Cautionary Tales with Tim Harford - Death on the Dance Floor
Episode Date: March 25, 2022With its splendid modern architecture, the Hyatt Regency was the place to be seen in Kansas City in 1981. Beneath space-age walkways, guests drank, laughed and danced... not realizing that the 60 tons... of of glass, concrete and steel hanging above their heads was about to come crashing down.One hundred and fourteen people died. But why? Was it cheap materials? Shoddy construction? Or a tiny error that seemed so insignificant that no one paid it any attention?For a full list of sources go to timharford.comIf you’d like to keep up with the most recent news from this and other Pushkin podcasts be sure to sign up for our email list at Pushkin.fm. Learn more about your ad-choices at https://www.iheartpodcastnetwork.comSee omnystudio.com/listener for privacy information.
Transcript
Discussion (0)
Pushkin
Friday night at the Hyatt Regency, one of Kansas City's most popular dates since the hotel
opened just a year earlier.
There are cues of the lobby bar, the place is buzzing.
It's the summer of 1981, but the orchestra is belting out the big band sounds of the
forties and fifties.
Dance contestants with numbers pinned to their backs are doing the manbo and the East Coast swing.
But while the music is nostalgic,
the hotel lobby is distinctly space-age.
It's big and airy, with a glass wall letting in the light
and three walkways, skywalks crossing the space
on the second, third and fourth floors.
They're suspended from the ceiling so that the lobby itself is
unobstructed by columns, all the more room for dancing.
Cindy Paulson briefly looks down from the terrace restaurant
to see if she can see her father out on the dance floor.
The restaurant on a large mezzanine overlooking the lobby provides the perfect view.
But the lobby floor is packed and she can't pick him out of the crowd.
After a moment she goes back to work.
Cindy's a college student. Her family live in town.
Being a hostess at the Hyatt is her summer job.
It's 5 past 7 in the evening.
Cindy will always remember that, because she glances across the lobby, out through the
floor to sealing glass wall, and sees the time on the clock, on the bank just across the
street.
Then, as a sharp crack, clearly audible, even above the music and the loud buzz of conversation.
Cindy's eyes refocus from the distant clock to the sky walks in front of her,
but she can't believe what she's seeing.
I'm Tim Halford, and you're listening to caution retail. Dr. Joseph Wackley got the call almost immediately from emergency dispatchers.
We need you at the highest.
He grabbed a stethoscope and scrubs and ran to his car.
Wackley, just 35 years old, had already served as the director of Kansas City's emergency
medical system.
He rushed to the scene.
When Wackley arrived at the Hyatt, it was lit by the flashing lights of ambulances.
Amid the crowd, there were dozens of injured people.
Some were bleeding, others had broken bones, some were lying on stretchers.
Wackley sprang into action, checking who needed assistance most urgently.
Then a paramedic grabbed him.
He said,
the real casualties are inside the hotel.
You need to get inside.
The lobby was no longer airy and bathed in evening sunlight.
It was coated in concrete dust.
The skywalks no longer levitated across the atrium.
Two of them had fallen, crunching onto the busy dance floor.
Voices were calling out for help.
Electric cables were swinging loose,
arcs of electricity sending flashes of light across the dim space.
Somewhere high above, a pipe had fractured and water was gushing onto the lobby floor.
It was six inches deep already and threatened to submerge those who had trapped.
If anyone had survived the impact that is.
Wackely could see arms and legs sticking out of the rubble.
Dr Wackely took charge of the medical operation.
It was bolstered by a huge voluntary effort
with taxis, buses and private cars
helping the ambulances to get injured people to hospitals.
Locals were doing whatever they could
to tend to the walking wounded,
free the trapped, and comfort families. Among them was Cindy Poulson, the restaurant hostess.
She'd helped evacuate the diners of the restaurant, but then she'd returned to the lobby.
Somewhere in the middle of that mess was her father.
She moved through the lobby doing what she could.
She comforted one couple, hind under the wreckage.
After firefighters were able to free them,
Cindy kept looking for her father.
