914 Names, 340 Beds, And A 1962 Assumption Nobody Revisited
The fire certificate permits 512 persons. The catering contract is priced for 400 covers. The gap between the 1,124 people the Instruction names and the bunks is a planning assumption taken from a Tuesday in September 1962.
Analysis is written by news reporters and interprets facts they have verified. It is not opinion, and it takes no position on what should be done.
CORSHAM, United Kingdom — The fire authority permits 512 persons on the site at any one time. The dormitory schedule counts 340 bunks in four blocks. The catering contract is priced to feed 400.
The Instruction names 1,124 people: 914 post-holders in the roster this newspaper counted in June, and a further 210 in Annex D — drivers, signallers, two cooks a shift, a chaplain.
Documents nobody classifies
We did not go looking for the accommodation plan, which we assumed we would not get and did not get.
We went looking for the things a building generates whatever is inside it. A fire certificate, because the authority has no interest in the Instruction and every interest in exits. A catering contract, because food is bought on a framework and frameworks publish. A laundry round, because sheets are washed by a firm in Chippenham that answers its telephone.
The three agree. Four blocks: 96 bunks, 84, 64, and 96. Three hundred and forty sets of linen a week.
The fire officer's schedule of conditions explains the 512: 340 asleep, 172 on watch. It is the only document in this story that assumes shift working, and the only one written by somebody outside the department.
Eleven letters
We wrote to eleven departments with one question: how does the arithmetic close?
Nine did not answer. One replied that bed allocation is held elsewhere and declined to say where.
The eleventh sent us forty-one pages of the current Instruction, including page 11, apparently by attaching the wrong file. We took advice, transcribed it, and told the department eleven days later. It has not asked for it back.
Assumption 4(b)
Page 11 answers the question in a single sentence, and it is not concealed. It has never been concealed.
Planning Assumption 4(b) is that 62 per cent of designated post-holders will not reach the site.
Thirty-eight per cent of 914 is 347. There are 340 bunks. The plan is seven short, and the shortfall is not the interesting number.
The interesting number is the 62.
A Tuesday in September
It is not a judgement about roads, or warning time, or what people do when they are frightened. Nobody modelled it.
It is an observed arrival rate. Exercise BROADLANDS, September 1962. Seventy-one participants were called forward. Forty-four did not arrive. Forty-four of seventy-one is 61.97 per cent.
The appendix gives a reason for every one of the forty-four. Thirty-one of them are the word leave. Six were not notified. Four are transport. Three have no reason recorded.
The exercise was held on a Tuesday in the second half of September.
Six revisions
The Instruction has been revised six times since. The assumption has been carried through all six without being re-derived, and the mechanism is not sinister. Each revision's terms of reference cover changes since the previous edition. Assumption 4(b) has not changed. Therefore it is not in scope. Therefore it is not looked at.
That is the whole of it. An unexamined figure is not always protected. Sometimes it is merely never the subject of the meeting.
Block D
The plan shows Block D as 96 bunks and has since 1974.
It has a concrete floor, power, a lighting circuit and no partitions. There are no frames in it. Stacked against the north wall are ninety-six mattresses, still in polythene printed with a 1988 date.
Take Block D out and the site sleeps 244 against an assumed arrival of 347.
What this desk takes from it
Hester Cadwaladr-Nnamdi, who drafted the fourth revision, read this piece and told us we had the wrong subject: that 1962 is doing the work of an argument, that no arrival study has ever been done by anybody including us, and that if nine hundred people came to that gate nine hundred would be let in and put on the floor.
She then told us something worse than anything we found, which is that she raised 4(b) in 1988 and was told that reopening it would reopen the accommodation schedule, and the accommodation schedule would reopen the capital line.
We asked for that minute. It is not in the series. Her letter is below.
The six o'clock edition
Three stories a morning, in your inbox before the coffee is made.
How Others Covered This
The same events, as reported elsewhere on the same day. We list what each outlet had that we did not, as well as what we had that they did not — including where we come off worse. Why we print this.
-
The Hollow PostTHE BEDS DO NOT ADD UP. SO WHO GETS LEFT ON THE SURFACE?
Read the shortfall as a selection already made, and spent 2,000 words asking which 776 names are on the list that decides it.
