0:00
/
0:00

Steven Levitan: Lessons from Writing 100s of TV Episodes

The co-creator of Modern Family

Steven Levitan is the creator of Modern Family, one of the most popular sitcoms in history, which won 22 Emmy Awards.

This conversation is all about how he wrote it. Some highlights below:

  1. Know the theme of what you’re writing, as it’ll give you a north star for every decision you make.

  2. Better stories come from bigger conflicts, but bigger doesn’t mean screaming and explosions, but rather the high stakes that come from a conflict that’ll cost the characters emotionally if it isn’t resolved.

  3. What kinds of conflicts work best? The stuff that never changes: love, loss, heartbreak… relationships between parents and kids, couples working through their problems.

  4. The story is everything. Once you crack what’s actually going to happen, the script practically writes itself. That’s why there’s an old screenwriter saying that the first draft is 90% finished once you’re done with the outline.

  5. Cut the half-jokes: A lukewarm joke kills momentum faster than no joke at all.

  6. Be real. Take it down a notch. Steven gets annoyed by “big.” When things feel heightened to the point where they don’t feel human anymore, or when they’re performing on stage, to an audience, instead of just being real people.

  7. Emotional moments must be earned. You want them to sneak up on you, to be understated.

  8. Steven would spy on his family for TV gold. Sometimes, he’d be sitting at the table, take his phone out to frantically capture the dialogue. When friends came over, his kids would joke: “Be careful what you say around the Levitan house. It might end up on Modern Family."

  9. A writer’s antenna needs to be up at all times. If you're observant, there's almost nowhere you can't go where you don't think: "That's funny."

  10. Some questions for writers to ask themselves: Does it feel real? Are they talking the way people talk? Are you tapping into what you're thinking about today?

  11. Conflict is everything, but how you do it matters. Don't just slap "these two hate each other" on your characters and call it conflict. Dig into who they actually are: what specific traits, backgrounds, or worldviews make them rub each other the wrong way? The clash has to come from something real.

  12. Steven’s mantra for writing Modern Family episodes: Keep it moving, but don’t race through them (he thinks the episodes became too frenetic in the later years).

  13. How was modern family written? ~7 writers. 8-10 weeks of story breaking at the beginning of a new season. Then the weekly cycle began: Mondays and Tuesdays were for rewrites, Wednesdays were for table reads, and then the final versions were due by Friday.

  14. What happens before a show is launched? Writers spend weeks doing nothing but talking about characters and the dynamics between them, with the goal of establishing relationships that’ll play out over the course of hundreds of episodes (if things go well).

  15. Give your characters a comedic lens to see the word through. On Modern Family, Phil is the Dad and he thinks he’s the “cool dad” with misguided confidence that he’s hipper than he really is, so the writers of the show knew how he’d react to things before they even wrote the scene.

Transcript

David Perell:

Well, across Frasier, Just Shoot Me!, Modern Family, you've written hundreds, hundreds of TV episodes.

Steven Levitan:

Yes, I have.

David Perell:

And there's gotta be a speed, a speed of writing that you've learned to cultivate. How have you done that?

Steven Levitan:

Well, it's actually interesting because my first job out of school, I was a TV reporter.

I would have to go out to a location. Something's going on, I have to take it in the location. I have to figure out what it is, who I'm talking to, shoot some stuff in my mind, know what I'm going to be saying on air, and sometimes have to say it two minutes later, live. Sometimes I have to rush back, write a script, edit it in time for the six o' clock news. And that was a pace that I got very used to.

So I remember when I got my first TV writing job, they're like, "Oh my God, we're going to need the script so fast. Can anybody write something really fast?"

And I said, "I probably can. When do you need it?"

They're like, "In a week."

And I'm like, "Well, that's an eternity to me right now."

So I've always been kind of a fast writer.

Part of that is I just like to get it done. I'm anxious to get the process over with.

So I jump in and I can't stop thinking about it until it's finished, and then there's a giant relief for me when it's finished.

David Perell:

Do you feel like there's fast twitch and slow twitch writers and you're like a fast twitch writer?

Steven Levitan:

Thousand percent.

I've worked closely with some slow twitch writers, and it's both really good and at times really frustrating to have those different speeds, because sometimes I'm just like, "Let's keep moving, let's keep moving."

There's a certain momentum you gain when you're really moving towards something, and then you can go back.

My attitude is you can go back and fix and you can go back and look at it, and other people are much more, "Slow down. Let's really think this through. Let's be deliberate about this."

I think that there are advantages to both, and it's probably good for me to have at least somebody with that voice in the room.

David Perell:

So then with an episode like the one where you do the whole one on FaceTime and the Modern Family episode. So with that, I presume that was kind of a big risk to take. So are you just like, "Hey, I got this idea, let's go for it?"

Did that idea come up and you're like, "Hey, let's let's make it happen?"

Or was that something that you had been stirring on over some time where like now it's time to shoot the shot?

Steven Levitan:

So I remember I had that idea to do one of the... you know how we also always had multiple storylines going on. And I said, "Wouldn't it be kind of cool if we can do one storyline that took place just on the computer, so it'd be like a third of the episode."

I just thought that would be cool, and it was in the back of my head.

Then I saw this short film that these guys had done where it was a story told totally on screen. And I'm like, "Well, if they can do that, we can do that."

So I came in one day with, told everybody in my room, "This is the goal. Let's see if we can come up with a storyline that works."

And sort of miraculously, we stumbled upon an idea that really fed that really nicely. The idea of like we can't find somebody and there's a little bit of panic and she lost her phone.

We worked really hard on that. That was a very time consuming episode, but I was really, really proud of the way that that turned out.

David Perell:

Well, this is a good chance to dive into the way that writing and directing kind of weave together, right? So with that, how much of a script like that is done in the writing versus, "Hey, now we're actually making it and all these new ideas are developing?"

Steven Levitan:

It was pretty much in the writing because it was a really production intensive episode because everything you saw on the screens had to be manufactured by us. It wasn't like we were just taking screenshots.

Every little side thing that was on somebody's screen, every little background behind everybody, everything had to be.

David Perell:

It says "porn" on her top right corner and that it's like her Pinterest board of organization porn.

Steven Levitan:

All those jokes had to be written. They had to be thought of. I worked on that one, the room. We worked on it together as a group and really broke it out. Then I worked with Megan Ganz.

I chose her to write it with me because she's prone to geek out the way that I am over those sorts of things, and she got really deep into populating the screens in a really impressive way. So it was a fun process.

I had done so much by then that I was always looking for something that would be a giant challenge, that would keep me interested in doing this. I was thinking, "I like this. This is going to be hard, but let's see if we can really pull this off." And it was very fun to do.

David Perell:

Geeking out? Tell me about that.

Steven Levitan:

Well, I'm prone to geek out on things. I'm very techy, and I take deep, obsessive dives into really weird, obscure things.

The idea for that one, in particular, came to me when I was having a FaceTime conversation with my daughter, who may have been in college. I remember my wife came behind me, and then I was getting a text from somebody on my screen at the same time. I remember thinking, "Boy, you can really tell a lot about my life by what's on this screen." That's how that originally started.

I'm often looking at new tech. How can we work that in? Like, Phil being on an iPad that's on a thing that moves around the room? How can we make that? That seems like something Phil would do. Phil was nerdy and tech the way I was.

In my first show, "Just Shoot Me," which I created, Jack Gallo was into Sharper Image and all those stupid gadgets from there. I was at the time, so it was a way for me to sort of indulge all my nerdy interests.

David Perell:

Well, the thing that's cool about Modern Family is, obviously, it's a fictional show, but this feels more real than reality TV.

Steven Levitan:

Yeah, the idea for it came out of our lives. I wrote that. I created that with Christopher Lloyd. We were on a deal together at the time, and we were working on another idea that we liked, but we would come in on Monday mornings and go, "How was your weekend?" "Oh, my God, I did this and this. My son..." You know, it's crazy.

I'm on a phone call, and some people came over, and I walked out in the backyard, and my son is bouncing naked on the trampoline with a box on his head. Things like that. You would just tell funny little stories: an argument you got in with your wife or some trouble your kids got into or whatever.

