November 25, 2014

Paternal lineages and languages in the Caucasus

An interesting new study on Y chromosome and languages in the Caucasus. The distribution of haplogroups is on the left. The authors make some associations of haplogroups with language families:

  • R1b: Indo-European
  • R1a: Scytho-Sarmatian
  • J2: Hurro-Urartian
  • G2: Kartvelian

Hum Biol. 2014 May;86(2):113-30.

Human paternal lineages, languages, and environment in the caucasus.

Tarkhnishvili D1, Gavashelishvili A1, Murtskhvaladze M1, Gabelaia M1, Tevzadze G2.

Abstract

Publications that describe the composition of the human Y-DNA haplogroup in diffferent ethnic or linguistic groups and geographic regions provide no explicit explanation of the distribution of human paternal lineages in relation to specific ecological conditions. Our research attempts to address this topic for the Caucasus, a geographic region that encompasses a relatively small area but harbors high linguistic, ethnic, and Y-DNA haplogroup diversity. We genotyped 224 men that identified themselves as ethnic Georgian for 23 Y-chromosome short tandem-repeat markers and assigned them to their geographic places of origin. The genotyped data were supplemented with published data on haplogroup composition and location of other ethnic groups of the Caucasus. We used multivariate statistical methods to see if linguistics, climate, and landscape accounted for geographical diffferences in frequencies of the Y-DNA haplogroups G2, R1a, R1b, J1, and J2. The analysis showed significant associations of (1) G2 with wellforested mountains, (2) J2 with warm areas or poorly forested mountains, and (3) J1 with poorly forested mountains. R1b showed no association with environment. Haplogroups J1 and R1a were significantly associated with Daghestanian and Kipchak speakers, respectively, but the other haplogroups showed no such simple associations with languages. Climate and landscape in the context of competition over productive areas among diffferent paternal lineages, arriving in the Caucasus in diffferent times, have played an important role in shaping the present-day spatial distribution of patrilineages in the Caucasus. This spatial pattern had formed before linguistic subdivisions were finally shaped, probably in the Neolithic to Bronze Age. Later historical turmoil had little influence on the patrilineage composition and spatial distribution. Based on our results, the scenario of postglacial expansions of humans and their languages to the Caucasus from the Middle East, western Eurasia, and the East European Plain is plausible.

Link (pdf)

46 comments:

andrew said...

The raw data is quite good. The historical linguistic analysis is dubious and strongly influenced by Russian school concepts that have little or no empirical basis.

R1a also looks like a fairly recent intruder to the region as it is found near the Eastern Plains in Turkic speaking populations.

The analysis of R1b which is one of the trickier distributions - found in Georgians (sometimes up to 50%), in Caspian Sea residents, and in Armenians, but dominant in no population, is particularly ham handed.

It largely ignores the fact that we know when Turkic languages arrived in the region (in the last two thousand years) and that it is most likely that Armenian languages probably arrived in the vicinity around the time of Bronze Age collapse (around 3200 years BP).

It also stubbornly insists that R1b was probably Indo-European, which is actually unlikely, and ignores the evidence pointing to its spread in Western Europe as a second wave Neolithic phenomena. This and the large phylogeny of R1b points to the Caucuses as likely geographically close to the source of R1b in Western Europe probably around the Copper Age. Since this was pre-Indo-European and pre-Turkic, and is geographically almost all S. Caucasian rather than N. Caucasian, it points to a potential linguistic R1b affinity with the S. Caucasian languages in preference to the N. Caucasian languages - although R1b could have been intrusive to the S. Caucasus from either Anatolia or the Persian highlands instead, with a pre-Copper Age European First Farmer G2 composition.

J2 suggests a link between NE Caucasian (Chechen and Inguish) and Anatolia.

Valikhan said...

Scytho-Sarmatian? Is it a new terminology for Altaic/Turkic languages? Or authors cannot speak loudly that Caucasian R1a is very common for Turkic speakers and arrived relatively late?

alpha-eagle said...

Ydna R1b were come from Central Asian invaders a.k.a Oghuz Turks and huns

Kurti said...

