Podchaser Logo
Home
797: What I Was Thinking As We Were Sinking

797: What I Was Thinking As We Were Sinking

Released Sunday, 23rd June 2024
 1 person rated this episode
797: What I Was Thinking As We Were Sinking

797: What I Was Thinking As We Were Sinking

797: What I Was Thinking As We Were Sinking

797: What I Was Thinking As We Were Sinking

Sunday, 23rd June 2024
 1 person rated this episode
Rate Episode

Episode Transcript

Transcripts are displayed as originally observed. Some content, including advertisements may have changed.

Use Ctrl + F to search

0:00

Support for this American life comes from

0:02

Indeed. Indeed is driven by the search

0:04

for better. But when it comes

0:06

to hiring, the best way to search for

0:08

a candidate isn't to search at all. Don't

0:10

search, match with Indeed. And Indeed doesn't just

0:13

help you hire faster. 93%

0:15

of employers agree Indeed delivers the highest

0:17

quality matches compared to other job sites,

0:20

according to a recent Indeed survey. And

0:22

listeners of this show will get a $75 sponsored job

0:25

credit to get your

0:28

job's more visibility at

0:30

indeed.com/American. That's indeed.com/American. Terms

0:32

and conditions apply. Need

0:35

to hire? You need Indeed. A

0:38

quick warning, there are curse words that are un-beeped

0:40

in today's episode of the show. If

0:42

you prefer a beeped version, you can

0:45

find that at our website, thisamericanlife.org. I've

0:49

had a bunch of bike accidents lately. And the

0:51

moment that things go wrong, it's always the same.

0:54

These accidents happen during my morning commute. And

0:57

they always happen when it rains. I have this

0:59

very light bicycle I got.

1:01

And when it rains, something about this bike on

1:03

the city streets, a metal grate, a

1:06

slippery patch of any sort at all can make

1:08

the bike slip out from under me. I've

1:11

broken a clavicle, I've banged up my wrist, I've

1:13

given myself some kind of bruise on my right arm

1:15

that still wakes me up at night months later. And

1:19

in each of these three accidents, you

1:22

know, like when you're falling, there's that second or two

1:24

when you realize you're going down and you're on your

1:26

way down, but you haven't actually hit the ground yet.

1:28

Weirdly, it is enough time for a bunch of thoughts

1:30

to go through your head. And all three times, like

1:33

I say, it's been the same. Somehow, I get

1:35

very calm and I think, oh,

1:38

this isn't gonna be so bad. I'm not

1:40

going very fast, the ground isn't that far. Is

1:42

a car coming? Okay, no car's coming. I

1:44

figured out how to fall. It's gonna be fine

1:47

this time. And

1:49

I think everybody knows how

1:51

a person reacts under duress. I can tell you

1:53

something about them. And I think what

1:55

this set of thoughts says about me is, I

1:58

am optimistic even when there

2:00

is. There's no factual basis for it at

2:02

all. Like at the exact

2:04

moment when I'm going to collide with the

2:06

wet surface of the street, a moment when things are definitely

2:08

not going to be fine, I'm thinking things will be fine.

2:12

I think this optimism has kept me

2:14

in faltering relationships. When

2:16

I mentioned all this at our radio show story

2:18

meeting, one of my co-workers, somebody who I've worked

2:21

with closely for years, told me,

2:23

he has never said this, that

2:25

he thinks my optimism is the single most

2:27

important quality I have. He

2:29

said, oh, tell me there are 17 things that can go wrong

2:31

with some plan that we have, and I'll just go like, eh,

2:34

let's go for it. Though I

2:36

have to say after three accidents, I have learned not to

2:38

ride this particular bike in the rain. Thank you very much.

2:41

Today on our program, we have stories where things

2:43

go very, very badly for people. They

2:46

are on the way down and in that moment of

2:48

crisis. What they're thinking reveals

2:50

so much about who they are. Some

2:52

people, it leads to new insights. Some

2:55

leads to no insights at all. And

2:57

some are hit with a glaiting boat

3:00

of inspiration and figure out how to save themselves.

3:03

From WBC Chicago, this is American Life. I'm

3:05

Aaron Glass. Stay with us. Equine, Pirates of

3:07

the Caribbean Town. Oh,

3:10

before we

3:12

go any further, I

3:14

should say today's

3:16

show is a rerun. One

3:18

summer a while back, a group of friends

3:21

got in trouble. Got in over

3:23

their heads, almost literally, in the middle of

3:26

Boston Harbor. Our producer, Ike Srees-Kandaraja, tells the

3:28

tale. Though for the purposes

3:30

of today's story, call him a schmail.

3:33

I've been friends with you this whole time, and I didn't know that

3:35

this is your most told story.

3:38

Well, I wouldn't tell it to you. You were there.

3:40

I was there. It

3:42

was over 10 years ago, and I was a new

3:45

resident in the Boston Harbor. I

3:48

was a new resident of the Boston Harbor. 10

3:50

years ago, and I was a new resident in

3:52

the Boston area. My good

3:54

friend Sophie Tintorey introduced me to some

3:56

people she knew well. We

3:58

liked the same kinds of jokes. and

4:00

could only afford the same kind of rent.

4:03

So, we moved into a big house

4:05

together. And soon, we

4:08

got to the stage in any relationship

4:10

where someone suggests taking things to the

4:12

next level. So I don't

4:14

know exactly how it came up, but maybe

4:16

eight friends or so decided to

4:18

go in collectively and

4:21

buy a used boat. And

4:23

to my understanding, none of them knew a

4:25

single thing about boats. True. But

4:28

did you know the word catamaran comes from

4:30

Tamil? Kattu, marim, wood tied

4:32

together? Those are my guys. So

4:37

that's one thing I know about boats and maybe more ancestral knowledge

4:41

with just waiting to surface

4:43

if I owned a boat. Here's

4:46

how it happened. Our

4:48

friend Max brought us the opportunity. He knew a person

4:51

who knew a person who

4:54

needed to get rid of their boat. For

4:57

$400 each, Max could make us

5:00

Massachusetts boat people. Having

5:02

doubled that in my bank account, I

5:05

signed a check, sight unseen. And

5:08

when I finally got to meet our boat down

5:10

at the harbor, it

5:12

felt like we had gotten away with larceny.

5:15

I wish you could have seen it. A

5:18

1950s highliner, a glamorous

5:20

little wooden motorboat with an engine hanging

5:22

off the back, polished

5:25

wood, two rows of white vinyl

5:27

seats, chrome finishes. Think

5:30

Italian vacation, like the talented

5:33

Mr. Ripley. But

5:35

right here on the Boston Riviera.

5:38

Also, like Boston's own Matt Damon

5:41

in the talented Mr. Ripley, a

5:43

new high society life was waiting

5:46

for us. The

5:53

first time I got on the boat, it really felt

5:55

like magic. We honked just

5:57

the right number of times in a bridge.

6:00

opened up for us like a

6:02

bouncer unclipping a red velvet rope

6:04

to the VIP section. We

6:07

entered the Charles River and

6:09

stopped to admire the sailboats swirling around

6:11

us. A small motorboat

6:13

pulled up to us. The man

6:15

asked if we were stranded. No,

6:18

sir, just enjoying this beautiful day.