We started to realize that those were in many ways more compelling stories than the ones we were trying to figure out for this other show. Then it occurred to us at that time, because Raymond had just ended, that there wasn't really a good Family show at that point that we really liked, there wasn't one that was sort of speaking to what was happening with families today. So then we said, well, how has family changed and how can we play with that idea?

Family has changed because we have very different looking families now, thus a gay couple raising a kid.

David Perell:

Immigrant.

Steven Levitan:

Immigrant, cross-generational, cross-cultural families. Then smartphones, technology, the way that that all works, and the way that that's changed the way that traditional families operate.

David Perell:

You were talking about the speed of developing episodes, but in the episode in Vegas with the bathtub, I was like, how did you keep all this in your head?

Steven Levitan:

I ran, I oversaw half the episodes. So that's kind of the way we did it. Early on we realized, because Chris and I often saw things from very different points of view and there were only two of us.

If there had been three of us, then we could solve arguments with a tiebreaker. We had two of us, so we quickly realized let's just decide that every week we alternate who wins the battle so we can move forward.

Because there's no right answers or wrong answers. It's just we've got to pick a direction. Otherwise we can sit here and argue about this for an hour, and that's not good for anybody.

It started that way, and then it eventually evolved into: because it's also quite efficient, we would shoot a show, and let's say Chris was on set shooting his episode. He'd be there with the writer of that episode, and then I would be back with everybody else that's left for that week at my disposal.

We would come in, we would rewrite whatever script was going to be up that week. We would be working on future stories. Maybe there'd be a writer or two off doing their future episodes, but that's the way we kind of worked it. So the staff kind of mingled between the two of us, and then it would switch the next week.

David Perell:

So then the question behind that, though, is there's still a lot of complexity in the episodes of just the different threads and how they come together. So how did you manage that?

Did you whiteboard things? Do you have pieces of paper? How does that work?

Steven Levitan:

One of the tricky things about Modern Family is we were juggling three or four stories in an episode. You realize, okay, well, who's not in this one? And then, okay, well, we have an idea over here for another for a secondary story.

But wait a minute. This story takes place. It's got to be over two days. There's no way to do it over one day. We need that time or whatever, and this one takes place over a week.

That doesn't work. So okay, we'll save that one maybe for a later one, or figure out a way to make this one longer. It was a lot of math.

David Perell:

That's what it feels like.

Steven Levitan:

You're weaving, and we had ten characters or more, and we had to make sure that everybody had something to do and had a point of view and how are they working into this story.

I think that's what gave Modern Family its pace and its energy and made it feel a little bit, along with writing that was modern and observational. It gave it a more contemporary feeling than a more traditional show.

David Perell:

Tell me more about pace and energy.

Steven Levitan:

We try to keep it moving. I mean, when you're telling three or four stories in an episode, you've got to get to it, and you've got to be very, very economical.

At the same time, you don't want to race through it, and that was also a challenge. I feel like many of the episodes in the later part of the show became too frenetic, and I wanted to sometimes do less stories so that we could settle in and just enjoy a moment and stop for some dialogue. As I would always say, we need to let it breathe a little bit or just stop for some real conversation. So it was always finding that balancing act.

Modern Family was a network sitcom or a network comedy, and it was 21 minutes and 30 seconds, including our little main title, which was 10 seconds. That's a short amount of time to tell three or four stories. You're basically telling a seven or eight-minute story every week in a particular episode. So you just have to keep it moving.

The talking heads are a wonderful device. I remember one of the things we said early on is, "Let's do everything we can to make this easy on us, to write it well and to produce it well. So let's do talking heads." Even though The Office had already been doing them, we're going to do them more as couples, and that'll make ours a little bit different because it's more about the dynamic than it is just about Michael Scott sitting there talking.

Let's set it in LA so we don't have to fake everything. We don't have to hide the palm trees or the license plates that drove by or whatever. Let's just set it where we live. That's another sort of an interesting little thing, which is we took it really from our lives in many, many, many ways.

David Perell:

Well, that's why I feel so real.

Steven Levitan:

Yeah. So, for example, I thought in my head that Jay had gotten divorced. He probably lived in a traditional house when he was married to Shelley Long's character, and she probably had tchotchkes everywhere. When he got divorced, he just wanted something very, very different, so he went out, and he probably doesn't have great taste, and he went out and he bought a modern house.

I said, "It's one of just like a modern house," and they brought me some pictures. I'm like, "No, no, no, not like that. There's one down the street from me." I described where it was so that they could see it as an example. They went to that house. That's the house we used.

For example, the little circle where he taught Gloria to ride a bike, that's where I taught my kids to ride a bike. Exactly the spots. There were many, many places, all sorts of little bits that were just so real, based on what we were going through at that time—kids getting driver's licenses or things happening at school. We were constantly stealing from our wives and kids.

David Perell:

I read somewhere that one of your daughter's friends said, "Hey, be careful what you say around the house. It might end up in Modern Family."

Steven Levitan:

Yeah, there were things that happened. I had two older daughters and then a son, so it was very much Dunphy-esque in terms of the ages and all that. They would be getting into a fight at the kitchen table, and I would just pull out my phone and start writing dialogue: "Oh my god, okay, the sweater."

In the first couple years when it first came on, we had a tradition where my whole family would watch it live. I remember one particular time where Alex and Haley are coming down the stairs in a major fight over something, and it's literally the lines that my kids had said, and I just look over, and they're both looking at me like this.

David Perell:

Giving you the death stare.

Steven Levitan:

We did things like that all the time. There was another time where I probably crossed lines. I used a kid's name that my daughter was worried everybody would think she has a crush on him, because I just liked his name and I'd heard his name. So, I used that name as a character in an episode, and she was worried that everybody was going to think that she liked that kid at school.

I gave my son some advice about how to break up with a girl, and he did it, and then it didn't go so well. So, then I told that story as part of an Alex story, and that didn't go over well with my son that that was eventually on TV. But by and large, they did love it because it was a show that their friends watched and really liked. So, I think it felt very special to them.

David Perell:

How do you do this? The show has a sense of character, and you're not the only one writing. There's a writer's room. So, as you are a leader of this team, what do you tell writers so that they can understand your vision and your voice? How do you make that explicit?

Steven Levitan:

What was tricky about that is that they had two masters. They had me one week and Chris the next week, and we had very different tastes.

I favored things that just felt extremely real, extremely natural. I would come in and say, "By the way, the stories came from everybody. All the writers came in with great ideas. Some of the best episodes came from the writers' lives." I don't say I always came in with the idea, but I'd come in with a particular notion: this happened to me this weekend, and I had this thought, and I think there's an episode there.

That to me was typically enough to start. I liked it coming from a real place. I'm thinking about this; this is the idea that I like here. Then it's a matter of taste, tone. I always say tone is set by 10,000 little decisions. No, we don't want to do that; it feels too jokey. No, that makes the character unlikable if they do that. Oh, I love that. Oh, my God, that little thing there brings a sense of hope, which I like in that moment. That's how tone is set.

David Perell:

Tom Ford was once asked how he built his company, and he said something to the effect of, "I just said yes, no, yes, no, yes, no, 100,000 times." That's how Tom Ford became Tom Ford. It's just a bunch of yes and no decisions.

Steven Levitan:

That's it. You got to have just sort of a sense of instinct. I always tell people, if you're ever hiring a second in command when you're putting together a staff, the most important quality that that person can have is that they share your tone, that they get what you're trying to do.

David Perell:

Like intuitively or that they understand it over time, and they can adapt to it.

Steven Levitan:

Well, you don't always know, but it's best if they just share it. They naturally go that way. Some people naturally, and there's nothing wrong with it. They naturally like something that's more broad.

Chris tends to like things that are very, and I don't mean this in a pejorative way at all, like very clever. Like, "Oh, isn't it clever that that thing that was there comes back later on, and then it builds to this moment?"

There's a very thoughtful process about it, where I was kind of approaching things like I just remember sitting there watching my kid in this moment and thinking this. And that's how I sometimes came at that. It didn't have to be super logical, because life sometimes takes these weird turns, and you have to be open for that. That ending for that episode that's so poignant just came out of nowhere there, but in a way that felt like it would happen to me in real life.

