What are the ancient origins of Kurdish DNA? The deep ancestry of the Zagros, northern Mesopotamia and eastern Anatolia: Neolithic Zagros farmer-related ancestry blended with Anatolian and Caucasus-related streams, with modest later additions. No ancient transect ties specific ancient individuals to specific Kurdish communities. Ancestrify models each era from your raw DNA.
Key facts
Figures as published, one source per row.
| The Zagros foundation, about 8000 BCE | The early farmers of Ganj Dareh, in today's Kermanshah province inside the Kurdish cultural zone, were a separate people from the Anatolian farmers who colonised Europe, related instead to the Caucasus hunter-gatherers.Source: Early Neolithic genomes from the eastern Fertile Crescent (2016), Broushaki et al. |
|---|---|
| Why the layer is local | The region's founding farmer ancestry is not an import. It remains the single largest deep component in Kurdish genomes in every modelling exercise.Source: Genomic insights into the origin of farming in the ancient Near East (2016), Lazaridis et al. |
| Bronze and Iron Age | Kura-Araxes, Hurrian-era and Urartian-era samples on the northern rim are recombinations of the same local streams, with only limited steppe-related ancestry reaching south of the Araxes.Source: The genetic history of the Southern Arc: A bridge between West Asia and Europe (2022), Lazaridis et al. |
| The Iranian languages | Their arrival is conventionally tied to Iron Age movements from the east, and their genetic footprint in the highlands appears modest. Language spread here with little gene flow. |
| Where modern samples sit | Nearest their geographic neighbours: Iranians, Assyrians, eastern Anatolians, Armenians and Azeris, with the proportions of the three deep streams varying along the mountain arc. |
| The sampling limit | No published ancient transect runs through Kurdish-majority regions with the density that Anatolia or the Levant enjoy, so continuity here is inferred from neighbouring transects rather than measured directly. |
| What Ancestrify reports | An era-by-era qpAdm model of your raw DNA against these ancient sources, published with its p-value, every source's standard error and the full right set, from from €29.99.See Ancient Ancestry |
Kurds — some 30–40 million people across the mountain arc where Turkey, Iraq, Iran and Syria meet — live almost exactly where farming, herding and some of the deepest ancestry streams of West Eurasia were forged. That geography writes the genetic story in advance more than most national narratives would like: Kurdish ancestry is, to first order, the ancestry of the Zagros and its foothills, held with notable continuity. Here is what the published genetics actually supports, and where its limits are.
The short answer: Kurdish genomes are dominated by the deep ancestry of the Zagros–northern Mesopotamia–eastern Anatolia triangle — Neolithic Zagros farmer-related ancestry blended with Anatolian- and Caucasus-related streams — with modest later additions. Modern samples place Kurds nearest their geographic neighbours (Iranians, Assyrians, eastern Anatolians, Armenians and Azeris), and no published ancient transect yet ties specific ancient individuals to specific modern Kurdish communities.
The Zagros foundation#
In 2016, genomes from the eastern Fertile Crescent rewrote the Neolithic map: the early farmers of the Zagros (Ganj Dareh, ~8000 BCE, in today's Kermanshah province — inside the Kurdish cultural zone) were a separate people from the Anatolian farmers who colonised Europe — strongly differentiated, related instead to the Caucasus hunter-gatherers. This Zagros/Iranian Neolithic-related ancestry became one of West Eurasia's great streams: it expanded west across the Iranian plateau and Mesopotamia, east toward South Asia, and north into the Bronze Age Caucasus.
For Kurdish ancestry the point is direct: the region's founding farmer layer is not an import — it is local, it is old, and it remains the single largest deep component in Kurdish genomes in every modelling exercise, joined by Anatolian farmer-related ancestry from the west and CHG-related ancestry from the north in proportions that vary along the mountain arc.
