Grooved ware (ակոսավոր խեցեղեն), the missing piece of puzzle.

 Grooved Ware (ակոսավոր խեցեղեն), the missing piece of the puzzle

We have had many discussions in this group about the modern genetic profile of Armenians—how, where, and when it formed. Nareg Asatrian has frequently posted genetic qpAdm models showing the modern Armenian profile as a mixture of two populations. Similar models have been reproduced by other commentators and have started to appear in academic papers as well, most notably Hovhannisyan et al. 2025.

Although these models are statistically feasible and useful as rapid abstractions, the real historical events behind the formation of the modern genetic profile were more complex. Two key events are usually emphasized: the Middle Bronze Age migration from the Steppe and the Urartu expansion (840–600 BCE) with its policies of relocations. More than 1,500 years separate these two events, and it would be unusual if nothing remarkable occurred between them. In reality, a significant demographic event affected historic Armenia around 1200 BCE, known as the spread of Grooved Ware, also called Nairi Ware.

Based on the comprehensive study by Guido Guarducci, the oldest forms of Grooved Ware, which he calls Proto-Nairi, appear in the South Caucasus, mostly in Lchashen culture sites, and in northwestern Iran (1400–1300 BCE). The attached map is from Guarducci’s book. The red areas represent Proto-Nairi. The color coding can be misleading, as western Georgia is included even though it did not have Grooved Ware. The best way to interpret the map is to focus on the archaeological sites (dots) rather than the colors.

Around 1200 BCE, all the regions colored green suddenly witness the appearance of this new type of grooved pottery. Some scholars believe this event was associated with large migrations. However, there is considerable debate about the geographic origin of the migrants. Summarizing the opinions, they can be grouped into three categories:

1. Grooved Ware in the green zone came from the South Caucasus.

2. Grooved Ware in the green zone expanded from the Van basin.

3. The origins of Grooved Ware in the green zone were in the upper Euphrates, where a large number of sites with this type of pottery are found.

The pottery found in the red and green zones is not completely identical but shares important features. It is clear that the initial impulse came from the South Caucasus. However, a direct migration of Lchashen people into all the green lands would imply that modern Armenians should have higher Steppe ancestry than they do today. Another possibility is that the initial impulse from the South Caucasus affected groups living in the Van basin or upper Euphrates, who later expanded independently, spreading a genetic profile different from that of Lchashen. Such a scenario aligns better with the modern Armenian genetic profile.

Of course, later Urartian activities were also important in shaping the modern Armenian genetic profile. However, the Grooved Ware period is crucial for understanding the genetic history of the Highlands prior to the formation of Urartu. Unfortunately, the currently available ancient DNA is insufficient to favor any of the aforementioned theories, and, as usual, we will have to wait for new data.

If the aceramic nature of this site is confirmed then this will be the oldest Neolithic site in South Caucasus and probably in a wider region.

If the aceramic nature of this site is confirmed, it will be the oldest Neolithic site in the South Caucasus and probably in the wider region. Following the Lernagog 1 discovery, dated to the 7th millennium BCE, this would be the next major Neolithic site—most likely even more important than Lernagog 1.

https://www.facebook.com/share/p/1DBu1jBTtj/

The possibility that Sumerians originated from north Mesopotamia raise the probability that Proto-Sumerians had the occasion to contact the Fertile Crescent farmers who moved to South Caucasus and further north of Caucasus.

 The possibility that the Sumerians originated from northern Mesopotamia increases the likelihood that Proto-Sumerians had contact with Fertile Crescent farmers who moved into the South Caucasus and beyond. This could mean that some lexical parallels between Sumerian and Proto-Indo-European (PIE) are genuine. Aleksi Sahala from the Helsinki Institute summarized proposed parallels. Not all cases are of high quality, but some have strong chances of being real cognates.

Here are some examples:

Sum. gud, gu4 ‘ox, bull; cattle’ ~ PIE *gwou(s)- ‘cow; ox’; Hitt. *kuṷāu-;
Sanskrit: go; Greek: bous (βοῦς); Tocharian B: keŭ; Old Norse: *kú; Armenian: kov.
Cattle were domesticated in the core Fertile Crescent region of West Asia and spread to the South and North Caucasus. It is plausible that the term used by pastoralists in the Pontic and Caspian steppe came from northern Mesopotamia, as it did for the Sumerians. The Egyptian words ka ‘ox’ and kaut ‘cow’ may share this origin. Notably, ancient Egyptian farmers carried T1a1a and H2 haplotypes, which were also present among historic Armenian and South Caucasus farmers.