To be honest I wouldn't connect Haplogroup G2a exclusively to Kartvelians. A recent study on Alanic burials on the Don River has shown Haplogroup G2a to be the dominant Haplogroup. Since in all of the Cacausus only among the Northwestern Caucasian groups such as Adyghe/Abkhazians and Georgians (Basically groups which have been historically in contact with Alans ) Haplogroup G2a is dominant. Could it even be that Haplogroup G2a was spred to this particular region by Alans?
Even if not, thats still a prove that Haplogroup G2a in the Caucasus is not all exlcusive to Kartvelians.

http://en.wikipedia.org/wiki/Alans#Genetics

eurologist said...

As I have stated numerous times, for the past several tens of millennia, the Caucasus has been a sink - not a source - of populations. People arrived and settled a particular mountain valley, and then defended it with all their might.

This is also why you find very recent, modestly ancient, and very old haplogroups, there. But, just because some very old haplogroups are found in the Caucasus, this does not mean they originated from there to eventually occupy much of Europe. The latter were actually the majority of people, who simply moved on.

I am not surprised that there is a significant correlation between y-DNA and habitat, and that the correlation with languages is not as clear-cut. We know that in many instances, it is the women (mothers) who either will give their language to their children, completely, or at least the part used most in a daily home routine (see, e.g., everyday Germanic English versus French expressions in court and in formal gatherings).

terryt said...

"significant associations of (1) G2 with wellforested mountains, (2) J2 with warm areas or poorly forested mountains, and (3) J1 with poorly forested mountains. R1b showed no association with environment".

That is most likely not actually a product of ecological preference. Far more likely to be the product of sequence of arrival with G2 being the oldest and being forced into the most heavily forested regions. J1 and J2 have not been pushed into such isolated regions and R1b being the latest arrival and spreading everywhere.

Lappa said...

That picture is not consistent with DNA data:
http://3.bp.blogspot.com/.../wu2JBDE8uM0/s1600/diagram.jpg

R1a is shown as haplo of Ossetians, when R1a has highest % in Turkic nations like Karachays and Balkars.
In the other way Ossetians are mostly in Haplo G:
http://www.science.org.ge/2-1/Nasidze.pdf
that's why I'm very astonished that such charts were placed in that article
P.S.
Scytho-Sarmatians are not Indo-Europeans?

Kurti said...

@Valikhan & Andrew

it doesn't matter if R1a was brought to Caucasus by Turkic speakers or not, because at the end of the day the R1a (which is exclusively z93 and speaks for Bottleneck effect) in Turkic speakers stems from integrated Indo_Iranians in Central Asia.

Lathdrinor said...

@eurologist While I've come to the same conclusion about the neglect of mtDNA when it comes to languages, it is in fact even harder to trace languages by individual mtDNA haplogroups. Ultimately, the whole project of trying to draw equations between haplogroups and languages is flawed. I am prepared to believe that the proto-Indo-European population was high on XYZ haplogroup. But I am not prepared to believe that XYZ haplogroup, is therefore, connected to Indo-Europeans wherever it is found. It's pretty obvious that while the timing of certain linguistic expansions match with certain genetic expansions, other such expansions do not.

Jim said...

Valikhan,
"Scytho-Sarmatian? Is it a new terminology for Altaic/Turkic languages? Or authors cannot speak loudly that Caucasian R1a is very common for Turkic speakers and arrived relatively late?"

No, the term refers to Iranian-speaking people who language-shifted to Turkic languages in the middle ages. The same happened in the now Turkic-speaking cities in Central Asia.

Andrew,
"J2 suggests a link between NE Caucasian (Chechen and Ingush) and Anatolia."

That chimes with what some people say was the linguistic situation before the area shifted to Armenian. http://en.wikipedia.org/wiki/Hurro-Urartian_languages
http://en.wikipedia.org/wiki/Northeast_Caucasian_languages#Morphology

I know Diakonoff et al are not accepted universally, but they seem to have some evidence for their claim.

Nathan Paul said...

Heading is languages so talking about script is not out of place.

Look at Kartvelian script, old avestan script and Indian scripts. Look at their phonetics also. Giving them Aramac origin is the biggest religious and cultural based conspiracy. All the Indic and Persian writing systems were there before their roots.

