Origins

This is the part of Limbu history where the most is claimed and the least is settled. The page below takes the four evidence streams one at a time, says what each can carry, and ends with two lists: what can be said with confidence, and what cannot.

The classical texts

The Sanskrit term Kirāta is genuinely ancient. It appears in the Yajurveda (Śukla XXX.16; Kṛṣṇa III.4.12.1) and the Atharvaveda (X.4.14), commonly paired with the Cīnas and distinguished from the Niṣādas. The Manusmṛti (X.43–45) treats the group as fallen Kṣatriyas “noted for their failure to respect Brahmanas” and calls their speech mleccha, barbarous. Kiratas appear in the Mahābhārata as a martial people to the east, and Ptolemy’s Geographia records a “Kirrhadia” Ninglekhu/ku heritage documentation project 2026 A.

What none of those attestations establishes is who the Kiratas were.

The Kirat dynasty of the Kathmandu Valley

The tradition of a Kirat dynasty ruling the Kathmandu Valley before the Licchavis — with Yalambar as its founder and a list of twenty-nine or thirty-two kings — comes from the late vaṃśāvalī chronicle tradition, principally the Gopālarājavaṃśāvalī, and not from contemporary epigraphy. Modern historians (D. R. Regmi, Mary Slusser’s Nepal Mandala, Luciano Petech’s Mediaeval History of Nepal) treat the king-list as chronicle material of uncertain historical value; Camillo Formigatti’s philological study of the Gopālarājavaṃśāvalī is the best current treatment of what such a chronicle can and cannot support Formigatti, Camillo A. 2022 B.

There is a second step that popular accounts routinely skip. Even if the Valley dynasty were historical, its ancestry is a separate question from that of the Limbu of the eastern hills. The two are frequently run together Ninglekhu/ku heritage documentation project 2026 B.

The linguistic evidence

Limbu (ISO 639-3 lif; Glottolog limb1266) is a Tibeto-Burman language of the Kiranti group, conventionally placed as the sole member of Eastern Kiranti alongside Central Kiranti — the Rai languages, Bantawa, Chamling, Kulung, Thulung and others — and Western Kiranti. George van Driem’s A Grammar of Limbu (Mouton de Gruyter, 1987) remains the standard reference grammar van Driem, George 1987 B.

Two features of this literature deserve to be better known. First, “Kiranti” is not a settled genetic node: historical linguists since about 2012 have treated it as paraphyletic rather than a coherent group, and Gerber and Grollmann (2018) accept a narrower “Central-Eastern Kiranti” while leaving “Western Kiranti” unclassified. Popular accounts routinely use linguistic kinship as proof of a single ancestral Kirat people — an inference the linguistics, as the report found it in August 2026, does not license Ninglekhu/ku heritage documentation project 2026 B.

Second, van Driem publicly retracted his own “Mahakiranti” hypothesis — which had grouped Kiranti with Newar, Baram and Thangmi — in 2004, after fieldwork on Gongduk in Bhutan convinced him that the shared morphology was “a shared retention of a far older trait” rather than evidence of a common node Ninglekhu/ku heritage documentation project 2026 B.

The genetic evidence

Population genomics now speaks directly to this question. The most relevant study — “Whole-genome sequences provide insights into the formation and adaptation of human populations in the Himalayas,” a bioRxiv preprint of November 2024 published in Current Biology in 2025 — generated “high-coverage whole-genome sequences from 87 samples of 16 populations from Bhutan, Nepal, India and the Tibetan plateau,” and Limbu was one of the sixteen sampled populations. Its headline findings: Himalayan populations diverged from East Asians around 10,000–12,000 years ago, at the Neolithic transition; there is “surprisingly high” population structure within the Himalayas; and Himalayan subclades of Y-chromosome haplogroups O2 and D1 and mtDNA lineage M9 “all coalesce around 2,000–5,000 ya” 2025 B.

Supporting work fills in the model. Gnecchi-Ruscone and colleagues found that Nepalese Tibeto-Burman groups other than the Sherpa derived their East Asian ancestry “not from the Tibetan/Sherpa lineage, but from low-altitude ancestors who migrated from China plausibly across Northern India/Myanmar” Gnecchi-Ruscone, G. A., Jeong, C., De Fanti, S., et al. 2017 B. Basnet and colleagues analysed 999 Nepalese mitochondrial sequences and found clear genetic substructure across Nepal’s ethnic groups Basnet, R., Rai, N., Tamang, R., et al. 2023 B. Liu and colleagues showed from ancient DNA that present-day mid-elevation Tibeto-Burman speakers “form a genetic cline that reflects a distinct genetic history” from the ancient high-altitude highlanders Liu, C.-C., Witonsky, D., Gosling, A., et al. 2022 B.

The Mundhum’s own accounts

The tradition offers more than one origin story, and that is the most interesting thing about it.

The best-known is the gotra scheme: some Limbu lineages traced to migration from Lhasa in the north (Lhasa gotra), others from Kashi/Varanasi in the south (Kashi gotra), with counts varying between sources — thirteen groups from Lhasa and five from Kashi in one telling, eight and five in another. A related tripartite classification names Khambongba, Tangsangba and Munaphen, with Khambongba glossed as the bhumiputra, the autochthonous stratum Subba, Nawa Raj D. A variant framing substitutes a “Yunan gotra” for the Lhasa one, tracing a route through Yunnan, north Burma and Assam.

Set alongside these is a wholly non-migratory account: Dr Chandra Laksamba records a tradition in which the Limbu people “were created by the god Sigera Yabhundin Mang” through divine instruction — a cosmogony with no geographic content at all Ninglekhu/ku heritage documentation project 2026 D.

Synthesis

What can be said with confidence:

• The Limbu are a Tibeto-Burman-speaking people of the eastern Nepal hills whose language belongs to the Kiranti group van Driem, George 1987 B.
• Their ancestral population separated from East Asian populations at a time-depth of roughly ten to twelve millennia 2025 B.
• They arrived in the southern Himalaya by a low-altitude southern route distinct from the Tibetan plateau route Gnecchi-Ruscone, G. A., Jeong, C., De Fanti, S., et al. 2017 B.
• They have been settled in Limbuwan long enough to have generated a dense toponymic clan system Ninglekhu/ku heritage documentation project 2026 B.
• They have been described by outsiders using the Sanskrit label Kirāta for a very long time Ninglekhu/ku heritage documentation project 2026 A.

What cannot be said with confidence:

• That the Kiratas of the Vedas were Limbus Ninglekhu/ku heritage documentation project 2026 B.
• That the Kathmandu Valley’s Kirat dynasty was Limbu Formigatti, Camillo A. 2022 B.
• That any particular migration route in the Mundhum corresponds to a documented event Ninglekhu/ku heritage documentation project 2026 E.

These are matters of tradition and of contested scholarship, and they are more interesting presented as such.