Where did proto-IE language start?

Source of proto-Indo-European language

  • R1a

    Votes: 23 31.9%
  • R1b

    Votes: 22 30.6%
  • Cucuteni-Tripolye

    Votes: 10 13.9%
  • Caucasus-Mykop

    Votes: 17 23.6%

  • Total voters
    72

Oh God, the more I read from that blog the more I think he's tremendously confused particularly after he became obsessed with establishing this unlikely close connection between Proto-Uralic and Proto-IE as the similar languages of neighbor cultures in the Pontic-Caspian steppe. I wonder who told him the "Indo-Uralic" hypothesis is even remotely about a close, direct relationship between IE and Uralic, the connections, if you take out all probable loanwords, are at best VERY tenuous and mostly rest on extremely basic and stable parts of the lexicon and syntax, mainly the forms of some (some, not most) personal and relative pronouns. And here he is trying to make us believe that Khvalynsk-Sredny Stog I spoke Indo-Uralic in 5,000 BC, that late Sredny Stog (which he assumes is the only genetic AND cultural ancestor of CWC) was already speaking fully formed Proto-Uralic by 4,000-3,500 BC (Sredny Stog was absorbed by Yamna right after 3,500 BC), and that Yamna was already speaking Late PIE in 3,500-3,000 BC! Come on, even from just a basic linguistic point of view, this is extremely improbable, if not dowright impossible. Proto-Anatolian split from PIE centuries earlier than the other branches, probably as early as 4,000-3,500 BC, so are we to assume that the split of Proto-Anatolian was just ~1,000 years after the split of Proto-Uralic, which is, compared to PIE, an absurdly divergent language with few traces of possible common roots?

And then there is all that obsession with CWC being the ancestor of Uralic languages even though virtually half of Uralic branches appeared where CWC never even existed, east of the Urals, and it is Siberian admixture (even if a minor one) and N1c, not R1a-M417, that are genetic markers really found in ALL Uralic-speaking regions without excepton... This idea of "Uralic Sredny Stog", even though Sredny Stog and Yamna clearly had much closer origins than any reasonable Indo-Uralic proposal imagined, seems to have become his fixture. :-O
 
And then there is all that obsession with CWC being the ancestor of Uralic languages even though virtually half of Uralic branches appeared where CWC never even existed, east of the Urals, and it is Siberian admixture (even if a minor one) and N1c, not R1a-M417, that are genetic markers really found in ALL Uralic-speaking regions without excepton... This idea of "Uralic Sredny Stog", even though Sredny Stog and Yamna clearly had much closer origins than any reasonable Indo-Uralic proposal imagined, seems to have become his fixture. :-O

Irrespective of what he says (I don't read that blog), some R1a subclades could have existed in the proto-Uralic homeland. That depends on how old the proto-language is, but some support that it is ~5000 years old and it expanded during the last 4000 years, from West to East (from the forest steppe?).

There are many Uralic groups which have 35-40% R1a.

Also Tatars from Tatarstan have 37.9% (according to Underhill) but they also have relatively high diversity. There is high diversity even in places with low frequency like Turkey, Sweden (Malmo), Italy, Crete or Qatar but in Tatarstan there is relatively high diversity with relatively high frequency.

That can be explained in various ways and irrespective of what their ancestors originally were, some R1a subclades could have expanded with Turkic groups too.

Since that is evident, it makes no sense giving haplogroups labels like 'Iranic', 'Slavic' or 'Celtic', 'Germanic' etc.
 
Irrespective of what he says (I don't read that blog), some R1a subclades could have existed in the proto-Uralic homeland. That depends on how old the proto-language is, but some support that it is ~5000 years old and it expanded during the last 4000 years, from West to East (from the forest steppe?).

There are many Uralic groups which have 35-40% R1a.

Also Tatars from Tatarstan have 37.9% (according to Underhill) but they also have relatively high diversity. There is high diversity even in places with low frequency like Turkey, Sweden (Malmo), Italy, Crete or Qatar but in Tatarstan there is relatively high diversity with relatively high frequency.

That can be explained in various ways and irrespective of what their ancestors originally were, some R1a subclades could have expanded with Turkic groups too.

Since that is evident, it makes no sense giving haplogroups labels like 'Iranic', 'Slavic' or 'Celtic', 'Germanic' etc.

Yes, but what you're saying is much broader and vaguer - and thus, yes, more certainly correct (there was R1a in Proto-Uralic homelands, which doesn't mean it came with males carrying it, and is also very defensible since Proto-Uralic expansion looks like coming from the language of a mix of Siberian incomers with peoples already living in and around the Urals).

