Disturbing History - The Challenger Disaster

Episode Date: August 26, 2026

On the night of January 27, 1986, a Morton Thiokol engineer named Allan McDonald refused to sign a document saying it was safe to launch the space shuttle Challenger the next morning.The engineers who built the shuttle's solid rocket boosters had recommended against flying in temperatures colder than 53 degrees, then reversed themselves after a 30-minute private caucus in which a vice president was told to take off his engineering hat and put on his management hat.The launch went ahead at 11:38 a.m. in 36-degree weather, and 73 seconds later Challenger came apart on live television.In this episode, Brian opens with his own memory of watching the launch as a 12-year-old home sick from school, on one of the few channels carrying it live, and then goes into the record most people have never read. He follows the paper trail from Roger Boisjoly's July 1985 memo warning of a catastrophe of the highest order, through the launch constraint that NASA's own booster manager imposed and then waived five flights in a row, to the ice team's disputed 8-degree reading on the right booster before dawn.He walks through the teleconference and the caucus using the sworn testimony of the men who were in the room, and explains why the people with authority to scrub the launch that morning had never heard the words O-ring in connection with it.The second half of the episode covers what the official record says about the crew after the breakup, drawn from the July 1986 report by physician and astronaut Joseph Kerwin.The crew compartment stayed intact, climbed to 65,000 feet, and took 2 minutes and 45 seconds to reach the water. Four emergency air packs were recovered and three had been switched on, including pilot Mike Smith's, which someone seated behind him would have had to reach back and activate. Three lever-lock switches on Smith's panel had been moved from their launch positions. Brian keeps to what the physical evidence supports and stops where Kerwin stopped.The episode also tells the story of how the Rogers Commission found out about the Thiokol recommendation only after a NASA budget analyst leaked documents to the New York Times and McDonald stood up in a closed hearing to say NASA wasn't telling the whole story. It covers Richard Feynman's ice water demonstration and the phone call from Sally Ride that set it up, a detail General Donald Kutyna kept secret for 26 years. It follows McDonald and Boisjoly home to Utah, where both men were reassigned within weeks of testifying, and traces what the rest of their lives looked like. Bob Ebeling, the engineer who first raised the alarm about the forecast, spent 30 years blaming himself and received a letter from one of the NASA managers who had overruled him two months before he died.The episode closes on the wreckage sealed inside two abandoned Minuteman missile silos at Cape Canaveral, the pieces the ocean has kept returning as recently as 2022, and Barbara Morgan, the backup teacher who watched from the ground with the families and flew to the International Space Station 21 years later.Email BrianJoin Our FREE NewsletterGet Brian's Books Leave Us A VoicemailVisit Our WebsiteHave a forgotten historical mystery, disturbing event, unsolved crime, or hidden conspiracy you think deserves investigation?Send your suggestions to [e-mail] History is a dark history podcast exploring unsolved mysteries, secret societies, historical conspiracies, lost civilizations, and the shadowy stories buried beneath the surface of the past.Follow the show and enable automatic downloads so you never miss a deep dive into history’s most unsettling secrets.Because sometimes the truth is darker than fiction.

Some stories were never meant to be told. Others were buried on purpose. This podcast digs them all up. Disturbing history peels back the layers of the past to uncover the strange, the sinister, and the stories that were never supposed to survive. From shadowy presidential secrets to government experiments that sound more like fiction than fact,

this is history they hoped you'd forget. I'm Brian, investigator, author, and your guide through the dark corner, of our collective memory. Each week I'll narrate some of the most chilling and little-known tales from history that will make you question everything you thought you knew.

And here's the twist. Sometimes the history is disturbing to us. And sometimes, we have to disturb history itself, just to get to the truth. If you like your facts with the side of fear, if you're not afraid to pull at threads, others leave alone. You're in the right place.

History isn't just written by the victors. Victors. Sometimes, it's rewritten by the disturbed. I was 12 years old, and I was home sick from school that day. Honestly, it's hard to say whether I was actually sick. Because the truth is, I pretended to be sick a lot in those days, because I didn't like school. But no matter if I was really sick or only pretending to be, that's the only reason I saw it live. I've talked to a lot of people my age who swear they watched the Challenger explode in real time, and most of them didn't. The big three networks weren't carrying the launch. Shuttle flights had become routine enough by 1986 that CBS, NBC, and ABC had all gone back to regular daytime programming.

CNN carried it, so did NASA's own satellite feed, which had been piped into thousands of classrooms because a schoolteacher was on board the shuttle that day. If you were a kid in a school with that feed, you saw it. But almost everyone else saw the replay that played over and over in the day. days after the explosion. I remember the countdown more than the launch. I remember the long hold and the announcer talking about ice on the pad. I remember the shuttle going up and looking like every other shuttle launch I'd ever seen, which by then was a lot of them, because I was the kind of kid who watched shuttle launches. Then there was a shape in the sky I didn't recognize, two white trails going off in different directions and a cloud in the middle that kept getting bigger. The announcer said

something about a major malfunction. I remember that phrase exactly, because it seemed like such a strange thing to call what I was looking at. I didn't understand what I'd seen for a while. I think I assumed the crew had ejected somehow, because in the world of a 12-year-old, that's what happens. Everybody bails out, right? It took my mother sitting down next to me and telling me what happened. Before I understood that seven people had just died on television, and that one of them was a teacher. I've spent the better part of four decades since then, assuming I knew this story. Faulty O-ring, cold morning, NASA pushed ahead when they shouldn't have. That's true, as far as it goes.

But when I finally sat down with the actual record, the commission report, and the testimony and the engineering memos, I found that the version I'd been carrying around was missing most of what makes this thing hard to sit with. The people who tried to stop it, and what happened to them afterward. what the evidence says about those 73 seconds, and what it says about the two minutes and 45 seconds that came after them, which is the part almost nobody talks about. On the night of January 27, 1986, a little after 11 o'clock Eastern Time, a man named Alan McDonald refused to sign a piece of paper.