I can remember certain episodes where I was like, "I think this will work," and people weren't sure. I felt like, "No, let's just go on this ride with me."

And then you said, "That's what you have to do."

I always would tell people when you would be debating things, I'd say, "I may not be right, but we have to make a decision. We have to move forward, and this is what my gut's telling me, and I don't want to have to second guess it tomorrow. So if I'm wrong, we'll fix it, but this is the way we're gonna go, and let's just get on board."

David Perell:

Can you paint a picture for me of any given Tuesday? Like, what is a Tuesday like for you as a writer in the writer's room?

Steven Levitan:

What is going on on Modern Family?

David Perell:

Yeah.

Steven Levitan:

When you first start the season, you're doing eight weeks or more, 10 weeks of story breaking, and you're just breaking stories and sending off writers and start to get scripts.

David Perell:

So what's that story breaking?

Steven Levitan:

You're just going, what if? What if? What if this happens? What if that happens? Okay, we have a good, okay, that's a good Phil and Claire story. We need a Jay and Gloria story to go here.

David Perell:

And you're doing it for the entire season at once.

Steven Levitan:

No, you're just doing it one episode at a time. There were certain seasons that we would talk about, "Let's have a seasonal arc," like Mitch and Cam getting married, or Gloria's gonna have a baby, or whatever that is. Things with Haley and her boyfriends and things.

But we were kind of in that old school of being somewhat episodic. So we just need a good story. Like, who's got a good story? And so you just start like this. We need another one.

Stories are gold. Stories are everything.

There's an old story about two screenwriters talking, and one of them says, "How's your new screenplay coming?" He goes, "Well, I just finished the outline and I'm about to start the first draft, so I'm 90% done, right?" And that's the way I think we all looked at it. Like, the drafts are easy. It's just coming up with what happens here.

So it all works. When you feel like you have a great story, like, "Oh, my God, that's gotta be so funny," the script is in some way, it's not secondary, but it's just like the story works and you can feel it. That's everything. The hardest part of the job. So I'd come home.

David Perell:

Do most screenwriters feel that way, or is that really particular to you?

Steven Levitan:

That's surprising to me. I think that's more the case than not.

I remember coming home in that time, and I always equated it to sort of, we'd start in June 1st, and it would be the summer. So it was a terrible way to spend the summer because it was just endless story breaking. It felt like doing math problems all day, like math word problems. "If he's six dozen miles away from this," that's what it felt like.

So I'd come home at the end of the day and I would be fried. My brain would just be throbbing. And my wife would say, "How's it going? What are you working on? Tell me about what you did."

David Perell:

Don't get me started.

Steven Levitan:

I just need to not think about it for a while. I'd go watch TV, do whatever, go out, whatever, and then I'd wake up at three in the morning, thinking about that story because if we didn't fix it, we can't get it. Sometimes a story would break really quickly, occasionally. Sometimes you'd be five days and you have nothing, which is really frustrating and by far the hardest part.

To answer your first question, a week would look like this: I'm in the room, Chris is on stage shooting an episode. Monday, we start rewriting the script that we're going to read on Wednesday at the table, a table read. Monday we'll do a rewrite on that. That will quite often go into Tuesday.

Typically, because we had a pretty large writing staff, I was able to take half the people to do the rewrite with me, and this group would go break story, try to break some stories for the upcoming episodes. We'd have six or seven people doing the rewrite in the room. We'd go through the script, and that would typically take two days, or it had to be finished in two days because we would read it on Wednesday.

David Perell:

You had no choice.

Steven Levitan:

Yeah. So Wednesday lunchtime, we always did a table read. You'd do that, you'd come back, we would all talk, everybody give each other notes, like, here was what we're thinking, this is how it felt, good. Then we would start the rewrite, and that had to be done by Friday when we left. We would start shooting that one on Monday.

There would be production meetings going on, editing of previous ones you're shooting. You're running around, bouncing around and putting out fires. What do you think about this car for Jay's car in the next episode? It's just constant decisions.

David Perell:

How do you think about the relationship between story and comedy? The thing that surprised me most about talking to comedy writers is they're like, dude, story is king over and over and over again. I don't know why, just me being naive, I was like, oh no, joke is king, you're a comedy writer.

Steven Levitan:

The best jokes, the best laughs come out of character, first of all. So that it's just natural behavior. You just know there's going to be good, clever turns of phrase. I think that that sort of thing was a little bit more in vogue a few years ago than it is today.

I always say comedy is like fashion and music. It really evolves, it changes. It's just very different, like the old setup punchline, setup punchline.

David Perell:

That's like, oh shucks, 1960s, 1970s.

Steven Levitan:

That really feels old. That's true. It really feels old. And there is a generation of comedy writers who are really allergic to jokes. They don't want jokes because they say it feels sitcom-y.

I understand that. A bad joke or a joke that feels a little bit forced just takes you out of the moment. Something that's much more naturalistic, I think that just came along.

Larry Sanders, I think, was much more trying to be much more natural than Frasier was very much about the joke. A beautifully written, well-crafted joke. When it's beautifully written and it's well-crafted and it's smart, those actually hold up.

It's the ones that don't, that are a little bit like set up, punchline, set up, punchline, that age terribly. Story is the thing that moves you along. I'm interested in this. I'm caring about this. I want to see what happens next. I did not see that twist coming. That's funny.

Just walking in and seeing that person. Mitch and Cam, they're going to apply to a private school, and they realize that they're gay and they have an adopted Asian daughter. They realize that they're, for the first time in both of their lives, very much in demand as Mitch, who was always picked last on the playground. At these private schools, they're a goal. They're told, oh, my God, you can get in anywhere.

Then they're sitting in the waiting room of the private school thinking they got it locked. Then in walks a lesbian with a black baby, and then the partner comes in and she's in a wheelchair, and they just look at it going, oh, my God, with the handicap kicker. She wins this hand. That's just good.

It's story. It's a twist in the story that suddenly they went from being on top to, oh, my God, we're in second place.

David Perell:

We became the royal flush.

Steven Levitan:

Those kind of moments are very important.

David Perell:

Tell me about how character leads to laughter. One of the things that comes to mind is in the pilot episode when one of the dad comes in to present the adopted baby, and we have the Lion King in the background, so the music.

Then his husband says, turn it off. He goes, I can't. It's just who I am. That's all out of character. It's exactly what you're saying it's not a joke, but it just reveals so much. There's all these different layers there, and you kind of bring that all together and get the laugh from that.

Steven Levitan:

There's very sweet things. In that moment, Jay saying something along the lines of, let's see the little pot sticker. And Phil, in that moment, going when he learns that her name is Lily, he goes, isn't that going to be hard for her to say? So you're getting little character insights in those moments.

Originally, we had written a scene that involved Claire and Mitchell going into the baby's room to change the baby. The Lion King thing wasn't in it at all. There was a baby monitor on, and everybody was out there listening to their conversation. That was originally what we had written, and it was well written and all that. I remember thinking, I wish this was more visceral.

David Perell:

Visceral?

Steven Levitan:

Yeah. I wish there was something we could do that just has a bigger feeling to it that captures the characters without it being a talky scene. Then we just started talking about, what could it be?

I don't know how. I don't remember how the actual Lion King thing came up, but it felt like it built.

I remember when Modern Family was going to. We had shot the pilot. We were in contention for the fall. They told us they were going to pick it up. I went to New York for the upfronts, which used to be such a big deal. It was a big thing.

David Perell:

That's where you would sell the show.

Steven Levitan:

Yeah, they would show the lineup to all the advertisers. It was a giant gathering.

David Perell:

Oh, so that's when advertisers would see it.

Steven Levitan:

Yeah, and a lot of press would see it for the first time. It was going to be at Lincoln Center, full Lincoln Center.

And I got a call in Ms. Streets in New York, and I got a call from Steve McPherson, who said, "Hey, we're not going to show clips from Modern Family in the upfronts."

I'm like, "You're not?"

We're going to show the entire pilot.

Oh, wow.

That had been done very few times in history of the upfronts anywhere. The last time it had been done was for the show Joey, which was a Friends spin-off, and it didn't—apparently, I wasn't there—it didn't play well.

And so it was a scary thing.

