Doppelganger in Japan: The Double Said to Foretell Death

In a short story published in 1917, Akutagawa Ryunosuke put a university teacher at the centre of a quiet catastrophe: three separate times, the man sees doubles of himself and his wife. Nine years later, Kajii Motojiro built a fantasy story on a doppelganger experience he had already set down, in a 1925 essay, as something that had happened to him. Both writers were reaching for an idea their readers already carried — that a double of a living person is not a curiosity but a warning.

Quick facts

• Verdict: Folklore, not an internet invention. The belief that seeing one’s own double foretells death is recorded in Japan long before the web, with Edo-period antecedents and early twentieth-century literary treatments; newer school-ghost variants were layered on top of an older core.
• Name: The German loanword doppelganger (ドッペルゲンガー) is the usual modern Japanese term; older Japanese writing places the phenomenon under rikonbyo (離魂病), a term for the soul parting from the body.
• Type: Folk belief about a death omen, later reworked as school ghost stories and urban legends.
• Earliest known record: No single first attestation is established. Edo-period texts are cited as containing similar accounts, and one of them, the tale of the shadow sickness, was copied by Akutagawa Ryunosuke into a student notebook.
• Where it spread: From folk belief into modern Japanese literature, then into schoolyard legend and Japanese occult media.

What the story says

The core of the Japanese version is short enough to fit in a sentence: if a person’s double appears, in the belief, that person will soon die. In older Japanese usage the phenomenon was filed under rikonbyo, a category covering cases in which the spirit is understood to have separated from the body — the double is not an impostor or a twin but something shed by the person it copies.

Later retellings give the idea rules. In one school ghost story, called Double, a pupil who goes to the lavatory late at night finds someone there with the same face; if that figure turns around and smiles, the pupil is said to die of a cold within three days. Another figure, Koito-san, is said to appear twice in a person’s life, the second appearance marking the moment of death, and is credited with a small catalogue of warning signs: a five-yen coin vanishing from a wallet, a pet dying within two months, a pinprick hole appearing in the left ring finger with blood seeping out. A third story, Doppe-chan, is tied to an anecdote from October 1995 in which a fifth-year primary school pupil was reportedly seen in two places in the same school at once; when the pupil mentioned it, friends replied that meeting the double in person would have been fatal. The variants differ in detail, but the arithmetic is constant: the double is a countdown.

Earliest known record

There is no fixed earliest record for this legend, and the materials behind this article do not establish one. The oldest pointers named are two Edo-period texts cited in webMu’s column on the subject: the account of the shadow sickness in Oshu Hanashi, written by Tadano Makuzu, and the account of rikonbyo in Sorori Monogatari, both described as containing similar phenomena. That is a genealogy rather than a dated first appearance — it shows the idea was in circulation in written Japanese well before the modern era, without telling us when it began or where it first surfaced.

One link in that genealogy can be checked more closely. A reference case by Osaka Prefectural Central Library, in the National Diet Library’s Collaborative Reference Database, confirms that Kage no Yamai (影の病), sometimes attributed to Akutagawa, is not his work but a tale in Oshu Hanashi, Tadano Makuzu’s collection of strange tales from the Tohoku region. As a student at the First Higher School and the Imperial University, Akutagawa copied it into Shozu Shii (椒図志異), his notebook of strange tales, leaving out roughly the last four lines and noting Oshu Hanashi as the source.

The firmest modern datings are literary. Akutagawa’s short story Futatsu no Tegami appeared in the magazine Kuroshio in September 1917. Kajii’s essay Deinei, in which he recorded a doppelganger experience as his own, dates to 1925, and the fantasy story he built from it, K no Shoten, to 1926. Those are publication dates for works of literature, not documentation that anything happened.

What is documented

Three things are solidly on the record.

