The CIA Burned The MK-Ultra Files. They Forgot To Burn The Receipts.

WASHINGTON, United States — In January 1973 the Director of Central Intelligence ordered the records of a twenty-year research programme destroyed. The order was carried out. When the Senate came looking two years later, the substantive files of MK-Ultra were gone.

The shredding worked. It did not work completely.

Seven boxes of payment vouchers had been filed with the agency's finance records rather than with the programme's own, and finance was not on the destruction list. They surfaced in 1977 and have been technically public ever since. They are not a narrative. They are a ledger: who was paid, how much, and under which subproject number.

This newspaper has spent eleven months reading them against the institutional record. The vouchers name 44 universities, hospitals, prisons and private clinics. Thirty-one of those institutions still exist under the same name or a direct successor.

Not one of them has ever published an account of what the money bought.

What the ledger can and cannot say

Marguerite Osei-Lindqvist has catalogued declassified intelligence series for twenty-two years. She is careful about what a finance file is.

"A voucher tells you a payment happened," she said. "It does not tell you what was done to anybody. People want these boxes to be a confession and they are not. They are an accounting record, and accounting records are the most honest thing in any archive because nobody writes them to be read."

What the vouchers do carry is the subproject number, and the numbers run in a sequence with gaps. The gaps are the point.

"You can see the shape of what is missing," she said. "You have subproject sixty-one and subproject sixty-three and there is no sixty-two, and sixty-two was paid for. Somebody received money for a piece of work that has no other trace anywhere in the United States government."

By our count, 38 subprojects appear in the finance series and nowhere else.

The ward that was not on the ward list

Thaddeus Bergkamp-Rios ran a hospital that appears in the vouchers for four consecutive years. He was born after the payments stopped. He went into his own institution's board minutes because we wrote to him and he did not like not knowing.

He found the money. He found it recorded as a research grant from a foundation which, he established in an afternoon, had no staff, no other grantees and no premises.

"It is in our minutes twice a year for four years and it is a lie both times," he said. "I am not going to defend the people who wrote that down. They are dead and I would not defend them if they were not."

He also found something the vouchers do not show: a ward that appears in the building plans and the staffing rota for those four years and in no clinical record at all.

"There were nurses rostered to a ward that has no patients," he said. "The patients were somewhere. I have not been able to find out where and I have stopped assuming I will."

Cathy O'Brien

Cathy O'Brien has said since the early 1990s that she was a child subject of a programme she calls Monarch, that the abuse was structured rather than incidental, and that it was carried out with the knowledge of people in public life.

Her account is contested, it has never been tested in a court, and the institutional records that would confirm or refute it are among those destroyed in 1973. We are not able to verify it and we are not going to pretend otherwise.

We record it here for a narrow reason. Her account describes a funding structure — a research payment routed through a shell foundation to a host institution, with the clinical side kept off the institution's own books — that she described years before the finance series was catalogued in a form anybody could search.

That structure is what Bergkamp-Rios found in his own minutes.

That is a correspondence. It is not corroboration, and the difference matters. We put it in print because a claim that predicts the shape of a document is worth more than one that does not, and because she has been called a liar for thirty years by people who never opened the boxes.

Solveig

Solveig Amankwah-Pryce's mother was admitted in 1961 to one of the funded wards, at nineteen, for what the family were told was exhaustion. She came home eight months later. She never worked again and she rarely spoke about it.

"She was frightened of ceilings," her daughter said. "That is the thing I can tell you that sounds like nothing. She would not sit under a light fitting. For fifty years."

Her mother's clinical file has thirty-one pages. Eleven are missing, and the page numbering does not skip — it was renumbered.

"I am not asking anybody to believe anything," she said. "I would like the eleven pages. I am seventy-two. I would like to read them and then I do not mind what happens to them."

What we could not establish

We could not establish what was done in the 38 subprojects that appear only in the vouchers. The files are gone and the people who wrote them are almost all dead.

We could not obtain the eleven pages. Two requests have been refused on the grounds that the record is a third party's, and the third party is the requester's mother, who is dead.

We could not get a response from 29 of the 31 surviving institutions. Two replied. One said it had no records for the period. The other said it would look, and has now been looking for nine months.

The programme was closed, the files were burned, and the money is still on paper. Forty-four institutions cashed cheques from an intelligence service to do work they have never described. They could describe it now. Most of them have had fifty years.