But what is really hard to believe, for now, is the assumption involving this very specific combination - R1a-M417, CWC and Sredny Stog-related admixture - are all THE origins of Proto-Uralic, and not a very distant origin, but right there neighboring Indo-European Yamna and coming from the same immediate source (that's the part that I find most unlikely in this hypothesis). Everything about this "new model" looks very far-fetched to me right now, at least until we find much better evidences, especially ones that explain how and why CWC only overlaps with some - not even most - Uralic territories and what role the Siberian-related admixture and N1c had in this process, especially since that blogger assumes that N1c men adopted the local Uralic CWC languages, but that clearly didn't happen with the high % of N1c men who speak Baltic languages, not Uralic, and also not Paleo-Siberian languages.
 
Yes, but what you're saying is much broader and vaguer - and thus, yes, more certainly correct.
Yes. Ok. I just made that post because, very often, people assume that proto-Uralic speakers just had one haplogroup.Imo, they should sample Fatyonovo & Abashevo.

Fatyanovo is a culture where it is said that many Baltic hydronyms existed (I haven't checked the data myself and I am not sure if we can definitely say when those hydronyms were first used but it is a culture where a Baltic or Baltic-like language could have been spoken)

It is considered an Eastern extension of Corded Ware but I personally expect they would have more EEF/ANF admixture than typical Corded Ware samples because there are cultural similarities with the Globula Amphora culture.

Abashevo also seems important because it is thought that around there some Iranic loans entered proto-Uralic.

Then Seima-Turbino samples would have been interesting since some associate the expansion of Uralic languages with it.
 
This whole argument boils down to the very mechanics of what you believe about language development. Personally I believe that languages have two bases of transmission, natural genetically or innovation and wave. Just as early forms of writing were copied by other cultures/societies that were capable and could find utility in them, so too would be useful lingual concepts. I think this would be especially true in pre-history. The Vedics are supposed to have memorized their works and passed them down unchanged into history. I have no reason to believe this was not also true of many other complex pre-historical civilizations or societies.

Surely there were polyglots throughout all ages who would translate and trade far and wide. We know extensive ancient trade networks existed long before the start of any historical record. When one encountered a language or groups of languages with interesting and adaptable innovations that might make ones own more efficient, it would be natural to do so, such innovations that “strike a chord” could spread throughout a region very quickly. And that’s the thing, more than just words were shared between languages. I’d think this would be more the case with languages with no written form, language was less formal, blended across cultural borders and more prone to adaptation. I think there were more language innovation waves historically than trees though trees went with invasions and changes of lifestyle so, of course, they make the bigger splashes. We ignore and kind of overlook the regulating effects of the language habits/practices/accents of native or extant populations on the language of the invaders or new lifestyle.

I believe the Basque and their apparent mass exogamy event hold a massive piece of the tree versus wave puzzle. If you want to understand what the evolution of pre-historical languages might have been like though you must pay attention to the historical evolution of languages. A lot of isolates were eaten up in the envelopment of empires but left their mark on adapted IE languages which later Romanticized or Germanicized throughout Western Europe. When I look at historical language development, I don’t see too many trees, I see blending of languages, influences and waves of innovation... why would we expect the past to have been any different? Italian, French, Spanish, English, despite many attempts to standardize remain quite diverse in terms of regional dialects... many of the differences now are due in part to recent events but at the root they’re based in long standing differences of influences between regions.

without pronostic concerning the PIE heimat, I would say:
"waves" in languages are in the wind to date; I don't refuse this insight, but I moderate its input: when "brother" languages are spoken on a large space or separated since a short time adn still in contact by some of their speakers, this notion of wave can apply; but in far descendants of a previously common language, separated by time and space, the seldom new tied contacts allow loans of words, rarest loans of bits of sentances ("frozen") but they don't allow exchanges of syntaxic and pronounciation tendancies, IMO. The commonest influence onto languages is the sbstratum, and it concerns languages shifts, here we could be in accord - but I see still the "tree" in our languages, and concerning innovation on the syntaxic level I see more fragmentation than osmosis spite the weak input of polyglot people of the academic world or of the trading world.
It would be interesting to have the opinion of somebody aware of the syntaxic evolution of American English and Spanish in states where the so called 'latino' minority is strong?
 