David Tarkhnishvili said...

andrew,
"R1a also looks like a fairly recent intruder to the region as it is found near the Eastern Plains in Turkic speaking populations."
- this is true but only if we reject the option that current Turkic speaking populations of the region, in fact, to great extent descend from the earlier settlers of the Eastern European steppe, very much like Slavic speakers - Russians, Ukrainians, and Poles.
"ound in Georgians (sometimes up to 50%), in Caspian Sea residents, and in Armenians"- in fact, in central southern Caucasus (besides Daghestan) - both in Armenians and Georgians from this geographic area.
"It largely ignores the fact that we know when Turkic languages arrived in the region (in the last two thousand years)"
- it does not ignore this fact. Turkic languages, indeed, arrived relatively recently, which does not mean that the genetic ancestors of current Turkic speakers did not live in the region before that.
"and that it is most likely that Armenian languages probably arrived in the vicinity around the time of Bronze Age collapse (around 3200 years BP)." - Please read the paper carefully. We did consider this. Sorry if we could not make this clear enough.
"It also stubbornly insists that R1b was probably Indo-European, which is actually unlikely" - I still do not see sufficient evidence of this. Will be thankful for the newest references.

David Tarkhnishvili said...

Vakikhan,
"Scytho-Sarmatian? Is it a new terminology for Altaic/Turkic languages?"
- No. This is not a synonym of Altaic languages. This is a synonym of Eatern Iranian languages, spoken today by a wide range of ethnic groups, including Ossetians and Pushtu.

David Tarkhnishvili said...

alpha-eagle,
"Ydna R1b were come from Central Asian invaders a.k.a Oghuz Turks and huns" - I don't see sufficient evidence for this. Presence of a high proportion of R1b in present-day Turkmens is not a sufficient argument.

David Tarkhnishvili said...

Kurti,
by no means the authors connect G2 "exclusively to Kartvelians". We just link it, after the exclusion method, to early speakers of proto-Kartvelian languages. Of course, if present-day ethnic groups are considered, all populations of the western Greater Caucasus largely descend from this lineage, including Ossetians, Kartvelians, and Adyghe-speakers, which all we consider, to a great extent, descendants of the glacial-time population of the region. Concerning record of G2 in Don burials you wrote about, I will be thankful for the exact reference. Is it certainly Alan, or could also be perhaps Circassian? Anyway, I don't think there is sufficient argument to rule out traditional view linking Iranian languages with R1a.

David Tarkhnishvili said...

Eurologist,
Thank you for this comment. "But, just because some very old haplogroups are found in the Caucasus, this does not mean they originated from there to eventually occupy much of Europe."
- of course. Concerning our interpretations, we stay far from believing that a conclusive evidences were provided - just discussion around the arguments. The focal point of the paper is, of course, linking haplogroups with glacial refugia and with a modern ecological environment.

David Tarkhnishvili said...

terryt
-"That is most likely not actually a product of ecological preference. Far more likely to be the product of sequence of arrival with G2 being the oldest and being forced into the most heavily forested regions' - of course! I thought this should be clear from the discussion. We did discuss the sequence of arrival as a reason for the observed spatial distribution.

David Tarkhnishvili said...

Lappa,
did not understand the comment. I know that Ossetians are largely G2, the same as Georgians and Adyghe-speakers. This is stated in the paper. Don't see the contradiction.
Scytho-Sarmatians are of course Indoeuropeans, however their language belonged to "eastern branch" of Indoeuropeans, along with other Iranian, Indian, Slavic, and Baltic languages. To be brief, we used "Indoeuropean" in a diagram to mark Armenian language that does not belong to this subgroup.

David Tarkhnishvili said...

PS "R1a is shown as haplo of Ossetians"

Please look at this diagram carefully. R1a is NOT shown as haplo of Ossetians

Alshynkazak said...

No. If you belive that Scythians were fully Iranian, then you are brain dead. Scythians were a confederation with a Turkic dominant elite( royal scythians) and included some Iranian, and finno ugric tribes

David Tarkhnishvili said...

Lathdrinor -"Ultimately, the whole project of trying to draw equations between haplogroups and languages is flawed" - totally true if the modern populations are considered. However, there are some reasons in geological past that caused divergence of the haplogroups (btw, we speak about Y-DNA haplogroups and not mitochondrial haplogeoups) AND there are some reasons that caused divergence of the language groups. Most likely these reasons are past isolation among human-populated areas in Eurasia, which obviously did exist during and before the LGM.