250 miles away in St. Louis, Jack, Gillum, and his wife arrived home to find the telephone ringing.
The call was from one of the architects at the Hyatt Regency Hotel.
There's been a collapse at the Hyatt.
Jack Gillum was stunned. He was the structural engineer of record for the Hyatt Regency Hotel's construction.
That meant he was the one who'd signed off on all the designs, certifying that they were safe.
And now Gillum was being told that those elegant walkways across the hotel lobby
had fallen out of the sky and smashed into the crowd of people below.
fallen out of the sky and smashed into the crowd of people below. Jack, the rescue teams are trying to get equipment that can move the fallen sections of
warquake.
They need to know how heavy the sections are.
Reeling, Gillum could hardly think.
He stammered out an approximate answer.
Each of those airy-looking skywalks weighed over 30 tons.
A single walkway section weighed about 8 tons.
They were going to need cranes.
Gillum managed to charter a plane and fly over, arriving a few hours into the rescue effort.
Whatever he expected to see, the disaster was worse.
But his job now was to try to figure out how the
structure had failed. This was a brand new building. It looked bold but the
engineering was simple enough. Somehow the structure had failed catastrophically.
Somehow they had missed something. But what?
Three Skywalks had spanned the atrium of the Hyatt Regency, flat decks of steel and concrete
with glass balustrates, hanging on steel rods from the roof.
The third floor skywalk hung by itself and was undamaged.
But the second floor skywalk had hung directly beneath the one on the fourth floor.
When Cindy Paulson had gazed absolutely over the lobby at a clock across the street,
her attention had been caught by the sound of...
Skywalks, lurching downwards a few inches.
There were a few people on the upper skywalk and more on the lower
one, 50, 60 maybe, but there wasn't time for them to react. A moment later, both Skywalks
simultaneously tore free of one of the supporting ceiling rods. The extra load was suddenly
transferred to nearby rods, ripping the Skywalks free of those connections too.
In the next.
In the next.
The Skywalks are simply unzip themselves from the ceiling logs, and the Skywalks no longer.
More than 60 tonnes of steel and concrete fell onto the dancers below, spraying bystanders with deadly glass
raffles. The entire chain reaction took seconds.
Dr. Joseph Wackley and the other rescuers were now trying to deal with the aftermath.
One man was half buried, his legs crushed by a piece of skywalk. Wackily told him he wouldn't survive unless
they could amputate. The man said no.
Others didn't get the choice. Medics were forced to tell people, we can't get you out
in time. You'll die before they get the cranes in here. All we can do is give you something
to ease the pain. Then they moved on to the next patient.
Family members were on the scene almost immediately,
begging the medics and the fire crews to help.
Wackley was trying to decide who most urgently needed care
with the spouses and parents and children of the victims
right there, crying, pleadingading screaming as he did it.
He went back to the man who had turned down the amputation.
I've changed my mind doc, he said, you can take my legs off, I just want to live.
Wackily back and over a surgeon with a chainsaw to perform the amputation.
They got him out, but in vain, he didn't survive the night.
Cindy Paulson was outside looking for her father, hoping he'd been able to walk away.
And deep, deep under the rubble was a man called Mark Williams.
Williams was an old school friend of Joseph
Wackley. While Wackley hurried around directing the medical response, Williams
was doing his best to contribute. He was trapped in an 18-inch pocket of air
underneath a section of fallen skywalk. His legs had been slightly askew when
the skywalks fell and were wrenched into the splits and beyond.
Both his legs were dislocated. His left foot was up beside his right ear.
But other trapped people could hear his voice calmly talking to them, trying to reassure
everyone that help was on the way, that they'd get through this. But with his face pressed against the floor and
water seeping through the rubble all around him, Williams realized that after surviving
over 60 tons falling on him, he was now all too likely to drown. While the medics and the firefighters worked through the night, the structural engineer Jack
Gillum wasn't getting any sleep either. He was the man who'd officially declared the skywalks were safe to build.
What had gone wrong was a very simple and surprising mistake.
Gillum spent some time doing some calculations to confirm his suspicions.
