The small temple complex on the slopes of Thorung La in the upper Mustang valley has two names. From the south, from the Hindu side of the Himalaya, it is called Muktinath — mukti for liberation, natha for lord, the Sanskrit-derived name under which the temple appears in the Vishnu Purana, in the Nalayira Divya Prabandham, and in the Satsangi Jeevan of the Swaminarayan sampraday.
From the north, from the Tibetan plateau and the monasteries of Lo Manthang, the same temple is called Chumig Gyatsa — chumig, “springs,” gyatsa, “hundred,” the Tibetan name meaning “the Hundred Springs.” It has been Chumig Gyatsa to Tibetan and Himalayan Buddhists for at least twelve hundred years.
The temple has one physical existence and two liturgical identities. The same pagoda is circumambulated by Hindu brahmin priests and by Tibetan Buddhist nuns. The same 108 water spouts are understood by Hindus as purifying from samsara and by Buddhists as cleansing the 108 defilements of the mind that obscure awakening. The same eternal flame is worshipped by Hindus as Jwala Mai and by Buddhists as the abode of the dakinis. A Hindu ashtadhatu murti of Vishnu, flanked by Lakshmi and Saraswati, is at the centre of the sanctum; the same murti is worshipped by Buddhists as Avalokiteshvara, the Bodhisattva of compassion.1
This is not syncretism in the loose sense. Muktinath-Chumig Gyatsa is not a blended, diluted meeting-place of two traditions that have softened their edges to accommodate each other. It is a site where two distinct, internally coherent, fully-articulated religious systems recognise the same ground as central — each for its own reasons — and have maintained this recognition, side by side, for more than a thousand years, without either tradition ceding any of its doctrinal commitments.
For a Vaishnava pilgrim arriving at Muktinath, understanding what the temple is called on its Tibetan side is not an optional cultural detail. It is part of the meaning of the ground. A site that is claimed by only one tradition can be the property of that tradition. A site that is claimed by two traditions, each for reasons internal to itself, must be — at a level deeper than either tradition’s claim — the kind of ground that was already sacred before either of them arrived to name it.
The name and the number
Chumig Gyatsa is sometimes translated as “the Hundred Springs” and sometimes as “the Hundred Waters.” Tibetan Buddhist literature uses the number gyatsa — one hundred — in the conventional sense that Sanskrit uses shata: a round figure indicating fullness or totality, not a precise arithmetic count. The actual site has 108 water-spouts, not 100, and 108 is itself a sacred number in both traditions.2
The precise count of 108 is striking. It appears here in water and also at every Tibetan Buddhist monastery in the number of beads on a mala, the number of volumes of the Kangyur, the number of names of Avalokiteshvara. It appears in the Hindu tradition as the number of Upanishads, the number of Divya Desams, the number of beads on a japa mala, the number of Mukti Dhara at this very temple. The convergence is not accidental. Both traditions drew, in their early formative periods, on shared Indo-Tibetan siddha lineages in which 108 had already acquired its cosmological weight.
The Tibetan Buddhist tradition attributes the formation of the 108 springs to a specific miraculous event. According to one prominent legend — paralleling the story of Chumi Gyatse Falls in Arunachal Pradesh — Guru Padmasambhava, on his eighth-century journey from India to Tibet, flung his mala (rosary) against a rock in the Muktinath valley. From the point of impact, 108 streams of water gushed forth, one for each bead. This is the Buddhist origin story of the 108 spouts. The Hindu tradition, parallel and independent, attributes the 108 waters to the Vishnu Purana’s description of Mukti Kshetra as a ground where the elements manifest in their purest form, with water emerging in the number sacred to the cosmos itself.
Both stories hold that the water is not ordinary glacial runoff. It is svayam-prakata, self-manifested. It arose from a sacred cause and it continues to flow for a sacred reason.