First, the belief itself. In Japan the appearance of a double has been believed since early times to be an omen of that person’s death, and it was understood as a form of rikonbyo, a phenomenon connected with the separation of body and soul. That is the load-bearing fact: the death-omen reading is Japanese folk belief, not a later import dressed up as one. The belief has also been recorded in recent form: a 1998 paper on young people’s folk beliefs in the journal of the Shizuoka Folklore Society, indexed in the yokai database of the International Research Center for Japanese Studies (Nichibunken), sets down the saying that whoever sees their doppelganger will be killed by it within a week. It is catalogued as a collected belief, not an account of events.

Second, the literary trail. Akutagawa’s 1917 story takes the form of a letter to a police chief from Sasaki Shin’ichiro, a thirty-five-year-old university teacher. It turns the motif into psychological horror by repetition: the persistent “second self” appears before him three times, and the doubling extends to his wife. The letter-writer also cites European cases — Becker, Dr. Werner, Professor Stilling — and remarks that the appearance of a doppelgaenger seems to foretell death, but not always.

Kajii’s route runs the other way, from an experience he wrote down as personal in 1925 to a fictional treatment the following year. The text of Deinei, first published in the magazine Aozora in July 1925, never uses the word doppelganger; it describes a first-person narrator on a moonlit, snowy road whose shadow becomes a vivid, living self walking ahead, watched from a position like the moon’s. Separately, Akutagawa is recorded as having said at a roundtable that his own double appeared once at the Teikoku Gekijo and once in Ginza, and as not having regarded the matter as simple illusion. That remark is documented as a remark; it belongs to literary history, and nothing establishes that any such event took place.

Third, the newer variants. Doppe-chan, Double and Koito-san are recorded in a column by Asazato Itsuki published by webMu, the online arm of Gakken’s occult magazine. They are set down there as legends — stories that circulate, with their own internal rules — and that is how they should be read.

What is contested

The word itself is not Japanese, and the concept is not exclusively Japanese either. English-language reference work records the doppelganger as a figure long documented in Western folklore and literature: the exact double of a living person. What the Japanese material adds is the specific fatal reading, and its own older vocabulary for it. Whether the Edo-period shadow-sickness accounts and the imported German term describe the same thing, or two similar things that merged once the loanword arrived, is a question the available sources do not settle.

The individual stories are on weaker ground still. The 1995 school case is an anecdote recorded in a magazine column, not an investigated incident; no source confirms that a pupil was seen in two places at once. The warning signs attributed to Koito-san — the missing coin, the dying pet, the marked finger — are legend material, not observations, and nothing in the record treats any of this as a medically established phenomenon. Reading the stories as folklore is not a hedge; it is the only reading the sources support.

One more distinction is worth holding onto. The old belief is a way of explaining something after the fact: a person was seen where they could not have been, and later something happened. The school versions invert that, turning the double into a rule-bound game with conditions and deadlines. The same image is doing very different work in the two cases.

What the sources add

The primary texts and reference records consulted for this revision agree with the familiar account in outline, but they differ from it at several points of detail.

The first concerns Kajii. Deinei is usually summarised as a doppelganger account, but its text never uses the word. What it presents is a moonlit shadow that becomes a living self, watched from a position likened to the moon’s. The doppelganger label is a way of classifying that scene, not the scene’s own vocabulary.

The second concerns Akutagawa’s story. Summaries tend to reduce Futatsu no Tegami to its three sightings. The text is more argumentative than that. Its letter-writer assembles European cases in the manner of a case file and says outright that a double does not always foretell death. The story does not simply restate the death-omen belief; within the fiction, the omen is one possibility among several.

The third concerns the Edo-period material. Earlier summaries can only say that Oshu Hanashi is cited as containing something similar. The library record adds a concrete, checkable fact: Akutagawa copied its shadow-sickness tale by hand into his student notebook Shozu Shii and recorded where it came from. That establishes the copy and its source note. It does not give a date for the Edo text, and the sources used here do not go further than that.

Finally, the 1998 folklore record shows the belief in a sharper form than the old phrasing of an omen, with a fixed deadline of one week attached. It was recorded as something young people said. Like the school stories, it is best read as a record of how the idea was told, not as evidence that it describes anything real.