0:15

Pushkin. Hello

0:19

Tim here. Shortly after we

0:21

recorded this episode, I received the sad

0:23

news that Charles Perrault, the great

0:25

sociologist whose work we discuss, had

0:28

just died. I was sorry to hear

0:30

it. I do hope that Charles

0:32

would like what we've done with his ideas.

0:39

As the night draws in and the

0:42

fire blazes on the hearth, we

0:44

warn the children by telling them stories.

0:48

Star Wars teaches them always

0:50

scan for droids. But

0:53

my stories are for the education

0:56

of the grown ups, and my

0:58

stories are all true. I'm

1:01

Tim Harford. Gather close

1:04

and listen to my cautionary

1:06

tale.

1:18

You've heard this story before. You

1:21

might even have watched it happening live.

1:23

Thirty three million people did. There

1:27

They stand together on stage,

1:30

Fade Unaway and Warren, Beatty, Bonnie

1:33

and Clyde together again after fifty

1:35

years. Even for a pair

1:37

of veteran Hollywood stars, it

1:40

must have been a nerve racking moment. Yet

1:42

their task seemed simple. Open

1:44

a red envelope, take out

1:47

a card, and read out the title

1:49

of the film that had won the Academy

1:51

Award for Best Picture. Betty

1:54

opens the envelope, So far, so

1:57

good. Then he looks

1:59

at the card in his hand. He

2:01

hesitates, then he looks

2:03

inside the envelope again, as though checking

2:05

to see if there was a cover note. He

2:08

looks over at Away, then raises

2:10

both eyebrows. She reaches

2:12

over and touches his arm affectionately.

2:15

The Academy Award Betty begins.

2:18

He pauses again. He looks

2:20

at the card again. He slips

2:22

his hand inside the envelope one more time.

2:25

The audience laugh at the moment of

2:27

maximum tension that old rogue

2:29

is playing with everyone's emotions. He

2:32

continues for a Best Picture. He

2:34

stops again. Faye Dunaway

2:37

thinks he's goofing around too. You're impossible,

2:39

she says, come on, He shows

2:41

her the card. She doesn't

2:44

hesitate for a moment. La

2:47

La The

2:52

audience erupts and applause, and

2:54

the producers of La La Land's take to

2:56

the stage to give their acceptance speeches.

2:59

Meanwhile, off in the wings, an

3:02

accountant named Brian Cullenan

3:04

knows that the biggest screw up in the

3:07

history of the Academy Awards is

3:09

in full flow. The twenty

3:11

sixteen winner of Best Picture isn't

3:14

La La Land. It's Moonlight.

3:18

So yes, You've heard this story,

3:22

Have you understood what it really means?

3:25

You're listening to another cautionary

3:27

tale.

3:51

There's a simple way to tell the story of this fiasco.

3:54

As Warren Beatty walked on stage, Brian

3:57

Cullenan's job was to hand him

3:59

the envelope for Best Picture. Instead,

4:02

Cullenan handed over the envelope for Best

4:05

Actress. A few moments

4:07

earlier, that award had been

4:09

won by Emma Stone for her performance

4:11

in La La Land, and so when

4:14

Betty opened the envelope, he saw Emma

4:17

Stone La La Land. Jimmy

4:20

Kimmel, the host of the Oscars that evening,

4:23

jokingly blamed it all on Beatty. Larren,

4:25

what did you do? I

4:28

wanted to tell you what happened, but

4:31

it really wasn't Beata's fault. He

4:33

didn't understand what he was looking at, and

4:35

rather than say the wrong thing, he

4:37

hesitated. Then he turned to Faye

4:40

Dunaway. She was the person

4:42

who actually uttered the title of the wrong

4:44

film, but it wasn't her fault either.

4:47

Imagine her situation. She's on

4:49

stage in front of the most star studded

4:51

audience imaginable, plus tens of

4:53

millions watching live on television. Betty

4:56

seems to be messing around and she doesn't know why.

