Language trees support hybrid model for origin of Indo-European languages

The latest archaeogenetic data must be used and this is not visible on this map. The main R1a Z93 branch archaeogenetically shows migration route north of the Caspian Sea and not directly from Caucasus through Anatolia to India. I think that such an approach is frivolous.

Indian geneticists have proved that R1a-Z93 came to India about 1,000 years after the appearance of Indo-Aryan culture in India, so it doesn't relate to the spread of Indo-Aryan languages in the south Asia, their paper will be published soon.
 
Indian geneticists have proved that R1a-Z93 came to India about 1,000 years after the appearance of Indo-Aryan culture in India, so it doesn't relate to the spread of Indo-Aryan languages in the south Asia, their paper will be published soon.
Looking at the clade-cross-correlation (a statistical probe that I use to evaluate migrations time) for R-Z93 between Russia and India, I got that :

Capture-d-e-cran-2024-01-12-a-16-27-02.png

Mainly, this probe indicates that India and Russia got sourced by R-Z93 populations that "splitted" a first time around ~1800 BCE and a second time around ~1200 BCE.
I think we would agree that R-Z93 around 2500 BCE started from ~Russian steppe.
Therefore, this probe imply that the populations that would source India with R-Z93 left Russian lands in two waves, one around 1800 BCE and one around 1200 BCE (PS: I didn't check the structure of these two splits, because I'm too lazy, to evaluate if we are dealing with two injections of R1a or one injection and a backflow ... ).

At least, the dates would align well with Cemetery H culture.
On a general note, I'm always cautious about works that use "absence of signal" as a proof (because our current coverage of aDNA is awfull). I prefer using aDNA only as a "positive" proof (that's why I also didn't consider the current absence of WSH in BA-Anatolia as some definite proof).

But I hope the paper you speak about will air soon, I'm interested about the methodology behind the conclusions you report.
 
Indian geneticists have proved that R1a-Z93 came to India about 1,000 years after the appearance of Indo-Aryan culture in India, so it doesn't relate to the spread of Indo-Aryan languages in the south Asia, their paper will be published soon.
Indian geneticists can't prove anything for now because they have no confirmation in archaeogenetics or living YFull genetics. Therefore, what we have now is R1a Z93 direction of migration north of the Caspian Sea.

Archaeogenetic evidence, which would indicate closer living of R1a and R1b people, for now is in southeastern Ukraine. This place could be a candidate for a common house or area of Proto-Indo-European language at this point.

In Anatolia, for now, we do not have any archaeogenetic evidence of R1b and R1a groups living in the same place or next to each other.

By the way, on this new map, location from where they supposedly start their migration starts from one house. However, the main Indo-European waves towards Europe do not show that R1b came from a common source with R1a Z280 M458 branches, which means that separation of R1b-R1a tribe happened before 5000 ybp.

R1a migration from the steppe to Europe begins no earlier than the emergence of R1a Z93, which is TMRCA 4500-5000 ybp.

Considering that R1b also has a similar time period when it starts from the Russian Steppe 5000, 4500 ybp, it means that the migrations of R1a, R1a Z93 and R1b groups are connected and that they must all live together near each other at that time. At this point, the candidate for the Proto-Indo-European homeland can only be the Russian Steppe.

 
Looking at the clade-cross-correlation (a statistical probe that I use to evaluate migrations time) for R-Z93 between Russia and India, I got that :

Capture-d-e-cran-2024-01-12-a-16-27-02.png

Mainly, this probe indicates that India and Russia got sourced by R-Z93 populations that "splitted" a first time around ~1800 BCE and a second time around ~1200 BCE.
I think we would agree that R-Z93 around 2500 BCE started from ~Russian steppe.
Therefore, this probe imply that the populations that would source India with R-Z93 left Russian lands in two waves, one around 1800 BCE and one around 1200 BCE (PS: I didn't check the structure of these two splits, because I'm too lazy, to evaluate if we are dealing with two injections of R1a or one injection and a backflow ... ).

