search this blog

Tuesday, February 17, 2015

Latest linguistics research backs the Indo-European steppe hypothesis


Most of the action on this blog in recent weeks has revolved around the Indo-European urheimat question. So it's probably not a coincidence that I just got this press release in the mail:

Linguists have long agreed that languages from English to Greek to Hindi, known as 'Indo-European languages', are part of a language family which first emerged from a common ancestor spoken thousands of years ago. Now, a new study gives us more information on when and where it was most likely used. Using data from over 150 languages, linguists at the University of California, Berkeley provide evidence that this ancestor language originated 5,500 - 6,500 years ago, on the Pontic-Caspian steppe stretching from Moldova to Russia and western Kazakhstan.

"Ancestry-constrained phylogenetic analysis supports the Indo-European steppe hypothesis", by Will Chang, Chundra Cathcart, David Hall and Andrew Garrett, will appear in the March issue of the academic journal Language. A pre-print version of the article is available on the LSA website [see HERE].

This article provides new support for the "steppe hypothesis" or "Kurgan hypothesis", which proposes that Indo-European languages first spread with cultural developments in animal husbandry around 4500 - 3500 BCE. (An alternate theory proposes that they spread much earlier, around 7500 - 6000 BCE, in Anatolia in modern-day Turkey.)

Chang et al. examined over 200 sets of words from living and historical Indo-European languages; after determining how quickly these words changed over time through statistical modeling, they concluded that the rate of change indicated that the languages which first used these words began to diverge approximately 6,500 years ago, in accordance with the steppe hypothesis.

This is one of the first quantitatively-based academic papers in support of the steppe hypothesis, and the first to use a model with "ancestry constraints" which more directly incorporate previously discovered relationships between languages. Discussion of prior studies in favor of and against the steppe hypothesis can be found in the paper.

I'm reading the paper now, and it'll probably take me a while to get my head around it. Admittedly, linguistics is not my strong point, but I might post some observations in the comments if I feel up to it.

In any case, here's one of the phylogenetic trees from the paper. It'd be interesting to see how it lines up with thousands of complete Y-chromosome sequences from these language groups, particularly from Y-haplogroup R1; I have a feeling we'd see some very nice correlations.


Citation...

Chang et al., Ancestry-constrained phylogenetic analysis supports the Indo-European steppe hypothesis. Manuscript to be published in Language, (Vol. 91, No. 1) March 2015.

See also...

Massive migration from the steppe is a source for Indo-European languages in Europe (Haak et al. 2015 preprint)

Eastern Europe as a bifurcation hotspot for Y-hg R1

126 comments:

Krefter said...

Wow, they group Indo Iranian, Balto-Slavic, Celtic, Germanic, and Italic together. Then group Balto-Slavic, German, Celtic, and Italic together. Then German, Italic, and Celtic, Then Celtic and Italic

This fits very well with Y DNA.

Mike Thomas said...

Great Paper by Andrew Garret and his team.

But IMO, it only proves the chronolgoy, ie disproves a Neolithic time frame. Nothing in what they state proves the Ponto -Caspian steppe over the south Caspian pathway just south of it, which was otherwise relatively similar.

It seems the authors just fall back to those two mutualy exclusive models as an either/ or situation, and do not consider more nuanced axplanations of langaueg origins and spread. Also partly to blame is, being solely linguistcs, they do not understand the Kurgan model has problems from an archaeological POV. Their conclusion is ultimatley partial only - the steppe model is right because the Neolithic model is wrong. Go figure !

They write: "The evidence for contact between early
Uralic languages and the Indo-Iranian branch of IE is uncontroversial, supported by
dozens of unambiguous loanwords, and accepted by specialists (Rédei 1986, Lubotsky
2001, Mallory 2002). There is similarly clear evidence for contact with Balto-Slavic
(Kallio 2005, 2006, 2008) and Germanic (Hahmo et al. 1991–2012). The evidence for
contact with PIE itself is weaker (Kallio 2009), perhaps because Uralic languages
spread from the east into northern Europe and Proto-Uralic itself was not spoken in
proximity to the steppe."

Which means PIE could have come from anywhere, and Uralic only has uncontroversial contact with later, "mature" Iranian, Slavic Germanic as they subsequently spread further north. At least for the latter two - that might have been considerably later than 3000 BC.

Nirjhar007 said...

LOL!
They just don't get tired beating the dead horse do they????....

Nirjhar007 said...

@Anyone
As i don't have time now so i request would any one please point me the main 5 arguments at least by the authors in Support of the Steppe Hypothesis? I will analyse them in about an hour.....

postneo said...

Its interesting but Scratching my head on the titles of both this and Anthony's paper.

Now I have understood: Steppe hypothesis is just a term and does not indicate a location.

All they are trying to do is disprove the early chronology of Renfrew's Anatolian hypothesis, as if thats all they need to do. They have not even bothered to refute a more recent Anatolian hypothesis leave alone numerous other possibilities.

They offer nothing in terms of geographical location. Ringe's paper did not even emphasize much the shared word for horse which by far the strongest support for a Steppe hypothesis.


Mike Thomas said...

Yes postneo
See my above post

wagg said...

Mike Thomas: "Which means PIE could have come from anywhere, and Uralic only has uncontroversial contact with later, "mature" Iranian, Slavic Germanic as they subsequently spread further north. At least for the latter two - that might have been considerably later than 3000 BC."

That's just wrong. Not only it's been shown for MANY years, that Finno-ugric (Hungarian included) langauges have got _Proto-indo-iranian_, but also _old Indic_ (or something closely related), and indeed Iranic, but also NON-Indo-iranian Indo-european languages! (there are several studies (for instance one of Jakko Häkkinen) and it's explicitly mentionned in David Anthony's main book (the wheel, the horse, etc..)).

Also, NO, with so much IE (including OLD non-Indo-Iranian) loanwords in Proto-Uralic, the most likely hypothesis is BY FAR that PIE was pretty close to the south of Urals.

The Ural area was not exactly a highway in the neolithic time... it's was more a very forlorn peripheral area from where civilization grew and spreaded. It was a bit different after the domestication of the horse and the rise of the fully nomadic way of life (scythians/Turks, etc...). But at the time we're talking of? not exactly...

Mike Thomas said...

Nirj
Don;t worry, there is nothing in the article which "proves' anything. It doens;t even focus on the question of the IE homeland apart from a couple of chapters on chronology.

February 18, 2015 at 12:59 AM Delete

jaakkeli said...

Mike Thomas: "Which means PIE could have come from anywhere, and Uralic only has uncontroversial contact with later, "mature" Iranian, Slavic Germanic as they subsequently spread further north. At least for the latter two - that might have been considerably later than 3000 BC."

This is hardly true. There is more shared between IE and Uralic than the later loanwords identified with IE sub-branches and this existence is not controversial, only the meaning is (ie whether we have evidence of an "Indo-Uralic" genetic relation or not).

There are also words that are simply shared between proto-Uralic and proto-Indo-European and by definition we cannot identify the direction or timing of loaning here or even decide whether they are loanwords as it's perfectly possible that Uralic and IE are related. It is tautologically true that there are no uncontroversial loanwords between two proto-languages because we can only identify loaning between language families once they have sub-branch developments to give us evidence for the direction of loaning.

Many of the unexplained matches are between very basic words such as

name / nimi in modern Finnish
water / vesi in modern Finnish

which are close matches in p-U and p-IE reconstructions. There are more and more than enough to rule out chance resemblance. However we cannot declare them loanwords because they're just shared with p-U and p-IE and we can't identify the direction of loanwords or whether they come from a common "Indo-Uralic" ancestor which is somewhat plausible given that such very basic vocabulary, pronouns etc seem to match.

A lot of people seem to make some big deal out of stating that "identified contact loanwords in Uralic start from Indo-Iranian..." which is just tautologically true: we can't identify the shared words as loanwords until we get to sub-branches that have diverged enough that we can identify the specific source. In other words the "logical deduction" is like this

We have evidence for what seems to be *either* early IE and Uralic contact *or* a Indo-Uralic common ancestor

Therefore I conclude that there is no evidence for early IE contact with Uralic and I can put my homeland far away from the other language family in my pet theory...

This is an extremely lousy attempt to get rid of evidence. No. If you decide against early IE / Uralic loaning, the default option becomes an Indo-Uralic relation. If you decide against Indo-Uralic, the default option becomes loan contact between p-IE and p-U.

Mike Thomas said...

Wagg you're an idiot

Read what I wrote, and what the study quotes.