Oh God, the more I read from that blog the more I think he's tremendously confused particularly after he became obsessed with establishing this unlikely close connection between Proto-Uralic and Proto-IE as the similar languages of neighbor cultures in the Pontic-Caspian steppe. I wonder who told him the "Indo-Uralic" hypothesis is even remotely about a close, direct relationship between IE and Uralic, the connections, if you take out all probable loanwords, are at best VERY tenuous and mostly rest on extremely basic and stable parts of the lexicon and syntax, mainly the forms of some (some, not most) personal and relative pronouns. ... Come on, even from just a basic linguistic point of view, this is extremely improbable, if not dowright impossible. Proto-Anatolian split from PIE centuries earlier than the other branches, probably as early as 4,000-3,500 BC, so are we to assume that the split of Proto-Anatolian was just ~1,000 years after the split of Proto-Uralic, which is, compared to PIE, an absurdly divergent language with few traces of possible common roots?

The writing style on that blog is very confusing indeed. But I do think that a Volga-Steppe connection for Indo-Uralic is plausible. There is enough commonality between IE and Uralic that many linguists assume that they must be related in some way. But most of these linguists assume that such a connection must be more distant in time, going back to the end of the ice age. And therefore almost nobody bothers to compare these two language families to see if a common proto-language can be reconstructed. Even the Moscow school of linguistics, with its Altaic and Nostratic language families, only compares the two at a Eurasiatic level together with Altaic and a bunch of other languages. But one example where a linguist does make such a direct comparison is Adam Hyllested's 'Internal Reconstruction vs. External Comparison: The Case of the Indo-Uralic Laryngeals'. And it paints a completely different picture. I myself (as an amateur) am involved in a comparison of IE and Uralic, based on that work. And much of what I see confirms that picture. If there was a common Indo-Uralic language, they split much later than is commonly assumed.

Also, we also have to keep in mind that IE is reconstructed based on languages from antiquity and even the late bronze age. But Uralic is reconstructed from languages of the modern age. Suppose we did not have written records of all these ancient Indo-European languages. So we would not have had Sanskrit, Ancent Greek, Latin, Anatolian and Tocharian. If we had to reconstruct IE from only the languages of the modern age, we would not reconstruct ablaut and laryngeals. Sure there is the Germanic strong verb which reflects the original IE ablaut (but with many innovations). But without a strong parallel from other modern IE languages, we would assume this was a Germanic innovation. There are also vowel alternations in Ob-Ugric (especially Khanty), which are similar in scope to the ones in Germanic. Uralicists generally assume that this is an innovation of Ob-Ugric. But it could also be that both IE and Uralic had ablaut. For IE we can reconstruct ablaut and have to reconstruct ablaut because it is attested in the ancient languages.
 
The writing style on that blog is very confusing indeed. But I do think that a Volga-Steppe connection for Indo-Uralic is plausible. There is enough commonality between IE and Uralic that many linguists assume that they must be related in some way. But most of these linguists assume that such a connection must be more distant in time, going back to the end of the ice age. And therefore almost nobody bothers to compare these two language families to see if a common proto-language can be reconstructed. Even the Moscow school of linguistics, with its Altaic and Nostratic language families, only compares the two at a Eurasiatic level together with Altaic and a bunch of other languages. But one example where a linguist does make such a direct comparison is Adam Hyllested's 'Internal Reconstruction vs. External Comparison: The Case of the Indo-Uralic Laryngeals'. And it paints a completely different picture. I myself (as an amateur) am involved in a comparison of IE and Uralic, based on that work. And much of what I see confirms that picture. If there was a common Indo-Uralic language, they split much later than is commonly assumed.

Also, we also have to keep in mind that IE is reconstructed based on languages from antiquity and even the late bronze age. But Uralic is reconstructed from languages of the modern age. Suppose we did not have written records of all these ancient Indo-European languages. So we would not have had Sanskrit, Ancent Greek, Latin, Anatolian and Tocharian. If we had to reconstruct IE from only the languages of the modern age, we would not reconstruct ablaut and laryngeals. Sure there is the Germanic strong verb which reflects the original IE ablaut (but with many innovations). But without a strong parallel from other modern IE languages, we would assume this was a Germanic innovation. There are also vowel alternations in Ob-Ugric (especially Khanty), which are similar in scope to the ones in Germanic. Uralicists generally assume that this is an innovation of Ob-Ugric. But it could also be that both IE and Uralic had ablaut. For IE we can reconstruct ablaut and have to reconstruct ablaut because it is attested in the ancient languages.

