The Genetic Origin of the Kurds: A Comprehensive Haplogroup-Based Analysis

1. Introduction

The genetic origin of a people should be read not as a single "pure lineage" but as a stratification of ancestral populations arriving in the region at different times. Haplogroups are the most powerful tools for tracking these layers: Y-DNA is passed only from father to son, mtDNA only from mother to child, and neither undergoes recombination. This allows uninterrupted lineages reaching back thousands of years to be traced.

Geographically, Kurds live in the Zagros–Taurus mountain belt — the eastern wing of the Fertile Crescent, where agriculture first emerged. This location is decisive in understanding their genetic depth: the region hosted some of the oldest settled farming populations in human history.

2. Paternal (Y-DNA) Haplogroups

2.1. General Table

A Y-STR study of 162 Sorani Kurds in Sulaymaniyah gives the following paternal distribution:

Source: BMC Genomics (2022), Sorani sample. Figures are from a single study; they vary by region and tribe.

2.2. Neolithic Farmer Lineages: J2, G2a, E1b1b

The largest portion of the Kurdish paternal line consists of Neolithic farmer lineages. J2 (especially J2a-M410) and G2a are classic "first farmer" lineages directly associated with the spread of agriculture from the Fertile Crescent. These two haplogroups are also dominant in the oldest Neolithic skeletons of the Zagros and Anatolia. E1b1b-M35 is a lineage that spread via the East Mediterranean, likewise linked to the Neolithic.

In a different regional study (Kurdistan-wide), values such as J2-M172 19.9%, J1-M267 11.2%, and G-M201 2% confirm this Neolithic weight. Even when the sample changes, the dominance of farmer lineages remains constant.

2.3. The Semitic Layer: J1-M267

J1-M267 is a pastoral-nomadic lineage typically associated with the Arabian Peninsula and Semitic peoples. Its presence in Kurds at around 11–14% reflects the region's millennia-old Mesopotamian neighbourhood and continuous genetic exchange with Semitic peoples (Assyrian, Arab).

2.4. The Iranic / Steppe Layer: R1a-Z93

The genetic signature of the language-bringing layer in Kurdish ethnogenesis is R1a — but the detail is critical. The R1a in Kurds is not the R1a-Z282 common in Europe, but predominantly the R1a-Z93 (and its subclade Z94/Z2124) lineage. This subclade is precisely the one associated with the Indo-Iranian language carriers descending south from the Central Asian steppe. In tribe-based tests (FTDNA), subclades R1a-Z94>Z2124>Z2125 have been documented in Kurmanji and Sorani samples from regions such as Dêrsim, Yerevan, and Kirkuk.

Important point: the R1a rate in Kurds varies between 7–20% (Kurmanji groups, especially in Turkey and the Caucasus, tend to be slightly higher than Sorani). Thus, although the Iranic migration layer is a minority in genetics, it is the decisive layer that brought the language and the ethnic name. This is the clearest evidence of the "language–gene divergence" phenomenon.

2.5. Kurmanji – Sorani Differences

Subgroup comparisons show that E1b1b and J1 are somewhat more prominent in Sorani groups, while R1a and R1b rates are slightly higher in Kurmanji groups (especially in the Caucasus and Turkey). Some Kurmanji communities in the Caucasus show a "bottleneck" effect and low genetic diversity due to historical isolation. This illustrates how migration and isolation can alter haplogroup distributions in small groups.

3. Maternal (mtDNA) Haplogroups

3.1. General Table

A comprehensive mtDNA study of Iranian Kurds characterizes the maternal pool by the dominance of Western Eurasian haplogroups:

Source: Journal of Genetics (2016), Iranian Kurds. In the Iraqi (Sorani) sample the order differs slightly: U 22.5%, H 19.2%, J 14.5%, HV 12.9%.

3.2. Interpretation: Ancient Indigenous Layer and European Affinity

Three things stand out in the maternal line. First, the dominance of haplogroup H (37%) — the deepest and most widespread Western Eurasian maternal lineage, pointing to both Palaeolithic and Neolithic roots. Second, the high frequency of haplogroup U is particularly meaningful: U carries the trace of the region's oldest hunter-gatherers and supports the thesis that the Kurds' ancestors also assimilated and admixed with pre-Neolithic hunter-gatherers. Third, some studies (Comas et al.) found that the mtDNA lineages of Kurmanji groups show typical affinity with the European gene pool — stemming from the shared Neolithic maternal heritage of the Near East and Europe.

3.3. East Eurasian Traces

East Eurasian lineages are quite low in Kurdish groups (total ~2%). Small fractions such as M1a (6.66%) in Iraqi Kurds and G3b (4.76%) in the Hawrami group appear. This low rate shows that Kurds are predominantly of Western Eurasian origin and that the eastern (steppe) influence remained limited on the maternal side — meaning the steppe migration was most likely a male-driven one.

3.4. Time of Demographic Expansion

mtDNA-based demographic analyses date the Kurdish expansion to roughly 9,500 years ago, linking it to the Neolithic transition in the Near East (from hunting-gathering to farming and animal husbandry). This date agrees with archaeological data and concretizes the statement that the Kurdish root is "as old as the soil."

4. Autosomal (Genome-Wide) Evidence

While haplogroups trace a single lineage, autosomal DNA gives the total ancestral admixture. qpAdm models successfully explain modern Kurds with the components: Zagros/Iran Neolithic farmers + Anatolian Neolithic + Levant Neolithic + Iron Age Scythian/Sarmatian (steppe). In one single-sample model, values such as Iran Neolithic ~44%, Anatolian Neolithic ~33%, and Levant Neolithic ~4% were found.

The general tendency is that modern Kurds carry about 75–80% indigenous (Neolithic + earlier) component, with the remainder being the steppe/Iranic layer. In the Bronze Age Hajji Firuz Tepe sample, a near 50-50 mixture of Zagros natives and Pontic-Caspian steppe migrants was modeled.

5. Language–Gene Divergence

All this data points to a single fact: Kurds took their language from Iranic migrants but preserved their genetic weight from the region's most ancient indigenous people. While the R1a-Z93 lineage (language carrier) is a minority in genetics, the J2/G2a/E1b1b (indigenous farmer) and mtDNA H/U (ancient maternal) layers are dominant. Linguistically, Kurdish rests on a Parthian base with a Median substratum, bearing eastern/central Iranic features.

6. Limitations

All the percentages given are not definitive: sample size, geographic region (Sorani/Kurmanji/Zaza/Feyli), tribal structure, and haplogroup prediction method change results significantly. Furthermore, claims of direct ancestral links with ancient peoples such as the Gutians and Lullubi remain speculative, since no language or DNA data exist for these peoples. FTDNA/community data are not of scientific sampling standard and should be read as indications of tendency.

7. Conclusion

Haplogroup-based data clearly reveal that Kurds have a multi-layered ethnogenesis: a dominant ancient Zagros/Neolithic indigenous core (J2-G2a-E1b1b in the paternal line, H-U-K in the maternal line), an Iranic/steppe layer laid over it (R1a-Z93), and ongoing admixture with neighbouring Mesopotamian, Caucasian and Anatolian peoples. This picture positions the Kurds as both an ancient and indigenous people and an admixed one — and these two qualities do not contradict each other.

(References: same as the Turkish version — BMC Genomics 2022; Journal of Genetics 2016; Egyptian Journal of Forensic Sciences 2021; Broushaki et al., Science 2016; Rojava Journal 2026; FTDNA community data; Eurasian DNA 2023.)