Padmasambhava’s passage
The figure at the centre of the Buddhist claim on Chumig Gyatsa is Padmasambhava, known throughout the Tibetan tradition as Guru Rinpoche — “the Precious Master.” Padmasambhava is not a minor figure in Tibetan Buddhism; he is the founder. Before his arrival in Tibet in the eighth century, the tradition holds, the Tibetan plateau was resistant to the Buddhist dharma — spiritually contested by the indigenous Bon shamanic tradition, physically difficult, and politically resistant. Padmasambhava, invited from India by the Tibetan king Trisong Detsen, travelled across the Himalaya, subduing obstacles, converting adversarial local deities into dharmapala (protectors of the dharma), and establishing the conditions under which Vajrayana Buddhism could take root.3
The Tibetan tradition names a specific number of sites — twenty-four — at which Padmasambhava performed tantric practices during his northward journey. These are the twenty-four Tantric sites, held within Tibetan Buddhism as the foundational geography of Vajrayana. Chumig Gyatsa is one of them.4
The detail matters. Twenty-four is not a large number. Across the entire span of Padmasambhava’s travels — from his origin in Oddiyana (likely in the Swat valley of present-day Pakistan) through India, through the Himalaya, and into Tibet — only twenty-four sites are singled out as sites of direct tantric accomplishment. Chumig Gyatsa is one of these twenty-four. Not a place Padmasambhava passed through. Not a place he stayed briefly. A place he practised, in the full technical Vajrayana sense, leaving behind the spiritual residue that Tibetan tradition holds such practices imprint on the ground.
The physical trace of Padmasambhava’s presence is preserved in the Narsingh Gompa (also called Mharme Lha Khang Gompa, the “monastery of a thousand lamps”) located adjacent to the Muktinath temple complex. The gompa houses a clay statue of Padmasambhava that, according to Tibetan tradition, was made by Padmasambhava himself in his own likeness. It has been cared for continuously, by a lineage of nuns, for twelve centuries.5
The dakinis
The Tibetan Buddhist tradition identifies Muktinath-Chumig Gyatsa not only as a site of Padmasambhava’s practice but as a residence of the dakinis — khandromas in Tibetan, literally “sky-goers,” the female awakened beings who embody enlightened energy in Vajrayana Buddhist cosmology.6
In Vajrayana literature, certain geographical sites are held to be physically inhabited by dakinis. These are not metaphorical presences. The tradition holds that khandromas are real, active, perceivable-under-the-right-conditions beings who congregate at sites where the conditions of practice are sufficiently refined. Chumig Gyatsa is classified as such a site — specifically, as the residence of the 21 Taras and of a community of dakinis who are said to be direct spiritual descendants of the women whom Padmasambhava taught during his eighth-century stay.
The nuns who have maintained the Narsingh Gompa for twelve centuries hold themselves, by tradition, to be the spiritual descendants of these dakini-lineages. The daily ritual of the gompa — the lighting of the thousand lamps, the reading of the Vajrayana sadhanas, the care of Padmasambhava’s self-made clay image — is performed within this specific understanding. The nuns are not priestesses of a generic Tibetan Buddhism. They are the continuators of a specific transmission, from a specific moment, at a specific ground.
For a Hindu pilgrim visiting Muktinath, this presence is worth noting. The Vaishnava pilgrim who performs darshan at the main temple and then walks the few metres to the 108 spouts is passing, on the way, the Narsingh Gompa and the daughter-dakini nuns who have held this ground as long as any Vaishnava institution anywhere has held any of its tirthas. The two traditions at Muktinath are not parallel. They are intertwined.
How the same stone is two deities
The philosophical question that Muktinath poses — how the same physical object can be, simultaneously, Vishnu for one tradition and Avalokiteshvara for another — is worth addressing directly, because it is not resolvable by compromise. Each tradition holds its claim with full seriousness.
The Vaishnava claim is that the central murti of the Muktinath temple, the life-size golden form in the sanctum, is Vishnu as Sri Mukti Natha Perumal, with his consorts Lakshmi and Bhumi. The Alvars sang this in the seventh century. The Sri Vaishnava tradition has maintained this identification without interruption. The Swaminarayan sampraday, two centuries ago, arrived at the same identification through the Satsangi Jeevan’s account of Nilkanth Varni’s worship of “Lord Vishnu in the holy place Muktinath.” The murti is Vishnu.
The Vajrayana claim is that the same murti is Avalokiteshvara, the Bodhisattva whose compassion is embodied in the mantra Om Mani Padme Hum. The Tibetan Buddhist tradition, inheriting the earlier Vajrayana reading of Himalayan Vishnu-forms as Avalokiteshvara-forms, has maintained this identification for over a thousand years. The murti is Avalokiteshvara.
How do both claims remain true?
