All My Friends Are in My Head

by Rachel Monroe

I used to talk to myself all the time. Now that I have a cat, I mostly talk to him. It’s not because I live alone; I did the same thing back when I had a half-dozen roommates, too. It’s great if you can find someone who’ll talk back, of course, but it isn’t entirely necessary; you can get by with a pet, an imaginary friend, or perhaps a pony-shaped thought-creature that you’ve manifested with your own mind.

As a young woman growing up in Brussels in the eighteen seventies, Alexandra David-Neel loved secrets: secret societies, esoteric religions, all things taboo and forbidden. She dabbled in Freemasonry, Theosophy, opera singing, and anarchist pamphleteering; she dressed up like a man to hang out with a French cult that smoked hash and saw visions. In her twenties, she began traveling to Asia. In Sikkim, she met the Dalai Lama; in India, she studied Sanskrit and took part in tantric rites. She spent two years living in a cave in Tibet with a hermit who wore an apron carved of human bone; he taught her telepathy and tumo, a Tibetan technique for generating body heat through breath, which they used to avoid freezing to death. Between adventures, she returned to France and published books about her exploits, which were devoured by a public eager for tales from the exotic east.

In With Mystics and Magicians in Tibet, first published in English in 1931, David-Neel described two uncanny encounters. The first was with a Tibetan painter followed around by a fantastic creature that looked just like the monsters in his paintings; the second was with another with a monk who had made an exact, living-and-breathing replica of himself to confuse his enemies. Both creatures were tulpas, which David-Neel defined as “magic formations generated by a powerful concentration of thought.” Stemming from a syncretic blend of Tibetan Buddhism and indigenous shamanic traditions, Tibetan tulpas were thoughtforms brought into actual existence through ritual meditation, a sort of mental counterpart to the Talmud’s mindless, mud-born golem.

Admitting her own “habitual incredulity,” David-Neel decided to try to make a tulpa herself. “I chose for my experiment a most insignificant character: a monk, short and fat, of an innocent and jolly type,” she wrote. After a few months of seclusion and diligent ritual prayer, her phantom monk began to take shape. Over time, his form grew increasingly fixed and life-like. Months later, when David-Neel left for an extended horseback tour through the countryside, her tulpa tagged along. By this point, she no longer had to focus her thoughts to make him manifest; he was just there, performing actions that appeared to be autonomous: walking quietly, looking around at the view. At times, it was as if his robe had brushed against her. Once, she felt his hand softly rest on her shoulder.

For the half-century or so after the publication of David-Neel’s account, tulpas remained an intermittent topic of interest for the cryptid/UFO/fringe science crowd. Shortly before his death, Philip K. Dick claimed to have manifested a tulpa of Walt Disney, who sat down at the table and ate the food placed in front of him, but hadn’t spoken — yet. In the 2001 updated edition of the Mothman Prophecies, John Keel posited that the titular cryptid might not have been an extraterrestrial after all, but rather an uncontrolled tulpa, a thoughtform gone rogue.

If there’s one lesson to take from all the self-published tulpa horror novels and message board creepypasta out there, it’s that tulpas have a tendency to get out of hand and exert their will on the world in troublesome ways. But maybe that’s what you get for creating a creature with a mind of its own, even if that mind was created inside your own head. Of course they’ll go rogue. Of course you’ll find them in places you never expected them to be.

To access the Tulpas Gone Wild board sub-reddit, you have to click a button certifying that you’re over 18. Then you’ll find a dozen or so threads with headlines like “Feeling a bit [f]risky tonight 🙂” and “Up for a little [m]ischief?” and “[F]ucked in public,” all of which are marked NSFW, and all of which link to images that are not only eminently SFW, but so dull that they look as though they might have been taken by mistake: a beige couch, its cushions slightly askew; the upper corner of a white ceiling. This is the world of the twenty-first century tulpa — an empty room, a digital camera, an online forum.

Tulpas started cropping up as a popular topic of conversation on 4chan around 2009. The growing community was eventually trolled out of 4chan; over the next few years, they moved onto scores of message boards, tumblrs, and blogs. Currently, the tulpa forum on Reddit, the central online gathering place for people engaged in creating “intelligent companions imagined into existence,” boasts nearly seven thousand subscribers (although far fewer active posters). Tulpa-creation as practiced on the internet would likely be unrecognizable to Tibetan mystics who conceived of tulpas as magically created manifestations. Redditors creating tulpas don’t think of themselves as shamans engaged in metaphysical edge-play; they tend to frame the practice as an exploration of human cognition, relying on an odd contemporary blend of mindfulness practice, folk neuroscience, and role-playing-game character creation.