The study states that linguists only unequivocally agree that the earliest loans are from Indo-Iranian, the issue of 'early PIE; loans is more contentious.

Moreover, loans can only be dated relatively, not absolutely. So it shows that Uralic had 'early; contact with I-I, but cannot state definitively when this actually occurred.

Whatever the case, it doens;t prove that PIE came from Samara anymore than it shows that it came a mere 150 Miles south of that, eg Bactria.

Matt said...

Phylogeny and dates seem like the most interesting elements?

Repeats Grey and Atkinson's tree structure, while not their Anatolian Hypothesis supporting dates - http://armchairprehistory.com/wp-content/uploads/2012/05/atkinson-gray.jpg.

Like a sped up version of the same tree, although the Germanic-Italic vs Italic-Celtic grouping resolves slightly differently (they're all just part of a subclade on both trees though, so its hardly important for ancient origins).

Divergence of Hittite at 4000BC, Tocharian at 3500BC, then Greek-Armenian at 3200BC, Indo-Aryan divergence from remaining European branches at 3000BC.

So that's a divergence of Hittite before the earliest Yamnaya culture dates, Tocharian at around the beginning of the earliest Yamnaya culture and Greek-Armenian during the early Yamnaya culture.

Seems at first blush (without me having really read any of their work) that fits with Marija Gimbutas who saw PIE as pre-Yamnaya, I think (but who also thought a lot of crazy stuff), sort of still fits date wise with Anthony who sees PIE as Yamnaya, I think?

Doesn't fit with any idea that Hittite branched off around any time near the beginning of the Bronze Age Hittite Empire circa 1600 BC. Seems the Hittites were well separated from any other IE peoples for approx 2000 years by this time, and diverged earlier than any of other branches, well before the Bronze Age that is often reconstructed for PIE.

Seems like this is different from David W Anthony's previous preferred tree, the Ringe Tree (by his long term collaborator Don Ringe), where the split off of Armenian, Greek, and Indo-Aryan is fairly late down the tree (http://www.jolr.ru/files/(104)jlr2013-9(1-21).pdf and http://language.cs.auckland.ac.nz/files/2012/08/RingePaperTree2.png). http://www.academia.edu/10597023/The_Indo-European_homeland_from_linguistic_and_archaeological_evidence.

That tree has always seemed dubious to me, as a non-linguist, to have divergence between the European branches at all further up the tree than the Asian branches. Hopefully there is more consensus it can now be dispensed with. Anthony has mapped archaeological horizons to this tree (and I think this is substantially important in his model), so I guess those may need to be revised, if this is considered the most solid tree.

Maybe now we have more of a consensus on the tree, an Indo-Hittite tree, so urheimat estimates can focus on whether any lexical items with a purported environmental bias are particular to any subclade on the tree. Maybe this will eventually lead to graphics where each colour region has an approximately equal branch off date can now be revised in favour of this "true" branch order - http://en.wikipedia.org/wiki/Hittites#mediaviewer/File:IE_expansion.png

jaakkeli said...

Mike Thomas: "The study states that linguists only unequivocally agree that the earliest loans are from Indo-Iranian, the issue of 'early PIE; loans is more contentious."

Again, this is nonsense. The existence of early loanwords is controversial because some people want to argue that the early shared words reflect a common ancestor ie an Indo-Uralic language family instead of being simply loanwords. If that is so then the IE and Uralic homelands are not just close together, they are the same.

There is an unfortunate misunderstanding that has spread from this debate and the authors have fallen into this trap of assuming that the controversy over common origin vs early loaning means that there is a controversy about whether Uralic and IE have early ties that ask for an explanation in any homeland theory of either language family. There is not.

Mike Thomas said...

Perhaps, but I quote from the study:
"The evidence for contact with PIE itself is weaker (Kallio 2009),"

Im not sure of your expertise, but I;ll stick to what the expert quotes in his paper.

And, it doesn;t refute a homeland in central Asia or the steppe. Either way, it shows PIE and esp, Indo-Iranian spread north at some point.

wagg said...

Mike Thomas : "Wagg you're an idiot"

I must be on the internet.

Look cyber badass,
what you quoted didn't say "The study states that linguists __only__ unequivocally agree that the earliest loans are from Indo-Iranian", and several previous mainstream studies did show that some proto-indo-iranian loanwords, (unmistakenly) old Indic-like and NON Indo-iranian loanwords WERE effectvely present in Finno-Ugric (didn't YOU say "Which means PIE could have come from anywhere, and Uralic only has uncontroversial contact with later, "mature" Iranian".
Good luck disproving for example that Hungarian tehen (cow) is closer to Avestan dainu than Indic dhenu.

As for "The evidence for contact with PIE itself is weaker", I can't remember mentionning PIE, only old non-Indo-Iranian IE loanwords.

I was commenting of the fact that a forlorn peripheral region receiving through the time so much IE loanwords from different IE language families adds support to PIE appearing not too far from it.

That you want to agree with it or not is none of my problem. The validity of the words of a poster are not measured by his level of arrogance, the last time I checked.


"it doens;t prove that PIE came from Samara anymore than it shows that it came a mere 150 Miles south of that, eg Bactria"

I was not talking of a proof but of the most likely (by far) probability. I have no Idea why your Bactrians would have left their southern Paradise to end up chatting with Finno-ugric hunter-gatherers in their deep snowy cold forests... but whatever floats your boat...

Mike Thomas said...

Matt,

I think all modern IE languages, perhaps with the exception of Greek and some older, now extinct 'Balkan IE' descend from one singe branch, and then took hundreds of years more to evolve into actual Celtic, Germanic, etc.


Newer perspectives on Anatolian, no longer see it to be that . divergent. Granted it has some odd features, but these might be as much to do with the heavy influences of Non-IE languages in its vicinity as its age.

Mike Thomas said...

Then, wagg, you misunderstood both what I and you said.

I was referring to I-I generically. I know that Old Indic forms are attested surprisingly early, eg also in the Mitanni texts, as well as later Scythian names in the Black Sea area, a/p Trubachev.

As to why people moved from central Asia north, its probably similar to why they had done so since the Paeleolithic .

wagg said...

Darn*t!
Oops, I made a mistake.
I meant that hungarian "Tehen" is closer to old Indic (sanskrit) dhenu, not avestan dainu, obviously - that was precisely my point.

@ Mike Thomas:

"I was referring to I-I generically. I know that Old Indic forms are attested surprisingly early, eg also in the Mitanni texts, as well as later Scythian names in the Black Sea area, a/p Trubachev."

My English must be quite flawed. I was referring to Indic-like in _FINNO-UGRIC_ (You do know that Hungarian is part of UGRIC, right? What do have mitanni to do with anything in the context of our discussion? Are you trying me to tell that the vicinity of north Syria makes just as much sense as a source for early Indo-iranian? You do realize that the Urals are completely peripheral to the centers of human civilization of the time, right? while Syria was on the contrary within it. So it's easier to explain old Indic in northern Syria (especially as a few words concerning an Indo-iranian elite having a military role linked with horses and chariots) than in the Urals).

Also you explicitly said that "Uralic only has uncontroversial contact with later, "mature" Iranian", which is incorrect, so don't try to lie your way out of it: you weren't "referring to I-I generically".

Mike Thomas said...

Wagg I wrote ""mature" Iranian, Slavic Germanic as they subsequently spread further north"

Referring to all in the list. Admittedly a sloppy term, so my bad. I simply meant Indo-Iranian, Germanic and Slavic propper, c.f. still proto-INdo-European

And if you want respect and courtesy, dont YELL in your comments and start your posts with NO !!!

jaakkeli said...

Mike Thomas: "Im not sure of your expertise, but I;ll stick to what the expert quotes in his paper."

You are not even sticking to what your own quote says. It's about loaning.

There is no evidence of any direct loan contact between Icelandic and Yiddish (AFAIK). Shall we now conclude that Icelandic and Yiddish can be unrelated languages that could have come from any two different corners of the world with no origins in common?

You can repeat a million times that there's no evidence of direct Yiddish-Icelandic contact and it still won't make them unconnected languages.

You can repeat a million times that it's controversial whether p-U and p-IE have direct loaning but it still won't make them unconnected. Yes, there are plausible explanations for the shared p-U / p-IE features that don't require any loaning, but they put requirements on homelands as well. An Indo-Uralic language family, for example, could explain everything without loaning but then IE would go to the Urals.

If you have some theory that puts p-IE very far from p-Uralic you really do have contradictory evidence to explain and it won't go away no matter how many times you repeat quotes stating that it can all be explained in some other way than direct loaning. Yes, it can, so if you have some alternative theory you simply need to start by stating which non-loanword explanation for p-IE / p-U similarities you believe in.