4:58

The first thing she sees on the card is La

5:00

La Land, and straightaway that's what she says.

5:04

Some people blamed Betty, some

5:06

people blamed Dunaway, most

5:08

people blame Brian Cullenan, the

5:10

accountant who handed over the wrong envelope.

5:13

But all of those people, I think

5:16

are making a mistake. Almost

5:20

a decade ago, years before this

5:22

fiasco, I interviewed

5:24

one of the world's most important thinkers

5:27

on how accidents happen. He's

5:29

a sociologist named Charles

5:31

Perrault, and he told me

5:33

something that stuck in my mind that

5:36

we always blame the operator.

5:39

It's always, he says, a case

5:41

of pilot error. Charles

5:44

Perrault is a wise old man. He's

5:46

older than the Academy Awards themselves.

5:48

He was four back in nineteen twenty

5:50

nine when they were first presented. And

5:53

Perrault is absolutely right. We

5:56

inever to be looked for some one to blame,

5:59

and that inclination leads

6:01

us astray. Our instinct

6:03

is to blame the accountant, Brian Cullenan.

6:06

He did give Batty the wrong envelope. He

6:08

was distracted. He'd tweeted a photograph

6:11

of Emma Stone holding her Oscar

6:13

statuette at a time he should have

6:15

been focusing on giving Batty the right envelope.

6:18

He was on his phone, enjoying

6:20

being close to one of the world's most beautiful

6:22

people at her moment of triumph.

6:26

We can all learn a lesson from that. Get

6:28

off your phone, get off your phone when you're

6:30

supposed to be having dinner with your family, get

6:32

off your phone when you're supposed to be driving, and

6:34

get off your phone when you're supposed to be giving

6:36

the envelope containing the best picture

6:38

card to Warren Beatty. But

6:41

the reason that it's a mistake to simply blame

6:43

Culonan is because Cullinan was

6:45

just being human. Humans

6:48

are always getting distracted, Humans

6:50

are always making mistakes. If

6:52

our systems can't cope with those mistakes,

6:54

then it's hopeless to demand better humans.

6:57

We need better systems. When

7:01

La La land fleetingly won Moonlight's

7:03

Oscar, some rich and successful

7:05

people suffered some embarrassment and some

7:07

heartache, but nobody died.

7:10

Yet the lesson of the fiasco is

7:13

far from trivial. The same

7:15

kind of mistake in different situations

7:18

can be catastrophic. Such

7:20

mistakes can lead to nuclear accidents

7:22

and financial meltdowns. And

7:25

when I say can lead, I

7:27

mean have led. Such disasters

7:30

have already occurred. So

7:33

let's try to understand what really

7:35

happened that strange night at the oscars.

7:38

Perhaps we can use that insight to

7:40

prevent far more serious calamities.

7:54

Let's talk about problem number one, bad

7:58

typography. I know it sounds

8:00

strange, but it's true. The card

8:02

that Betty took out of the envelope had

8:05

nine words on it, and the largest

8:07

word is oscars. Really

8:11

is that really the most important

8:13

piece of information? Are we worried

8:15

that without it? Warren bate He's going to think

8:17

he's at the Iowa State Fair handing

8:19

out a medallion for the best hog. The

8:23

words La La Land and

8:25

Emma Stone are printed with equal

8:28

weight, even though the winner is

8:30

Emma Stone. The rest is detail.

8:33

Meanwhile, the important words

8:35

best Actress are tucked

8:38

away at the bottom of the card, and

8:40

they're tiny. If

8:42

best Actress had been prominent,

8:45

Warren Batey wouldn't have been confused.

8:47

He would have known that he had

8:49

the wrong card in his hand. And

8:52

if Emma Stone had been

8:55

in larger type than La La land

8:57

Faye Dunaway wouldn't have blurted out

8:59

the name of the wrong film. It

9:01

would have been awkward for Baiting and done Away

9:04

to walk off stage to get the right envelope,

9:06

but not nearly as awkward as not

9:09

walking off stage to get the right envelope.