There are a number of tulpa-creation FAQs out there, offering practical — even strict — advice: “You should spend at least twenty hours of just sitting down and visualizing your tulpa perfectly in your head. The very minimum you should be forcing for at one period is forty minutes. Also, don’t force for longer than three hours at a time or else you will get horrific headaches.” Forcing a tulpa, to use the preferred verb of contemporary online tulpamancers, is not for the wishy-washy. Getting your tulpa to the point of imposition — that is, where she appears as a fully 3D creature perceivable by all five senses, and capable of independent thought and action — can take months of daily practice. Judging from Reddit’s Tulpa Art Tuesdays, human tulpas tend to look right out of an anime casting call: all big, wet eyes, rainbow-colored hair, and dainty noses. Non-human tulpas include flirty squirrels, eyeless reptilians, and, yes, plenty of ponies. (The brony and tulpa communities overlap; around 2012, after the tulpamancers were trolled off the main 4chan boards, they found refuge in /mlp/, the My Little Pony board.)

Tulpa FAQs recommend spending three to ten hours concentrating on your tulpa’s smell alone, and five to ten hours sitting quietly, picturing its particular gait and gestures. After two months of daily practice, your tulpa might begin to show faint signs of independence — first communicating with you through flashes of shared emotion, then parroting your words back to you, and finally, hopefully, saying something on his own. “I did not get an emotional response until I hit around 50 hours in, so really, don’t hold your breath,” one prolific poster notes. “Remember that if a sentient being could be made in a day then everyone would have one.”

Even after achieving sentience, tulpas must be tended or else they’ll wither away, Tamagotchi-like, from lack of attention. The forums are full of suggestions for fun tulpa-host activities to sustain and deepen their connection. Some sound dull (addition flashcards); some seem pulled from a romantic comedy dating montage (cook dinner together, feed ducks at a pond); and some are outright bizarre (“Lie down, set a stuffed animal (with arms, sorry no snakes) and set it on your chest with its arms facing you, and ask your tulpa to ‘control’ the arms through possessing your hands. Then they can slap you like they’ve always wanted to!”). Some tulpas even post on Reddit, presumably through the bodies of their hosts, offering insider tips to new users: “When I was younger the host exposed us both to new experiences. We did not stay cooped up in his house, but went exploring on his days off from work. Then I could react to new things alongside him.”

As far as internet subcultures go, the tulpa crowd isn’t very popular. Outsiders are quick to assume that tulpamancers are at best hopelessly awkward and at worst somehow icky, as if despite all the “best friend” rhetoric, what they’re really after is a zombie sex-slave or yes-man (or yes-pony). The Encyclopedia Dramatic defines a tulpa as “an imaginary friend or imaginary girlfriend/boyfriend for batshit insane,lonely /x/philes, bronies, and weeaboos.” The hours tulpamancers spend forcing can seem not only pathetic but also unfair, as if they’re trying to reap the benefits of companionship without ever emerging from their rooms and having to do the hard, messy work of negotiating a real relationship with a real person.

Preliminary data from a study on the tulpa community dovetails nicely (in some ways, at least) with stereotypes of the undersocialized, over-internetted young man: Tulpamancers are mostly white, middle to upper-middle class urban youth, mostly between the ages of nineteen and twenty-three. There are three men for every woman in the community, although around ten percent of tulpamancers identify as gender-fluid. The prototypical tulpamancer is cerebral, shy, and socially anxious.

But as the study’s author, McGill anthropologist Dr. Samuel Viessiere, points out, dismissing tulpamancers as antisocial creepers is a mistake. While they report social anxiety and limited opportunities for in-person social interaction, they also display average to above-average levels of empathy and interest in other people. Being a person is sometimes inescapably lonely, even for the most socially skilled and likable among us; for the awkward and slightly askew, it’s exhausting. Human consciousness is intrinsically social and collective, Viessiere points out, but contemporary social processes are often isolating. Not finding the connection they crave in the world, the tulpamancers create it for themselves.

It’s telling, too, that the tulpa community’s most prized quality is autonomy. In true subculture fashion, there are lots of in-group words for all the wrong paths you might wander down. Are you trying to create a thought-creature for primarily sexual purposes? Then you’re making a mind doll, not a tulpa. Does your tulpa exist only to serve your own needs, without a will or an opinion of her own? Then she’s a servitor, and you might as well start all over again. Does your tulpa repeat your own thoughts and opinions back to you? Then he’s parroting, and while that may be a step on the path to imposition, it’s certainly not autonomy. For some posters, the practice of cultivating and nurturing a tulpa seems almost like rehearsing for a relationship: Ask your tulpa how she’s feeling, listen to her favorite music even if it’s not to your taste, go out and find some ducks to feed.

Ultimately, then, tulpamancy is the practice of creating an other within your own mind. This isn’t as crazy or improbable as it may sound at first. Our selves are fractured and plural, made up of parts that may seem external or even alien to us. Otherwise, sentences like I got so mad at myself wouldn’t make any sense. Some tulpamancers even sound like they’re working to manifest a Redditized version of their Jungian anima: “I am a white male, 27, single/foreveralone, and have a masters in computer science. I currently don’t have a ‘real’ job, so I have plenty of time for this. I’ve decided that my tupa will be a human female who I hope will be my closest friend, help me stay motivated, and help me explore my mind… I chose to make her female primarily because I feel that some part of my mind is feminine. This will give that part a chance to exist rather than be repressed by society’s rules about what it is to be male.”