It is simply not true that IE can go anywhere - sure, there are many explanations for the p-Uralic / p-IE similarities, but they will all put constraints on the IE homeland.

Grey said...

@Mike Thomas

You seem to be wrong

again.

Nirjhar007 said...

The Case of Aryan loan words in Uralic Just correlates with the BMAC migrations to area like Arkaim, Sintastha etc...

capra internetensis said...

@postneo

I don't know that the shared word for "horse" (~*ekwos) is a very strong argument for steppe origin. Wild horses were found over a very wide area, and then there are wild asses and such. Even if the word originally applied to wild horses, it could plausibly be independently shifted to mean the domestic horse in multiple branches, since this is a common and natural semantic shift.

And if it originally refers to the domestic horse, that would not allow us to distinguish between a homeland in the Pontic steppe and one in Central Asia somewhere. Possibly even further afield. There doesn't seem to be a consensus on when and where the horse was domesticated, and it could have happened more than once.

That said, the preservation of the word in practically every branch does suggest that they were familiar and salient animals for members of every branch, and having domestic horses from the beginning clearly fits the situation well.

andrew said...

FWIW, I think that the date for Hittite is way too old and that the apparent age of Hittite is a function of a substrate influence that is quite different from the substrate influences in other IE languages.

It is also hard to be confident about the Tocharian branch since the only archaeological culture definitively associated with it dates to about 1800 BCE. Obviously, it takes some time for the Tocharians to get from the PIE homeland to the Tarim Basin, but a 1700 year trek pushes the envelope. Tocharian is likely closer to PIE than any of the other IE languages, however, because it probably had the least substrate influence and the fewest areal influence, in much the same way that Icelandic is the most conservative of the Germanic languages for the same reasons.

Krefter said...

Chad,

I tend to think Bell beaker and Unetice will place right in north Europe, while CWC, Yamna, and the others will be southeast of them. F-statistics in Haak 2015 suggest Unetice and Bell beaker had pretty high WHG.

Mike Thomas said...

@ Jaakeli

There is no evidence of any direct loan contact between Icelandic and Yiddish (AFAIK). Shall we now conclude that Icelandic and Yiddish can be unrelated languages that could have come from any two different corners of the world with no origins in common?"

Irrelevant. Icelandic and Yiddish are both clearly Germanic.

"You can repeat a million times that it's controversial whether p-U and p-IE have direct loaning but it still won't make them unconnected"

Thats not what I was saying. You have misunderstood.

"It is simply not true that IE can go anywhere - sure, there are many explanations for the p-Uralic / p-IE similarities, but they will all put constraints on the IE homeland."

Yes, Clearly PIE did not come from Egypt or Melanesia. Well done.

What I was stating, is that the predominant loans into FU from the earliest contacts are Indo-Iranian, whether one accepts there were also genuinely "PIE" loans also.

My further point was that these loans can only be dated in terms of relative chronology and no honest scholar can attach absolute dates. So the Indo-Iranian -FU contacts could have come from 3000 BC or from 1000 BC. There is no point lying to yourself and using FU as a basis to 'anchor' PIE , because we don;t have a crystal ball view of FU itself. So we have essentially two floating variable.

3rd point:
I wholly accept FU - PIE contacts. But this does not resolve the specific location of PIE homeland to a great degree other than it must have been close-ish to FU. Now, if you consider that FU was once spread much further south to the forest zone, and not merely the sub-arctic boreal zone on the one hand, and the fluctuative character of steppe occupation on the other (eg zome periods the steppe was often depopulated), then there is no a priori reason why PIE could not have originally expanded from, say central Asia rather than the steppe.

Keep in mind also, that other languages might have been spoken on the steppe. Given that they are now extinct, we have no way of highlighting them in modern Uralic. So Uralic could have earlier stratas of language contact with other , non-IE languages.

Whatever the case, the strong FU - PIE contact is undeniable, but lets not be overconfident with what this means. It means that at some point, there was close and uninhibited contact between these two languages, without the need for an intermediary. This means that at any point between 3000 BC and 1000 BC (say), the two had become contiguous. But it doesn't pin-point the location of PIE homeland to a precision of less than 500 KM, but rather excludes proposterous scenarios - none of which anyone remotely entertains.

Roy King said...

I would wager that M269*(xL23) which is concentrated in SE Anatolia near the epicenter of Luwian and also found in the Balkans tracks the Anolian languages whereas L23-Z2105 and M412 track Greek/Phrygia/Armenian/Albanian (?Illyrian). M269* is "older" than say Z2105 and could have left the PIE homeland somewaht earlier than the Yamnaya horizon.

Mike Thomas said...

Perhaps, Roy.

But, if you are proposing that M-269* arose in the Ponto-Caspian steppe, then moved into Anatolia and the Balkans, all in the pre-yamnaya period mind you, are n't you worried about the lack of an explanatory theoretical motive ?

The pre-Yamnaya steppe (eg Repin and such "cultures"), dating from early to mid M4, show no evidence to posses expansionist abilities, or indeed motives. At that point, they were still mostly foragers, the steppe was sparsely populated, and the Balkan and Anatolian civilizations were still in their Copper Age zenith.

For me, its highly improbable.

What spurred on the Yamnaya culture was the appearance of Majkop. Quite clearly, the former modelled the latter, and the importance of the Ukrainian steppe revolved around its status as intermediaries between the Caucasus and the Balkans.

Roy King said...

@Mike
You could be right, but the distribution of M-269* fits the linguistic geography of the Anatolian languages. Also, the dating of Anatolian is likely pre-Yamnaya at 4500 BCE. The culture of the Anatolians seems quite different from the later IE languages. The Luwian healing goddess, Kamrusepa is a major player in Hittite mythology and derives from kammara (mist smoke, akin to himmel(German--heaven)/camisia(Latin)--shirt from a proto-IE root kem--to cover. Either the Anatolian languages and religion and culture were so transformed by substrate languages and culture or the proto-Anatolians never saw the strong horse/patriarchal cultures of Yamnaya, either originating in a different location or at an earlier time horizon.

Marnie said...

@Mike,

I just spent half an hour looking at currently archaeology of the Mesolithic in Albania, Macedonia, Thessaly, Epiros, and Serbia. There's quite a lot of evidence that people were living in these areas in the UP and Mesolithic. With the lower sea level, the contact zone between both Italy and Anatolia would have been a lot larger.

Anyway, I just put up a post on this on Dienekes site. It probably won't post until tomorrow.

Again, I think with high probability that G, R, I, and J hg rich pops could have inhabited these regions in the late UP and Mesolithic. They wouldn't have been completed isolated and could have moved between the Southern Balkans, Central Europe, Italy and the Anatolian Plateau.

Marnie said...

@Mike
@Matt

The funny thing about this tree is that root for Arvanitika (Albanian), Tosk (Albanian), and Ancient Greek split with all the other living IE languages more than 5000 years ago.

Greece and Albania were habitable 5000 years ago.

I just looked, and it looks like there were sites in Greece, Macedonia and Albania that were inhabited in the Mesolithic.

And lots of new archaeology showing a very early Neolithic in Northern Central Greece.

And ecozone continuity from Northern Greece and the central Balkan plateau to all the places where PIE spread.

Marnie said...

@Roy

"Either the Anatolian languages and religion and culture were so transformed by substrate languages and culture or the proto-Anatolians never saw the strong horse/patriarchal cultures of Yamnaya, either originating in a different location or at an earlier time horizon. "

It would be a tough ride for horses in Albania. Most shepherds don't use horses. It would scare the sheep.

:)

Richard Rocca said...

@Roy King Roy said..."
You could be right, but the distribution of M-269* fits the linguistic geography of the Anatolian languages"

Actually, M269(xL23) does not fit Anatolian languages. See my map here:

http://r1b.org/imgs/M269_without_L23.png

Also, M269(xL23) is actually L23's brother clade PF7562 and may actually be slightly younger than L23 based on STR signature.

Mike Thomas said...

Hi Roy

"Also, the dating of Anatolian is likely pre-Yamnaya at 4500 BCE."

I don;t think we need to push it that far, (although, sure pre-proto-Anatolian existed then) especially in light of what you stated " Anatolian languages and religion and culture were so transformed by substrate languages".

In fact, I'd flip that, and state that other I-E langauges and 'cultures' are a later subset of original diversity> hence their apparent greater homogeneity. Too much has been made of the supposedly Universality of IE culture and religion. This is simply false and is a lingering paradigm from the days of Dumezil.

