Pod ldungs sdu khlad kyi pya khyi ma

The Tibetan roll that carries her name came out of a sealed cave at Dunhuang and is held now in the British Library as IOL Tib J 734: eight sheets cut from a single scroll, the opening lost, 354 lines on the face. Near the end of the eighth sheet the text reaches the land of mChims and gives its lord a demoness. Rje mchims rje ne ‘u bzhugs. Srin mchims srin pod [gap] du khlad kyi pya khyi ma’i mchid nas. The lord of mChims is seated. The demoness of mChims speaks. Her line, in the formula the scroll repeats for demoness after demoness down the list, is a threat to carry him off to the demon land, where fire is not hot and water is not wet. The answer the scroll gives is a payment.

Appearance

She is known by her name alone. The scroll never describes her, and no image of her survives in Tibetan art. What survives is a word.

The word is damaged in the middle. The letters still on the page run pod [gap] du khlad kyi pya khyi ma, and whatever stood in the gap is gone. F. W. Thomas, who worked through the roll in the 1950s, read the name in full as Pod ldungs sdu khlad kyi pya khyi ma, and the reading has held since. Janet Gyatso lists her in 1989 beside two more demonesses from the same stretch of text, Phyag na yed mo and gNag brang ma, with the line numbers to match. The restoration is Thomas’s. What the restored syllables mean is another question.

The end of the name is the part that can be read plainly. Pya is the manuscript’s spelling of bya, bird. Khyi is the ordinary word for dog, and ma marks a female. Taken that way, the demoness of mChims is the bird-dog woman.

The compound is an ordinary Tibetan word. Dan Martin’s Zhangzhung dictionary lists bya khyi, “bird dog”, and records two accounts of what the animal is. In the first it is a tiny dog that lives in the nests of cliff-dwelling birds and can detect poison in food. In the second it is a large bird of prey and no dog at all. He notes the interest of the problem and leaves it open. So the creature behind the name has no settled body either, and the demoness named for it has none.

The rest of the catalogue suggests the compiler named his demons after what they looked like. One is ra lpags, goatskin. One is phlad po mgo dgu, the nine-headed. Others carry names like phyag na yed mo, pyo pyo mo, yug mo and tsa lung zha. These names are the only portraits the scroll provides, and where the compiler of the mChims entry reached the demoness of his own district, he used two animals and left the reader to work it out.

Function

The demoness belongs to a ritual that repeats an identical shape for land after land. Each passage opens on a local lord seated in his own country. The demon of that country then threatens him. The local priest, called gshen, sets up an effigy of the local mountain god. The passage closes with the god’s effigy handed over to the demon as a ransom, called glud, after which the lord is safe.

For mChims the passage names the lord as Ne’u. It names the priest as shen do rab ‘bring rab, and it names no god, because the line breaks there. Neighbouring Dunhuang manuscripts fill the gap: P. T. 1060 places a deity called mChims la’i Than ’tsho in mChims yul dgu sul beside the same lord. Reading Than ’tsho as the god whose effigy was set up for this demoness is an inference, since the scroll itself is damaged at exactly that point, and a reasonable one.

The threat is the same in every entry. The demon says it will carry the lord away to the country where fire is not hot and water is not wet. The scholar Yuewei Wang, working through the manuscript, notes that this formula is the reason the ransom is needed: the demon has announced a removal, and the ritual arranges for the removal to fall on a substitute. Samten Karmay describes the underlying idea as an exchange of equals, mnyam gnyis brje ba, something given for something taken. Nothing is killed and nothing is bound. The demoness is paid.

Her neighbours on the list reach across the world the Tibetan Empire knew, from rTsang and gNubs and sKyi to the nine-headed demoness of Dags, the demoness of Myang, and finally the demoness of rGya, China itself. Each has a lord of her own, and a priest who knew what to do about her. The scroll calls the gods of this ritual the nine mountain gods, a crowd rather than a count: Wang notes that the group it actually lists comes to fourteen.

Cross-Cultural Connections

The pattern of one demoness per district outlived the scroll. Janet Gyatso, writing on the feminine ground in Tibet, notes that the srin appear in many origin myths and are often attached to particular regions of the country, and she points to this scroll as the early text in which a crowd of them is named. P. T. 307 carries seven demonesses titled bod khams kyi bdag mo, the mistresses of the land of Tibet, named region by region and subdued in a single campaign by the Indian master Padmasambhava. The logic is the same and the ending is not: the older text buys its demoness off, while the later one puts her under oath. At full scale the same idea gives the supine demoness whose body is the map of Tibet, pinned by temples built at her joints. The mChims entry is an early and local instance of a habit that runs through Tibetan religious history.

By the time Tibetan Buddhism had reorganised its spirits into a system, the srin po sat inside the eight classes of gods and demons, the lha srin sde brgyad, a set whose membership shifts with the lineage doing the counting. Christopher Bell’s survey of Tibetan demonology opens with that taxonomy of named categories. The Dunhuang catalogue is older and more concrete: it pins its demonesses to places where people lived and had priests, one to a district.

