A lecture course of the 1890s, listing things people in the provinces believed, gives this one in a single clause.
In the Kōriyama district of Iwashiro, around 1873 or 1874, when people there saw photographs — everybody said at the time that a person who is photographed has his life shortened.
That is the whole entry, and it is the right place to begin, because it fixes the date. Photography reached ordinary Japanese towns within living memory of everyone in this note. It arrived with a superstition already attached. And within about twenty years the same technology was producing the opposite kind of story: not that the camera takes something out of the living, but that it puts something in from the dead.
The lecturer had a pen name — Fushigi-an, the hermitage of wonders — and a newspaper column. A sequel to his collection of a hundred tales of the uncanny was running in a Tokyo daily.
While it was running, somebody wrote in with a challenge. He cited cases of a phantom appearing in a photograph, offered them as evidence that ghosts exist, and asked how the master of the hermitage of wonders proposed to explain that.
Somebody else answered on his behalf at the time. He notes this in passing, and it is a detail worth keeping: by the 1890s the argument had enough people in it that a question addressed to him could be fielded by a stranger.
He adds that he had been asked the same thing repeatedly for years — and then he gives away the best fact in the section. He had seen two of the photographs himself.
The one he printed in the column has a source and a place.
According to a report in a Sendai newspaper the previous year, there had been a strange affair in the Kamikita district of Aomori prefecture: a ghost that appeared in a photograph.
Several people had been photographed together, and when the picture was looked at, there was something in among the figures. Examined closely, it was somebody who had died of an illness some years earlier — a person the family could identify. They stood about looking at the photograph and wondering.
So they asked the photographer.
His answer is the whole reason this note exists, and it takes about two sentences.
This is no strange affair, he said. If a glass plate that has already been exposed once is not properly cleaned off, and is then used again, the earlier image can remain on it faintly. So this sort of thing can happen by accident. It is not worth calling a wonder.
The lecturer, independently, had gone and asked a photographer of his own, and got the same account in almost the same words: a plate not polished clean, used a second time, keeps the dim shape of what was on it before.
Two photographers, in different parts of the country, each asked by a different person, each giving the same technical explanation without hesitation.
Which means the ghost in that family’s photograph was almost certainly a portrait of somebody else — an earlier sitter, in a working studio, on a plate that was scraped down between customers and that morning was not scraped enough.
It is worth being exact about what that does and does not settle. It explains how a second face got onto the plate. It does not explain why the second face was recognised as one particular dead person rather than as a stranger, or a smudge, or nothing at all — and that recognition is the part the family actually experienced.
The family recognised the face of their own dead. That part I do not think anyone could talk them out of, then or now, and it is not really a claim about photography.
Having disposed of the mechanics, he goes on to the part he cares about, and it is not what a modern sceptic would say.
He does not argue that the photograph was faked. He argues that a ghost is not the kind of thing that could be in one.
Even supposing ghosts exist, he writes, there is no reason why a ghost should be furnished with a body. If it had a body, it would not be a ghost. Ordinary people think ghosts have shapes, and therefore think they can appear in pictures, and this is a large mistake.
That is a premise about what the word means, not a finding about silver halides. Our note on Inoue Enryō sets out his method and its cost; this is a clean sample of it, and it is doing more work here than the glass plate.
Then he does something I have not seen anywhere else in this material. He compares the two national ghosts as though reviewing two pieces of engineering.
The Japanese ghost, he says — a thing of the middle ages onward — has hands but no feet, and is not transparent. The Western ghost has both hands and feet, and its body is transparent like glass.
Compare the two, and the Western one has rather more value as a ghost.
His reasoning: a Japanese ghost with hands and no feet is extremely hard to make sense of — if it has hands, one would expect it to have feet. Whereas transparency is at least somewhat close to being formless, which is what a ghost ought to be.
And then the last line, which is an instruction to artists: in future, if you are going to paint a ghost, paint it as dim as possible, and in the air.
There is one thing in the passage he does not appear to notice, and it is the reason to put these two arguments on the same page.
The Western ghost is praised for being transparent like glass. The Japanese ghost photograph is produced by a sheet of glass that was not wiped clean.
The same substance supplies both the ideal ghost and the fake one, four paragraphs apart, in a lecture arguing that neither is real. Spirit photography in Japan did not need debunking later; it arrived with its explanation attached, given by the tradesman who made the print, and the explanation and the phenomenon are in the same newspaper report.
What outlived both, as usual, was the picture: a dim figure, in the air, with no feet — which is very nearly the specification he wrote down for painters at the end of the section, as a joke.
The readings in “The Word Doing Two Jobs” are KWAIDANOTE’s own.
Found an error? Write to kwaidanote@gmail.com — corrections are made on this page and dated.