6:20

Thanks for asking. Well,

6:23

you know, you're parked in the

6:25

middle of a regatta. Oh,

6:29

he looked us over. Do

6:31

you guys work for Google? The

6:34

plan to cosplay our way to high

6:36

society was already working. Max

6:48

knew the next step. We

6:50

had to announce our old boat

6:52

as the newest member of Boston's watery

6:54

glitterati, which meant a

6:57

christening party. Max had

6:59

been practicing for this moment his whole

7:01

life. The man

7:03

loves extravagant gestures. He

7:06

invited friends from near and far

7:09

to the Charles River Esplanade and

7:11

asked everyone to wear white. He

7:14

showed up in black, a

7:16

rented black tuxedo. Max

7:19

stood on the front of the

7:21

boat, named it Marjorie after his

7:23

own mother, and smashed a purple bottle

7:25

of Andre sparkling wine on the

7:27

nose. Now,

7:35

I don't know if you believe in curses

7:37

and we didn't know it right away, but

7:40

Max invoked roughly three curses

7:42

in that moment. Wearing

7:44

a black tuxedo to a white party

7:46

of your own design, that's

7:49

more of a faux pas, not a

7:51

curse. But number one,

7:53

renaming a boat, that's a curse.

7:57

Number two, Jews like

7:59

Max don't name children after living

8:01

relatives. Now the

8:03

prohibition doesn't specify boats so

8:05

let's call that a half

8:07

curse. Number three,

8:10

smashing champagne on a used

8:12

boat. I heard that's a

8:14

curse but only if

8:16

you consider Andre sparkling wine to be

8:19

champagne. It's

8:23

impossible to say which curse attached

8:26

itself to the Marjorie but

8:28

moments later the sky

8:30

darkened and we all scattered

8:33

as sheets of rain fell upon us.

8:45

The next day the clouds cleared. Our

8:48

friends were still in town so we

8:50

tried to take them out again. Now

8:52

this is where the story

8:54

really begins. It was a

8:57

beautiful day I think it was early

8:59

summer and really sunny and gorgeous and

9:02

there were maybe 12 of

9:06

us on the boat. Does that

9:08

sound right? Nine in a boat

9:10

that was maybe made for six but

9:14

our Captain Max had just completed a

9:16

boat safety course and he

9:18

told us the only limit for boat

9:20

capacity is having enough life jackets for

9:22

everyone which isn't

9:24

true but we didn't know

9:26

that and Max denies ever

9:29

saying it. Squished

9:31

together two in the front seat four

9:33

in the back three others

9:35

perched on the sides we

9:38

headed out. People are taking turns

9:40

driving which is really fun I got to take

9:42

the wheel for a little bit and I was

9:44

like cutting across the waves in a way where

9:46

we'd like catch a little bit of air and

9:49

slam down and catch some air and slam down

9:51

which I thought was very fun and then at

9:53

one point somebody leaned over and said to me

9:56

Sophie this is a like a wooden

9:58

lake boat. The

10:00

Marjorie wasn't built for high-octane

10:02

thrills. Someone says,

10:04

we should get back soon. But

10:07

Sophie overrules them. And

10:09

then, just like the American

10:11

munitions that repelled British ships in

10:13

these very waters, cannonball!

10:17

I jump off the boat because it's

10:19

a beautiful day, and I like to swim, and

10:22

I'm kind of messing around and looking at the boat

10:24

from the water. And when I climbed back

10:26

into the boat, I was like, I climbed in on

10:28

the back, and I just hoisted myself up over the

10:30

edge of the boat. And at the point where the

10:32

most force was being pushed down on the boat, a

10:35

whole group of water just flowed right

10:38

in. I

10:40

was like, whoa, OK. The

10:42

back of the boat dipped below the water

10:44

line, and Boston Harbor began

10:47

to enter. And then, almost

10:50

instantaneously, the boat was like thigh-deep in water.

10:53

This is Ben Ewenkampen, a

10:55

moral pillar of our group, the

10:57

kind of friend your mom might ask, is Ben going

10:59

to be at the party? We

11:02

lived together and co-owned this boat. My

11:05

memory is that it was really, really

11:07

fast. Just it went from regular

11:09

life, you're having fun, to all of

11:12

a sudden emergency mode. It's

11:14

still a lovely summer day. We're still

11:16

on our beautiful boat. But

11:18

suddenly, I'm tearing an aluminum

11:20

can in half to bail us out. Turns

11:24

out, half a soda can only

11:26

scoops as much water as your own

11:29

cupped hands. I look

11:31

over at Sophie, the most MacGyvery of us,

11:33

to see if she has a better idea.

11:36

I saw a empty

11:38

gallon, like plastic water

11:41

gallon. And I was like, oh,

11:43

if I could tear the top off of this, then I can

11:45

scoop water out really fast. But I'm

11:47

wondering, OK, I don't have

11:49

anything to cut it with. That's fine. I'll cut it with

11:51

my teeth. I'll just crack a little hole. And once I

11:53

get a little hole going, they'll be purchased and I can

11:56

tear the whole thing open. And I try to bite it

11:58

with my teeth, and my teeth are just so good. weak.

12:01

My hands are so weak

12:03

and I like can't I can't do

12:05

anything. I don't think I'm feeling experiencing

12:09

anxiety but suddenly it's like one of

12:11

those nightmares where like none of your

12:13

limbs work. Maybe the

12:15

weight of nine passengers was too

12:17

much or maybe Sophie's

12:19

full throttle aerials had split

12:22

the seams or

12:24

maybe it was a

12:26

curse. Here's

12:29

what we know. We're alone.

12:31

We're in a small boat. We're

12:34

far from shore. The

12:36

nearest land to us is Logan Airport

12:39

and that's when it begins to dawn on me. If

12:42

we have to jump off this boat

12:44

and swim to the closest land we

12:47

would be army crawling up the

12:49

banks of a government controlled airspace.

12:52

Look it's so close we

12:54

can just swim to Logan

12:56

and very quickly and like

12:58

quietly but quite definitively you

13:01

said I'm not gonna do that. I'm

13:05

picturing my white friends being

13:07

wrapped in tinfoil blankets and

13:09

fed cocoa and me being

13:12

perp walked off the tarmac

13:14

by Homeland Security. Not

13:16

today government black site. Our

13:19

options are shrinking. Our bales

13:21

aren't bailing. The engine's not

13:23

turning. There's water

13:25

water everywhere. I think we're

13:28

going to sink. We were

13:30

sinking. We were in a sinking boat. Toby

13:33

David is visiting from Philly. A

13:35

gifted speech maker he helped christen

13:37

the Marjorie just the day before.

13:40

Now he reminds us he is

13:42

not a good swimmer. Asmatic

13:44

and does not like cold water.

13:47

The boat is going down and

13:50

it's tilting and we're

13:52

in the middle of the water and we have no idea what to do and

13:55

I'm freaking out. I'm really fully freaking out.

13:58

I'm not hearing much but through

14:00

the ether, Ben's voice

14:02

comes through and says, is

14:05

anybody else noticing that occasionally one of

14:07

us will ask a question and nobody

14:09

will respond? A

14:12

silent panic falls over us. It

14:15

is a moment that demands leadership, decisive

14:18

action. We turn

14:20

our heads to our captain, Max.

14:24

He found this boat, named it

14:26

after his mother, sold us on

14:28

the dream. And now,

14:31

like a monomaniacal Ahab

14:33

in a Red Sox cap, Max

14:35

denies there's any problem at all. Should

14:39

we call for help? And Max saying, no.