McDonald was the director of the Solid Rocket Motor Project for Morton Thia Call, the Utah company that built the two booster rockets strapped to the side, of every space shuttle. He was at the Kennedy Space Center in Florida that night, sitting in a conference room with NASA managers on a three-way telephone conference with Thia Calls engineers back in Utah and NASA's Marshall Space Flight Center in Alabama. The subject of the call was whether it was safe to launch Challenger the next morning. With temperatures forecast to drop into the low 20s overnight, Thia Call's engineers had said no. Then, after a break in the call during which the next day,

NASA side wasn't listening in. Thiacol's managers came back on and said yes. NASA's marshal managers asked Thiacol to put that yes in writing, signed by someone with authority. And McDonald, who was the senior Thiacall person in the room in Florida, told them he wouldn't be the one to sign it. He said, according to his own testimony to the Rogers Commission, a few weeks later, that if anything happened to that launch, he wouldn't want to be the one who had to explain it to a board of inquiry.

He also said, in words he repeated under oath, that he'd be surprised if NASA would want to fly a solid rocket motor outside its qualification range and then have to explain that decision to the public. The paper got signed anyway. It was signed by Joe Kilminster, a Thia Call vice president in Utah, and faxed to Florida a little before midnight. The launch went ahead about 12 hours later. The shape of what McDonald did matters. He didn't have the authority to stop the launch. nobody at Thiacol did, technically. NASA was the customer. What McDonald had was the ability to put his own name on the recommendation or keep it off, and he kept it off.

That didn't save anyone. But it created a record, and that record is a large part of why we know what we know. To understand what McDonald was refusing to sign, I have to back up about six months. In July of 1985, a Thiacol engineer named Roger Boisjele, wrote a memo to his vice president of engineering. Boisjoli was a seal specialist.

He'd spent most of the previous year looking at what happened to the rubber O-rings inside the joints of the solid rocket boosters after they came back from flight. Each booster was built in segments, stacked on top of each other like sections of pipe, and shipped to Florida by rail from Utah. The joints between the segments were sealed by two O-rings,

a primary and a secondary. Each about a quarter-inch thick, and roughly 37 feet around. When the booster fired, the pressure inside was supposed to push the primary O-ring into its groove and sealed the joint within a fraction of a second. If it didn't seal, hot gas at more than 5,000 degrees would start burning past it. Bois Jolie had seen the evidence of that burning. Theacall called it blow by.

On the boosters recovered from a discovery flight in January of 1985, one year almost to the day before Challenger, He'd found that the primary O-ring in one joint had been eroded and that hot gas had gotten past it to the secondary. That flight had launched at 53 degrees, the coldest in the program's history up to that point. Boisjoli's judgment which he stated in his memo and later under oath was that the cold had made the rubber stiffer and slower to seal, and that the slower it sealed, the more time the gas had to get past it. The memo he wrote that July said that if the O-ring problem wasn't fixed, The result could be a catastrophe of the highest order, which he defined as the loss of human life.

He wrote that he was worried Thiakal was in jeopardy of losing a flight along with the launch pad facilities. Thiakal's response was to form a task force. Boyzoli was on it. By October, he was writing internal memos complaining that the task force wasn't getting resources or attention. He used the word helpless. That's his word. from a memo dated October 1985 three months before the launch. I read a lot of internal memos in my old job.

People don't usually write helpless in a memo unless they mean it. And here's the thing. Thiakol and NASA both knew about the O-ring problem. This wasn't hidden. The problem had been documented on flight after flight going back to 1981, and in December of 1982, NASA had formerly changed the joint's classification to what they called criticality, meaning that a failure would mean the loss of the vehicle and crew.

NASA's own engineers at Marshall had written memos about it. There'd been a briefing at NASA headquarters in August of 1985, where Thuyacol laid out the erosion history. The issue wasn't that nobody knew. The issue was that the shuttle had flown 24 times with this joint and come home every time. And somewhere along the way, the fact that it had kept coming home had started to feel like proof that it would keep coming home.

From headquarters, the problem looked settled. On August 19, 1985, Thiakol and Marshall presented the O-ring erosion history to NASA's associate administrator for space flight and his staff in Washington. The briefing charts, which the commission reproduced, laid out every case of erosion and blow by on every flight. The last chart concluded that the joint was safe to fly

while the company worked on a fix, and that the seal problem was, in the words on the chart, acceptable. Nobody at the meeting disagreed. The task force Bojoli was complaining about two months later was the fix that the briefing had promised. General Katina, on the commission, was the one who pushed hardest on why this kind of thing didn't travel.

He'd run launch programs for the Air Force, and what he saw in Marshall's paperwork was an organization that reported problems sideways and downward, but rarely upward. The commission's report used the phrase management isolation to describe it. Marshall solved its own problems or accepted them, and the shuttle program office in Houston and headquarters in Washington

heard about it in summaries that had been written to sound settled. Richard Feynman, the physicist who served on the Rogers Commission, later compared that reasoning to playing Russian roulette and concluding from each empty chamber that the gun was safe. There's a piece of paperwork from the summer of 1985 that I think is the most damning single item in the file, and it has Larry Malloy's name on it. In April of 1985, a Challenger flight came back with something worse than what Bois Jolie had found in January.

In one of the nozzle joints, the primary O-ring had been eroded through, and hot gas had reached the secondary O-ring and eroded that too. Both seals in a joint that was rated Criticality 1 had been damaged on the same flight. Under NASA's own rules, that kind of finding was supposed to trigger what they called a launch constraint. A formal flag that had to be cleared before any subsequent flight could go. Malloy imposed the constraint in July. Then, for each of the next five flights, he signed a waiver of it. The waivers were legal under NASA procedure.

The rationale written on them was that the erosion was understood and that the flights could be accepted with the risk. The commission found that these waivers weren't reported up the chain in any way that would have told the shuttle program's senior managers, there was a constraint in place at all. Malloy testified that he'd considered the constraint a routine tracking mechanism. The last of those waivers was signed for Challenger's flight. It wasn't discussed at the Flight Readiness Review on January 15th,

where the mission was formally certified as ready, because the O-ring issue had by then been carried on the books as an accepted risk for so long that it didn't rise to the level of the review. Jesse Moore, who chaired that review as head of the shuttle program, told the commission that he'd never been briefed on the launch constraint. The joint also had a second problem that gets less attention than the O-rings. Between the segments and the O-rings sat a layer of insulating putty, and the putty was supposed to protect the rings from the heat.