Then he went to bed, staring at the ceiling of his hotel room until the morning.
Meanwhile, the rest of the world was left to speculate.
One popular theory was that the Skywalks were overcrowded
with people dancing,
and that their rhythmic movements
had set up dangerous resonances
which destroyed the structure.
But the numbers didn't add up.
The people on the Skywalks would have had to have been
frenetically go-go dancing, not gently nodding along to big band swing.
And the weight of the people was small compared with the weight of the Skywalks themselves.
There were 63 people, on walkways that by coincidence weighed 63 tons.
Kansas City Building Codes demanded that the Skywalks
could have carried nearly 10 times as many people,
more like 500 people than 63.
Then, torque turned to the quality of construction.
Had someone tried to save money by using cheap materials.
A reasonable guess, but no.
Within days, the world had figured out the Skywalk's fatal flaw, thanks to a young engineer, a telephoto lens, and the Kansas City Star.
I've relied on reporting from the Kansas City Star for the eyewitness accounts of the
disaster, but the newspaper also won a Pulitzer Prize
for its investigation into exactly why the Skywalks collapsed.
After loud protests about a cover-up, the press were allowed to visit the Hyatt's shattered
atrium, the star sent its best reporters, and at the last minute persuaded a young structural engineer named Wayne Lyshka to join them.
The visit was a shock. Wayne Lyshka and his wife had spent evenings dancing in the hotel lobby.
The disaster seemed unnervingly close, but it was also frustrating. Journalists were only allowed
the view that the hostess Cindy Paulson had had, looking down at the lobby floor from the elevated restaurant.
It wasn't close enough to see much. Unless, of course, Lyshka asked the
stars photographer to use a long zoom lens. Even from a distance, Lyshka could
point out the details he wanted to see up close. Take a shot of that, he said.
And that.
And that.
When the photographs were developed, they showed the end of the vertical rods, uselessly hanging
from the ceiling down to the fourth floor.
Lyshka could see clearly that the rods were basically undamaged.
They hadn't thinned or stretched.
The rods hadn't failed.
The skywalks had fallen off them.
Examining a few more photographs, Lishka and the Kansas City Star had their answer.
And on the front page of the newspaper, they published technical drawings highlighting a tiny detail of the design, a tiny detail that had been overlooked with the most appalling
consequences.
The detail concerned the connection between the long vertical rods and the horizontal beams
that supported the walkways.
These steel beams were hollow big square steel pipes.
The vertical rods pierced each horizontal beam and the beam was supported by a steel nut
threaded onto each rod. The beam couldn't slide down the rod because the nut was in the way.
That might sound flimsy, but in fact a well-designed steel nut could support a huge weight.
Now, here comes the fatal detail.
The original drawings for the Skywalks showed both of them suspended from the same long
rods.
The fourth floor skywalk was supported by nuts halfway up each rod, and the second floor skywalk
was fixed by nuts at the each rod, and the second floor skywalk was fixed by nuts at
the bottom of the same rods. But that is not what was built. Instead of hanging from the same
long steel rods, the upper skywalk hung from shorter rods, and the lower skywalk hung from rods
which were attached to the upper one. Describing it, that change seems trivial.
And if I were to show you a technical drawing, the chances are you're still say that the
change was trivial.
It seems hard to see why the change makes a difference.
So picture this.
Imagine that you and a friend are planning to swing on a rope, hanging from a high tree
branch.
You'll hold on to a knot in the middle of the rope, and your friend will hold on to a
knot near the bottom.
But then you realise you don't have a long enough piece of rope.
You've only got two shorter lengths.
No problem, says your friend.
She ties one length of rope, kn at the bottom around the tree branch.
You swing from that rope, she says.
Then she ties the other length, firmly around your waist.
I'll swing from this one.
That isn't going to work, is it?
Your grip was strong enough to support your own weight, but it's not strong enough to
support both of you.
It won't be long before your clasped hands lose their grip, slipping around the knot.
And because your friend is hanging from a rope attached to you, when you fall, she'll fall too.