I remember I was very nervous. I came in, the lights dimmed, or we came in, and then Ty Burrell and Sarah Hyland came, and we all sat together. So we get to this. They start showing the Modern Family pilot, and we're worried. It's not built for the studio audience. Are the laughs going to step over things? Are there going to be any laughs? Can they hear it? Is it going to play? It's not really a punchline setup thing. Is it going to play?

Anyway, it just started to build. The laughs just started to build. You could just tell people were laughing and they were into it. It was just rolling.

It just kept, and we're looking at each other like, "Holy shit, this is working!"

Then we get to the Lion King moment, and he goes like this, and the place erupts.

I've never heard an upfront erupt before, like massive applause. People, this is during the show, and people were laughing and massive applause. That's right near the end, and it ends in huge ovation. Then they say, "That's our lineup, everybody. Let's go to the party."

We get up, and Sarah Hyland and Ty Burrell, who had walked in anonymous, were walking in celebrities. Everybody wanted to congratulate them, take their pictures. Oh, my God, that was amazing. I watched their lives change before my eyes. It was crazy, but that is truly one of the great moments.

David Perell:

Tell me about making a pilot. There must be—this is my best summary of what a pilot is. You basically make one episode so that you can show the studios, "Hey, this is the kind of thing we're going for." It's concrete enough that it can then become the first episode of the show if it's good. Then you use it to get the money to finance the first season. Is that right?

Steven Levitan:

Essentially, yeah.

The studio is financing it, but they're going to make a bunch of pilots. Of those pilots, a select few get picked, and then of those select few that get picked, an even more select couple survive. Every once in a while, one of them becomes a giant hit. So there are many pilots.

David Perell:

But then, for that one, I want to get a sense of how you think about the character development coming into it. With this show in particular, there were a lot of different characters to develop. Also, I know that that's something that you learned to do over the course of your career. That isn't something that you had at the beginning, but then I think you nailed it. Some of the characters from Modern Family are just so good.

Steven Levitan:

We spent a lot of time on it. Chris and I probably spent six weeks doing nothing but talking about the characters and their dynamics.

David Perell:

Was that math brain too?

Steven Levitan:

Part of it is because I talk about this a lot when I'm talking to people about doing a pilot. I'm always surprised that people don't think about it in these terms typically.

The most important thing you're doing is establishing dynamics that will play for hopefully hundreds of episodes.

David Perell:

Dynamics, yes.

Steven Levitan:

The way I look at it now is I have this theory; I call it "two points in a line." You start with a point, and that's a character. Let's say that's Oscar. Oscar is messy, single, doesn't care about much, and has a lot of other habits.

Now, we're going to have somebody move in, and it's going to be about their relationship. We need to make that line between them crackle like a high-power electrical line. If it's just, "Oh, he's got a friend, and they're buddies, and they like to hang out and talk sports," it's a line. They're friends, but is it crackling? So, how do you make it crackle?

Oscar's the messiest person you've ever seen. Felix is the neatest person you've ever seen and persnickety. Oscar smokes cigars, and Felix has allergies, and on and on and on. You just keep adding attributes to each character to make it more alive, to make that line more alive.

Sam and Diane. Let's start by just reducing them: dumb jock running a bar, a womanizing jock, a snobby academic woman. Okay, that's starting to feel like there's some. She looks down on him for being a womanizer, and she's smarter than him, but he's more street smart than she is. That line is just getting better and better.

Now, we're going to add some other characters. We're going to add Carla. We need to draw a line between Carla and Diane and Carla and Sam, and those lines need to be crackling, too. How do you do that? Carla is kind of in love with Sam, idolizes Sam. He's her god.

What about Diane? She's got to hate Diane because Sam has a thing for Diane, and she doesn't understand it. She's got to be very blue collar next to the erudite academic. That line is starting to get crackling, and they don't like each other.

Then you add all these other characters, and you do the same thing. You keep drawing lines between every character that you've added until it starts to look like this weird bicycle spoke kind of thing. You're putting everybody there and connecting everybody. If two of them are stuck in an elevator, to use an old corny sitcom setup, or just sitting there having a conversation about something that just happened, do I know what's going to be funny about the two of them before I'm even starting? Because I know their points of view, their tensions, and what's going on there. It doesn't have to be that they're always opposites.

It could be an unrequited love, Niles and Daphne. Niles is in love with Daphne. There wasn't conflict between them. It was just he was in love with her, and she didn't quite know it for a long time.

Any form, the line just needs to be compelling. It could be compelling for love reasons, Pam and Jim on The Office, or it could be compelling for comedic reasons, Jim and Dwight on The Office. That's a thing that takes time.

I feel like it's skipped over quite often because when I read a lot of pilots, they tend to be either premise-y. So, we've set up a nice premise.

It's a pilot. "Oh, he's a superhero." Things that are based on premise get old.

David Perell:

It's not enough to sustain a show.

Steven Levitan:

No, you got to care about the characters, and you got to know. I mean, we watched Sam and Diane evolve over time. It was fun to watch that dance because they came from such different places.

But at the same time, you could bring in somebody like Frasier, who was another version of Diane. The brilliance of the setup for that was the initial instinct probably was, "Well, we'll give Frasier his father, and we'll give him a brother who's sort of the same, cut from the same cloth as the father, and Frasier is the odd man out." That would have been probably fine. But the brilliance was we're going to bring in Niles, who was more Frazier than Frasier. Now they're competitive, and now Frazier's really in the middle, weirdly.

There's a snobby, comparative thing here. There's a competition for father's love, for who's more successful, who's more sophisticated, who's more knowledgeable. And then you have the dynamic between Frasier and Niles and their father, who's blue collar. How did this guy get these two kids? So those dynamics were just there.

We spend a lot of time working through that. What's the difference between Haley, Alex, and Luke? By the way, it's a choice. Let's give them three kids. Let's not give them two kids. Let's give him one kid. Let's make him two girls and a boy. Whatever.

Haley is very kind of shallow, and she's great, popular, great with boys, love her. She's very sort of street smart but not book smart. Alex is a book smart nerd, and very tense. She overthinks everything, whereas Haley overthinks nothing. And then Luke is just sort of like the dumb, kind of, in the beginning, just sort of the dumb, funny, comic relief kid who is the boy. He's just a boy, very different energy from the girls. And then, how do Phil and Luke interact? What's the difference between Phil and Claire? They're not just a couple that are raising two kids. Phil's like the fourth child, and she's got to always feel like she's got to be the responsible one.

David Perell:

Right.

Steven Levitan:

He loves Claire and is impressed by Claire, but she's a little rigid, and it's his job to make life more fun for Claire. Therein lies that dynamic.

David Perell:

So round out my takeaway from what you've said. I think it's really reductive, because I think it's going to spark something in you that what you want is strong characters who have tension and harmony between them, and it's an art to figure out how that tension and harmony works.

Steven Levitan:

Yeah, because you're not just adding, "Oh, these two don't like each other. That's the dynamic." You've got to go back to the two points. What do you have to do to the two points to make them not like each other? So that's interesting.

You're trying to just not start a scene from zero. We have nothing.

I can remember I did a series a long time ago, and I had a lot of the characters well thought out. Then there was a bartender who worked downstairs at the bar downstairs. I said, "Well, she's just going to be sort of a sarcastic bartender type." I can't tell you how hard it was to write that character once we got into it, because I didn't do the homework. I let it slide, and I could see when that happens. I watch certain shows, and I go, "Oh, they didn't figure that character out." They're searching, and I bet you that's the character that's keeping them two hours later every night because they don't know what's funny, what.

David Perell:

Makes a character funny.

Steven Levitan:

They could be anything. Take a brilliant scientist and put them into a brilliant science lab. Maybe he's not so funny now. Put him in a bar.

David Perell:

So relationships really matter in terms of dealing.

Steven Levitan:

Humor, context, that could be it, or maybe he thinks he's smarter than everybody in the room. So that is funny, because he has very little patience for everybody.

He is maybe the smartest, but he doesn't hide it. So everybody hates him, but they also respect him. Okay, but this one's in love with them because of the way he is.

I mean, you could go on and on. It's also just a funny point of view. Phil always had a funny take on anything. I remember one of the first lines that we wrote that defined Phil was, "I'm the cool dad. That's my thing."