The short answer is that each tradition operates with a distinct but non-conflicting theory of what a murti is. In the Vaishnava tradition, the murti is Vishnu because Vishnu has manifested in that form. In the Vajrayana tradition, the murti is Avalokiteshvara because Avalokiteshvara — as the universal principle of compassion — appears in whatever form is accessible to the practitioner. Vishnu and Avalokiteshvara are not, in the Vajrayana reading, the same personal deity. They are different names for what the Vajrayana understands as a single underlying reality that expresses itself differently in different cultural forms.
Neither tradition requires the other to accept its framing. The Vaishnava need not concede that Vishnu is “really” Avalokiteshvara. The Buddhist need not concede that Avalokiteshvara is “really” Vishnu. At Muktinath, both simply worship what is present, in the terms available to their tradition, and walk out. The coexistence is not achieved through agreement. It is achieved through the absence of a requirement to agree. Each tradition’s worship is internally complete without reference to the other.
This, historically, is the most philosophically elegant aspect of Muktinath. The ground does not demand that its pilgrims resolve their metaphysical differences. It offers, instead, a meeting-place where difference can be held without conflict. Vaishnavas and Buddhists at Muktinath have, across a millennium, demonstrated that deeply incompatible theological claims can share a single piece of ground if — and only if — neither claim is made contingent on the other’s acceptance.
The Bon substrate
Beneath both the Vaishnava and the Vajrayana identifications of Muktinath, there is an older layer. The indigenous religion of the Tibetan plateau before the arrival of Buddhism — and before the Hindu Puranic canon had crystallised — was Bon, a shamanic tradition that recognised particular sites as the dwelling-places of powerful local deities, earth-spirits, and ancestral forces.
The Mustang region was Bon territory. It still is, in part; Bon practitioners continue to operate in upper Mustang, and the ethnographic record of Holly Walters (treated in a separate article in this series) documents their continued involvement with the Shaligram pilgrimage. The specific identification of Muktinath as a powerful site pre-dates both Vaishnavism and Buddhism; both later traditions arrived at this ground and inscribed their own readings onto a substrate that was already understood, locally, to be sacred for its own reasons.
This is why Muktinath is sometimes described as a “three-tradition” site rather than a two-tradition site. The Bon element is quieter than the Vaishnava and Buddhist elements — it does not have a temple, an enumerated pilgrimage canon, or a written textual tradition to the same extent — but it is materially older. The ground was, in the most literal chronological sense, Bon before it was Hindu, and Hindu before it was Buddhist.
For the contemporary pilgrim, this triple inscription matters because it reinforces a point that a single-tradition reading of Muktinath would obscure: the ground itself is prior to its names. The various traditions identify, in their own idioms, what the ground already was. None of them constructed it. They recognised it, each from their own angle. What they recognised is the same.
How the two traditions meet on the ground
The daily practice at Muktinath today bears out this co-existence without conflict. A Hindu brahmin priest performs the morning abhisheka of the main Vishnu murti with Ganga-jal, offers tulsi, performs the Vishnu Sahasranama recitation. A Tibetan Buddhist nun performs the morning puja of Padmasambhava at the Narsingh Gompa, lights the thousand butter lamps, recites the Guru Rinpoche sadhana. Pilgrims of both traditions pass the 108 Mukti Dhara — the Hindu reads them as washing away the sins of many lifetimes; the Buddhist reads them as cleansing the 108 defilements of the mind.
What is striking, to anyone who visits the site, is the absence of conflict. There are no signs saying “Hindu entrance here, Buddhist entrance there.” There are no separated queues. The murti is one. The 108 spouts are one set. The Jwala Mai is one flame. A Hindu pilgrim and a Buddhist pilgrim standing shoulder to shoulder at the 108 Dhara are not performing two different rituals that have been brought into external conformity. They are performing the same act — passing beneath cold glacial water — with two different internal theological frameworks, and both are entirely satisfied that the act has accomplished what their respective traditions say it accomplishes.
For the Bodha Retreat mission, this mutual recognition is not an obstacle to be managed. It is a feature of the site that strengthens the Vaishnava claim, not weakens it. The fact that a second major religious tradition, entirely independently, has for twelve centuries held this same ground as a site of its own foundational figure’s direct practice — this is external corroboration that something about the ground is real. The ground is not a Vaishnava invention, not a Buddhist invention, not a local projection. It is what these traditions, separately, came to in recognising a reality that was already present.
