Wednesday, October 7, 2026
They Buried
We Dug It Up
⌕ Search

Hilde Marchmont-Owusu Copied One Register In 1994 And Told Nobody For 31 Years.

She was a county records archivist told to prepare accession series 44 for withdrawal. Over eleven evenings she copied its 61 pages. Three independent tests say the copy is genuine; 297 of its 341 case files survive nowhere else.

Eleven evenings at a dry-toner machine. What she took out was a copy. What was taken away, on 2 April 1994, was everything else. Engraved for They Buried

The 61 pages are the accession register for series 44 of a county records centre. A register of that kind is not a case file. It is the index to case files: what came in, under what programme, on what date, shelved where.

Series 44 lists 341 of them, numbered, under the programme's own name.

The originals left the building by van on 2 April 1994 and have not been produced since.

The instruction

In March 1994 she was told to prepare series 44 for "withdrawal for conservation" by a state agency. The instruction came by telephone and was confirmed on a single sheet she was not permitted to keep.

She was 52, eleven years into a job she would hold for another seven. She had processed withdrawals before. This one carried no numbered receipt, which every previous one had.

She did not object, refuse, or report it. She stayed late.

Figure Accession register, series 44 — a section through 61 pages, and what each part can still be checked against
Page 1 · Cover sheet and county seal0Pp. 2–11 · Series description, authorityNames the programme in full. Namesthe agency that asked.−11p.Pp. 12–19 · Microfilm targets in marginsThe county's own camera marks.Second of three tests.−19p.Pp. 20–52 · Case list, 341 files numberedNumber, intake date, facilitycode, two-letter disposition.−52p.DISPUTEDPp. 53–58 · Disposition column, inserts297 entries carry codes nopublished county key defines.−58p.IN NO PUBLISHED RECORDPp. 59–61 · Withdrawal note and receiptOne initialled signature. Thestate says the form is not theirs.−61p.DISPUTED
The register is a finding aid, not a case file. It tells you what existed, under what name, and where it was shelved. Two of its five parts can be corroborated against surviving county paper; two cannot; one is a receipt the state archives says is not its receipt. Depths are page numbers, counted from the cover sheet. They Buried, from the Marchmont-Owusu copy and the surviving series 41 and series 52 registers

Eleven evenings

The machine was a dry-toner copier in a corridor off the reading room, and it took the register a page at a time because the binding would not lie flat.

Her sister remembers the arithmetic being the point: she worked out how many evenings she could stay without it becoming a thing anyone remarked on, and then stayed exactly that many.

Eleven evenings. Sixty-one pages. She put the copy in a stationery box and the box under the bed in the back room, and that is where it stayed until August 2025, when she wrote the instructions.

The three tests

We did not want to publish a photocopy on trust, so we had it examined three ways by three people who did not know of one another.

The toner is a dry two-component styrene-acrylate consistent with machines sold before 1996. Dr. Sunniva Achterberg-Rho, who was shown the pages with the cover sheet masked, cannot date it closer than a decade and says so.

Pages 12 to 19 carry registration targets in the margins — camera marks, from filming, that a copyist would have no reason to draw. Delmar Ostrowski-Nnaji ran the county's microfilm camera from 1979 to 1998 and named them as his own equipment before he was told what series he was looking at.

The accession stamps sit in sequence between two unrelated registers the county still holds, series 41 and series 52, with the same platen wear developing in the right direction on the right dates.

Three tests, three methods, three people. All three land.

The 297

Forty-four of the 341 numbers can be found somewhere else — in a docket index, a transfer slip, a surviving box list. Two hundred and ninety-seven cannot.

Not in the county's finding aids, not in the state's, not in the federal depository schedule, not in the two published concordances. We looked for eleven weeks and found nothing.

Her copy is now the only index of who was in Project Monarch.

Thirty-one years

She never tried to sell it. Two dealers in the 1990s bought county paper by the pound in that part of Ohio and one of them, still trading, has no record of her name. She approached no publisher, no lawyer and no broadcaster.