Jim - thank you for the comment you did on Scytho-Sarmatian. Concerning Diakonoff, we did not mention him, but Starostin. I think Starostin made a sound contribution into evolutionary linguistics. Cavalli-Sforza makes quite a bit of references on him.

Alshynkazak - please avoid the expressions such as "brain dead" if you wish to discuss the problem.

terryt said...

"I am prepared to believe that the proto-Indo-European population was high on XYZ haplogroup. But I am not prepared to believe that XYZ haplogroup, is therefore, connected to Indo-Europeans wherever it is found. It's pretty obvious that while the timing of certain linguistic expansions match with certain genetic expansions, other such expansions do not".

I'm sure that is correct.

Kurti said...

@David Tarkhnishvili

I am definitely not ruling out R1a* for typical Iranian marker. That would be absurd. I am just against the idea. That only R1a can be linked to a whole language family, while we see quite diversity in all Iranian speakers.

About the G2 in Alanic burials. Here is the link. It is very long andf It is in Russian language.

https://www.academia.edu/7061155/%D0%90%D1%84%D0%B0%D0%BD%D0%B0%D1%81%D1%8C%D0%B5%D0%B2_%D0%93.%D0%95._%D0%94%D0%BE%D0%B1%D1%80%D0%BE%D0%B2%D0%BE%D0%BB%D1%8C%D1%81%D0%BA%D0%B0%D1%8F_%D0%9C.%D0%92._%D0%9A%D0%BE%D1%80%D0%BE%D0%B1%D0%BE%D0%B2_%D0%94.%D0%A1._%D0%A0%D0%B5%D1%88%D0%B5%D1%82%D0%BE%D0%B2%D0%B0_%D0%98.%D0%9A._%D0%9E_%D0%BA%D1%83%D0%BB%D1%8C%D1%82%D1%83%D1%80%D0%BD%D0%BE%D0%B9_%D0%B0%D0%BD%D1%82%D1%80%D0%BE%D0%BF%D0%BE%D0%BB%D0%BE%D0%B3%D0%B8%D1%87%D0%B5%D1%81%D0%BA%D0%BE%D0%B9_%D0%B8_%D0%B3%D0%B5%D0%BD%D0%B5%D1%82%D0%B8%D1%87%D0%B5%D1%81%D0%BA%D0%BE%D0%B9_%D1%81%D0%BF%D0%B5%D1%86%D0%B8%D1%84%D0%B8%D0%BA%D0%B5_%D0%B4%D0%BE%D0%BD%D1%81%D0%BA%D0%B8%D1%85_%D0%B0%D0%BB%D0%B0%D0%BD_%D0%95.%D0%98._%D0%9A%D1%80%D1%83%D0%BF%D0%BD%D0%BE%D0%B2_%D0%B8_%D1%80%D0%B0%D0%B7%D0%B2%D0%B8%D1%82%D0%B8%D0%B5_%D0%B0%D1%80%D1%85%D0%B5%D0%BE%D0%BB%D0%BE%D0%B3%D0%B8%D0%B8_%D0%A1%D0%B5%D0%B2%D0%B5%D1%80%D0%BD%D0%BE%D0%B3%D0%BE_%D0%9A%D0%B0%D0%B2%D0%BA%D0%B0%D0%B7%D0%B0._%D0%9C._2014._%D0%A1._312-315

Kurti said...

@ Alshynkazak

The Scythians existed during a time when there was nothing heard or known about people called Turkic.

The first Turkic speakers appeared far later, when the last Scythians and Sarmatians disappeared. If there is someone brain dead than it is probably you. Absolutely no one would come to the idea that the Royal Scythians of East Ukraine/Northwest Caucasus were Turkic speakers.

capra internetensis said...

@Alshynkazak

What is the basis for thinking Royal Scythians were Turkic? There's no reason they couldn't have been, of course, but the great Turkic expansion that we know of was considerably later. An interesting study I saw recently argued that the Scythians familiar to the Ancient Greeks were actually Circassian speakers. But I agree it's very unlikely that they were all Iranian speakers.

I wonder if branches of almost every nearby ethnolinguistic group didn't end up on the steppe at some point - as with the Plains Indians.