Of course, belief in an independently functioning being that lives inside your own mind requires some logistical and metaphysical gymnastics: “Tulpas don’t go tend anywhere you can’t find them. You may only think they did. You’re in the same head after all. At most they’re off thinking by themselves for a while or ‘asleep’. People need their space.” Not all tulpamancers would agree with the above statement; the community is split between those who see their tulpas as metaphysical emanations and those who frame phenomenon in the language of neuroscience, with the latter group seemingly outnumbering the former; the subhead on is “For Science!,” and there is occasional excited talk about experimenting with EEG machines. For this group, tulpas are real in the way that lucid dreams or “advanced hallucinations” are real; your tulpa exists in your mind, and your mind creates reality, ergo your tulpa is real. (There are also some complicated theories involving neural networks and black boxes, if that’s what you prefer.)

The scientific tulpamancers are optimistic to the point of naiveté about the mind’s ability to act upon — or reshape — the world. In this way, they’re another link in the chain reaching from nineteenth-century New Thought to positive-thinking their way to a new car, vis-a-vis Oprah: If you visualize it, you can will it into being. The science-minded tulpamancers are not so different, then, than their more mystical-minded counterparts, only with neuroscience standing in as a kind of twenty-first century sorcery.

After a period of peaceful coexistence, Alexandra David-Neel’s tulpa, the jolly monk, began to change. “The fat, chubby-cheeked fellow grew leaner, his face assumed a vaguely mocking, sly, malignant look. He became more troublesome and bold,” she wrote. “In brief, he escaped my control.” Her former companion was now a burden; his presence began to feel like “a daynightmare.” She decided to dissolve him, but he didn’t want to go: “My mind-creature was tenacious of life,” she wrote. It took her half a year to get rid of him.

This could be read as a warning against errant tulpamancy, but I don’t think so. After all, aren’t we all always building mind creatures every day, whether we mean to or not? As a Tibetan lama tells David-Neel, “One must know how to protect oneself against the tigers to which one has given birth, as well as against those that have been begotten by others.” This is good advice for non-tulpamancers, too — even if you haven’t birthed any tigers recently, there’s a decent chance you’ve got something predatory living inside your head. Mindfulness meditation, tulpamancy, therapy; they’re all ways to try and be careful about what you’re nurturing up there. is full of stories of tulpamancers whose tulpas have been invaluable to their well-being. A girl who’s prone to panic attacks calms down when her tulpa runs his fingers through her hair; a guy who tends to stay up all night is convinced by his tulpa to go to bed at a reasonable hour. Sometimes, the tulpas intervene in more serious cases: “I was diagnosed and put on mood stabilizers when I was eight years old. After fourteen years, it seemed normal. But with Aura’s help, I have been able to come off of all my medicines, and have been free of them for the past five months. I would have never been able to if it hadn’t been for her helping me.” “I was actually on Lexapro for depression/anxiety for a while, and I went off of it cold-turkey about the time I created Flutters… [While on meds] I quit my job, stopped trying at school, would just sit at home all day and be extremely anxious if I had to leave. Luckily, I have been off of them, and Flutters has helped me improve immensely. I now have a decent job, am doing better in school, [and am] making music again.”

My tulpa’s name would be Cam, I decided on Day One, and he would be male, and human. His first few characteristics came easily: he’s a couple inches shorter than me, and he has a big nose. But when I tried to believe in him existing in the world, I was less successful; sitting on my couch, I tried to picture him sitting next to me, taking up an approximately human sized slice of space, but he remained featureless and eighty-five percent transparent, like a ghost or a failing hologram. “Cam,” I said, experimentally. “Hello, Cam. It’s nice to meet you.” Cam didn’t say anything back, maybe because I was doing a bad job of believing in him, but also because he hardly existed yet; according to typical tulpa timelines, he was barely a zygote. I fell silent, deciding that it was okay that Cam and I didn’t have much to say to each other at this point. Instead, I spent forty minutes projecting a vague sense of friendliness and camaraderie toward a human-sized emptiness on my couch.

On Day Four, following some of the tulpa FAQ advice, Cam and I went for a walk. Faced with all the visual distractions of the outside world, I had an even harder time imagining him there, but I made space for him on the path nonetheless. When there was a puddle to walk around, I went first, then paused and waited for him to catch up. This felt ridiculous, but it also felt fine. West Texas had a wetter-than-normal summer this year, and the path by the railroad tracks was bordered by thick grass as high as my hip, interrupted by occasional patches of happy-faced sunflowers. The grass was hectic with the sound of insects, and grasshoppers flashed the red undersides of their wings as they leapt out of the way of my — our — footsteps.

And so for an hour we walked like that, in companionable science, me and the inside of my own brain.

Photos by Keith Harrison, Rob Carlson, and Scott Rubin.