I agree with you and also believe that there were very distant relationships (some people really overestimate the similarities, they're few, unsystematic and in very basic parts of the language) between the ancestors of Uralic and Indo-European, but that's all I think there is. To create an entire hypothesis around CWC being Uralic and deriving its language and genetics from a Indo-Uralic Sredny Stog that became Uralic in CWC and Indo-European already in Yamnaya (less than 1000 years later), is just too much for me and, I'd guess, for virtually every professional linguist in the world. That's assuming a relationship that is much much much closer than can be inferred and estimated, actually that would be assuming that Proto-Uralic and Proto-Indo-European would've been as apart from each other as, for instance, Portuguese is from Spanish nowadays (these two started to diverge around 900-1000 AD). Uralic and Indo-European are extremely unlike each other, they aren't even the same kinds of languages (one agluttinative, the other fusional; one genderless, the other gendered and with strong evidences that even in its earliest stage, without a masculine vs. feminine distinction, it was already gendered), and this kind of differentiation in the fundamental way a language works takes milennia to fully develop.
 
I don't buy the Indo-Uralic Sredny Stog or CWC theories. We simply do not know what they spoke. But Iet's assume that the Indo-Uralic split-up was just before the time when the first Kurgan cultures (Khavalynsk and Samara) appeared south of the Volga-Kama area. That would be ~5500 BCE. The common estimate for when Uralic began to disintegrate is around 3000 BCE. So a more apt comparison would be between English or Afrikaans 2000 CE (entirely genderless, and with little inflection) and Gothic 500 CE (three genders and plenty of inflection) which both came from common Germanic 500 BCE.

The first group to split up was Samoyedic. It might not be well known, but there are only about 200 words that Samoyedic and the rest of Uralic have in common under regular sound correspondences. Uralic itself is kind of a long-range reconstruction. But it is a very successful one. Even the worst splitters won't risk their reputations attacking the genetic relationship between Samoyedic and the rest of Uralic (Finno-Ugric).
 
I don't buy the Indo-Uralic Sredny Stog or CWC theories. We simply do not know what they spoke. But Iet's assume that the Indo-Uralic split-up was just before the time when the first Kurgan cultures (Khavalynsk and Samara) appeared south of the Volga-Kama area. That would be ~5500 BCE. The common estimate for when Uralic began to disintegrate is around 3000 BCE. So a more apt comparison would be between English or Afrikaans 2000 CE (entirely genderless, and with little inflection) and Gothic 500 CE (three genders and plenty of inflection) which both came from common Germanic 500 BCE.

The first group to split up was Samoyedic. It might not be well known, but there are only about 200 words that Samoyedic and the rest of Uralic have in common under regular sound correspondences. Uralic itself is kind of a long-range reconstruction. But it is a very successful one. Even the worst splitters won't risk their reputations attacking the genetic relationship between Samoyedic and the rest of Uralic (Finno-Ugric).

A Late Mesolithic relationship is plausible, but nothing later than that. Also, I think that a Proto-Uralic vis à vis Proto-Indo-European relationship would be more . English and Afrikaans are still Germanic in the way the grammar functions and, particularly, are still essentially fusional languages (they just reduced and regularized inflections, but didn't get rid of them in favor of an entirely new structure - it was a simplification, not yet a complete transformation), and most linguists would consider that the transformation from agluttinative to fusional is a very slow and profound one, one that would usually require many thousands of years and a numerous succession of changes in the language. The distinction (in lexicon and also in syntax) between PIE and PU as they're reconstructed is very significant, suggestive of a split way earlier than the times where they must've existed (~4500-6000 years ago).

If I had to bet, I'd place the latest possible date of an Indo-Uralic language around 3000-3500 years before the existence of PIE. If it seems increasingly likely that even Late PIE in Yamnaya times wasn't the earliest stage of PIE, due to the presence of Anatolian, then the Early PIE could be as ancient as Sredny Stog (~4200-4000 BC).

So Indo-Uralic would've been spoken in 7000-8000 BC at the latest (emphasis on "latest") - and that is simply assuming that such a Indo-Uralic common language ever existed, which is not totally certain. I wouldn't entertain any possibility of linking "Indo-Uralic" to any post-Neolithic revolution culture of the steppes/forest-stepp. But a Mesolithic , okay, that can be discussed.
 
