Neolithic ancestry in South Asia

If R1a-Z93 tells the story of the Bronze Age pastoralists who brought Indo-European languages to South Asia, haplogroup L tells an older story — one rooted in the Neolithic farmers who spread agriculture from the Iranian plateau into the Indus Valley thousands of years before the Steppe migrations began.^[1]^[2]

Deep origins

Haplogroup L (L-M20) is a Y-chromosome lineage that originated in West Asia roughly 30,000 years ago. Its highest concentration today is along the Indus River in Pakistan, where the Indus Valley Civilization flourished between 3300 and 1300 BCE — a coincidence that has led geneticists to hypothesise that L-M20 may be one of the paternal lineages of the civilisation's creators.^[4]^[5]

Time estimates for L-M20 place its expansion at approximately 26,000 years ago in Pakistan, with subsequent spread into India during the Late Glacial Maximum recovery period. The haplogroup's diversity is highest in Pakistan and declines toward both Iran and southern India, suggesting a central Asian cradle of diversification.^[4]

The L1a2-M357 branch

Haplogroup L divides into several sub-branches. The most relevant to Afghanistan and the wider region is L1a2-M357 (also defined by marker L1307), which a 2024 study traced to a West Asian origin approximately 8,600 years ago.^[1]

From this West Asian root, two population groups emerged. One remained in West Asia and diversified into lineages found today among the Chechens and Ingushes of the Northeast Caucasus, as well as populations in Anatolia and the Armenian Highlands. The other migrated to South Asia between 8,000 and 6,000 years ago, where it diversified into the lineages found today in Pakistan, northern India, and Afghanistan.^[1]

An ancient DNA confirmation came from Shahr-i-Sokhta, an Early Bronze Age site in southeastern Iran, where a male individual dating to approximately 4,900 years ago was assigned to the L1a2-M357 branch. His genome-wide ancestry was consistent with populations related to the Iranian plateau — a direct link between the modern L1a2 carriers and the ancient farmers of West Asia.^[1]

Afghanistan at the interface

Afghanistan sits at the geographic and genetic interface between the West Asian and South Asian branches of haplogroup L. The frequency of L-M357 varies across the region's ethnic groups, reflecting the layered demographic history of the Hindu Kush.^[2]^[3]

Among Afghan Pashtuns, L lineages account for roughly 7–9% of paternal lineages, with L1a2-M357 making up the majority. The frequency is higher in northern Pashtun populations (around 20% carry L-M20) than in the south (around 5%), suggesting differential assimilation of the lineage across the Hindu Kush range.^[3]

The Burusho people of Hunza and the Kalash of Chitral — isolated populations of uncertain linguistic affiliation — carry the highest frequencies of L1a2-M357 in the region, at approximately 12–15% and 25% respectively. These groups may preserve a signal of the ancient Neolithic population that once inhabited the mountain valleys before the arrival of Steppe pastoralists.^[2]

The agricultural connection

The distribution of haplogroup L correlates strongly with the spread of Neolithic agriculture from the Iranian plateau into South Asia. A 2024 study of Y-chromosome variation across South Asian populations found that traditional farming communities carry significantly higher frequencies of L-M27/M76 than pastoralist or hunter-gatherer groups, suggesting that the lineage was carried by the early agriculturists who transformed the subcontinent's economy.^[6]

The same study proposed a link between L1-M22 lineages and the spread of Dravidian languages, with a possible connection to the ancient Elamite language spoken in southwestern Iran. If confirmed, this would mean that haplogroup L traces not only a genetic migration but a linguistic one — the movement of Dravidian-speaking farmers into the Indian subcontinent from the west, thousands of years before the Indo-Aryans arrived from the north.^[1]^[6]

Two layers of paternal ancestry

Afghanistan's paternal gene pool is shaped by at least two major layers. The older layer, represented by haplogroup L and its sub-branches, arrived with Neolithic farmers from West Asia between 8,000 and 6,000 years ago. The younger layer, represented by R1a-Z93, arrived with Bronze Age Steppe pastoralists between 2,000 and 1,500 years ago.^[1]^[2]

The proportions vary by group. Pashtuns carry both layers in significant quantities — roughly 50–65% R1a-Z93 and 7–9% L lineages — making them a population shaped by both the Neolithic and Bronze Age migrations. Tajiks, Hazaras, and Uzbeks carry different balances of the same two layers, reflecting their distinct paths of assimilation into the Afghan gene pool.^[2]^[3]