9:13

So the Academy should have hired

9:15

a designer, but they're

9:17

not the only ones. Imagine

9:20

that you're the night shift supervisor

9:23

at a nuclear power plant. It's

9:25

four o'clock in the morning and you

9:27

don't know it yet, but the turbine

9:30

system that draws away the heat from

9:32

the reactor core has just shut

9:34

down. You're going to have to make some quick

9:36

decisions, assuming you can

9:38

figure out what's going wrong and

9:41

why? What's

9:43

that? Ah, let's

9:46

see, I think it's

9:49

he Can

9:51

we shut it down? I do no,

9:56

can you. During

10:00

the first few minutes of the accident, more

10:02

than one hundred alarms went off, and

10:05

there was no system for suppressing the unimportant

10:07

signals so that operators could concentrate

10:09

on the significant alarms. Information

10:12

was not presented in a clear and sufficiently

10:14

understandable form that from

10:17

the overview of the official inquiry,

10:20

into what was then arguably the world's

10:22

most serious nuclear accident at

10:25

Three Mile Island in nineteen seventy

10:27

nine. It destroyed the reactor,

10:30

came close to a serious release of radioactive

10:32

material on the Eastern Seaboard, and

10:35

shattered the reputation of the American

10:37

nuclear industry. It's striking

10:40

how quickly the inquiry focused

10:42

in on the question of design, But

10:44

to anyone who studies how accidents

10:47

happen, it's not surprising at all.

10:49

The plan's control rooms were so poorly

10:52

designed that error was the inevitable.

10:55

Don Norman, the director of the design lab

10:58

at UC San Diego, was asked

11:00

to help analyze the problems at Three Mile

11:02

Island. Design was at fault,

11:05

not the operators. The

11:07

control panels were baffling. They

11:09

displayed almost seven hundred and

11:11

fifty lights, some next to the

11:14

relevant switches, or above them, or

11:16

below them, sometimes nowhere near

11:18

the relevant switch at all. Red

11:20

lights indicated open valves

11:23

or active equipment. Green indicated

11:26

closed valves or inactive equipment.

11:28

But since some of the lights were typically

11:31

red and others were normally green, the

11:33

overall effect was dizzying.

11:37

So yes, the operators

11:39

made mistakes at three Mile island,

11:41

but with better design they might not

11:44

have. Warren Beatty and

11:46

Fade Unaway would sympathize.

11:51

But something else went wrong that night at the

11:53

oscars, something deeper and

11:55

more surprising. It's a strange

11:58

effect, and it was first observed

12:00

by a man less famous for looking at

12:02

why things go wrong, and

12:04

more famous for looking at the stars. Galileo

12:08

Galilei is known for his astronomy

12:11

and because his work was consigned

12:13

to the Church's Index Liborum

12:16

Prohibitorum, the list

12:18

of forbidden books. But

12:20

the great Man's final work opens

12:23

with a less provocative topic, the

12:25

correct method of storing a stone

12:27

column on a building site. There

12:30

with me, This book from sixteen

12:33

thirty eight is going to explain

12:35

the Oscar fiasco and much

12:37

more, a master related circumstance

12:39

which is worthy of your attention, as

12:42

indeed are ol events which happened

12:44

contrary to expectation, especially

12:47

when a precautionary measure turns out

12:49

to be a cause of disaster.

12:52

A precautionary measure turns out to be a

12:54

cause of disaster. That's very interesting,

12:57

Galileo. Please go on. A

12:59

large marble column was laid

13:01

out so that its two ends

13:04

rested each upon a piece of beam.

13:07

I can picture that in my mind support

13:09

at the column while it's being stored horizontally

13:11

ready for use. If you lay it on the

13:13

ground, it may get stained and you'll

13:15

probably break it when you try to get ropes

13:17

underneath it to pull it upright. So

13:20

yes, store it flat, but

13:22

propped up by a support at one end and

13:25

a support at the other. But

13:27

what if the column can't support its own

13:29

weight like that and simply snaps

13:31

in half? Galileo has

13:34

thought of that. A little

13:36

later, it occurred to a mechanic

13:38

that, in order to be doubly sure

13:41

of it's not breaking in the middle, it

13:43

would be wise to lay a third support

13:46

midway. This seemed

13:48

to all an excellent idea.