Had that we did not
The bunk count, which is right, and which they got the same way we did, from the linen contractor.
Left out
Planning Assumption 4(b), which is printed in the Instruction they quote from, on the page before the page they quote.
-
ApertureThe Mattresses In Block D
An essay on polythene, preparedness and the English talent for building a room and then not entering it.
Had that we did not
The best sentence anybody has written about this site, which is about the smell of a store that has been sealed since 1988.
Left out
The arithmetic. The essay never establishes how many bunks exist, and the number is obtainable in one telephone call to a laundry contractor in Chippenham.
-
They Buriedthis newspaper914 Names, 340 Beds, And A 1962 Assumption Nobody Revisited
Went looking for beds rather than names, and found the answer in a fire certificate, a catering contract and a laundry round.
Had that we did not
All four documents in full, the eleven letters we sent, and the one department that answered the question by accident.
Left out
We transcribed a document a department sent us in error and told the department eleven days later, once it was safely typed up. We would do it again. That is not the same as it being nothing. — V. Ashcombe-Doyle, standards editor
Right of Reply
They Buried contacted Hester Cadwaladr-Nnamdi, formerly of the relocation secretariat, drafter of the fourth revision on 9 June 2026 in person, and on 3 July with the full draft, the plan, the fire schedule and the 1962 appendix, with four weeks and no conditions. Replied in writing on 29 July 2026, asking that it be printed whole or not at all. Printed unedited.
I will take your points in order, and then I will tell you what I think your story is actually about, which is not what you think it is about.
First. A planning assumption is not a prediction. It is a number you agree to design against so that the design can be finished. Nobody in that secretariat believed that 62 per cent of post-holders would fail to arrive, in the way that one believes a weather forecast. We believed that a building has to have a size, that the size has to be defended in a capital bid against a Treasury which is very good at its job, and that a figure with a document behind it survives that meeting and a figure with a judgement behind it does not. You have found the document. It is a thin document. I have never pretended otherwise and neither did the four people who used it before me.
Second, and I would ask you to print this part particularly. You write that the shortfall means somebody is left out, and your rival paper writes it in capital letters. There is no list of who is left out. There has never been a list of who is left out. The Instruction names posts and states an arrival rate; it does not select, it does not rank, and there is no annex that decides between a permanent secretary and a signaller. If nine hundred people arrived at that gate, nine hundred people would be accommodated, and they would be accommodated appallingly, on floors, in corridors and in the vehicle bay, and the fire officer would have a great deal to say about it afterwards, and everyone would be alive. Beds are the cheapest thing on that site. What binds is water, air handling and the twenty-one days of tinned food you have correctly identified and then walked straight past.
Third. You have written a story in which the year 1962 does the work of an argument. It is a very effective device and it is not evidence. The question is not whether the figure is old. The question is whether it is wrong, and to answer that you would need an arrival study, and no arrival study has ever been done, by us or by you. You had the resources to commission one. You commissioned a genealogist for your last piece on this subject. You did not commission this, because a story about a Tuesday in 1962 writes itself and a story about arrival modelling does not.
Fourth, the revisions. You are right that the terms of reference of each revision cover changes since the previous edition, and you are right that this is how an assumption survives without being re-derived, and it is a real defect. I said so in 1988 in a minute you have not obtained, and I was told that reopening 4(b) would reopen the accommodation schedule, and reopening the accommodation schedule would reopen the capital line, and that was the end of that. Print that. It is worse than what you have written, and it is the true answer to the question you asked eleven departments.
Lastly, Block D. You end on the polythene. It is a fine paragraph and I understood exactly what it was doing, which is carrying the weight your evidence will not carry. Ninety-six mattresses in a dry room are ninety-six mattresses. I would rather you had ended on the minute I have just described, which you do not have, than on a photograph of a store.
Published unedited under our right-of-reply guarantee.
How was this story?
We publish the result, whatever it is. Reader verdicts appear on the front page and in our newsroom metrics.
Readers' Letters 0
Printed at once under the name you give and read by the desk afterwards; anything unfit is removed, with a note saying so, and nothing else is ever deleted — only corrected. Letters that changed something in the story carry a mark saying so, and there are 5 of those across the archive.