Unknown said...

The idea of mapping against the environment/food production technology undoubtably advances the discussion. The premise of going back to the refuge is a logical place to start, though of course it may end up not telling the whole story.

A couple of points that may be helpful.

Kurgan advocates like Anthony and Mallory insist that the Caucasus were an impenetrable barrier to linguistic movement between the steppes north of the Black Sea and Anatolia, Armenia, Iran, etc. This has been one of the main arguments against IE languages originally coming out of the south. So one wonders if there are corridors running north-south and if they reflect some special genetic identity or diversity.

Archaeology has not been keen on connecting the western shore of the Caucasus with the Balkans via the Black Sea before the Iron Age. I’m wondering if there is genetic uniqueness about the Georgian coast. That coast should have been of particular interest to hunter-gatherers. It would also be a logical avenue in later times.

Finally, you’ll see in the link below that the area from Iran to eastern Anatolia has been suggested as the origin point of R1a some time about 5000 years before the present.
If we premise that this group was associated with cattle-raising, is there an environment in the Caucasus that would favor this kind of agriculture?

SEE http://dienekes.blogspot.com/2014/03/major-new-article-on-deep-origins-of-y.html

thanks

Unknown said...

terryt wrote:
""I am prepared to believe that the proto-Indo-European population was high on XYZ haplogroup"

And perhaps it would be wise to not even believe that. If "Proto-Indo-European" did not fall full-blown from the sky, it would have had to have sister dialects that were part of a Pre-Indo-European group of languages. If that particular dialect suddenly became the popular one, it could already have been shared by different Y-dna groups who found it worthwhile talking the same language. As it was being adopted common language, the original speakers may have had nothing to do with its spread.

Jim said...

David,
"Concerning Diakonoff, we did not mention him, but Starostin. I think Starostin made a sound contribution into evolutionary linguistics."

Yes. I referred to Diakonoff simply because he also is prominent among Nostraticists.

I think you will turn out to be right about Starostin and his contributions. He gets lumped in with people like Greenberg and Ruhlen because his claims are expansive, but the difference his detractors are missing is that he is careful in his methods, and I expect them to withstand scrutiny.

David,
"Of course, if present-day ethnic groups are considered, all populations of the western Greater Caucasus largely descend from this lineage, including Ossetians, Kartvelians, and Adyghe-speakers, which all we consider, to a great extent, descendants of the glacial-time population of the region."

You repeat in your paper time after time that the distribution of these haplotypes follows geography more closely than language group. There is really no other way to read that.

"It also stubbornly insists that R1b was probably Indo-European, which is actually unlikely" - I still do not see sufficient evidence of this. Will be thankful for the newest references."

R1B is a big question. Why would anyone dogmatically associate it only with IE, when it is so widespread in Chad and that region?

George said...

LBK(Central Europe) and Starchevo(Balkans) cultures were mostly G2, are these cultures Kartvelian?

Kurti said...

@ capra internetensis

Scythia was more of an Empire. An Empire similar to Rome or Mongol Empire. And as such it of course included territory settled by non Scythian speakers. Most of them Iranic speakers anyways such as the Dahae, Massagatae, Sarmatians, Cimmerians. And these were never claimed to be Scythian by any author.

Than you had Mongols in the East and Finno_Ugric Speakers in the North included in the Scythian empire. But than No author, neither ancient nore modern EVER claimed them to be Scythians and they were clearly distinguished from the Scythians. So to claim Scythians were not all Iranic, because of the fact that some non Scythian tribes were conquered by Scythians, is as logical or fair as claiming "Romans" were not all Italic Speakers or Mongols were not all Mongols because they managed to conquer vast non native land.

And I don't see anyone doing the sais to any other historic tribe. What is the obsession about the Scythians that from every corner you see people coming and make absurd claims against every scientific facts. And some of the worst from my experience are pan Turkic ideologists who go that far to even claim the Medes and their successors as Turks. I have even seen some Turks claim Native Americans as lost Turkic tribe. Just based on basic nomadic similarities.This is simply incredibly absurd. There is clear description of Early Scythians by Greek Historians like Heredotus, who describe them as relatives of Sogdians (which was identified by linguists as Iranic). There are Scythian inscriptions of Khotanese Scythian in Western China which were identified as Iranic.