The apparently trivial design change to the Skywalks had the same fatal effect.
Instead of two walkways hanging from the same set of rods, now the upper walkway was supporting
its own weight and the weight of the lower Skywalk 2.
For a year, the steel strained under double the intended weight, absolutely at the limit. What makes this an engineering
tragedy, rather than an engineering curiosity, was how finally balanced the skywalks were.
Any weaker, and they would have collapsed during construction. Any stronger, and they might
have lasted indefinitely. But they were just strong enough to stay aloft until one busy, buzzing Friday evening in the
summer of 81, when hundreds of people danced underneath the skywalks, and a few dozen
more came to stand on them and watch.
The first responders to the disaster were weeping in frustration as they tried to pry up
the huge concrete slabs.
People underneath were begging for help,
but the standard rescue equipment was all but useless.
As the word spread across Kansas City
of what was happening at the Hyatt,
construction workers began to arrive at the site,
bringing jackhammers and oxyacetylene torches, portable generators and lights.
The deputy fire chief used to work part-time at a construction crane operator just down the street.
He made the call and before long the huge cranes were on their way to the scene.
Cindy Paulson was outside still looking for her father.
When across the chaos she saw her mother instead.
The two rushed towards one another.
How did you get here?
Our Cindy.
Your father called. Her mother applied.
He was perfectly safe. He'd left the hotel
before the skywalks fell, and only heard about the accident on the radio. He'd called
home to reassure his family, and Cindy's mother had rushed to the scene to reassure Cindy.
Cindy told her mother to head home without her. She was going back in to help.
without her, she was going back in to help. From time to time, Dr. Wackley would whistle as a signal for the rescue effort to fall silent.
The rescuers, hundreds of them, would listen as trapped people called out their names from
under the rubble.
The rescuers did their best to mark the location of each one.
After the first hour or so, there wasn't much good news. Most of the survivors were out,
most of those who remained were beyond help. But there was the occasional moment to lift the spirits. At two o'clock in the morning, a man, a woman and a boy were pulled alive from an air pocket under the wreckage.
Mark Williams, his legs wrenched around his ears, was still waiting.
The water around him had receded, but the voices of those buried nearby had also fallen silent. At three in the morning, two massive cranes smashed through the atrium's beautiful glass
wall, sharrowing the shards across the blood and the dust on the floor.
Those all too heavy concrete slabs were lifted a little, letting rescuers call in and shine
their torches. The excited rescuers were now very close to Mark Williams, which posed new and unexpected
risks.
He could hear a jackhammer getting ever closer.
Its chisel tip brushed his ribs, then stabbed down near his crock.
Shuttered jackhammer off you idiots, he yelled during a brief pause.
They did.
And a few minutes later, Mark Williams twisted like a pretzel that very much alive was pulled
out of the rubble.
After that, there would be no more miracles.
The rescuers had spent the night cutting through dead bodies to reach the living, trying
to stabilize people with catastrophic injuries, digging out the spouses who died together as
they danced.
18 married couples, and finding the body of 11-year-old Pamela Coffey, the only child to die in
the collapse.
Dawn had broken when the cranes lifted the final fallen section of Skywalk.
For some of the rescuers, it was the moment when the long struggle became too much to bear.
There were thirty-one people underneath.
Every single one of them, was dead.
Over the years, engineers have found a horrid fascination in the Skywalk collapse.
It's been a case study in a thousand engineering classes,
a kind of logic puzzle, spot the floor in the innocuous seeming design change.
It's because the era is so surprising, so easy to overlook, and then so obvious in hindsight.
Letting the lower Skywalk hang off the upper skywalk was the simple, fatal error.
So clear only when you imagine swinging off a rope and trying to bear the weight of someone
swinging off you. The steel cross beams couldn't bear the double weight, focused through that
small nut on the steel rod. The beams buckled and slipped around the nut and off the rods, and disaster followed
almost instantly.
After the tragedy, there was plenty of commentary in places such as the letters page of the
engineering news record.
Some pointed out that it was no surprise that the original design concept had been modified.