In 2011 she wrote 400 words to this newspaper's letters desk, describing a withdrawal she had witnessed and not naming the series. She received a form reply. That reply is in our file and it is our fault, and it is printed in the coverage table above under our own name.

She married Kwesi Owusu in 1976 and he died in 2009. There were no children. She kept bees, badly, by her own account, and was for nineteen years the treasurer of a hospice league.

What she was afraid of

Her sister is 81 and does not soften it. She was afraid of losing the pension, and later of being made to look foolish, and later still of being the only person left who remembered why it mattered.

"She used to say the box was doing the waiting for her," Mrs. Lisle said. "I think she meant she couldn't."

The written instructions are one paragraph. They give a newspaper's postal address, no individual's name, and a single condition: that the copy go afterwards to a public institution where a member of the public can ask for it by reference.

The state archives, in its reply below, says the 1994 withdrawal was not its withdrawal — that it never used the phrase, never used vans, and can find no authorisation on any form it is able to inspect. We have printed that entire.

What it means is that a van came, somebody signed, and the only page in the world that records it was copied eleven evenings running by a woman who then said nothing for thirty-one years and left instructions.

Sources & Method

We were left the copy, not given it: her written instructions named this newspaper's postal address and no individual. Before reading it as evidence we had it tested three ways by three people who did not know of each other, two of whom were shown the pages with the cover sheet masked. We then took the copy to the county reading room and measured its accession stamps against two unrelated registers the county still holds. The state archives had the draft, the figure and all three reports for eleven days before publication.

Who we spoke to

  1. Junie Marchmont Lisle, Her sister; the only person told before 2025. Interviewed in person at her home over two afternoons, and again by telephone 26–27 July and 4 August 2026 Produced the box, the bed it stood under, and the sealed envelope of written instructions.
  2. Dr. Sunniva Achterberg-Rho, Forensic document examiner. Commissioned by this newspaper; given the copy with the cover sheet masked and no account of where it came from May–June 2026 Toner is a dry two-component styrene-acrylate consistent with machines sold before 1996. Cannot date it closer than a decade.
  3. Delmar Ostrowski-Nnaji, Retired county microfilm camera operator, 1979–1998. Interviewed twice by video call, then shown pages 12 to 19 without being told what they were June 2026 Identified the marginal registration targets as the county's own camera before he was told the series number.
  4. Surviving accession registers, series 41 and series 52, County records centre, public reading room. Examined on site; accession-stamp impressions photographed under raking light and measured July 2026 Stamp sequence and platen wear on the series 44 copy fall between the two, in the right order, on the right dates.
  5. Halvard Ossory-Bennett, Deputy State Archivist for Collections. Sent the full draft, the figure and all three test reports; replied in writing after eleven days August 2026 Denies the withdrawal was his agency's. Printed in full under Right of Reply.

What we could not confirm

  • Where the originals went on 2 April 1994. The county gate log records a van, a time and a plate. The plate traces to a haulage firm dissolved in 2003, whose records passed to a successor company that has acknowledged our letters twice and answered neither. That company still trades. This is the next door, and it is not locked so much as unopened.
  • What the two-letter disposition codes mean. Two hundred and ninety-seven entries carry them. The published key for series 41 exists, we have it, and it does not fit — under that key, four of the codes are not codes at all. Somebody wrote a second key. We have not found it.
  • Whether 341 files means 341 people. A case number is a file, not a person. Re-openings, transfers between facilities and duplicate intakes all generate numbers. Until the codes are read, the honest figure is 341 files and an unknown number of human beings behind them, and we will not round it up.
Disclosure. This newspaper paid Dr. Achterberg-Rho $2,400 and Mr. Ostrowski-Nnaji $600, in advance, with no clause tying either fee to any finding. Nothing was paid to the family, who declined a fee and asked that the copy be lodged with a public institution once the reporting is finished. It will be.

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.

  1. The Meridian Telegraph
    The Woman Who Slept On The Monarch List For Thirty-One Years

    Ran her as a heroine of suspense and printed a photograph of the bed.