Sum. šáḫ(a) ‘pig; boar’ → Akkadian šaḫû ‘pig’; Ugaritic šeḫû ‘pig’ ~ PIE *suh₁- ‘swine’; Sanskrit: sūkara (सुक्र); Tocharian B: suwo; Latin: sūs; Gothic: swein.
Variants include šaḫ, šúḫ (ŠUBUR). The reading with is widely accepted and supported by Akkadian. Similar words are found in Kartvelian languages, e.g., GZ ešw- ‘wild boar, pig’. These likely share a prehistoric etymology. The Armenian xoz is not directly derived from PIE but may be related to the same areal term or borrowed from Parthian.

Sum. sí-sí ‘horse’ ↔? Akkadian sisium ‘horse’ ← Hurrian issi(a) ‘horse’ ~ PIE *h₁ekwos ‘horse’; Hittite: aśuwas; C. Luwian: a-aš-šu; Sanskrit: áśva (अश); PIE Anatolian: *aĉwa-; Latin: equus; Albanian: sasë; Armenian: eš.
The trajectory of this word is likely from PIE to Sumerian. The author conjectures Indo-Iranian mediation, but the Armenian eš / išoy, formerly meaning ‘horse’, is the best source for the Hurrian form (Petrosyan 2002). The Armenian term later underwent a semantic shift (Martirosyan 2009). Sumerians and Akkadians likely received the word from Hurrians. Most Sumerian attestations are from the Ur III period, after the Akkadian period.

Sum. urud(a) ‘copper’ → Akkadian erû ‘copper’ ~ PIE *h₁reudh-ó- ‘red’; Sanskrit: rudhira; Avestan: raoðita; Tocharian A: rtär; Greek: eruthros (ἐρυθρός); Lithuanian: raudonas; Gaulish: roudos; Old Norse: rjóðr.
This is another notable example. The semantic shift is probably from a color term to a metal, similar to argentum ‘silver’. This word is attested in Sumerian from the 4th millennium BCE. Copper use predates Mesopotamia in the Highlands. If the word for ‘red’ was originally an Indo-European term, the most likely period is the Chaff-Faced Ware culture (Late Chalcolithic, 4200–3500 BCE), a mixed population almost certainly including IE speakers.

For other linguistic parallels, see the link in the comments. Also relevant are studies on PIE–Semitic lexical parallels.

The exact branch of G2 found in Middle Bronze Age kurgan in Atskuri near Akhaltsikhe is now available. Its the G2-Y258395.

 The exact branch of G2 found in a Middle Bronze Age kurgan at Atskuri near Akhaltsikhe is now identified as G2-Y258395.

Today, it is a minor branch found among Armenians, Turks, and Georgians. The parallel branch PH1780 is predominantly Georgian and Kartvelian, except for the subbranch PH311, which is more cosmopolitan. A slightly more distant parallel branch, L1264, is mainly found in the northwest Caucasus among Abkhaz–Adyghean speakers.

Given that the center of gravity of G2-Y258395 is located in what is now Turkey, it is reasonable to assume that the original ancestor of this branch likely lived somewhere between the Ardahan and Erzurum regions, rather than at Atskuri itself. Ancient samples from northeastern Turkey are needed to confirm this.

All three of these branches are part of the larger haplotype G2-L1266, which is around 7,800 years old. It is now reasonable to conclude that this haplotype was a Neolithic lineage that moved from the south—probably from the Upper Euphrates region—into the western and northwestern Caucasus during the Neolithic period, where it is abundant today. Another possible route, not yet fully ruled out, is a migration via Daghestan to the North Caucasus and the Maykop culture, followed by a backflow to the South Caucasus through the western Caucasus. However, this latter scenario is less likely than the first.

More information: G2-Y258395 YFull tree

In Armenian legends modern Armenia is an important place where many patriarchs lived.

In Armenian legends, modern Armenia is presented as an important region where many patriarchs lived. Beginning with Aramaneak, the son of Hayk, many generations lived and ruled in the Araratian Plain and its surrounding areas. Aramayis, Shara, Amasya, Gelam, Harma, Sisak, and others left their names in local toponyms. It was also in the Araratian Plain that the legendary capital Armavir — which was also a real city — was located. This fact is used as an argument for the Etiuni theory, since Etiuni was likewise situated in what is now the Republic of Armenia.

However, Hayk himself, who represents the eponymous patriarch of the Hay people (Armenians), was primarily active near Lake Van. His main domain was in Hark in Taron, west of Lake Van. He battled Bel in Hayoc' Dzor, which was located southeast of Lake Van. At least two settlements associated with him — Haykashen and Haykaberd — are also situated near Lake Van. The latter, Haykaberd, is today known as Cavushtepe, from where several ancient DNA samples have been obtained.

