Kokkuri-san: Japan's coin-and-paper Ouija, explained

Readers who met the name in the first chapter of Jujutsu Kaisen, or in the comedy series Gugure! Kokkuri-san, usually meet it as a piece of set dressing: a name you call, and something answers. What almost never comes across in those appearances is that Kokkuri-san (こっくりさん) is not an ancient rite at all. The trail leads back to a damaged American sailing ship on the Izu coast in the 1880s, and to a Meiji-era philosopher who, within a few years, set out to explain the craze away.

Quick facts

• Verdict: folklore, not an internet invention — a custom traced to the Meiji era, when Western table-turning arrived in Japan.
• Name: Kokkuri-san, from kokkuri, later written with the characters 狐狗狸.
• Type: a divination game, based in part on Western table-turning.
• Earliest known record: 1887, Inoue Enryo’s Yokai Gendan, which places the practice’s arrival at Shimoda about three years earlier (around 1884); the word itself is in a newspaper headline by July 1886.
• Where it spread: from Izu along the coast to central Japan, the Kyoto–Osaka region and Boso, then to Tokyo and beyond, according to Inoue; a second and much larger wave came among schoolchildren in the 1970s.

What the story says

In its present-day form the game is simple. A sheet of paper is marked with a small fixed vocabulary — “yes” and “no”, a torii gate, numbers, and the hiragana syllabary — and a coin is placed on it. The participants rest their fingertips lightly on the coin, a question is addressed aloud to Kokkuri-san, and the coin moves: to one of the fixed answers, or from character to character, spelling something out.

That is essentially the whole of the form. There is no special equipment beyond what sits in a school bag, which is a large part of why the game travelled the way it did. The paper-and-coin version is usually presented as a close relative of the Western Ouija board, and the practice as a whole as partly based on Western table-turning — the parlour séance in which sitters place their hands on a table and wait for it to tip.

What these descriptions give is the procedure, not the cause. Whether the coin moves of its own accord is exactly the question the game exists to pose. In Japan it has been taken seriously for a long time: an early systematic answer was published in the 1880s (see below), and a 2004 book by Anzai Ikuro still asks, in its title, why Kokkuri-san gets things right.

Earliest known record

The date usually given for the practice’s arrival is 1884, and it can now be traced to its source, a book of 1887, rather than to a summary of it. Inoue Enryo’s Yokai Gendan, first published in May 1887 and reissued in October 1900, records that “about three years ago” an American sailing ship was damaged near Shimoda, on the Izu Peninsula, and that Americans who stayed on in connection with the wreck taught the practice to the local people. Counting back from 1887 gives 1884. Later retellings, including a Toyo University article and a Gendai Business feature, repeat the essentials: foreign sailors at Shimoda, local people copying them, and the practice spreading through the country’s ports.

The same passage explains the name. According to Inoue, the Americans called the practice by an English name the local people could not say, so it was named after the way the lid on its bamboo stand tipped — kokkuri, a word for a nodding or tipping motion. The characters 狐狗狸 were attached to the name afterwards.

There is also an independent check on the timing. The Nihon Kokugo Daijiten, as reproduced on Kotobank, cites as its earliest example of the word a headline in the Asahi Shimbun of 16 July 1886 that names kokkuri alongside kami-oroshi, the calling down of a deity. That puts the word in print about ten months before Inoue’s book appeared. The newspaper item itself was not examined for this article, and it confirms the word and the craze rather than the Shimoda story.

Two caveats still apply. The year 1884 is an inference from Inoue’s phrase “about three years ago”, not a dated record, and he gives the shipwreck story as something he was told. A retrospective account by one investigator, however careful, is a different kind of evidence from a report filed at the time. Some references also date Yokai Gendan to 1892/1900; the colophon of the Aozora Bunko edition gives May 1887 for the first edition and October 1900 for the reprint.

What is documented

Several details make the 1884 account look plausible rather than merely tidy. The trademark for the modern Ouija board dates from the 1880s — the same decade in which Kokkuri-san appears in Japan. The two practices are not simply similar; they surface at the same moment, on opposite sides of the Pacific, out of the same Victorian enthusiasm for table-turning.

