On January 28, 1986, I watched the Space Shuttle Challenger explode live on television from work. I’d spent seventeen years watching space exploration as a story of triumph.
Challenger was the first time I watched it kill people in real time.
Seven people died that morning. Christa McAuliffe, a high school teacher from New Hampshire, was supposed to teach lessons from orbit, and NASA had arranged for students across the country to watch her launch at school. A lot of children saw it happen.
What most of them never learned is that the disaster had been predicted the night before, in writing, by the people who built the rocket boosters. One of those engineers was handed the launch recommendation and refused to sign it. His name was Allan McDonald, and he deserves to be remembered on January 28 along with the crew.
Who were the seven astronauts aboard the Space Shuttle Challenger?
Dick Scobee commanded the mission, STS-51-L. Michael J. Smith was the pilot, on his first spaceflight. The mission specialists were Judith Resnik, Ellison Onizuka and Ronald McNair. Gregory Jarvis, an engineer from Hughes Aircraft, flew as a payload specialist.
Christa McAuliffe was the seventh. NASA had picked her from more than 11,000 applicants for its Teacher in Space Project, and she was going to be the first teacher in space. Her lessons from orbit were going to be broadcast to classrooms.
They were test pilots, scientists, an engineer and a schoolteacher.
Every one of them trusted that the people on the ground had checked the work.
What happened to the Space Shuttle Challenger on January 28, 1986?
Challenger lifted off from Kennedy Space Center at 11:38 a.m. Eastern time. It was 36 degrees Fahrenheit at launch, and the ice team had found icicles hanging from the launch tower that morning. A seal in a joint of the right solid rocket booster failed almost immediately. Puffs of black smoke came out of the joint in the first second of flight. Then a flame burned through the side of the booster and into the external fuel tank. At 73 seconds, about 46,000 feet up, the vehicle broke apart.
The crew cabin came through the breakup largely in one piece and fell to the Atlantic. It hit the water about two minutes and 45 seconds later.
Nobody survived.
President Reagan postponed his State of the Union address and spoke to the country from the Oval Office that evening. He ended with lines from the poem “High Flight”: the crew had “slipped the surly bonds of earth” to “touch the face of God.”
Why did the Challenger O-rings fail in the cold?
Each solid rocket booster was built in sections, and the joints between them were sealed with rubber O-rings, a primary and a backup. When the boosters lit, the joints flexed. The O-rings had to spring into the gap fast enough to keep burning gas inside.
Cold rubber doesn’t spring. It stiffens.
The coldest shuttle launch before Challenger, in January 1985, had a calculated O-ring temperature of 53 degrees, and it came back with hot gas blowing past the primary seal. Engineer Roger Boisjoly had already warned his managers in a memo dated July 31, 1985, that if a joint failed, the result would be “a catastrophe of the highest order.”
On the morning of the launch the seals were far colder than anything they’d flown at. Overnight, ice team readings near the right booster fell as low as 8 degrees.
From Roger Boisjoly's 1985 memo to the shuttle's return to flight in 1988.
Who was Allan McDonald, the engineer who refused to sign the Challenger launch?
McDonald was director of the Space Shuttle Solid Rocket Motor Project at Morton Thiokol, the Utah company that built the boosters. He was at the Cape as Thiokol’s senior representative for the launch.
On the evening of January 27, Thiokol engineers joined a teleconference with NASA. They went through their charts and recommended against launching below 53 degrees. NASA’s booster project manager, Lawrence Mulloy, didn’t like it. McDonald later recalled him asking whether Thiokol wanted him to wait until next April to launch.
Thiokol’s managers asked for a recess and talked it over among themselves. During that recess, one of the senior executives told the vice president of engineering, Robert Lund, to take off his engineering hat and put on his management hat. When the call resumed, Thiokol said the data was inconclusive and recommended launching.
Nothing about the cold had changed during that recess. Only the people in charge of the answer had.
Nothing about the cold had changed during that recess. Only the people in charge of the answer had.
NASA wanted the recommendation in writing, signed by Thiokol’s man at the Cape. McDonald refused to sign it. “I wouldn’t want to be the person that has to stand in front of a board of inquiry,” he said. Thiokol sent a recommendation signed by another manager in Utah, and the countdown went on.
McDonald watched the launch the next morning. Seventy-three seconds later, he knew exactly what had happened.
What did the Rogers Commission find about the Challenger disaster?
President Reagan appointed a commission led by former Secretary of State William Rogers. Neil Armstrong, Sally Ride, Chuck Yeager and the physicist Richard Feynman were among its members. In a closed hearing a couple of weeks after the accident, a NASA official walked the commission through the launch decision and left out that the engineers had objected. McDonald was sitting in the back. He raised his hand and told the commission, “we recommended not launching below 53 degrees Fahrenheit.” Rogers asked him to come down to the floor and repeat it.
On February 11, Feynman dropped a piece of O-ring material into a glass of ice water at a televised hearing and showed that it lost its spring.