At least, the dates would align well with Cemetery H culture.
On a general note, I'm always cautious about works that use "absence of signal" as a proof (because our current coverage of aDNA is awfull). I prefer using aDNA only as a "positive" proof (that's why I also didn't consider the current absence of WSH in BA-Anatolia as some definite proof).

But I hope the paper you speak about will air soon, I'm interested about the methodology behind the conclusions you report.
I know about some new papers regarding ancient Iranian-speaking people which say not only ancient Medes and Persians but even ancient Parthians had no steppe ancestry, there are new samples from Mersinchal, a Parthian cemetery near Hecatompylos, the first capital of Parthian empire.

Parthia.jpg
 
"In summary, we demonstrate a higher proportion of genomic sharing between PNWI populations and ancient EHG and Steppe-related populations than we observe in other South Asians. We report that the Ror are the modern population that is closest to the first prehistorical and early historical South Asian ancient samples near the Indus Valley, and they also harbor the highest Steppe-related, EHG, and Neolithic Anatolian ancestry. However, compared to other adjoining groups, the Ror show less affinity with the Neolithic Iranians."

The Genetic Ancestry of Modern Indus Valley Populations from Northwest India​

 
I know about some new papers regarding ancient Iranian-speaking people which say not only ancient Medes and Persians but even ancient Parthians had no steppe ancestry, there are new samples from Mersinchal, a Parthian cemetery near Hecatompylos, the first capital of Parthian empire.

The main issue, that make me stick to my position that the question will be hard to settle (if even possible), is that diffusing a language didn't require a significant injection of DNA (and on the opposite significant DNA-injection didn't insure that a language will be diffused).
We have many documented exemples.
Thus there is room for a CHG model, but I think a WSH model is also not out of the picture.
 
I know about some new papers regarding ancient Iranian-speaking people which say not only ancient Medes and Persians but even ancient Parthians had no steppe ancestry, there are new samples from Mersinchal, a Parthian cemetery near Hecatompylos, the first capital of Parthian empire.

Maybe you're talking about this scientific paper, "An Ancient Harappan Genome Lacks Ancestry from Steppe Pastoralists or Iranian Farmers"

"Our results also have linguistic implications. One theory for the origins of the now-widespread Indo-European languages in South Asia is the “Anatolian hypothesis,” which posits that the spread of these languages was propelled by movements of people from Anatolia across the Iranian plateau and into South Asia associated with the spread of farming. However, we have shown that the ancient South Asian farmers represented in the IVC Cline had negligible ancestry related to ancient Anatolian farmers as well as an Iranian-related ancestry component distinct from sampled ancient farmers and herders in Iran. Since language proxy spreads in pre-state societies are often accompanied by large-scale movements of people (Bellwood, 2013), these results argue against the model (Heggarty, 2019) of a trans-Iranian-plateau route for Indo-European language spread into South Asia. However, a natural route for Indo-European languages to have spread into South Asia is from Eastern Europe via Central Asia in the first half of the 2nd millennium BCE, a chain of transmission that did occur as has been documented in detail with ancient DNA. The fact that the Steppe pastoralist ancestry in South Asia matches that in Bronze Age Eastern Europe (but not Western Europe [de Barros Damgaard et al., 2018, Narasimhan et al., 2019]) provides additional evidence for this theory, as it elegantly explains the shared distinctive features of Balto-Slavic and Indo-Iranian languages (Ringe et al., 2002)."


So the Balto-Slavic language according to the YFull tree has a common connection with the ancestors of R1a Z93, and that is their common source, which is separated about 5000 ybp, but the words that exist in Sanskrit, which are common to the Balto-Slavic words, exist 5000 ybp. Migration through Anatolia is not possible because it would be genetically visible in the area of Harappa. But that is not the case for now.
 
Last edited:
"Thus linguistically speaking Heggarty et al. 2023 claims that the Indo-Iranic/Balto-Slavic node is unlikely."

