PDA

View Full Version : The R1b full circle anticlockwise circumpontic theory



alan
06-10-2013, 04:48 AM
You know this is the first time I think I really think this works. Here is what could be the story:

1. R1b regugium in Caspian area.

2. R1b is on the edge of farming expansion and doesnt join in until late Neolithic-either based in Caucasus area or north Iran. The lack of take off of P297 R1b prior to the rise of M269 and M73 c. 4 or 5 thousand BC would make it very likely it was based in a late farming arival area like the Caucasus or similar area of north Iran rather than to the south. The people involved are either local Caucasus-steppe interface farmers, migrants from north Iran or a blend of both.

3. Maykop copper age wonder culture starts c. 4000BC in the north Caucasus. It is linked to by network to northern Iran and also Uruk to the south (this could explain both L23 Assyrians and the suggested IE Eurphratic aspect in Sumerian). Everything would point to M269/L23 being involved and probably M73 too. The Maykop people may have been bilingual located as they are (were?) on an interface between Caucasian and PIE due to their position at the northern piedmont of the Caucasus and facing the steppes. They may even have needed to know a bit of Sumerian too.

4. Maykop is highly influencial on the western steppe peoples introducing developed metallurgy, the idea of Kurgans, social complexity and possibly even things like the wheel etc that were crucial for mobility. M269 and M73 becomes a part of the steppe population c. 3500BC.

5. c. 4000-3000BC various maves of steppes peoples arrive in the Balkans. Some of these include M269 and L23.

6. Over time L23 steppe elements move up from the Lower Danube to the Alps giving birth to L51 around Austria.

7. Other L23 elements in the Balkans head into Anatolia and the southern Caucasus in waves in the Bronze Age -Hittites, Greeks, Amenians, Phyrigians etc. This essentially completes the circle of the Black Sea that L23 had carried out. This model has the circle being completed at the Caucasus. This could explain why Armenian variance is not that high while steppe, east Balkan and Anatolian is higher.

8. It is possible that some of the groups of Maykop/CMP people more set in from the steppes were more Caucasian speaking. It is not impossible that the wave of new apparenty CMP derived copper working that swepped as far as Italy by 3400BC and Iberia by 3000BC could have included Maykop derived elements in the CMP Vanguard who had left the Maykop homeland when the Maykop-steppe culture interaction was barely starting and could have been primarily NE Caucasian speakers. That could potentially explain why some see a Caucasian linguistic link to Basque.

The full circle of the Black Sea idea has been coming to me for a few weeks but for the first time I think it really does beautifully fit the evidence.

lgmayka
06-10-2013, 02:08 PM
The full circle of the Black Sea idea has been coming to me for a few weeks but for the first time I think it really does beautifully fit the evidence.
If it fits the R1b evidence, how do you then fit it into the R1a evidence? And that of Indo-European?

alan
06-10-2013, 03:08 PM
If it fits the R1b evidence, how do you then fit it into the R1a evidence? And that of Indo-European?

Well I have never been convinced by the tendancy to try and make their stories utterly contrasting. I think that is overegged by some posters on some sites who like to treat R1b and R1a as some sort of Maykop was located in what essentially was the steppe-Caucasus interface with the steppes far more open than the area to the south. The connection to the Capsian and Iran was also probably much stronger than a direct connection south. There were steppe groups within Maykop territory too. What seems to have created Maykop is the link to metals and cultural influence from and via iran. However, that doesnt change the fact Maykop culture essentially ran into the southern part of the steppe. Its influence was massivee on the western steppes (metal, idea of Kurgans, wheels?) and supplied most of the metal for the cultures of the Ukraine steppe and probably had a role in the extension of CMP metallurgy to the Urals and to the Yamnaya people. So there is plenty of scope for mixing in the zone around the steppe-north Caucasus interface from Azov to the Capsian and of course then following trade routes etc from there. I dont think inference can quite tease out the exact details of this. It is possible that R1b was local to the steppes-north Caucasus interface or that it arrived from Iran with the high culture and metallurgy influences that created Maykop. I think there is room for a mix there but I dont really know. One thing I am clear about is that if R1b (M269, L23, M73) was associated with Maykop then it would certainly have filtered deeper into the steppes along its routes of influence that are very obvous in the Ukraine steppe especially but affected a much wider area too.

Also I think most would agree that R1a is very strongly associated with the Slavic, Baltic and Indo-Iranian branches of IE. However, Anthony links them with a rather different chain of cultures than the ones around the Caucasus-steppe interface. He links them with middle Dnieper-Fatyanov-Abashevo-Sintashta moving west to east. They featured both Corded Ware and steppes elements. I think an Abashevo-Andronovo linkup is also mentioned. The earlier link with Afansievo is unproven and is speculated from the Tarim dummies which are much later (I think keep an open mind on that). The question is whether the element that brought R1a into the middle Dnieper-Fatyanov-Abashevo-Sintashta chain of saetem IE associated cultures was down to the shared corded ware or the shared steppe element. I have played devils advocate on the corded ware element but it is also as possible that it came from the steppe element. Lets say for a moment that R1a is closely associated with Yamnaya. That is also a reasonable guess. Anyway it is possible that this derived from an R1a-dominated steppe group around the Volga-Urals who were subject to Maykop influences but essentially (other than a minority) remained R1a dominated. The timing of all of that actually fits very well. I think both the Maykop and true steppe groups must have been able to communicate. If IE was associated with Maykop I dont think it could have been down to the Iranian influence and more likely the steppe substrate. I say that because Iran has a whole range of animals that I am pretty sure are not in PIE. The question remains open whether M269 was a local north Caucasus-steppe interface or down to the influences from Iran c. 4000BC.

R.Rocca
06-10-2013, 05:44 PM
Alan, on the whole, I like your R1b-CMP theory. However, I do have some concerns around its chronology. From Chernykh (2008)...

The first phase covers the initial formation of the province, and the author prefers to call it the “proto-CMP”. This represents a return to the terminology formulated in the first large work applying radiocarbon chronology to the metalproducing centers of the Carpatho-Balkan and Circumpontic provinces (Chernykh et al. 2000: 14-18, 37-38). In fact the chronological range of this early phase includes the whole IV millennium BC. The prefix “proto” indicates that the territory of the productive centers of the CMP initially did not include all subsequent “Circumpontic” areas (Fig. 5). The northern Balkan peninsula, the Carpathian and Danube basins and the steppe zone of the northern Black Sea Coast remained within the borders of the dying Carpatho-Balkan province.

In the second phase a true Circumpontic province develops: its productive centers completely encircle the basin of the Black Sea. By then the disintegration of the CBMP is completed and its former territories are occupied by metallurgical and metal processing centers in which the morphological and technological standards of the CMP are completely dominant. This second phase covers the entire III millennium BC.