... English and Afrikaans are still Germanic in the way the grammar functions and, particularly, are still essentially fusional languages (they just reduced and regularized inflections, but didn't get rid of them in favor of an entirely new structure - it was a simplification, not yet a complete transformation), and most linguists would consider that the transformation from agluttinative to fusional is a very slow and profound one, one that would usually require many thousands of years and a numerous succession of changes in the language. ....

This is bullshit. Terms like analytical, isolating, fusional and agglutinating are very vague at best and misleading at worst. And those 'most linguists' that you cite are certainly aware of that. But when you do apply more precise definitions, like WALS does, you see that the Uralic languages score more like fusional languages than agglutinative ones.
 
This is bullshit. Terms like analytical, isolating, fusional and agglutinating are very vague at best and misleading at worst. And those 'most linguists' that you cite are certainly aware of that. But when you do apply more precise definitions, like WALS does, you see that the Uralic languages score more like fusional languages than agglutinative ones.

Well, I happen to have read exactly that from a linguist, who pointed out that PU is reconstructed to have been very purely agglutinative, unlike PIE which was extremely fusional. Those terms are not "vague", they have definite meanings even though they aren't rigid and actually form a kind of continuum from one extreme to another. But they still exist, even though there is, as linguists often claim, a general tendency to shift gradually from one extreme position to another along that continuum.

Besides, as I've said, the very specific, and not "vague", ways that the morphology and syntax of reconstructed Proto-Uralic and PIE worked were extremely unlike each other and, apart from very few morphemes (like *-ti vs. *-si for the 2nd person), based themselves on affixes that are clearly not cognates of each other. It is not just that one language is fusional and the other is agglutinative. The ways those different structures function to form words and build sentences are very distinct and do not look like they have preserved much in common. PU or PIE are not just "simplified" or "reduced" versions of the same system used by an earlier language (e.g. the current structure of English preserves "original" Germanic things as basic as prepositions and pronouns, and the simpler syntax derives mostly from regularizations and simplifications of earlier Germanic declensions, e.g. the "s" plural), which was better preserved by the other family. They are, compared to each other, distinct on deep levels, which points out to a very early split.

AFAIK there is virtually no linguist in the world proposing that if Indo-Uralic existed it existed immediately before the last stage of undivided PIE (that would be around the time of Sredny Stog). Those that do accept (it's most definitely not a consensus) the Indo-Uralic hypothesis talk about a distant relationship. The correspondences between PU and PIE are just not enough to affirm any strong resemblance to each other, and I'd actually be very glad to see you giving us more evidences that in fact they were closely related in many more aspects than we have learned yet.

P.S.: Next time avoid making your very 1st comment on a topic by simply calling what another poster said "bullshit". That's uncalled for and, honestly, sounds like bad and baseless rhetoric. We want to have civilized debates here, not heated online arguments.
 
One of the very fusional ways to form a present in IE is by plopping a *n?/*n in the middle of the root. This is known as the nasal infix present. For example PIE *leikʷ 'to leave' (~ Uralic *l?kti 'to depart') had a present *lin?kʷti/linkʷ?nti, which is reflected in Latin relinquō, relictus.

Many IE languages have remnants of these forms. For example, the English word 'lamp' comes from the Greek lamp?s 'torch, light' which is derived from l?mpō 'to shine', which is a nasal infix present from PIE *leh₂p 'to glow'. Greek is the only language that has a nasal infix present from this root.. except Uralic which has *l?mpi 'warmth'. The distance between where Ancient Greek and Uralic were spoken plus the semantic divergence make a loan between IE and Uralic nearly impossible here.
And there are more examples of this: PU *w?ntV 'to see, to look at, to watch' ~ PIE *weid 'to see' > Latin videō 'to see'. PU *śaŋka- 'to sting, to stick' ~ PIE *stegʰ- 'to sting, to stick'.

When I found that, and I found that there were no need for 'wild' sound correspondences between the PIE consonants and PU consonants, that's when I realized that these language families must have split much later than most linguists assumed. One common assumption is that PIE gained its fusional traits as ablaut because of its later close interaction with Caucasian peoples. But if the language from which both Uralic and IE derive already clearly had these traits, then this hypothesis is baseless.