Thiacol had changed putty suppliers in 1982, after the original manufacturer stopped making it because of asbestos regulations. Marshall's own engineers had documented that the new putty formed blowholes under pressure, channeling hot gas directly at the O-rings in specific spots, rather than spreading it across the joint. The commission found that the putty, the joint's tendency to rotate open slightly under pressure, and the cold together had produced the failure. No one of them alone had done it. That's part of why Malloy's argument on the teleconference wasn't crazy on its face.

He was right that blowby had happened at 75 degrees. What he didn't have, and what Thiakaw's engineers didn't have eye. was a way to tell how much worse it got at 36. The Commission's answer, after the fact, was that not having that data was itself the reason not to fly. The forecast for the morning of January 28th called for temperatures in the low 20s

at the pad before dawn. Launch was scheduled for 9.38 in the morning. The coldest previous launch had been 53 degrees. On the afternoon of the 27th, Thiakal's engineers in Utah heard about the forecast and started putting together a case that they shouldn't fly. Bois Jolie and another engineer named Arnie Thompson did most of the work. What they had was thin in a specific way that ended up mattering.

They had test data on how the O-ring rubber behaved at different temperatures, and they had flight data on blowby from the boosters that had come back. What they didn't have was a test of a full-scale joint at anything close to 20 degrees, because nobody had ever expected to launch and weather that cold. The teleconference started at about 8.45 that night, Eastern Time. Thea Call's engineers presented about a dozen charts. The core of their argument was the 53-degree flight and the blow-by they'd found on it. Their recommendation, delivered by Bob Lund, Tha-Call's vice president of engineering,

was not to launch below 53 degrees. The pushback came from two NASA managers at Marshall. One was Larry Malloy, who ran the solid rocket boot, booster project for NASA. The other was George Hardy, the deputy director of science and engineering at Marshall. Malloy's objection, which he stated on the call and later repeated to the commission, was that Thiakol's data didn't prove temperature was the cause. He pointed out that there'd been blow by on a flight at 75 degrees too. He asked how Thiakol could set a launch limit of 53 degrees when the booster had been formally qualified to fly down to 40. Maloy also said something on that call that's

been quoted in every account of the disaster since. Several people who were on the line remembered it the same way, and McDonald testified to it. Malloy asked, more or less, my God, Thiochal, when do you want me to launch? Next April, stay tuned for more disturbing history. We'll be back after these messages. Hardy said he was appalled by Thiakal's recommendation. He confirmed the word under oath. He also said, in the same breath that he wouldn't launch over the contractor's objection. Both halves of that sentence are true, and I think both of them are load-bearing. Because after Hardy said he was appalled, Thaikal asked for a five-minute break to talk offline.

The break lasted about 30 minutes. What happened in that room in Utah during those 30 minutes comes to us almost entirely from the testimony of the people who were in it, given to the commission a few weeks later. The accounts line up on the important points. Four Thiacol managers were in the room, along with the engineers. The senior manager was Jerry Mason, a senior vice president. Boisjoli and Thompson made their argument again.

Boisjoli said he got up, walked to the table where the managers were sitting, and put the photographs of the blowby from the previous year's flight in front of them. Thompson, by his own account in Boisjolies, sketched the joint on a notepad to try to explain the ceiling problem one more time. Both men testified that at some point they realized nobody was listening to them anymore, and they sat back down. Mason then said, according to several witnesses including Boyce Jolie, that a management decision was needed.

He polled the four managers. Three said yes. The fourth, Bob Lund, who as vice president of engineering, had delivered the original no-launch recommendation an hour earlier, hesitated. What Mason said to Lund next is the single most quoted line to come out, of the Challenger investigation. Lund testified to it himself, and Bois Jolie and others confirmed it. Mason told him to take off his engineering hat and put on his management hat.

Lund voted yes. The engineers weren't polled. Bois Jolie testified that it was clear to him he was not being asked. When the teleconference resumed, Kilminster read out a new recommendation. It acknowledged that cold temperatures were a concern, and that the primary O-ring might take longer to see it. Then it said that the secondary O-ring would be in position to seal if the primary didn't, and that Thia Call recommended proceeding with the launch.

Malloy asked for it in writing, Kilminster signed it, and sent the fax. I've read that faxed recommendation. It's about a page long, and it's a strange document, because the first two-thirds of it are the engineering case for not flying. The reversal happens in the last few lines. Down in Florida, McDonald wasn't done. after he refused to sign, he kept arguing with the NASA managers in the room. He testified that he raised two other issues that night that had nothing to do with O-rings.

One was the ice that was expected to form on the launch tower overnight. The other was that the seas in the booster recovery area were running high enough that the two ships assigned to retrieve the boosters after launch had been ordered back toward port. McDonnell told the commission that he'd said if he were the launch director, he'd wait. and that he couldn't understand how NASA would fly under any one of those three conditions, let alone all three. He also testified that one of the NASA managers told him the ice and the sea state weren't his concern.

And there's one more piece of this that I think gets lost. The people who ran the launch that morning didn't know any of it had happened. The teleconference was a Marshall-level conversation, what NASA called level three in its flight readiness chain. Maloy and Hardy had the authority to accept the contractor's recommendation without kicking the question upstairs. So they didn't. Arnold Aldrich, the shuttle program manager who was at the Cape running the launch decision, testified that he wasn't told about Thia Call's objections.

Neither was Jesse Moore, the head of the shuttle program at NASA headquarters. The people with the authority to scrub the launch that morning had never heard the words O-ring in connection with it. The commission report put it this way. the decision makers were unaware of the history of problems with the O-rings and the joint, and unaware of the initial written recommendation of the contractor advising against the launch. By the time the launch team came in before dawn, the pad looked like something out of a different climate. Overnight temperatures had dropped into the low 20s. NASA had kept water trickling through the pad's fire suppression system to keep the pipes from freezing,

and the water had run out over the structure and frozen. Photographs taken by NASA's ice inspection team that morning show icicles hanging off the launch tower. Some of them were more than a foot long. The ice team was led by a NASA engineer named Charlie Stevenson. They went out to the pad three times that morning. On their first inspection, around two in the morning, they measured the air temperature at 24 degrees.

