The rulings
The Only Blinded Test on the Record Came Back Negative, and the Case Proceeded Anyway
Eight rulings graded against the register: the contract is always signed and was enforced without challenge once in four hundred years, the best-known sign is younger than the moon landing, and the one piece of real evidence on file was treated as noise.
By Casimir Vane The Registers · Filed Under Protest 2026-10-07
These are the rulings on the Possession Register, which is where the evidence is.
Fifty-four filings is enough to stop taking the folklore on trust and start grading it, which is the one advantage of a pile this size: a rule that independent tellers keep arriving at separately has done something a rule invented once has not. I have graded the rumours under the paper’s standards rather than mine, which is to say I could not put a thumb on the scales even where I badly wanted to. A rule is established when the record carries it consistently and can say why, attested when everybody repeats it and nobody explains it, contested when the filings disagree and both sides have documents, and unsourced when it turns out to have been invented once and believed ever since.
One ruling I would have preferred not to print at all, and it is the first one below.
The rulings
Confirmed, contested, or tested once and ignored
Every citation is a filing on the register.
| The rumour | Verdict | The desk's finding |
|---|---|---|
| Holy water burns | ONE BLINDED TEST, NEGATIVE, IGNORED Attested, unexplained | The register's only controlled trial came back negative, and the case proceeded anyway. The long versionTap water is used in place of holy water, and the subject reacts anyway. The only filing that contradicts the rule is the same filing that carries it, which is a strange sentence to write and an accurate one. The desk records the finding without satisfaction: the record contains one piece of real evidence about its own central claim, and the record ignored it. Enters the record DM-1973-EXO, in a scene that nobody in the film acts on Carried by Contradicted by |
| The rite works | NO CONTROLLED TRIAL ON FILE Contested | A published procedure, four centuries of use, and nothing anybody could call a trial. The long versionThe desk is grading a procedure and not a belief, and it wants the distinction on the record. What the procedure has going for it is real: a promulgation date, a revision history, a delegated permission, and an exclusion requirement stated as caution in 1614 and as an instruction in 1999. None of that is evidence that it works, and the register declines to cite it as though it were. A document existing, being revised, or being taught on a course establishes that somebody wrote it down, which is a different finding and is filed as one. On the evidence side there is a single filing, and even that one resolves on a self-sacrifice rather than on the procedure. Against it: two filings carrying a case in which the subject died, one in which the exorcist is a fraud filming his own method, one in which an assessment unit finds most of what it assesses mundane, and a possession case of 1634 that ended in a conviction and a sentence carried out. Enters the record DM-1614-RIT, as a written procedure with prerequisites Carried by Contradicted by |
| The head turns all the way round | NOTHING READ BEFORE 1971 Unsourced | The best-known sign on the register appears in nothing the desk has read from before the 1970s. The long versionThe single most confidently known fact about possession appears in no filing the desk has read on this register dated before 1971. The Faust line does not have it. Milton does not have it. The chapbook, the plays, the hearing in the shtetl, the letters and the apartment building do not have it, and neither procedural document supplies it. The desk states two limits and states them as limits rather than as evidence. It has not established whether the novel of 1971 or the film of 1973 introduces the sign, and does not assign it to either. It has also not read the press report of 1949 in the original, which is why that filing is cited nowhere on this finding: a card the desk cannot vouch for is not evidence, and labelling it a contradiction would be borrowing an authority it has not got. What is left is still a result. The best-known physical sign in the entire file is younger than the moon landing, and everything carrying it is downstream of one case. Enters the record the 1971 and 1973 filings; the desk has not established which of the two supplies it Carried by Contradicted by |
| It has to be let in | INHERITED FROM THE VAMPIRE FILE Contested | True of the contract half of the register and untrue of the possession half. The long versionThe Undead Register established that the vampire must be invited. This file inherits the rule and then splits down the middle over it. On the contract side it holds absolutely: there is an offer, and there is an instrument, three times over. It holds in a room where a passage is read aloud, which is the register's clearest case of an admission being performed rather than signed. And it holds, in a fashion, in the two filings that damage it most — a consent obtained in 1967 by telling somebody a different story, and a host in 2018 supplied willingly by a household on behalf of the person it concerns. Against it: one subject attached in childhood, before any question of consent could arise; one arrival arranged at a birth by other people; and one case the desk will not reason about at all, in which a young woman died. The desk's finding is that consent is a real mechanism and a worthless safeguard, and that the two are not the same claim. Enters the record DM-1587-HIS, where a signature is the whole of the consent Carried by Contradicted by |