Anyone watching could see it.
The commission’s report, delivered June 6, 1986, blamed the failure of the O-ring seals in the right booster’s aft joint, and it blamed the decision-making process that sent Challenger up anyway. Feynman added his own appendix, and it ends with a line every manager should have on the wall: “For a successful technology, reality must take precedence over public relations, for nature cannot be fooled.”
What happened to Allan McDonald after he told the truth?
Thiokol demoted him.
Boisjoly told the commission about it, and Representative Edward Markey of Massachusetts introduced a joint resolution that would have barred Thiokol from future NASA contracts over the way it treated its engineers.
The company backed down.
It promoted McDonald to vice president and put him in charge of redesigning the booster joints. His redesign flew from the shuttle’s return to flight in 1988 until the program ended in 2011.
He retired in 2001 and spent the rest of his life teaching engineers and managers about ethical decisions. In 2009 he published Truth, Lies, and O-Rings with the historian James R. Hansen. He died on March 6, 2021, in Ogden, Utah, at 83.
He had a line about regret that I think about. “Regret for things we did is tempered by time,” he said. “But regret for things we did not do is inconsolable.”
What can managers learn from the Challenger launch decision?
I ran a retailer’s technology department for twenty years, and I cracked down on weasel words. “I think I can have it by Friday” got the same answer every time: that’s a weasel word, stop it. My staff didn’t like it. I didn’t care, because a hedge is how bad news gets lost on its way up the chain.
The Challenger decision is the worst version of that I know of.
The engineers said no. By the time the answer came back from the recess, no had become “inconclusive,” and inconclusive turned into go. Nobody lied, exactly. They softened the answer until it said what the schedule needed. I wrote about that habit in The Hidden Cost of Weasel Words.
The Challenger engineers said don’t launch below 53 degrees. It was 36. Their managers called the data inconclusive and launched anyway. Seven people died.Share on X
The sociologist Diane Vaughan called the bigger problem the normalization of deviance. The boosters had been coming back with damaged seals for years. Each time nothing blew up, the damage became a little more acceptable, until a known flaw was just part of flying.
I’ve watched the same thing happen in IT. A warning that fires every week stops getting read. Security folded into IT reports to people who are judged on uptime and budget, so the person saying “stop” works for the person who wants to go. I cover why that structure fails in the Cybersecurity Hub. The engineers at Thiokol had the same problem. They were right, and they didn’t have the authority to make right stick.
The lesson is simple to say and hard to live.
The person who knows the most about the risk needs the power to stop the launch, and the people above that person need to want to hear it.
How do people remember Challenger on January 28?
NASA holds its Day of Remembrance on the last Thursday in January. It honors the crews of Apollo 1, Challenger and Columbia, along with everyone else who died in the space program. In 2027 it falls on January 28, the anniversary itself.
The crew’s families founded the Challenger Center for Space Science Education to carry on the mission’s teaching. Unidentified remains of the crew were buried at the Space Shuttle Challenger Memorial at Arlington National Cemetery in May 1986. In 2022, divers searching for a World War II aircraft off the Florida coast found a 20-foot piece of Challenger on the sea floor.
I live in Florida now, and I watch SpaceX launches rise from Cape Canaveral. I still think grounding the shuttle for 32 months after Challenger was the wrong call. SpaceX works the way I wished NASA had: when one blows up, they fix what broke and launch another one. What broke has to get fixed, and Allan McDonald is the reason the shuttle’s boosters did.
If you mark the day, say the seven names out loud. Then tell somebody about the engineer who wouldn’t sign. If you’ve lived a story like his and want it told, I’ve written about how a ghostwriter can help with a whistleblower memoir.
My take
Challenger wasn’t an accident in the way people use the word. The engineers called it the night before, in writing, and their managers talked themselves out of hearing it. Allan McDonald refused to sign, and then he refused to stay quiet, and his company punished him for both. Remember the seven on January 28. Remember him too. If someone who knows the risk tells you no, don’t ask them to change hats.
Seventy-three seconds
Seventy-three seconds is about how long it takes to read this paragraph out loud. Seven people had that long. The warning that could have saved them had been given hours earlier, by a man who then had to watch.
There are more days worth marking on the Literary Holidays Calendar, including Petrov Day, for another man whose judgment mattered more than the machine. I wrote about the space launches that shaped me, Challenger among them, in I Watched the Moon Landing at Eight Years Old. On January 28, think about the seven, and about the engineer who wouldn’t sign.
The Thiokol engineers kept their judgment and got overruled anyway. My book covers how people and organizations give up judgment one reasonable shortcut at a time, and what it costs.
- Space Shuttle Challenger disaster, Wikipedia
- Allan J. McDonald, Wikipedia
- Remembering Allan McDonald: He Refused To Approve Challenger Launch, Exposed Cover-Up, NPR
- Report of the Presidential Commission on the Space Shuttle Challenger Accident, NASA
- Roger Boisjoly and the Challenger Disaster, Online Ethics Center