The Woman Who Gave the West the Tulpa

The word tulpa reached English through a single source. Alexandra David-Neel used it in Mystiques et Magiciens du Tibet, published in French in 1929 and in English in 1931 as With Mystics and Magicians in Tibet, later reprinted as Magic and Mystery in Tibet. She glossed it as thought-form, and defined tulpas as magic formations generated by a powerful concentration of thought.

Every later use runs through that book: the horror fiction, the paranormal television, the online tulpamancy communities. When a forum post today explains what a tulpa is, it is quoting a French woman's travel memoir at somewhere between two and six removes.

David-Neel was born on 24 October 1868 and died on 8 September 1969, a few weeks short of 101. Between 1912 and 1916 she lived in a cave in northern Sikkim at over 4,000 metres, studying under a hermit lama. She met Aphur Yongden there in 1914, when he was fifteen; he travelled with her for the next four decades and she adopted him as her son in 1929. He died on 7 October 1955.

In 1924, at 55, disguised as a beggar with Yongden posing as a monk, she walked into Lhasa, which was closed to foreigners, and stayed two months. The detail that gives away how she was finally identified is the best line in her biography: she was unmasked because she washed at the river every morning, and nobody that clean was a beggar.

This matters because the usual dismissal of the tulpa story is that some credulous tourist wrote it down. She spent years inside the tradition, learned the language, and was capable of saying she did not know.

She chose the least suggestive subject she could think of, and said so: a most insignificant character, a monk, short and fat, of an innocent and jolly type. The choice is deliberate. A terrifying figure would prove nothing except that she could frighten herself.

Months of concentrated visualization in seclusion followed. Her account is that the figure gradually stabilised, then began to appear when she had not summoned it, and finally started acting on its own. Her words for the drift are precise: it became more troublesome and bold, and it escaped her control.

The detail she leans on hardest is that a herdsman who came to deliver butter saw the monk and took him for a real lama. That is the whole evidential weight of the story, and she knew it.

The monk also changed. The jolly fat face thinned and turned, in her description, sly and malignant. She decided to dissolve it, because she needed a quiet brain free of other preoccupations. It took, in her phrase, six months of hard struggle.

She did not present this as proof of anything. She wrote that she might have created her own hallucination.

That single clause is missing from most modern retellings, and it is the most useful line in the chapter. She offers two readings of her own experience and refuses to pick: either something with a degree of independent existence was formed, or she had been visualising so intensively for so long that her perception gave way. She points out that her own senses are not an instrument she can check with her own senses.

Hold both readings while you read the rest of her book, and it becomes much better evidence than it is usually treated as, because you can see where she draws her own lines.

Tulpa is an anglicised spelling. The Tibetan is sprul pa, and David-Neel noted the spelling herself in a footnote. In Buddhist doctrine the family of terms around sprul pa concerns emanation: the nirmanakaya, the emanation body of a buddha, and by extension the tulku, the recognised emanation of a realised teacher. It is a doctrine about how enlightened activity appears in the world, not a technique for manufacturing a private companion.

Scholars have made this point in detail. Natasha Mikles and Joseph Laycock, in Tracking the Tulpa, argue that the Western concept owes more to Theosophy, which was busy with thought-forms in Europe well before 1929, than to anything a Tibetan monastery would recognise. The vocabulary is Tibetan. The idea being described in English is largely a European one wearing Tibetan clothes.

This is not an accusation against David-Neel. She was translating an unfamiliar practice for readers who already had a word waiting for it, and the waiting word pulled her account toward itself.

The route is traceable. Her book stayed in print. Paranormal writers picked the word up. Television and film in the 1990s and 2000s used it for a monster-of-the-week, which is where most people first heard it.

Then it turned into a practice again. Online communities formed on 4chan and later on Reddit in the early 2010s, with an unlikely engine: adult fans of My Little Pony discussing whether sustained visualization of a character could produce something that answered back. The term for the practice, tulpamancy, and the term for the practitioner, tulpamancer, are inventions of those boards, not translations of anything.

What those communities describe is close to what David-Neel described and far from what a Tibetan text describes: long daily visualization, a figure that gradually acquires its own voice, and a relationship that continues for years.

Take the practitioner seriously enough to take the warning seriously. The person with the most detailed first-hand account of this in Western literature spent six months undoing it and called that a struggle.

Three things follow from her account. First, the figure you choose is the figure you get, and hers changed on its own, in an unpleasant direction, over time. Second, the effort to end it was much larger than the effort to start it. Third, she never resolved whether she had built something or destabilised her own perception, and she had more training than almost anyone reading this.

If your interest is in the mechanism rather than the companion, the honest version is available without any of the risk: sustained daily visualization of a fixed image is a documented concentration practice, it sharpens attention, and you can stop whenever you like. If you find a visualization starting to appear uninvited, that is the point David-Neel's account says to stop, not the point it says to continue.