They reported ice on the tower, ice on the walkways, and ice in the water troughs at the base of the pad that were meant to dampen the sound. of the engines. Almost nobody on the launch team that morning was thinking about O-Rinks. The concern about the ice was that chunks of it would break off at ignition and strike the orbiter, either damaging the heat shield tiles or getting sucked into the main engines. Rockwell International, the company that built the orbiter, had that concern in a serious way.

Rockwell's people had been watching the ice reports from their facility in California. Rocco Patron, who ran Rockwell's space division, testified that he told his representatives at the Cape to tell NASA that Rockwell could not assure it was safe to fly. Two Rockwell managers at the Cape, Bob Glacier and Marty Shelfa Letty, said that's what they told NASA at the final pre-launch meeting. NASA heard it differently. Aldridge testified that he understood Rockwell to be expressing concern,

but not to be recommending against launch. He said that if Rockwell had said don't launch, don't launch, he'd have listened. The commission looked at both sides of this and concluded that Rockwell's position had been ambiguous enough that NASA's reading wasn't unreasonable. But it also found that the contractor's position should have been clearer and that NASA should have treated a contractor unable to say it was safe as a contractor saying it wasn't. I've sat in enough meetings where somebody was trying to say no, without saying no, to recognize what happened there. Both sides walked out with a different understanding,

and nobody stopped to make sure they'd heard the same thing. The ice team made one other measurement that morning that's been argued about ever since. Using a handheld infrared thermometer, they took readings on the surface of the boosters. On the left booster, they got readings in the mid-20s, which matched the air. On the right booster, near the aft field joint, the reading was somewhere around 8 degrees. The right booster's aft field joint is where the failure started. The commission investigated the reading and reported that the infrared device may not have been properly calibrated for the conditions, and that the low reading may have been caused by supercooled air flowing off the vent for the liquid oxygen tank,

which was on that side of the vehicle. Nobody's ever settled it. What's not disputed is that the reading was taken, that it was far lower than anything else on the pad, and that it wasn't passed up to the people making the launch decision as anything other than a data point. The launch had been scheduled for 9.38. Because of the ice, it was held. The ice team went out again and again.

Launch director Gene Thomas eventually cleared it for 1138, when the temperature at the pad had climbed to 36 degrees. 36 was still 17 degrees colder than any shuttle had ever launched in. The people in the firing room that morning were also carrying the previous week. Challenger's launch had already been scrubbed several times. the mission had originally been set for late January, and a chain of delays on the previous flight, a Columbia mission that had itself scrubbed seven times before getting off the ground,

had pushed everything back. Then Challenger was delayed for weather at an emergency landing site in Africa. Then, on the morning of January 27th, the crew was strapped in and ready when a technician couldn't get a handle off the orbiter's side hatch. A bolt had stripped. They tried to drill. The drill's battery was dead. They tried other batteries and those were dead too.

Eventually they cut the bolt off with a hacksaw. By then the winds at the Cape had come up past the limits for an emergency return to the runway and the launch was scrubbed again. Every one of those delays was covered on television and by the 27th the coverage had started to take on a tone. The shuttle that was supposed to fly a teacher into space couldn't get off the pad.

I'm not going to claim that pressure decided anything, because the commission looked for that and didn't find it. But the commission did write that the schedule NASA had set for itself for 1986, 15 flights in a single year, had left the agency without the margin it needed, and that the pressure of that schedule was part of the environment the launch decision was made in. One piece of pressure that's been repeated for decades doesn't hold up. The story goes that the White House pushed for the launch because President Reagan was giving the state of the union address that night.

and wanted to mention the teacher in space. The commission specifically investigated that and found no evidence of any White House contact with NASA about the launch schedule. The state of the union was postponed after the accident. Reagan spoke to the country from the Oval Office instead, and his speechwriter, Peggy Noonan, built the address around lines from a poem written by a pilot in 1941.

That speech is the reason a lot of people my age remember the phrase surly bonds. But the idea that the speech caused the launch is one of those claims that gets repeated, because it sounds like it should be true. And the record doesn't support it. The crew of seven walked out to the pad late that morning. Dick Scobie was the commander. Michael Smith was the pilot on his first flight.

Judith Resnick, Ellison Anizuka, and Ronald McNair were the three mission specialists. Gregory Jarvis was a payload specialist from Hughes Aircraft, and Krista McAuliffe was a social studies teacher from Concord, New Hampshire. Hampshire, picked out of more than 11,000 applicants to be the first ordinary citizen NASA sent to orbit. McCalliffe's backup, a teacher from Idaho named Barbara Morgan, was at the Cape that morning, too, watching from the ground with McCallif's family and the families of the other astronauts. Morgan's part in this story doesn't end there, and I'll come back to her. The countdown resumed. At 1138 in the morning, Eastern time, the three main engines lit,

followed six seconds later by the two solid rocket boosters and Challenger lifted off Pad 39B. The cameras caught what nobody in the firing room could see in real time. At 678,000th of a second after ignition, a puff of dark gray smoke appeared from the aft field joint on the right solid rocket booster on the side facing the external tank. Over the next two seconds, eight more puffs came out of the same spot. The smoke was dark because it was burning. burning grease and insulation and rubber. The primary O-ring in that joint hadn't sealed.

It had been too cold and too stiff to move into its groove fast enough, and hot gas had blown past it and passed the secondary as well. Then the puff stopped. The commission's engineers worked out what had probably happened. As the propellant burned, aluminum oxide slag from the exhaust had been thrown into the gap and packed itself into the breach, forming a temporary plug.

