia during the Paleolithic, then moved around the Black Sea with all the Neolithic and Bronze Age technology from the Caucasus region, and conquered the steppes, then once the horse domesticated (itself probably requiring the prior domestication of other tamer animals, like cattle, goats and sheep), R1b would have had an open window on the Danube valley and the metal rich Chalcolithic towns of the Balkans ("Old Europe"). Massive and destructive migrations into heavily populated regions don't happen just like that. They require a motive (e.g. plunder rich towns, seize copper resources) and capability to conquer (more advanced weapons, in this case bronze axes and swords and horses).

Bronze-age R1b1b and R1a1a people from the Pontic Steppe were essentially a semi-nomadic (no towns or cities) patriarchal and ware-like pastoralist people who craved gold, copper and tin and needed green pasture for their herds. Their dream land was a place with plenty of green grass and abundant mineral resources. Although the Caucasus was rich in metal for a while, the resources dwindled with time, became harder to extract and increasingly more coveted by Mesopotamian and Anatolian neighbours. The Balkans were rich and militarily weak, ripe for plunder, but too populated to settle and too farmed for herding. This is why the bulk of R1b moved on along the Danube till the Black Forest, then expanded in every direction in the relatively low populated and green lands of Western Europe. They settled most heavily in places least suitable for farming, and therefore least inhabited, such as the Landes in Southwest France, the marshes of the Netherlands, or the rocky terrains of Brittany, Cornwall, Wales, Scotland and Ireland.

R1b never migrated in big number eastward to Central Asia. A small minority did blend with the R1a tribes from the Volga-Urals region, perhaps an elite that brought bronze-working to them and converted them to their Proto-Indo-European language. The original language of these R1a1a people could have influenced the pronunciation of these new Indo-European speakers, becoming the Satem branch of Indo-European languages.

After over 5000 years of constant migrations from the Volga-Urals and Central Asia towards the Pontic Steppe and the Balkans, R1b eventually became a minority haplogroup in the region. It is even likely that no Bronze-age R1b from the Pontic Steppe survives today in that region, and that the R1b lineages found there today are back migrations from Central Asia, where R1b is also a small minority.

I have wondered for a long time if it wasn't simpler for R1b to migrate directly from Anatolia to the Balkans, then to Central and Western Europe. But there is no archaeological evidence supporting such a migration, and it wouldn't explain how about 10% of Russians are R1b, and how R1b got into Central Asia alongside R1a, as far as Xinjiang. It also bypasses the two most important factors for the conquest of Europe : bronze weapons and horses. All the archaeological evidence points at the Maykop culture as the very first to develop bronze swords and axes, and it is undeniable that horses were first domesticated in the Volga-Urals Steppe, in the continuum of the Pontic Steppe. Therefore it is unthinkable that R1b forced its away directly from Anatolia to the Balkans, without a passage through the North Caucasus and the Eurasian Steppes beforehand.
 
It makes much more sense that R1b came from the Middle East from Central-South Asia during the Paleolithic, then moved around the Black Sea with all the Neolithic and Bronze Age technology from the Caucasus region, and conquered the steppes

Do you support the idea that R1b1b2 was among the Yamna tribes? I saw this on your migrations map. But the Corded Ware culture which has its R1a from the Yamna lacks R1b1b2.
 
Do you support the idea that R1b1b2 was among the Yamna tribes? I saw this on your migrations map. But the Corded Ware culture which has its R1a from the Yamna lacks R1b1b2.

It's hard to say exactly when most of the R1b left the Pontic Steppe and R1a filled the void. It was a progressive process taking place over thousands of years. I would say that the Pontic Steppe still had a fair number of R1b at least until the Yamna period. R1a might have become the majority lineage during the Catacomb or Timber-grave culture. But the Corded Ware, which originated in the forest-steppe north of the Pontic Steppe, was almost certainly in vast majority R1a. The Corded Ware Culture correspond to an extension of the Abashevo Culture, not of the Yamna Culture. You can see on my migration maps that the yellow cultures correspond to a strongly dominant R1a element, the red ones to strongly dominant R1b, and the orange ones a blend of both, or uncertain.
 