It seems that if Proto-CMP was very much localized in an area to the east of the Black Sea circa 4000-3000 BC and did not expand into the former Carpatho-Balkan province until 3000-2000 BC. This chronology is later than Italian Copper Age and would leave very little time for the development of Beakers all the way out west in Iberia starting around 2800 BC.

The CMP metallurgic entrance into the Balkans circa 3000 BC would then have to be the same as Heyd's Kurgan Yamnaya advance into the Danube, no?

alan
06-10-2013, 06:42 PM
Alan, on the whole, I like your R1b-CMP theory. However, I do have some concerns around its chronology. From Chernykh (2008)...

The first phase covers the initial formation of the province, and the author prefers to call it the “proto-CMP”. This represents a return to the terminology formulated in the first large work applying radiocarbon chronology to the metalproducing centers of the Carpatho-Balkan and Circumpontic provinces (Chernykh et al. 2000: 14-18, 37-38). In fact the chronological range of this early phase includes the whole IV millennium BC. The prefix “proto” indicates that the territory of the productive centers of the CMP initially did not include all subsequent “Circumpontic” areas (Fig. 5). The northern Balkan peninsula, the Carpathian and Danube basins and the steppe zone of the northern Black Sea Coast remained within the borders of the dying Carpatho-Balkan province.

In the second phase a true Circumpontic province develops: its productive centers completely encircle the basin of the Black Sea. By then the disintegration of the CBMP is completed and its former territories are occupied by metallurgical and metal processing centers in which the morphological and technological standards of the CMP are completely dominant. This second phase covers the entire III millennium BC.

It seems that if Proto-CMP was very much localized in an area to the east of the Black Sea circa 4000-3000 BC and did not expand into the former Carpatho-Balkan province until 3000-2000 BC. This chronology is later than Italian Copper Age and would leave very little time for the development of Beakers all the way out west in Iberia starting around 2800 BC.

The CMP metallurgic entrance into the Balkans circa 3000 BC would then have to be the same as Heyd's Kurgan Yamnaya advance into the Danube, no?

These metallurgical papers seem to use ridiculously wide millenial intervals. I noticed that in Amzallag and Chernykh. I think that partly became a convention among archaeometallurgists because of a lack of dating but would generally be considered totally useless in most fields of archaeology. Russian archaeologists also took ages to accept and apply calibration too. So many of the statements in papers are extremly vague.

I am not sure of all the details of internal chronology of CMP but I do know that the extension of CMP knowledge into the Volga-Urals area including the mining at Kargaly is dated to at least c. 3300BC with early Yamnaya material but my feeling is that is not likely to be the first generation settlement there as an external group simply had to have arrived (no matter how small the numbers) to transfer such specific knowledge and methods (that is what several metallurgists have said anyway). I would assume CMP metallurgy arrived earlier in the Ukraine steppes. However, I agree the chronology remains badly defined.

I realise too that the earliest steppe groups west like Suvoravo (linked to the Anatolian branch by Anthony) c. 4200BC were pre-CMP and metal poor - themselves contributing to the end of the Carpathto-Balkans group. I agreee there is a fine chronological line between the theory working or not. I dont know about all Italian copper age groups (such as Rinaldone etc) but Remedello is the one that is linked to the earliest mines in north Italy and has strong representation on the pre-beaker stelae in the Alps. I assume it is a major plank in Jean's book (I should know but I have a poor memory). I believe a stage of Jean's book links the Alpine stelae to Maykop influenced groups along the Black Sea shore like Kemi-Obi I think. I am inclined to like that idea now as the Crimea sort of area should have been one of the easiest and earliest places for Maykop to reach and they supplied most of that areas arsenic copper metalwork for a considerable time after the Urals groups had started using Kargaly pure copper.

The chronology is tight but I dont think given the issues with RC dating and the constant reviews the model is impossible. I think the model is fairly tweakable though. It may be that a leapfrog was done by CMP people moving up the Danube somehow pretty early using the river to move quick to the Alps. it is even possible that CMP lineages could move separately from actual larger steppe folk movements. I think CMP started infiltrating areas before the Yamnaya culture arose. So if it was partly integrated into and spread by steppe cultures then it had to be pre-Yamnaya ones if we are to link it with the spread of copperworking into the Alps. The dating of the mines and metalwork generally is not really precise if its from for instance carbon from a hearth or something like that. Even human bone dates can be significantly effected by diet. So, I think it will take a good selection and accumulation of dates to build up a true chronology. I think last time this was discussed Jean had a minor panic before it was concluded that there was no definate chronological knock out blow against her model which did indeed involved the spread of lineages along the trail of stelae (many of which have Remedello stuff). I was thinking of a date for a spread to the Danube and a leap up it of c. 3500BC which in steppe terms falls in between the really early Suvoravo type groups and the Yamnaya groups.

As I said though CMP moved through non-steppic areas too so it could move independant of steppe cultures. It early had access to Azov and the Black Sea so if they had any maritime technology they could outstrip actual steppe type wanderings west. That is even true in the steppe because the Maykop culture spread had access to Azov, the Black Sea, the Caspian Sea, the Don, the Volga etc and could theoretically reach the Balkans in one direction or the Urals in another without putting a foot on land.

It would be good if Jean can comment on the chronological issue, Kemi Oba etc as she settled on that sort of model some time ago. I think in light of the data I am now inclined to agree with most of Jean's model for R1b. I also agree with the alternative possibility of R1b getting into the steppes via the Balkans from NW Anatolian pastoralists although I have got a little less into that idea than I once was.

alan
06-10-2013, 07:09 PM
Alan, on the whole, I like your R1b-CMP theory. However, I do have some concerns around its chronology. From Chernykh (2008)...

The first phase covers the initial formation of the province, and the author prefers to call it the “proto-CMP”. This represents a return to the terminology formulated in the first large work applying radiocarbon chronology to the metalproducing centers of the Carpatho-Balkan and Circumpontic provinces (Chernykh et al. 2000: 14-18, 37-38). In fact the chronological range of this early phase includes the whole IV millennium BC. The prefix “proto” indicates that the territory of the productive centers of the CMP initially did not include all subsequent “Circumpontic” areas (Fig. 5). The northern Balkan peninsula, the Carpathian and Danube basins and the steppe zone of the northern Black Sea Coast remained within the borders of the dying Carpatho-Balkan province.

In the second phase a true Circumpontic province develops: its productive centers completely encircle the basin of the Black Sea. By then the disintegration of the CBMP is completed and its former territories are occupied by metallurgical and metal processing centers in which the morphological and technological standards of the CMP are completely dominant. This second phase covers the entire III millennium BC.

It seems that if Proto-CMP was very much localized in an area to the east of the Black Sea circa 4000-3000 BC and did not expand into the former Carpatho-Balkan province until 3000-2000 BC. This chronology is later than Italian Copper Age and would leave very little time for the development of Beakers all the way out west in Iberia starting around 2800 BC.