That concept secured the upper skywalk by resting it on nuts halfway up the vertical steel rods.
Simple.
Until you think about it.
How does each nut get there?
One possibility is to machine 15 feet of each rod with a fine screw thread, and hope that those threads remain
undamaged, while these six rods were installed in a busy construction site. Or perhaps custom
made rods slightly slimmer at one end than another, so that the nut can slide along until it
reaches the thread in the middle. There are other ways to get a nut onto the middle of a long rod, but the original design
doesn't acknowledge that there's even an issue.
No wonder the contractor suggested a change.
As one engineer commented, a design that begs a change cannot be completely without blame
when the change is made.
The letters page was filled with clever alternative solutions,
but it's not hard to find a solution once you clearly understand that switching to an
upper and lower rod posed a problem in the first place. Nobody did.
I worry that in focusing on the Skywalk collapse as a floor in engineering design,
we risk losing sight of the real problem. Engineers will forever more pay close attention
to the design of Rod Gurdder connections, but this isn't really about the structure of a steel
connection. It's about the structure of a project. What failed wasn't just a beam.
What failed was a process.
Turn away from the powerful eyewitness reporting of the Kansas City Star.
And pick up instead a special May 2000 symposium in the Journal of Performance of Constructed
Facilities.
In that symposium, numerous engineers, including Jack Gillum himself, analyzed what went wrong.
Gillum listed 17 different factors contributing to the error.
The project was rushed, details weren't checked by external reviewers, senior people resigned
from the engineering company halfway through,
important jobs were outsourced.
Gillum himself protested that we requested from the architect, owner and construction manager
on three separate occasions that we be permitted to provide full time inspection during the
construction, yet each time we would anide.
Of course, Gillum was pleading his own case there, but the overall impression is clear.
It was a complicated project.
With lots of ins and outs, lots of people coming and going.
It's easy to see how things could get missed.
What's unnerving about this is that, well, every big building is a complicated project.
We have to trust the project managers to cope.
In his book, The Checklist Manifesto, The Surgeon at Olga Wande, looks at how to prevent accidents
in modern medicine.
It's also complicated these days, he argues.
There are large medical teams, complex medical procedures, lots of ways in which things can
go wrong, often because of a simple oversight.
Goanday makes a compelling case that this complexity can be tamed with checklists.
Agree a list of what has to be done?
Everything from a surgical team introducing themselves to each other, to ensuring that there are the right supplies and spares on hand, to making
sure that you have the right patient on the operating table in front of you.
Checklists sound absurdly simple, but in many cases, they've been proved to work.
But what's striking to me is where Atul Goande originally got the inspiration
for this idea. It was talking to the team who were building an 11 story extension to
the hospital where he worked. 3,885 tons of steel, 13,000 cubic yards of concrete, 19 air handling units, 16 elevators, one cooling tower,
and one backup emergency generator.
And to make sure nothing was missed, checklists, lots of checklists, checklists for schedules,
checklists for communication, checklists for checklists.
But when it came to building the Hyatt Regency in Kansas City, some vital boxes never got ticked.
There's one pivotal moment worth highlighting. It happened early in January 1979, two and a half years before the disaster.
At the time, it didn't seem like much. And that was the problem.
It was simply a phone call, which was a common enough way of communicating in the days before
email. A manager at the Steel Fabricator phoned the structural engineers project manager
and suggested that fateful switch from the long single rods to the shorter pairs of rods. The structural engineer thought it through while still on the phone, but missed the critical
realization that the change would double the load that the upper beams were supporting.
He said that the change sounded fine, but asked the steel fabricator to submit the
request through formal channels.
That formal request was never made.
Do you see why I say this was a process error?
As much as a failure of structural engineering design,
someone makes a phone call with a request.
Someone else says it sounds okay, but put it in writing.
It never gets put in writing.
Nobody follows up.
This sort of thing happens all the time.
In offices all over the world.
Meeting notes are unclear, or nonexistent, or lost.
Emails disappear down into the depths of an overloaded inbox.
People agree that some task needs to be done, but it's not quite clear who's responsible.