"The closest relatives of Balto-Slavic are Albanian and Indo-Iranian. The large majority of special correspondences between Balto-Slavic and IndoIranian are archaisms, not innovations. This is important because it implies that a comparison of Balto-Slavic with Indo-Iranian leads to a reconstruction of an early stage of Indo-European."

Balto-Slavic and Indo-Iranian Frederik Kortlandt

Therefore, Balto-Slavic and Indo-Iranian in their common homeland, the Russian steppe, are at that place important for reconstruction of an early stage of Indo-European language.
 
Recent peer reviewed paper Heggarty et al. 2023 (suppl.) said (about Indo-Iranic and Balto-Slavis sharing a recent node) :
REF : https://www.science.org/doi/10.1126/science.abg0818
The claim that the Ringe tree topology supports the Steppe hypothesis rests especially on one particular node,which has also been taken as particularly important for contextualizing ancient DNA findings. The last major nodein the Ringe tree keeps Indo-Iranic and Balto-Slavic together in a single common branch until very late in thefamily’s divergence history, after all other main branches have split off. This has been taken to entail that theancestor of Indo-Iranic languages would likely have still been outside India and Iran, and further north on theSteppe, until a correspondingly late date. In Ringe’s chronology, the lineage ancestral to all Indic and Iraniclanguages remained undistinguished from the lineage ancestral to Baltic and Slavic languages until as late as c.2450 BC, i.e. less than a millennium before the time of Vedic, set by Ringe at 1500 BC. (It is not in fact universallyaccepted that Balto-Slavic itself is a valid node, although this is the majority view and in our analysis has a highposterior probability. For discussion, see (154).)
Even in Ringe’s own data-set, support for this last major node of [Indo-Iranic + Balto-Slavic] is very thin: it consistsof only three binary characters (while other characters conflict with any single tree). Each of these three charactersis highly questionable, and not considered probative by many other specialists. It is no widespread view that Indo-Iranic is most closely related to Balto-Slavic, or to any other particular branch (155).

One of Ringe’s three characters is cognacy in the meaning ‘all (plural)’. Cognacy between Indo-Iranic and Balto-Slavic languages is only partial here, however. Their ‘all’ words share only the morpheme *wi, followed by differentextensions: Indo-Iranic has *wikw􏰀 o, Balto-Slavic has *wiso. Ringe opts to code both as state 5 (differentiated onlyby his 5a and 5b labels), rather than as different states 5 and 6. Furthermore, *wi reconstructs to Proto-Indo-European, so its use in this meaning may be a shared retention or a parallel semantic shift, and does not prove aunique, late [Indo-Iranic + Balto-Slavic] node.

The other two characters have been long debated in Indo-European research: one aspect of the centum/satemcontrast, and the ‘ruki rule’. Ringe does not follow the basic contrast between centum and satem languages. Hedefines his character P2 narrowly, as “full ‘satem’ development of dorsals”, i.e. in such a way as to apply not to allsatem languages, but only to Indo-Iranic and Balto-Slavic (and to explicitly exclude Armenian and Albanian: p. 104in (50)). This is taken as a definitive character, which specifically separates [Indo-Iranic + Balto-Slavic] from [Greek+ Armenian]. So in effect, Ringe’s preferred definition separates [Indo-Iranic + Balto-Slavic], to set the satemlanguage Armenian together with the non-satem Greek instead. (This is because the state for P2 full satem is absentin both, even though basic satem is present in Armenian and absent in Greek.) This contradicts and overrides thebasic centum/satem contrast, then, which is not used as a character in the Ringe data‐set. There is long-standingdebate on this issue, and a range of alternative hypotheses on how the patterns in the language data arose. Ringe’schoice of definition is subjective, as is his confidence that that particular definition can be taken as a probativephylogenetic character — necessarily inherited, rather than a parallel development or areal phenomenon — whileother definitions cannot. This interpretation does not enjoy consensus in the field.