David Tarkhnishvili said...

Kurti
OK, I looked through the paper you mention. Those are burials of VIII-X centuries AD of Alans that "migrated there from the northern Caucasus". Hence, we speak about much earlier times. Of course, Ossetians/Alans of VIII century AD were not different from present-day Ossetians genetically. This does not mean at all they were genetically close to prto-Iranian cultures that lived in the Eastern European Steppe in Neolitics or Bronze Age.

David Tarkhnishvili said...

"LBK(Central Europe) and Starchevo(Balkans) cultures were mostly G2, are these cultures Kartvelian?"

- nowhere it is said G2 are "Kartvelian". It is said that we anticipate that pre-historical, glacial-time G2 population could speak a proto-Kartvelian language more likely than proto-east-Caucasian or proto-IndoEuropean.

Rabbi Sardar said...

I have stated numerous times, for the past several tens of millennia, the Caucasus has been a sink - not a source - of populations. People arrived and settled a particular mountain valley, and then defended it with all their might.corporate entertainer

andrew said...

@Eurologist

I agree with your point on the Caucasus likely being a sink rather than a source. G2 is characteristically European first wave Neolithic and the Caucasus appears to be one of the few places in Europe (Sardinia being another refuge) where they were not largely superseded in the early metal ages.

But, I don't agree with your suggestion on mothers as an important linguistic source. The anthropological evidence quite strongly points towards fathers and peers having much more influence than mothers on their children's adult primary language in mixed language couples, even when some of a mother's language may be passed on to her children (but not grandchildren). Male dominated demic change is generally associated with language shift. Cases where the conqueror adopts the conquered's language such as the retention of Greek in the Eastern Roman Empire, are rare indeed historically.

@alpha-eagle

"Ydna R1b were come from Central Asian invaders a.k.a Oghuz Turks and huns"

Unlikely. In the Caucasus, R1b is mostly found on the south side of the "continental divide" distant from Central Asia, and Central Asia was overwhelmingly R1a rather than R1b. The R1b in the Turks and huns was probably picked up in Europe rather than brought from an Altaic Turkic homeland where R1b is rare.

@Valikhan

Scytho-Samaritan is a reference to pre-Turkic people of the Southern Steppe who spoke languages in the Iranian family of the Indo-European languages. They were replaced or assimilated by later Turkic folk migrations. Ossestian is the closest modern relative of this language.

@Jim

The linguistic link between Hurro-Uratian and NE Caucasian proposed by Diakonoff and Starostin is plausible. But, Starostin's Dene-Caucasian hypothesis, his Nostratic hypothesis, his Eurasiatic family, and his Borean hypothesis are all exceedingly dubious. The Russian school is definitely guilty of indiscriminate lumping in historical linguistics, even though the American and Western European linguists err too strongly in the splitting and overskeptical direction at times. Genetic evidence provides helpful guidance in these disputes, even though language is not equal to genes since language shifts happen under well established specific circumstances.

andrew said...

@ George

G2 is found in multiple Caucasian linguistic groups not just Kartvelian. But, it is a very fair guess that the first farmers were not Indo-European linguistically and spoke an ergative language in the same linguistic family as one of more of the Caucasian languages.

@ David Tarkhnishvili re linguistic association of R1b.

R1b is predominant in Western Europe and appears from ancient DNA to have arrived in the Copper Age expanding out of Iberia, probably with the Bell Beaker culture whose range it closely matches. The earliest archaeological culture in Western Europe in archaeological continuity with known Indo-European language speakers are the Urnfield people ca. 1300 BCE whose successor cultures speak Celtic languages. The Bell Beaker range is home to a Vasconic substrate in toponymns and in linguistic features like base twenty numbers. Critically, the non-Indo-European Basque who predate Indo-Europeans are the most R1b people in Europe and their ethnogenesis can be traced archaeologically to the Bell Beaker culture. Using the Basque as a comparison to Indo-European neighbors, the demic contribution of the arrival of Indo-Europeans is on the order of 8%, about the same as the Turkic contribution to Anatolia.