13:51

Yes, if two supports are good, surely

13:54

three supports are better. It was quite

13:56

the opposite, for not many months

13:59

past before the column was found cracked

14:01

and broken exactly

14:03

above the new middle support. How

14:07

did that happen? One of the end support's

14:09

head, after a long while, become

14:11

decayed and sunken, but the

14:14

middle one remained hard and strong,

14:17

thus causing one half of the column

14:19

to project in the year without

14:22

any support. The

14:24

central support didn't make

14:26

the column safer. It pressed

14:29

into it like the central pivot of a

14:31

seesaw, snapping it in half. Galileo's

14:34

tail isn't really about storing

14:36

columns, and neither as mine. It's

14:39

about what I'm going to call Galileo's

14:42

principle. The steps we

14:44

take to make ourselves safe sometimes

14:47

lead us into danger. The

14:50

problem Galileo described is

14:52

well known to safety engineers. There's

14:55

an article in the fine scholarly

14:57

journal Process Safety Progress

15:00

titled no good deed Goes

15:02

unpunished Case studies

15:04

of incidence and potential incidents

15:07

caused by protective systems. These

15:11

case studies are magnificent.

15:14

There's a chemical plant where

15:16

two pressure relief systems interact

15:18

in a way that means neither

15:20

of them work. There's

15:22

a flare designed to destroy pollutants,

15:25

which ends up causing the release of

15:28

toxic gases. There's

15:30

an explosion suppression device that

15:33

causes an explosion. This

15:35

kind of thing happens more than you might

15:38

think. An example

15:40

is one of the earliest nuclear

15:42

accidents at Fermi One,

15:45

an experimental reactor in Michigan.

15:48

It's barely remembered now except in

15:50

gil Scott Heron's song We

15:52

almost lost Detroit. This time,

15:55

the operators at Fermi one lost

15:57

control of the nuclear reaction for

15:59

reasons that were baffling to them. Some

16:02

of the reactor fuel melted. It was all

16:04

touch and go. Eventually

16:06

they got the reactor under control, shut

16:08

it all down, and waited until

16:11

it was safe to take it all apart. It

16:13

was almost a year before the reactor

16:15

had cooled enough to identify the

16:18

culprit. A piece of metal

16:20

the size of a crushed beer can

16:23

had blocked the circulation of the coolant

16:25

and the reactor core. It

16:28

was a filter that had been installed

16:30

at the last moment for safety reasons,

16:33

at the express request of

16:35

the Nuclear Regulatory Commission. It

16:37

had come loose and caused the entire

16:40

problem. Galileo's

16:42

principle strikes again. Why

16:47

do safety systems sometimes backfire?

16:50

The wise old sociologist Charles

16:52

Perrault's most famous book is

16:54

called Normal Accidents. It's

16:57

about how certain kinds of

16:59

system are vulnerable to catastrophic

17:01

failure, so vulnerable, in

17:04

fact, that we should view accidents

17:06

in those systems as inevitable.

17:09

The vulnerable systems have two features.

17:12

The first is that they're what Pero calls

17:14

tightly coupled. In a tightly

17:16

coupled system, one thing leads to another,

17:19

and another and another. It's like domino

17:22

toppling, which is actually a great

17:24

example of a tightly coupled system. Once

17:26

you start, it's hard to hit the pores

17:28

button. A second feature

17:31

is complexity. A complex

17:33

system has elements that interact

17:36

in unexpected ways. A

17:38

rainforest is a complex system. So

17:40

is Harvard University. But Harvard

17:43

University usually isn't tightly

17:45

coupled. If there's a problem, there's

17:47

also time to find a solution. A

17:50

system that's both complex

17:53

and tightly coupled is

17:55

dangerous. The complexity

17:57

means there will occasionally be surprises.

