Jinmensō is three pieces: jin, person; men, face; sō, a sore or ulcer. A human-face sore.
The name is a description of an appearance, and nothing in it claims an agent. There is no demon in the word, no possession, no curse — only a lesion that has taken a shape it should not have.
That is where this subject starts, and where most English accounts of it do not, because the figure now reaches the anglophone internet as a monster entry: a talking sore on a knee or an upper arm, demanding to be fed, usually illustrated. The older Japanese handling of it is medical, and the difference is not a technicality.
The clearest evidence of where the term belonged is a passage in which it is not the subject at all.
A study of Japanese writing on mental and anomalous illness reproduces an Edo author’s discussion of a nun who was said to have eaten nothing for twenty years. The writer went to look, concluded she was an ordinary woman, took lodgings in the village and asked around.
Then he reasons about it, and the reasoning is what matters here. This, he says, is an abnormal illness. The appetites can be suppressed. And then, to establish that a body can behave in unexpected ways, he reaches for two examples: the ōseichū and the jinmensō and other such strange diseases.
Notice the grammar. He is not telling a story about a face-sore. He is using the face-sore as an agreed fact in order to argue about something else — the way one might say people do sleepwalk, after all. For that move to work, his reader has to accept the category without explanation.
So by the Edo period jinmensō was a thing an educated writer could cite in passing, in a medical argument, expecting to be understood. It was not a wonder to be reported. It was a precedent to be invoked.
The companion term in that sentence is the better clue, and it is worth following.
The ōseichū — the echo insect — names a condition in which something inside the body answers when the person speaks. The same study of Japanese medical writing places it in a bibliography of Chinese sources, citing a Song-dynasty miscellany for the insect itself and an earlier Tang collection for the related ōseibyō, the echo sickness.
Put the pair together and the shared structure is obvious. Both conditions consist of a second party inside one body: one that answers, and one that has a face. Neither is described as an invader from outside; both are described as something the body has produced.
That is why they travel together in Japanese writing, and it is the strongest argument for reading jinmensō as a medical idea rather than a monster. The two are varieties of the same thought, and the other one was indexed to Chinese medical literature with authors and titles attached.
One question the category raises and does not answer: why a face.
A lesion can look like a great many things, and the pattern-finding that produces a face out of a sore is the same faculty that produces one out of a knot in a plank — our note on the blood ceiling deals with that at length, including a tradesman’s account of how the marks actually got there.
But a face is not just one pattern among many. It is the pattern human vision is most strongly tuned to find, and the only one that carries an implication: a face implies somebody. A sore shaped like a fish is a curiosity. A sore shaped like a face is a second occupant.
Which is why the explanations that attached to it are about identity and debt rather than about tissue. Once the lesion has a face, the question stops being what it is and becomes whose it is — and that question has no medical answer at all.
The explanation usually offered in Japan was karmic: gōbyō, an illness that is the working-out of an act.
It is worth being precise about what that framing does, because it is often read as mere moralising. A karmic account answers a question that a purely physical account cannot: not what the lesion is, but why this person.
And it has a consequence that is easy to miss. If a sore is the visible return of something done, then the sore is evidence, and the body carrying it is a document that can be read by anybody who looks. That is a far heavier thing to have on your arm than an infection.
Our note on the inugami describes the same logic working at the level of a household rather than a body, with the same result: a condition treated as a disclosure, and the people around it reading it as one.
One more piece of evidence about the category, from the other end of the timeline.
A writer of the 1920s, describing his own obsessive collecting of strange material, sets down the contents of his notebook as a long list. It runs through eclipses and comets, through charms and divination, through named creatures, and then into a run of specific wonders: the face-sore, the killing stone, the blood ceiling, the flesh-adhering mask, the night-weeping stone, the soul-separation sickness.
That is not a medical list. It is a list of famous individual marvels, and by the 1920s jinmensō had joined it.
There is a further wrinkle in that list worth noting. It also contains the soul-separation sickness — a condition rather than a creature — sitting between two place-bound marvels. So the 1920s collector was not sorting by kind at all; he was sorting by fame. Anything sufficiently well known went into the same column regardless of whether it was an illness, a stone or a stain.
So the term moved. In the Edo medical writing it is a disease class; in the modern enthusiast’s notebook it is a famous item alongside the blood ceiling, which our note on that subject deals with. The same word is doing two different jobs two centuries apart, and the English-language monster entry is descended from the second one.
A closing observation about what is missing, which is the thing that makes this subject hard to write about and easy to sensationalise.
There is no standard story. The great yōkai categories come with a shape: the fox deceives, the kappa takes, the vengeful dead demand. Jinmensō comes with an appearance and a cause-type and no plot at all. Nothing in the category tells you what the sore does next.
Which means almost everything the modern treatment adds — the speech, the appetite, the bargaining — is supplied from outside the term. The record gives a face where there should not be one, and an explanation for why it is on that particular person, and stops.
People set down that this had been seen on a body, and what was thought to have brought it. Those records exist in some number; they are short, they report the condition and the reckoning attached to it, and they are not curious about the person.
Five records of that kind became the five stories in There Is No Column For It. The book is at /books/there-is-no-column-for-it/.
The readings in “The Thing Beside It”, “What a Karmic Disease Is For” and “What the Category Will Not Supply” are KWAIDANOTE’s own.
Found an error? Write to kwaidanote@gmail.com — corrections are made on this page and dated.