The claim, examined
A fabrication with receipts
Another page with no retelling, and for the same reason as Komayō. This one is a confirmed modern invention with a named author and a year, and the useful thing about it is not the story — it's the method, which is documented at every step.
The lore, as it circulates
A person starves to death because others refused them food.
They come back — and they go to the graves of the people who refused, dig up the corpses, and gnaw the heads.
It's a good grudge. It has the right shape: a specific injustice, a proportionate and grotesque revenge, and a target list.
That shape is exactly the problem.
The receipts
Created by Saitō Morihiro, for Bessatsu Shōjo Friend, in 1966.
Same author. Same magazine. Same year as the giant skeleton assembled from the famine dead — the single most successful fake in Japanese folklore.
One workshop, one year, two monsters that are now in encyclopedias.
And then the art-laundering, which is the part worth studying.
Saitō attached an Edo woodblock print as his "source" — an image of a male spirit with a severed head in its mouth. Real period art. Genuinely old.
And, per Murakami, never intended to depict any such creature. A painting with its own unrelated subject, conscripted as evidence for a monster invented in 1966.
That's precisely what happened with the giant skeleton, whose "ancient depiction" is a Kuniyoshi print about a completely different legend.
Then the propagation chain, with dates:
1968 — a magazine yokai encyclopedia picks it up. 1972 — an illustrated guide reprints it. After that — general reference.
Two reprints. That's all it took to become canon.
The full technique
Set the steps out and you have the clearest teaching case in the collection:
1. Find an orphaned Edo image — something real, striking, and unexplained. 2. Write it a grievance — a motive with the right folk grammar. 3. Publish in a children's magazine, where nobody checks. 4. Wait for the encyclopedias.
Every stage has a name and a year attached to it, which is rare. Usually you can prove that something is invented but not how.
Why it worked
And here's the thing I keep coming back to.
Saitō's invented grudge — food withheld, heads collected — is a pastiche of real doctrine. The tradition genuinely does run on karmic retribution in the medium of the offence: the oil thief who returns as fire, the priest who buried for money and must eat the parish's dead, the coin thief whose arms grow the eyes of the coins she took.
He wasn't making something up out of nothing. He was writing fluently in a tradition he understood well.
Which is the uncomfortable lesson of this entry, and worth stating plainly:
Good fakes are fluent. The thing that makes a fabrication spread isn't sloppiness — it's that somebody competent understood the grammar and wrote a sentence that belonged.
Field guide
At a glance
- Behavior
- The circulating lore: a person starved while others withheld food returns to dig up the withholders' graves and gnaw the heads. A perfect-sounding folk grudge — which is exactly the problem.
- Powers
- Fluency: the invented grievance pastiches real karmic-retribution grammar, which is why it passed.
- Weaknesses
- Its receipts — created by Saitō Morihiro for a 1966 girls' magazine, the gashadokuro author's same workshop, same year.
- Oddities
- The method study of fabrication: find an orphaned Edo image, write it a grievance, publish in a children's magazine, wait for the encyclopedias — every step documented with names and years.
- The "source" painting was never meant to depict any such yokai — a period artwork conscripted as evidence.
Overview
The circulating lore: a person who starved to death while others withheld food becomes a spirit that visits the graves of those who refused it, digs up their corpses, and gnaws the heads [ja-kubikajiri]. A perfect-sounding folk grudge — which is exactly the problem.
The dossier
- Created by Saitō Morihiro for Bessatsu Shōjo Friend, 1966 [ja-kubikajiri] — the same author-and-year as the gashadokuro fabrication: the fake-lore dossier's most productive workshop.
- The art-laundering: Saitō's "source" image — an Edo ukiyo-e of "a male spirit with a severed head in its mouth" (Ippitsusai Bunchō lineage) — was, per Murakami, never meant to depict any such yokai [ja-kubikajiri]. A period painting conscripted as evidence.
- The propagation chain, dated: Mizuki's magazine yokai encyclopedia (1968) → Satō Yūmon's illustrated guide (1972) → general reference [ja-kubikajiri]. Two reprints made it canon.
- Modern taxonomy: "fictional yokai" [ja-kubikajiri].
The reading
The fake-lore dossier gains its clearest method study. The gashadokuro showed Saitō inventing from whole cloth; the kubikajiri shows the fuller technique — find an orphaned Edo image, write it a grievance, publish in a children's magazine, wait for the encyclopedias. Every step is documented with names and years, which makes this entry the authenticity-spectrum's best teaching case: put the Bunchō painting, the 1966 caption, and the 1972 "reference book" side by side, and you have fabrication with its receipts. And note what Saitō understood: his invented grudge (food withheld, heads collected) is a pastiche of real doctrine — the tradition's actual karmic-retribution grammar (shio-no-chōji, kasha) — which is why it passed. Good fakes are fluent.
Two-track labeling
The image
Citations
The record
- [ja-kubikajiri]「首かじり」ja.wikipedia via WebFetch (retrieved 2026-07-26), Tier C carrying Murakami.modern reference Japanese Wikipedia's article on this subject, as retrieved in 2026 — often carries regional detail the English article lacks.All-source record
Where the evidence thins
Open questions
- Saitō Morihiro dossier consolidation (gashadokuro + kubikajiri; scan for a third).
- Bunchō painting identification (artist name renderings vary — verify Ippitsusai Bunchō vs summary's garble).