West Asia has been overlooked as the PIE homeland becuase of the complex mix of langauges and cultures there. But I'd say this isnt a real reason. If you look at the Copper Age and EBA archaeology you'd see a landscape of more or less indendent regions focussed around one or two central places. Each would have had its own idiom, which would have had linguistic relatives, but not neccssarily in any territorially contiguous way.

Im not saying that Anatolia was the homeland,and certainly not suggesting a Neolithic time frame, but if you look at where the earliest and most manifest ;networks' which might have facilitated langauge spread occured, then it was on the northern periphery ("arc") around Mesopotamia, and not the steppe. The steppe only really opened up after c. 2000 BC. Before that, Each steppe group (western , central, eastern) were their own entities rather than the nice, eat west-> east spread Anthony suggests. In fact, each respective steppe region had more to do with the southern agricultural lands than the neighbouring steppe zone to its east or west.

To me the trail is clear - simply follow where the earliest IE Languages are- Aegean, Anatolia, northern Mesopotamia (Indic Mittani). Yes, they have no direct sub-familial relationship, but thats precisely becuase of their antiquity there and the fact that the attested forms were themselves latter developments resulting from secondary linguistic processes (such as convergence, substratum effects, etc).

Mike Thomas said...

yes Marnie
I don;t doubt that the Balkans will produce a wonderful array of genetic diversity.

Chad Rohlfsen said...

Uhm... of course the Balkans were populated in the Mesolithic.. who said it wasn't? Yes, the Neolithic too... I'm sure us regulars are familiar with Thessaly. Again, all Neolithics are G, I, C, and a lone R1b that isn't important or ancestral to anyone outside that clade.

Chad Rohlfsen said...

Plus, you can't group all of pre-Yamnaya with what went on in Samara. It's the most extreme NE of Yamnaya. I wouldn't label the whole thing as some backwater nobodies. They obviously were involved in bringing down the Balkan Chalcolithic. Doesn't sound like simple foragers.

Marnie said...

@Chad

And no ancient DNA from Thessaly, the very place you acknowledge as early Neolithic, in the right ecozone and central to be spread of IE.

Yet, you continue with an assertion of a direct connection between MA1 - > Samara.

Like I said, go for it! Cook up a bogus theory and publish a bogus paper.

How original!

Chad Rohlfsen said...

Ugh... I'm not even going to break down how that can't work. Go do some reading on Gumelinita/Karanova, Varna, CT, Hamangia, etc. Make a connection with Western Europe... and good luck, because it won't happen. You're arguing Balkans with the wrong guy!

Marnie said...

@Chad

Gumelinita/Karanova: Romania
Varna: Bulgaria
Hamangia: Black Sea

Wrong ecozone, wrong countries.

Chad Rohlfsen said...

Gumelnita-Karanova extends from Romania to Thrace. You clearly don't know much about them. Vinca goes down into Macedonia and Greece. Please show me how either of those are central to IE and link to Western Europe... Please do..

Chad Rohlfsen said...

We can go straight up Sitagroi if you like... You won't connect either to an IE spread to Central Europe and Western Europe. Where are the cultural parallels?

Roy King said...

@Rocca
"Actually, M269(xL23) does not fit Anatolian languages. See my map here:

http://r1b.org/imgs/M269_without_L23.png"

Lovely image, but it is the distribution within Anatolia that conforms to the distribution of M269*(xl23). Looking at the actual samples from Turkey from Myers et al and Cinnioglu et al, we found M269* to be concentrated in Se Anatolia, like Luwian. What is significant is their low frequency in East Anatolia/Armenia which are more concentrated with Z2105. About the route to Anatolia from the Balkans or the South Caspian for M269*, I'm pretty agnostic--Balkans tends to have more M269* as is apparent from your map with a concentration among the Bashkirs--close to the Volga. The reduce variation in YSTRs could be do to a lack of demographic expansion for this presumptive Anatolian language source, fitting well with a strong substrate influence on Anatolian.

Marnie said...

@Roy,

What about upstream from M269*?

Roy King said...

@Marnie,
Good question. I have been trying to argue that V88 is a Levantine/African sister to P297. V88 is found all the way to the DRC--Democratic Republic of the Congo.
I think that V88, P297, and its offspring M269 and M73 are all hunger-gatherer/mesolithic cultures that arrive in Africa/Urals/?Balkans independent of the Neolithic from the Near East. I am uncertain about M269 originating in the Balkans--could easily be the case with the Yamnaya being a Z2105 offshoot. Also Mike could be correct with M343/P297 originating from say Tajikistan/Hindu Kush/South Caspian etc... and moving north to the Urals. It cannot be the case that M343 originated the the Urals/Volga area IMO---it moved there with Mesolithic hunter-gatherers from either a Balkan or South Capsian or Hindu Kush refugium.

Davidski said...

The AG-2 genome, which was very similar to MA-1, the main ANE proxy, belonged to an individual who lived in Central Siberia after the LGM peak. Now, MA-1 of course carried Y-HG R*.

So I think it's unlikely that R-rich ANE populations re-expanded from the Balkans, South Caspian or Hindu Kush back to Siberia.

It's more likely that there was a refugium somewhere further north, where a pure ANE and R-rich population lived during the Ice Age, and this is essentially where R1a and R1b have their roots. How about the Altai or Western Siberia?

Krefter said...

I just can't wrap my mind around R* evolving solely around Siberia, while the most basal forms of P1 and R are found south of Siberia.

Maybe this is because Euro-Russian Y DNA is mostly the result of post-bronze age migrations from other regions and Siberians have a complex history added with not being sampled very much.

I don't know. It's just I would expect overwhelming evidence in modern pops of R, especially R1 diversifying in north Eurasia.

Krefter said...

Wasn't MA-1 R3 while all(most) modern R is R1'2, and therefore a brother not father to modern R?

Davidski said...

Siberia wasn't repopulated by ANE populations from the Balkans or the Near East after the Ice Age. If that was the case we wouldn't be talking about ANE-rich Yamnaya groups replacing Neolithic farmers of mostly Near Eastern origin in most of Europe, and bringing Y-HG R with them.

So R1, R1a and R1b don't come from the Balkans or the Near East, that much is obvious.

Mike Thomas said...

R* evolved in South Central Asia
Just look at the Hg K phylogeny

Mike Thomas said...

R1 somewhat over to the Indus - Caspian line
M269 is more debatable

Davidski said...

I have no faith in phylogeographic trees based on modern samples.

We already know K was in Western Siberia 45,000 years ago, R in South Siberia 24,000 years ago, and both R1a and R1b in pre-Neolithic Eastern Europe. But that's all we really know.

Grey said...

@Krefter

"I just can't wrap my mind around R* evolving solely around Siberia, while the most basal forms of P1 and R are found south of Siberia."

Another possibility is they mated with an archaic population that had evolved in the far north and that's what gave them an edge in those regions - like the Tibetans getting altitude genes from Denisova.

capra internetensis said...

@Mike

The Aegean, Mesopotamia, Anatolia naturally have the first attested Indo-European languages; no one else had (decipherable) writing.

There's that sample distribution thing again.

The considerable divergence between them indicates that they *didn't* develop in close contact - not that the distance need be very large in absolute terms, of course. The interesting point geographically is that Greek and Indo-Iranian, which are distributed on either side of Anatolian, are closer to each other than to the latter. (We can add Phrygian and Armenian to this pile as well.)

Mike Thomas said...

Very true Dave
We've all been led up the garden path with solely modern results
But I think this one is difficult to argue against
Esp given that Malta boy was a failed early northern dispersion
Between 24 and 7 kya there could have been other dispersals from Central Asia to the north although a place for other refugia like ukraine and the South caspian also must be considered ; althouh id include the latter in Central Asia anyway

capra internetensis said...

If you don't trust modern samples, then you will find *everything* comes from Siberia, because that's where the DNA survives. They couldn't even get Harappan DNA, and that's less than 5000 years old. Hopefully we will recover it eventually, but we can't ever expect to get samples as old as we can from cold climates.

aDNA is great when we have it, but we can't rely on it to the exclusion of other evidence.

Fortunately, we probably *will* get lots more aDNA from Northern Eurasia, and we should be able to trace the dispersal of R, if it originated there, or its arrival if it didn't.

Nirjhar007 said...