Hello,

I just joined this site which is extremely interesting to me. I would like to contribute the result of my analysis of various recent publications that I have read, most importantly the well known Myres et al article as well as the book :

"The horse, the wheel, and language" by David Anthony, the extensive analysis of the PIE problem.

I put pieces of my findings on the following link: (unfortunately only in French for now)



.bertrandjost.tk/Francais/Monog-famille/Jost/Asie/nouvelle-naissance.html



But in essence my theory about the origin of R1b, S28 and R1a is the following:



60K ago: Haplogroup CT crosses gulf of Aden from Ethiopia and follows the Arabian coast eastward (Neanderthal occupies Middle east, including Irak and western Iran, blocking progress north)

50K: Modern Humans arrive in Western India: appearance of Haplogroup F

40-35K: Haplogroup K appears in central India (very few individuals, hence little remains)

25-18K: Haplogroup R, R1, R1b appear in Northern Pakistan, right below the limit of the steppes at that time (max ice age)

This appears clearly on Myres et Al study

Later R2 will migrate back south to Southeast India; other R and R1b will migrate west toward Iran and Jordan

12-14K: Most R1b migrate from Pakistan to the caves of Bashkortostan, south the Ural mountains; this where U152 will appear during this period. Myres et al also shows this.

10K: R1b will follow Volga and other rivers to Western Ukraine, followed by R1a settling Eastern Ukraine. Both R1b and R1a are part of the Yamna PIE culture that appears around 4500BC in this area. (See Anthony). Main advantages of the culture: herding, horse, mobility (wagons) and lactose tolerance.

In the west, R1b language will give birth to the centum group of PIE languages

In the east R1a will give birth to the satem languages.

Ca 4000BC, some r1b migrate to rumania, bulgaria and cross over to Anatolia (ancestors of the Hittite)

Ca3700BC tocharian will migrate from R1a group toward urals in the direction of Western China (proven by DNA tests on mummies)

Ca3300BC migration of R1b will begin up the Danube (Celtic/Italic) and Dniester (Germanic). these are the two primary waves: Germanic toward Denmark, celtic/Italic toward switzerland

Ca3100BC: R1a will migrate northwest up the Dnieper toward western russia and Poland and back east toward Ural, Pakistan and India.



The key next step will be to find ancient R1b in the balkans to verify this but I believe the fact that Ossesians (descendants of Scythians of Western Ukraine) have a lot of R1bs among them is good indication of this theory. Also no ancient R1b was found in Western Europe prior to 3000BC.

As for the book of David Anthony he clearly shows the various steps of evolution of PIE culture and the later migration with likely dates.
 
First off, welcome on the forum.

Honestly, I am afraid to say this, but these dates are absolutely impossible. This gets very clear from the adherence/non-adherence of sound laws. It is absolutely impossible that the Indo-European languages diverged before the latest Neolithic / Chalcolithic due to the fact that there are common words for agriculture, horses, cattle, and metal-working. Likewise, that the Centum and Satem splits are 10,000 years old is completely impossible. How could there be a common word for "horse" in Indo-European (ek´wos) if the Indo-European language family was older than the domestication of the horse?

Likewise, the association of R1a and R1b with the Indo-European languages is dubious, because only specific subclades (R1b-M269 and R1a1a) appears to be tied with the spread of the Indo-European languages, and even this produces a very incomplete and inaccurate picture. For instance is the distribution of R1a not limited to areas where Satem-IE languages are spoken. For instance, R1a reaches ~20% in Scandinavia, ~10% in Greece. Both the Germanic languages and Greek are Centum.

Also, never mind that with high likelihood R1b and R1a weren't in Europe that early.
 