The joint had sealed itself with something that was never designed to seal it. For the next 50-some seconds, Challenger flew normally. Around the 52nd mark, the shuttle hit the strongest wind shear any shuttle had encountered up to that point. The commission's report is careful about this. It says the wind shear caused steering loads on the vehicle that were larger than any prior flight, and that the loads may have been what dislodged the slag plug. It doesn't say for certain. What it says for certain is that at 58.788 seconds, a small flame appeared on the right booster at the aft field joint,

and that from that point on, the vehicle had roughly 14 seconds left. At 64 seconds, the flame changed shape, which the commission read as the moment it burned through into the external tank's hydrogen compartment. At about 72 seconds, the lower strut holding the right booster to the tank gave way. The booster pivoted around its upper attachment, and its nose struck the tank. At 73 seconds, the hydrogen tank ruptured, then the oxygen tank above it, and the whole stack came apart. What happened next wasn't an explosion, at least not in the way most people use the word. The cloud that filled the television screen was mostly the tank's liquid hydrogen and liquid oxygen burning off. The orbiter itself wasn't destroyed by a blast.

It was traveling at nearly twice the speed of sound, and when the tank came apart under it, the orbiter got turned sideways into the airstream and torn apart by aerodynamic forces. The two boosters, no longer attached to anything, kept flying on their own, which is what made the two white trails I saw arcing off in different directions. The last words on the cockpit voice recorder came from Mike Smith at 73 seconds. The recorder was recovered from the ocean floor in March, and the transcript was released in July. Smith said, uh-oh, that was all.

On the ground, the first hour went the way you'd expect, and one way you wouldn't. In the firing room at the Cape, the flight controllers' consoles had gone to no data at 73 seconds. Steve Nesbett, the NASA Public Affairs officer whose voice was on the broadcast, kept reading from his script for several seconds because he was working from a printed timeline

and hadn't yet looked up at the screen. When he did, what he said, and I'm quoting the broadcast here, was, Flight controllers here looking very carefully at the situation. Obviously a major malfunction. About a minute later, he said that the vehicle had exploded and that the flight director had confirmed it. At Concord High School in New Hampshire, students had been gathered in the auditorium to watch. Contemporaneous reporting from the Associated Press and the Boston Globe describes the students cheering at liftoff, some of them wearing.

party hats and blowing noise makers the school had handed out, and then going quiet, and then teachers turning off the television. Stay tuned for more disturbing history. We'll be back after these messages. The principals sent the students home for the day. The Coast Guard and NASA had recovery ships on standby, but they couldn't move in. Debris kept falling from the cloud for close to an hour, some of it from as high as 100,000 feet, and the range was closed to surface vessels until it stopped. When the first ships did get in, they were recovering floating wreckage and nothing else. There was never a rescue phase, because there was never anything to rescue. That was known within minutes on the ground and wasn't stated publicly for hours.

The two boosters, meanwhile, had flown on. Photographs show them corkscrewing away from the cloud under their own power, still burning. The range safety officer at the Cape, whose job was to make sure nothing landed where people were, sent the Destruct command at about 110 seconds after launch. The boosters broke apart and fell into the ocean. Some of the pieces that came up from the bottom in the following weeks, including the section of the right booster with the burn hole in it, were from that debris.

I mentioned the boosters because it's the one thing the range safety system was designed to do that day, and it did it. Everything else in the system that was supposed to protect the crew had already failed by the time anyone pushed a button. The hardest fact in this story is what the crew compartment did after the vehicle came apart, and I put off reading that section of the file for two days. The crew compartment on the shuttle was a reinforced aluminum pressure vessel bolted inside the forward fuselage.

When the orbiter broke up, the crew compartment separated from the rest of the fuselage more or less intact. Photographic analysis by the commission showed it emerging from the cloud on a ballistic arc, still climbing. It reached a peak altitude of about 65,000 feet, then fell. It took two minutes and 45 seconds to reach the water. The compartment hit the Atlantic at about 207 miles an hour. The commission estimated the deceleration at more than 200 times the force of gravity.

Nothing survived that impact. The question the investigators had to answer was what had happened in the two minutes and 45 seconds before it. NASA assigned that question to Joseph Kerwin. a physician and former astronaut, who was then the director of life sciences at the Johnson Space Center. Kerwin's report was delivered to NASA Administrator James Fletcher on July 28, 1986, six months to the day after the accident, and released publicly. It's about four pages long and it's written in the flattest language I've ever seen used to describe something like this. Kerwin's findings were these. The cause of death could not be positively determined.

The forces the crew were subjected to during the breakup of the vehicle were probably not sufficient to cause death or serious injury. And the crew possibly, but not certainly, lost consciousness in the seconds after the breakup due to loss of cabin pressure. That word possibly is doing a great deal of work, and I don't think Kerwin was hiding behind it. He explains why he couldn't do better. The crew compartment had been on the ocean floor for six weeks before it was found,

in about 100 feet of water. The remains were too badly damaged by the impact and the time in the water to say medically what had happened to them before the impact. The compartment itself was too badly damaged to say for certain whether it had held pressure. If the cabin held pressure, the crew would have been conscious the whole way down. If it lost pressure quickly, at 65,000 feet, they'd have lost consciousness within seconds, unless they'd gotten their emergency air on, which is where the evidence gets specific. Each crew member had a device called a personal egress airpack, a small emergency air supply mounted behind the seat

and meant to be used during a ground evacuation if the cabin filled with toxic gas. It wasn't a pressure system. It supplied air, not pressurized oxygen, so it couldn't have kept anyone conscious at high altitude if the cabin had lost pressure. But turning one on required reaching back and pulling a lever, and that lever had to be pulled by someone.

Four of the seven air packs were recovered from the ocean floor. Three of them had been activated. One of the activated packs was identified as Mike Smith's, the pilots. Smith was in the right-hand seat on the flight deck. His air pack was mounted behind him, where he couldn't easily reach it. To turn it on, someone behind him would have had to do it. The two crew members seated behind the pilot on the flight deck were Ellison on Azuka and Judith Resnick. Kerwin's report doesn't say which one of them activated Smith's pack.

