Genetics and population history
Origins of the Pashtuns: tradition and evidence
How Pashtun origins are explained: the genealogy from Qais Abdur Rashid, the Bani Israel tradition, the Pactyes of Herodotus, the evidence of the Pashto language and what genetic studies can and cannot show.
Published 9 October 20265 sources
On this page
Tradition, hypothesis and evidence
Questions about where the Pashtuns came from have three different kinds of answer, and it helps to keep them apart. The first is tradition: genealogies and origin stories that were written down in Persian chronicles in the early seventeenth century. The second is hypothesis: attempts to connect the Pashtuns with peoples named in ancient Greek writing. The third is evidence that can be tested, mainly from the history of the Pashto language and, more recently, from population genetics. [1] [2] [4]
The genealogy from Qais Abdur Rashid
The best-known written account of Pashtun genealogy is the Makhzan-i Afghani (مخزن افغاني), compiled by Nimatullah of Herat, a chronicler at the court of the Mughal emperor Jahangir. Its English translator, Bernhard Dorn, dated the work to 1018 to 1020 AH, about 1609 to 1611 CE. [1]
In this account every Pashtun tribe descends from Qais, who was given the name Abdur Rashid and the title Pathan. Qais had three sons: Sarban, Bitan (Batni) and Ghurghusht. Sarban in turn had two sons, Sharkhbun and Kharshbun. The tribes Dorn calls the Kerranians, usually known as the Karlanri, are treated in the same section as the Ghurghusht line rather than as a fourth son's descendants. Dorn noted that the British writer Mountstuart Elphinstone had recorded a fourth son of Qais named Kurleh, a name Dorn said he found nowhere else. [1]
The genealogy also records marriages and changes of allegiance. Bitan's daughter, Bibi Matu, married Shah Husain, described as a prince from Ghor, and their sons were remembered as the ancestors of the Ghilzai and the Lodi. Because Shah Husain was not of Afghan descent, his line was called Mati, after its mother. In another story Sheerani, a son of Sharkhbun brought up by his maternal grandfather Kakar, vowed to count himself with the Ghurghusht line. [1]
The Bani Israel tradition
The Makhzan gives Qais a much longer ancestry. It traces his line back through Afghana, a grandson of King Talut (Saul), to Yakub Israel (Jacob), Ibrahim (Abraham) and Adam. In its story, the descendants of Afghana settled in the mountains around Ghor after the Israelites were driven out by Bukhtnasar (Nebuchadnezzar). Much later, Qais led a party of Afghan chiefs to Medina, accepted Islam, and was given the name Abdur Rashid by the Prophet. [1]
This account is usually called the Bani Israel (بني اسرائيل) tradition. It should be read as tradition rather than as a record that can be checked: it was compiled many centuries after the events it describes, and Dorn's notes point out places where its genealogies differ from other accounts. The linguistic evidence described below points to the Iranian languages of the region, not to Hebrew, as the background of Pashto itself. [1] [2]
The Pactyes of Herodotus
The Greek historian Herodotus, writing in the fifth century BCE, mentions a land he calls the Pactyic country. He says that the explorer Scylax, sent by the Persian king Darius to find where the Indus reaches the sea, set out "from the city of Caspatyrus and the Pactyic country" and sailed down the river. Herodotus also uses a tribal name, Paktues, usually rendered in English as Pactyes. [3] [2]
Because the names sound alike, many writers have linked the Pactyes with the Pashtuns. The linguist Georg Morgenstierne called this derivation "phonetically untenable": in his view the Greek letters could not have represented the sounds the Iranian name would have had at that time. He thought it more likely that Pashto and Pashtun go back to older forms reconstructed as *Parsuwā and *Parswāna-. The link with Herodotus is therefore a hypothesis, and a disputed one. [2]
What the language shows
Firmer evidence comes from Pashto (پښتو) itself. Pashto is an Iranian language of the Indo-European family, spoken in southern and south-eastern Afghanistan, in north-western and western Pakistan and on Iran's eastern border. Morgenstierne placed it firmly in the north-eastern branch of Iranian and noted a few features it shares with Munji. He also judged that the name Pashto, though recorded later than Afghan, is the original native name. [2]