@ Guys
I have a suggestion to make of which i'm not staunch of so please help if i'm incorrect:).
A friend of mine just discovered an interesting detail on mtDNA linking Iran and Karelia, the Hg V2a which this
Iranian in the Genographic project has: "My mother is from Mazanderan province in far north-central Iran. Her mother was from Sari, Iran, in the same province. My maternal great-grandmother was from Kashan, also in northern Iran, but south of Mazanderan. My maternal great-grandfather was Azeri (Turkic) Iranian from Zanjan province, which is more northwesterly. My mother's father was from Sangsar in Semnan province, Iran. His parents, I am unsure of, but I believe were also from Semnan province, which is to the east and south of Tehran. All of this makes it logical that my maternal line's geographically closest deep ancestry heat map is northwest Iranian, and perhaps why I am closer to the Georgian (Caucasus) reference population than the average Iranian reference population. But my maternal haplogroup is found in populations all throughout Europe. An interesting note is that the hottest part of the heatmap for my maternal haplogroup is northern Scandinavia and specifically the Karelia peninsula of Finland and Russia! Less hot areas are just north of the Black Sea, and parts of northwest Africa and west-central African Sahel. Interestingly, there is very little shared with Anatolia and southeast Europe. All these whys and hows are stories waiting to be discovered."

So, maybe also a maternal line went from NW Iran to Karelia... actually, the Saami people has the highest concentration
of the V Hg: http://en.wikipedia.org/wiki/Haplogroup_V_%28mtDNA%29
It is also significant that it has been found in two Early Neolithic sites in Germany, already in 5000 BC, and this should
be of Near Eastern origin. Also in Reich's paper (p.19, p.25) there are instances of V mtDNA in EN and MN sites in Europe,
including Portugal and Spain, and never in Hunter-Gatherers sites. However, this suggests that V arrived there from Europe
itself: http://en.wikipedia.org/wiki/Population_genetics_of_the_Sami#V The Karelian HG belongs to C1g, as you guys know.
On the other hand, both V and U5b1b, found in the Sami, are shared with the Berbers... paradoxes of genetics! in Reich's
paper U5b is strong in Hunter-Gatherers, but even more in MN Germany (50%).
So whats the judgement?

Davidski said...

I think you're clutching at straws again. C1g is obviously North Eurasian, so it goes well with the R1a.

By the way, I'll try and blog about this new paper tomorrow.

http://www.biorxiv.org/content/early/2015/02/18/015396

Marnie said...

@Roy,

Thanks so much for taking the time to try to answer my difficult question.

"V88 is found all the way to the DRC--Democratic Republic of the Congo. I think that V88, P297, and its offspring M269 and M73 are all hunter-gatherer/mesolithic cultures that arrive in Africa/Urals/?Balkans independent of the Neolithic from the Near East."

Roy, look at this map for the range of capra aegragus:

http://www.lhnet.org/bezoar-wild-goat/

C. aegragus used to extend through Lebanon and Syria, into Jordan:

http://www.iucnredlist.org/details/3786/0

Furthermore, the Nubian Ibex range extends from Jordan, down the Great Rift to Ethiopia:

http://en.wikipedia.org/wiki/Nubian_ibex

Going east, there are also species of goats and sheep that extend into Kazahkstan and the Altai, such as the Siberian ibex.

If early R1b* were hunting in the zone of C. aegragus and the Siberian Ibex, this would explain his range, from Anatolia, to Jordan, to Afghanistan, Kyrgyzstan and even to the western Himalayas.

Appearance of R1b in the Balkans and the rest of Europe comes when R1b follows the chamois and ibex. Early in the Neolithic, C. aegragus is *brought* to Greece from Anatolia (around 8600 and 7000 BC), according this paper:

http://www.researchgate.net/publication/257802955_Past_and_present_distribution_of_the_genus_Capra_in_Greece

Lookind at the C. aegragus range, it has to have been brought by shepherds from Anatolia.

So that tells you that even 10,000-9,000 years ago, shepherds where moving between Anatolia and the Balkans. And they may have been moving over a much wider range than that.

"I am uncertain about M269 originating in the Balkans--could easily be the case with the Yamnaya being a Z2105 offshoot."

"Also Mike could be correct with M343/P297 originating from say Tajikistan/Hindu Kush/South Caspian etc... and moving north to the Urals."

Again, this is in the range of Ibex and capra.

"It cannot be the case that M343 originated the the Urals/Volga area IMO---it moved there with Mesolithic hunter-gatherers from either a Balkan or South Capsian or Hindu Kush refugium."

In Russia, the range of the Siberian ibex:

http://www.lhnet.org/siberian-ibex/

It does not extend to the Urals.

Just an idea.

Marnie said...

Roy,

I know it sounds a bit nuts, but more than 60% of the rock art at Ughtasar, a ritual site in Armenia, is of male C. aegagrus., dated starting about 7000 thousand years ago.

There are some rock art sites in Kazahkstan that appear to be either ibex or c. aegagrus.

postneo said...


"The AG-2 genome, which was very similar to MA-1, the main ANE proxy, belonged to an individual who lived in Central Siberia after the LGM peak. Now, MA-1 of course carried Y-HG R*."

Denisovans were found in siberia. Did they spread from there to Papua new guniea? How did they get there? Did they evolve from some endemic primate species of siberia?

ANE is a basal lineage and not north east eurasian.

Kristiina said...

Jaakkeli, at least I understand you and what you are saying! :-) 
Water (*wete) and name (*nime) are not Indo-Iranian loanwords, e.g. Sanskrit Water jala, jalá, vāri, vā́ri; name nā́man.
Uralic protoword for water *wete is closest to Proto-Germanic *watan and Uralic protoword for name *nime is closest to Tokharian B ñem. There are also other proto-level cognate words such as proto-IE *wīr-o- man, proto-Uralic *urV man, proto-IE *(s)kerǝ- bark, skin, proto-Uralic *kärnä bark.

However, indoeuropeanists usually want to take to themselves all innovations, word roots in more developed areas from Spain to China, huge areas extending from Central Asia to Europe and ydnas R1a, R1b, I and even J and sometimes also G. Uralic speakers are expelled to huddle up in Arctic tundra or at most in the underdeveloped forest zone and all this is sealed with yDNA N introducing small, ugly and dark Siberian & East Asian looks to Indo-Aryan race.

Now, it is somewhat funny that a R1a1 haplotype that could be an ancestor of Indo-Euroean R1a1 in Europe was found in Karelian forest zone and not in the centre of pastoralist developement.

I hope that ancient yDNA will bring more light to these issues and help resolve the homeland questions. However, yDNAs, such as R1a1 and N may prove to be much less important in the formation protolanguages than we now assume.

Davidski said...

Ahh, OK, so AG-2 and Y-HG R moved up north again after the Ice Age from the south Caspian.

Thanks, I almost forgot about the magical south Caspian; home of the Indo-Europeans and everyone else.

postneo said...

@Marnie
So far domestication of wild Ovi-Caprids points to to south and south west caspian. Other potential wild progenitors did not contribute.

The key point is that Wild Caprids are highly specialized for steep elevations. They can barely outrun a human on flat land, leave alone a cheetah.

Mike Thomas said...

@ Post-Neo

"ANE is a basal lineage and not north east eurasian."

That makes sense to me. Nothing I have seen to invalidate that, apart from no aDNA data from central and south Asia to yet confirm it.

@ Capra

Im not saying that IE began in in the south simply because that's where all the earliest IE are attested; rather due to the fact that there is where the earliest and most prominent networks first appeared: in a peripheral arc north of Mesopotamia- from western Anatolia to central Asia. That is undeniable; and the 'steppe highway' opened considerably later. Certainly, the simplistic archaeological equation of Yamnaya -> Afansievo and -> central -south Asia is not only unproven, but decidedly wrong. Rather, each region of the steppe (the western (Pontic), central (Kazakhstan), and east (Mongolia) developed their own trajectories of pastoral nomadism. They soon developed a many elements of pan-Eurasian steppe traditions, but this happened rather late - after 2000 BC, and certainly not from a wave of advance from Yamnaya. There wa nothing special about Yamnaya apart from its own uniqueness.

Mike Thomas said...

The fact that Greek, Anatolian, Indo-Iranian don't have any higher order of relationship beyond PIE itself doesn't invalidate the proposal, for reasons which I have already explained.

Mike Thomas said...

@ Kristiina

"However, indoeuropeanists usually want to take to themselves all innovations, word roots in more developed areas from Spain to China, huge areas extending from Central Asia to Europe and ydnas R1a, R1b, I and even J and sometimes also G. Uralic speakers are expelled to huddle up in Arctic tundra or at most in the underdeveloped forest zone and all this is sealed with yDNA N introducing small, ugly and dark Siberian & East Asian looks to Indo-Aryan race.