The ‘ruki’ rule (Ringe’s character P3) is a sound change named for the phonological context that triggers it, namelythe set of sounds /r/, /u/, /k/ and /i/ (or in fact not just /k/, but also other dorsal stops). The sound change is thatoriginal Proto-Indo-European */s/ came to be articulated slightly further back in the mouth (‘retracted’), as /ʃ/instead. Cross-linguistically, /s/ > /ʃ/ is in itself a frequent sound change, but what most Indo-Iranic and Balto-Slavic languages share is that this change generally applied also in the same set of contexts, namely where */s/followed immediately after any of the ‘ruki’ set of sounds. Some analysts deem the /i/ and /u/ vowels, and theconsonants /r/ and dorsal stops (like /k/), to form a particular combination of contexts idiosyncratic and unusualenough that it can be considered unlikely that the same sound change in these same contexts could have arisenindependently twice. These analysts take it as more plausible that the retraction change arose only once, that is,along a single branch to a common ancestor of [Indo-Iranic + Balto-Slavic]. This is Ringe’s position.

However, /i/ and /u/ have in common that they are the two highest cardinal vowels, where the tongue comesclosest to the palate and velum respectively. And /k/ is also formed at the velum, by the back of the tongue actuallymaking full closure there. Other ‘dorsal’ stops are also defined as being articulated by the back of the tongue (the‘dorsum’) against the palate, velum or even further back at the uvula. As for /r/-sounds, in phonology these form anotoriously diverse set, including sounds articulated from palatal to uvular as the location of stricture. The precisephonetic realization of the /r/ phoneme in early Indo-European branches is uncertain. So in fact, /r/, /u/, /k/ and/i/ all lie on the ‘close’ and ‘back’ limits to the vowel space, where it maps onto the corresponding places ofarticulation for consonants (see Fig. 10.6 in (156), and (157)). Nor is the mixed set of vowel and consonant triggersso unusual, particularly given that in Proto-Indo-European /i/ and /u/ functioned largely as (sonorants) in anycase, i.e. as consonants. The ‘ruki’ contexts all have in common a location of stricture that can naturally draw thearticulation of a preceding /s/ sound to be retracted to /ʃ/, closer where dorsal stops and high vowels arethemselves articulated with the tongue (i.e. partial assimilation in place of articulation).

The ‘ruki’ set of contexts is thus not phonetically unmotivated or unusual for triggering a retraction of /s/. Otheranalysts consider the outputs to reflect a set of similar retraction changes that arose independently in parallel,rather than a single rule operating just once in a putative common Proto-Indo-Iranic-Balto-Slavic. There aremultiple indications that somewhat different, separate changes did indeed arise. According to some analyses, the‘ruki’ retraction of */s/ may also have applied partly in Albanian and to individual words in Armenian.Furthermore, the retraction did not operate identically or universally across Balto-Slavic and Indo-Iranic in anycase. Ringe himself recognizes several complications (p. 109 in (50)), while other authors discuss details andexceptions in (158) for Indic and in (159) for Iranic, for example.