The Uralic reconstruction is so agglutinating because it assumes that all the fusional traits in the daughter languages are innovations. This makes some sense for the ablaut in Ob-Ugric, since there is only one branch that has it. But there is another typical Uralic stem alternation for which this makes less sense. I am talking about Consonant Gradation. Saami, Finnic, and some Samoyedic languages (Nganasan and Selkup) have it. The places where these languages are spoken are about a thousand miles apart. And yet, they have the same system. The thing is that consonant gradation does not leave many traces in Proto-Uralic, since Proto-Uralic basically has one stop series. But I have one good example where we see what could be remnants of both:
PU *j?k? 'river' / PU *juka 'river' (~ PIE h₂ep- 'river' (?) /h₂ekʷeh₂ 'river, water')
PU *j?x? 'to drink' (~ PIE h₁egʷʰ 'to drink')

And here you see that Consonant Gradation is something that may also have been active in PIE. In other words, Proto-Indo-Uralic might even have been more fusional, [with both consonant and vowel alternations] than reconstructed PIE. An possible alternation between unvoiced - voiced aspirate is something that has been noticed in more more instances within PIE. See the paper 'Consonantal Alternations in Indo-European Roots: Diatopic and/or Diachronic Variants or Functional Mechanism?' by 'Aldo Luiz Bizzocchi'.
 
Last edited:
Very interesting. So when would you date that split, how long before the existence of already completely differentiated Proto-Uralic and Proto-Indo-European in the Chalcolithic/Early Bronze Age? I cannot see it happening in the immediately preceding Neolithic of the Pontic-Caspian region, so I'd estimate a split before 7000 BC, possibly more between 7500-8500 BC, if there are glottochronological and increasingly linguistic and also archaeological evidences of an Anatolian IE (already very distinct from Proto-Uralic) split as early as ~4000 BC, which indicates that by the mid 5th milennium BC there was already a PU/PIE division. That would be more or less like the (reasonably high, but still traceable) degree of divergence between Irish and Italian, or Finnish and Hungarian.
 
I would go for somewhere in the 5500 BCE-7000 BCE range. But this is a topic that needs more research. Another important point in light of the most recent genetics papers is that the language families that have the most in common with PIE (Uralic, Yukaghir, Turkic, Mongolic, Tungusic, ...) are all in the north-east. This makes a PIE homeland south of the Caucasus very improbable to me. See also the papers 'Inferring The World Tree Of Languages From Word Lists' and 'Support for linguistic macrofamilies from weighted sequence alignment' by Gerhard J?ger.
 
Where are the "None of the Above" and "All of the Above" options?
 
Ancient languages from R1a people (as Balto-Slavic and Indo-Iranian) consists exclusively Indo-European words, 100% percent of them. While among languages, which are influenced by R1b (or even were totally carriers of R1b), the non-Indo-European caracteristics are foundable. It doesn't mean that those languages are non-Indo-European, but just because of that, i think that IE language is originally from R1a people.
 
I agree with that. All the most innovative Indo-European languages are majority R1b. Sanskrit and Lithuanian are the most archaic Indo-European languages, and they are both R1a.
 
I agree with that. All the most innovative Indo-European languages are majority R1b. Sanskrit and Lithuanian are the most archaic Indo-European languages, and they are both R1a.

Lithuanian is the most archaic of the modern Indo-European languages, but i would not compare it with Sanskrit. It's a bit exaggerated about Lithuanian archaism imho. Phonologically Lithuanian is the most archaic IE-language in some obvious respects, e.g. preserving diphtongs and word-final syllables, but in other respects Lithuanian greatly innovated. For example, verbal system of Old Church Slavonic (also proto-Slavic) is more archaic than Lithuanian. But anyways, R1a languages are in general more archaic, and this is true.
 
Ancient languages from R1a people (as Balto-Slavic and Indo-Iranian) consists exclusively Indo-European words, 100% percent of them. While among languages, which are influenced by R1b (or even were totally carriers of R1b), the non-Indo-European caracteristics are foundable. It doesn't mean that those languages are non-Indo-European, but just because of that, i think that IE language is originally from R1a people.

Hi, Srbadija. I'm afraid I disagree on that. The objection lies in the centum/satem split itself. PIE "h1ekwos", very close to Latin "equus", can also evolve into Sanskrit "as'va". But there is no way the Sanskrit "sh" sound can evolve into "k" or "kw". Which means that Centum forms have to be the original ones, and Satem the derived ones. In other words, if you associate Centum with (some) R1b tribes, and Satem with (some) R1a tribes, then R1b must have been the speakers of the most "primitive" forms of PIE. Or at least they were the clans that preserved the original phonology over a longer period.

This said, I am quite prepared to agree with you that as Centum PIE moved west, it was gradually altered by the lexical additions and phonological influences from the substrate languages it superseded.
 

This thread has been viewed 182284 times.

Back
Top