The CMP metallurgic entrance into the Balkans circa 3000 BC would then have to be the same as Heyd's Kurgan Yamnaya advance into the Danube, no?

I think something to bear in mind too is Maykop looks like its raison d'etre was to link metallurgical skills and source around Iran first to the peoples on the steppe-Caucasus interface and then into the steppes. Its wealth looks like its dependent on trade. Geography makes the steppes the most likely market for this. The steppes may well have lost their Carpatho-Balkan supply already when steppe groups spread into the Danube mouth and the climate did its worst around 4000BC or so. So, I really cannot believe that Maykop people (who basically could look into the steppe from there piedmont steppe edge homes from some point in the earlier 4th millenium) would not have been supplying this lack to some degree. It has long been thought that Maykop wealth was based on north Caucasus middlemen acting between the steppe and metallurgical areas to the south (or northern Iran as the latest paper suggests). If they were not supplying someone then what was the middleman wealth of the Maykop chiefs?

R.Rocca
06-10-2013, 07:23 PM
The below map shows a nice contrasting view between 'landtilling' and 'cattlebreeding' zones during the EBA and MBA.

From Černyh (1991) The Circumpontic Metallurgical Province as a System.

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

alan
06-10-2013, 07:59 PM
Actually I dug out the old thread where we were fretting over the chronology of making it all fit together

http://eng.molgen.org/viewtopic.php?f=85&t=823&start=100

alan
06-10-2013, 09:19 PM
I have tried hard to find info on Kemi Obi other than Anthony and Mallory's books but I am stuggling to find any.

alan
06-10-2013, 09:28 PM
looking at the map of the Caucasus and adjacent the Crimean must have been very tempting for Maykop people. They must have controlled the Sea of Azov anyway.

http://en.wikipedia.org/wiki/File:Maykop_culture-en.svg

The Crimean peninsula must have been a major temptation as it is the natural port area of the Black Sea. a rare non-steppe environment in this area and tactically useful raised land. Once of the peninsula the west of the Black Sea and the Danube would have been easy to reach by sea or land. I wonder what the balance was between sea/river travel and land travel in cultures like Maykop and Kemi Oba? It would seem on the surface to make sense to use the rivers for metals etc although less so for large cattle herds. As I posted before, most of the north Black Sea shore of the Neolithic is lost but there are hints of maritime arrivals like a sherd or two of cardial type pottery. The locals probably in maritime areas presumably had some boats and skills after thousands of years there anyway.

Mikewww
06-11-2013, 12:42 AM
This fairly new paper is interesting in terms of R1b in Iran
http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0041252

The very rare root R-M343* has a better representation than normal in Iran. It rises to a spectacular 4.3% among the ethnic Pesians of Yazd a little north of centre and and 3.2 among Azeri of Azerbaijan in the north-west. It also shows among the Gilaks of Gilan in the extreme north, the Persians of Khorasan in the north-east and the Kurds of Kurdestan in the north-west in lower numbers. It is absent in the southern half of Iran. Persians, Kurds, Azeri (formerly) and Gilaks are/were Iranian speakers....
We can shake off some of that as potential noise but when you add in the Grugni finding that R1b-M269* showing up at 15.4% among the Zoroasterian population, it encouraged me to think again about the break up early IE languages and how they might help us understand R1b. It is true that the linguistic tree and arguments are a cause in themselves but they ultimately do tie back to people movements and exchanges, which may be instructive for understanding R1b.

The red lines/writing are mine but Warnow's analysis does not consider Armenian as a Satem language.
https://dl.dropboxusercontent.com/u/17907527/IE-Language-Tree_Reconstruction_by_Warnow_2013.jpg

There is a some disagreement about all of this, but it appears Greek and Armenian lineages broke late from the PIE homeland/vicinity in comparison to Italo-Celtic and Germanic.

Wikipedia excerpts:
"The hypothetical Proto-Graeco-Armenian stage would need to date to the 3rd millennium BC, only barely differentiated from either late PIE or Graeco-Armeno-Aryan.
...
Proto-Greek would have been spoken in the late 3rd millennium BC, most probably in the Balkans
...
Close similarities between Ancient Greek and Vedic Sanskrit suggest that both Proto-Greek and Proto-Indo-Iranian were still quite similar to either late Proto-Indo-European, which would place the latter somewhere in the late 4th millennium BC
...
Greek is a Centum language, which would place a possible Graeco-Aryan protolanguage before Satemization, making it identical to late PIE. Proto-Greek does appear to have been affected by the general trend of palatalization characteristic of the Satem group, evidenced for
...
Armenian shares a number of major innovations with Greek, and some linguists group these two languages together with Phrygian and the Indo-Iranian family into a higher-level subgroup of Indo-European which is defined by such shared innovations as the augment. More recently, others have proposed a Balkan grouping including Greek, Armenian, Phrygian and Albanian.
...
The large percentage of loans from Iranian languages initially led linguists to erroneously classify Armenian as an Iranian language. The distinctness of Armenian was only recognized when Hübschmann (1875)[7] used the comparative method to distinguish two layers of Iranian loans from the older Armenian vocabulary.
...
In his paper, "Hurro-Urartian Borrowings in Old Armenian", Soviet linguist Igor Mikhailovich Diakonov notes the presence in Old Armenian of what he calls a Caucasian substratum, identified by earlier scholars, consisting of loans from the Kartvelian and Northeast Caucasian languages such as Udi. Noting that the Hurro-Urartian peoples inhabited the Armenian homeland in the second millennium BC, Diakonov identifies in Armenian a Hurro-Urartian substratum of social, cultural, and zoological and biological terms such as ałaxin ('slavegirl') and xnjor ('apple(tree)'). Some of the terms he gives admittedly have an Akkadian or Sumerian provenance, but he suggests they were borrowed through Hurrian or Urartu. Given that these borrowings do not undergo sound changes characteristic of the development of Armenian from Proto-Indo-European, he dates their borrowing to a time before the written record but after the Proto-Armenian language stage."

Could the proposed Hellenic (pre-Greek/Armenian) speakers carried some R1b-L23xL51 due south into Greece as well as southeast/east into Anatolia?

However, did this whole group break away just as the Satemization process was getting started. If so that might tell us a little more about where L23xL51 came from?

Also, back to the M269* in Iran... we also see this due north (Rocca's map) just northeast of the Caspian. This should be the territory where Satemization was taking place, right?


P.S. In the U.S. we might call this a counter-clockwise movement. Is that not a common word in the Isles?

Rathna
06-11-2013, 06:17 AM
"Could the proposed Hellenic (pre-Greek/Armenian) speakers carried some R1b-L23xL51 due south into Greece as well as southeast/east into Anatolia?"