Now, it is somewhat funny that a R1a1 haplotype that could be an ancestor of Indo-Euroean R1a1 in Europe was found in Karelian forest zone and not in the centre of pastoralist developement.

I hope that ancient yDNA will bring more light to these issues and help resolve the homeland questions. However, yDNAs, such as R1a1 and N may prove to be much less important in the formation protolanguages than we now assume."

I entirely agree with what you say.
In an above comment, I described that FU likely existed far to the south of its current range. As Lamberg-Karlovsky pointed out, we have no evidence that the supposedly "Kurgan IE" cultures in much of Eurasia aren't also those of pre-Mongolo-Turkic, FU-speaking peoples etc. Its an arrogant and stupid proposition, but one which is the currently dominant one.

Marnie said...

@postneo

yes, I know where sheep and goats were domesticated.

However, the paper that I put up shows that domesticated goats were brought from Anatolia to Greece about 10,000 years ago.

Roy made a comment about R1b Mesolithic. I'm noting what could have been hunted by R1b prior to domestication.

wagg said...

@ mike thomas : "we have no evidence that the supposedly "Kurgan IE" cultures in much of Eurasia aren't also those of pre-Mongolo-Turkic, FU-speaking peoples etc. Its an arrogant and stupid proposition, but one which is the currently dominant one"

It might have something to do with the fact that we don't have evidences of ancient pre-mongol/-pre-turkic loanwords in Uralic (or in europe) - while we have diverse IE ones -, and no (pre-)Uralic loanwords in IE languages and hydronymia in Europe (cultures deriving from each others should at least have parts of their languages in common, right?), etc..., while there seem to be a correlation between kurgan-derived cultures (from which we are pretty sure the corded ware culture is derived now (at least the north-west Yamnaya which has more in common with (starting with cord impression on potteries, and similar axes))) both in Europe and in Asia, at the right timeframe, with IE languages (IE is spoken everywhere where these Kurgan-derived cultures had popped up).
We can relatively well correlate archaeological cultures and the presence of IE languages (even in the Tarim Basin, even if it's still quite controversial. We see, chronologically where they spread and what kind of languages were speaking the first known historical locutors in this place.

We can also wonder why there would be such a difference of life between for instance Uralic speakers on the steppes and Uralic speakers in the snowy forest just above on the map.

Also, the word Arya has been borrowed in Uralic and it seems to imply that the Aryas were enemies of the south, from the Uralic speakers (cf. David Anthony's book), so this also tend to support, among the huge heap of convincing elements, the association of the Kurgan-derived cultures of Bronze age central Asia with early Indo-iranian-speaking culture.

But yes, arrogance must be the main reason, and stupidity from a large part of the academic world, instead of common sense, another good reason. Obviously.

Kristiina said...

Forgot to add this:
"Now, it is somewhat funny that a R1a1 haplotype that could be an ancestor of Indo-Euroean R1a1 in Europe was found in Karelian forest zone and not in the centre of pastoralist developement."

I have seen people arguing that this Karelian man spoke Uralic protolanguage and others who claim that he spoke Indo-European protolanguage. Either way, here we have genetic evidence for Indo-Uralic language family. Who knows what is the next surprise to come!

Kristiina said...

Wagg, by the way, Arya in Uralic languages means "slave". In modern Finnish slave is orja.

Davidski said...

The Karelian forger didn't belong to the common Indo-European lineage R1a-M417. He belonged to the sister clade and nowadays very rare R1a-M459. So he probably wasn't a pre-Indo-European, but one of his cousins was.

Nirjhar007 said...

"'Also, the word Arya has been borrowed in Uralic and it seems to imply that the Aryas were enemies of the south, from the Uralic speakers (cf. David Anthony's book), so this also tend to support, among the huge heap of convincing elements, the association of the Kurgan-derived cultures of Bronze age central Asia with early Indo-iranian-speaking culture.''
Its just BMAC migrations towards Urals bringing Aryan language influx.
The SSC and BMAC were trading civilizations which created several trading posts all over Eurasia...

Nirjhar007 said...

@Kristiina
'' Arya in Uralic languages means "slave". In modern Finnish slave is orja.''
Sounds similar yes its interesting what is the self designation term in Finnish?

Nirjhar007 said...

@David
'' So he probably wasn't a pre-Indo-European, but one of his cousins was.''
WOW!You Once told me he was Pre-PIE!!!
Ah you were quite emotional i guess;)...

Kristiina said...

Nirjhar, do you want to know how I say I am Finnish in Finnish? That is "olen suomalainen". This word "suomalainen" has a myriad of different etymologies.

Nirjhar007 said...

@Kristiina
Thank you ! i will now hunt for cognates in ALL I-I branches for a term similar to suomalainen....

Mike Thomas said...

Yeah not quite
My point was there is no reaon to argue; and indeed no evidence to claim that all kurgans in eurasia only represent IE speaking peoples; on the one hand, and there are no kurhans in the Tarim, on the other . So your simple little explanation is wrong. And no one one mentioned Mongol ; which existed only since the high middle ages (a bit off with ur timing).

Nirjhar007 said...

@MT
To whom You are referring to?

Shaikorth said...

Nirjhar, looking for Indo-Iranian cognates with that might take a while since proto-Uralic self-designation term probably wasn't anything close to it.

Proto-Ugric mańć (man or person) from which comes Magyar (self-designation for Hungarians) might be closer, but don't count on even that. I couldn't find the term in proto-Finnic never mind proto-Uralic which is what you should be looking for.

Nirjhar007 said...

Corrected-
Guys,
I just took a look on Anthonys book at Page 385 let me inform you that the Words which were borrowed in F-U seems to be belonging to Indo-Aryan, the Iranian characteristics is absent and there is NO need to fabricate them with Proto-II etc they are just Indic also they Show clear vowel changes from Original A->O and E
Like in Arya to Orjo, Madhu to Mete we must remember that original PIE had AIU as vowels and E and O were derivatives of the Trio a reverse of what is popularly known.
http://new-indology.blogspot.in/2013/07/indo-european-linguistics-indo-iranian.html

Kristiina said...

Nirjhar, what words you mean when you say "at Page 385 let me inform you that the Words which were borrowed in F-U seems to be belonging to Indo-Aryan"? Does that book argue that water (wete), name (nim) and man (uro) are Indo-Aryan loanwords?

Nirjhar007 said...

@Shai
''Proto-Ugric mańć (man or person) from which comes Magyar (self-designation for Hungarians) might be closer, but don't count on even that. I couldn't find the term in proto-Finnic never mind proto-Uralic which is what you should be looking for.''
Thanks i know that anyway i didn't find anything so far....
It ONLY gives the same conclusion I-I/I-A Intruded the F-U Zone and there were some clashes and one way Influences on the other hand the Vedic and Avestan Tradition had nothing to do with those distant happenings...

Nirjhar007 said...

@Kristiina
''Nirjhar, what words you mean when you say "at Page 385 let me inform you that the Words which were borrowed in F-U seems to be belonging to Indo-Aryan"? Does that book argue that water (wete), name (nim) and man (uro) are Indo-Aryan loanwords?''
Except Wete Nim and Uro etc which i think are results of Previous migrations from Zagros-Zarzian or more simply a common Nostratic heritage i think and see there is NO word which can be labeled as Pre-Aryan or Indo-Iranian at all!, all aryan related terms can simply be termed as Indo-Aryan without any doubt...
We shall also know that morphological similarity also exists with PIE and Semitic as shown by Sergeant i think again pointing an Ancient Nostratic type structure.

Kristiina said...

People have their pet theories. Mikko K. Heikkilä in his recent academic work argues that ethnonyms Suomi, Saame and Häme are all connected with the Germanic word meaning dark *sǣmaz. I do not readily buy his theory of Germanic origin but I think that the idea "dark" is not that bad. Another idea is that these words are derived from the Baltic word for earth *ǯem-i-. However, I prefer to think that this word has deep roots on the area and is connected with the pre-Uralic/Uralic lexicon.

Nirjhar007 said...

Can you link please it may help on my research:)...

Nirjhar007 said...

Interesting Kristiina Suomi, Saame and Häme is similar to Sanskrit Shyama meaning also dark!

Nirjhar007 said...

see here-http://spokensanskrit.de/index.php?script=HK&beginning=0+&tinput=+shyama&trans=Translate&direction=AU
Now i have to check a Similar PIE term Exist or not ....

Nirjhar007 said...

@Kristiina and others
Nope! So it appears at the moment Germanic *sǣmaz Aryan Zyama both meaning Dark has no other IE cognates but its application on Suomi, Saame is tempting, if someone has other related and similar IE cognates PLEASE refer....