Last edited:
R1b - R1a - and PIE problem

Hello Taranis,
You must have misunderstood my point. I did not say that the centum and satem are 10,000 years old. The Tocharians (R1a) were clearly part of the centum group because the split had not occured yet. The split occured around 3500 BC;

What i said is that Myres et Al clearly shows that R1b originates in northern Pakistan and that the following areas of settlement for the Majority (not all) of R1b and R1a were bashkortostan, and Ukraine (also shown by Myres et Al). The PIE language only developed in the Ponthic-Caspian steppes around 4500-4000BC. At that time, according to my assumption, both R1b-269 (in Western Ukraine) and R1a (in eastern Ukraine) spoke the PIE language that split a few centuries later, as western and eastern group started to drift apart.

Of course since R1b originated much earlier in Pakistan (for R1a it is harder to tell since there was a massive reverse migration in that region), some small sub-groups split on the way to create what is now the small groups of R1b in Central Asia, Middle East and Africa.
 
The Anatolian languages are generally thought to be first branch of Indo-European languages to have diverged. In particular predate the Centum/Satem split:

- the treatment of the so-called labiovelar sounds in PIE (, and g´h, which are essentially what the Centum-Satem split is all about) is unique.
- the Anatolian languages were the only branch to preserve the pharyngal sounds lost in all other branches Proto-Indo-European.

At that point, I have speculated in the past if the Anatolian hypothesis might be sort of correct if we assume that there was a migration of Proto-Indo-Europeans from Anatolia into the Pontic-Caspian region. However, I am nonetheless of the opinion that the Kurgan hypothesis is more or less correct because from the linguistic perspective it's prettymuch impossible that the IE languages would have spread into Europe during the Neolithic.

Regarding the Centum-Satem split, what is important to consider is that you must assume that this is not an early split, but it's more probable that it was a common innovation occuring in a 'core area' of Indo-European languages which were in proximity (Balto-Slavic, Armenian, the Paleo-Balkan languages, and Indo-Iranic) did participate, whereas the IE branches in the greater distance (Italo-Celtic, Germanic, Greek and Tocharian) did not.

In my opinion, the Centum-Satem split must have occured some time in the late 3rd to early 2nd millennium BC.
 
Last edited:
First off, welcome on the forum.

Honestly, I am afraid to say this, but these dates are absolutely impossible. This gets very clear from the adherence/non-adherence of sound laws. It is absolutely impossible that the Indo-European languages diverged before the latest Neolithic / Chalcolithic due to the fact that there are common words for agriculture, horses, cattle, and metal-working. Likewise, that the Centum and Satem splits are 10,000 years old is completely impossible. How could there be a common word for "horse" in Indo-European (ek´wos) if the Indo-European language family was older than the domestication of the horse?

.

Out of curiosity about the "horse" word (leaving aside the issue of this in English and German (unless you have ideas there too!)
(1) Have you read any info as to why the Hittites didn't have a proper IE "horse word"? The Tocharians did.
(2) The Balts also have a satemized version of the IE horse word. But why don't the Slavs? Do you have any info as to whence the "kon'" (or older "komon'") of the Slavs (="horse") comes from?
 
I've read somewhere that old Slavic kon was kobn, the horse female is kobyla in polish. Now we are closer to Italian caballo. Also horse run called trot is called kłus (eng phonetic: kwus, qwus) in polish, now we are close to IE eq'wos.
 
I've read somewhere that old Slavic kon was kobn, the horse female is kobyla in polish. Now we are closer to Italian caballo. Also horse run called trot is called kłus (eng phonetic: kwus, qwus) in polish, now we are close to IE eq'wos.

Thank you. After your post, I checked out Vasmer, and that is indeed his explanation. He considers "kobn" "kobyla" "komon" to derive from the same old root which produced the Latin "caballus". Then we hit obscurity. Vasmer mentions theories that this root goes back to some ancient Asia Minor or Danubian language, He does not deny this, but rejects all actually proferred proofs... At least we know where "pferd" (German) comes from. But the English "horse" is from a root as obscure as the Slavic "kon'/kin'". I find it interesting that the Baltic languages have retained the ancient satemized version of "equus", while Slavic languages have not. Especially since the Slavs borrowed the "God" word from the same peoples who used this satemized equivalent of "equus", and Balts kept the older term (as of course did the Latins)...
 