14:42

Max had this really strong instinct, don't call

14:44

for help. Don't call for, I know

14:46

what I'm doing. I think Max was

14:48

trying to cover up for this

14:51

massive fuckup. And so he

14:54

was sort of in denial that there was

14:56

a problem. Even up to

14:58

the point where the entire back end

15:01

of the boat was fully underwater, the

15:03

entire motor was underwater, Max was still

15:05

saying, it's fine,

15:07

there's no problem. But

15:09

there are two. One, we're

15:11

stranded in the water and

15:13

two, our captain has gone

15:16

mad. We only know how

15:18

to deal with the last one. Everyone

15:20

shouts at Max. Admit

15:23

it, we are in trouble. You

15:25

need to call for help. But

15:28

Max is sitting in the captain's seat,

15:31

one hand clenched around the steering wheel

15:33

and the other turning the key in

15:35

the ignition, like the

15:38

engine wasn't underwater. The

15:40

expression, hope floats, not

15:43

true and very dangerous. On

15:46

one end of the spectrum of willingness

15:48

to acknowledge this situation, there was Max

15:50

who was just like face forward, pretending

15:53

he was still driving a boat. And

15:56

then at the other end of the spectrum, the

15:58

other extreme was there's this one. wild

16:00

card. Cat, oh my

16:02

god, I don't know anything

16:04

about her. She just like

16:07

sort of appeared in a

16:09

beam of clarity. Cat

16:11

Spangler went to college with Max. She

16:14

is the ninth passenger and the one

16:16

outsider. When Max invited

16:18

her on a ride, he had not mentioned

16:20

there would be so many of us. For

16:23

most of the trip, she was shy and off

16:25

to the side. But now,

16:28

the strangers surrounding her are screaming

16:30

at the one person she knows,

16:32

and he has lost control.

16:35

I remember just really clearly like looking

16:37

over at Max, sitting there in

16:40

the captain's seat with his

16:42

eyes kind of downcast, frozen,

16:44

and something, there

16:47

was like a switch that flipped in me.

16:50

If the boat sinks and

16:53

all of us are in the harbor,

16:55

we're no longer visible to anyone. We're just

16:57

little heads, no one can see us. We

17:00

can tread water for a while. What

17:03

if we get a

17:05

cramp, get tired, swallow water? How

17:08

long can we do that for? What if the current

17:11

takes us out further? I

17:13

didn't want to let myself go there,

17:15

but I remember thinking that

17:18

people were going to die. If

17:20

we don't do something, people are going to die today. And

17:24

in that moment, Captain

17:26

Cap is born. My

17:28

mind went really blank and calm, and

17:31

it was like

17:34

a calculator filled my brain, and I was

17:36

just very logical, and I started

17:38

focusing on actions. Where

17:40

are the life vests? How many do you have? I

17:43

took on this voice that was very

17:45

bossy and direct. Like

17:48

Boston's own Matt Damon in Good Will

17:50

Hunting, she begins

17:52

solving the unsolvable problem. Namely,

17:56

if nine passengers get in a boat,

17:58

which turns out to only have

18:01

eight life jackets, not one per

18:03

person as we have been led

18:05

to believe, Max. And

18:07

now, some of those life jackets

18:09

are unreachable in the underwater part

18:11

of the boat. How

18:14

many passengers are totally screwed?

18:17

I remember looking over and there was this

18:19

guy who looked really pale. He

18:22

just didn't look like he was doing very well. And

18:24

I asked him, I said, can you swim

18:26

well? And he said, no,

18:28

I can't. And I just gave

18:31

him the life vest. I took it off my neck,

18:33

put it on his, and said, well, I can. Kath

18:36

hands out the remaining life jackets based

18:38

on need. The water

18:40

is rising. We are

18:42

coming apart and preparing to

18:45

abandon ship. And that's when I

18:47

got on top of the bow of the boat. She

18:50

climbs to the front of the boat and

18:52

starts waving her arms to the

18:54

pilots in airplanes over our heads to

18:57

nearby boats in the water, anyone

18:59

who might be able to come to our

19:01

rescue. She was the one person

19:03

who was sort of like, what the fuck is wrong

19:05

with you people? Of course we need help. And she

19:07

like stood up on the boat and started waving her

19:09

arms and trying to get the attention of a

19:12

boat that I didn't realize was also quite close. That

19:14

was like the rescue boat.

19:17

An actual rescue boat just

19:20

within waving distance. It

19:22

was so much taller than us off

19:24

the waterline. And there were guys on

19:26

deck kind of looking down and almost

19:28

chuckling at us, like making light of

19:30

the situation. And I remember

19:33

feeling like they have no idea what we've

19:36

been through. They

19:38

must have radioed for a towboat

19:40

because one arrives within minutes. The

19:43

captain calls down, do I

19:46

have permission to perform a life

19:48

rescue? Who

19:50

says no to that? I have to

19:52

confess one thing really fast. Toby is

19:55

admirably honest about what happened

19:57

next. I just have to

19:59

say, I'm sure that when the rescue boat

20:01

arrived that I, I may

20:03

have been the first person off of the sinking ship. I

20:06

fucking levitated to that ladder to

20:08

get off that boat. So

20:10

that's not great for me. We

20:13

climbed off the boat to safety. And

20:16

just like Matt Damon in Saving

20:18

Private Ryan, we were going

20:20

home. Matt

20:27

says she looks back at this moment a

20:29

lot, that it's become

20:31

pivotal to her understanding of herself.

20:35

She's proud that she jumped in and saved

20:37

us. It

20:43

was a tragic ending for our captain. From

20:46

the safety of our towboat, I can still

20:49

picture him down there, alone

20:51

on the Marjorie, refusing

20:53

to abandon ship. And

20:55

that was the last time I ever saw him. He

20:58

died that day. Yeah,

21:01

that's the only honorable thing I could have done.

21:03

I went down with the ship. Captain

21:06

Max died. My friend Max

21:09

survived. A little bruised,

21:11

his ego, and

21:13

his reputation for big schemes

21:15

with little consequences. And

21:18

while the boat became the story Sophie

21:21

likes to tell more than any other,

21:23

not Max. This is

21:25

a story I have told the least. I

21:28

can't overstate to you how big

21:30

a failure this was. One of

21:32

the most embarrassing failures of my

21:34

entire life. It's a

21:36

thing I look back at and

21:39

shudder. Max spent

21:41

more than a decade feeling terrible. The

21:44

rest of us spent that same

21:46

amount of time wondering why our friend

21:49

Max was trying to gaslight us and

21:51

sink us. But

21:53

we'd never talked about it. None

21:55

of us have. And when

21:57

Max and I catch up for this story,

22:00

it becomes comes clear that he has a

22:02

completely different memory of how it all went

22:04

down. He says he

22:07

actually did the one thing we were

22:09

begging him to do. I

22:11

remember a lot of things. I remember calling

22:14

Seto and telling them where we were.

22:17

I remember that like I did- Wait, can

22:19

I just back up once? Go ahead. Did

22:21

you just say you called Seto? Seto

22:25

is boat AAA, the boat that

22:27

performed the life rescue. Max

22:30

says he called them on his phone

22:33

way back when the engine first stalled

22:35

out. That's the reason

22:37

we got saved that day. Yeah,

22:40

I don't think I made a

22:42

big deal about calling Seto. No,

22:46

I didn't believe him. I

22:49

just interviewed five of our friends

22:51

who had no memory of this

22:53

at all. So I

22:55

followed up with Seto. Turns

22:57

out they keep meticulous records.

23:00

And there it was, a

23:02

call for a tow, July 17th, 2010,

23:06

charged to our membership under

23:09

Max's name. It cost $50. Help

23:13

was on the way. I

23:15

guess in the chaos, Max just

23:17

forgot to tell everybody on the

23:19

boat. Look, I

23:23

mean, clearly I owe them an explanation,

23:25

but a lot of our passive aggressive

23:27

ass friends have never brought up this

23:29

day with me. Our

23:33

story had hardened over a decade, but

23:35

was built on a lie. Max

23:38

had kind of done the right thing,

23:41

and Kath wasn't the one who

23:43

saved us. The

23:45

sadistic editors at This American

23:47

Life made me invite Kath

23:49

back to share our findings.