| A signed contract binds | ALWAYS SIGNED, RARELY ENFORCED Contested | Executed in every filing that has one. Enforced without challenge, on this register, once. The long versionEvery instrument on this register is properly executed and almost none of them survives contact with anybody willing to read it back. The chapbook of 1587 supplies the form. Exactly one filing on this page then enforces it without challenge, on a stage, in 1604, on the date stated. After that the filings that reach an outcome are six escapes: a wager redrafted so the wording can be attacked, a jury setting one aside, an escape clause written into the agreement and honoured, seven wishes granted to the letter and failing on assumptions the client never stated, a shop with its terms in the small print, and a contest treated as a tribunal. One unchallenged enforcement, six escapes, four centuries, and two filings the desk counts on neither side: a transferable instrument of 1820 that finds no taker, and a fifteen-year sample whose enforcements it has not counted. The desk offers that to readers as practical advice rather than as literary criticism: the counterparty in this file has an excellent standard form and an atrocious record in front of anybody who reads it. Enters the record DM-1587-HIS, with disclosed terms and a signature in the debtor's own hand Carried by Contradicted by |
| The Antichrist arrives by clerical error | DELIVERY ERROR, ATTESTED Attested, unexplained | Three stories, three hospitals, three failures at the point of handover. The long versionWidely carried, never explained, and unanimous as far as it goes. Three separate stories on this register arrange the arrival of the Antichrist, one of them filed twice as a novel and as its film, and all three arrange it through a hospital. One is a substitution at birth. One is a swap performed by nuns that sends the child to the wrong household. One comes by way of neighbours in the same building who are a coven. The desk has no mechanism to offer and does not pretend to one. What it records is that the most consequential delivery in the literature is placed, three times out of three, in the one part of a household's life where the paperwork is done entirely by strangers and nobody is given a copy. Enters the record DM-1967-ROS, where the neighbours are a coven Carried by |
| The Devil is a landlord, an employer or a shopkeeper | CONFIRMED: IT IS PROCUREMENT Established | The counterparty in this file is almost always somebody you have a commercial relationship with. The long versionThe best-supported finding on the register, and the one the desk went looking for because it is this paper's beat. The desk enters no contradictions against this one, and says why rather than leaving the blank to be read as confidence. The filings that do not fit are not filings that disagree; they are filings with no commercial relationship in them at all, which is an absence of evidence and not evidence of absence. Enters the record DM-1966-MAS, where the plot of a novel is an apartment number Carried by |
| The crossroads deal | NO CONTEMPORARY SOURCE ON FILE Unsourced | It arrives on this register fully formed, with nothing behind it. The long versionThe register carries two filings that run on this story and not one document that supports it. There is no contemporary source on this page naming the musician the legend is now attached to, and the desk has seen the attribution disputed — the claim being that it was told of a different man first — but it has not checked that claim against a source and therefore does not print it as fact. What can be printed is the structure, which is the register's familiar one: a story with no founding document, adopted by repetition, acquiring a named subject somewhere along the way, and arriving in the popular record so completely formed that the works using it feel no need to explain where it came from. The desk has now filed this pattern on three registers running and has stopped being surprised by it. Enters the record nowhere on this page; the story is already finished when the earliest filing that uses it picks it up Carried by |
Findings
Tap water. I have spent a whole register of my own complaining that an unsourced claim survives on seniority, and I stand by it. This is worse, and I concede it with no grace whatever. A claim never tested is merely old. A claim tested, failed, and carried forward regardless has been decided, and no repetition after that is honest. The failure is not credulity: somebody thought of the test and ran it. The failure is what was done with the answer.
The procedure asks the right question and nobody has ever answered the second one. I approve of exclusion criteria, since they are the whole of this paper’s method for mould, smells, noises and damp, and since nobody has ever extended me one. The manual is better than its record. That is a sentence I could not write about a single institution in my own file, and I have looked.
The best-known sign in the whole subject is younger than the moon landing, as far as the desk has read, and the table says how far that is. The result is familiar. My sunlight rule was invented in 1922. The outbreak file’s brains were invented in 1985. Here is the third, on the third subject, in exactly the same shape, and I am beginning to think the monsters are not the problem.
Always signed, and hardly ever enforced. The old line is that the devil is in the detail. On this record he wrote the detail, and the table has what became of it. This register turns on consent at a signature; mine turns on consent at a door. I am obliged to report that the signature is better documented, better drafted, and considerably easier to get out of.
The landlord ruling is the one this paper went looking for. I file the registers, so I will say the unwelcome part out loud: I have been one.
The one perimeter that holds anywhere on the register is a protective circle, in a filing from 1968. It is not crossed. That is the same result the outbreak register reached from the other direction across ninety years of film, and neither of us enjoyed reaching it.
The reading list is at Selected Possession Readings.
— Casimir Vane