Kristiina said...

Nirjhar, I was aware of those Sanskrit correspondencies!

Heikkilä's work is in Swedish, so you might need the Google Translator to figure out the meaning of the phrases. https://helda.helsinki.fi/bitstream/handle/10138/135714/bidragti.pdf?sequence=1

Nirjhar007 said...

@Kristiina
Thanks though google translate is very stupid many times any way it will be great! also this that Sanskrit is unbelievable as Maxmueller concluded Long Ago-
“[A]s in his language and in his grammar [the Indian] has preserved something of
what seems peculiar to each of the northern [Indo-european] dialects singly, as he
agrees with the Greek and the German where the Greek and the German seem to
differ from all the rest … no other language has carried off so large a share of the
common Aryan heirloom – whether roots, grammar, words, myths or
legends”
Even after the discovery of Hittite and Tokharian etc the practical value of that Centuries old conclusion is quite the case.

Roy King said...

@Marnie,
Great maps and info on ibex and wild goat in the Near East and Greece/Balkans, Marnie! I do remember reading about a high density of petroglyphs depicting caprids in the Caucasus. It makes sense that mesolithic hunter-foragers would pursue and hunt caprids in the regions of the map. Other proto-Neolithic groups achieved domestication of wild goats in the hilly flanks of the Zagros/Taurus mountains. But what if the R1b's--P297 and V88 continued to hunt ibex and wild goat? I am reminded of the indigenous people of California who chose to remain hunter-forager-fishers despite trading with and having knowledge of the large corn-based cultures of the Southwest and Mexico.

Richard Rocca said...

@Roy King said...

http://r1b.org/imgs/M269_without_L23.png"

Lovely image, but it is the distribution within Anatolia that conforms to the distribution of M269*(xl23). Looking at the actual samples from Turkey from Myers et al and Cinnioglu et al, we found M269* to be concentrated in Se Anatolia, like Luwian. What is significant is their low frequency in East Anatolia/Armenia which are more concentrated with Z2105. About the route to Anatolia from the Balkans or the South Caspian for M269*, I'm pretty agnostic--Balkans tends to have more M269* as is apparent from your map with a concentration among the Bashkirs--close to the Volga. The reduce variation in YSTRs could be do to a lack of demographic expansion for this presumptive Anatolian language source, fitting well with a strong substrate influence on Anatolian.

My point is, arguing that the extremely light modern day frequency of M269+PF7562+ is an indicator of direction flow is like placing the origin of M269 in Ireland because M269+L23+ is heaviest there.

capra internetensis said...

Nirjhar:

You know the sacred verses that the Druids, that took 20 years to memorize?

Imagine how much of the Indo-European "roots, grammar, words, myths, and legends" *those* would have had.

But they are all gone, like the rest of Indo-European religion in Europe, except a few bits like the Homeric hymns. Meanwhile in India oral tradition retained a huge corpus of texts of very great age.

That sample distribution thing again.

Also, your link is just unconvincing speculation. Do you think these sort of things have not been thought up and debated many times before? It's basically just saying that everything *could* work backwards from how it usually does. Yes, it could, but parsimony is a thing.

And judging by your comments on that post, I question your ability to distinguish between and Indo-Iranian and an Indo-Aryan loan. Have you brushed up on your IE linguistics since then?

DDeden said...

Black Sea was EurAsia's central Oasis, so Anatolia and Crimea/Steppe are both right.

Marnie said...

@Roy,

"But what if the R1b's--P297 and V88 continued to hunt ibex and wild goat?"

The rock art at Ughtasar indicates ritualistic hunting of male goats. Some of the rock art reflects elements of pastoralism, rather than hunting.

I did some snooping around last night and found a post forum (I don't frequent these by the way) that had some very good recent archaeological information connecting Western Upper Macedonia with both Southern Greece and Phrygia in Asia Minor. Apparently, Nicolas Hammond noted a striking similarity between the archaeology of Phrygia and Upper Macedonia in what is now Grevena, Mount Voion, and Kastoria.

Again, with the sheep, I'm just trying to point out that shepherds could have been moving back and forth through the sub-alpine zones with their domesticated sheep, between the Central Southern Balkan Plateau, Asia Minor, and points further south and east, for quite a long time.

Which could mean that R1b hgs have been "sloshing back and forth" in this sub-alpine zone for thousands of years [as opposed to very gradually progressing from Kazahkstan to Europe.]

Actually, the early divergence of "Irish" tells us that R1b was on the move.

postneo said...

@marnie
Some of the first Hunter G. to hunt and domesticate goats in the Caucasus and Zagros may have been R1b.

Perhaps they were cold adapted(culturally) since they had to stick to mountains at first.

Heres a map that superimposes topography and some early neolithic sites.

http://postneolithic.blogspot.com/2015/02/first-farmers-and-ungulate-dung.html

Chad Rohlfsen said...

You guys are missing important links. Early Mako and Vucedol are considered ancestral to German Beakers. Look at our Beakers plot with Mako BR1 and to the east. European R1b is out of the Carpathian Yamnaya descendants. But, go ahead and stumble through stuff that has no link to German Beakers. It makes good entertainment.

Mike Thomas said...

@ Chad
" European R1b is out of the Carpathian Yamnaya descendants"

Well yeah out of carpathians ; but who's 'descendants' they really were remains to be demonstrated. To me; the steppe is not yet self-evident

capra internetensis said...

@Mike

I agree that the West-Central Asian zone of long-distance trade and interaction is a good place to look for PIE origins. However, someone needs to go beyond the first step and make concrete proposals about the correlates of the spread of Indo-European languages.

How and when did Greek speakers get to Greece? How and when did Indic speakers get to India?

This something that proponents of the Kurgan theory have done in considerable detail.

Gamkrelidze and Ivanov did postulate dates and offer a few vague archaeological correlates in their magnum opus, but most of their evidence concerns loanwords, toponyms, and mythology.

Mike Thomas said...

Capra
... await my paper ..
It may take a few months

Marnie said...

@postneo

I read your post.

Very interesting about the sights for early Neolithic Revolution.

The observation that sedentary agriculture would not have been competitive without animal dung is key. I agree, and on the hillsides where they farm in Greece, they rely on animal dung for fertilizer.

The first farmers must also have kept sheep or goats.

Marnie said...

... correction ...

They farmed with animal dung until about 50 years ago. [Now they use commercial fertilizer.]

Laziz said...

"Linguists have long agreed that languages from English to Greek to Hindi, known as 'Indo-European languages', are part of a language family which first emerged from a common ancestor spoken thousands of years ago."

Right from the beginning that statement is false. Never was such common ancestor found; it is only speculation.

Roy King said...

@Richard Rocca,
I agree and that's why I said I'm agnostic about M269(xl23) origin. There is not enough data as Marnie pointed out. What I'm trying to do is deduce which lineages covary with the Anatolian languages vs. Italo-Celtic, Greek-Armenian etc... Your maps indicate a concentration of M269(xl23) in Western/Central Anatolia--data which come from our Turkish study of Cinnioglu et al. I think M269* better tracks the Anatolian languages than Z2105, which better track Greek/Armenian.

Marnie said...

@Roy

The Phrygians of Anatolia were definitely connected to both Thrace and Upper Macedonia and were considered to speak a language of the Southern Balkans.

Some info here, for instance:
http://www.ancient.eu/phrygia/

The Phyrgian position in Anatolia/Asia Minor would have brought them into close contact with Armenia, which extended further west than its boundary with Turkey today.

Mike Thomas said...

Laziz
Of course it hasn't been "found". But the admittedly abstracted and reconstructed proto langauge is acceoted unless youre wanting to throw the entire comaparative method out the window

Mike Thomas said...

Marnie
The theory that Phrygians came from the Balkans has little solid evidence apart from Herodotus playing on puns (Brygi - Phrygi) and outdated archaeology
There are recent perspectives whicj view the Phrygians as local developments in westenr anatolia from the Bronze Age; which certainly raises questions about linguistic prehistory given their well known connections to ancient balkan IE languages like greek; Paeonians and ancient Macedonian

capra internetensis said...

The statement is "Linguists have long agreed" - which is perfectly accurate - not "time travelers have confirmed the existence of PIE".

Looking forward to the paper, Mike

Laziz said...

To Mike Thomas
The only thing we have is some similarities amongst different languages and this is quite normal. But if you stick, for some unclear reasons, to an original home or a language well spread all over half of the world, then you must come with some good arguments which go with it. But when we read your guys, what we have are some tribes without nothing, only what you imagination want to see and find. One thing we can say for sure is that it is not scientific. This is my opinion.