23:53

Max called for help, and

23:56

apparently he was the one who

23:58

got help to arrive. not you

24:01

waving. That just

24:06

feels impossible. But I think it's true. And

24:10

I'm so sorry that's the way

24:12

that it happened. I'm

24:17

just completely confused.

24:19

So this whole time, he

24:22

has been the hero of this story, but

24:26

no one has known it? It's

24:32

so perfect that this is the

24:34

finding. It's

24:36

so unlike Max 2 to not kind

24:38

of claim that, like

24:40

claim the credit of it and

24:43

say, no, I did that. The

24:46

final curse on the Marjorie had the longest

24:48

views and a

24:51

devastating payload. It

24:53

added an asterisk to a heroic

24:55

deed and turned a

24:57

failed captain into a guy who

24:59

kind of did the right thing

25:02

but was really confusing about

25:04

it. Please

25:07

heed this precautionary tale.

25:10

Beware of curses. Don't

25:12

try to be a Massachusetts boat person

25:15

for just $400. And

25:18

if you already called Cito,

25:22

tell your friends. We'll

25:28

gather round and hear this tale,

25:30

all ye sons and daughters of

25:32

a cursed motorboat at the bottom

25:34

of the Boston Harbor. Captain

25:36

Max sold him on a dream of heading

25:39

out to CEC. They emptied

25:41

out their bank accounts and

25:43

bought the Marjorie. Marjorie,

25:47

Marjorie. They

25:50

emptied out their bank accounts and

25:53

bought the Marjorie. Captain

25:56

Max was standing proud on

25:58

that fateful morning. Torso

26:00

bearing the wind in his hair

26:02

Suddenly without warning The engine died,

26:04

this r and

26:40

the ship. How do

26:42

you like them apples? So

26:46

that story was by Ike Suisse Kanderaja. The

26:49

sea shanty was written for us by Sam

26:51

Geller, who was there that fateful day. He

26:53

has Captain Max's brother. He performs as Samson,

26:55

the truest. Other writers on the

26:57

song, Ariel East, Chances with Wolves, and

26:59

Glasser, Gordon Manette, played the accordion. Coming

27:02

up, ship gets a new captain who

27:04

immediately throws half the crew overboard. He

27:07

says to save the ship from sinking. We

27:09

hear an insider's true real life account. That's

27:12

in a minute from Chicago Public Radio

27:14

when our program continues. This

27:19

is American Life from Ira Glass. Today's program,

27:21

What I Was Thinking as we were Sinking.

27:24

We have stories today about what goes through your head

27:26

in the middle of calamities, big and small, and what

27:28

those thoughts tell us. We have arrived at Act 2

27:30

of our program, Act 2, going

27:32

down with the censorship. So

27:35

the SS Marjorie cost our producer Ike and his friends

27:37

about 400 bucks each. The

27:40

ship we're going to talk about next is significantly

27:42

more expensive by $44 billion. That's

27:45

what Elon Musk paid for Twitter back in

27:47

October 2022, a few months

27:50

before we first aired today's episode. That was

27:52

back when we still called it Twitter. Now,

27:54

of course, it's called X. The company

27:56

has been foundering in the water ever since Musk

27:58

bought it, kind of famously. Just

28:00

this past week, Bloomberg News uncovered

28:02

documents that show that the company's revenues were

28:05

down 40% in the

28:07

first half of 2023. And

28:09

one year after Musk purchased Twitter, the

28:12

overall value of the company had dropped by

28:14

half. Its competitor Threads is

28:16

now passing it in daily US

28:18

users. And some of the

28:20

most revealing details about what it has been like

28:22

on the inside of the company. Among the people

28:24

who work there have come from reporter Casey

28:26

Newton and his colleague, Zoe Schiffer. They

28:28

make the newsletter platformer. They've

28:31

had so many scoops. Casey

28:33

also co-hosts a podcast called Hard Fork that The

28:35

New York Times puts out about the tech industry

28:38

that, can I say, I really love, especially their

28:40

deep and often very funny reporting on the latest

28:42

AI news, highest recommendation. But

28:44

to get back to our story, there's been

28:46

one person in particular at Twitter that

28:49

Casey has been wanting to talk to, a very

28:51

senior employee at the company who,

28:53

while just doing his job, ended

28:55

up having to take on two of the most powerful

28:57

people on the internet and in the world.

29:00

Those two people, Elon Musk and

29:03

the former president of the United States, Donald Trump.

29:06

Casey wanted to hear all about that. And

29:09

also what it was like for the guy, what

29:11

he was thinking, what he was doing, once

29:13

Elon took over and the place started taking on

29:15

water. Here's Casey Newton.

29:18

Yoel Roth did a lot of jobs at Twitter

29:20

over the years, but it was always the same

29:22

kind of job. He was in

29:24

the content moderation business. One

29:26

of those people who decides which of your posts can

29:28

stay up on the internet and which ones need to

29:30

come down. And he got

29:33

his first glimpse at what life as a

29:35

content moderator would be like while he was

29:37

in college on a date. He's

29:39

gay. So am I. I went

29:42

out for drinks with somebody without knowing where

29:44

he worked. And he volunteered

29:46

that he actually worked for the

29:48

parent company of the website Manhunt,

29:51

which was one of the

29:54

kind of early gay websites

29:56

that was very specifically sexually

29:58

focused. these early

30:00

days of the web, there was already a

30:02

team of people who were deciding what you could and

30:04

couldn't post there. They had a

30:06

set of kind of convoluted rules

30:08

about what types of nudity you

30:10

were allowed to show in which

30:13

places. So nudity, fine, but

30:15

not all nudity. So there were specifics.

30:18

And he described to me

30:20

a system of color-coding images of

30:22

red, yellow, green, and

30:25

then a team of people who were responsible for

30:27

making those designations. And I'll never forget, he said,

30:30

the people doing these reviews are

30:32

almost entirely straight women. And

30:35

I was just floored in that moment

30:37

of thinking, God, there's a team of

30:39

heterosexual women who have to look at

30:41

the depraved things that gay men are

30:43

posting on the internet. I'm so sorry.

30:47

And right. The senior whole

30:49

pick specialist at Manhunt was some poor

30:51

woman. That's not

30:53

an exaggeration. Yeah. Yeah. We

30:56

hope she's doing okay. Are you out there calling to this

30:58

American life? I'm so sorry. Yeah. I'm

31:00

sorry for what you

31:02

saw. After the date,

31:04

Yoel had one thought. I was like, aha,

31:07

that's my dissertation topic. Yoel

31:10

was in grad school. He got his PhD. And

31:13

soon after, a job at Twitter. They

31:15

gave him a small desk. This was 2015. The

31:19

office's most striking feature was probably

31:21

a giant life-size cardboard

31:24

cutout of Justin Bieber sat

31:26

directly behind my desk. Justin

31:29

Bieber obviously being a major figure in

31:31

early Twitter. Maybe the most

31:33

popular user, at least for some period of time. Yes. There

31:36

were rumors that Twitter had entire

31:38

servers just dedicated to serving Justin

31:41

Bieber-related traffic. Besides

31:43

Bieber, what Twitter was really known for back

31:45

then was its trolls. The site

31:48

was plagued by users harassing

31:50

other users, particularly women. That

31:53

year, I co-reported a story about how the

31:55

site's then CEO, Dick Costolo, wrote a memo

31:58

saying, quote, we suck. at

32:00

dealing with abuse on trolls on the platform, and

32:02

we've sucked at it for years. That

32:05

was the backdrop for Yoel's new job. As

32:08

an intern at Twitter the previous year, he spent

32:10

part of his time moderating content. He'd

32:12

seen this video of a dog getting abused.

