Who Actually Found the Challenger O-Ring Problem

On 11 February 1986, at the first public hearing into the Challenger explosion, a physicist took a piece of rubber O-ring, clamped it, and dropped it into a glass of ice water. He pulled it out. It had lost its spring.

That was the cause of the accident. That is the moment everyone remembers. And it is probably the most famous act of individual scientific insight of the last fifty years.

Richard Feynman was handed it.

The chain

Sally Ride, the first American woman in space, was on the commission. So was General Donald Kutyna of the US Air Force. Early in the investigation, Ride came into possession of an internal NASA document.

Kutyna described it years later, in an oral history for Popular Mechanics: she opened her notebook and gave him a piece of paper. A NASA document, two columns. Temperature, and the resiliency of the O-rings.

That is the accident — NASA’s own data, sitting in an internal file while a presidential commission searched for a cause.

Ride could not say it. She was a serving NASA astronaut and the most recognisable one alive; there was no quiet version. Kutyna could not say it either — the Air Force and NASA were bound together on the shuttle programme.

So Kutyna gave it to Feynman. Not directly. He invited him over and said:

“Professor, these carburetors have O-rings in them. And when it gets cold, they leak. Do you suppose that has anything to do with our situation?”

There was no carburetor. Kutyna wrote Feynman’s line for him and left the room. Two weeks later it was on national television, and it was a glass of ice water, and it was genius.

Kutyna kept Ride’s name to himself for twenty-six years, and said it publicly only after she died in 2012. Feynman knew an astronaut had supplied the document; he never learned which one. He wrote his own account of the investigation and died in 1988 without the story he was at the centre of.

The man who did say it

There is a comfortable reading of this: a system couldn’t speak, so it needed an outsider. That reading is wrong, and one name is enough to break it.

Roger Boisjoly was an engineer at Morton Thiokol, the contractor that built the solid rocket boosters. In July 1985 — six months before the accident — he wrote a memo about the seal design warning that the result would be, as the Associated Press reported it, “a catastrophe of the highest order — loss of human life.”

On the night of 27 January 1986, with the forecast below freezing, Boisjoly and his colleagues argued on a teleconference with NASA to stop the launch. Thiokol’s initial recommendation was: don’t fly.

Then a vice president called a five-minute caucus, and the engineers were excluded from it. Inside that break, according to Adam Higginbotham’s Challenger (2024), Jerry Mason turned to engineering manager Bob Lund and said:

“Now, Bob, you take off your engineering hat and put on your management hat.”

The recommendation flipped. They went back on the line and told NASA the data was inconclusive. Challenger launched the next morning at about minus two degrees.

Boisjoly testified to the commission and called the caucus “an unethical decision-making forum resulting from intense customer intimidation” — an extraordinary thing to say in public, under oath, about the people who still employed you.

They took his work away. He was reassigned, colleagues stopped speaking to him, and he had insomnia, depression and headaches for years. Two lawsuits were dismissed. He told the AP in 1988:

“When I realized what was happening, it absolutely destroyed me. It destroyed my career, my life, everything else.”

That same year the American Association for the Advancement of Science gave him its Award for Scientific Freedom and Responsibility.

He was right. He said so early, in writing, and under oath. It cost him everything and it saved nobody.

What the inner scorecard doesn’t cover

There is an old and appealing idea that what sustains you through something like that is knowing why you did it. Keep your own scorecard. Don’t worry what anyone thinks. You know what you did, and that is enough.

Boisjoly had that. Completely. He never had to revise a word of it.

It was not enough — not for him, and not for the seven people on that shuttle. An inner scorecard is what is left when you have already lost. It is a consolation, and a real one. It is not a strategy, and it should not be sold as one.

But the opposite conclusion is worse, and it is where this argument usually ends up: that integrity is naïve and the smart play is silence. That isn’t what happened here either. There were three people, not two.

The third option

Ride did neither. She didn’t stand up and she didn’t stay quiet. She read the room, worked out who in it could carry the thing further than she could, and handed it sideways. Kutyna did the same. Two weeks later it was on the evening news and NASA could not take it back.

Boisjoly did it the way we tell children to do it, and seven people died anyway. Ride did it the way nobody puts in a film, and it landed. The price of it landing was that no account of the investigation has her name anywhere near it — until a general decided, twenty-six years later, after she was dead and it could no longer cost her anything, that he was not going to carry it alone.

That is not cowardice. It is reading a room correctly and choosing the outcome over the credit. It is badly underrated, and nobody thanks you for it.

One caution, and it matters: Ride never explained herself. Kutyna’s account is on the record; hers is not. Everything anyone says about why she did it — including the generous reading above — is inference. Her reasons died with her.

What Feynman actually contributed

It would be easy to finish by taking the episode away from him. That would also be wrong.

Plenty of people are unpunishable and do nothing with it. Feynman was sixty-seven and being treated for a rare cancer during the three months of the investigation; he had two years to live. He built the demonstration, and he put it on live television where it could not be walked back. The commission chairman complained that “Feynman is becoming a real pain.”

At the end he threatened to withdraw his name from the report unless his own observations were published. They were, as Appendix F — which has held up far better than the glass of water. He asked people across NASA what they thought the odds were of losing a shuttle:

“The estimates range from roughly 1 in 100 to 1 in 100,000. The higher figures come from the working engineers, and the very low figures from management.”

A thousandfold gap inside one organisation. And the closing line, which earned its place on every poster it has ever appeared on:

“For a successful technology, reality must take precedence over public relations, for nature cannot be fooled.”

His contribution was not the insight. It was that he could afford to write that down — and that he did. Those are two different things, and only the second one comes from the person.

Seventeen years later

In 2003, Columbia came apart on re-entry. Seven more people died. A new board was convened, and Sally Ride sat on it. She is the only person who served on both.

That board wrote, in its own report, that “the management practices overseeing the Space Shuttle Program were as much a cause of the accident as the foam that struck the left wing.” In its list of cultural failures is a phrase describing organisational barriers that “stifled professional differences of opinion.”

That is a description of what happened to Roger Boisjoly, written seventeen years afterwards as though it were a new finding.

Ride read that. Having done what she did in 1986, and having watched what happened to the man who did it the other way.

Every organisation produces knowledge it cannot say out loud. The evidence is rarely the scarce part. What’s scarce is someone who can afford to say it — and most of the time, that person is not going to be you. Which makes the question less would I be brave and more who in this room could carry it, and do they know I have it.

This is a US story throughout — NASA, a presidential commission, a US contractor. The organisational pattern is the claim; none of it is offered as a transferable figure.

---

Sources: Dean, “An Oral History of the Space Shuttle Challenger Disaster,” Popular Mechanics, 2016 · Feynman, “Appendix F — Personal Observations on the Reliability of the Shuttle,” Report of the Presidential Commission on the Space Shuttle Challenger Accident, 1986 · Higginbotham, Challenger, 2024 · Cook, The Burning Blue, 2021 · Columbia Accident Investigation Board Report Vol. I, 2003.