Alternative analyses are possible because various more recent sound changes have obscured the evidence of whatmay or may not have happened at earlier stages. In many languages, original Proto-Indo-European */s/ in ‘ruki’contexts is found as other sounds, not /ʃ/, and it is not always knowable whether an original */s/ > /ʃ/ retractionhad or had not already happened first. The retraction is observable in Lithuanian, for instance, but not in otherlanguages of the Baltic branch, so it is not clear whether the change had operated by the Proto-Balto-Slavic stage. InSlavic, the retracted /ʃ/ shifted again to /x/, whereas in Indic /ʃ/ shifted to contrastive retroflex /ʂ/. In Nuristani,Morgenstierne first noted that although original /s/ is often found changed to /ʂ/ as in Indic languages, this doesnot apply after the /u/ context, where original /s/ remained (160). As examples from Morgenstierne (withcorrections for Nuristani added here), in words for ‘mouse’ derived from Proto-Indo-European *múh2s with */s/,Indic languages have changed that /s/ ultimately to retroflex /ʂ/ (spelt in Indic as dotted ṣ), as for example in Vedicmūṣ- or Gawarbati muṣa. The cognate word in Nuristani languages, however, has /s/, without retraction to /ʂ/:Kâmviri and Kâta-vari both have /musˈa/ (realized phonetically as [muzˈɨ] and [musˈɨ] respectively), and Vâsi-varihas /müs/. Also in Nuristani, both after and before the /i/ context, original */s/ is found as /ʃ/, but through arelatively late sound change. For these /i/ contexts, strictly it cannot be known whether original ‘ruki’ changes hadalready applied, i.e. in a chain of /s/ > /ʃ/ > ‘Indic’ /ʂ/ and then back to Nuristani /ʃ/, but this raises questions ofparsimony and phonological motivation.

To be consistent with ‘ruki’ as evidence for an Indo-Iranic + Balto-Slavic node, one has to entertain the idea thatNuristani languages did actually inherit all the results of a full original ‘ruki’ change, but then reversed some of

those results by later sound changes: e.g. where found between vowels, /ʂ/ changed (‘back’) to /s/. Although thishas been hypothesized (161), the analysis includes faulty data on Nuristani languages, and proposes a sound changewhich — although convenient for defending a sweeping ruki rule — is unmotivated phonologically. It is unclearwhat phonetic preconditions should have caused such a reversal, nor why other instances of intervocalic *ṣ did nottherefore also become *s in Nuristani. For a more phonologically plausible alternative analysis of Nuristani “non-ruki *u”, not consistent with ruki as validly demonstrating an Indo-Iranic + Balto-Slavic node, see §4.2 in (162).

In sum, it is highly controversial whether the ‘ruki’ sound change is a genuine, single shared innovation (at aputative common Proto-Indo-Iranic-Balto-Slavic stage); many specialists consider it a series of independentparallel changes instead.

So the only three characters in (50) that ostensibly support a unique [Indo-Iranic + Balto-Slavic] node are all highlyproblematic, much debated in Indo-European linguistics, and not widely accepted as probative that the node didactually exist. Ringe himself warns that overall, even his preferred tree is supported by very few characters, andcontradicted by more: “18 characters are incompatible with the ‘best’ tree returned ... in computational terms ourresult is a total failure” (p. 85 in (50)). Ringe concedes that “the higher-order subgrouping of the lE family hasremained an unsolved problem for so many generations partly because the evidence is genuinely meagre” (p. 98 in(50)). In his own analysis, “the evidence for virtually all the larger, non-traditional subgroups that our algorithmposits is so slender” (p. 104 in (50)): no more than a handful of binary characters for any higher node. This is asreflected in the very low (>0.3) support values for the first three splits in our results here, and only 0.15 support for[Indo-Iranic + Balto-Slavic]. We take the linguistic data — qualitative as well as quantitative — as far fromsupporting such a node, especially not as the last major node to break up in the model for the Steppe hypothesis setout in (5), and when taken to have remained intact until less than a millennium before Vedic. We therefore alsotake it as premature for interpretations of aDNA data to presume the Ringe tree and its chronology as if in supportof a late presence of Indo-Iranic on the Steppe, before a late entry into India and Iran, in order to fit the Steppehypothesis.
This claims (being accepted in a peer-reviewed paper) illustrates the scientific debate about Balto-Slavic and Indo-Iranic sharing a recent node. Clearly showing that the question is not settle and is an open topic of discussion in the scientific community.

Anyone having issues with this content should publish a paper in a peer-reviewed journal to "help" the scientific community to settle the question (this is how science progress, not with "cherry-picking" biaised claims made by randoms on an internet forum). Such paper should adress the arguments of Heggarty et al. 2023 as a recent major paper on the topic.