What it says is that the activation happened, and that it would have required a deliberate act by someone who was conscious after the breakup. The other two activated packs couldn't be tied to specific crew members. The one that hadn't been activated was identifiable, and it was Dick Scobies, the commanders, mounted behind the left seat where he'd have had no more reach than Smith did. And there's one more piece of physical evidence, which I keep coming back to because it's so small.

When the flight deck was recovered, investigators examined the switches on the pilot's right-hand instrument panel. Three of the electrical system switches had been moved from their normal launch positions. They were lever-lock switches, the kind that have to be pulled out against a spring before they'll move, so a jolt from the breakup couldn't have thrown them. Investigators concluded that the switches had been moved deliberately, after the breakup, by Smith,

in what appears to have been an effort to restore power to the cockpit. I've thought about what it would mean to be a pilot who's just watched every instrument in front of him go dark and to start working the problem. I'm not going to speculate any further than that, because Kerwin didn't, and he had the evidence in front of him. What the record supports is this. At least one person on that flight deck was conscious after the breakup,

reached back and turned on a crewmate's emergency air, and at least one person, probably Smith, worked the switches. How long any of them stayed conscious after that, and whether any of them were awake at the end, is a question the evidence can't answer. NASA's own conclusion stated in Kerwin's cover letter was that the crew possibly lost consciousness within seconds due to depressurization,

and that even if they hadn't, the impact was unsurvivable. That's where the official record stops. The search for the crew compartment took six weeks, and for most of that time NASA said almost nothing publicly about it. The debris field covered hundreds of square miles of ocean off the Florida coast. The Navy, the Coast Guard, and NASA put more than a dozen ships and submersibles on it. In the first days, they were recovering floating wreckage, and the priority was the right solid rocket booster, because that's where the investigators already suspected the failure had started. Pieces of the booster casing with the

burned through hole in them were pulled up from the bottom in February and March, and those pieces are the physical proof of everything the commission later concluded. The crew compartment was found on March 7th by divers from the Navy salvage ship USS Preserver, working from sonar contacts about 18 miles northeast of the launch site in roughly 100 feet of water. NASA confirmed the find publicly on March 9th. The recovery of the remains took the rest of March and into April, and it was done by Navy divers working in poor visibility inside wreckage that had hit the water at 200 miles an hour. NASA's handling of this was criticized at the time, and I think some of that criticism holds up.

The agency refused for weeks to say whether remains had been found, citing the family's privacy. Reporters who flew over the recovery site were told to leave. The Kerwin report wasn't released until July. It's understandable that NASA didn't want the families learning details from a news broadcast. But the effect was also that the public didn't learn until months later that the crew hadn't died at 73 seconds. And by then the story had settled into the version most people still carry. On April 29, 1986, the remains were transferred from the Cape to Dover Air Force Base in Delaware

in seven flag-draped caskets. Individual remains that could be identified went to the families. Remains that couldn't be separated were buried together under a single memorial at Arlington National Cemetery on May 20th. Scobie and Smith were also buried at Arlington. McNair was buried in his hometown in South Carolina. Onazuka in Hawaii, Resnick was cremated, and McAuliffe was buried in Concord. President Reagan named the Investigating Commission on February 3rd, six days after the accident.

He picked William Rogers to chair it, a former Secretary of State and Attorney General. The membership included Neil Armstrong as vice chair, Sally Ride, who'd flown on the shuttle twice, an Air Force General named Donald Kottina who'd overseen military space launches, and Richard Feynman, the Caltech physicist, who'd won a Nobel Prize and had, by his own account, never wanted to be on a government committee. The commission was given 120 days, and it used almost all of them. It held public hearings, took testimony from more than 160 people, and produced a report of 256 pages plus appendices.

And the way it got to the O-ring is a story of its own, because NASA didn't lead them there. In the first hearings in early February, NASA's presentation to the commission was led by Molloy and others from Marshall, and it treated the O-ring history as one possibility among many. The commission members weren't told about the Thea-Call teleconference, or that the contractor had recommended against launch. Two things broke that open, and they happened within days of each other.

The first was a newspaper. On February 9th, the New York Times published a story reporting that NASA had been warned about the O-ring joints months before the accident. The reporting was based on documents leaked by a NASA budget analyst named Richard Cook, who'd written a memo in July of 1985, the same month as Boisoli's, warning that the joints could fail and that it was, in his phrase, the most critical safety issue on the shuttle. Cook wasn't an engineer.

He was a resource analyst who'd asked engineers what kept them up at night and written down what they told him. After the accident, he decided the commission needed to see it and gave it to the times. The second was a phone call and a car engine. General Kutina later told the story this way, in interviews over the following decades.

A few days into the commission's work, he got a call from a fellow astronaut who'd been talking to someone at NASA, and the astronaut told him that NASA had test data showing the O-rings lost their resilience in the cold. Katina couldn't use it directly without exposing his source, so he invited Feynman to his house, and while they were standing in his garage

looking at the carburetor of an old car, Katina asked Feynman, more or less, what does cold weather do to an O-ring in a carburetor? He let Feynman work the rest out. Katina didn't name the astronaut for more than 20, 25 years. After Sally Ride died in 2012, he said publicly that it had been her. Ride never confirmed it while she was alive, and there's no document that ties her to it. What we have is Kootina's account and the timing, plus the fact that Ride was the one commission

member who'd have had the personal contacts inside the astronaut office to get that data quietly. Her biographer and her NASA colleagues have generally accepted the account. On February 11th at a public hearing, Feynman asked for a glass of ice water. He took a small piece of O-ring rubber that had been clamp in a sea clamp, dropped it in the water, and left it there while Malloy was testifying. When he took it out and released the clamp, the rubber didn't spring back. He said, into the microphone, that he believed this had some significance for the problem. I've watched that clip a lot.

Feynman's biographers have argued about how much of it was spontaneous and how much was set up by Katina. Katina himself later said he'd steered Feynman toward it. Either way, the demonstration did in about 30 seconds what NASA's charts hadn't done in two weeks, which was make the O-ring problem something a person could see. The Thial Teleconference came to the commission's attention through McDonald. On February 14th at a closed session at the Kennedy Space Center,

NASA was again presenting its account of the launch decision. Maloy was describing the pre-launch review. McDonald was in the room as a Thiacol representative, sitting in the back. By his own account and by the accounts of commission members who were there, he stood up and told the commission that what they were hearing wasn't the whole story. He told them Thiakol had recommended against launch, that NASA had pushed back, that Thiakol had reversed under pressure, and that he had refused to sign.