And where they could have come from if not from the Balkans?

"However, did this whole group break away just as the Satemization process was getting started. If so that might tell us a little more about where L23xL51 came from?"

I think that probably you ignore the basic concepts of linguistic: an innovation (and there is no doubt that satem is an innovation as to centum) reaches the region all around it, till it may expand, and there we have no need of a separation, but this process may have happened already in the IE unity also if we take in consideration that samples of satem-outcomes we have also in non satem languages and it seems that these languages were expanding from West to East and not the other way around. Everything carries us to central Europe (and you know that for me before to the Alpine zone/Italy).

"Also, back to the M269* in Iran... we also see this due north (Rocca's map) just northeast of the Caspian. This should be the territory where Satemization was taking place, right?"

Indo-Europeans arrive there late, and were, as to hg-R, above all R1a etc. rather than R1b. We haven't found R1b in ancient DNA from central Asia.

Mikewww
06-11-2013, 02:23 PM
I think that probably you ignore the basic concepts of linguistic: an innovation (and there is no doubt that satem is an innovation as to centum) reaches the region all around it, till it may expand, and there we have no need of a separation, but this process may have happened already in the IE unity
I don't understand why you think I'm ignoring the concept of an innovation. That is what I relied upon to suggest the Hellenic linguistic predecessors broke late from the area where Satemization was beginning to occur but before it was completed. Anyway, we probably just see things a little differently. Nothing wrong with that.


Everything carries us to central Europe
I think you are actually saying carries us to a Central Europe origin. I'm not seeing that but I guess I never will until I understand the evidence. I do follow your Rathna's assessment of the early R1b subclades. Let's discuss that over there.


Indo-Europeans arrive there late, and were, as to hg-R, above all R1a etc. rather than R1b. We haven't found R1b in ancient DNA from central Asia.
I am not trying to equate the haplogroups 1 for 1 with the languages. I agree that there were probably R1a people that reached Central Asia that were speaking IE languages. Likewise, I'm not trying to say all R1b always spoke IE languages. We just know a lot do now and/or in the historical timeframe.
As far as ancient DNA goes, we've found very little for R1b, period... must have had a late expansion and/or it was exterminated to some degree in its original haunts.

Rathna
06-11-2013, 05:12 PM
I alluded to the theory of the "aree laterali" of Bartoli. The Satemization seems just happened in a "central area", seen that we have centum languages in the "aree laterali" (Western Indo-European and Tocharian), but Albanian should say many things to us and it probably comes from somewhere in the Balkans:
1) it conserves the laryngeals (hurdhe, herdhe etc.)
2) it has a phenomenon linked to the Satemization: th from ky and dh from gy etc.
and Albanian survived by chance. Probably many languages went extinct, but the "Stele of Novilara" may say so much about the origin of the Italic-Celtic branch. And we always are between Italy and the Balkans.

newtoboard
06-11-2013, 06:01 PM
I alluded to the theory of the "aree laterali" of Bartoli. The Satemization seems just happened in a "central area", seen that we have centum languages in the "aree laterali" (Western Indo-European and Tocharian), but Albanian should say many things to us and it probably comes from somewhere in the Balkans:
1) it conserves the laryngeals (hurdhe, herdhe etc.)
2) it has a phenomenon linked to the Satemization: th from ky and dh from gy etc.
and Albanian survived by chance. Probably many languages went extinct, but the "Stele of Novilara" may say so much about the origin of the Italic-Celtic branch. And we always are between Italy and the Balkans.


Tocharians ended up in the east but that is obviously not where they begin. Satemization likely occurred at the SE edges. There's plenty of evidence that only Indo-Iranian is fully satemized and Balto-Slavic is not. And it also wouldn't surprise me if the Satemization of Albanian and Armenian occurred due to Scythians and Persians respectively.

The dark red area is the most likely place of origin.

http://upload.wikimedia.org/wikipedia/commons/a/ae/Centum_Satem_map.png

http://en.wikipedia.org/wiki/Centum-satem_isogloss

newtoboard
06-11-2013, 06:09 PM
"Could the proposed Hellenic (pre-Greek/Armenian) speakers carried some R1b-L23xL51 due south into Greece as well as southeast/east into Anatolia?"

And where they could have come from if not from the Balkans?

"However, did this whole group break away just as the Satemization process was getting started. If so that might tell us a little more about where L23xL51 came from?"

I think that probably you ignore the basic concepts of linguistic: an innovation (and there is no doubt that satem is an innovation as to centum) reaches the region all around it, till it may expand, and there we have no need of a separation, but this process may have happened already in the IE unity also if we take in consideration that samples of satem-outcomes we have also in non satem languages and it seems that these languages were expanding from West to East and not the other way around. Everything carries us to central Europe (and you know that for me before to the Alpine zone/Italy).

"Also, back to the M269* in Iran... we also see this due north (Rocca's map) just northeast of the Caspian. This should be the territory where Satemization was taking place, right?"

Indo-Europeans arrive there late, and were, as to hg-R, above all R1a etc. rather than R1b. We haven't found R1b in ancient DNA from central Asia.

Of course Satemization could also just be an influence on Balto-Slavic, Albanian and Armenian. At a certain point the movements of all steepe Iranians (and these movements would have brought Satem speakers into contact with Balto-Slavs and Albanians) and settled Iranians speakers was East to West(bring Persians and Medes into contact with Armenians) and not the other way around(which actually continued deep into in the Turkic period).

alan
06-11-2013, 09:28 PM
We can shake off some of that as potential noise but when you add in the Grugni finding that R1b-M269* showing up at 15.4% among the Zoroasterian population, it encouraged me to think again about the break up early IE languages and how they might help us understand R1b. It is true that the linguistic tree and arguments are a cause in themselves but they ultimately do tie back to people movements and exchanges, which may be instructive for understanding R1b.

The red lines/writing are mine but Warnow's analysis does not consider Armenian as a Satem language.
https://dl.dropboxusercontent.com/u/17907527/IE-Language-Tree_Reconstruction_by_Warnow_2013.jpg

There is a some disagreement about all of this, but it appears Greek and Armenian lineages broke late from the PIE homeland/vicinity in comparison to Italo-Celtic and Germanic.