32:15

He removed it from the site, but for years it

32:17

haunted him. It was never

32:19

even like the specific image. I couldn't, I

32:21

couldn't tell you what the dog looked like

32:24

or what the video was. I just

32:26

remember its existence. And I remember that

32:28

feeling of seeing it, and then of

32:30

clicking like I think the button said

32:32

no. More

32:35

than anyone ever talks about, it's

32:37

this mostly invisible job of content

32:39

moderation that makes Twitter usable for

32:42

the average person. It's what makes

32:44

every forum on the internet usable at all.

32:47

And Yoel was good at the job. He got

32:49

promotion after promotion in his department, what

32:51

Twitter and a lot of other tech companies now call

32:53

trust and safety. It's a hard job

32:56

and it just kept getting more complicated. The

32:59

way Yoel tells it, there was a wild new

33:01

case to examine almost every day. Foreign

33:03

governments impersonating their enemies, real

33:06

people organizing harassment campaigns, impossible

33:09

debates over what should count as hate speech,

33:12

and regular meetings over whether to put labels on

33:14

tweets that didn't quite violate the company's

33:16

rules, but would benefit from more context,

33:19

like about COVID. In

33:21

2020, the biggest case yet landed on Yoel's

33:23

desk. It was a case about

33:26

a user who kept causing problems. And

33:28

this guy's fans were even more rabid than

33:30

Justin Bieber's. It was the

33:32

president of the United States, Donald Trump.

33:36

This is a couple months into the pandemic. Trump

33:38

had tweeted that mail-in ballots in that year's election

33:41

were going to lead to widespread fraud. And

33:43

just to lay my own cards on the table,

33:46

I thought that was really bad, because they won't

33:48

lead to widespread fraud. Anyway,

33:50

Twitter's policies prohibited misleading people about

33:52

the voting process the way Trump

33:54

was doing. But the company had

33:56

never taken action against the president's tweets before. Yoel

34:00

had to decide what to do. I

34:02

didn't see a basis for changing

34:04

the policy, modifying it,

34:07

winking at it, squinting

34:10

and finding a violation. Like there was

34:12

no way around it. It was clearly

34:14

a violation of our policy. Truthfully,

34:16

there was a lot of nervousness

34:18

about crossing this line for the

34:20

first time taking action on a

34:23

tweet from the president of the United States.

34:25

The company decided that instead of removing the president's

34:28

post, it would put a label under it. A

34:30

label that just said, get the facts about

34:32

mail-in ballots, with a link to a

34:34

page that pushed back on Trump's claims. At

34:37

a certain point when it became clear that

34:39

yes, this was going to happen, it

34:42

became a question of who could push the button. On

34:45

some level, we probably understand that in a

34:47

moment like this, someone has to take a

34:49

physical action to type the words, get the

34:51

facts about mail-in ballots and click the button

34:53

to attach the label to the post. I've

34:56

talked to dozens of content moderators over the years,

34:58

but I've never talked to someone who had moderated

35:00

the president of the United States. When

35:03

it came time to take action, only a handful of

35:05

people at Twitter had the power to do it. The

35:08

company had locked down access after an

35:10

incident where a former contractor on his

35:13

last day working there briefly deactivated Trump's

35:15

account. Shout out to

35:17

bodyarduisek, who says it was an accident? Also,

35:20

Twitter had just introduced this idea of putting

35:22

labels on misinformation a couple of weeks before.

35:25

And so it was this perfect storm where

35:28

it required elevated access

35:30

and knowledge of this incredibly

35:33

convoluted system for applying these

35:35

labels. And I was the only one who

35:37

knew how to do it. And so I

35:40

got an instruction from my boss that said, all

35:42

right, we're going to do this. Also,

35:45

because this is how life goes, Yo-El and

35:47

his husband were moving houses the day all

35:49

this happened. I excused

35:52

myself from wrangling the

35:54

dog and the movers and the relocation

35:56

of stuff and sat in... the

36:00

front seat of the car with my

36:02

cell phone tethered to my work

36:04

laptop. I was on a video

36:07

call with some of the other leaders at

36:09

the company who were making this decision. And

36:12

I remember a countdown

36:15

where I was going

36:17

to push the button that

36:19

would apply the label to this tweet at

36:21

that same moment, Twitter's communication staff was going

36:24

to announce the decision. And it

36:26

felt very important in that moment for

36:28

the timing to be exactly joined up

36:30

for some reason. We

36:32

counted down, I clicked

36:35

the button, and then I refreshed

36:37

the public view of the tweet and saw

36:39

the label. And

36:41

the communications team said, we've got it from

36:43

here. And I said, okay, I have

36:45

to go back and deal with the movers now. And

36:48

I hung up the call and I closed my laptop

36:50

and I crossed the street back

36:53

into my apartment. If they

36:55

made a movie about Trump and Twitter, you

36:57

can imagine how they'd shoot this scene with

36:59

the Twitter employees hunched over a console in

37:01

a control room, high fiving. But

37:03

in reality, of course, it's the opposite. Most

37:06

content moderators try really hard not to bring their

37:08

own political beliefs into the job. In

37:11

a way, the legitimacy of the whole company they

37:13

work for depends on it. Shortly

37:16

before that Trump tweet, Twitter had

37:18

explained its reasoning for adding labels

37:20

to misleading information with a blog

37:22

post. Importantly, the post

37:24

was signed with Yoel's name. And

37:27

soon after that first label showed up on Trump's

37:29

tweet, his name was everywhere. I

37:32

wake up one morning, the third day that my husband

37:34

and I are in our new home, to my

37:37

phone exploding because Kellyanne Conway has

37:39

just talked about me on Fox

37:41

News and has said that I'm

37:44

responsible for the censorship of the

37:46

president's account and I'm responsible for

37:48

censorship at Twitter more generally. And

37:51

in that moment, everything exploded. Thank you

37:53

very much. We're here today to defend

37:56

free speech from one of the gravest

37:58

dangers. up

38:00

a copy of the New York

38:02

Post with me on it in

38:04

the Oval Office as he announced

38:07

an executive order restricting censorship by

38:09

Silicon Valley companies. His name is

38:11

Yoel Roth and he's

38:13

the one that said that mail-in

38:15

balloting, you look mail-in. No

38:18

fraud? No fraud, really? And

38:20

for weeks, discussion of

38:22

me and my political opinions and

38:24

my beliefs became a symbol

38:29

of everything that was allegedly wrong with Silicon

38:31

Valley and with the decisions that companies have

38:33

made. Twitter had to hire

38:35

security to protect Yoel and his husband. It

38:38

had all taken him by surprise. He'd

38:40

expected the criticism, but not that he would

38:42

be the target. In

38:44

cases like this, people would usually come after the

38:47

CEO or the company itself, but

38:49

soon Yoel realized that what his harassers were

38:51

doing was much more effective. If

38:54

you make companies believe that their employees could

38:56

be hurt for enforcing the rules, they

38:59

might be more reluctant to enforce them. Twitter

39:01

didn't stop though. They kept putting labels

39:03

on his tweets. And Trump, of course,

39:05

lost the election. Though that's probably

39:08

not how he would describe what happened. And

39:10

after the January 6th attacks of the Capitol, he

39:13

lost his Twitter account too. Yoel

39:15

did not press the button on that one, but

39:18

here's a detail about that day that I love. Yeah,

39:20

there was a technical question about whether it

39:22

would work or whether Twitter would crash. Can

39:25

you actually ban Donald Trump's account?