Assuming that legendary memory is not an arbitrary construct, this could suggest that the regions around Lake Van were as important for the ethnogenesis of Armenians as modern Armenia.

We have already observed indications of this in the available ancient DNA. However, the existing data is not sufficient to reach definitive conclusions. In any case, this does not change the broader picture of Armenian origins derived from Middle Bronze Age migrations. Most likely, however, important details will emerge with new data from the Lake Van region.

Syunik particularity.

Syunik Particularity

Davidski has produced new G25 coordinates based on the Skourtanioti et al. (2025) paper, including the 25 new samples from Syunik. He removed noisy SNPs that were almost certainly the result of postmortem damage. After this correction, the PCA of the “Hellenistic” era Syunik samples appears more compact and less heterogeneous, while modern Syunik individuals plot within the Aghitu Hellenistic cluster.

The conclusions remain the same as in the previous post regarding the Aghitu cave samples.

Some samples that show a shift toward the Iranian Plateau are probably from the Achaemenid/Orontid period (600–300 BC) rather than the Hellenistic period (300 BC–1 AD). The shift toward the Iranian Plateau is even stronger in a Late Antique sample from Caucasian Albania, specifically from the Shamakhi region. Therefore, it may be argued that Iranian influence was probably stronger in the eastern parts of the South Caucasus than in other regions of historic Armenia and the Caucasus. However, this influence appears to have been transient in Armenia, given that modern Syunik Armenians do not show a strong shift toward the Iranian Plateau.

The reason why Syunik does not display an abrupt genetic change in the post-Urartian period — similar to what is observed in the Araratian Plain, Sevan, and Shirak regions — is probably that Urartu did not exercise strong control over Syunik and did not establish new cities there with relocated populations. As a result, the Etiuni genetic profile persisted longer, and the transition to the modern genetic profile was gradual.

A similar pattern of gradual change may have occurred in other parts of historic Armenia that did not experience strong control by the Urartian Empire. Examples include Artsakh, Utik, and parts of Gugark.

These data support the theory proposed in Petrosyan and Palyan (2023) regarding the post-Urartian genetic shift. We suggested that this shift was a consequence of Urartu’s population relocation policies.

Additionally, it is possible that after the fall of the Biainean dynasty in Van, the Biainean elite relocated to the Araratian Plain — a theory suggested by some scholars. However, this hypothesis requires additional data in order to be confirmed or falsified.

The Middle Bronze Age (2400-1500BCE) in South Caucasus

The Middle Bronze Age (2400–1500 BCE) in the South Caucasus

Now that the DNA files from the Skourtanioti et al. (2025) paper on Georgia and the South Caucasus are available, we can examine them more closely. The focus of this thread is the crucial period of the Middle Bronze Age (MBA), when new Y-DNA lineages of steppe origin (R1b-Z2103, I2a2b) appeared in the South Caucasus and throughout historic Armenia.

The MBA data from Georgia show that only eastern Georgia was affected by steppe migrants. The blue dots indicate sites where steppe ancestry and steppe-associated Y-DNA were present, while the red dots mark sites where they were mostly absent. The second chart presents the same pattern numerically. This distribution is possible only if steppe migrants crossed the Caucasus via Dagestan and the eastern Caucasian passes.

Theoretically, migration through the Central Caucasus is also possible; however, in that case, the migrants would likely have acquired additional CHG (Caucasus Hunter-Gatherer) ancestry, which is not observed. On the contrary, MBA samples from both Georgia and Armenia show an additional shift toward Anatolian farmers. This shift is possible only if the migrants crossed what is now Azerbaijan, where an Anatolian farmer–rich population had been present since at least the Late Chalcolithic and apparently persisted into the Early Bronze Age.

This scenario explains two key findings of the paper:

1. The reason why the average steppe ancestry is lower in MBA Georgia than in MBA Armenia. This difference is due to geographic heterogeneity. Western Georgia, which lacked or had very low levels of steppe ancestry, likely harbored a non–Indo-European population, almost certainly Kartvelian tribes.

2. The paper also notes an excessive shift toward Anatolia during the MBA. This can be explained by migration through regions inhabited by Anatolian-shifted populations. The most plausible candidate is the Leyla-Tepe culture in what is now the Republic of Azerbaijan, which may also explain the presence of haplogroup E1b in MBA Armenia.

P.S. Thanks to Tigran Sg for providing the labeled G25 coordinates and for first identifying this pattern.