What the Swaminarayan tradition inherits
When Nilkanth Varni performed his tapasya at Pulhashram in 1792, the Narsingh Gompa was already 1,000 years old. Padmasambhava’s clay statue had already been cared for continuously by forty generations of nuns. The Tibetan Buddhist identification of the site as Chumig Gyatsa, as the residence of the 21 Taras, as one of the twenty-four tantric grounds of Padmasambhava’s eighth-century passage — all of this was already there.
Nilkanth Varni arrived as a Vaishnava ascetic, into a site he recognised through the Puranic and Satsangi Jeevan frame as Pulhashram, as the ground of Bharat’s penance, as the temple of Sri Mukti Natha Perumal. His own worship was unambiguously Vaishnava. The Satsangi Jeevan’s account does not mention the Buddhist dimension of the site — not because the text was concealing it, but because the frame of the biography is the story of Nilkanth Varni, and the frame he was operating in was the Vaishnava frame.
But the Buddhist dimension of Muktinath is not, for this reason, outside the Swaminarayan sampraday’s inheritance. When contemporary satsangis arrive at Muktinath, they arrive at a site that is, by the full facts of its history, both Muktinath and Chumig Gyatsa. The experience of the place includes both. The 108 Mukti Dhara are, simultaneously, the 108 springs of Padmasambhava’s mala. The eternal flame is, simultaneously, Jwala Mai and the Buddhist me-bar. The ground on which one stands has been a site of accumulated tapas not only from Pulaha and Bharat and Nilkanth Varni, but from Padmasambhava and a thousand years of dakini-tradition nuns.
This is not dilution. This is deposit. The ground has held more, not less. The pilgrim who arrives aware of the full history of the site arrives at a deeper ground than the pilgrim who arrives aware of only one of its traditions. The sampraday’s own teaching — that sacred ground is a parampara, a succession of attention across lifetimes — is, if anything, more fully supported at Muktinath than at any other single-tradition tirtha the sampraday holds.
Chumig Gyatsa is not the name of Muktinath under a different religion. It is the name of one more layer of attention that has been added to the same ground, by a different community, for a thousand years. The Hindu pilgrim who climbs the path to Muktinath climbs through that layer. They cannot do otherwise. The layer is the path.
Notes
-
The dual identification of the central murti is attested in multiple independent sources, including the Wikipedia article on Muktinath, the Nepal Tourism Board’s official documentation of the site, and Holly Walters’ ethnographic work. The Buddhist reading of the Vishnu murti as Avalokiteshvara is not a later reinterpretation but a foundational element of the Tibetan Buddhist classification of the site as one of the 24 Tantric grounds. ↩
-
The 108 spouts are technically 108, not 100, though the Tibetan gyatsa in the site’s name means “hundred.” The slippage between “hundred” as a round figure and 108 as the actual count is common in both Tibetan Buddhist and Sanskrit-derived Hindu traditions and reflects the shared cultural use of 108 as a sacred cosmological number. ↩
-
The outline of Padmasambhava’s biography is given across extensive Tibetan Buddhist hagiographical literature, most notably the Padma bka’ thang (the “Testament of Padma”). The specific claim that Chumig Gyatsa is one of the 24 Tantric sites is preserved in multiple Tibetan canonical enumerations; the precise list varies slightly across lineages but Muktinath appears in all major enumerations. ↩
-
The 24 Tantric sites of Vajrayana Buddhism (also called pitha-s in Vajrayana usage, echoing but distinct from the Hindu Shakta pitha system) are enumerated across the Vajrayana canon. Muktinath-Chumig Gyatsa appears consistently in the Nyingma tradition’s list. ↩
-
The Padmasambhava clay image at Narsingh Gompa is attested across multiple ethnographic and travel sources, including Holly Walters’ Shaligram Pilgrimage in the Nepal Himalayas (Amsterdam University Press, 2020) and Nepal Tourism Board documentation. ↩
-
The identification of Muktinath as a site of the 21 Taras and the resident dakini community is preserved in the Vajrayana literature of the site and is corroborated in contemporary ethnographic sources. The lineage claim of the current nuns at Narsingh Gompa as spiritual descendants of Padmasambhava’s eighth-century female disciples is a claim internal to the Tibetan Buddhist tradition and worth preserving verbatim from the tradition’s own sources. ↩