Wikipedia excerpts:
"The hypothetical Proto-Graeco-Armenian stage would need to date to the 3rd millennium BC, only barely differentiated from either late PIE or Graeco-Armeno-Aryan.
...
Proto-Greek would have been spoken in the late 3rd millennium BC, most probably in the Balkans
...
Close similarities between Ancient Greek and Vedic Sanskrit suggest that both Proto-Greek and Proto-Indo-Iranian were still quite similar to either late Proto-Indo-European, which would place the latter somewhere in the late 4th millennium BC
...
Greek is a Centum language, which would place a possible Graeco-Aryan protolanguage before Satemization, making it identical to late PIE. Proto-Greek does appear to have been affected by the general trend of palatalization characteristic of the Satem group, evidenced for
...
Armenian shares a number of major innovations with Greek, and some linguists group these two languages together with Phrygian and the Indo-Iranian family into a higher-level subgroup of Indo-European which is defined by such shared innovations as the augment. More recently, others have proposed a Balkan grouping including Greek, Armenian, Phrygian and Albanian.
...
The large percentage of loans from Iranian languages initially led linguists to erroneously classify Armenian as an Iranian language. The distinctness of Armenian was only recognized when Hübschmann (1875)[7] used the comparative method to distinguish two layers of Iranian loans from the older Armenian vocabulary.
...
In his paper, "Hurro-Urartian Borrowings in Old Armenian", Soviet linguist Igor Mikhailovich Diakonov notes the presence in Old Armenian of what he calls a Caucasian substratum, identified by earlier scholars, consisting of loans from the Kartvelian and Northeast Caucasian languages such as Udi. Noting that the Hurro-Urartian peoples inhabited the Armenian homeland in the second millennium BC, Diakonov identifies in Armenian a Hurro-Urartian substratum of social, cultural, and zoological and biological terms such as ałaxin ('slavegirl') and xnjor ('apple(tree)'). Some of the terms he gives admittedly have an Akkadian or Sumerian provenance, but he suggests they were borrowed through Hurrian or Urartu. Given that these borrowings do not undergo sound changes characteristic of the development of Armenian from Proto-Indo-European, he dates their borrowing to a time before the written record but after the Proto-Armenian language stage."

Could the proposed Hellenic (pre-Greek/Armenian) speakers carried some R1b-L23xL51 due south into Greece as well as southeast/east into Anatolia?

However, did this whole group break away just as the Satemization process was getting started. If so that might tell us a little more about where L23xL51 came from?

Also, back to the M269* in Iran... we also see this due north (Rocca's map) just northeast of the Caspian. This should be the territory where Satemization was taking place, right?


P.S. In the U.S. we might call this a counter-clockwise movement. Is that not a common word in the Isles?

Both terms would be well understood but anti-clockwise would seem to me more common. Probably depends on your age too!

alan
06-11-2013, 09:41 PM
Tocharians ended up in the east but that is obviously not where they begin. Satemization likely occurred at the SE edges. There's plenty of evidence that only Indo-Iranian is fully satemized and Balto-Slavic is not. And it also wouldn't surprise me if the Satemization of Albanian and Armenian occurred due to Scythians and Persians respectively.

The dark red area is the most likely place of origin.

http://upload.wikimedia.org/wikipedia/commons/a/ae/Centum_Satem_map.png

http://en.wikipedia.org/wiki/Centum-satem_isogloss

I think that you could be right. Its similar to the way old books have overegged the P-Q split in Celtic not realising it probably spread in a large part through aerial means between elites without hardly any population movement being required and doesnt really represent any real division among Celts.

Mikewww
06-11-2013, 11:25 PM
Tocharians ended up in the east but that is obviously not where they begin. Satemization likely occurred at the SE edges. There's plenty of evidence that only Indo-Iranian is fully satemized and Balto-Slavic is not. And it also wouldn't surprise me if the Satemization of Albanian and Armenian occurred due to Scythians and Persians respectively.
http://upload.wikimedia.org/wikipedia/commons/a/ae/Centum_Satem_map.png This map may be wrong!

Perhaps we should consider Satemization in phases, which could be very instructive if considered in context of the expected area and timing of Satemization.

First, it is not clear that Armenian is full Satem and some classify it as Centum. Likewise Greek is generally classified as Centum but there are systems that claim it is Satem. As far as the Wikipedia map goes in may mean that some of the shades are not concretely red or blue but more purplish shades of blue or red.

I was not aware of the point that Balto-Slavic languages did not have the full Satem set of innovations that Indo-Iranian languages do.

At the same time, the latest and greated computational cladistics analyses has Greek and Armenian on the same branch more closely related to each other than to Balto-Slavic, Indo-Iranian, Germanic and Italo-Celtic branches.

So we could have
1) pure Centum languages like the pre-Germanic and Italo-Celtic branches,
2) partially Satemized/Centum languages like the pre-Greek and Armenian languages that some call the Hellenic or the Balkans branch,
3) early Satem languages like the Balto-Slavic branch, and
4) full Satem languages like the Indo-Iranian branch.

If we have some decent idea of where and when the Satem innovations occurred that is great news.

The #1 grouping of pure Centum languages either broke away early, before Satem started or on the outskirts of the PIE homeland where Satem innovations had not yet reached. I would guess the Italo-Celtic and pre-Germanic branches broke away from the western edge of PIE, and perhaps early. I would guess the Anatolian branch broke away very early since it has some pre-PIE characteristics. This means it could have more easily been close to the core PIE homeland.

#2 partially Satemized/Centum languages must have been in interaction with the zone where Satemization was growing. This implies they broke away later, at least compared to Anatolian and quite possibly when compared to italo-Celtic and pre-Germanic. This probably supports Alan's counter-clockwise Circumpontic movements. In the case of Armenian, the implication is close to a full circle.

#3 types of early Satem languages like the Balto-Slavic must have broken away later than #2 or #1 types, or at least their break-away point was NOT from the edge of PIE but they broke away very close to the core, where Satemization was occurring. I know that later, there was supposed to be some west to east movements of these languages back towards the core so I don't know how that works out.

#4 of course the full Satem languages would have appeared to have broken late (later than #1 and #2 anyway) from the core area of Satemization and obviously in a different direction.

alan
06-12-2013, 12:28 PM
I wonder if satemisation at its original core was a substrate effect. Then its influence spread by becoming an adstrate in other areas. I think the IE trees show that, like the P-Q split in Celtic it may be a later effect and not originally a feature of most IE branches. As someone noted above, a lot of it could be down to the massive amount of Iranian impact in Eurasia at different times. Its easier to see that group as being influential in satemisation more than others because its about as close to a common denomenator as we get in the Satem areas. That of course raised the question of the origins of Iranian. I vaguely recall Sintashta and/or some sort of meeting of Afanasievo with Andronovo being involved.

alan
06-12-2013, 01:18 PM
This map may be wrong!

Perhaps we should consider Satemization in phases, which could be very instructive if considered in context of the expected area and timing of Satemization.

First, it is not clear that Armenian is full Satem and some classify it as Centum. Likewise Greek is generally classified as Centum but there are systems that claim it is Satem. As far as the Wikipedia map goes in may mean that some of the shades are not concretely red or blue but more purplish shades of blue or red.

I was not aware of the point that Balto-Slavic languages did not have the full Satem set of innovations that Indo-Iranian languages do.