39:28

Banning somebody with that many followers

39:30

is actually technically very complicated. When

39:32

you suspend somebody, Twitter

39:35

systems have to figure out what to do with

39:38

all of the people who followed them. And

39:41

in other words, if you follow Trump, Twitter

39:43

has to remove him from your list of

39:45

followers. Which sounds very straightforward, but when you

39:47

have to do that tens of millions of

39:49

times immediately, we had

39:52

to think about like, if we push this button,

39:54

is the site going to go down? As

39:57

it turned out, the site stayed up and Trump was

39:59

banned. for a while, anyway. It

40:03

was such a strange moment. With

40:05

the click of a mouse, Twitter had managed to

40:07

do something that Congress attempted twice and failed. To

40:10

punish Donald Trump in a way that had real

40:12

and immediate consequences for him. Trump

40:21

headed off to Mar-a-Lago. You all

40:23

got promoted. He was running the whole

40:25

department. And that's when another mouthy

40:27

rich guy started to complain about all the rules on

40:29

Twitter. The guy was Elon

40:31

Musk. In April 2022, Musk

40:33

announced he'd acquired a big stake in the company.

40:37

A few days after that, he announced his intention to buy

40:39

an outright. As

40:41

soon as the news broke, EOL's employees started asking

40:43

what it meant for them. Elon

40:46

had been tweeting a lot about free speech, and

40:48

his feeling that Twitter didn't have enough of it. He

40:51

posted a photo of six people in dark robes with the caption,

40:54

Shadowban Council Reviewing Tweet, and Truth

40:58

Social Exists Because Twitter Censored Free Speech.

41:01

Also stuff like, next I'm gonna buy Coca-Cola and put

41:04

the cocaine back in. And

41:06

let's make Twitter maximum fun.

41:10

Some employees working in trust and safety worried that maximum

41:13

fun might mean Elon would dismantle their

41:15

whole operation. You all

41:17

was willing to give them a chance, though. What

41:19

I told them and what I sincerely believed was,

41:21

it's too soon to tell. People

41:24

are frequently caricatured and villainized

41:26

in the media. Certainly I was.

41:30

And that's not a reflection of who

41:32

they actually are. And so don't prejudge.

41:35

At the same time, Yuel knew that his

41:37

more concerned employees might be right. That

41:40

he was aboard a ship that might be about to sink. He

41:43

knew he needed to be alert for the sides. His

41:46

solution was to make a list. To

41:49

write down the red lines that he would not

41:51

cross, no matter what. Most

41:53

days his job was to enforce other people's rules.

41:56

But with Elon coming in, he wanted to write

41:58

down some rules for himself. You have

42:01

to have written policies and procedures

42:03

so that when the moment comes

42:05

to make that decision, you

42:07

just follow the procedure

42:09

that you had laid out before.

42:11

Your whole job was about trying

42:13

to not make decisions out

42:15

of impulse and emotion, but sort of

42:17

by following a playbook. And

42:20

that meant that before Elon took over, you actually had

42:22

to give yourself a playbook. That's right. And

42:25

so on a notepad by his desk at his

42:27

house, he wrote down his red lines. I

42:30

will not break the law. I

42:32

will not lie for him. I

42:34

will not undermine the integrity of an election.

42:37

By the way, if you ever find yourself making a

42:39

list like this, your job is insane. Then

42:43

you all wrote down one more rule. This

42:45

was like a big one. I will not

42:47

take arbitrary or unilateral

42:49

content moderation action. So

42:53

if Elon came up to you and just said, ban this person,

42:55

you were gonna do that. That was the limit. Did

42:58

people on your team show you the

43:00

list that they were making too, or talk to you about them? We did. Yoel's

43:09

list of rules got its first test

43:11

pretty quickly. On the day Elon officially

43:13

took over Twitter, it was

43:15

the end of October, lawyers were finalizing

43:17

paperwork, and Twitter staff was attempting to

43:20

enjoy the annual company Halloween party. The

43:23

scene was surreal. Were you

43:25

there for the Halloween party? I was. Were

43:27

you dressed up? I was not.

43:30

Lots of people did dress up though. Employees

43:32

brought their kids. There were balloons and

43:34

face painting. I've talked

43:36

to so many people who went to this party,

43:38

and every one of them has added some bizarre

43:40

new detail. Some people saw

43:43

a guy dressed as a scarecrow walking around with

43:45

what appeared to be a handler. They

43:47

wondered if it was Musk. It turned out

43:49

to be a hired performer. As the

43:51

Halloween party had started, I was sitting

43:54

in a conference room doing

43:56

some work, and we start

43:58

hearing rumors. that

44:00

not only has the deal closed, but

44:03

also the company's executives have been fired. And

44:07

at first it's unconfirmed.

44:09

I get texts from a

44:11

couple of reporters who asked me, is it true

44:13

that Vigia has been fired? And I said, no, I

44:15

just saw her. She's

44:19

still online in the company's like

44:21

Slack and Gmail. Like she's of

44:23

course not like your sources are

44:25

lying to you. And

44:29

then it was true. Such

44:31

an important lesson. Always trust the

44:33

reporters. Pretty

44:38

soon afterward, Yoel gets summoned over to the part

44:40

of headquarters where Elon and his team had set

44:42

up shop. He was nervous. And

44:44

I thought, okay, I'm about to be fired. So

44:47

I walk past a number of my employees and I

44:49

don't let on that any of this is happening because

44:52

I don't want to panic them because they're there with

44:54

their kids. And so I smile and make

44:56

jokes about Halloween costumes and walk over

44:59

to this other part of the office

45:02

where somebody who

45:05

I gather works for Elon Musk in

45:07

some capacity, but they don't introduce themselves.

45:10

They just say, how do

45:12

I get access to Twitter's internal

45:14

content moderation systems? And

45:17

I kind of pause and blink and

45:19

say, you don't. That's not going to

45:21

happen. I explained that Twitter

45:24

is operating under an FTC consent

45:26

decree, that access to internal systems

45:29

is regarded as highly sensitive and

45:31

that there are both legal

45:33

and policy reasons why we simply

45:36

couldn't grant access to somebody. Elon's

45:39

aid explains that they're worried about an insider threat,

45:42

someone who might try to sabotage the site on

45:44

their way out. Yoel

45:46

tells them, sure, I can help with that. He

45:49

explains some steps they can take to protect the company. And

45:52

to Yoel's surprise, the aid says, okay, you're going

45:55

to tell that to Elon. And

45:58

then he leaves and comes back. with

46:00

Elon Musk. Who at this point I've like seen on

46:02

the internet, but I had not met

46:04

in person. And so Elon

46:07

sits down and asks,

46:12

well, let me see our tools, our

46:14

tools. He owns the company at this point. And

46:17

so I show him his own

46:20

account in Twitter's set

46:22

of enforcement tools. And I explained to him

46:24

what the basic capabilities are. And

46:26

then I make a recommendation to him of what I

46:28

think Twitter should do to prevent

46:31

insider misuse of tools during

46:33

the corporate transition. You

46:36

will also had recommendations about the midterms

46:38

and the upcoming presidential election in Brazil.

46:40

And as I start to explain some of

46:42

the rationale related to the Brazilian election, Elon

46:45

interrupts me and says, yes,

46:47

Brazil, Bolsonaro and Lula, very

46:49

dangerous. We need to protect

46:51

that. And I was floored.

46:54

I came into that conversation expecting him

46:57

to fire me and instead he jumps

47:00

ahead of me to say that he

47:02

is sensitive to the risks of offline

47:04

violence in the context of the Brazilian

47:06

election and wants to make sure that

47:08

we don't interrupt Twitter's content moderation capabilities.

47:11

It was like a dream come true. You're

47:13

thinking maybe I'm actually aligned with this person.