David Perell:

T H A N G. You just knew.

Steven Levitan:

This is a guy who thinks he's up to speed on everything, and he's not. He's a dad. You know, he's a nerdy, embarrassing dad. You're a boomer, man. But he thinks he's as cool as anybody, and that's part of it.

Phil is like a puppy in a meadow. He was just happy and open and onto the next thing, and that's a funny energy against Claire's energy. He was trying to keep things controlled. Phil, we need to control these kids.

David Perell:

I need you to watch the kids.

Steven Levitan:

So many different ways that a character can be funny.

David Perell:

How do you think about the interplay between depth, and then when things feel overly sentimental? It's cheesy.

Steven Levitan:

That's part of the art of it, is to earn those moments and not go too over the top.

David Perell:

Got to earn them.

Steven Levitan:

Yeah, you want them to sneak up on you. You want them to be understated, and you want to feel them. You just want to take those moments.

I mean, I think I tend to write that way. My favorite kind of comedy is the comedy that has both big laughs, smart laughs, and heart that you feel. So what's underneath that character?

There's a great—one of my favorite scenes of all time—by the way, two weeks ago, I was at a charity event, and I'd never met Danny DeVito. I met him and I said, "I have to tell you, I quote this scene all the time as one of the best sitcom TV comedy scenes of all time."

It's when Elaine—Louis had been a well-established character at this point, Louis De Palma on Taxi.

He was the acerbic guy who ran the cab company, and he was hateable, and he was always hitting on everybody, and he was insulting everybody. Danny DeVito, of course, is very short, so that's part of the joke and all that.

He has to go to a wedding or something, and he has Elaine help him go buy a suit at the department store. So Elaine goes with him, and he reveals to her in the midst of all this stuff going on that the reason he's uncomfortable being there at the store is because he has to shop in the boys department. It's very painful for him.

She's like, "Oh, my God, Louis!" This woman who's been the subject of all of his advances and sexual comments—"Oh, my God, Louis." She gives him a big hug, and it's this unbelievably powerful moment.

Then he slowly reaches around and grabs her ass, and it's the biggest laugh because you had—you just took a ride. You went from laugh, laugh, laugh, laugh, laugh to, "Oh, my God, that's so powerful. I just learned so much about this guy, and I'm feeling things I never felt before for this guy, of all people." She felt it, too, and then laugh again, and he's back. It was perfect.

David Perell:

Yeah.

Steven Levitan:

That's a North Star kind of a scene where you're looking to really feel things. Sometimes you don't have to cut it with a laugh.

I could look back on certain Modern Family moments that we did. I don't watch the show very much, but my wife has it on from time to time, and I'll stumble into a scene and I'll see something, and I will immediately start to get teary-eyed from a moment.

I can remember being on set and thinking, "That really got me," and often it was very understated.

An example would be: Jay is going to take Gloria to Napa Valley, go drink wine, have a great weekend. Manny's dad is going to pick him up for the weekend. They're all ready to go, the whole thing. The dad's going to take him to Disneyland, and it's going to be great.

He gets a call at the last minute that dad's still in Vegas; he's not coming. Jay has this moment where he's got to tell Manny. Manny's sitting on the curb waiting for his dad. I still get choked up.

He's got his little backpack on, and he goes out there and says, "Kid, I got some bad news. Your dad can't make it." You can see it on Manny's face because the father is not dependable.

He says, "He's not coming. He wanted to be here. He was doing something heroic. He had to save somebody; he had to do something. He couldn't get away. It's killing him, but they needed him to fix this thing and they can't get him away." You can see the sadness in his face.

Then the limo pulls up to take Jay and Gloria to the airport. Jay says, "It's going to be work out just fine." He cuts to Jay and Gloria taking Manny to Disneyland, or coming home from Disneyland, whatever it was. He sucked it up without saying anything.

He did the right thing for the kid. It was very understated. It wasn't like, "I love you, kid, and you're always a kid's son to me." He didn't say any of that. He's just like, "He wanted me to take you, so we're going to take you. He even got you a limo. Look at this." He just did something that was beautiful, and it was within his character and no more. It was a very poignant moment.

David Perell:

You're really good at the father-kid dynamics. It reminds me at the end of the pilot for Just Shoot Me, there's the conversation with the newspaper owner dad and the daughter. He's about to have a kid, and then she says something to him almost like, if you're scared that you're going to be a bad dad, then just don't do the things that make you a bad dad, or something like that.

There's that beautiful line where the daughter says she's going to want to push you away, but she doesn't actually want that.

Steven Levitan:

It took a lot to get to that, and it was the key. That scene was the key to the whole series because I was having trouble. The Just Shoot Me pilot went through a lot of machinations.

I was a little bit worried about this adult child complaining to her adult dad about, "You're never this, and you're never that."

David Perell:

Because it's annoying.

Steven Levitan:

It's annoying, and you're an adult. There's a part of me that's like, get over it. You're an adult. And part of that was also in the construction.

Originally, it was set in a modeling agency, and you can't come into a modeling agency. She, as a character who was a feminist, can't come into a modeling agency and say, "We're going to do this differently so that it's more pro-women's rights or whatever." It's a modeling agency; it is what it is.

So then, very late in the game, I changed it to a Cosmopolitan, Cosmo magazine. Then she's a journalist, so she could actually have something to aim for, which is to make a difference, to help steer it in a better direction, to be able to try to make it more substantial.

With the whole device of him having this baby and it being sprung on her, it wasn't her saying, "You screwed me over. You weren't there for me. You were never there for me." She said, "Now, that's a selfish way."

She got in a selfless way and said, "Be there for this kid. Be there for her." So she's both giving and revealing what she went through. And that was the key to that character, to that moment, to that relationship. She was always there to try to help him be a better person, not just to put him down.

David Perell:

If I were to show up, and now I'm on your writer's team and it's day one, and you're like, "Hey, here are some of the mantras, some of the frameworks that I'm always thinking about. I'm always going to come back to these. So know them by heart. Get to know the examples." What would those be?

Steven Levitan:

Does it feel real?

David Perell:

Does it feel real?

Steven Levitan:

First and foremost, are they talking the way people talk? Are you tapping into what you're thinking about today?

Writing is so interesting because my wife had never seen this old sitcom from the late 1970s that was a classic, and she'd never seen it. So I said, "Let me turn on the pilot for you. You just watch the pilot." I think of it as a classic, beautiful, unbelievable pilot. It's so dated in the way it looks, the way it's shot, the way sometimes just the attitude.

That's fine to do. Be of the moment, tap into the zeitgeist. But at the same time, what are classic things that couples deal with or fathers and sons deal with? What are timeless things that can live on always?

Let's not get over our skis. There's a tendency that happens more in multicam than it does in single cam to some degree, where you get on the stage and something doesn't work. Then you start pitching and start pushing, and it gets bigger and bigger until you get a big laugh from that big hyped-up audience. Then you see it on film or on digital now, on screen, and it's big. It feels big.

I'm trying to think about, I'm watching this three years from now and what's my attitude about it? Certain things don't change. That's why I think so far, Modern Family is aging well. I'm happy with that. It's aging pretty well because that stuff is still going on between parents and kids.

David Perell:

Till the end of time.

Steven Levitan:

Yeah, till the end of time. By the way, across the world, we were shocked at markets. Here we are setting a show on the west side of Los Angeles, a very affluent area, and it's resonating.

David Perell:

The guy who edits these podcasts, he's Brazilian. Johnny loves Modern Family.

Steven Levitan:

Johnny, thank you.

David Perell:

Down in Sao Paulo, he's like, "I've just watched so many episodes. I just love that show, man."

Steven Levitan:

Yeah. Because family is universal. So try to tap into those universal truths.

David Perell:

The thing that makes it feel real is that it's like the inverse of trying. A lot of what you're railing against here is being there on set with this thing, this thing, this thing. A lot of how Modern Family was written was just, you're at the table and you're just writing down exactly what they said.

That's just how it is. And there's a kind of trust in the mundane parts of life that I think you've had to have and just be like, "That's just how it is. It's real." And just trusting that because it's mundane, it's going to have a lot of relatability too.

Steven Levitan:

Yeah. I think that much of really good writing, and nobody's better than Larry David, is having your antenna up. I always say, as a writer, your antenna needs to be up at all times.