And so in the end, nobody is.
We've all seen it happen.
It's just that usually, when someone's a bit sloppy with their to-do list, nobody dies.
This time, people did.
A lot of people.
Jack Gillum, the structural engineer with overall responsibility for the project, explained
the connection that failed was never designed.
The original detail of the connection, as shown on the contract structural drawings,
was only conceptual in nature,
neither the original detail nor the double rod connection as built was designed.
So not a failure of design, an absence of design.
Because of a failure of process.
When the official investigation was completed by the National Bureau of Standards, it concluded
that yes, the switch from long single rods to short double rods was probably the difference
between the skywalks staying up and the skywalks falling. But it also concluded that even the
original design wasn't strong enough to meet the requirements of Kansas City's building code.
The original design would almost certainly not have collapsed, which of course is the
important thing, but the original design was still too weak.
Nobody ever seems to have checked that.
And nobody had noticed, not the structural engineers, not the steel fabricators, not
the building inspectors.
In the arguments and legal disputes that followed, the attorney representing the state licensing
board said, it wasn't a matter of doing something wrong. They just never did it at all.
Nobody ever did any calculations to figure out whether or not the particular connection that held the skywalks up would work?
It got built without anybody ever figuring out if it would be strong enough.
It just slipped through the cracks.
Literally true, of course.
Under the tremendous weight of two skywalks, the nut ripped through the weakest point of
the box beams, a welding
seam that cracked open.
Metaphorically true as well.
In between the two-ing and throwing, the resignations and the outsourcing, everyone assumed that
someone had designed and checked the steel joint.
Nobody had.
It slipped through the cracks.
114 people died in the disaster.
Almost 200 were injured.
No criminal charges were filed, but Jack Gillum and one of his colleagues lost their license
to practice in Missouri.
Gillum spent his twilight years retelling the story of the collapse
as a cautionary tale for other construction professionals. He often spoke to engineers about the
disaster, hoping to scare the daylights out of them. The buck stops with me, he would tell them,
that was my engineering seal on the plans.
Student hostess Cindy Paulson spent that night at the Hyatt, helping with the rescue
effort. It took a toll on her. Her grades, once outstanding, started to fall. She took
time out of her studies to try to reset, but that didn't help either. She was later
diagnosed with depression.
But she got on with life and found a career selling real estate.
She tries not to think too much about the Hyatt, but she can't help but get nervous on
walkways.
Mark Williams, the last person to be pulled out of the rubble alive, spent two months in
hospital, much of it on
dialysis because his kidneys had given up under the strain. He spent a further two years
learning to walk again, a little unevenly, but not too bad. He gave an upbeat assessment
to the Kansas City Star. We don't know what we can accomplish, what we can withstand, until we're tested.
As a slogan for life, it's a good one. But it was never supposed to be a principle
of engineering design. We rely on engineers to get things right on the drawing board.
We don't expect to find out the hard way that 60 tons of steel and concrete are floating
in the sky on a joint that nobody designed and nobody checked. A key source for this episode was Donna McGuire's reporting for the Kansas City Star in 2001. For a full list of references check out the show notes at TimHalford.com.
Corsinary Tales is written by me Tim Halford with Andrew Wright. It's produced by Ryan
Dilly with support from Courtney Garino and Emily Vaughn. The sound design and original
music is the work of Pascal Weiss.
Julia Barton edited the scripts.
It features the voice talents of Ben Crow, Melanie Guthridge, Stella Hartford, and Rufus Wright.
The show also wouldn't have been possible without the work of Mia LeBelle, Jacob Weissberg,
Heather Fein, John Schnarrs, Carly Mjiori, Eric Sandler,
Royston Besserv, Maggie Taylor, Nicole Marano,
Daniela Lakan, and Maya Canig.
Corsionary Tales is a production of Pushkin Industries.
If you like the show, please remember to share, rate,
and review.
And if you want to hear the show add free
and listen to four exclusive cautionary
tail shorts, then sign up for Pushkin Plus on the show page in Apple podcasts or at
Pushkin.fm-plus.
you