David Reich speech on steppe migrations-April 29, 2017

Indo-European lineages were either R1a or R1b, but if you remove Slavic, Germanic, Celtic and Roman R1a and R1b in Greece, there is not much left. Mostly R1b-Z2103 and R1a-Z93, but that's just a few percents of the population. If the Dorians were R1b -Z2103 from the Balkans, then all that is left for the Mycenaeans is R1a. However R1a-Z93 probably came from the Near East.

there may have been a 3rd IE clade, apart from R1a and R1b, certain sublades of I2a2a
not for Myceneans, but for the expansion toward the Carpathian Basin from the steppe
 
Concerning that tree which as far as I understand is based on Chang et al computational model. Which can create exotic branchings.
Any tree that place the split between Indo Aryan and Iranian after 2000 BC is wrong wrong wrong.
You can have dozens of ancient dna from Sintashta and Andronovo You will not find the Indian L657.

Because it was not there and it made a leap frog directly from maybe Abashevo to Indus Valley.

Albanian being a Germanic language is also very funny.

I am surprised with the new classification of Albanian as Germanic language, but its not really funny. Albanian original words, not the borrowed one, contain a number of German words (but not too many) that can not be explained with Gothic or Vandal invasions. Also 2 Austrian linguists published a work saying that Albanian is the base of all European languages including Germanic, and Albanian language is not related to Illyrian. Albanians were annoyed to their suggestion of Albanian not related to Illyrian since there is evidence of relation and ignored their findings. To make my point, even if Albanian is not Germanic signs of some kind of relation are there. But the question is who brought that language to Albanians since genetically Albanians are Balkanic and R1b Germanic is not common among Albanians.
 
Milan.M

You need to read that Cyprus paper. Cyprus was Hellenized by E-V13, R1b, and J2-M67. No R1a and I2a1 which both are recent Slavic settlers in Greece during Byzantine epoch.
yes you are right about that, but a simple explanation could be that when cyprus was hellenized the myceneans were already settled in greece for centuries so even if the original myceneans were predominately R1a the invading armies of cyprus had all greek haplotypes in the normal percentages.Then if you consider the fact that this army was mingled with the locals then ydna differences are to be aspected anyway.
 
I think that this is just a problem of sample bias. There were lots of people living in the Balkans from 4200 to 3000 BCE. How many did they test? Ten? One hundred? Was it from different locations and different periods? If they tested over one hundred from at least 10 different locations and periods, including cultures that archaeologists thought had Steppe influence like Cernavodă, Ezero, Glina or Bubanj-Hum, then it would be rather surprising. But if they took samples from cultures still belonging to Old Europe like Boian or Karanovo, then there is nothing odd about it. Even in the former case, there was surely a strong segregation between the Steppe invaders and the indigenous population, so not finding any Steppe ancestry could just mean that they got samples belonging to unmixed indigenous people in the conquered population.

he stated over 2000 tested
 
I am surprised with the new classification of Albanian as Germanic language, but its not really funny. Albanian original words, not the borrowed one, contain a number of German words (but not too many) that can not be explained with Gothic or Vandal invasions. Also 2 Austrian linguists published a work saying that Albanian is the base of all European languages including Germanic, and Albanian language is not related to Illyrian. Albanians were annoyed to their suggestion of Albanian not related to Illyrian since there is evidence of relation and ignored their findings. To make my point, even if Albanian is not Germanic signs of some kind of relation are there. But the question is who brought that language to Albanians since genetically Albanians are Balkanic and R1b Germanic is not common among Albanians.

gothic and vandal invasion for this albanian is too late at ~ 400 AD

R1b is not Germanic
 
Personally I believe that Myceneans cause they are connected with Vucocar since we found many common with should be R1b
and Hellenes-Dorians should be R1a since Makedonians were Dorians but also consider as cousins the Bryges a Thracian nation, the Mygdones

Cyprus was first colonised by Myceneans and Achaioi,
much later only 2 areas colonised by Makedonians at Ammochostos/famagusta and North at Solotoi if remember correct,

but linguistic Cyprus belongs to Arcado-Cypriot Dialect,
which means is Connected with minor-Asian population the known at Hettite as Arzawa-Asuwa the ones in Centum Greek called Arcadia
 
This is refreshing - it's great to see that David Reich is so willing to adjust his hypotheses to accomodate new evidence. I'd think much of Gramkelidze's & Ivanov's work has been vindicated by recent DNA, though there's still quite some inertia when it comes to more popular theories. Anecdotally, what converted me from being Gimbutas' most loyal acolyte was G&I demonstration that early attested Indo-European societies fit quite perfectly in the pan-West-Asian cultural horizon. That always seemed like a stronger argument to me than the Y-DNA monomania and such and made me doubt their intrusive nature. This is in no contradiction to an intermediate steppe episode, which G&I discuss in several chapters.

Incidentally they also attempt to demonstrate the West Asian centrality of Greek culture. They even hint at the possibility that the extant Greeks might have been but an offshoot of a larger population inhabiting Anatolia and the Transcaucasus, which was displaced in its original homeland. If there was no early steppe incursion into the Balkans that suggestion might not be so outrageous after all.
 