Rogers by several accounts was furious. The commission, which had been set up to work with NASA, shifted that day to investigating it. Rogers announced publicly that the launched decision process appeared to have been flawed and that NASA officials involved in the decision would no longer participate in the commission's investigation. Boyce Jolie testified on February 25th. He brought his memos. He told the commission about the caucus, the hat comment, the photographs he'd put on the table.

His testimony and McDonald's are the core of the commission's chapter on the, the launch decision. Then they both went home to Utah. And that's the part of the story I really didn't know until I found it while doing research for this episode. Within weeks of their testimony, both McDonald and Boy Jolie had been moved. McDonald was removed from the solid rocket motor program and reassigned. Boy Jolie was taken off the SEAL Task Force and given a job with no connection to the redesign. Neither was fired. The IACAL described the moves as routine reorganization. Both men described them as punishment for having told the truth in public. It came out in May. The commission had a hearing on May 2nd, and McDonald and Boisjoli

told the commission what had happened to them. Roger said publicly that he was deeply troubled by it. Then a group of members of Congress got involved. Senator Ernest Hollings of South Carolina and others made clear to Thia a call that its future contracts would be a subject of interest if the witnesses were being retaliated against. Stay tuned. for more disturbing history. We'll be back after these messages. Thiacol reversed course within days. It also, by McDonald's account,

tried to have him and Boisjoli barred from further testimony on the grounds that they were now speaking for themselves and not the company. That didn't work either. McDonald was restored and then more than restored. He was put in charge of the redesign of the booster joint, and he ran that program until the shuttle flew again

in September of 1988. He stayed at Theacol until he retired. He wrote a book about it called Truth, Lies, and O-Rings, and he spent the rest of his life giving talks about engineering ethics. He died in 2021 at 83. Boisoli's story went differently. He was restored to the program on paper, but he later said that the environment had become unworkable, that colleagues blamed him for the reputational damage to the company, and that he couldn't do the work. He went on medical leave in July of 1986 with what he described as severe stress,

and depression, and he never went back. He sued Thiacall and NASA in federal court. The suits were dismissed. In 1988, the American Association for the Advancement of Science gave him its award for scientific freedom and responsibility. He spent the following decades speaking to engineering students about what had happened in that caucus room, and he was open in those talks about the toll it had taken. He died in 2012 at 73, the men who'd overruled them mostly retired. Malloy left NASA in 1986. Hardy retired that year too. Mason, Lund, and Kilminster were all gone from Thia Call's booster program by the end of the year.

None of them was ever charged with anything, and the commission didn't recommend that anyone be. One more Thial call engineer belongs in this part of the story, and his name doesn't come up as often. Bob Ebeling was Boazoli's manager and was on the teleconference. He was the one who'd first raised the alarm, on the afternoon of the 27th when he heard the forecast. He told an interviewer decades later that he'd gone home that night and told his wife that Challenger was going to blow up and that he'd driven to work the next morning

to watch the launch on a screen in a conference room with Bois Jolie because he couldn't stand to be anywhere else. Ebelin gave an interview to National Public Radio in January of 2016 on the 30th anniversary. He was 89 and in poor health. He said he'd spent 30 years blaming himself for not having done more. and that he thought God had picked the wrong person for the job. Hundreds of people wrote to him after it aired.

So did NASA. So did George Hardy, who wrote him a letter saying he'd made the right call. Ebeling said in a follow-up interview that it had helped. He died about two months later. The commission's report came out on June 9, 1986. Its finding on the physical cause was unambiguous. The accident was caused by a failure in the joint between the two lower segments of the right solid rocket booster.

The specific failure was the destruction of the seals meant to keep hot gas from getting through the joint. The cold weather was named as a contributing factor, along with the design of the joint itself, the effects of the wind shear, and the process by which the joint had been assembled. Its finding on the decision was just as direct. The commission wrote that the decision to launch had been flawed, that the people who made it were operating without the information they needed. and that if the decision makers had known everything Thiocauls engineers knew, it was unlikely the launch would have gone ahead.

But the section of the report that I think has aged best is the one on what the commission called the Silent Safety Program. NASA had, on paper, a safety, reliability, and quality assurance organization. The commission found that it had been cut back through the early 80s, that it reported to the same managers whose decisions it was supposed to check, and that no one from it had been on the teleconference. Nobody from safety had been in the room, because there wasn't really a safety room to be in.

Feynman wrote his own appendix to the report, and he had to fight to get it included. Rogers wanted the commission to speak with one voice. Feynman's view, which he wrote in his own account later, was that some of what he'd found didn't fit into a consensus document. So it became Appendix F, under his name alone. The part of Appendix F that gets quoted most is about probability. Feynman had asked NASA managers and NASA engineers separately

what they thought the odds were of losing a shuttle on any given flight. The engineers, he wrote, gave him numbers on the order of 1 in 100. The managers gave him 1 in 100,000. He pointed out that 1 in 100,000 would mean you could launch a shuttle every day for 300 years and expect to lose one. He asked where the number had come from. Nobody could tell him. His conclusion was that the manager's figure had been arrived at

because the program needed it to be that low, and that calling it an estimate was generous. The appendix ends with a sentence that's become the most famous line Feynman ever wrote about anything, and I'll paraphrase it, rather than quote it, because I think the sense is what matters. He wrote that when a technology is going to succeed, the truth has to come before public relations,

because nature can't be fooled. The commission made nine recommendations. The joint was redesigned with a third O-ring and a heater to keep the seals warm. The management structure was reorganized so that the kind of level three decision Malloy and Hardy had made on their own would be reported upward. An independent safety office was created reporting to the administrator. The flight rate was cut, and astronauts were brought back into management. On the theory that people who'd actually fly the vehicle would weigh risk differently, One of the recommendations that gets less attention was about escape.