Anyone, falsely trying to present this question has being solved (as of early 2024), is on purpuse trying to lure readers and is performing an intellectual fraud.
 
Recent peer reviewed paper Heggarty et al. 2023 (suppl.) said (about Indo-Iranic and Balto-Slavis sharing a recent node) :
REF : https://www.science.org/doi/10.1126/science.abg0818

This claims (being accepted in a peer-reviewed paper) illustrates the scientific debate about Balto-Slavic and Indo-Iranic sharing a recent node. Clearly showing that the question is not settle and is an open topic of discussion in the scientific community.

Anyone having issues with this content should publish a paper in a peer-reviewed journal to "help" the scientific community to settle the question (this is how science progress, not with "cherry-picking" biaised claims made by randoms on an internet forum). Such paper should adress the arguments of Heggarty et al. 2023 as a recent major paper on the topic.

Anyone, falsely trying to present this question has being solved (as of early 2024), is on purpuse trying to lure readers and is performing an intellectual fraud.
There is not a single archaeogenetic sample(y dna Indo-European candidate) which would prove migration of someone from the area of Anatolia, whether towards the east or the west, and whose bearers would be speakers or transmitters of Indo-European language. I already mentioned that.

" South Asian farmers represented in the IVC Cline had negligible ancestry related to ancient Anatolian farmers as well as an Iranian-related ancestry component distinct from sampled ancient farmers and herders in Iran. Since language proxy spreads in pre-state societies are often accompanied by large-scale movements of people (Bellwood, 2013), these results argue against the model (Heggarty, 2019) of a trans-Iranian-plateau route for Indo-European language spread into South Asia."
An Ancient Harappan Genome Lacks Ancestry from Steppe Pastoralists or Iranian Farmers (2019)
This is information for the eastern direction, I don't know what we have for the western direction, it is possible that it is in the scientific works concerning the Albanians. What we know about Albanian R1b branches is that the main branch BY611 or Z2103 does not show archaeogenetic ancestral mutations in Anatolia but in Central Europe.

 
Anyone having issues with this content should publish a paper in a peer-reviewed journal to "help" the scientific community to settle the question (this is how science progress, not with "cherry-picking" biaised claims made by randoms on an internet forum). Such paper should adress the arguments of Heggarty et al. 2023 as a recent major paper on the topic.

On top of the satetement above, any person trying to ignore the shared CHG ancestry of WSH and Indus Valley periphery is likely only pushing for a given model only for ideological reasons.

PS: language diffusion didn't require injection of DNA and can be achieved by a "political/military domination". Anyone trying to deny this statement is confusing DNA and culture ...

PPS : Cultural diffusion of Copper Age mettalurgy is known to have occured with dates consistent with the Heggarty et al. 2023 model, showing that this timing is decent. If cultural innovation spread, language could have been carried aswell. REF : Nikolsky et al. 2020 (page 29) https://psycnet.apa.org/record/2020-15982-001
 
Last edited:
The main issue, that make me stick to my position that the question will be hard to settle (if even possible), is that diffusing a language didn't require a significant injection of DNA (and on the opposite significant DNA-injection didn't insure that a language will be diffused).
We have many documented exemples.
Thus there is room for a CHG model, but I think a WSH model is also not out of the picture.
There are some great scholars who strongly believe Indo-Iranian culture existed in the Iranian Plateau in the 3rd millennium BC , for example look at this article by Prof. Massimo Vidale: Protohistory of the vara. Exploring the Proto-Indo-Iranian Background of an Early Mytheme of the Iranian Plateau
 
There are some great scholars who strongly believe Indo-Iranian culture existed in the Iranian Plateau in the 3rd millennium BC , for example look at this article by Prof. Massimo Vidale: Protohistory of the vara. Exploring the Proto-Indo-Iranian Background of an Early Mytheme of the Iranian Plateau
I completely agree that IE finding its root around modern ~Iran is a possibility. But I also stay open to other solutions, because the WSH signal is also uniting nearly all IE-speaking cultures. Neglecting this possibility wouldn't sound serious to me.