18:00

The tight coupling means that there will be

18:02

no time to deal with the surprises.

18:07

Charles Pero's theory explains

18:09

Galileo's principle. Every

18:12

time you add a feature that's designed to prevent

18:14

a problem, you're adding complexity.

18:17

The middle support for the column added

18:19

complexity. So did the safety

18:21

filter that damaged the Fermi one reactor.

18:24

Safety systems don't always

18:27

make us safe. Here's

18:35

the question we should be asking about that

18:37

bizarre evening at the OSCARS. How

18:40

was it even possible for

18:42

the distracted accountant Brian cullenan

18:45

to give Warren Beatty the wrong

18:47

envelope. A few minutes earlier,

18:50

the envelope for Best Actress, the

18:52

envelope containing the card that read

18:54

Emma Stone La La Land. That

18:57

envelope had been in the hands of Leonardo

19:00

DiCaprio as he stood on

19:02

stage announcing her win. How

19:04

could that envelope have made

19:07

its way into the hands of Warren Beatty.

19:09

The answer it

19:12

didn't. There were two

19:14

envelopes. Every envelope

19:17

for every category had a duplicate

19:20

version waiting in the wings. These

19:22

duplicate envelopes were there as

19:25

a safety measure, and that safety

19:27

measure is what made the fiasco

19:29

possible. Galileo's

19:32

principle had bitten hard.

19:36

Charles Perrot's argument is that

19:38

when systems are both complex

19:40

and tightly coupled, we should

19:42

expect catastrophic accidents.

19:45

Does the Academy Awards ceremony fit

19:47

that description. It's certainly

19:50

tightly coupled. You can't easily interrupt

19:52

a live TV spectacular in

19:54

front of millions of people to ask

19:57

for advice. The show must

19:59

go on. Yet the

20:01

ceremony doesn't have to be complex.

20:04

Giving an envelope to Warren Beatty. Doesn't

20:07

have to be complex, but

20:09

you can make it complex if you try.

20:12

Brian Cullenan's partner in crime that

20:14

evening was Martha Ruise. Like

20:17

Cullan, she was a senior accountant.

20:20

The pair of them carried identical

20:22

briefcases with an identical

20:24

set of envelopes. On

20:26

the day of the show, we'll get the ballots

20:28

and Brian and I will go to the theater on

20:31

two separate roads. He'll go one

20:33

route and I'll go another route. That's

20:36

how Martha Ruise helpfully explain

20:38

things to journalists. Just before Oscar

20:41

night, both she and Cullinan

20:43

had been proudly giving interviews about

20:45

the full proof system. We do

20:47

that to ensure that in case

20:49

anything happens to one, the

20:52

other will be there on time and

20:54

delivering what's needed with the full

20:56

set. We do have security

20:58

measures up until we're at the theater

21:01

and delivering that envelope to the

21:03

presenter just seconds before

21:05

they walk on stage. We'll be

21:07

in two different locations. Brian will be

21:10

on stage right and all

21:12

the on stage left. It all

21:14

sounds sensible, and in many

21:17

ways it is sensible.

21:19

It's also complicated. The

21:22

system of twin envelopes meant

21:25

that every time an envelope was opened

21:27

on stage, its duplicate

21:29

in the wings had to be set aside.

21:32

Martha Ruiz stage left

21:34

handed the envelope for Best Actress

21:36

to Leonardo DiCaprio, leaving

21:39

Brian cullenan stage right with

21:41

a job of discarding the duplicate, the

21:44

job he failed to do. If

21:46

there hadn't been that set of twin envelopes,

21:49

Warren Beatty could never have been

21:51

given the wrong one. So

21:55

bad design helped cause

21:57

the Oscar fiasco. It also

22:00

helped cause the accident at Three Mile

22:02

Island, and a

22:04

complicated safety system was

22:06

at the root of the Oscar fast It

22:08

was also at the root of the Fermi

22:11

one accident. But it's

22:13

not just oscars and nuclear power.