47:16

Yes. Yeah. And

47:22

so you all stayed. He

47:24

was surprised in a good way. On

47:27

Twitter, Elon talked about the company as if it should

47:29

barely have any rules at all. But

47:31

in that moment, one on one, you all thought

47:34

he might turn out to be more reasonable. Maybe

47:36

spending some time inside the company would

47:38

show Elon the real value of those

47:40

rules, which is that without them, you

47:43

lose your users and you lose your

47:45

advertisers. And you all

47:47

felt like Elon could be sensible. One

47:49

of his first requests was to restore the account

47:51

of the Babylon B, a right wing satire site.

47:54

But you well explained how it had broken Twitter's

47:56

rules and Elon backed off. I

47:59

found him to be. funny, I

48:01

found him to be reasonable. I found

48:03

that he responded well to having evidence-backed

48:06

recommendations be put in front of him.

48:09

And I, for

48:12

a moment, felt that it might

48:14

be possible for Twitter's

48:17

trust and safety work to not

48:19

just continue, but also to get

48:21

better. After that,

48:23

things began to move really quickly. About

48:26

a week later, Elon laid off half the

48:28

staff. Suddenly, Yoel was

48:30

one of the highest-ranking employees from the old

48:32

Twitter, who was still working at the new

48:35

one. And it seemed like Elon liked him.

48:38

After some trolls went after Yoel for some

48:40

of his old tweets, Elon tweeted

48:42

that he supported him. The

48:49

U.S. midterm elections took place mostly without

48:51

incident. Same for the election in Brazil.

48:54

Elon kept pushing his teams to move faster,

48:56

even as he was laying them off. At

48:59

first, Yoel said that Twitter still had enough content moderators

49:01

to keep the site safe. But

49:03

the cuts kept coming, and the work got harder

49:06

and harder. Soon, Elon

49:08

unveiled his first big idea for making lots

49:10

of money and recouping the $44 billion

49:13

he had spent to buy the company. Yoel

49:16

and his team thought it was insane. The

49:18

plan? To let anyone get a

49:20

blue verified badge for their profile for $8

49:23

a month. The

49:25

company called it Twitter Blue. The

49:28

risks seemed obvious. People would

49:30

just make new accounts to impersonate brands

49:32

and politicians and other celebrities. Yoel

49:35

and his team wrote a seven-page document

49:37

outlining the risks. But

49:39

their badges went on sale anyway. And almost

49:41

immediately, impersonators started buying them and wreaking

49:43

havoc. In

49:45

maybe the most famous case, someone impersonated the

49:48

drugmaker Eli Lilly and said that insulin would

49:50

now be free. The

49:52

real Eli Lilly's stock price dropped more than 5%. It

49:56

was a vivid illustration of why companies like Twitter make rules

49:58

in the first place. impersonators

50:00

were suddenly all over the site. And

50:03

so, okay, we have to ban them,

50:05

but somebody has to review them. We can't

50:07

just like ban everyone. And

50:10

so you do that

50:12

with content moderators. And

50:14

we had instructions to fire more of

50:16

our contract content moderation staff to cut

50:18

costs. All this seems really self-evident to

50:21

me, and I think it would have

50:23

seemed self-evident even before you launched this.

50:27

What was Elon's take on this? Like, how

50:29

did he respond to you raising these concerns? Do

50:34

it anyway. And that

50:37

was a breaking point for me. We

50:40

reached out to Twitter for comment, but didn't hear back.

50:43

Reporters have sometimes gotten automated poop emojis,

50:45

but I didn't even get that. Yoel

50:52

had spent a long time gaming out scenarios for

50:54

what might make him leave Twitter. He

50:56

made that whole list. He wouldn't

50:58

break the law for Elon. He wouldn't undermine an

51:01

election. But ultimately what

51:03

got to him was something he didn't foresee.

51:05

It wasn't on the list. It

51:07

was something more personal. He

51:10

knew this bizarre plan wouldn't just make people lose

51:12

trust in Twitter. They would lose trust in

51:14

him. Behind Elon Musk, I was

51:17

the most prominent representative of

51:19

the company, period. And

51:21

I became aware that

51:23

when Twitter Blue turned into the

51:26

predictable hot mess that it was,

51:29

that people would ask, why

51:31

didn't the trust and safety team see

51:33

this coming? Yoel, why are you so bad

51:35

at your job? The

51:38

day after the launch, Yoel and Elon got on the

51:40

phone. Elon thought the

51:42

problem could be fixed if Apple would just

51:44

hand over all the credit card information of

51:47

the people doing the impersonations. Yoel

51:49

had to explain that Apple would never do that. He

51:52

also asked Elon to slow down the rollout of

51:54

Blue so that they would have time to hire

51:56

and train more content moderators to look for impersonators.

51:59

Elon did didn't understand why that would take longer than a day.

52:02

I got off that phone call and thought,

52:06

I can't solve this problem. I

52:09

will spend the rest of my time

52:11

at this company trying to bail out

52:13

a ship that might sink more slowly

52:15

because I'm there bailing it out, but

52:18

I don't want to spend the rest of my life bailing

52:20

out a sinking ship. Yoel

52:22

had made up his mind to leave. He

52:25

called a couple of his employees to let them know. I

52:27

knew that that day I did not

52:30

want to be walked out of Twitter

52:32

after almost eight years by corporate security.

52:34

I wanted to leave on my own

52:36

terms. And

52:38

there was an all hands going

52:40

on at the time, Elon's first time addressing

52:42

the company in person. And

52:46

during that all hands meeting, I hit

52:48

send on my resignation email, put

52:50

my laptop in my bag and walked out of the building

52:52

for the last time. Did you

52:56

purposefully send it when you knew he was

52:58

on stage? Yes, absolutely. I

53:00

knew that it would take some time for the

53:02

HR team to see it and process it

53:05

for that to get to him, for him to react

53:07

to it. And in that time

53:09

I knew I wanted to be back at home

53:11

and not be in the office. Was it a

53:14

long email? It was

53:16

one sentence. What was the sentence? I

53:22

am no longer able to perform the responsibilities

53:24

of my job and resign it as of

53:26

today at 5 p.m. I

53:34

remember feeling two things. On

53:37

one hand I felt relieved. And

53:40

then I also just felt deeply sad. I

53:42

just wanted to get home. So

53:45

I left Twitter's garage and was

53:47

driving. And I was about halfway

53:49

across the Bay Bridge when I think

53:52

Zoe broke the news that I left Twitter

53:54

and my phone exploded. And

53:56

I get that you didn't even get across the

53:58

bridge before Zoe broke the news. God, I

54:00

love her. Zoe is my coworker.

54:02

I'm immensely proud of her, even

54:05

if she did kind of mess up Yoel's plans. The

54:07

car Yoel was driving that day was a Tesla,

54:10

by the way. He was leasing it.

54:12

He'd been trying to return it, but couldn't get anyone

54:14

to respond to him. Maybe they'd

54:16

all been drafted to work at Twitter. ["The

54:24

World's Best Day of the

54:26

Year"] Yoel laid low for a few days.