Just to bring some "water" to your PoV on the topic :

Let look at the J2 subclade-cross-correlation for Iran/Caucasus versus India/Pakistan (personal computation based on YFULL v11.05),
Capture-d-e-cran-2024-01-14-a-22-55-39.png

It was to be expected, the main decoupling time (look at the black curve massive drop) for India/Pakistan J2s versus Caucasian/Iranian J2s is around ~3500 BCE. Some earlier signal can also be seen around ~13000 BCE, that could be related with the main source of autosomal DNA observed in the IVC-cline derived from Iranian-HG (Shinde et al. 2019).

As a remainder, here is the copper age diffusion map from Nikolsky et al. 2020, the key information to observe is the date at which copper age is reaching India/Pakistan :
Capture-d-e-cran-2023-08-02-a-16-18-48.png


Interestingly, Heggarty et al. 2023 are dating the Indic/Iranic node to around 3500 BCE ... if they didn't "cheated", their model have everything : cultural-elements diffusion & DNA-connection. Yet it won't "prove" that IE-languages took this path, but it could have.
Therefore we have clades dispersal matching the Ringe-tree topology but we also have clades dispersal matching with the Heggarty-topology.

On the other hand, if we look at the European side :
In the Heggarty et al. 2023 model, a potential lineage involved in the diffusion of early IE would be J-CTS1026 (J1). It have the potential to have been injected toward Europe as early as ~5000 BCE and to have been injected north of the Caucasus around 4500 BCE or few centuries before.
A close lineage have been found among Khvalynsk culture around 4500 BCE.

Let check again clade-cross-correlation but for J1 this time, looking at Caucasus-Anatolia vs South-Eastern-Europe,
Capture-d-e-cran-2024-01-14-a-23-32-36.png

Some stuff (decoupling of clades) is hapening between 5000 and 3500 BCE.
It is also quite well fitting with Heggarty et al. 2023 nodes for Balto-Slavic/Italo-Germano-Celtic around 5000 BCE and for Albanian/Graeco-Armenian around 5000 BCE. It is also interesting to note the Graeco-Armenian node around 3500 BCE.
We do have J2 that shows a small ~3500 BCE Caucasian/Europe decoupling.

To track the early stages, clade-cross-correlation might not be well suited due to the limited concerned geographical area. In such case I perfer to use global diversification rate as a proxy, and J1 and J2 are entering significant diversification events around ~6000 and ~6500 BCE respectively.

With a ~CHG homeland of IE, do we observe genetic connections on the Y-tree that occur at the good "time" and between the good locations to fit the model proposed by Heggarty et al. 2023 ? Yes, but involving J haplogroup.
Such connection is even backuped via ancient DNA in the Steppe.

The key point is the timing and the topology of the linguistic tree, but there is no consensus on the topic.
Because with a lower chronology of the tree, and a slightly different topology, you can also find genetic conections that would support a WSH-model.

Thus, I prefer to stay carefull, because timing language divergence is clearly not a simple task ... and more and more I'm reading on the topic, less and less convinced that there is (and will be) enough data to definitevely settle the question.
 
A great news from India: https://scroll.in/latest/1066223/nc...k-to-cast-doubts-about-aryan-migration-theory

NCERT revises Class 12 history textbook to cast doubts about Aryan migration theory​

The National Council for Education Research and Training has revised a chapter in its Class 12 history textbook for the Central Board of Secondary Education to cast doubts on the theory of Aryan migration, instead emphasising that Harappans are the “indigenous people” of the Indus Valley, reported The Indian Express.

The theory of Aryan migration was first put forward by Western scholars during the colonial age. The theory postulated that a race of European or Central Asian “Aryans” swept into the Indian subcontinent, displacing the indigenous Indus Valley Civilisation.
 
Back
Top