Out of curiosity about the "horse" word (leaving aside the issue of this in English and German (unless you have ideas there too!)
(1) Have you read any info as to why the Hittites didn't have a proper IE "horse word"? The Tocharians did.
(2) The Balts also have a satemized version of the IE horse word. But why don't the Slavs? Do you have any info as to whence the "kon'" (or older "komon'") of the Slavs (="horse") comes from?

Yes, cognates for ek´wos are not preserved in all branches of Indo-European, especially few of the modern languages preserve it.

Regarding the Germanic languages, the lost has been lost in all the modern languages, but it is attested in two older ones, namely Anglo-Saxon 'Eoh' and Gothic 'Aihws' (consider that in Germanic, PIE *k´ is first merged with *k like in the other Centum languages but later shifted to *h).

As for the Hittites, while it isn't attested there, the word is attested in another Anatolian languages, Luwian as 'A-su-wa'.
 
As for the Hittites, while it isn't attested there, the word is attested in another Anatolian languages, Luwian as 'A-su-wa'.

Where it seems a borrowing (from the Mitanni?) It's the satemized version, and closer to Indic or Thracian than to Iranic, which substitutes a "p" for the "v" or "w".
 
Last edited:
Where it seems a borrowing (from the Mitanni?) It's the satemized version, and closer to Indic or Thracian than to Iranic, which substitutes a "p" for the "v" or "w".

Actually, I don't think so. The shift from *k´ > *s (or *z) is found throughout Luwian. You might call Luwian a Satem language, but the critical part is that the Anatolian languages were the first part to diverge from PIE, and as a result the treatment of the Palatovelars (k´, g´ and g´h) occurs individually in the Anatolian languages.
 
Actually, I don't think so. The shift from *k´ > *s (or *z) is found throughout Luwian. You might call Luwian a Satem language, but the critical part is that the Anatolian languages were the first part to diverge from PIE, and as a result the treatment of the Palatovelars (k´, g´ and g´h) occurs individually in the Anatolian languages.

Thank you. I didn't realize the ubiquity of the satem shift in Luwian. I'm not a linguist and simply assumed that the deep comparative antiquity of the Anatolian languages (Luwian included) meant that the appearance of a "horse" word absent in Hittite implied a borrowing etc. But I am still a bit bemused by this. The satemization process is still, as far as I know, imperfectly dated and understood. I take it that Hittite and Palaic are NOT satemized?
 
Thank you. I didn't realize the ubiquity of the satem shift in Luwian. I'm not a linguist and simply assumed that the deep comparative antiquity of the Anatolian languages (Luwian included) meant that the appearance of a "horse" word absent in Hittite implied a borrowing etc. But I am still a bit bemused by this. The satemization process is still, as far as I know, imperfectly dated and understood. I take it that Hittite and Palaic are NOT satemized?

Don't worry, I can understand your reasoning here: if the Hittite language had (hypothetically) diverged before the adoption of the horse by the Indo-Europeans, then one would expect them not to have the common word for horse, and it would indeed make sense if the word was borrowed into Luwian.

I can't tell you about Palaic without looking it up, but yes, Hittite might be considered a Centum language. As I said, the Anatolian languages are generally thought to have diverged before the Centum/Satem split, and as such, the development of the palatovelars occurs in the individual Anatolian languages, not in Proto-Anatolian.

In contrast, in the Centum and Satem languages, the treatment of the Palatovelars (merger with plain velars in the former case, and shift to fricative sounds in the latter case) occurs very early proto-stages of the respective languages.
 

This thread has been viewed 69918 times.

Back
Top