A language is not the same thing as ancestry, but it shows where a community's speech comes from. Morgenstierne also studied the Wanetsi dialect, whose speakers he thought might represent "the forerunners of the main Paṣ̌tūn movement towards the east". He added that when and where they split off was impossible to say. [2]
What genetics has and has not shown
In 2012 a team led by Marc Haber studied markers on the Y chromosome, which passes from father to son, in 204 men from Afghanistan's four largest groups: Pashtun, Tajik, Hazara and Uzbek. Comparing them with over 8,500 samples from the wider region, they concluded that modern Afghans largely share a heritage from a common ancestral population that may date to the spread of farming, and that the groups began to diverge in the Bronze Age. Pashtun and Tajik samples were closer to populations of northern and western India than to Hazara and Uzbek samples. [4]
A 2007 study by Sadaf Firasat and colleagues tested a different tradition: the claim, made by some Pathans as well as by the Kalash and Burusho of northern Pakistan, of descent from the soldiers of Alexander the Great. They found two Pathan men with a Y-chromosome type common in Greece and read this as evidence of a European origin for "a small proportion" of Pathan Y chromosomes, not for the population as a whole. [5]
These studies have limits. They follow the male line only and describe populations rather than named ancestors. Genetics can show broad regional connections, but it cannot confirm or disprove a figure such as Qais. Neither study reports evidence for an Israelite origin of the Pashtuns. [4] [5]
Reading the family tree
The family tree on this site follows the traditional genealogy, mainly as recorded in Dorn's translation of the Makhzan-i Afghani. It shows how Pashtun tribes have described their relationships to one another, and it should not be read as a proven biological pedigree. Versions differ between sources: the place of the Karlanri, the descent of particular tribes and even the number of Qais's sons vary from one account to another. [1]
Sources
- Bernhard Dorn (trans.) — History of the Afghans, translated from the Persian of Neamet Ullah [Makhzan-i Afghani], Parts I and II (London, 1829–1836), Internet Archive scan
Used for the date and compilers of the Makhzan-i Afghani, the genealogy from Qais (Sarban, Batni, Ghurghusht, Sharkhbun, Kharshbun, Taren, Avdal, Kerranians), the Bani Israel narrative, the Bibi Matu and Sheerani stories, and Dorn's note on Elphinstone's fourth son. Consulted 9 October 2026.
- Georg Morgenstierne — "Afghanistan vi. Paṣ̌tō", Encyclopaedia Iranica (1983)
Used for Pashto as a north-eastern Iranian language, its geographical spread, the rejection of the Páktues derivation, the *Parsuwā etymology and the Wanetsi dialect. Consulted 9 October 2026.
- Herodotus — Histories 4.44, trans. A. D. Godley, Perseus Digital Library
Used for the passage on Scylax setting out from Caspatyrus and the Pactyic country. Consulted 9 October 2026.
- Haber, M. et al. — "Afghanistan's Ethnic Groups Share a Y-Chromosomal Heritage Structured by Historical Events", PLoS ONE 7(3): e34288 (2012)
Used for the sample design, shared Neolithic-era heritage, Bronze Age divergence and the Pashtun-Tajik affinity with northern and western Indian populations. Consulted 9 October 2026.
- Firasat, S. et al. — "Y-chromosomal evidence for a limited Greek contribution to the Pathan population of Pakistan", European Journal of Human Genetics (2007), PubMed Central
Used for the test of Greek-descent claims and the finding of a small European-origin share of Pathan Y chromosomes. Consulted 9 October 2026.
Noticed a mistake or a missing source? Here's how we correct things
Related articles
Timeline
c. 1609–1611
Makhzan-i Afghani compiled
Nimatullah of Herat compiles the Makhzan-i Afghani at Jahangir's court, recording the genealogy from Qais Abdur Rashid.
1829
Dorn's English translation
Bernhard Dorn publishes the first part of his English translation as History of the Afghans in London.
2012
Afghan Y-chromosome study
Haber and colleagues publish a Y-chromosome study of Pashtun, Tajik, Hazara and Uzbek men in PLoS ONE.