At the same time, the latest and greated computational cladistics analyses has Greek and Armenian on the same branch more closely related to each other than to Balto-Slavic, Indo-Iranian, Germanic and Italo-Celtic branches.

So we could have
1) pure Centum languages like the pre-Germanic and Italo-Celtic branches,
2) partially Satemized/Centum languages like the pre-Greek, Albanian, Armenian languages that some call the Hellenic or the Balkans branch,
3) early Satem languages like the Balto-Slavic branch, and
4) full Satem languages like the Indo-Iranian branch.

If we have some decent idea of where and when the Satem innovations occurred that is great news.

The #1 grouping of pure Centum languages either broke away early, before Satem started or on the outskirts of the PIE homeland where Satem innovations had not yet reached. I would guess the Italo-Celtic and pre-Germanic branches broke away from the western edge of PIE, and perhaps early. I would guess the Anatolian branch broke away very early since it has some pre-PIE characteristics. This means it could have more easily been close to the core PIE homeland.

#2 partially Satemized/Centum languages must have been in interaction with the zone where Satemization was growing. This implies they broke away later, at least compared to Anatolian and quite possibly when compared to italo-Celtic and pre-Germanic. This probably supports Alan's counter-clockwise Circumpontic movements. In the case of Armenian, the implication is close to a full circle.

#3 types of early Satem languages like the Balto-Slavic must have broken away later than #2 or #1 types, or at least their break-away point was NOT from the edge of PIE but they broke away very close to the core, where Satemization was occurring. I know that later, there was supposed to be some west to east movements of these languages back towards the core so I don't know how that works out.

#4 of course the full Satem languages would have appeared to have broken late (later than #1 and #2 anyway) from the core area of Satemization and obviously in a different direction.

The fact that Balkans group IE languages (if one believes is such a group - I tend to) saw only partial satemisation is interesting when it is considered that most theories place this group in the eastern half of the Balkans in much of the period 3000-2000BC. It at least shows that the only partial satemisation of this group may have been very late. After all the Greeks are centum but are considered to have only moved from the Balkans into Greece fairly late after 2000BC in several waves. Albanian is also believed by many linguists to have arrived on the Adriatic from the east Balkans rather late. In contrast the satem Armenians are believed to have remained in the Balkans until somewhat later. That suggests to me that part satemisation of the Balkans took place rather late in the day in the late 2nd or even 1st millenium BC. The influence of Iranian speakers seems a likely source to me. They touched both the Balkans and passed through the Caucasus from before 700BC sort of pincer style. So the satemisation of Armenian could have taken place in either place before or after their alleged migration. I am maybe being silly here but is there any actual evidence of satem before this other than in Iranian? Iranian speakers had an incredible impact on eastern Europe and SW Asia, far wider than any other group in prehistory (the Celts probably were the next most prolific).

lgmayka
06-12-2013, 03:43 PM
Both terms would be well understood but anti-clockwise would seem to me more common. Probably depends on your age too!
No, it is a difference between British and American English (https://en.wikipedia.org/wiki/Clockwise):
---
The opposite sense of rotation or revolution is (in Commonwealth English) anticlockwise (ACW), or (in North American English) counterclockwise (CCW).
---

alan
06-12-2013, 04:09 PM
No, it is a difference between British and American English (https://en.wikipedia.org/wiki/Clockwise):
---
The opposite sense of rotation or revolution is (in Commonwealth English) anticlockwise (ACW), or (in North American English) counterclockwise (CCW).
---

Oh I realise that. Its just that American English has been having a major impact on younger people's British English for decades and never more so than in the internet age. So the American terms will also be used by some people. British English is heading towards being a hybrid and even American English spelling are common to see because people learn them on net and through using the wrong spellchecker.

Mikewww
06-15-2013, 05:13 PM
Perhaps we should consider Satemization in phases, which could be very instructive if considered in context of the expected area and timing of Satemization.

First, it is not clear that Armenian is full Satem and some classify it as Centum. Likewise Greek is generally classified as Centum but there are systems that claim it is Satem. As far as the Wikipedia map goes in may mean that some of the shades are not concretely red or blue but more purplish shades of blue or red.

I was not aware of the point that Balto-Slavic languages did not have the full Satem set of innovations that Indo-Iranian languages do.

At the same time, the latest and greated computational cladistics analyses has Greek and Armenian on the same branch more closely related to each other than to Balto-Slavic, Indo-Iranian, Germanic and Italo-Celtic branches.

So we could have
1) pure Centum languages like the pre-Germanic and Italo-Celtic branches,
2) partially Satemized/Centum languages like the pre-Greek and Armenian languages that some call the Hellenic or the Balkans branch,
3) early Satem languages like the Balto-Slavic branch, and
4) full Satem languages like the Indo-Iranian branch.
....

I'm in an IE discussion linguistic group, most of which is way too in depth for me, but I asked what they thought of the relationships of Armenian and Greek.

Is there a good case that Armenian and Greek predecessors are of the same branch?

Here is the first answer. I'll let you know of if there is disagreement on this stuff.
Richard Wordingham answered,

"There are several signs of mutual influence between Indo-Iranian, Armenian and Greek. While it is quite possible that Greek and Armenian were of a common branch, the evidence is not very strong. Much IE subgrouping evidence is weak, suggesting rapid fragmentation into multiple groups. Gray & Atkinson quoted a 'Bayesian posterior probability of 40%' for the grouping, which is not overwhelming."

Gray and Atkinson are in the Anatolian origin camp. Warnow was the one putting them together, but we'll see if can get other opinions and logic.

The instructive thing that I got out of this response was the "rapid fragmentation" of IE branching. Earlier on another thread, Jean M mentioned that the IE branching of Italo-Celtic and Germanic demonstrate an early split with a period of long development in "isolation."

If we try to map languages and R1b subclades (not necessarily 1 for 1 though) to see if correlations can help us fit the puzzle together, I think the initial question is - Where and when did the various subclades pick up their primary respective IE dialects?

To me, it is hard to think that U106 picked up pre-Germanic or Germanic, P312 elements Italic, other P312 elements Celtic, L23xL51/Z2103 elements Armenian, etc., etc. all in situ close to where they are now. That seems amazingly coincidental, particularly since much of the old Celtic lands are very, very high in P312 while devoid R1a.

If R1b picked up IE before these major breakaways, it would seem L11* was there at that location and time as U106 (or pre-U106) went with the pre-Germanic branch while P312 mostly went another way.

alan
06-15-2013, 07:04 PM
I'm in an IE discussion linguistic group, most of which is way too in depth for me, but I asked what they thought of the relationships of Armenian and Greek.