R1b is also found in the ancient DNA of Minoans whose high culture was contemporaneous with the Iberian Bell Beaker culture, and is as common in the older non-Indo-European Kartevlian layer in the Southern Caucasus as it is in the more recent Indo-European Armenian layer. The affinity of Armenian and Greek favors an arrival of Armenian post-2000 BCE when the Indo-European languages arrived in the Aegean area, and there is admittedly thin historical evidence pointing to a folk migration from the Aegean to the subsequent Armenian range following the collapse of the Hittite empire that is not contradicted by any strong evidence of any kind. The difficulty of fitting Armenian in the Indo-European tree from linguistic evidence alone can be attributed to areal impacts from Eastern Indo-European languages that muddied its original Western Indo-European language origin.

Alshynkazak said...

@ Kurti

You have the worst logic ever. So, just because there was no term Turkic at a certain point of time, that means that those people suddenly dropped from the sky? There were no Iranians until establishment of Iran which was later than the establishment of Scythians. Scythians are one of the ancestors of Turkic people, and spoke a similar language. I understand that your Euro-centrist mind cannot apprehend that fact, but face it, a large amount of Scythians were Turkic, as I mentioned, historians have agreed upon including Royal Scythians as Turkic people. Also, your Iranian Scythian agenda has so many flaws. At first R1a is Scytho-Sarmatian, yet when Turkic speaking Karachays end up as R1a as opposed to Ossetians who are G2, now G2 is suddenly Scytho-Sarmatian. Again, where is the logic in that?

George said...

Scythian culture is an Iron age culture, and there is no adna results from Scythians

andrew said...

@Alshynkazak

The Scythian language was part of the Indo-European language family and in particular the Iranian subfamily within it. Its most familiar modern relative is Persian which is about as closely related to Scythian as Swedish is to German.

The Turkic language are not Indo-European languages. The originated near the Altai Mountains. The Turkic people migrated from East to West across the Southern Eurasian Steppe starting a few centuries BCE and continuing through a few centuries CE. The genetic trace of the Turkic people's migration gradually declines from East to West, reaching a low point of about 8% in the modern day nation-state of Turkey.

Towards the later end of that migration period (close to the BCE/CE divide) the Turkic people arrived in the central Asian steppe occupied by the Scythians, there was population genetic admixture with the Turkic superstrate leaving a low double digit Turkic population genetic signature and there was language shift with left us with the ethnicities that went on to form of the national basis of central Asia's "-stans".

Prior to the arrival of this Turkic folk migration, the Scythians did not speak a Turkic language similar to the one currently spoken there.

Kurti said...

@Alshynkazak.

Normally I should simply ignotr you by that much nonsense. I can imagine who your "Scientists are".

R1a* in ALL Turkic speakers fall exclusively under the z93* branch. Which spealers for a huge founder effect nothing more and nothing else. Haplogroup R1a* at the other hand has as high frequency in the East Iranic speaking Pashtuns as in other Turkic speakers, but yet contrary to Turkic speakers they do also have other clades of R1a.

The paternal origin of Kazakhs, Uzbeks(heck even the majority of Uzbeks was once Tajiks), Turkmens etc is Iranic (Sogdian, Dahae, Massagaeta).

We see that on the fact that the very first Turkic Dynasty, the Ashina were original Iranic speaking people.

Even the term Ashina derives from Iranic and means blue. It is equivalent to the Kurdish/Pashtun word "shin" which also means blue.
http://en.wikipedia.org/wiki/Ashina_%28clan%29

Haplogroup R1b* in Turkomans is mostly L23. Ironically the only high frequency of R1b* m343 (basal R1b) is found among the Kurds of southeastern Kazakahstan who have been deported there by Stalin. Otherwise this rare sub clade is only found throughout the Middle East.

http://en.wikipedia.org/wiki/Haplogroup_R1b_%28Y-DNA%29#R1b_.28R-M343.29


It's not about beeing "Eurocentric" it's about having scientific knowledge and common sense.

Onur said...

Haplogroup R1b* in Turkomans is mostly L23.

As far as I know, most R1b in Turkmens are M73.

Kurti said...

@onur

So it is M73. Didn't knew that.
Still the point stays the same. R1b in Turkmens is founder effect. Almost no diversity.

Alshynkazak said...