If you're observant, there's almost nowhere you can't go where you just go, "Well, that's funny that they did that." It annoys me that that guy was double dipping into the dip. That's a thing that nobody else would see and make note of, but a great writer like Larry sees that and says, "Oh, that's funny, that annoys me. That's going in the book."

Wouldn't it be interesting if this happened here? That's going in his little notepad, and that's going to become something. We would all say to each other in the room, "Be observant this weekend. Find something that we need a story for." That's part of it.

David Perell:

Well, it's funny, I was in Portland recently, and I just finished the day, and I just love Guinness. So I always try to get a nice pint of Guinness whenever I'm traveling.

Steven Levitan:

I was just in Ireland.

David Perell:

It's just the greatest drink ever made. It's basically the only alcohol I drink, but I could just drink it over and over. So good.

I'm having this pint, and I'm like, "You know what? I'm going to try something." I went up to the bartender and I go, "Hey, do you got a piece of paper?" He goes to the cash register and gets a long piece of paper, probably five inches big. I'm just going to fill out this whole piece of paper with thoughts, thoughts that I'd had that day. I did it. It was just like, "Hey, the roses in Portland were like this. These people are like that. Hey, there are these weird fentanyl ads on the street."

You realize that under the register of consciousness, there's so many thoughts that you have that are just kind of there waiting to be mined. But in the speed of everyday life, you just skip over those things.

Steven Levitan:

Well, especially now. I think one of the problems with this moment we're living in is we're all flooded with information. It's just never ending.

David Perell:

Flooded.

Steven Levitan 63:27 - 65:29

It's exhausting. It's very important to just be aware. If you get stuck on something, go work on something else. Creativity is connections. You're making new connections.

We're going to do a family show, a classic family show, and we're going to mix it up with sort of a documentary-looking thing with interviews, but we're also going to add a gay couple. That's a new twist to that old thing. We're just connecting things that have all been there. People have talked about technology before, but maybe not in the context with a family and how it's affecting family lives or whatever.

Just being aware of all that's happening in the world right now and trying to turn down the noise and walk around. If you're writing about a couple, you're writing a love story or something, being aware, thinking about some of the people that you've known and the relationships you've seen, or you're at an airport or you're at a restaurant, look at that couple and think: isn't it interesting the way that he keeps looking over at that woman across the way, and she seems completely annoyed by him or whatever that is? I wonder if that's an interesting moment for something here.

So it's hard to do. It takes discipline, but stay very aware. When you take your mind off of something and then you're working on something else, you go, oh, wait a minute, that thing he just did in there, what if that guy did something like that? That would change everything there. That's how connections are made.

David Perell 65:30 - 65:33

You said, does it feel real? What else would you say to people?

Steven Levitan 65:34 - 66:01

I think knowing the theme is really important. You may not know it at the beginning, and it may change, but to be able to have a theme. Some of our best episodes in Modern Family, for example, we had three stories, and we would try to stick to a theme. I wrote one called 15%, and it's about: can people change?

David Perell 66:02 - 66:08

The other one is Caught in the Act. I think that one has the stain on the carpet and the kids walking.

Steven Levitan 66:08 - 66:09

In on the parents, yes.

David Perell 66:10 - 66:14

So Caught in the Act actually refers to multiple things in the episode.

Steven Levitan 66:14 - 67:16

Yeah. So we had that idea that came from a real life story. Then we had the idea, like, what's another version of getting caught in the act?

Thematically, we came up with the idea that if they try really, really hard, people can change maybe 15%, but that can be enough. That was the idea of it. Sometimes being a good dad is just showing up, things like that. What do we want to say with this episode?

If you can have that, and then all three of those stories in some way feed that little voiceover at the end, which we didn't always do because we got worried about being too predictable, that here comes the voiceover. So we would try to mix it up. But really knowing that, I think, strengthens everything. I mean, I think you have to know it at some point. It's best to know it from the beginning.

David Perell 67:16 - 67:16

The theme.

Steven Levitan 67:16 - 67:49

The theme. You're never going to think about things the same way someone else is, so you need to be more open to someone who completely disagrees with you. Whatever that is, fine, that's a theme.

Whatever it is, it could be more heartfelt, it could be more just a nice observational moment. What are we trying to say with this episode? I think knowing that from the beginning is really, really helpful.

David Perell 67:50 - 68:17

How much of writing a show like this comes back to the word joy? That's the word that I think of when I think of what I want to get from a show like that. I want to sit down for 21 minutes, 30 seconds, and I want to have a sense of joy. Is that what you feel as the creator? That is the one word—joy—which kind of mixes laughter, a sense of fun. I don't know if you resonate with it.

Steven Levitan:

With that, it's funny. I don't watch a lot of television, and there have been some amazing shows that have been on that I haven't seen because they're angst-producing. They're really intense, and you're on the edge of your seat, and oh my God, I can't watch this.

Contrary to popular belief, I'm very sensitive, and if I see something that I find very upsetting or I'm just uncomfortable, I want to change the channel. It just bothers me and I'm not enjoying it. So I tend to like to do shows that bring people joy, that make people happy.

David Perell:

I had so much fun prepping for the episode. I woke up 5 a.m. this morning, and I just watched Modern Family all morning. It was so fun.

That was the word; it's like there's so much joy. I watched no TV and it was just like I was just smiling the entire time.

Steven Levitan:

I think that there's just something. There's nothing wrong with those kinds of shows, and they're brilliant, but at the end of the day, Modern Family was supposed to be a joy-inducing show. It was a show that made people happy. It was a show that celebrated family.

It was a love letter to our families. I can't tell you, we've gotten a lot of feedback over the years. My two favorite versions of those would be the ones from either gay teens or gay young people who said because of your show, my parents, who were very conservative, they liked Mitch and Cam, and they would laugh at Mitch and Cam. It opened the door for me to be able to tell them that I'm gay, and it saved my life.

Or from the parents who would say, I got used to having the idea of having a gay kid through Jay, through what Jay was going through because he wasn't entirely comfortable with it. Because of that, when my son came out to me, I handled it way better than I would have, and it made us closer. We got a lot of thank yous for that.

The other one, which is just as meaningful for me, is I was going through hell. I was sick. I had just lost my parent. I lost my job. I was at the end of my rope, and I would turn on Modern Family, and I would smile for a half hour. I can't tell you how much that meant to me. I'm so proud and happy to be part of that and to put that into the world because there's not enough of that.

David Perell:

You mentioned conflict and how core that is to a show. Tell me about that.

Steven Levitan:

It's hard to have a good story without some conflict.

David Perell:

It's the tinder for the flame.

Steven Levitan:

You need to have something go wrong. You need to have somebody to have a goal, and there's an obstacle in the goal, and then that conflict is where the comedy comes from.

Jim and Pam on The Office are such an important part of that show for giving it depth and heart, but there's not a lot of conflict between them. In the beginning, there was the conflict of she was engaged to somebody else, but once they were together, they were a loving couple, and there wasn't a tremendous amount of conflict. The conflict came from elsewhere in the show, and that's where someone has to.

If someone's faced with conflict, they do extreme things. They have to take extreme measures. Sometimes you look at a story and you go, nothing happens. There's nothing happening here.

The conflict is this big. Can we make the conflict bigger? That's a big part of this.

David Perell:

Bigger.

Steven Levitan:

Yeah, bigger, and not in an over-the-top kind of way, but can we make it more core to this is really causing some problems in their lives?

David Perell:

Bigger, not in terms of giant, but in terms of meaningful.

Steven Levitan:

Meaningful, like, "Oh, this is not just an easy thing to go. I'm sorry, and it's over with."

It's, "Oh, no. I really think that this person did something wrong for me, and I'm going to say something to them, and then that's going to have ramifications."

David Perell:

How, over the years, have you gone about improving your craft?

And the arc there of, "I'm going to get better, I'm going to get better, I'm going to get better."

Steven Levitan:

I've tried to be more thoughtful about things, and maybe now I'm probably slowing down a little bit.

Like, I could, in the beginning, just race through and, "Here's the draft," and dah, dah.

Now I'm trying to slow down and be a little bit more deliberate about it.