22:16

I promised you a financial catastrophe

22:18

two and in this one, both

22:21

safety systems and bad typography

22:24

at a blame. He might even

22:26

start to see the banking crisis in

22:28

a very different light. In

22:34

September two thousand and eight, at

22:36

the height of the financial crisis, an

22:39

insurance executive named Robert

22:41

Willemstat requested a meeting

22:43

with Tim Geitner Geitner

22:46

would later be the Treasury Secretary. At

22:48

the time, he was the president of the

22:50

New York Federal Reserve. That

22:53

meant he was responsible for supervising

22:56

Wall Streets banks, including

22:58

Lehman Brothers, which was on the brink

23:00

of collapse. Robert

23:02

Willemstadt was the boss of an insurance

23:05

company called AIG, and

23:07

since AIG wasn't a bank, it

23:10

was far from obvious why Willemstadt

23:13

was Geitner's problem. In

23:15

his book Too Big to Fail, the

23:18

journalist Andrew Ross Sorkin reports

23:21

the intimate details of this

23:23

ill fated meeting. I'm

23:25

really sorry, mister Willemstad. Mister Gattner's

23:27

going to be a few minutes, no problem,

23:30

I have time. I'm sorry.

23:32

I know you've been waiting a long time. Mister

23:35

Gattner's on the phone to the bars

23:37

of Lehman Brothers. He's up

23:39

to his arboles and layman. Tim

23:42

Geitner was also exhausted. He'd

23:45

been on an overnight flight from a banking

23:47

conference in Switzerland. He must have felt

23:49

completely overwhelmed. Who wouldn't

23:51

have barbas Sorry

23:54

to keep you waiting coming, Willemstat got

23:56

his moment. He badly needed

23:59

to be able to borrow from the FED, not

24:02

normally something AIG would be allowed

24:04

to do. But he also didn't

24:06

want to panic Geitner. He needed

24:08

to walk a tight rope to suggest

24:11

that AIG could use some help

24:13

but wasn't actually bankrupt.

24:16

Is this a critical or emergency situation,

24:18

Bob, Well, you know, let

24:21

me just say that it would be very

24:23

beneficial to AIG, mister Gagner.

24:26

Perhaps I can leave this with you. William

24:29

Stat handed Gaigner a briefing note.

24:32

Buried deep within it was

24:34

a fast ticking time bomb.

24:37

The largest firms on Wall Street

24:39

were relying on AIG

24:42

to pay out on insurance against

24:44

financial trouble. The total

24:46

sum insured was a truly

24:49

ludicrous two thousand,

24:51

seven hundred billion dollars.

24:55

AIG couldn't possibly pay

24:57

if all the claims came in, and

24:59

it was starting to look as though they might,

25:03

But that meant that the big Wall Street

25:05

banks wouldn't get the insurance payments

25:08

they were relying AIG

25:11

was both a bigger threat to the financial system

25:13

than Lehman Brothers and a far

25:15

more surprising one. If

25:17

AIG was a safety net, it

25:19

was one that wasn't going to break. Anyone's

25:22

fall. But to realize

25:24

that Tim Geitner would actually

25:27

have to read and absorb the information

25:29

in the note, and he was busy,

25:33

really busy, So

25:35

instead he filed it away and

25:37

turned back to the Lehman Brothers problem.

25:40

AIG would melt down a

25:42

few days later. The

25:44

parallels with Oscar Nite are uncanny.

25:48

For one thing, Geitner, who's no fool,

25:51

had no idea how to interpret

25:53

what he was looking at. It was unexpected,

25:55

and the key information was buried

25:57

in the small print. Fade unaway

26:00

and Warren Beatty know the feeling. Then

26:03

there's Galileo's principle. Safety

26:06

systems don't always make us safe.