Davidski said...

You're wrong.

There are too many coincidences, like the <6,000 year-old split between the Balto-Slavic-Norse R-Z282 and Indo-Iranian R-Z93.

That's the Indo-European expansion from the steppe right there.

Marnie said...

@Mike

"The theory that Phrygians came from the Balkans has little solid evidence apart from Herodotus "

I didn't say that the Phrygians "came from" the Balkans. I said "connected to".

And I didn't say exactly when.

However, I see a recent paper by the archaeologist C. Brian Rose which says:

"The beginning of the Early Phrygian Period is generally dated to the 10th century BC, although the Phrygians' ancestors appear to arrived in central Anatolia from Thrace during the 12th century, at roughly the same time in which another Thracian group began to occupy the citadel of Troy . . ."

https://books.google.com/books?id=nQ8hAQAAQBAJ&pg=PA1&dq=phrygian+rose&hl=en&sa=X&ei=sInmVJfRFszvoATV8IDQDg&ved=0CB8Q6AEwAA#v=onepage&q=Thrace&f=false

Hammond also thought there was a connection. Beyond that, I don't know.

Marnie said...

@Mike

"There are recent perspectives which view the Phrygians as local developments in westenr anatolia from the Bronze Age; which certainly raises questions about linguistic prehistory given their well known connections to ancient balkan IE languages like greek; Paeonians and ancient Macedonian."


I agree.

Mike Thomas said...

Yes implying a greater antiquity of language contact between balkans and Asia minor thanngenerally assumed if one (falsely) takes the alleged Herodetean Phrygian migration at face value

Marnie said...

@Mike

" (falsely) takes the alleged Herodetean Phrygian migration at face value "

Yeah. Herodetean account obviously too simplistic.

Have many thoughts about something Roy said about the Luwians yesterday, and about the religious beliefs of the Early Greeks. I don't think the early Greeks were strictly patriarchal in their religious beliefs - given that there were so many female dieties. Athena? Demetra? Hera?

This seems at odds with the Steppe hypothesis.

Mike Thomas said...

Laz
I'm not saying the current "homeland " models and their explanations are right ; but the modern IE really are related and it's not an artefact of the analysis proces itself. However the tree model as currently constructed is not correct

Krefter said...

@Marnie,
"Have many thoughts about something Roy said about the Luwians yesterday, and about the religious beliefs of the Early Greeks. I don't think the early Greeks were strictly patriarchal in their religious beliefs - given that there were so many female dieties. Athena? Demetra? Hera?

This seems at odds with the Steppe hypothesis."

I lack knowledge on studying the structure of societies, but based on my experiences I tend to think the idea of "male-lead' and "female-lead" groups is too simplistic and a recent invention by feminist and the like.

This relates to what you guys are discussing because maybe you shouldn't take male-lead vs female-lead society so seriously when looking through evidence.

So, Marnie I doubt IEs from the steppe consciously thought "We're lead by males and must have mostly male gods, and will never change in this way later in history." Importance of goddesses in ancient Greece isn't good evidence against the steppe hypothesis. We don't know what religion they had for one thing.

I'm always confused by the discussion of "gender-roles", etc. It is usually presented as black and white topic but in my opinion it has many shades of grey. When people discuss gender roles it usually revolves around women being ambitious, etc.

There are general trends when it comes to how males and females behave in society, but it usually goes under the microscope of feminist ideals which gives a biased view of history.

Marnie said...

@Krefter

I'm not interested in getting in another long and drawn out discussion about "gender roles" with you.

It's just a comment about mythology. That's all.

God forbid that anyone should question the all sacred "Steppe Only Hypothesis."

Grey said...

@Marnie

"I don't think the early Greeks were strictly patriarchal in their religious beliefs - given that there were so many female dieties. Athena? Demetra? Hera?

This seems at odds with the Steppe hypothesis."

I think the steppe hypothesis would imply that where the IE took over as a new elite over a pre-existing society you should look for a god who was originally married to the main fertility goddess but got downgraded (and probably physically maimed) and replaced by a new thunder/sky god

e.g.

Vulcan&Venus
->
Zeus
Venus
Vulcan

and alternatively if you have a male/female fertility pair amalgamated into a pantheon as a pair that might imply an amalgamation of tribes

e.g.

Frey&Freya
vs
Thor&Odin
->
Frey&Freya&Thor&Odin

(maybe)

Colin Welling said...

I don't think the early Greeks were strictly patriarchal in their religious beliefs - given that there were so many female dieties. Athena? Demetra? Hera?

This seems at odds with the Steppe hypothesis.


I think the steppe hypothesis would imply that where the IE took over as a new elite over a pre-existing society you should look for a god who was originally married to the main fertility goddess but got downgraded (and probably physically maimed) and replaced by a new thunder/sky god

What's with these simplifications of an entire horizon of people. You think based on some archeological patters and related languages they all act the way you say they do?

You think all germans share a religion? Do you think all the spanish colonists to the americas mated with the locals (argentina)? Do you think all british colonies were prison camps.

You're wrong grey, a special elite did not change the autosomal landscape of europe, a mass migration did.

Marnie, nowhere is it dictated that indo europeans must worship male gods. Things are not black and white like feminists insist, especially for our own culture. Now that the term patriarchy has become so twisted to suite a political agenda I feel it is important to point out that women typically have more influence of the children including the sexist ideals they learn. The very gender norms that dictate a great deal of life for men and women are probably more influenced by women.

Grey said...

@Colin Welling

"You're wrong grey, a special elite did not change the autosomal landscape of europe, a mass migration did."

I only mentioned gods and a possible model - a maimed god married to the prime fertility goddess.

That part might only require an elite replacement.

Marnie said...
This comment has been removed by the author.
Marnie said...

@Collin

Like I said, its just a comment about mythology. I won't be sharing my thoughts further, given the close minded attitude that both you and Krefter seem to have.

Rather than wasting my time quibling about gender roles and mythology last night, instead I read a great recent paper about ceramics and the Mesolithic/Neolithic transition in the Balkans.

But I'm not going to share it here, given the agendas that so many on either side of the Steppe / Anatolian hypothesis debate seem to have.

By the way, if you keep going off on me whenever I make a comment vaguely related to gender, I will be sending a letter to Athene Donald about your online behavior.

capra internetensis said...

@Marnie

You are so weird. Or trolling hard, I can't decide.

Anyway, re: Phrygians, have you seen the Gordion Archaeological Project?

http://sites.museum.upenn.edu/gordion

According to them, at the transition from the Hittite-influenced Late Bronze Age to the Early Iron Age at Gordion (later capital of the Phrygians), there are "clear changes in architecture, domestic features, ceramics, and animal remains.... The ceramic finds do not support a gradual transition... instead, the evidence strongly suggests a population change at this time, rather than simply a shift in political and economic organization. Few Early Iron Age sites have been excavated in Anatolia, but similarities in pottery (handmade shapes and specific details of incised or impressed decoration) point to a Thracian or more distant southeastern European origin fro the immigrant group. This has been taken to mark the beginning of the presence of the Phyrgians...."

It's a very interesting website.

Simon_W said...

By the way, interesting that Greek, Armenian and Albanian split from the mainstream right after the Hittites and Tocharians. In all previous linguistic trees I've seen it was Italo-Celtic that split off third.

If Italo-Celtic came from the Yamnaya in the Carpathian Basin (which is the most likely hypothesis imo) then an early split of Italo-Celtic from the rest of IE would mean that Greek, Armenian and Albanian cannot be from West Asia - they would have to be from the steppe too - unless there were several migrations of IEs across the Caucasus from West Asia to the steppe, which is unlikely.

However, if the Armenian-Greek-Albanian branch split off earlier than most other IE languages, then it would be at least possible that they had simply stayed in West Asia, while all the younger branches were from the steppe. And then it would be possible to explain the Bronze Age expansion of haplogroup J2 and West Asian admixture into the Balkans and southeastern Europe as associated with the arrival of some of these languages. Which makes more sense than ascribing it to the expansion of some undocumented non-IE languages.

sujay rao said...

You are invited to read my publications on the Indo-European issues

https://www.researchgate.net/profile/Sujay_Rao_Mandavilli/publications

Sujay Rao Mandavilli


DDeden said...

Musing: (Earlier from Dead Sea, NW to Anatolia, split north around Black Sea, east branch gets horse/xorsa/Kazhkar(Kazak-Khazar) and loops around to south Urals meeting Balkanized west branch?