The Inner Asian neighbours ran their demonologies the same way, as classes to be named and moved along. The Shimnus and the Mangus Demons of Mongolia and Buryatia are the black host of Buddhist cosmological warfare, expelled and defeated by heroes; the Albasty of the Turkic world is a female demon met at the bedside with iron and forty days of vigilance. Where the Turkic answer was iron and the Mongolian answer was a hero, the answer recorded in this scroll was an object of equivalent value, handed over at the boundary.

The scholarly work on the ransom section runs from the 1950s to the present. F. W. Thomas published the first complete transliteration and translation in 1957. Ariane Macdonald read the text in 1971 as evidence for the political mythology of the small kingdoms, while Rolf Stein examined the effigy as a ritual object. Later work by Brandon Dotson compared the healing specialists of IOL Tib J 734 with the priests of P. T. 1285. On the mechanism the ritual uses to name and to pay, see Exorcism Across Religions and Cultures, where the demand for a demon’s name recurs in every tradition covered. For the Tibetan afterlife system that grew up on the far side of this world, see The Tibetan Book of the Dead.

Modern Survival

The ransom still gets paid. Robert Barnett, working in Lhasa in the 2000s, documented the glud gtor in ordinary use, with a living trade in effigies and open argument about what the rite means in a modern city. Tibetan monasteries keep printing ritual collections for the removal of the srin mo by thread-cross. The vocabulary of the Dunhuang scroll stayed in work for twelve hundred years.

The demoness herself did not. She has no living cult or festival; her name is spoken nowhere outside the manuscript and the handful of studies built on it. She survives as a damaged line on the eighth sheet of a scroll in London, in a list of demons that a Tibetan official or priest had copied out while the empire held the oases of the north-eastern frontier.

Two things stay open. The name’s middle, pod ldungs sdu khlad, resists translation, and the gap in the line means nobody can be certain the name was written in full there. And the animal at the end of it, the bird-dog, is itself a creature Tibetans could not agree on. She was named after something that may have flown.

Sources

• IOL Tib J 734, Dunhuang Tibetan manuscript, British Library. Transliteration in the Old Tibetan Documents Online (OTDO), Tokyo University of Foreign Studies, https://otdo.aa-ken.jp/archives?p=ITJ_0734
• Thomas, F. W. Ancient Folk-Literature from North-Eastern Tibet: Introductions, Texts, Translations and Notes. Abhandlungen der Deutschen Akademie der Wissenschaften zu Berlin, Klasse für Sprachen, Literatur und Kunst, Jahrgang 1952, Nr. 3. Berlin: Akademie-Verlag, 1957
• Wang, Yuewei. ‘Mountain Gods, Celestial Gods, and Kingship: Non-Buddhist Traditions in Tibetan Dunhuang Manuscripts.’ BuddhistRoad Paper 3.4. Bochum: Center for Religious Studies, Ruhr-Universität Bochum, 2025. DOI 10.46586/rub.br.457
• Dotson, Brandon. ‘Complementarity and Opposition in Early Tibetan Ritual.’ Journal of the American Oriental Society 128, no. 1 (2008): 41-67
• Karmay, Samten G. The Arrow and the Spindle: Studies in History, Myths, Rituals and Beliefs in Tibet. Kathmandu: Mandala Book Point, 2009
• Macdonald, Ariane. ‘Une lecture des Pelliot tibétain 1286, 1287, 1038, 1047 et 1290.’ In Études tibétaines dédiées à la mémoire de Marcelle Lalou, 357-373. Paris: Maisonneuve, 1971
• Stein, Rolf A. ‘Du récit au rituel dans les manuscrits tibétains de Touen-houang.’ In Études tibétaines dédiées à la mémoire de Marcelle Lalou, 479-547. Paris: Maisonneuve, 1971
• Martin, Dan. ‘Zhang zhung dictionary.’ Revue d’Études Tibétaines 18 (April 2010): 33-253
• Barnett, Robert. ‘Notes on Contemporary Ransom Rituals in Lhasa.’ In Katia Buffetrille (ed.), Revisiting Rituals in a Changing Tibetan World, 273-374. Leiden: Brill, 2012
• Bell, Christopher. Tibetan Demonology. Elements in Religion and Violence. Cambridge: Cambridge University Press, 2020
• Gyatso, Janet. ‘Down with the Demoness: Reflections on a Feminine Ground in Tibet.’ In Janice D. Willis (ed.), Feminine Ground: Essays on Women and Tibet, 33-51. Ithaca: Snow Lion, 1989
• Prosopography of the Tibetan Empire, entries ‘mChims rje Ne’u’ (00098) and ‘Than ’tsho’ (00771), compiled from the Old Tibetan documents
• Jäschke, H. A. A Tibetan-English Dictionary. London: Routledge & Kegan Paul, 1881