Here is the first answer. I'll let you know of if there is disagreement on this stuff.
Richard Wordingham answered,

"There are several signs of mutual influence between Indo-Iranian, Armenian and Greek. While it is quite possible that Greek and Armenian were of a common branch, the evidence is not very strong. Much IE subgrouping evidence is weak, suggesting rapid fragmentation into multiple groups. Gray & Atkinson quoted a 'Bayesian posterior probability of 40%' for the grouping, which is not overwhelming."

Gray and Atkinson are in the Anatolian origin camp. Warnow was the one putting them together, but we'll see if can get other opinions and logic.

The instructive thing that I got out of this response was the "rapid fragmentation" of IE branching. Earlier on another thread, Jean M mentioned that the IE branching of Italo-Celtic and Germanic demonstrate an early split with a period of long development in "isolation."

If we try to map languages and R1b subclades (not necessarily 1 for 1 though) to see if correlations can help us fit the puzzle together, I think the initial question is - Where and when did the various subclades pick up their primary respective IE dialects?

To me, it is hard to think that U106 picked up pre-Germanic or Germanic, P312 elements Italic, other P312 elements Celtic, L23xL51/Z2103 elements Armenian, etc., etc. all in situ close to where they are now. That seems amazingly coincidental, particularly since much of the old Celtic lands are very, very high in P312 while devoid R1a.

If R1b picked up IE before these major breakaways, it would seem L11* was there at that location and time as U106 (or pre-U106) went with the pre-Germanic branch while P312 mostly went another way.

I see it most likely that M269* and L23* was involved in the Anatolian and also the PIE linguistic and especially cultural formative stages of the latter and that it spilled from there into a number of lineages in different directions. Most L23* lineages seem to have spilled a shorter distance via Ukraine into the western shore of the Black Sea and from there through the Balkans and ultimately Anatolia. I think this is most likely linked to both Anatolian and the hypothetical Balkans group in different waves. There were several waves spilling into that area from 4200-3000BC so there is plenty of scope. So I would link the non-L51 L23 clades with languages like Hittite, Luwain, Albanian, Greek and Armenian and I would think they had a significant period in the east Balkans before they got their final positions. I am by no means claiming exclusivity to R1b on this. It could well have been a mixed group including R1a too.

One branch of true L23* or early L51* may have made (from somewhere along the north Black Sea shore) a rapid jump along the Danube using that river c. 3500BC. That is significantly further west than others. That to me would be the arrival of Italo-Celtic. I get the impression that a far flung offshoot of CMP (possibly transferred by a steppe-Maykop rooted group) took place.

I am a believer that M73 had a role in the spread east with Afanasievo and Tocharian. This is dated to about 3500BC.

I never go near the whole pre-Germanic subject because it is too uncertain and obscure. I think its a hybrid between elements that arrived from the east via corded ware and elements that arrived in the beaker area from the west. I think that resptively consisted of R1a and L11*.

As for Slavic, Baltic and Indo-Iranian I think based on Anthony's chain of west to east moving cultures (middle Dnieper-Fatyanovo-Abashevo-Sintashta) c. 3200BC onwards that they were linked to a forrest steppe group with steppic influence and were probably spread by the complex mixing and arising of new cultures in the Carpathian area. I think its pretty clear that R1a relates well to the satem group.

So, in terms of a rapid breakup, I think that most non-Anatolian dialects can be seen as moving away from each other around 3500-3200BC.

I would no longer worry about this sort of chronology and Yamnaya. It is clear to me that in the last few years the start of a retreat from the emphasis on that one culture has begun.

newtoboard
06-15-2013, 07:08 PM
I wonder if satemisation at its original core was a substrate effect. Then its influence spread by becoming an adstrate in other areas. I think the IE trees show that, like the P-Q split in Celtic it may be a later effect and not originally a feature of most IE branches. As someone noted above, a lot of it could be down to the massive amount of Iranian impact in Eurasia at different times. Its easier to see that group as being influential in satemisation more than others because its about as close to a common denomenator as we get in the Satem areas. That of course raised the question of the origins of Iranian. I vaguely recall Sintashta and/or some sort of meeting of Afanasievo with Andronovo being involved.

I've considered the possibility that that Satemization occurred when Andronovo met with the BMAC. It quickly backmigrated to the steepe and then pushed west. Remember all Iranian (and likely Iranian nomad) steepe movements were east to west. The Cimmerians move west due to Scythians who are then pushed west by Sarmatians.

alan
06-15-2013, 07:13 PM
I've considered the possibility that that Satemization occurred when Andronovo met with the BMAC. It quickly backmigrated to the steepe and then pushed west. Remember all Iranian (and likely Iranian nomad) steepe movements were east to west. The Cimmerians move west due to Scythians who are then pushed west by Sarmatians.

I think its a strong possibility. I am surprised I have never read anything on the likely linguistic cause of satem in terms of substrate, something along the lines of the new idea that the Q-P change in Celtic occurred under Etruscan influence. It is a deviation from the original after all.

alan
06-15-2013, 07:23 PM
I have to say that the more I have read in recent papers the less I think we need to worry about Yamnaya so specifically. It was just part of a turmoil of mixing and movements of peoples from in and bordering the steppe in this period. It is also clearly secondary to the Maykop influences that effected a much wider, earlier and more westerly steppe groups than Yamnaya. I think it was part of this but it wasnt the whole thing. I also dont believe that PIE developed out of Yamnaya specifically. I think they need to look a little further south and west to the Caucasus-Ukraine interface area a few centuries earlier to see the origins of PIE. It was those slighly earlier groups that introduced much of what we think of as Kurgan culture. Yamnanya was a recipient of this not the originator. Yamnaya just looks like a slightly late more easterly impulse west but by the time it arrived I think some of the languages had already headed west c. 4000-3500BC.

newtoboard
06-15-2013, 07:29 PM
I see it most likely that M269* and L23* was involved in the Anatolian and also the PIE linguistic and especially cultural formative stages of the latter and that it spilled from there into a number of lineages in different directions. Most L23* lineages seem to have spilled a shorter distance via Ukraine into the western shore of the Black Sea and from there through the Balkans and ultimately Anatolia. I think this is most likely linked to both Anatolian and the hypothetical Balkans group in different waves. There were several waves spilling into that area from 4200-3000BC so there is plenty of scope. So I would link the non-L51 L23 clades with languages like Hittite, Luwain, Albanian, Greek and Armenian and I would think they had a significant period in the east Balkans before they got their final positions. I am by no means claiming exclusivity to R1b on this. It could well have been a mixed group including R1a too.

One branch of true L23* or early L51* may have made (from somewhere along the north Black Sea shore) a rapid jump along the Danube using that river c. 3500BC. That is significantly further west than others. That to me would be the arrival of Italo-Celtic. I get the impression that a far flung offshoot of CMP (possibly transferred by a steppe-Maykop rooted group) took place.

I am a believer that M73 had a role in the spread east with Afanasievo and Tocharian. This is dated to about 3500BC.