54:28

He spent some time writing and published an op-ed in

54:30

the New York Times. It explained in

54:33

a very dry and principled way why

54:35

he'd left. That's

54:37

when some rando account re-shared something Yoel had

54:39

tweeted from 2010 about

54:41

relationships between adults and minors. Around

54:44

that time, he'd been working on his dissertation, which

54:46

called for tech companies to do more to protect

54:48

minors at gay hookup sites like Grindr. But

54:51

Elon replied with a tweet, quote, "'This

54:54

explains a lot.'" Then he

54:56

linked to Yoel's dissertation, quote,

54:59

"'Looks like Yoel is arguing in favor of

55:01

children being able to access adult internet services

55:03

in his PhD thesis.'" Not

55:06

true, but Yoel's phone exploded

55:08

with abusive messages. It

55:10

made the backlash to labeling a Trump tweet

55:12

look minor by comparison. Hundreds

55:14

of messages per hour, homophobic,

55:17

anti-Semitic, and

55:19

also violent, just deeply,

55:21

endlessly violent. And

55:24

he only had to tweet once. He

55:26

didn't even have to say directly, Yoel is

55:28

a pedophile. He just had to wink and

55:30

nod in that direction and people took his

55:32

lead. When

55:40

Yoel had first used the internet, it

55:42

felt like a small, self-contained space, separate

55:45

from what we used to call real life. But

55:48

by the time Yoel quit Twitter, the

55:50

distinction between online and off had

55:52

collapsed. And it had collapsed in

55:54

large part because of the company he worked at, Twitter.

55:59

The site brought together... so many of the

56:01

world's most influential people and then pitted

56:03

them against each other in these all-consuming

56:05

daily battles. And the anger coming

56:07

out of that could drive people to do things. Violent

56:10

things. Pretty soon,

56:12

Yoel and his husband were overwhelmed with

56:14

death threats. My husband

56:17

turned to me one day and

56:19

said, I've seen you through a lot

56:21

of being targeted and

56:23

being harassed. I've never

56:26

seen you look scared before. And

56:29

that was the moment that we decided to leave

56:31

our home. And

56:33

so, once again, they moved. I

56:37

met with Yoel at the temporary house that he and

56:39

his husband are staying at while they look for a

56:41

new place. After

56:43

all this, I thought Yoel might want a different kind of

56:45

job. I would have wanted a

56:47

different kind of job. The internet

56:49

had almost killed him, or threatened to

56:51

anyway. But still,

56:53

somehow, he's optimistic about what the

56:55

internet could be in a

56:57

way you almost never hear anymore. I

56:59

love the internet. I really do. I

57:02

think the

57:05

internet's power to bring

57:07

people together and help folks

57:10

all over the world find connections that

57:12

matter to them is

57:15

magical and is one

57:17

of humanity's greatest achievements. I

57:21

also think the internet can

57:23

be incredibly dangerous and scary.

57:25

And the work of trust and safety

57:27

is trying to push that back

57:29

a little bit and to make

57:31

the internet more of what it can be

57:34

and less of the dangers

57:36

of what it could turn into.

57:39

Yoel's idealism about the internet feels

57:42

radical, given how destabilizing it's been,

57:46

but I know what he means. Back

57:49

when he was a teenager, the internet gave

57:51

Yoel a place to discover other gay people,

57:53

the chance to talk to everyone in the

57:55

world instantaneously. It gave him a

57:57

career. It gave me all

57:59

those things too. I

58:02

remember life before the internet. It was

58:04

a less frantic time, but it was also a lonelier

58:06

one. Here's

58:13

how Twitter's doing since Yoel left. Hate

58:16

speech is on the rise. Advertisers

58:19

have fled. Banks

58:21

that funded Musk's takeover have marked their

58:23

investments down by more than half. Musk

58:26

himself has warned repeatedly that the site might

58:28

go bankrupt. I kind

58:30

of hope it does. Because what's

58:32

happening at Twitter right now is teaching us a

58:34

lesson it's taken us way too long to learn.

58:37

The people like Yoel, they're not the

58:40

enemies of free speech online, they're

58:42

the ones who make it possible. If

58:44

you get any value out of social media at all,

58:47

it's in part because of them. They

58:49

clean the place up, make it feel good to be

58:51

there. They pull us back when we

58:53

go too far. And they do

58:55

censor us. And like, of course

58:57

we hate them for it. We

59:00

convince ourselves we do a much better job if it

59:02

were us. That's what Yoel had

59:04

thought. Look what happened. Nobody

59:07

likes the guy enforcing the rules.

59:10

But watching Twitter sink into the ocean,

59:13

you can't help but notice how much you miss that

59:15

guy when he's gone. Casey

59:23

Newton. He's the co-host of the podcast

59:25

Hard Fork from the New York Times, who

59:27

were our collaborators on this. He also writes

59:29

the newsletter Platformer. His story was produced by

59:32

David Kestenbaum and Davis Land. Since

59:34

his story first aired last year, Yoel has taken on

59:36

a new job. He is now the head of trust

59:38

and safety for the parent company of

59:41

the dating apps Tinder and Hinge. For

1:00:00

his rockets and cars, he

1:00:02

had tons of acclaim But

1:00:04

Elon wanted more and it

1:00:06

ended in shame Ohhh Elon

1:00:13

wanted more and

1:00:15

it ended in shame He dragged

1:00:17

all the watchmen out from their

1:00:19

crow's nest And forced them to

1:00:21

walk down the plate to their

1:00:23

rest Leaving no one to look

1:00:25

out, no one who knew best

1:00:28

The ship fell apart in

1:00:31

a watery mess Ohhh

1:00:36

The ship fell apart in

1:00:38

a watery mess Diane

1:00:58

Wu, our managing editor Sara Abderaben,

1:01:01

our senior editors David Kestenbaum, our

1:01:03

executive editors Emmanuel Berry Help on

1:01:05

today's rerun from Henry Larson and

1:01:07

Safiya Riddle Special thanks today

1:01:09

to Kendra Stan Lee, David Pichalik, Emma

1:01:11

Lasherst, Shauna Lee, Jeremy Butman, Imrin Ahmed,

1:01:13

and the CTO captain who rescued the

1:01:16

Marjorie And spoke with Ike, who

1:01:18

would like to remain unnamed Our

1:01:20

second sea shanty in today's show was

1:01:23

also written by Samson the Truist, Sam

1:01:25

Geller, Strings and Engineering by Jessica Tansky

1:01:28

Our website thisamericanlife.org You

1:01:31

can stream our archive of over 800 shows for absolutely

1:01:33

free Also there's a

1:01:35

list of favorite shows if you're looking for something to listen to, all kinds

1:01:38

of other stuff there Again,

1:01:40

thisamericanlife.org This American Life is

1:01:42

delivered to public radio stations by PRX The

1:01:46

Public Radio Exchange Thanks as always to our programs

1:01:48

co-founder, Ms. Tori Malatia, you know He

1:01:50

and Kendall and Roman and Shiv were in

1:01:52

the deep end of the pool They

1:01:55

were nearly drowning So Tori looked

1:01:57

up and realized Look, it's so close, we can

1:01:59

just swim to low again. I'm Harold Glass, back

1:02:01

next week with more stories of This

1:02:03

American Life. Next

1:02:25

week on the podcast of This American Life, Bellin

1:02:28

thinks a huge part of who he is. Stuff

1:02:30

he is not so happy about himself. It can

1:02:32

be traced back to a moment before he was even born. The

1:02:35

moment when his dad met this guy named

1:02:37

Dave at the Oklahoma City Airport.

1:02:40

So one day, Bellin coughs Dave.

1:02:52

The devil he knows and the devil he

1:02:55

doesn't. Next week on the podcast

1:02:57

on your local public radio station. If

1:03:10

a friend asks how you're doing and

1:03:12

you say, I'm okay. When

1:03:14

the truth is, I don't want

1:03:17

my problems to burden anyone. Or

1:03:19

you say, hanging in there. Because if

1:03:21

I ask for help, they'll

1:03:24

just think I'm weak. Then

1:03:26

this is your sign to call, text

1:03:28

or chat 988 for free confidential

1:03:32

support. Anytime you

1:03:35

don't have to hide how you feel.

Unlock more with Podchaser Pro

  • Audience Insights
  • Contact Information
  • Demographics
  • Charts
  • Sponsor History
  • and More!
Pro Features