I think you just learn from doing the job.

It never gets old because there's just a new problem that you haven't faced every single story.

Oh my God, how am I going to write myself out of this one?

And then you do, or you see something that somebody did, and you go, "That's great." and you try to learn from that.

So, I'm trying to be more thoughtful about the setup of a series.

I mean, I was in Modern Family.

We certainly were.

I think prior to that, sometimes I had been taking a couple shortcuts, and so I'm really trying to put in the time in the beginning when it's not so sexy.

I sort of was on this idea that we're all professionals here, and we're writing, and we got to write hard, and you didn't deliver, and I'm not going to sugarcoat it.

I didn't think this draft was very good, and dah, dah, dah, dah.

I probably was a bit too direct at times.

I can be better at getting what I want out of a writer by taking the time to be more thoughtful about where they're coming from and what their particular psyche is.

I think I come from an environment; I'm very tough on myself as a writer, very hard on myself.

The person who runs my production company says I've never seen someone who's so hard on their own writing.

David Perell:

How does that show up?

Steven Levitan:

I just am brutal about it.

I don't like this scene.

I'm not happy with it yet, or I'm going to throw this whole thing out and start again.

I just do that all the time.

Okay?

And I am that way with other people because I demand it of myself, but it doesn't.

I'm tough, especially if it's coming from me.

So, I know it comes from a place of, "I just want this to be good," and I think that sometimes other people don't.

They have trouble seeing that in that moment where they just feel like, "I suck. You know, I've let them down."

I can undermine somebody's confidence as opposed to building it up to say, "Hey, let's really talk about this. Let's be more..."

So I'm trying very hard to be more open about the process in the very beginning before we even start.

This is how this process is going to go.

David Perell:

The initial conditions.

Steven Levitan:

If I'm going to say some stuff to you, it's not just because of you or what I've read here. I do this for everything, and just know this. I want to be more effusive when I see something that I like because I usually zero in on the problems we have to fix. That's the mentality of doing a network show for 11 years: Let's fix the problems; the rest of it works great.

So, you might go through the script and think, "That's a great joke." But I might say, "I got nothing on page three," which means I like page three. It should be more like, "Page three is great, page four is great. I love this thing you did here." Then, "Page five, I'm starting to have little questions about this, whatever." I think that's the area I've been thinking about the most lately.

How can I be a better partner to writers coming up who could use the guidance? I think I have a lot to offer in terms of understanding how networks work, how a show needs to work to keep the trains running, how to ask yourself the hard questions in the beginning, and then let them do what they do.

They bring their particular vision, their language, their comedy, their music, and their fashion because they're 25 now, and I'm not. They're going to bring something to that, which is different than what I'm going to do, and I have to honor that, respect that, and help them build that and make it stronger from underneath, not go, "Well, I wouldn't do it this way; I would do it this way." I might do it a different way, but let's make sure that the foundation is strong, and then from that, you should say it the way you want to say it.

David Perell:

Tell me, how is writing comedy different when you have laugh tracks like Frasier versus no laugh tracks with Modern Family?

Steven Levitan:

It used to be every comedy was a multicam.

David Perell:

Can you explain the significance of the difference between single cam and multicam from the writing and the director's perspective? That's not clear to me.

Steven Levitan:

Multicam is shot in front of a live studio audience. It's a three-wall set, so the audience is watching you from that fourth wall along with the cameras.

David Perell:

So, like when I went to go see Seth Meyers, it was sort of like that sort of thing. We were just there, and you're laughing. They got the cards, "Laugh now," and stuff like that.

Steven Levitan:

We don't do that, but there's an audience there, and they're hyped up because there's a stand-up warm-up guy there who's keeping the audience alive, and they're laughing, and everything's funny. It's all very much about the laughs.

Do we get big laughs? Does that scene feel like the audience is with us? Did they start to get bored, and all that? But you're writing jokes more.

Suddenly, there were some single camera comedies. The single camera came around, and then that's shot like a movie. So on a stage, you're shooting it; there's no audience, there's no one laughing, you're not leaving room for laughs, all that stuff. You're just doing the scene honestly, and hopefully it's funny.

In the beginning, the mistake I think that was made by many of those shows was, "Oh, we're single cam. We don't need to have jokes. We don't need it to be as funny because we're not that kind of a show."

Then you ended up with these half-comedy, dramedy, half-hour things that were not that funny.

What shows like The Office did, and certainly what we tried to do, we said, "No, we need to have even more jokes, even better jokes, smarter jokes all throughout this thing."

I didn't have to really say this, but I do remember saying this to one of the actors in the very beginning: "Pretend like you're a drug dealer, you're being chased by the cops, you have a bunch of drugs on you, and you're just going to crack open the door and drop the drugs out on the street as the car's rolling along. That's the way I want you to deliver the jokes. Just throw them away."

Don't even let it because sometimes you hit a joke. Like, "Oh, here comes a great joke. Well, maybe that's why you should've..." Or you just say it, and you just throw it away until the audience can go, "Oh my God, that was so funny."

Let's see the little pot sticker, or Lily, wouldn't that be hard for her to? You just throw them away. The audience gets them; they're little chuckles; they appreciate it; they love them; they watch it; they go, "I missed that the first time through," and they get it the second time through. That's a different way of performing. It makes a big difference in the type of performance you're giving and the pressure of a joke.

If you go too many pages without a joke in a traditional multi-camera sitcom, people get nervous. But Frasier was among the first shows to kind of break through that a little bit.

No, we're going to purposely go three pages without a joke. And then I would remember saying a lot of times, "This is a half-joke. Let's get rid of it so that the real joke plays better. We don't need a half-joke just to get a chuckle."

David Perell:

When you watch bad comedy, and it's agitating you, what is it that you want to say to the TV? Shake the TV and be like, "If you only understood this, it would be so much better."

Steven Levitan:

Take it down. I get annoyed by big. I've said that before here, I think, but it feels when things just feel heightened to the point where I don't recognize this as human behavior, or they're talking like they're on a stage performing to an audience.

It's really hard, by the way, doing a multicam well. It's very hard because it is this weird hybrid, halfway between theater and movies. You're playing in front of an audience, so you tend to play to that audience, but it needs to be small enough that it looks good when that camera's on you like this, the way that 99.9999% of the people are going to see it.

So take it down. Be more natural. Don't force it. Don't go for the half jokes. Just wait for the good joke. Just talk like real people and then get to a good joke.

David Perell:

Why, at the risk of touching the third rail here, why is Hollywood so against AI as a culture? You know, I come to Hollywood, and it is just the hard-line anti-AI stance.

Steven Levitan:

I think people are very scared of it. If the studios can do certain things for a lot less money, just as well, then they're going to. If they can do it for a lot less money, 90% as well, many of them are going to. The good ones won't.

So I think we're all—everybody should be scared. Everyone should be amazed, and everybody should be frightened and cautious of it. I'm a little obsessed with AI and I read a lot about it. I think the big danger, of course, is that the natural brakes that we would put on such a powerful technology, we're not putting on because we're afraid to lose the race. And the net result is that we've all rushed through before we put in the appropriate thought safeguards.

I think it could be an incredibly helpful tool, even in writing, to use it as research, to use it as—I want to learn more about this character. What are the kinds of jobs people do? What would a person who is a—whatever, a roller coaster designer—what do they do? Where do they work? What's their job? What's their background? Asking things like that.

You could run your script through it and see, put this against Joseph Campbell's hero's journey, and tell me where I'm falling short. You could do things like that, and I don't think there's anything wrong with that if you're not a slave to it. If you're discerning and you say, okay, well, that's nonsense, or, okay, great, but I'm not doing that. But you know what? That's actually a good point. I don't do that here, and maybe I could. If you use it as a tool, it could be useful. It can't do what we do yet. I hope it doesn't.

David Perell:

Well, there's two things. So the first thing is I'm going to go out on a limb and say it. I think AI is funny now. Ever since ChatGPT 4.5 came out, I laugh all the time.

Steven Levitan:

I hate to say it, I think it's gotten funnier.

David Perell:

It makes me laugh so hard. The thing that AI doesn't have, which you need for a TV show, and I've really picked up on here, is conviction.