I never go near the whole pre-Germanic subject because it is too uncertain and obscure. I think its a hybrid between elements that arrived from the east via corded ware and elements that arrived in the beaker area from the west. I think that resptively consisted of R1a and L11*.

As for Slavic, Baltic and Indo-Iranian I think based on Anthony's chain of west to east moving cultures (middle Dnieper-Fatyanovo-Abashevo-Sintashta) c. 3200BC onwards that they were linked to a forrest steppe group with steppic influence and were probably spread by the complex mixing and arising of new cultures in the Carpathian area. I think its pretty clear that R1a relates well to the satem group.

So, in terms of a rapid breakup, I think that most non-Anatolian dialects can be seen as moving away from each other around 3500-3200BC.

I would no longer worry about this sort of chronology and Yamnaya. It is clear to me that in the last few years the start of a retreat from the emphasis on that one culture has begun.

Once again you are overstating the value of Satemization. There is nothing special about Indo-Iranian and Balto-Slavic, not linguistically and not in terms of genetics. It is just geographical luck that Balto-Slavic picked up Satemization. Also ignoring the concept of NW Indo-European which is what Corded Ware was likely speaking. Interesting that Centum Germanic R1a is predominantly Z284+ which is downstream of Z283+ while Indo-Iranian R1a is not downstream of Z283+.

The Z93+ and Z283+ split is likely older than that chain of movements. And Abashevo's Indo-Iranian identity is often seen as an intrusive element from the Poltavaka culture.



I have to say that the more I have read in recent papers the less I think we need to worry about Yamnaya so specifically. It was just part of a turmoil of mixing and movements of peoples from in and bordering the steppe in this period. It is also clearly secondary to the Maykop influences that effected a much wider, earlier and more westerly steppe groups than Yamnaya. I think it was part of this but it wasnt the whole thing. I also dont believe that PIE developed out of Yamnaya specifically. I think they need to look a little further south and west to the Caucasus-Ukraine interface area a few centuries earlier to see the origins of PIE. It was those slighly earlier groups that introduced much of what we think of as Kurgan culture. Yamnanya was a recipient of this not the originator. Yamnaya just looks like a slightly late more easterly impulse west but by the time it arrived I think some of the languages had already headed west c. 4000-3500BC.


Its more likely the PIE homeland is located where the Samara culture was. This would account for the close links between Proto Indo-European and Proto Uralic as well.

alan
06-15-2013, 08:37 PM
Once again you are overstating the value of Satemization. There is nothing special about Indo-Iranian and Balto-Slavic, not linguistically and not in terms of genetics. It is just geographical luck that Balto-Slavic picked up Satemization. Also ignoring the concept of NW Indo-European which is what Corded Ware was likely speaking. Interesting that Centum Germanic R1a is predominantly Z284+ which is downstream of Z283+ while Indo-Iranian R1a is not downstream of Z283+.

The Z93+ and Z283+ split is likely older than that chain of movements. And Abashevo's Indo-Iranian identity is often seen as an intrusive element from the Poltavaka culture.





Its more likely the PIE homeland is located where the Samara culture was. This would account for the close links between Proto Indo-European and Proto Uralic as well.

I thought Abashevo was more linked to Baltic. I was simply following Anthony who made that link. I actually do agree with you about satemisation.

Regaring the NW IE, I dont know. Its one of several concepts. The corded ware link works best for satem languages (I am simply using that as a convenient terms and not implying it is significant -I agree it is probably an adstrate in many areas). The exception is Germanic.

I suspect the corded ware zone that the later proto-Germanic language shared with the later Baltic and Slavic language zones is part of the makeup of Germanic but I have often heard that pre-Germanic is never going to be fully understood and is probably a complicated mix of different IE influences from east and west even from its earliest days. After all we know both some form of M269XU106 (IMO probably L11* or P312) and also some form of R1a were both present in central Germany c. 2600BC. These of course were associated with bell beaker and corded ware respectively, both at that time intrusive newly arrived cultures who had originated to the south-west and in south Poland respectively a few centuries earlier. The sample is insanely small but it fits the links between the haplogroups and cultures (although we know Corded Ware also included Neolithic pre-R1a lines now).

I think Germany was basically at the edges between west and eastern influences from the start of copper age R1 groups moving in and this pattern carried on for a long period. From my reading of archaeology the corded ware influences (a culture now dated to have began in south Poland) pre-dated beaker in this area but only by perhaps a couple of centuries (the earliest couple of centuries of radiocarbon dates for Corded Ware have been brought into doubt). I imagine that Germany experienced an initial period of close linguistic links with areas to the east for a couple of centuries during corded ware before the prevailing flow of influence shifted to a western one for a very very long time through the beaker and early Bronze Age periods. The part of Germanic that relates to the Celto-Italic branch clearly relates to western influences which were may have been ongoing from the beaker period right through to the Iron Age. I think that early couple of centuries linked to the east by corded ware followed by a long switch to the dominantion of western influences made pre-Germanic an oddball language. I think the tree model of language fission is of limited use when a language experiences very early shift in direction of influence like pre-Germanic would likely have done judging by the archaeological record.

alan
06-15-2013, 08:58 PM
Once again you are overstating the value of Satemization. There is nothing special about Indo-Iranian and Balto-Slavic, not linguistically and not in terms of genetics. It is just geographical luck that Balto-Slavic picked up Satemization. Also ignoring the concept of NW Indo-European which is what Corded Ware was likely speaking. Interesting that Centum Germanic R1a is predominantly Z284+ which is downstream of Z283+ while Indo-Iranian R1a is not downstream of Z283+.

The Z93+ and Z283+ split is likely older than that chain of movements. And Abashevo's Indo-Iranian identity is often seen as an intrusive element from the Poltavaka culture.





Its more likely the PIE homeland is located where the Samara culture was. This would account for the close links between Proto Indo-European and Proto Uralic as well.

Noone really knows but I have doubts that we really know the distribution of Uralic languages in 4000BC. My main quiery about this is that more than a simple east-west division along the Ural sort of area, I would rather see some sort of ecological division as corresponding with the linguistic one, possibly along the lines of the forrest-forrest steppe-steppe zones. Now a recent article I read makes it clear that these divisions were highly unstable with the zones fluctuating greatly through the Neolithic and copper age as dry and humid phases came and went. The paper essentially indicates that people tended to move as their environment moved rather than radically adapt. So, I retain a suspicion that Uralic and perhaps PIE's divided within Europe may have originally been a north-south one and that that division may have stretch well west of the urals. I dont know what that means for PIE. There is a theory that PIE is essentially a Uralic western steppe language that hybrided with a NW Caucasian one. If that is the case PIE would probably have come into existance in the north Caucasus-steppe interface over a period of perhaps 1000 years or so of interaction c. 5000-4000BC.