I seem to have missed the tree :confused:

That's a bit odd, isn't it? Albanian a Germanic language? :startled:
 
So Albanian is Germanic? No wonder the Germans are one of the founders of the greater Albania project. Kosovo was first and looks like parts of Macedonia might be next.
 
This is the relevant paper for anyone who can properly interpret it:
https://www.cs.rice.edu/~nakhleh/CPHL/RWT02.pdf

That only applies to Mycenaeans, and I didn't say that a lot of Mycenaean Y-DNA lineages survived. Indo-European lineages were either R1a or R1b, but if you remove Slavic, Germanic, Celtic and Roman R1a and R1b in Greece, there is not much left. Mostly R1b-Z2103 and R1a-Z93, but that's just a few percents of the population. If the Dorians were R1b -Z2103 from the Balkans, then all that is left for the Mycenaeans is R1a. However R1a-Z93 probably came from the Near East. So chances are that the Mycenaeans had quite Proto-Slavic looking lineages like R1a-Z282 (either Z280 or M458) that cannot easily be distinguished from those of later Slavic invasions. That would explain why there is R1a all over Greece even though the Slavs only really settled in the north. Those two separate sources of similar-looking R1a would also explain why R1a is considerably higher in the north, but present throughout Greece. It is also in agreement with the new tree proposed by David Reich, which places Greek in the R1a branch alongside the Balto-Slavic and Indo-Iranian branches.

Early Slavic speakers settled pretty much throughout the Balkans, all the way deep into southern Greece, going by toponymy (even in Albania, especially the area south of the Shkumbin river and Tosks do seem to have R1a and I2a associated with the Slavic expansion at higher levels than Ghegs so that's kinda corroborated) but it's true that there seems to be a south-north cline in their impact, if not the exact density of toponyms, as you said.

Based on what I've seen, and with my limited knowledge, don't Albanians and Greeks have similar levels of R1b (~ low teens) potentially associated with early Indo-European expansions (rather than later ones from Western Europe into the Balkans etc.) by the way?

Also, to keep to the particular tree, there's a Greco-Armenian node and Armenian is pretty associated with R1b overall, isn't it? You can make the "Mycenaeans" post-Greek Indo-Iranians arrivals to Greece, per the theories of Huxley and Makkay of course, who would be using an already existing Greek dialect. Unless you also have in mind something along the lines of what Garrett considers in 'Convergence 
in 
the 
Formation 
of 
Indo‑European
 Subgroups:
 Phylogeny 
and
 Chronology'

http://linguistics.berkeley.edu/~garrett/IEConvergence.pdf
 
Last edited:
This probably wasn't your original intent when posting this, but I think this is something important that's often omitted from these discussions:

The first stage of IE language spread is characterized by a distinctive lexical, dervational, and onomastic profile; this corresponds to urbanization and the use of indigenous sociocultural traditions by speaker of IE languages. In Anatolia, Greece, and Bactria‑Margiana respectively, compare the dominant role of Hattic elements in Old Hittite religion and cult and ideology of kingship’ (Melcher 2003, 17), including Hattic loanwords like halmaššui‑‘throne’; the elite semantic profile of 'Minoan’ loans' in Greek (Renfrew 1998),including the vocabulary of kingship (Mycenaean wanaks > ánaks, perhaps gwasileus>basiléus).

This corresponds to the picture that Robert Drews paints in his 'The Coming of the Greeks' regarding the spread of Hittite, which is that of rather peaceful plebeian usurpation of an advanced urban culture. I imagine that this transformation couldn't have been effected without significant gene flow.
 
There are quite a few entries on the Hittites in "Empires of the Silk Road". He agrees it was a peaceful, gradual infiltration, not a military invasion.
https://books.google.com/books?id=-...r&source=gbs_atb#v=onepage&q=Hittites&f=false

Even if they find a Hittite sample with steppe, given the Mitanni influence I don't know how that would be dispositive. I think we need lots of samples from south eastern Balkan cultures.
 
That only applies to Mycenaeans, and I didn't say that a lot of Mycenaean Y-DNA lineages survived. Indo-European lineages were either R1a or R1b, but if you remove Slavic, Germanic, Celtic and Roman R1a and R1b in Greece, there is not much left. Mostly R1b-Z2103 and R1a-Z93, but that's just a few percents of the population. If the Dorians were R1b -Z2103 from the Balkans, then all that is left for the Mycenaeans is R1a. However R1a-Z93 probably came from the Near East. So chances are that the Mycenaeans had quite Proto-Slavic looking lineages like R1a-Z282 (either Z280 or M458) that cannot easily be distinguished from those of later Slavic invasions. That would explain why there is R1a all over Greece even though the Slavs only really settled in the north. Those two separate sources of similar-looking R1a would also explain why R1a is considerably higher in the north, but present throughout Greece. It is also in agreement with the new tree proposed by David Reich, which places Greek in the R1a branch alongside the Balto-Slavic and Indo-Iranian branches.