The adaptation is documented too, and it is the kind of detail that is hard to invent. Desks were not common in Japanese homes at the time, so the table of “table-turning” had to be improvised. A standard Japanese dictionary defines kokkuri as a tray set on three bamboo rods tied into a tripod; other accounts describe a box in the same position. The practice, in other words, did not arrive intact. It arrived as a technique and was rebuilt out of whatever Japanese households actually had — which is also why the version most people know today involves a coin and a sheet of paper rather than furniture.

Inoue also mapped how the craze moved. By his account it began in Izu, reached Owari and Mino and the Kyoto–Osaka region some months later, and at the same time spread through the Boso provinces. It entered Tokyo through Fukagawa, moved through Nihonbashi and Kyobashi, and reached Ushigome and Koishikawa; from there it went on to Saitama, Gunma and Shinano, to Kyushu, and finally to Oshu in the northeast. It is an unusually detailed contemporary map of how a fad travelled in the 1880s.

He did not stop at history. Toyo University, which Inoue founded as the Tetsugakukan in 1887, describes how he set up a Fushigi Kenkyukai (mystery research society) in 1886, published his kokkuri findings first in the Tetsugakkai Zasshi and then collected them in Yokai Gendan. Both the university and Gendai Business describe his experiments, run with different combinations of participants by age, sex and temperament. His conclusion was that two things work together: yoki-iko, the expectation of an answer already held in the mind, and fukaku-kindo, unconscious muscle movement. The participants, in his view, move the lid themselves without knowing it.

The second documented turning point is much more recent. In the 1970s the manga artist Tsunojiro featured Kokkuri-san in Ushiro no Hyakutaro, and the game became enormously popular among children as a result — popular enough to draw stern warnings from educators. The Meiji import had become, within a century, a classroom phenomenon that adults felt obliged to push back against.

What is contested

The weak point is not the existence of the practice but the precision of its origin story. A named year, a named stretch of coast and a group of foreign sailors make for an unusually clean beginning, and clean beginnings are worth treating with care. Every retelling examined here goes back to one Meiji-era writer, and his own text frames the Shimoda episode as a report he received, dated only relative to when he was writing.

The other open question is the mechanism. The game is only partly based on table-turning, which leaves room for whatever else fed into it. Inoue’s answer — expectation plus unconscious movement — is the explanation these sources give, and none of them supports a supernatural cause. Even so, the game is still played precisely because players treat the question as open, and anyone telling you flatly what Kokkuri-san “really” is goes beyond what the sources support.

What the sources add

Reading Inoue’s own text changes the familiar summary in small but real ways. The popular version has a lone sailor washed ashore; Yokai Gendan describes a damaged American sailing ship and Americans who stayed on at Shimoda afterwards. The popular version gives 1884 as a fact; Inoue gives “about three years ago” in a book of 1887, so the year is a reasonable inference rather than a record. His book is also dated to 1892 in some references, while the edition used here gives 1887 for the first printing.

The primary text agrees with the modern telling on the essentials — Shimoda, Americans, table-turning, and spread through ports — but it adds what summaries leave out: a named route through Izu, central Japan, Kyoto–Osaka, Boso, Tokyo and the provinces beyond. It also shows that the original equipment was a lid or tray on a bamboo tripod, not a coin on paper.

Later retellings compress further. The Gendai Business feature places the Shimoda episode in the “early Meiji” period and speaks of an Ouija board, where Inoue describes table-turning in the mid-1880s. A dictionary entry supplies the one piece of evidence that does not depend on Inoue: the word kokkuri in an 1886 newspaper headline, before his book appeared.

The real people and works involved are few and well defined — Inoue, the port of Shimoda, Yokai Gendan, and in the 1970s Ushiro no Hyakutaro — which makes this history easier to check than most.

Adaptations

The 1970s manga boom is the most consequential real-world ripple on record: Tsunojiro’s Ushiro no Hyakutaro is credited with the craze among schoolchildren and the backlash from educators that followed. Since then the name has circulated widely in comics and animation, with appearances in several manga and anime works, among them the first chapter of Jujutsu Kaisen and Gugure! Kokkuri-san — the entry points through which most readers outside Japan first encounter it. It is worth keeping the order straight: the practice came first, in the 1880s, and the fiction has been drawing on it ever since rather than the other way round.