26:09

Those insurance contracts were supposed

26:11

to offset risk, not create

26:13

it, right, But by now

26:16

we know that safety systems also introduce

26:18

new ways for things to go wrong. The

26:21

insurance contracts that were about

26:24

to destroy AIG were

26:26

called credit default swaps. They'd

26:29

become popular as a way to offset

26:31

risk with the blessing of regulators.

26:34

They seemed a smart idea, just

26:36

as the third support for the columns

26:38

seems like a smart idea, and the metal

26:41

filters at the Fermi reactor and

26:43

the duplicate set of award envelopes,

26:46

but they backfired. Wall

26:48

Street banks were relying on these credit

26:51

default swaps to keep them safe if

26:53

there was trouble. When it became

26:55

clear that insurance companies such

26:57

as AIG couldn't possibly

27:00

pay out, the banks all

27:02

scrambled to sell off their risky investments

27:04

at the exact same time for

27:07

the exact same reason. A

27:09

few days after Willemstadt had met

27:12

Geitner, officials and bankers

27:14

worked through the weekend on the Lehman Brothers

27:16

problem. Only on Sunday

27:19

evening did one of those bankers

27:21

receive a request from a Treasury official

27:24

to ask if she could drop everything

27:27

and work on rescuing AIG instead.

27:31

The surprising phone call was greeted

27:33

with a response that was unsurprisingly

27:36

unsuitable for family ears. Hold

27:40

on, hold on, you're calling me on a

27:42

Sunday night, saying that we just spent

27:44

the entire weekend on Layman and

27:46

now we have this. How the fuck

27:48

did we spend the past forty eight hours

27:50

on the wrong thing? How

27:52

Indeed, for the same reason

27:54

they gave the oscar to the wrong movie,

27:57

Confusing communication and

27:59

above all, a safety system

28:02

that created a brand new way

28:04

to fail the

28:07

banking crisis of two thousand and an

28:09

eight shook the world financial

28:11

system and destroyed millions of

28:13

jobs. More than a decade

28:15

later, we're still living with the consequences.

28:18

It was, in its way, a

28:21

more serious crisis than any

28:23

nuclear accident. It was certainly

28:25

far graver than a bungle at the Oscars.

28:28

Yet the same problems were at

28:30

the roots of all these accidents.

28:34

After the La La Land shambles,

28:36

Vanity Fair reported the Oscars

28:39

have an intense sixth

28:41

step plan to avoid another

28:44

envelope disaster. The six

28:46

steps include getting rid

28:49

of the two accounting partners, Brian

28:51

Cullenan and Martha Ruise, and

28:54

the twin sets of envelopes. Instead,

28:57

says Vanity Fair, there will

28:59

be three partners. A

29:01

third partner will sit in the

29:03

show's control room with the producers. All

29:06

three partners will have a complete

29:08

set of envelopes. If

29:11

having two sets caused the problem,

29:14

having three sets is better, right.

29:17

I'm not sure Galileo would

29:19

agree. You've

29:26

been listening to Cautionary Tales.

29:29

I expand on some of the ideas in this episode

29:31

in my book adapt You might

29:34

Like It. Cautionary

29:36

Tales is written and presented by me Tim

29:38

Harford. Our producers

29:40

are Ryan Dilley and Marilyn Rust.

29:43

The sound designer and mixer was Pascal

29:45

Wise, who also composed the

29:48

amazing music. This

29:50

season stars Alan Cumming, Archie

29:53

Panjabi, Toby Stevens and Russell

29:55

Tovey, with enso Celenti, Ed

29:58

Gochen, Melanie Gutteridge, Mercia

30:00

Munroe, Rufus Wright and introducing

30:03

Malcolm Gladwell. Thanks

30:06

to the team at Pushkin Industries, Julia

30:09

Artin, Heather Fame, Mia LaBelle,

30:11

Carlie Migliori, Jacob Weisberg

30:14

and of course the mighty Malcolm

30:16

Gladwell. And thanks to my colleagues

30:18

at The Financial Times