@Kurti

Regarding the Turkmen R1b. M73 is found only in Turkmens, Altaian Kazakhs, Bashkirs, and Uyghurs. All very distant geographically, so this subclade is Turkic, but no, you will insist that it is Iranian.

Alshynkazak said...

@Kurti

You are such an Iranist that it is not even funny. You claim Ashina is an Iranian term, yet you don't even know a simple that fact that their true NON-SINICIZED name was AshinaS. So again and again your statements are bullcrap

Alshynkazak said...

@Kurti

Your statements completely contradict the findings of Cambridge University that state that Turkic presence in Scythia before Huns was not a myth.

[The Cambridge History of Early Inner Asia - Cambridge University Press, 1994 - p.285-287]

"It could be that the first mention of the name Turk was made in the middle of first century A.D. Pomponius Mela (I,116) refers to the Turcae in the forests north of the Azov Sea, and Pliny the Elder in his Natural History (VI, 19) gives a list of peoples living in the same area among whom figure the Tyrcae. Usually this information is discarded with the argument the forms are corrupt, and reflect the Herodotian' Ivpkai(IV.22) located in the same region, a possible explanation but not more likely than its converse, to wit, that Herodotus has the wrong form. The argument most often heard against the identification-namely that the Latin forms cannot be correct because Turks appear much later and in a different part of the world - is fallacious, a school-example of petitio principii. Another, perhaps far-fetched, possibility would be that the Latin authors emdended the Greek name to a form familiar to them through other sources. I see no compelling reason to impugn the Latin data. The presence of Turkic-speaking peoples in the Pontic steppe and in the forest belt to the north of it is well established in the fifth century- i.e., well before the Turks' appearance in Mongolia-and Mela's information would simply testify to such presence at an earlier period.

So there you go, do I listen to a Cambridge publication, or to Kurti on Dienekes'? I think the answer is obvious.

Alshynkazak said...

@Kurti

Have you not read your own WIKI link that you posted?

""The recent re-reading of the Bugut inscription, the oldest inscription of the Ashina dynasty, written in Sogdian, by a Japanese team of philologists has suggested that the name, known only with the Chinese transcription of Ashina, was in fact Ashinas"

Simon_W said...

My own conclusions from the paper:

R1a is from the PC steppe; most of all the Kiptchaks have it, somewhat less the Adygei too, and there is a slight presence in the Oguz; strangely the Ossets have virtually none, but instead a neighbouring Kartvelian group has quite a bit of it. So it looks like this Kartvelian group is more Indo-Iranian on the paternal line than the Indo-Iranian Ossets. R1a has been found in ancient DNA from the Eurasian steppe from an early date on, so it's no surprise that Turkic tribes assimilated some of it and took it to the Caucasus, but the R1a presence in the Adygei may be much older.

Both J1 and J2 came from the south. J1 dominates in Dagestanic groups. J2 is present virtually everywhere, except in Dagestanic groups. It dominates in Nakhic speakers and in immediately adjacent Kartvelian groups. Ancient relatives of the Nakh-Dagestanic linguistic group may be the Hurrian-Urartians. Perhaps these had both J1 and J2. And founder effects explain the almost mutually exclusive presence of these in Nakhic and Dagestanic.

G2 is found most of all in Adygei and in Kartvelians (but also in the IE Ossets and the Turkic Kiptchaks). Was it originally associated with Kartvelian or with Adygian? In Adygei it's constantly the most common haplogroup, but nowhere exceeding 50%. In Kartvelians it's extremely dominant in the mountains, but in some other groups it's rather weak.

R1b is present in the NW Caucasus, but rare. In Adygei R1a is clearly more common. There is a peak in a southern Kartvelian group, and a lesser one in a Dagestanic group near the Caspian Sea, but on a population level it's strongest in Armenians.
As this recent paper
http://www.investigativegenetics.com/content/5/1/15
shows, the STR diversity of R1b-M269 peaks in the area where ancient Hittite, Luwian and Palaic were spoken, and it decreases towards the east, although the frequency increases towards the east. In Lezgins the diversity is particularly low. I think this suggests that R1b-M269 spread from the west eastwards and was originally associated with IEs.

PS. Did you notice? The colour coding they used is the same as on Eupedia! :D

PPS., @ andrew
R1b has been found in ancient Minoans?? You made that up!