They Really Did Make It In A Lab. The Rumour, Not The Virus.

KINSHASA — More than forty million people have died of this disease. This newspaper is going to be careful for eleven hundred words and then say something flat at the end, and readers who want the flat thing now can have it: the virus was not made in a laboratory, and the rumour that it was, was.

There are two ways to know that. They were assembled by different people, for different reasons, and neither was built to answer this question.

The freezer

In 1959, a plasma sample was drawn from a man in Léopoldville — now Kinshasa — for a study about something else entirely. It was frozen and catalogued and nobody thought about it again for forty years.

In 1960 a lymph node biopsy was taken from a woman in the same city.

Both were eventually sequenced. Both contain the virus.

That alone puts it in central Africa two decades before the alleged laboratory work. But the part that does the real damage is subtler, and Dr. Odile Bakayoko-Renner spent an hour making sure our correspondent understood it.

The two samples are already very different from each other.

Why difference is the evidence

A virus made in a laboratory and released is a virus with one starting point. Its descendants would be closely related to each other for a long time afterwards, because they would all be recent cousins.

The 1959 and 1960 samples are not close cousins. They come from one city, one year apart, and the distance between them is large.

Run that distance backwards at the rate the virus mutates and the two lineages meet somewhere near the start of the twentieth century.

She was firm that the date is an estimate with a real confidence interval, that the interval is wide, and that we should print the width rather than a single reassuring number.

We have. The interval is decades across. Every part of it is before 1930.

The filing cabinet

The other way to know is that the people who started the rumour said so.

In 1983 a letter appeared in an Indian newspaper claiming the virus had been engineered at a United States army laboratory. The newspaper had been set up by the KGB. The letter was placed.

In 1986 a report gave the claim scientific-looking clothing, and it travelled: eighty countries before the decade was out.

In 1992 the head of Russian foreign intelligence acknowledged, publicly, that the KGB had run the campaign.

This is not a leak, a defector, or an inference. It is an intelligence chief describing an operation.

The two people we interviewed

Both were told the rumour by a doctor. Both believed it for years.

Neither is embarrassed, and this newspaper is not going to make them so. One of them said something our correspondent has been turning over since:

"You are telling me the Russians made it up. All right. Who told my doctor? And why did he believe it? He was not a Russian. He was a man in a room with me, and he was frightened of the same people I was frightened of."

The warning our historian insisted on

Professor Emeka Thorvaldsen-Achike read our draft and asked to reply, on the grounds that this piece will make readers feel clever and that feeling clever is the failure mode.

His argument is at the foot of this page and it is the reason to read the story.

The campaign did not work because its evidence was good. Its evidence was thin and was seen to be thin at the time. It worked because it was offered to people who had specific, documented, entirely real reasons to believe that medical institutions would experiment on them without telling them.

The KGB did not manufacture those reasons. Real institutions supplied them, over decades, to real people, and that record is not in dispute — some of it is in this newspaper's own archive, in the story about a programme whose expense vouchers survived when its files did not.

A state looked at a wound somebody else had made and put its thumb in it. That is the technique, and it does not work on a population with no wound.

The flat thing

So: the virus crossed from another species in west-central Africa around the beginning of the last century, and there are two samples in freezers that prove it was there by 1959.

And the story that it came from a laboratory was itself manufactured, deliberately, by a state, which later said so.

Both of those are the sort of sentence that sounds too neat. Both are checkable, by two entirely different methods, and this newspaper has printed where to check them.

The useful sentence is not the KGB did it. It is the one the professor ends on: a lie needs a true thing to stand on, and the true thing here was a history of medical betrayal that nobody had to invent.