    Had that we did not

    The eleven evenings, accurately, and the sister's account of the hiding place.

    Left out

    All three authentication tests, which are the only reason anybody should believe a word of it.

  2. The Hollow Post
    Ohio Archive 'Cannot Explain' Vanished 1994 Transfer

    Wrote the state archives as the culprit, on the strength of a records officer's inability to explain a transfer his agency says it did not make.

    Had that we did not

    That no conservation withdrawal for series 44 exists in the state treatment log.

    Left out

    The rest of that sentence — that the state archives used numbered couriers, not vans, and that the deputy archivist volunteered this against his own interest.

  3. They Buriedthis newspaper
    Hilde Marchmont-Owusu Copied One Register In 1994 And Told Nobody For 31 Years.

    Had the copy tested three ways by three people who were told nothing about each other, then printed the state's reply entire.

    Had that we did not

    The toner chemistry, the camera targets, the stamp sequence, and every test result including the inconclusive one.

    Left out

    We ran the tests in nine days and took thirty-one years to knock on her door. She wrote to this newspaper's letters desk in 2011 and got a form reply, which is in our own file. We printed her obituary faster than we ever answered her letter. — V. Ashcombe-Doyle, standards editor

Right of Reply

They Buried contacted Halvard Ossory-Bennett, Deputy State Archivist for Collections on 5 August 2026, in writing, with the full draft, the diagram and the three test reports attached. Replied 16 August 2026. Printed in full and unedited, including the parts about this newspaper.

I have looked for the thing you asked me to look for, and I did not find it, and I want to be exact about what that does and does not mean, because your draft is not exact.

There is no conservation withdrawal for accession series 44 in our treatment log for 1994. There is none for 1993 or 1995 either. I checked the bound log, the card duplicate and the microfilm of both, and I had a colleague check them again after me. That is not a discovery of a cover-up. It is the ordinary result of searching for something in the place where it would be if it had happened.

What I will say, because you would find it anyway and because I would rather it came from me: 'withdrawal for conservation' is not our phrase. It has never appeared in our retention schedule. Our category was, and remains, 'treatment transfer', and in 1994 a treatment transfer moved by contract courier under a numbered receipt in triplicate — one copy to the holding office, one to us, one carried. We did not operate vans. I have been asked this question three times in my career and the answer has been the same three times.

So I cannot tell you that the collection on 2 April 1994 was authorised, and I cannot tell you it was not. I can tell you that whoever signed that receipt did not sign it on a state archives form, and that if any officer of this agency arranged it, he did so outside every procedure I am able to inspect. I am aware how that sentence reads. I have declined to soften it, and I would ask you to print it whole or not at all.

Now the part you will like less. You have three tests and you have oversold them. Dr. Achterberg-Rho establishes that the toner is old. Mr. Ostrowski-Nnaji establishes that a county camera once photographed something with those margins. The stamp measurements establish that an impression consistent with our platen was made in a plausible order. Not one of those findings speaks to the content of the pages. A document can be genuinely from 1994 and wrong on every line, and archivists spend their lives on exactly that distinction. You have authenticated an artefact and reported it as an index.

And there is a duty here that is not mine alone. If those 341 numbers correspond to people, some of them are alive, and a newspaper that prints a case list is publishing the fact of their file before any of them has been asked. You have withheld the numbers, and I note it, and I am glad of it. I would ask you to keep withholding them until someone has read the disposition column properly, because the day that column is read is the day this becomes a story about people rather than paper, and neither of us is ready for it.

Mrs. Marchmont-Owusu worked in this profession for thirty years and, on the evidence you have shown me, did the one thing our whole system depends on and cannot compel. I would have liked to meet her. I am sorry that the first time this agency wrote to her about series 44 was eleven days ago, to her sister, in the form of a request for access.

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.

60 verdicts · 96.7% loved it

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.

  1. No letters yet on this story. Yours would be the first.

Write to the desk