The shuttle had no way for the crew to get out during ascent. The first four flights had ejection seats for the two-man crews, and those were removed once the crews got bigger, because there was no practical way to eject seven people. NASA had known this. The commission recommended that an escape system be studied for the phase of flight after the booster separated, and what came out of that was a bailout system.

A telescoping pole, the crew could see. slide down to clear the wing during a controlled glide. It would not have helped Challenger. There was no controlled glide. The commission acknowledged that, but it was the first time anyone had built anything. The shuttle didn't fly again for two years and eight months. Discovery launched on September 29, 1988, with the redesigned boosters. Every subsequent flight had a launch-commit criterion for temperature. A sociologist named Diane Vaughn spent nine years on this. Her book came out in 1996, and she'd started with the same assumption most people have, which is that managers had broken the rules to keep the schedule.

What she found in the record was that they hadn't. They'd followed the rules. The problem was what the rules had become. Vaughn's phrase for it was the normalization of deviance. Each time the joint had shown erosion and the shuttle had come home, the erosion had been analyzed, documented, and accepted. The acceptable amount of erosion crept up. Then blowby was seen, and blowby was analyzed, documented, and accepted.

By 1986, seeing hot gas get past the primary seal, had been reclassified from a failure of the seal to an expected behavior of the joint. Nobody had decided to accept the risk in a single meeting. They'd accepted it a little at a time, and each acceptance had been the basis for the next. I think about this in terms of my old job. where the same thing happens with procedure. You skip a step once because you're short-handed and nothing goes wrong. Then that's the way it's done.

Then it's in the training. Vaughn's contribution was showing that the people at Marshall weren't villains and weren't stupid, and that their decision was reasonable given what they taught themselves to believe. 17 years later in February of 2003, Columbia broke up during re-entry over Texas and killed its crew of seven. A piece of foam insulation had come. off the external tank during launch and punched a hole in the leading edge of the wing. Foam had been coming off the tank on flight after flight for years. It had been analyzed, documented,

and accepted. The Columbia Investigation Board quoted Vaughn directly and said that NASA's organizational culture had as much to do with that accident as the foam had. Barbara Morgan, the backup teacher, stayed with NASA. She spent the year after the accident traveling the country as the teacher in space designing, giving the talks McCallif would have given. Then she went back to teaching in Idaho. In 1998, NASA selected her as a full mission specialist, not a teacher passenger, but an astronaut. And on August 8, 2007, she flew on endeavor to the International Space Station. She spent nearly 13 days in orbit, operated the station's robotic arm during construction work and did answer questions from school children on a live link, which is the one part of the

original mission plan that survived. That's 21 years, six months and 11 days after she watched Challenger from the ground. McCallough had planned to teach two lessons from orbit. They were to be broadcast live to schools. One was a tour of the shuttle. The other was about why people go to space at all. In 2017, two astronauts aboard the station filmed the science demonstrations McCollough, had planned, using her lesson plans, and the Challenger Center that the families founded released them for classroom use. That's not a footnote I expected to find, and I'm including it, because it's the closest thing to the mission being completed that exists. The families reached settlements. In December of 1986, the Justice Department announced that four of the families had

settled with the government and Thiacol for a combined $7.7 million, with Thiacol paying about 60%. The Jarvis and McNair families settled separately. The Smith family, the pilots, refused the settlement and sued NASA. That suit was dismissed on the grounds that Smith, as a government employee killed on the job, was limited to federal compensation. The family later settled with Thiacol. Thiakol, for its part, was never barred from the program. It built the redesigned boosters and kept the contract for the life of the shuttle. NASA had been in the middle of a process to bring in a second booster supplier when the accident happened, and the commission noted that their awareness of that competition may have shaped its reluctance

to tell its only customer no. More than a hundred tons of Challenger were recovered from the Atlantic, about half of the orbiter by weight. After the investigation, NASA sealed it inside two abandoned Minuteman missile silos at Cape Canaveral Air Force Station, Complex 31 and 32. The silos were capped with concrete. Nothing was displayed and for nearly 30 years nothing from the vehicle was shown publicly. The ocean kept giving pieces back anyway. In December of 1996, two large sections of the orbiter, one about six feet long, washed up on Cocoa Beach after a storm. NASA identified them and added them to the silo. As recently as 2022, a documentary dive crew looking for a World War II aircraft off the Florida coast, found a 20-foot section.

of Challenger's underside, tiles still attached on the seabed. NASA confirmed it and left it where it was. In 2015, the Kennedy Space Center Visitor Complex opened an exhibit called Forever Remembered with a single piece of Challenger in it, a section of the left fuselage with the American flag still visible. The families had been asked first. It was the first time any of the vehicle had been on public view. Ultimately, the record supports a short list of things. It supports that a known flaw and a known joint failed in the way engineers had predicted it would. On the coldest morning, the shuttle had ever flown in. It supports that the people closest to the problem said so the night before, in riding,

and were overruled by managers who had the authority to do it and used it. It supports that the people with the authority to stop the launch were never told. And it supports that after the vehicle broke apart, at least one member of the crew was conscious and worked. and possibly more. For some part of the two minutes and 45 seconds it took to fall. It doesn't support that anyone acted with malice. It doesn't support the White House story. It doesn't tell us how long the crew was awake. And it doesn't tell us because it can't. Whether a different set of managers on the teleconference that night would have decided differently,

or whether the same reasoning that had accepted blowby for two years would have found some other reason to fly. I've spent a lot of time with the transcript of the flight deck audio. It's about two minutes long. Most of it is routine. Scobie and Smith calling out speeds and engine settings. At about a minute in, Resnick says something about the engines being at 104%. A few seconds later, Scobie acknowledges the throttle-up call from the ground. The last entry is Smith at 73 seconds.

Then the transcript ends, and the physical evidence begins. three switches on the right-hand panel pulled out against their springs and moved. An airpack behind the pilot seat that someone reached back and turned on for him. And a reading of eight degrees on the surface of a rocket, taken before dawn by a man with a handheld thermometer that went into a report and stayed there.