I get your point, but I feel we're limiting ourselves here by assuming that IE nations must have been either R1a dominant or R1b dominant, that could have been possible at the early days of Yamna maybe, but Indo-Europeans assimilated other groups in their migrations. In the case of Greece, E-V13 satisfies every criteria to be the lineage of proto-Hellenic people, it is a very young lineage that expanded during the bronze age, and not all of its branches are in the Balkans, so it is younger still, maybe a great patriarch lived as close as 2000 BC, R1b would be a secondary minor lineage, one Balkan variety is R1b-CTS9219.

By all means they could have expanded from the northern steppes of Russia if that was your argument for R1a, and I don't understand why we assume Dorians and Mycenaeans have different haplogroups, I don't know but didn't they speak Greek ? one language ? they weren't separate branches of IE, but different dialects of one language, that means they weren't much different genetically, E-V13 and R1b-CTS9219, and personally I would add some G-L13.

If Greek R1a is Mycenaean, two things must be true: 1- it would a clear and different branch than Slavic ones, maybe separated by a TMRCA of about 4000 ybp. 2- it would be present in all places settled by the Greeks, in their conquests and colonies, as well as Roman distribution, so a pan-European presence as well as middle eastern.
 
Do you think Sintashta, Andronovo, Srubnaya spoke Iranian not proto-Indo Iranian? And that the Indo Aryans with L657 were already moving towards South Asia in 2000BC?

Yes they were exclusively Iranian related culture. Who later became Scythians mixing with Siberians and W Asians.
Indo Aryans were already in Indus valley at 2000 - 1600 bc that is why they don't need Sintashta but Yamna.
 
Last edited:
Hittite DNA will be a hard issue as IIRC they cremated their deceased, if some DNA is got from just buried people what will prevent that he wouldn't be Hattic?

By the way if Greek Mycaenean and Hittite are attested from 1600 BC such languages might be in place at least in 2000 BC. Of course if Yamnayists are obsesswd to find steppe genes the genetists would'nt be capable to track the IE migration, but for those pointing to a northern homeland it's not so problematic. I recall that Pax Augusta found in RISE595 a 15% of North Slavic.

http://www.eupedia.com/forum/thread...-they-for-real?p=499438&viewfull=1#post499438

Even if I'm not confident comparing ancients with moderns (it is more like to find out how where Europeans in the Middle Ages comparing them with Southafrican coloureds), it is an interesting point, also to notice how the Yamnayist obsession can divert us from the right path.
 
@Markoz,

Yes, the whole paper is interesting I think, but this part is what I had in mind in this case:

"...it hardly makes sense to reconstruct Proto-Greek as such: a coherent IE dialect, spoken by some IE speech community, ancestral to all the later Greek dialects. It is just as likely that Greek was formed by the coalescence of dialects that originally formed part of a continuum with other NIE (Nuclear Indo-European) dialects, including some that went on to participate in the formation of other IE branches"

As for E, J etc. and their potential assimilation on the way to their final stop, various theories (that of Michel Sakellariou in 'Les Proto-Grecs' comes to mind - and sorry for the vagueness since it's been a while :embarassed:) that agree with a steppe urheimat have the steppe-associated "proto-Greeks" being accompanied to the south with a non-Indo-European Balkan element alongside them. So it's not unlikely that those lineages were just as important or even more important than any R lineage when those groups finally arrived to the south sometime in the 2200-1900 BC period, assuming such a scenario. But I guess we'll see!
 
Regarding the new tree, I had proposed since 2009 that the Mycenaeans were predominantly R1a and descended from the Srubna culture by a nearly direct migration from the Steppe around 1600 BCE. I believe that the later Dorian migration brought R1b (Z2103, U152?) to Greece. Therefore Greek is a hybrid IE branch, and this is hard to show on such a tree.

However I disagree that this should be the case of the Armenian branch too. It is clear that Armenians are predominantly R1b-Z2103 (and especially L584). They have 30% of R1b for only 5% of R1a. The Mitanni or other Indo-Iranian tribes surely brought R1a-Z93 and probably also some R1b-Z2103 (Y24543 clade, found in Armenia, South Asia and Ukraine). That's probably what they found with ancient DNA. The Mitanni came first to Armenia (from 1500 BCE), then the actual Proto-Armenians from the Balkans (from 1200 BCE). That's what archaeology and history say.

I also find it odd to place Albanian with the Germanic branch. I wonder what are their justification for that. Albanian belong to R1b-Z2103 (SE European CTS9219 clade), not even to R1b-L51. I would place Albanian split a bit before the split between Italo-Celtic and Germanic. Phylogenetically, the Albanian branch is more closely related to the Armenian branch.

This is most likely the case.
 
The "weird tree" comes from 'Indo-European and Computational Cladistics' by Ringe et al. Albanian assumes a more typical position when Germanic is removed.
Ah. This implies a latter influence of Germanic rather than generative.
 

This thread has been viewed 43280 times.

Back
Top