The layers that followed#
The Bronze and Iron Ages stirred the triangle without replacing it. The Southern Arc dataset (Lazaridis et al. 2022) shows the region's ancient populations — Kura-Araxes, Hurrian-era, Urartian-era samples on the northern rim — as recombinations of the same local streams, with only limited steppe-related ancestry reaching south of the Araxes, in sharp contrast to Europe's steppe transformation. The arrival of Iranian languages — Kurdish among their descendants — is conventionally tied to Iron Age movements from the east; their genetic footprint in the highlands appears modest, a reminder that language can spread with little gene flow.
The historical era added texture rather than turnover: Arab, Turkic and regional gene flow at the edges, strongest where geography is open, weakest in the mountain cores — the standard highland pattern of retention.
What modern Kurdish samples show#
Genome-wide studies that include Kurds — regional medical genetics, the Southern Arc's modern panel, and community datasets — agree on the essentials. Kurds cluster with their neighbours in the Zagros–eastern Anatolia block, nearest Iranians, Assyrians, Armenians and Azeris — genetic distance in the triangle tracks geography more than religion or language. Internal structure follows the arc: Kurmanji, Sorani and southern communities differ modestly, along a cline rather than across a boundary, with Zaza and Gorani speakers sitting inside the same regional variation. Y-DNA is the region's old mix — J2 and J1, E-M34, R1b of local (non-European) branches, T, with R1a present at moderate frequency — no lineage is a "Kurdish gene", and none supports an exotic-origin story. Yazidi communities, studied for their strict endogamy, show the expected tightening of the same regional profile rather than a different one.
What does not exist yet: a published ancient-DNA transect from the Kurdish heartland — Hasankeyf, Erbil citadel, the Zagros valleys — connecting dated ancient individuals to modern communities the way Balkan or Iberian transects now do. Continuity is the parsimonious reading of regional data; it has not been tested locally.
Reading your own results#
- G25 distances: expect modern-era neighbours to lead (Kurdish, Iranian, Assyrian, eastern Anatolian averages), and ancient eras to lead with Zagros- and Caucasus-related samples. Era discipline matters: "closest to Ganj Dareh's descendants" and "closest to modern Iranians" are different, equally true statements.
- Admixture fits: a defensible West Asian panel models a Kurdish coordinate mostly from Zagros/Iranian-, Anatolian- and CHG-related sources. Because those three streams are geometrically close, collinearity is the standing trap — distrust any tool that splits them to the decimal with confidence.
- qpAdm: the formal instrument for "is this source required" questions — with the regional caveat that closely related candidates (Zagros vs CHG vs eastern Anatolian) are exactly the case where the right set decides what can be resolved, and where an honest model record sometimes reports that the data cannot choose. That, too, is a result.
Frequently asked questions#
Are Kurds descended from the Medes?#
Genetics cannot test dynastic or tribal labels. What it supports: continuity with the Iron Age populations of the Zagros region generally, of which the Medes were one named part. No sampled "Mede genome" exists to anchor anything sharper.
Are Kurds closer to Iranians or to Turks?#
To Iranians and the rest of the Zagros–eastern Anatolia neighbourhood. Anatolian Turkish genomes themselves are mostly old Anatolian ancestry with a minority Central Asian layer — so the gap is smaller than the political framing suggests, but the Kurdish cluster sits clearly on the plateau side.
How much steppe ancestry do Kurds have?#
Modest — the highlands south of the Caucasus received only limited steppe-related ancestry, and models of Kurdish genomes typically need far less of it than any European population. A large steppe percentage from a hobby calculator is usually a panel artefact.
Terms used here are defined in the glossary.
Sources and further reading#
- Broushaki, F. et al. (2016). Early Neolithic genomes from the eastern Fertile Crescent. Science, 353, 499–503.
- Lazaridis, I. et al. (2016). Genomic insights into the origin of farming in the ancient Near East. Nature, 536, 419–424.
- Lazaridis, I. et al. (2022). The genetic history of the Southern Arc. Science, 377, eabm4247.
- Narasimhan, V. M. et al. (2019). The formation of human populations in South and Central Asia. Science, 365, eaat7487.