There is a conviction that is required in order to push forward on a show and in a story, and AI can't give you conviction. It can give you ideas. It's sycophantic, but there's a conviction that humans, only humans can have. AI gives you volume, but it does not really do a good job of helping you chart your path.

Steven Levitan:

Yeah, and it's not to say that it won't again. It's just staggering.

Listen, in production, there are some incredible uses for it. I'll give you an example. We have a set and we need to find an exterior shot to match that set. In the old days, it's like, oh, fuck, these windows don't make any sense here. There's a door there. Where's the door in the inside house? Well, we could literally say we need an exterior shot of this with this window configuration. And we need a person riding their bike by and it's morning, and get a stock shot, for example, and do that. That's available today. Easy. That's nothing.

Does it take a couple people out of a job? Probably. I think using AI for production to help make some very expensive scenes affordable is inevitable. I just do. We can all wish it away. I think it's inevitable. My hope is that by doing that, we can produce more. Maybe we can even start producing more in LA.

David Perell:

Because nothing gets filmed here anymore.

Steven Levitan:

Yeah, because we need more shows to be made, and instead of a show costing $11 million for a pilot, maybe we can do that show for six now because of all the savings that we can do using AI.

David Perell:

Interesting.

Steven Levitan:

Because a lot of that was all that post-production work, those people are going to be affected. By the way, lawyers are going to be affected, illustrators are going to be affected, and just everybody's going to be affected.

It's a little grand of us to say, well, they should be, but we shouldn't be because we're artists. It's going to affect everybody. How can we use it to maybe make more things so that more people can go to work on more projects instead of putting $11 million into one thing?

David Perell:

We were talking about the initial conditions, the time that you spend setting up characters early on in a show, and the web of relationships and the depth of the characters.

When I think of humans and AI working together, if you can make that explicit, Gloria is going to be like this, Jay's going to be like that. We're really going to spend time defining that. My goodness, AI's ability to be a collaborative sparring partner with you, to really understand what it is that you're going for, what are the things that I might not be seeing. That, to me, is the platonic ideal of what we're going for, which is you've created a show, you've defined the characters, and now you just have this kick-ass writing partner.

Even in groups, sometimes it gets harder and harder as the group gets bigger and bigger. I just think that we're going to enter a time when writing groups will probably be smaller, more collaborative. AI will be there with you in real time giving suggestions. A lot of the work is just going to be making things more explicit so that it can actually help you and understand what's going on.

Steven Levitan:

What worries me, and it just might be because of nostalgia and caring for other human beings. At this point, I don't like the idea of those three writers that were just eliminated from that group that had a job and now don't. I don't love that.

David Perell:

But the flip side is they can make films. It's going to be cheaper for them to do it.

Steven Levitan:

I think that if that's the case and that we're allowing more people to do more things, then great. I like the idea of it as a tool for humans and not a replacement for humans, the way that a computer.

When I started writing scripts, I was on a typewriter. The idea of then going to a sophisticated word processor where I could change and move things around was unbelievably important for my writing style. It was so helpful for me. That's what I'm hoping this is, that I can just sit there and go, "God, I need a good way. What's a good thing to do here?" I'm using it as that partner sitting next to me all the time.

Would I still want the partner? I would. Will I get it going forward? I don't know. I would like to have that because I think that humans, so far—and again, I don't know what's going to happen in the future. I'm so afraid of saying something like, "AI will never be as good as a good person," and then two months from now, you look like an idiot.

It's getting so good so fast so far. Humans, some of the brilliant, amazing, weird, damaged, hilarious people that I know come up with things out of left field sometimes that are so smart and so unpredictable that I don't know that AI would have come up with that.

David Perell:

That's why I come back to conviction because it requires a real sense of taste. Even if AI came up with it, to say, "That's the best idea, and I'm going to put my spine on that," that's the first thing.

The other thing that is standing out is I have pretty high confidence that it's going to take a lot of jobs and put a lot of people out of work. At the same time, I have very high confidence that it means that people can make films with less budget, with smaller teams. Someone who used to just be a writer can now do all of these other things, and they, on their own, can take a film, a sitcom, a documentary, whatever it is, from start to finish way easier than they used to be able to.

Steven Levitan:

I'm hoping that all those people before, when I was talking about the things that you can save from production to take the cost out, I'm hoping those people who you've put out of work, the ones doing that, can then be part of that movement to help those who are technologically challenged to make their movie.

Like, now the guy who is a special effects person today, working in that, can then become sort of the person who is the liaison between AI and the writer. There's a role for those kinds of people who are good technologically, where many people just aren't. They're going to be able to rely on somebody whose job is now a master of getting AI to do the things that you need it to do.

So I think that there's room for that. The people who are driving the trucks to those locations, the people who are shooting that stock shot, etcetera, etcetera, I worry about them. The musicians who are doing scoring, all sorts of jobs up and down, I worry.

David Perell:

To go back to your original point, the thing about AI that makes it such a head-scratcher is that the game theory is such between the different AI companies and between America and China and just the geopolitical game theory that it has now gone full accelerationist, almost like a modern Cold War type thing in certain ways.

Because of that, it's all happening so much faster than our ability to process what's going on and ask, wait, what's good for society? It's the technology that's leading the ship and not the humans themselves. That's like, ah.

Steven Levitan:

I don't know if you read AI 2027.

David Perell:

No, it's an amazing piece from Slate Star Codex.

Steven Levitan:

It's about feeling the pressure to move forward and not putting in the proper safeguards. Are we doing that just from self-preservation? Like literally, is AI going to try to wipe us out at some point?

It sounds like science fiction, but according to them, and they seem very smart, it's real. We have to be careful; we have to be thoughtful.

I don't think we can be. In the last writer's guild election, or the strike, AI was a big part of that. I was very nervous about that. I'm like, how do you stop progress? I don't know how you stop it.

If you said to any writer, I have something here that can help you do a better job on your script faster and better, and it's available to you right now, and you can work it, that writer is going to want to use that tool. It's just like saying, would you like to have a couple of smart people in the room with you to help you? Some people are going, nope, I'm on my own. I want to do this. The pride is going to be for me 1000% doing it myself. But I think that it's inevitable that we will be using AI in some way.

So the idea of like, we're going to stop everyone who says, nope, nope, nope, we're not going to do it. I think those people are being naive.

David Perell:

Tell me this. When I talk to writers, I have one friend who's a novelist who's basically using Cursor to do feedback on his novel drafts. He's like, I hired an editor. They told me it was going to be a six-week turnaround. It's been three months, and they haven't gotten back to me. I'm now in Cursor, and I've gamed Cursor so they can give me really good feedback. It takes about 20 to 30 minutes.

What I notice with the writers is that the people who are anti-AI are very loud. The people who are pro-AI are quite quiet because there's a sort of taboo. What you hear behind closed doors is very different from what people say in public. Is it like that in Hollywood?

Steven Levitan:

I haven't talked to a lot of other writers about how you're using it. I should ask some friends what they're doing with it. I'd be so curious.

I know that there are always two kinds of people: the ones who are ready to jump aboard anything that's new and helpful and cool; let's try it, let me see what this is all about. And there are those who are just going to be the Luddites, who are going to be more inclined to just want to do it themselves and do it the old way. I don't see that in a derogatory way. That's the way they want to work.

We don't talk enough about, and I've seen hints at it with AI, is how demoralizing it might become.

David Perell:

Oh. I mean, it's already there. I will look at stuff that AI writes, and I'm like, that's better than I could have written it. It gives me way less pride in the skills that I have.

Steven Levitan:

Yeah. I saw someone's post that said, I spent a week on this presentation, and AI did it in five minutes, and it was better.

David Perell:

It is demoralizing.

Steven Levitan:

How do you attack the next presentation with the same amount of vigor? I think that's something that worries me. I mean, there are many positives to it, and there are many cautions or things that we should be cautious or be afraid of.

David Perell:

I really enjoyed this. Thank you.

Steven Levitan:

My pleasure. I really enjoyed it. I hope this was not dull. I always worry I get into the weeds, but thank you.

I'm glad to be here. It was nice to meet you, and I think that anybody who is focusing on writers and writing is a friend of mine, so thank you.

David Perell:

Thank you.

Discussion about this video

User's avatar