Kuchisake-onna

The useful question here is not whether she existed. In the testimonies Itō examines nobody treats her as a ghost, and neither Iikura nor Itō reports an attack attributed to her. The useful question is what kind of thing this was, and on that there is a genuine and recent scholarly answer. It was not a story. In 2021, four decades after the event, Itō Ryūhei went back to Nomura Jun'ichi's corpus and noticed something nobody had made much of: almost every account is written in the present tense. Not "her mouth was slit to the ears" but "her mouth is slit to the ears"; not "she was standing there" but "she stands there". That is the tense of a general rule, the tense of "the sun rises in the morning". The reports have no characters, no setting, no beginning and no end — only a list of attributes: she wears a mask, she carries a sickle, she wears a coat, she asks "am I pretty?", she runs fast. Itō's comparison is deflating and exact: it is the same form as saying that a kappa has a dish on its head and a shell on its back. It is not a narrative. It is, in the technical vocabulary of Japanese folklore studies, folk belief (zokushin) rather than tale (setsuwa) — and Itō's conclusion is that this, and not anything about 1970s Japan's psyche, is why it spread as it did: "the reason the slit-mouthed woman spread across the whole country is, it seems, that this was not a tale but folk belief." The mechanism is simple once stated. A story has to be told from beginning to end; drop a motif and it collapses into nonsense. A list does not. A child in the first year of primary school, with no narrative skill at all, can pass on a list, and anybody can add an item to it without rebuilding anything. That is exactly what happened: the "three sisters", the failed cosmetic surgery, the traffic accident, the escape from a psychiatric hospital were all bolted on later, each one answering a child's question about why she does this. Itō reads these additions as the rationalising and shrinking of the uncanny, and the psychiatric version was a piece of ordinary stigma, doing damage to real people who had nothing to do with any of it. The countermeasures are the give-away. Nomura's own reading, which Itō reports, was that the rumour carried the machinery of the old folktale inside it: the sickle comes out of folk custom, the escape-tale (tōsōtan) is a recognised type — the same shape as Kuwazu nyōbō, Ushikata yamauba, Sanmai no ofuda — and, most tellingly, "pomade" functions as a jugon, a spell-word. A rumour that arrives with its own antidote is a rumour a child can use: a child could say "pomade", wear it, or carry bekkō candy. Fear plus a usable defence is a far more transmissible package than fear alone, and it lets the listener become a participant rather than an audience. She was not a monster, which is why adults believed it. Itō makes the point that the children of 1979 did not classify her as a yōkai at all; the popular idea of a yōkai, shaped by Mizuki Shigeru, was something in premodern dress. Strip out the one impossible detail — the running speed, "faster than a motorbike" — and every remaining attribute could belong to a human being. The novelist Kyōgoku Natsuhiko, a high-school student at the time, put it flatly: "When I first heard it, the context was that a woman with a mental illness was doing harm to children, so I really believed it. There was nothing yōkai or occult about it." Itō adds that attacks on children were genuinely happening around the country, adults read the rumour in that frame, and that is why police were deployed and group walks home were organised. The response was not credulity. It was a reasonable reaction to an unreasonable-sounding report about child safety. What the sources do not support. In English the figure is often given Edo-period roots. The English Wikipedia article, as read on 25 September 2026, rests that on Matthew Meyer's illustrated database Yokai.com, whose entry says, without a source, that "during the Edo period, a large number of kuchisake onna attacks were blamed on shape-changed kitsune playing pranks on young men"; the samurai who slit his unfaithful wife's mouth comes, in the same article, from magazine pieces in The Atlantic (2013) and The Daily Telegraph (2019). The folklorists who worked on the 1979 rumour treat her as new. Iikura calls her "the first purely Japanese urban legend", and Itō places her in a modern lineage of child-snatching bogeys known from children's rumour: Kamasu-shoi, Kakurezatō, Fukurokatsugi, Aburatori, Aka-manto. He proposes a category for them, kaijin — neither human, nor ghost, nor shape-shifting fox, but taken by children with a monster's weight. The modern embellishments can be checked against the earliest record. In the cases Itō transcribes from Nomura's corpus, pomade is something she hates, and nobody is told to say it three times; her question leads to a knife threat if it goes unanswered, to a chase that ends in death if the listener screams and is caught, and in one case to a second question — "even like this?" — once the mask is off. The tidy symmetry in which every answer is fatal is spelled out on Yokai.com. Psychological readings exist: Akiyama Satoko in 1988 read the mouth as a "toothed womb" and the figure as the mother's tenderness and cruelty at once; Itō, who is polite about it, says that psychological work on this case looks to him like fitting the evidence to a conclusion already chosen, though it is interesting as a mirror of its moment. The sociologist Misumi Jōji, writing in 1991, filed the slit-mouthed woman among the urban legends that mutate by what makes a good telling rather than by what can be checked — he lists her top speed at 60 km/h beside the tsuchinoko's 80 and the human-faced dog's 150, and argues that it would be wrong to read those figures as witnesses exaggerating: what shapes an urban legend is the double criterion of being interesting as a story and staying inside the bounds of the plausible.